title
stringlengths 1
827
⌀ | uuid
stringlengths 36
36
| pmc_id
stringlengths 5
8
| search_term
stringclasses 44
values | text
stringlengths 0
8.58M
|
|---|---|---|---|---|
Prognostic refinement of NSMP high-risk endometrial cancers using oestrogen receptor immunohistochemistry | 692dc1d4-2947-43d5-8b76-69e067855294 | 10050005 | Anatomy[mh] | Endometrial cancer (EC) is the most common gynaecological malignancy in postmenopausal women . Although the majority of patients present with early-stage disease and have a good prognosis, 15–20% of women with EC have unfavourable disease characteristics that are associated with an increased risk of distant metastases and EC-related death . In the 2016 ESMO-ESGO-ESTRO guideline, high-risk EC was defined as stage I, grade 3 endometrioid EC (EEC) with deep invasion, stage II or III EEC, or non-endometrioid EC (NEEC) . For these patients, adjuvant pelvic external beam radiotherapy (EBRT) was the standard of care to improve locoregional control . The randomised PORTEC-3 clinical trial showed that the addition of adjuvant chemotherapy to EBRT (CTRT) increased overall survival (OS) and failure-free survival (FFS) of patients with high-risk EC by 5% and 7% at 5 years, respectively . The greatest OS benefit of CTRT was observed in stage III EC and serous carcinomas (SEC) . Unfortunately, histotype and grade assignment of EC is subject to substantial interobserver variability, hampering the selection of patients that would benefit from CTRT and reducing overtreatment for those who do not . The EC molecular classification, consisting of the POLE ultra-mutated ( POLE mut), mismatch repair-deficient (MMRd), p53-abnormal (p53abn) and no specific molecular profile (NSMP) molecular subgroups, has repeatedly shown to have strong and independent prognostic value and is also predictive for response to chemotherapy . For this reason, the EC molecular classification was incorporated in the latest European treatment guidelines . The assessment of the molecular classification is encouraged in all EC, especially in high-risk tumours, and a novel risk stratification incorporating the molecular classification has been introduced . All stage I-II POLE mut EC are classified as low-risk EC and adjuvant treatment can be safely omitted. In contrast, all p53abn EC with myometrial invasion are now considered high-risk and adjuvant chemotherapy with or without EBRT is recommended . The risk-assessment of patients with MMRd and NSMP EC, however, still depends on clinicopathological features such as stage, histotype, FIGO grade and the presence of lymphovascular space invasion (LVSI). The excellent clinical outcomes of patients with POLE mut EC, the intermediate prognosis of MMRd EC and poor survival of p53abn EC has consistently been shown across different cohorts and clinical trials . In contrast, 5-year recurrence-free survival of NSMP EC has varied between intermediate and poor . This heterogeneity in clinical outcomes hampers adequate adjuvant treatment recommendations and suggests biological diversity. Several molecular alterations that are not included in the current risk stratification have shown to be associated with clinical outcomes in EC, such as CTNNB1 exon 3 mutations, overexpression of L1CAM, lack of oestrogen receptor (ER) and progesterone receptor (PR) expression, chromosome 1q amplification and other copy number alterations. However, the prognostic relevance of these molecular alterations in high-risk EC, in the context of the EC molecular classification, as well as in relation to each other, is less well understood. These molecular alterations may refine the molecular classification and identify subsets of NSMP EC with a distinct prognosis. Using a large set of molecularly classified high-risk EC from the PORTEC-3 trial and a prospective cohort study, we investigated how ER, PR, L1CAM and CTNNB1 mutations and established clinicopathologic and molecular risk factors can improve EC risk-assessment.
Patient and tissue selection This study included patients who participated in the international PORTEC-3 randomised clinical trial, and the prospective clinical cohort of Medisch Spectrum Twente (MST). The design and results of the PORTEC-3 trial have been published previously . In short, this international phase-III trial randomly assigned 660 eligible patients with high-risk EC (1:1) to postoperative chemoradiotherapy or external beam radiotherapy alone. Inclusion criteria for the trial were: International Federation of Gynecology and Obstetrics (FIGO) 2009 stage IA grade 3 EEC with LVSI; stage IB grade 3 EEC; stage II-IIIC EEC of any grade; or non-endometrioid EC with stages IA (with myometrial invasion), IB-IIIC. Upfront central pathology review confirmed the eligibility of all patients . The presence of LVSI was dichotomously scored as present or absent. The study was approved by the ethics committees at all participating centres. Written informed consent was obtained from all patients. The prospective cohort study MST included 271 high-risk EC patients who were treated with adjuvant radiotherapy between 1987 and 2015 at Medisch Spectrum Twente, Enschede, The Netherlands. Pathology review was performed by MB, SR and TB to confirm high-risk disease. In contrast to PORTEC-3, LVSI was scored using a 3-tiered scoring system (e.g., no LVSI, focal LVSI, substantial LVSI) . As focal LVSI was not associated with an increased risk of recurrence in previous study , we combined focal LVSI with no LVSI in a final dichotomous LVSI variable. The current study was approved by the Leiden-Den Haag-Delft medical ethics committee, and a waiver for informed consent for the MST cohort was given. Immunohistochemistry Formalin-fixed paraffin-embedded (FFPE) tumour material was available for molecular analyses from 424 (64.2%) PORTEC-3 and 256 (94.5%) MST patients. Whole slide (4 μm) immunohistochemical (IHC) staining for MMR proteins (MLH1, PMS2, MSH2, and MSH6) and p53 on all PORTEC-3 cases was performed and described previously . Similar IHC staining and scoring for MMR proteins and p53 were performed on cases from MST. When no slides were available for IHC or MMR IHC failed ( n = 11), MSI status was determined using the MSI analysis system, version 1.2 (Promega, Madison, WI). In addition, IHC staining for L1CAM, ER and PR was performed on whole slides for all cases. The percentage of positive staining for L1CAM, ER and PR was noted and a 10% cut-off for positivity was used for all three stains, as this cut-off is commonly used for the assessment of L1CAM, ER and PR expression in EC. A detailed description of all IHC procedures and scoring is available in the Data Supplement. Next-generation sequencing Isolation of tumour DNA for targeted next-generation sequencing (NGS) was performed as described previously . Samples were sequenced using the AmpliSeq Cancer Hotspot Panel version 5 (PORTEC-3) and version 6 (MST) (Thermo Fisher Scientific, Waltham, MA). The presence of pathogenic somatic mutations was evaluated, considering a minimum coverage of 100 reads and variant allele frequency of 10%. A detailed description of DNA isolation and sequencing is available in the Data Supplement. When no slides were available for IHC or p53 IHC failed ( n = 20), the final p53 status was determined by the TP53 mutation status. In cases with failed NGS, KASPar competitive allele-specific polymerase chain reaction (LGC Genomics, Berlin, Germany) assays were used to screen for hotspot mutations in POLE (including codons 286, 297, 411, 456, and 459) as previously reported . Evaluation of IHC and sequencing results was performed blinded to each other and patient outcome. Statistical analysis The primary endpoint was recurrence-free survival (RFS); calculated from the date of randomisation (PORTEC-3) or date of start of adjuvant treatment (MST) to the date of the event of interest, or date of the last follow-up in patients without events. Secondary endpoints were locoregional recurrence-free survival (including vaginal and pelvic recurrences), distant metastasis-free survival (including para-aortic, abdominal and other distant recurrences), and disease-specific survival (DSS). For locoregional, distant and overall recurrence-free survival, event-free patients who died due to other causes than EC were censored. Differences between groups were tested using the χ 2 test or Fisher’s exact test for categorical variables, and with the Mann-Whitney U -test for ordinal and non-normally distributed continuous variables. Median follow-up time was estimated using the reverse Kaplan–Meier method. Survival analyses were performed according to Kaplan–Meier’s method and groups were compared with the log-rank test. Cox’ proportional hazards models were used to evaluate the prognostic value of (established) clinicopathological and molecular features in the complete study population, as well as in the molecular subgroups separately. Step-wise backward likelihood ratio-based variable selection with stratification for cohort was applied to build multivariable models. The relative importance of variables included in the multivariable models was based on the variable’s proportion of the χ 2 statistic. Model validation was performed by analysis of discrimination and indices of optimism determined by means of model fitting to 1000 bootstrap resamples. In addition, internal validation using the leave-one-out method was performed by re-estimating on the two cohorts independently. Comparison of fit between multivariable models was performed by means of Akaike’s information criterion (AIC), model concordance (C-statistic) and likelihood ratio test for comparison of nested models. A two-sided p -value <0.05 was considered statistically significant. Statistical analyses were performed with SPSS (Statistical Package of Social Science) version 25 (IBM, Armonk, NY, USA) and R (version 3.6.3., https://r-project.org ) using the survival package.
This study included patients who participated in the international PORTEC-3 randomised clinical trial, and the prospective clinical cohort of Medisch Spectrum Twente (MST). The design and results of the PORTEC-3 trial have been published previously . In short, this international phase-III trial randomly assigned 660 eligible patients with high-risk EC (1:1) to postoperative chemoradiotherapy or external beam radiotherapy alone. Inclusion criteria for the trial were: International Federation of Gynecology and Obstetrics (FIGO) 2009 stage IA grade 3 EEC with LVSI; stage IB grade 3 EEC; stage II-IIIC EEC of any grade; or non-endometrioid EC with stages IA (with myometrial invasion), IB-IIIC. Upfront central pathology review confirmed the eligibility of all patients . The presence of LVSI was dichotomously scored as present or absent. The study was approved by the ethics committees at all participating centres. Written informed consent was obtained from all patients. The prospective cohort study MST included 271 high-risk EC patients who were treated with adjuvant radiotherapy between 1987 and 2015 at Medisch Spectrum Twente, Enschede, The Netherlands. Pathology review was performed by MB, SR and TB to confirm high-risk disease. In contrast to PORTEC-3, LVSI was scored using a 3-tiered scoring system (e.g., no LVSI, focal LVSI, substantial LVSI) . As focal LVSI was not associated with an increased risk of recurrence in previous study , we combined focal LVSI with no LVSI in a final dichotomous LVSI variable. The current study was approved by the Leiden-Den Haag-Delft medical ethics committee, and a waiver for informed consent for the MST cohort was given.
Formalin-fixed paraffin-embedded (FFPE) tumour material was available for molecular analyses from 424 (64.2%) PORTEC-3 and 256 (94.5%) MST patients. Whole slide (4 μm) immunohistochemical (IHC) staining for MMR proteins (MLH1, PMS2, MSH2, and MSH6) and p53 on all PORTEC-3 cases was performed and described previously . Similar IHC staining and scoring for MMR proteins and p53 were performed on cases from MST. When no slides were available for IHC or MMR IHC failed ( n = 11), MSI status was determined using the MSI analysis system, version 1.2 (Promega, Madison, WI). In addition, IHC staining for L1CAM, ER and PR was performed on whole slides for all cases. The percentage of positive staining for L1CAM, ER and PR was noted and a 10% cut-off for positivity was used for all three stains, as this cut-off is commonly used for the assessment of L1CAM, ER and PR expression in EC. A detailed description of all IHC procedures and scoring is available in the Data Supplement.
Isolation of tumour DNA for targeted next-generation sequencing (NGS) was performed as described previously . Samples were sequenced using the AmpliSeq Cancer Hotspot Panel version 5 (PORTEC-3) and version 6 (MST) (Thermo Fisher Scientific, Waltham, MA). The presence of pathogenic somatic mutations was evaluated, considering a minimum coverage of 100 reads and variant allele frequency of 10%. A detailed description of DNA isolation and sequencing is available in the Data Supplement. When no slides were available for IHC or p53 IHC failed ( n = 20), the final p53 status was determined by the TP53 mutation status. In cases with failed NGS, KASPar competitive allele-specific polymerase chain reaction (LGC Genomics, Berlin, Germany) assays were used to screen for hotspot mutations in POLE (including codons 286, 297, 411, 456, and 459) as previously reported . Evaluation of IHC and sequencing results was performed blinded to each other and patient outcome.
The primary endpoint was recurrence-free survival (RFS); calculated from the date of randomisation (PORTEC-3) or date of start of adjuvant treatment (MST) to the date of the event of interest, or date of the last follow-up in patients without events. Secondary endpoints were locoregional recurrence-free survival (including vaginal and pelvic recurrences), distant metastasis-free survival (including para-aortic, abdominal and other distant recurrences), and disease-specific survival (DSS). For locoregional, distant and overall recurrence-free survival, event-free patients who died due to other causes than EC were censored. Differences between groups were tested using the χ 2 test or Fisher’s exact test for categorical variables, and with the Mann-Whitney U -test for ordinal and non-normally distributed continuous variables. Median follow-up time was estimated using the reverse Kaplan–Meier method. Survival analyses were performed according to Kaplan–Meier’s method and groups were compared with the log-rank test. Cox’ proportional hazards models were used to evaluate the prognostic value of (established) clinicopathological and molecular features in the complete study population, as well as in the molecular subgroups separately. Step-wise backward likelihood ratio-based variable selection with stratification for cohort was applied to build multivariable models. The relative importance of variables included in the multivariable models was based on the variable’s proportion of the χ 2 statistic. Model validation was performed by analysis of discrimination and indices of optimism determined by means of model fitting to 1000 bootstrap resamples. In addition, internal validation using the leave-one-out method was performed by re-estimating on the two cohorts independently. Comparison of fit between multivariable models was performed by means of Akaike’s information criterion (AIC), model concordance (C-statistic) and likelihood ratio test for comparison of nested models. A two-sided p -value <0.05 was considered statistically significant. Statistical analyses were performed with SPSS (Statistical Package of Social Science) version 25 (IBM, Armonk, NY, USA) and R (version 3.6.3., https://r-project.org ) using the survival package.
Clinicopathologic characteristics Molecular classification was successfully determined in 411 EC from PORTEC-3 and 237 EC from MST, making a total of 648 molecularly classified high-risk EC eligible for analyses (Supplementary Fig. ). There were no significant differences in patient and tumour characteristics between included and excluded patients (Supplementary Table ), except that the included patients more frequently received EBRT and had a slightly lower 5-year overall survival (71.7% vs. 77.0%, p = 0.031) compared to the excluded patients (supplementary table ). Characteristics of the included patients from PORTEC-3 and MST are shown in Table . Although MST had inclusion criteria similar to PORTEC-3, minor differences between the cohorts were observed: patients from MST predominantly received EBRT ( n = 199, 85.0%), and some had carcinosarcomas ( n = 24, 10.1%). Median follow-up time of the complete cohort was 7.0 years (95% CI 6.7–7.2). Molecular and other prognostic factors and correlation with clinical outcome Prognostic value of the molecular classification for locoregional, distant and overall RFS and CSS was evaluated (Supplementary Fig. ). For all four outcomes, POLE mut EC showed an excellent prognosis; even among the 17 patients with stage III POLE mut disease, only 1 recurrence was observed. p53abn EC showed the poorest clinical outcomes, while MMRd and NSMP EC had intermediate clinical outcomes. Kaplan–Meier analysis of RFS stratified by cohort is provided in Supplementary Fig. . Next, we evaluated the prognostic value of ER, PR, L1CAM, and CTNNB1 and established risk factors across all cases (Table ). Independent predictors for lower RFS in multivariable analysis were age at diagnosis above 60 years (HR 1.43, 95% CI 1.02–2.01), stage II (HR 1.78, 95% CI 1.15–2.75) and III disease (HR 3.47, 95% CI 2.37–5.07), and the p53abn molecular subgroup (HR 2.43, 95% CI 1.65–3.57). Adjuvant CTRT and POLE mut molecular subgroup were independent predictors for better RFS (HR 0.65, 95% CI 0.47–0.91 and HR 0.11, 95% CI 0.03–0.46, respectively). ER, PR, L1CAM and CTNNB1 were not found to be predictive of recurrence in multivariable analysis, after correction for clinicopathological risk factors and molecular subgroup. Next, we investigated molecular subgroup-specific prognostic factors (Table and Supplementary Table ). As only 1 patient with a POLE mut EC experienced a recurrence, no multivariable analysis was performed for this molecular subgroup. Among MMRd EC, both uni- and multivariable analyses showed that stage was a significant predictor for recurrence (stage I–II vs. III, HR 2.33, 95%CI 1.36–3.98, p = 0.002) (Table and Supplementary Table ). Histotype and grade did not have prognostic value within MMRd, as shown in Supplementary Fig. . ER, PR, L1CAM, and CTNNB1 were also not associated with recurrence in multivariable analysis of MMRd EC. Within the subgroup of p53abn EC, uni- and multivariable analyses showed that more advanced stage was significantly associated with recurrence (stage I–II vs. III, HR 3.66, 95% CI 2.34–5.72, p < 0.001) (Table and Supplementary Table ). Furthermore, CTRT was associated with a decreased risk of recurrence compared to RT alone (HR 0.56, 95% CI 0.33–0.93, p = 0.025). No prognostic impact of histotype and grade, and ER, PR, L1CAM, and CTNNB1 was found. Within the subgroup of NSMP EC, ER- and PR-positivity were found to be independently associated with a more favourable RFS (Table and Supplementary Table ). As ER and PR expression were significantly correlated (Spearman’s rho 0.67, p < 0.001), we investigated by Kaplan–Meier analysis of RFS whether a combination of ER and PR status was relevant for prognosis. Figure shows that women with ER-positive NSMP EC have a better RFS than those with ER-negative NSMP EC, regardless of the PR status. Of note, no ER-negative and PR-positive NSMP EC were encountered. Further exploration of the relation of ER, PR and the landscape of pathological and molecular features of NSMP EC revealed that ER negativity, rather than PR negativity, was associated with aggressive characteristics such as high-grade, non-endometrioid histology and L1CAM overexpression (Fig. ). Based on these findings, ER and not PR status was analysed by multivariable regression, which showed strong and independent prognostic impact on RFS, corrected for stage, histotype and tumour grade and adjuvant therapy (Table ). In contrast, non-endometrioid histologic subtype did not have independent prognostic value (Table ). Internal validation confirmed the prognostic effect of ER in NSMP EC (Supplementary Table ). To evaluate the chosen cut-off of 10% for ER positivity within NSMP EC we performed a Kaplan–Meier analysis for RFS by percentage of ER expression in tumour tissue, which showed that a threshold of 10% has more discriminative power than a threshold of 1% (Supplementary Fig. ). Finally, we evaluated differences in adjuvant treatment effect (CTRT vs. RT) between ER-positive and ER-negative NSMP EC (Supplementary Fig. ). Both patients with ER-positive and ER-negative NSMP EC appeared to have a small non-significant benefit of CTRT compared to RT alone. Prognostic refinement of the EC molecular classification We tested the incorporation of ER-negative NSMP and ER-positive NSMP in the molecular classification by comparing our multivariable model for RFS, including the molecular classifier with four subgroups (Table ), with the same model including the molecular classifier with NSMP divided into ER-positive and ER-negative (Supplementary Table ). This improved model fit (AIC 2173.77 vs. 2162.38, C-index 0.712 vs. 0.726, p < .001). In the multivariable model with five (molecular) subgroups, the ER-negative NSMP group was independently associated with a significantly worse RFS (HR 2.27, 95% CI 1.33–3.90, p = 0.003), while the NSMP ER-positive group was not (HR 0.69 95% CI 0.45–1.06, p = 0.09), compared to the reference group MMRd.
Molecular classification was successfully determined in 411 EC from PORTEC-3 and 237 EC from MST, making a total of 648 molecularly classified high-risk EC eligible for analyses (Supplementary Fig. ). There were no significant differences in patient and tumour characteristics between included and excluded patients (Supplementary Table ), except that the included patients more frequently received EBRT and had a slightly lower 5-year overall survival (71.7% vs. 77.0%, p = 0.031) compared to the excluded patients (supplementary table ). Characteristics of the included patients from PORTEC-3 and MST are shown in Table . Although MST had inclusion criteria similar to PORTEC-3, minor differences between the cohorts were observed: patients from MST predominantly received EBRT ( n = 199, 85.0%), and some had carcinosarcomas ( n = 24, 10.1%). Median follow-up time of the complete cohort was 7.0 years (95% CI 6.7–7.2).
Prognostic value of the molecular classification for locoregional, distant and overall RFS and CSS was evaluated (Supplementary Fig. ). For all four outcomes, POLE mut EC showed an excellent prognosis; even among the 17 patients with stage III POLE mut disease, only 1 recurrence was observed. p53abn EC showed the poorest clinical outcomes, while MMRd and NSMP EC had intermediate clinical outcomes. Kaplan–Meier analysis of RFS stratified by cohort is provided in Supplementary Fig. . Next, we evaluated the prognostic value of ER, PR, L1CAM, and CTNNB1 and established risk factors across all cases (Table ). Independent predictors for lower RFS in multivariable analysis were age at diagnosis above 60 years (HR 1.43, 95% CI 1.02–2.01), stage II (HR 1.78, 95% CI 1.15–2.75) and III disease (HR 3.47, 95% CI 2.37–5.07), and the p53abn molecular subgroup (HR 2.43, 95% CI 1.65–3.57). Adjuvant CTRT and POLE mut molecular subgroup were independent predictors for better RFS (HR 0.65, 95% CI 0.47–0.91 and HR 0.11, 95% CI 0.03–0.46, respectively). ER, PR, L1CAM and CTNNB1 were not found to be predictive of recurrence in multivariable analysis, after correction for clinicopathological risk factors and molecular subgroup. Next, we investigated molecular subgroup-specific prognostic factors (Table and Supplementary Table ). As only 1 patient with a POLE mut EC experienced a recurrence, no multivariable analysis was performed for this molecular subgroup. Among MMRd EC, both uni- and multivariable analyses showed that stage was a significant predictor for recurrence (stage I–II vs. III, HR 2.33, 95%CI 1.36–3.98, p = 0.002) (Table and Supplementary Table ). Histotype and grade did not have prognostic value within MMRd, as shown in Supplementary Fig. . ER, PR, L1CAM, and CTNNB1 were also not associated with recurrence in multivariable analysis of MMRd EC. Within the subgroup of p53abn EC, uni- and multivariable analyses showed that more advanced stage was significantly associated with recurrence (stage I–II vs. III, HR 3.66, 95% CI 2.34–5.72, p < 0.001) (Table and Supplementary Table ). Furthermore, CTRT was associated with a decreased risk of recurrence compared to RT alone (HR 0.56, 95% CI 0.33–0.93, p = 0.025). No prognostic impact of histotype and grade, and ER, PR, L1CAM, and CTNNB1 was found. Within the subgroup of NSMP EC, ER- and PR-positivity were found to be independently associated with a more favourable RFS (Table and Supplementary Table ). As ER and PR expression were significantly correlated (Spearman’s rho 0.67, p < 0.001), we investigated by Kaplan–Meier analysis of RFS whether a combination of ER and PR status was relevant for prognosis. Figure shows that women with ER-positive NSMP EC have a better RFS than those with ER-negative NSMP EC, regardless of the PR status. Of note, no ER-negative and PR-positive NSMP EC were encountered. Further exploration of the relation of ER, PR and the landscape of pathological and molecular features of NSMP EC revealed that ER negativity, rather than PR negativity, was associated with aggressive characteristics such as high-grade, non-endometrioid histology and L1CAM overexpression (Fig. ). Based on these findings, ER and not PR status was analysed by multivariable regression, which showed strong and independent prognostic impact on RFS, corrected for stage, histotype and tumour grade and adjuvant therapy (Table ). In contrast, non-endometrioid histologic subtype did not have independent prognostic value (Table ). Internal validation confirmed the prognostic effect of ER in NSMP EC (Supplementary Table ). To evaluate the chosen cut-off of 10% for ER positivity within NSMP EC we performed a Kaplan–Meier analysis for RFS by percentage of ER expression in tumour tissue, which showed that a threshold of 10% has more discriminative power than a threshold of 1% (Supplementary Fig. ). Finally, we evaluated differences in adjuvant treatment effect (CTRT vs. RT) between ER-positive and ER-negative NSMP EC (Supplementary Fig. ). Both patients with ER-positive and ER-negative NSMP EC appeared to have a small non-significant benefit of CTRT compared to RT alone.
We tested the incorporation of ER-negative NSMP and ER-positive NSMP in the molecular classification by comparing our multivariable model for RFS, including the molecular classifier with four subgroups (Table ), with the same model including the molecular classifier with NSMP divided into ER-positive and ER-negative (Supplementary Table ). This improved model fit (AIC 2173.77 vs. 2162.38, C-index 0.712 vs. 0.726, p < .001). In the multivariable model with five (molecular) subgroups, the ER-negative NSMP group was independently associated with a significantly worse RFS (HR 2.27, 95% CI 1.33–3.90, p = 0.003), while the NSMP ER-positive group was not (HR 0.69 95% CI 0.45–1.06, p = 0.09), compared to the reference group MMRd.
In this comprehensive analysis of 648 high-risk EC, we evaluated the prognostic value of ER, PR, L1CAM and CTNNB1 mutations and established clinicopathologic and molecular risk factors in one of the largest cohorts of molecularly classified high-risk endometrial cancers worldwide. Overall, no independent prognostic value of ER, PR, L1CAM and CTNNB1 was found, while the known independent impact of age, stage, the EC molecular classification and CTRT on risk of recurrence was confirmed. Within the NSMP molecular subgroup prognosis was clearly different by stage and grade, and women with ER-positive tumours had a substantially reduced risk of recurrence compared to those with ER-negative tumours. ER status, which can easily be assessed in routine diagnostics with immunohistochemistry, has the potential to refine risk stratification of women with high-risk NSMP EC. In our complete study cohort we did not find independent prognostic relevance of ER, PR, L1CAM and CTNNB1 status. Subgroup-analysis by molecular subgroup did not show prognostic relevance of PR, L1CAM and CTNNB1 status either. Importantly, ER status was an important predictor for RFS specifically in NSMP EC, but not in POLE mut, MMRd and p53abn EC. ER positivity appeared to identify a largely homogeneous group of NSMP EC with (low-grade) endometrioid histology, frequent alterations in the PI3K- and Wnt-signalling pathways and relatively favourable clinical outcomes. In contrast, the small group of ER-negative NSMP EC remained morphologically and molecularly heterogeneous, albeit all associated with more aggressive features such as non-endometrioid histology and poor clinical outcomes. Internal validation confirmed the prognostic effect of ER in NSMP EC. Distinguishing ER-positive and ER-negative NSMP EC in the molecular classification diagnostic algorithm (Fig. ) significantly improved prognostication in our cohorts of high-risk EC, with a clinically relevant difference in 5-year RFS between ER-positive and ER-negative NSMP EC (80.9% vs. 45.3% respectively, p < 0.001). The small group of ER-negative NSMP EC remains morphologically and molecularly heterogeneous, albeit with a common association of more aggressive features. A notable proportion of ER-negative NSMP tumours in our cohort were clear cell carcinomas. This rare type of endometrial cancer is generally associated with aggressive clinical behaviour, although recent studies suggest that this is molecular subgroup-dependent, with only NSMP and p53abn clear cell carcinomas having poor clinical outcomes . Currently, NSMP clear cell carcinomas are excluded from the prognostic risk groups of the European clinical guidelines due to insufficient evidence . Incorporating ER status of NSMP EC into the prognostic risk groups will decrease the number of patients that cannot be classified. Mesonephric-like carcinoma is another rare and aggressive type of EC that has only recently been recognised. These tumours are often morphologically mistaken for more common EC histotypes, such as low-grade endometrioid EC. Mesonephric-like carcinomas show intact MMR proteins and wildtype p53 expression and are thus frequently molecularly classified as NSMP EC. They are typically characterised by KRAS mutations, absence of PTEN gene alterations, chromosome 1q gains, expression of TTF-1 and/or GATA-3, and lack of ER expression . Correct identification of mesonephric-like carcinomas is crucial because of their poor clinical outcomes, including frequent metastases to the lungs, especially when compared to low-grade endometrioid EC . By performing ER IHC on all NSMP EC, pathologists can be alerted when finding negative ER staining in an otherwise apparently low-grade endometrioid EC. Whether mesonephric-like carcinomas should be categorised using the molecular classification system is an interesting topic for follow-up studies. Finally, some ER-negative NSMP tumours may have high levels of copy number alterations without p53 abnormalities. In the TCGA analyses, pathogenic TP53 mutations were present in 90% of copy number-high tumours . As p53 IHC and/or TP53 mutation analysis are used as surrogate markers for the identification of copy number-high tumours, a small proportion will be classified as NSMP EC. Previous studies showed that relatively high copy number alterations, including chromosome 1q gain/amplification, is associated with negative ER expression and adverse clinical outcomes in NSMP EC . There is currently no consensus about the IHC expression threshold to define ER positivity in EC. We used a 10% cut-off, as this is a commonly used threshold in EC. However, some studies use a 1% threshold , which is also used for selecting patients for hormonal therapy in advanced EC . Our analysis of a large cohort of high-risk NSMP EC showed that using a 10% threshold yields the best distinction in terms of prognosis. Future studies are warranted to validate this 10% cut-off for prediction of prognosis and response to hormonal therapy in EC patients. The recent incorporation of the EC molecular classification into the clinical guidelines has improved the risk stratification of EC patients . For NSMP EC patients, risk group assignment depends on stage, histotype, grade and LVSI status. Our results suggest that the addition of ER status can improve risk stratification of patients with NSMP EC. ER-negative NSMP tumours showed poor clinical outcomes, even comparable to p53abn EC, independent of other risk factors. Another study, including only high-grade endometrioid and non-endometrioid EC, reported similar poor clinical outcomes for NSMP EC . In this study, half of the NSMP EC were non-endometrioid (16% serous EC and 33% clear cell carcinomas) and plausibly ER-negative. It is, therefore, likely that all ER-negative NSMP EC have a high risk of recurrence. Our proposed prognostic stratification of NSMP into ER-positive and ER-negative NSMP EC should be evaluated in future studies that also include lower risk NSMP EC. In addition to ER status, the tumour stage, histotype and grade were independent predictors for recurrence in NSMP EC and may therefore still be relevant in the risk stratification of ER-positive NSMP EC. In POLE mut, MMRd, and p53abn endometrioid EC tumour grading was not informative. Confirmation of this finding in other cohorts may lead to a simplification in diagnosing and classifying patients in risk groups by limiting tumour grading to NSMP EEC. Remarkably, we found no significant independent prognostic value of LVSI across all cases and within the four molecular subgroups. This is probably because only presence, and not extent of LVSI was registered in PORTEC-3, and only substantial LVSI has shown to be a strong prognostic factor . ER status within NSMP EC may also be predictive for response to adjuvant treatment. In this study, we found a small non-significant benefit of CTRT in both ER-positive and ER-negative NSMP EC. Radiotherapy combined with hormonal therapy instead of chemotherapy may be an equally effective but much less toxic alternative for women with high-risk ER-positive NSMP EC. Historical trials did not show a significant benefit of adjuvant hormonal therapy . However, these trials were done in unselected cohorts, and testing of hormonal therapy specifically among ER-positive NSMP tumours might be the way forward. This will be investigated in the RAINBO NSMP-ORANGE randomised clinical trial (NCT05255653), including women with ER-positive NSMP EC. In this study, CTNNB1 exon 3 mutations were not independently associated with recurrence. Previous studies showing an association between CTNNB1 mutations and adverse clinical outcomes included more women with low- and (high-)intermediate risk EC, potentially explaining the difference in prognostic relevance . Also, L1CAM was not an independent predictor for recurrence in our study. Overexpression of L1CAM was most prevalent in the clinically unfavourable p53abn molecular subgroup and did not further delineate clinical outcomes in this group. Also within NSMP EC, overexpression of L1CAM was not an independent predictor due to its’ association with negative ER and PR expression. It has been shown that expression of L1CAM is dependent on TGF-β signalling and Wnt/β-catenin activity, which in turn are inhibited by progesterone . Although we find a strong and independent prognostic impact of ER status in NSMP EC in our study, these findings were not validated in an external validation cohort. However, internal validation using the leave-one-out method and bootstrap resampling confirmed the independent prognostic relevance of ER in NSMP EC. We have investigated the molecular landscape of NSMP EC using IHC and a large targeted NGS panel, which showed significant differences between ER-positive and ER-negative NSMP tumours. Investigation of copy number alterations in these tumours could have improved our study as it likely adds molecular and potentially prognostic information. In conclusion, the prognostic impact of the molecular classification, age, stage, and adjuvant CTRT was confirmed in a large cohort of high-risk EC. The prognostic relevance of tumour grading was limited to NSMP high-risk EC. PR and L1CAM expression and CTNNB1 mutations had no independent significant prognostic impact. ER-positivity was independently associated with a lower risk of recurrence in NSMP EC and identified a large homogeneous subgroup of NSMP tumours that are characterised by a low-grade endometrioid histotype and a relatively good prognosis. Assessment of ER status in high-risk NSMP EC is feasible in clinical practice and has the potential to improve risk stratification and treatment of patients with NSMP EC.
Data Supplement REMARK checklist
|
Molecules promoting circulating clusters of cancer cells suggest novel therapeutic targets for treatment of metastatic cancers | 8a58b217-cac8-4ed4-8217-c219512736ad | 10050392 | Internal Medicine[mh] | Introduction One of the cancer hallmarks is cancer cells dissemination and metastasis which is a leading cause of cancer associated death . Metastasis develops as a consequence of changes within cancer cells that lead to an ability to move through the tissue, survive in the circulation, attach and grow in the distal site, meanwhile escaping immune surveillance . Research of the last decades revealed that a key factor which determines the ability of cancer cells to metastasize is pathological interactions with neighboring non-cancerous cells such as fibroblasts, mesenchymal and immune cells, so called cells of tumor microenvironment (TME). Therefore, development of drugs targeting key molecules involved in the TME interactions that can suppress metastasis is a hot theme of current investigations . Cancer associated stromal cells as well as circulating exosomes migrate from the primary tumor to distal sites and change local microenvironment forming so-called pre-metastatic niche permissive for the cancer cells recruitment and growth . At the same time, cancer cells might disseminate from the primary tumor in the circulation in clusters with cancer associated cells . These clusters are thought to be relatively rare in the cancer patient population , although they have strong metastatic potential , and their presence is associated with metastasis and worse prognosis in breast , lung cancers , renal cell carcinoma , colorectal cancer , and others . Our review of current literature revealed that cells involved in the metastasis-promoting heterotypic CTC interactions include platelets, cancer associated fibroblast (CAFs), white blood cells (WBCs), specific population of tumor-associated macrophages, neutrophils and polymorphonuclear myeloid-derived suppressor cells (PMN-MDSCs). A number of investigations identified several key molecules involved in the heterotypic cell interactions such as IL1R1 , IL6, NODA, NOTCH1 , CD44 , CXCR4 , TGFBR2 , CDH1 , EPCAM , ICAM1 , CCR1 . Their expression promotes formation of CTC clusters and metastasis by inducing adhesion , proliferation , by metabolic adaptation to oxidative stress , and through the epithelial-mesenchymal transition . Quite intriguing, in lung cancer most CTCs interacting with WBCs were polyploid thus, implying repression of the mitotic checkpoint, induction of cell survival and migration . The analysis of literature and public databases revealed that expression of some genes affecting CTC clusters and metastasis predicts prognosis in many cancer types. Some of these molecules are targeted by the approved or experimental anti-cancer drugs (such as plerixafor for CXCR4 or tocilizumab for IL6R). Altogether, our review suggests the existence of common and cancer tissue specific mechanisms of CTC complex formation with implication for drug development and cancer treatment.
Tumor microenvironment promotes formation of the CTC clusters and metastasis 2.1 Interactions with white blood cells The CTCs can interact with a variety of WBCs in the circulation such as neutrophils , PMN-MDSC , platelets , macrophages , and lymphocytes . 2.1.1 Interaction with neutrophils One of the mechanisms of neutrophil mediated metastasis is formation of the neutrophil extracellular traps (NETs) consisting of neutrophil DNA . As NETs interact with and provide a niche for CTCs, blocking NET formation by DNAse, e.g. coated with nanoparticles inhibits lung metastasis . Using in vivo metastasis models, Spicer et al. have demonstrated a novel role of neutrophils in the early adhesive steps of liver metastasis in the Lewis lung carcinoma mice model . Their findings suggest that neutrophils promote cancer cell adhesion within liver sinusoids, thus influencing metastasis. The neutrophil ITGAM/ICAM-1 mediated the adhesion of lipopolysaccharide-activated neutrophils to the cancer cells . In breast cancer, CTCs interact with WBCs and in out of 70 investigated patients with invasive disease, CTCs were found in 34 (49%) patients. Among them, homotypic CTC clusters were found in 14 (20%) patients, out of which 6 (9%) also had CTC-WBC clusters and 4 (6%) had CTC-WBC clusters only . On average, about 2 CTCs were found in the CTC-WBC clusters that represented about 10% of all circulating CTCs . Most of these WBCs (75%) were myeloid cells, specifically neutrophils and T-cells. The neutrophil-CTC interactions detected in blood were associated with worse prognosis of patients . Neutrophil-CTC clusters promoted cancer cell proliferation in vitro and were characterized by higher metastatic potential in mice upon tail vein injection. Analysis of gene expression from either CTC alone or in a complex with neutrophils revealed 41 upregulated genes involved in the DNA replication and cell cycle progression. Further analysis of genes dysregulated in cancer associated neutrophils revealed that TNF-α, Oncostatin M (OSM), IL-1β and IL-6 cytokines are expressed in the neutrophils with corresponding expression of their receptors in CTCs. Reciprocal experiment detected cytokines granulocyte colony-stimulating factor (G-CSF), TGF-β3 and IL-15 in the CTCs with corresponding expression of the receptors in neutrophils. CRISPR-Cas9 mediated knockout of IL6ST and IL1R1 in cancer cells suppressed the growth advantage of the neutrophil-CTC clusters without effect on their frequency . In addition, vascular cell adhesion molecule (VCAM1) was identified in a CRISPR-Cas9 screen in the CTC as a molecule required for formation of the neutrophil-CTC clusters . Neutrophil recruitment to the primary site and metastasis was dependent on expression of CXCL1/2 in 4T1 breast cancer cells. Among molecules that block cancer cell invasion mediated by neutrophils were also NADPH oxidase, neutrophil elastase inhibitors, and DNAse . 2.1.2 Interaction with PMN-MDSCs Another type of myeloid cell - PMN-MDSCs normally function as suppressors of the immune response and have profound pro-carcinogenic properties promoting angiogenesis, formation of the pre-metastatic niche and cell proliferation , It was predicted that PMN-MDSCs interact with CTCs and it was hypothesized (yet to be proven) that PMN-MDSCs shield CTCs from the T-cell mediated destruction . At that time, CTCs were usually isolated as CD45 negative cells, thereby clusters of CTC with leukocytes (including PMN-MDSCs) were missed from the analysis. Indeed, PMN-MDSC clusters with circulating tumor cells were detected in patients with melanoma or breast cancer . It was reported that the ratio of cancer and non-cancerous cells in the clusters varied in the range 1:1 to 1:4 in six out of eight patients tested . Interestingly, a previous paper from the same group revealed that aggressive triple negative breast and melanoma cancers overexpress Nodal, an embryonic morphogen of the TGF-β family and a a putative Notch/RBPJ signaling pathway target . The patients with aggressive breast cancer had higher levels of Nodal in serum and PMN-MDSCs could promote survival of the CTCs in culture by activating reactive oxygen species (ROS) and Jagged2 response . Accordingly, CTCs promote differentiation of the PMN-MDSCs in pro-cancerous “type-2” phenotype by the Nodal signaling . Arnoletti et al. investigated the effect of interactions between the CTCs, MDSCs and T-cells extracted from the portal blood of pancreatic adenocarcinoma patients on CTC and T-cell proliferation, apoptosis and activation. It was demonstrated that MDSCs tended to cooperate with CTCs by repressing T-cells proliferation, although no significant effects on activation and anergy were reported . The mathematical modeling and direct measurements of genomic aberrations in breast cancer CTC clusters isolated by filtration revealed that the fraction of cancer cells in the clusters is in the range of 8%-48% . In contrast, isolation of multicellular clusters from the blood of breast cancer patients followed by single cell RNA-seq analysis identified genes associated specifically with clusters, in comparison to single cells, but failed to identify other cell types except platelets . In agreement with other studies, cell clusters contributed to metastasis 23 times more actively than the single cells and the presence of clusters in breast and prostate cancers was associated with significantly worse prognosis . The differences in CTC isolation protocols might lead to the differences in cell populations detected within CTC clusters. The latter study utilized HBCTC-Chip coated with cocktail of EPCAM, EGFR and HER2 antibodies , whereas Parsortix microfluidic device using Cell Separation Cassettes (GEN3D6.5, ANGLE) was used in the subsequent study that characterized neutrophils-CTC , whereas PMN-MDSC-CTCs clusters were isolated by FACS . 2.1.3 Interaction with tumor associated macrophages Interaction of CTCs with tumor associated macrophage (TAMs) seems to promote metastasis. Nanomechanical characterization of tumor associated macrophage-CTC clusters isolated from blood of prostate cancer patients revealed that contact with the macrophages softens and promotes adhesiveness of CTCs, which corresponds to mixed epithelial - mesenchymal phenotype . Notably, previous publication of the same group reported softness, deformability, and adhesiveness of single CTCs as markers of aggressive metastatic prostate cancer . The presence of TAMs in the invasive front was associated with the mesenchymal phenotype of CTCs and poor prognosis in colorectal cancer . Mechanistically, the Il-6 produced by the TAMs induced JAK2/STAT3/miR-506-3p/FoxQ1 signaling in cancer cells, thus promoting epithelial mesenchymal transition (EMT), metastasis and further attraction of macrophages by secretion of CCl2 . 2.1.4 Interaction with lymphocytes We found only one report that mentions interaction of CTCs with lymphocytes . However, CTCs are associated with impairments of adaptive immunity. The quantity of CTCs correlates with the presence in peripheral blood of the CD95(FAS)-positive T-helper cells and stage 3 breast cancer as well as with lower percentage of the CD8+ T-cells with activated T-cell receptor , the absence of tumor associated antigen specific TCRs and low TCR heterogeneity , and positively associated with intratumoral populations of T-regs . 2.2 Interactions with cancer associated fibroblasts Aside from single CTCs and cancer associated fibroblasts (CAFs), the presence of homotypic and heterotypic clusters of CTCs and CAFs was reported in patients with stages 1-4 of breast cancer . In their study, Sharma et al. detected CTCs in 90% and circulating CAFs (cCAFs) in 80% of patients; homotypic CTC clusters were found in 50% and heterotypic - in 25% of patients in treatment naive stages 2-3. Interestingly, only 25% of patients in stage 4 had homotypic clusters and 25% had heterotypic CTC-CAF clusters. The number of cCAFs and CTCs was much higher in patient blood with metastatic breast cancer in comparison to localized cancers whereas nothing was detected in the control group. The effect of cancer treatment on these clusters was not yet addressed . Using MDA-MB-231 cells and CD44-enriched MCF7 cells, authors have been able to demonstrate involvement of the stem cell marker CD44 in the heterotypic clustering and that heterotypic clusters metastasize more efficiently . Accordingly, it was shown that tumor suppressor Rb represses CD44 dependent collective invasion, release of breast cancer cells in circulation and lung metastasis . Circulating CAFs and CTCs were also detected in small groups of colorectal and prostate cancer patients . Consistent with others, the paper shows images of the distinct multicellular CTC clusters with CAF and with leukocytes, which were obtained by the negative filtration through 10 µm filter . 2.3 Interaction with platelets Activation of the coagulation cascade and formation of platelet-rich thrombus around tumor cells in the vasculature have both been proposed to play major roles in physically shielding CTCs from the stress of blood flow and from lysis by the Natural killer cells . One of the mechanisms is substitution of cancer cell MHC1 by platelets-derived MHC1 carrying normal peptides thereby protecting cancer cells from both NK and T-cell recognition . Analysis of the single cell gene expression of the CTCs in the pancreatic cancer mouse model revealed that 32% of the circulating cells interact with platelets leading to suppression of epithelial markers and expression changes of many other genes . Accordingly, direct interaction with platelets promotes EMT in cancer cells and either inhibition of NF-kB in cancer cells or inhibition of TGF-β in platelets was sufficient to protect against lung metastasis . In turn, disruption of platelets interactions with cancer cell by S-nitrosocaptopril (CapNO) inhibits adhesion to endothelial cells and lung cancer metastasis in immunocompetent mouse models through multiple mechanisms including reduction of Sialyl-Lewis X (Slex) levels in cancer cells and ADP-induced P-selectin in platelets, IL-1b induced VCAM1, ICAM-1, and E-selectin by HUVECs .
Interactions with white blood cells The CTCs can interact with a variety of WBCs in the circulation such as neutrophils , PMN-MDSC , platelets , macrophages , and lymphocytes . 2.1.1 Interaction with neutrophils One of the mechanisms of neutrophil mediated metastasis is formation of the neutrophil extracellular traps (NETs) consisting of neutrophil DNA . As NETs interact with and provide a niche for CTCs, blocking NET formation by DNAse, e.g. coated with nanoparticles inhibits lung metastasis . Using in vivo metastasis models, Spicer et al. have demonstrated a novel role of neutrophils in the early adhesive steps of liver metastasis in the Lewis lung carcinoma mice model . Their findings suggest that neutrophils promote cancer cell adhesion within liver sinusoids, thus influencing metastasis. The neutrophil ITGAM/ICAM-1 mediated the adhesion of lipopolysaccharide-activated neutrophils to the cancer cells . In breast cancer, CTCs interact with WBCs and in out of 70 investigated patients with invasive disease, CTCs were found in 34 (49%) patients. Among them, homotypic CTC clusters were found in 14 (20%) patients, out of which 6 (9%) also had CTC-WBC clusters and 4 (6%) had CTC-WBC clusters only . On average, about 2 CTCs were found in the CTC-WBC clusters that represented about 10% of all circulating CTCs . Most of these WBCs (75%) were myeloid cells, specifically neutrophils and T-cells. The neutrophil-CTC interactions detected in blood were associated with worse prognosis of patients . Neutrophil-CTC clusters promoted cancer cell proliferation in vitro and were characterized by higher metastatic potential in mice upon tail vein injection. Analysis of gene expression from either CTC alone or in a complex with neutrophils revealed 41 upregulated genes involved in the DNA replication and cell cycle progression. Further analysis of genes dysregulated in cancer associated neutrophils revealed that TNF-α, Oncostatin M (OSM), IL-1β and IL-6 cytokines are expressed in the neutrophils with corresponding expression of their receptors in CTCs. Reciprocal experiment detected cytokines granulocyte colony-stimulating factor (G-CSF), TGF-β3 and IL-15 in the CTCs with corresponding expression of the receptors in neutrophils. CRISPR-Cas9 mediated knockout of IL6ST and IL1R1 in cancer cells suppressed the growth advantage of the neutrophil-CTC clusters without effect on their frequency . In addition, vascular cell adhesion molecule (VCAM1) was identified in a CRISPR-Cas9 screen in the CTC as a molecule required for formation of the neutrophil-CTC clusters . Neutrophil recruitment to the primary site and metastasis was dependent on expression of CXCL1/2 in 4T1 breast cancer cells. Among molecules that block cancer cell invasion mediated by neutrophils were also NADPH oxidase, neutrophil elastase inhibitors, and DNAse . 2.1.2 Interaction with PMN-MDSCs Another type of myeloid cell - PMN-MDSCs normally function as suppressors of the immune response and have profound pro-carcinogenic properties promoting angiogenesis, formation of the pre-metastatic niche and cell proliferation , It was predicted that PMN-MDSCs interact with CTCs and it was hypothesized (yet to be proven) that PMN-MDSCs shield CTCs from the T-cell mediated destruction . At that time, CTCs were usually isolated as CD45 negative cells, thereby clusters of CTC with leukocytes (including PMN-MDSCs) were missed from the analysis. Indeed, PMN-MDSC clusters with circulating tumor cells were detected in patients with melanoma or breast cancer . It was reported that the ratio of cancer and non-cancerous cells in the clusters varied in the range 1:1 to 1:4 in six out of eight patients tested . Interestingly, a previous paper from the same group revealed that aggressive triple negative breast and melanoma cancers overexpress Nodal, an embryonic morphogen of the TGF-β family and a a putative Notch/RBPJ signaling pathway target . The patients with aggressive breast cancer had higher levels of Nodal in serum and PMN-MDSCs could promote survival of the CTCs in culture by activating reactive oxygen species (ROS) and Jagged2 response . Accordingly, CTCs promote differentiation of the PMN-MDSCs in pro-cancerous “type-2” phenotype by the Nodal signaling . Arnoletti et al. investigated the effect of interactions between the CTCs, MDSCs and T-cells extracted from the portal blood of pancreatic adenocarcinoma patients on CTC and T-cell proliferation, apoptosis and activation. It was demonstrated that MDSCs tended to cooperate with CTCs by repressing T-cells proliferation, although no significant effects on activation and anergy were reported . The mathematical modeling and direct measurements of genomic aberrations in breast cancer CTC clusters isolated by filtration revealed that the fraction of cancer cells in the clusters is in the range of 8%-48% . In contrast, isolation of multicellular clusters from the blood of breast cancer patients followed by single cell RNA-seq analysis identified genes associated specifically with clusters, in comparison to single cells, but failed to identify other cell types except platelets . In agreement with other studies, cell clusters contributed to metastasis 23 times more actively than the single cells and the presence of clusters in breast and prostate cancers was associated with significantly worse prognosis . The differences in CTC isolation protocols might lead to the differences in cell populations detected within CTC clusters. The latter study utilized HBCTC-Chip coated with cocktail of EPCAM, EGFR and HER2 antibodies , whereas Parsortix microfluidic device using Cell Separation Cassettes (GEN3D6.5, ANGLE) was used in the subsequent study that characterized neutrophils-CTC , whereas PMN-MDSC-CTCs clusters were isolated by FACS . 2.1.3 Interaction with tumor associated macrophages Interaction of CTCs with tumor associated macrophage (TAMs) seems to promote metastasis. Nanomechanical characterization of tumor associated macrophage-CTC clusters isolated from blood of prostate cancer patients revealed that contact with the macrophages softens and promotes adhesiveness of CTCs, which corresponds to mixed epithelial - mesenchymal phenotype . Notably, previous publication of the same group reported softness, deformability, and adhesiveness of single CTCs as markers of aggressive metastatic prostate cancer . The presence of TAMs in the invasive front was associated with the mesenchymal phenotype of CTCs and poor prognosis in colorectal cancer . Mechanistically, the Il-6 produced by the TAMs induced JAK2/STAT3/miR-506-3p/FoxQ1 signaling in cancer cells, thus promoting epithelial mesenchymal transition (EMT), metastasis and further attraction of macrophages by secretion of CCl2 . 2.1.4 Interaction with lymphocytes We found only one report that mentions interaction of CTCs with lymphocytes . However, CTCs are associated with impairments of adaptive immunity. The quantity of CTCs correlates with the presence in peripheral blood of the CD95(FAS)-positive T-helper cells and stage 3 breast cancer as well as with lower percentage of the CD8+ T-cells with activated T-cell receptor , the absence of tumor associated antigen specific TCRs and low TCR heterogeneity , and positively associated with intratumoral populations of T-regs .
Interaction with neutrophils One of the mechanisms of neutrophil mediated metastasis is formation of the neutrophil extracellular traps (NETs) consisting of neutrophil DNA . As NETs interact with and provide a niche for CTCs, blocking NET formation by DNAse, e.g. coated with nanoparticles inhibits lung metastasis . Using in vivo metastasis models, Spicer et al. have demonstrated a novel role of neutrophils in the early adhesive steps of liver metastasis in the Lewis lung carcinoma mice model . Their findings suggest that neutrophils promote cancer cell adhesion within liver sinusoids, thus influencing metastasis. The neutrophil ITGAM/ICAM-1 mediated the adhesion of lipopolysaccharide-activated neutrophils to the cancer cells . In breast cancer, CTCs interact with WBCs and in out of 70 investigated patients with invasive disease, CTCs were found in 34 (49%) patients. Among them, homotypic CTC clusters were found in 14 (20%) patients, out of which 6 (9%) also had CTC-WBC clusters and 4 (6%) had CTC-WBC clusters only . On average, about 2 CTCs were found in the CTC-WBC clusters that represented about 10% of all circulating CTCs . Most of these WBCs (75%) were myeloid cells, specifically neutrophils and T-cells. The neutrophil-CTC interactions detected in blood were associated with worse prognosis of patients . Neutrophil-CTC clusters promoted cancer cell proliferation in vitro and were characterized by higher metastatic potential in mice upon tail vein injection. Analysis of gene expression from either CTC alone or in a complex with neutrophils revealed 41 upregulated genes involved in the DNA replication and cell cycle progression. Further analysis of genes dysregulated in cancer associated neutrophils revealed that TNF-α, Oncostatin M (OSM), IL-1β and IL-6 cytokines are expressed in the neutrophils with corresponding expression of their receptors in CTCs. Reciprocal experiment detected cytokines granulocyte colony-stimulating factor (G-CSF), TGF-β3 and IL-15 in the CTCs with corresponding expression of the receptors in neutrophils. CRISPR-Cas9 mediated knockout of IL6ST and IL1R1 in cancer cells suppressed the growth advantage of the neutrophil-CTC clusters without effect on their frequency . In addition, vascular cell adhesion molecule (VCAM1) was identified in a CRISPR-Cas9 screen in the CTC as a molecule required for formation of the neutrophil-CTC clusters . Neutrophil recruitment to the primary site and metastasis was dependent on expression of CXCL1/2 in 4T1 breast cancer cells. Among molecules that block cancer cell invasion mediated by neutrophils were also NADPH oxidase, neutrophil elastase inhibitors, and DNAse .
Interaction with PMN-MDSCs Another type of myeloid cell - PMN-MDSCs normally function as suppressors of the immune response and have profound pro-carcinogenic properties promoting angiogenesis, formation of the pre-metastatic niche and cell proliferation , It was predicted that PMN-MDSCs interact with CTCs and it was hypothesized (yet to be proven) that PMN-MDSCs shield CTCs from the T-cell mediated destruction . At that time, CTCs were usually isolated as CD45 negative cells, thereby clusters of CTC with leukocytes (including PMN-MDSCs) were missed from the analysis. Indeed, PMN-MDSC clusters with circulating tumor cells were detected in patients with melanoma or breast cancer . It was reported that the ratio of cancer and non-cancerous cells in the clusters varied in the range 1:1 to 1:4 in six out of eight patients tested . Interestingly, a previous paper from the same group revealed that aggressive triple negative breast and melanoma cancers overexpress Nodal, an embryonic morphogen of the TGF-β family and a a putative Notch/RBPJ signaling pathway target . The patients with aggressive breast cancer had higher levels of Nodal in serum and PMN-MDSCs could promote survival of the CTCs in culture by activating reactive oxygen species (ROS) and Jagged2 response . Accordingly, CTCs promote differentiation of the PMN-MDSCs in pro-cancerous “type-2” phenotype by the Nodal signaling . Arnoletti et al. investigated the effect of interactions between the CTCs, MDSCs and T-cells extracted from the portal blood of pancreatic adenocarcinoma patients on CTC and T-cell proliferation, apoptosis and activation. It was demonstrated that MDSCs tended to cooperate with CTCs by repressing T-cells proliferation, although no significant effects on activation and anergy were reported . The mathematical modeling and direct measurements of genomic aberrations in breast cancer CTC clusters isolated by filtration revealed that the fraction of cancer cells in the clusters is in the range of 8%-48% . In contrast, isolation of multicellular clusters from the blood of breast cancer patients followed by single cell RNA-seq analysis identified genes associated specifically with clusters, in comparison to single cells, but failed to identify other cell types except platelets . In agreement with other studies, cell clusters contributed to metastasis 23 times more actively than the single cells and the presence of clusters in breast and prostate cancers was associated with significantly worse prognosis . The differences in CTC isolation protocols might lead to the differences in cell populations detected within CTC clusters. The latter study utilized HBCTC-Chip coated with cocktail of EPCAM, EGFR and HER2 antibodies , whereas Parsortix microfluidic device using Cell Separation Cassettes (GEN3D6.5, ANGLE) was used in the subsequent study that characterized neutrophils-CTC , whereas PMN-MDSC-CTCs clusters were isolated by FACS .
Interaction with tumor associated macrophages Interaction of CTCs with tumor associated macrophage (TAMs) seems to promote metastasis. Nanomechanical characterization of tumor associated macrophage-CTC clusters isolated from blood of prostate cancer patients revealed that contact with the macrophages softens and promotes adhesiveness of CTCs, which corresponds to mixed epithelial - mesenchymal phenotype . Notably, previous publication of the same group reported softness, deformability, and adhesiveness of single CTCs as markers of aggressive metastatic prostate cancer . The presence of TAMs in the invasive front was associated with the mesenchymal phenotype of CTCs and poor prognosis in colorectal cancer . Mechanistically, the Il-6 produced by the TAMs induced JAK2/STAT3/miR-506-3p/FoxQ1 signaling in cancer cells, thus promoting epithelial mesenchymal transition (EMT), metastasis and further attraction of macrophages by secretion of CCl2 .
Interaction with lymphocytes We found only one report that mentions interaction of CTCs with lymphocytes . However, CTCs are associated with impairments of adaptive immunity. The quantity of CTCs correlates with the presence in peripheral blood of the CD95(FAS)-positive T-helper cells and stage 3 breast cancer as well as with lower percentage of the CD8+ T-cells with activated T-cell receptor , the absence of tumor associated antigen specific TCRs and low TCR heterogeneity , and positively associated with intratumoral populations of T-regs .
Interactions with cancer associated fibroblasts Aside from single CTCs and cancer associated fibroblasts (CAFs), the presence of homotypic and heterotypic clusters of CTCs and CAFs was reported in patients with stages 1-4 of breast cancer . In their study, Sharma et al. detected CTCs in 90% and circulating CAFs (cCAFs) in 80% of patients; homotypic CTC clusters were found in 50% and heterotypic - in 25% of patients in treatment naive stages 2-3. Interestingly, only 25% of patients in stage 4 had homotypic clusters and 25% had heterotypic CTC-CAF clusters. The number of cCAFs and CTCs was much higher in patient blood with metastatic breast cancer in comparison to localized cancers whereas nothing was detected in the control group. The effect of cancer treatment on these clusters was not yet addressed . Using MDA-MB-231 cells and CD44-enriched MCF7 cells, authors have been able to demonstrate involvement of the stem cell marker CD44 in the heterotypic clustering and that heterotypic clusters metastasize more efficiently . Accordingly, it was shown that tumor suppressor Rb represses CD44 dependent collective invasion, release of breast cancer cells in circulation and lung metastasis . Circulating CAFs and CTCs were also detected in small groups of colorectal and prostate cancer patients . Consistent with others, the paper shows images of the distinct multicellular CTC clusters with CAF and with leukocytes, which were obtained by the negative filtration through 10 µm filter .
Interaction with platelets Activation of the coagulation cascade and formation of platelet-rich thrombus around tumor cells in the vasculature have both been proposed to play major roles in physically shielding CTCs from the stress of blood flow and from lysis by the Natural killer cells . One of the mechanisms is substitution of cancer cell MHC1 by platelets-derived MHC1 carrying normal peptides thereby protecting cancer cells from both NK and T-cell recognition . Analysis of the single cell gene expression of the CTCs in the pancreatic cancer mouse model revealed that 32% of the circulating cells interact with platelets leading to suppression of epithelial markers and expression changes of many other genes . Accordingly, direct interaction with platelets promotes EMT in cancer cells and either inhibition of NF-kB in cancer cells or inhibition of TGF-β in platelets was sufficient to protect against lung metastasis . In turn, disruption of platelets interactions with cancer cell by S-nitrosocaptopril (CapNO) inhibits adhesion to endothelial cells and lung cancer metastasis in immunocompetent mouse models through multiple mechanisms including reduction of Sialyl-Lewis X (Slex) levels in cancer cells and ADP-induced P-selectin in platelets, IL-1b induced VCAM1, ICAM-1, and E-selectin by HUVECs .
Polyploidy and epithelial-to-mesenchymal transition in CTC clusters As it is discussed in the previous sections, interaction with TAM or platelets induced metastasis promoting EMT in cancer cells . EMT is associated with cancer progression and metastasis . During EMT epithelial cells lose contact with epithelial or endothelial cells, change their cytoskeleton and consequentially, become less rigid, acquiring an ability to move . In addition, EMT induces stem cell properties in cancer, regulates and is regulated by immunosuppressive cancer microenvironment . Notably, cancer stem cells are characterized by mixed epithelial – mesenchymal phenotype . Interestingly, interaction with white blood cells also correlates with mixed Epithelial-mesenchymal phenotype and cancer cells polyploidy that play a key role in cancer resistance to treatment and metastasis . The presence of CTC-WBC clusters was associated with worse prognosis in lung , breast cancers , and hepatocellular carcinoma . Remarkably, in lung cancer, CTCs in complex with WBCs were exclusively polyploid . In turn, in glioblastoma, examination of ploidy together with expression of endothelial marker CD31 revealed that pre-operative small triploid CD31 negative CTCs were predictive of inferior prognosis . A recent paper employed the iFISH method combining FISH DNA staining and immunofluorescence to create Atlas of Circulating Rare Cells . High throughput imaging analysis of circulating rare cells (CRCs) purified by WBC subtraction categorized cells into 71 subtypes based on the CD45 leukocyte staining, cell size, chromosome 8 ploidy and the presence of a few tumor cell markers including PD-L1 (EPCAM/CK18/PD-L1/AFP/HER2/CA19-9), endothelial CD31, mesenchymal Vimentin and stem cell CD133 markers . Authors presented a set of cell images with polyploid chromosome 8. There were cells double positive for CD31 and Vimentin staining with abnormal chromosomes which can coincide with cytokeratin CK18, and even CD45-/EPCAM+/CD31+/Vim+ “aneuploid mesenchymal epithelial-endothelial fusion clusters” were detected. These observations are consistent with the previous data generated by iFISH linking polyploidy with EMT . The presence of CD45 positive cells was detected in the clusters with polyploid or multinuclear cancer cells . Quite importantly, comparison of the total count of CTCs and/or circulating tumor endothelial cells between 31 conditions revealed that CTCs are present in multiple cancers, however, the highest frequency of “CTCs” is observed within the group of non-neoplastic infectious diseases, suggesting that the pure presence of cells with these markers could not be used as a diagnostic test itself . Consistent with the Atlas of Circulating Rare Cells , sequencing of CTC clusters and individual circulating cancer cells revealed the mixed epithelial-mesenchymal markers in hepatocellular carcinoma (Vimentin, epithelial: CDH1, EPCAM, ASGR2, Keratin 8, stemness: CD133, POU5F1, NOTCH1 and STAT3) prostate cancer (EPCAM, keratins, E-cad, Vimentin, CD44) and Vimentin in lung cancer . However, two major conceptual questions here currently remain not sufficiently addressed: (i) How heterotypic interactions of cancer cells with WBCs promote polyploidy? (ii) How does the combination of ploidy and mesenchymal phenotype enhance metastasis? Mechanisms of how heterotypic interactions promote mobility and mesenchymal phenotype are described in the subsequent section.
Heterotypic interactions within tumor microenvironment are pivotal for CTC cluster formation Interactions with cells of cancer microenvironment promote EMT, formation of CTC clusters and metastasis . Classically, EMT is accompanied by decrease of E-cadherin/N-cadherin ratio . A recent publication highlighted a novel role of the E-cadherin (E-cad, encoded by CDH1 gene) expressing cells in breast cancer metastasis . It turned out that when cancer cells grow in the presence of CAFs there is a gradient of the E-cad from low at the trailing edge of the invading cancer cells to high E-cad behind it . Furthermore, another paper demonstrated that in breast cancer spheroid model stem cells lead the collective invasion co-expressing mesenchymal and epithelial marks . Dermal implants of CAFs with MCF10 cells with low intrinsic metastatic potential promoted this low-high E-cad gradient, the CTC cell clustering and metastasis . High throughput RNA expression profiles revealed induction of carcinoembryonic antigen-related cell adhesion molecule 5 (CEACAM5; CAM5) and CEACAM6 (CAM6) in the presence of CAFs. This experiment revealed overexpression of 44 CAF-induced genes, whose expression is associated with poor prognosis in breast cancer. Mechanistically, E-cad, CAM5 and CAM6 interact with each other forming an adherent junction complex on the cell surface. Functional shRNA studies revealed attenuation of lung metastasis upon E-cad, CAM5, or CAM6 depletion. Other excellent functional investigations reported in this paper revealed that CAF produced SDF-1(encoded by CXCL12 gene) and TGF-β that through their cognate receptors CXCR4 and TGFBRII activate SRC kinase phosphorylation/Zeb1 axis altogether mediating tumor cell cluster formation that are also detected as CTC clusters. The caveat of this report for our purposes is that we do not know if fibroblasts travel in the bloodstream with cancer cells. However, this paper clearly demonstrates stromal-cancer cell molecular interactions that regulate the ability of cancer to metastasize . Importantly, CRCX4 mediates immunosuppressive tumor microenvironment not only in cancer cells, but also in the SMA positive stromal cells including myofibroblasts and pericytes . CRE-Lox mediated knockout of CRCX4 in SMA expressing cells improved survival in mice with breast cancer, and pharmacological inhibition of CRCX4 potentiated activity of immune checkpoint inhibitors in the nude mice bearing human metastatic breast cancer . Similarly, to observation in breast cancer, cells of the collective invasion packs were E-cad positive in lung adenocarcinoma . The role of CAFs in the metastasis was demonstrated by the fact that only surrounding CAFs express vimentin and in the vimentin knockout mice, the CAFs motility decreases in vitro and in vivo . Vimentin was required for the heterotypic cancer cell - CAFs interaction, collective invasion, and lung adenocarcinoma metastasis . Thus, formation of Epithelial-mesenchymal gradient during collective invasion is mediated by cancer cell – stromal cell interaction and pivotal for CTC formation and metastasis . We schematized major findings on CTC interactions and their molecular physiological effects on , .
Expression of molecules involved in the CTC cluster formation and metastasis correlate with cancer survival As it is discussed in the previous sections, the formation of the CTC clusters and metastasis in particular cancers depend on IL1R1 , IL6, NODAL, NOTCH1 , CD44 , CXCR4 , TGFBR2 , CDH1 , EPCAM , ICAM1 , and CCR1 . Theoretically, these molecules can impact cancer metastasis with little to no information on the mechanisms involved in CTC cluster formation. To address this possibility, we interrogated a publicly available The Cancer Genome Atlas project (TCGA) database and research papers to examine if high or low expression of molecules that are functionally important for the formation of CTC clusters may characterize patient survival in multiple cancers. For example, it was demonstrated that IL1R1 protein induces CTC proliferation in breast cancer (BRCA) , and high IL1R gene expression corresponds to inferior prognosis in the TCGA-BRCA cohort as well as in many other cancers . In turn, high expression of CXCR4 in BRCA corresponds to better prognosis, smaller yet significant difference between Kaplan-Meier curves predicting better prognosis was observed for lung adenocarcinoma (LUAD) and thyroid cancer (THCA) , whereas no difference was observed in lung squamous cell carcinoma (LUSC). The clustering analysis separated CTC marker genes into two major groups: ( i ) CD44, CXCR4, ICAM1, CCR1 , and IL1R1 where high expression correlated with poor survival for low grade gliomas (LGG), kidney renal clear cell carcinoma (KIRC), for glioblastoma (GBM), kidney renal papillary cell carcinoma (KIRP) or lung squamous cell carcinoma (LUSC) . The second gene cluster ( ii ) includes TGFBR2, IL6ST, IL6R, CDH1 , and IGFBP5 . In this group we observed a correlation between high expression and better prognosis in KIRC and worse prognosis in KIRP and LGG. As discussed in the previous sections, molecules included in the analysis promote CTC cluster formation or metastasis in functional studies. Indeed, the results of clustering analysis suggest that high expression of genes from the first cluster predicts a rapid disease progression in multiple cancers. Conversely, the second cluster contains more genes whose expression promotes cancer progression in a cancer specific manner. However, in some cases focused investigations contradict prediction of patient survival based on the TCGA dataset ( , upper right triangles in depict approximates for HR collected from the literature). Specifically, high expression of stem cell marker CD44 corresponded to poor prognosis in kidney cancers (KIRC, KIRP) in TCGA data and, accordingly, high CD44 and b-catenin immunostaining correlated with advanced stage, although no significant correlation with survival could be observed in a specific focused study . However, other reports communicated a correlation between high CD44 levels and decrease of progression free survival in renal cell carcinoma after treatment with multi-targeted tyrosine kinase inhibitor . Consistent with the literature, high CD44 expression predicts inferior prognosis in LGG and GBM TCGA cohorts . The only case of association between CD44 expression and positive thyroid cancer prognosis contradicts to the literature . It was reported that high CXCR4 expression corresponds to bad prognosis for breast , lung and colorectal cancers contradicting TCGA-based findings . It was recently reported that in advanced CD8 negative thyroid cancer, high expression of CXCR4 and its ligand CXCL12 (SDF-1) correlates with bad prognosis, thus contradicting to TCGA data . In contrast, ICAM1 expression is associated with favorable prognosis in the breast cancer TCGA cohort, consistent with similar survival analysis of NCBI GEO dataset and repression of the lung metastasis in spontaneous breast cancer metastasis model and contradicting another paper reporting pivotal role of the ICAM1 in the CTC cluster formation, trans-endothelial migration and metastasis in breast cancer . Thus, the positive associations between expression of CD44, CXCR4 and ICAM1 for thyroid, lung and breast cancers in TCGA dataset are not consistent with the literature suggesting that the first cluster is indeed represents genes whose high expression correlates with inferior prognosis consistent with their role in the CTCs biology. Further we compared TCGA prediction with the literature for a few genes from the second cluster to address the question if they have more tissue specific roles in cancer metastasis. One of such genes is CDH1 (E-cad protein) whose high expression was a predictor of better prognosis for colorectal cancer in agreement with TCGA data . Again, consistent with TCGA data, high protein staining of E-cad in kidney renal papillary cell carcinoma was associated with worse prognosis, and no association was detected for kidney renal clear cell carcinoma . Recent analysis of E-cad in the cohort of NSCLC with 66% cases representing squamous cell carcinoma identifies E-cad as a positive prognostic factor consistent with TCGA data . When astrocytomas, oligodendrogliomas and oligoastrocytomas were analyzed, the loss of E-cad immunostaining and hypermethylation of its promoter were associated with worse prognosis contradicting TCGA data, although, gene expression analysis was not performed . In contrast, consistent with TCGA data, a positive association between higher E-cad expression and worse prognosis was reported in the low-grade gliomas and in glioblastoma . IL6 receptors IL6ST and IL6R are involved in the CTC heterotypic interactions in breast and colorectal cancers . Consistent with the literature, expression of IL6R has strong prognostic value in glioblastoma and in lung adenocarcinoma . In contrast, in kidney clear cell carcinoma we found a contradiction between the literature and TCGA data concerning the biomarker potential of IL6R expression: good predictor according to the literature , and poor predictor according to TCGA data. Thus, the role of IL6R expression in cancer can be considered tissue specific. The member of TGFb receptor family - TGFBR2 is a tumor suppressor in lung cancer, and the loss of TGFBR2 expression is associated with worse prognosis of both squamous cell cancer and adenocarcinoma . Accordingly, TGFBR2 mutation predicts lung cancer resistance to checkpoint inhibitors . Thus, the literature supports prediction of TCGA dataset regarding the role of TGFBR2 in LUAD progression and contradicts association of high TGFBR2 with negative prognosis in LUSC. In breast cancer, reduced expression of TGFBR2 is associated with worse prognosis contradicting the TCGA data especially in ER positive patients , while the report by Gao and coauthors is in line with the TCGA data . Theoretically, these contradictions might be connected with the presence of TGFBR2 mutations which were not investigated in these published reports. Little is known about the influence of the TGFBR2 on glioma survival, however TCGA prediction of the negative association might be valuable since TGBFR2 compensates for inhibition of PDGFR, thereby promoting survival . NOTCH1 activation as measured by the immunostaining against NOTCH intracellular domain correlates with poor prognosis of kidney renal clear cell carcinoma (KIRC) . In turn, high total NOTCH1 immunostaining is associated with progression of kidney renal clear cell carcinoma contradicting TCGA prediction . Likewise, in contrast to TCGA data, literature suggests association of NOTCH1 expression and glioma progression by modulating CXCL12/CXCR4 . In contrast, a recent meta-analysis revealed that NOTCH1 expression does not correlate with overall survival in adenocarcinoma, although DLL4 and HES1 were associated with worse prognosis . Measurements of VCAM1 in KIRC revealed association of high expression with good prognosis consistent with TCGA data analysis . In turn, for KIRP we found no published data that can validate the association of VCAM1 high expression with good prognosis observed for the TCGA dataset. EPCAM expression was associated with favorable prognosis of breast cancer in TCGA data, however immunohistochemical analysis has shown that it is associated with worse prognosis specifically in the basal-like and luminal B HER2+ subtypes . However, in the HER2+ subtype, EPCAM was also reported to be associated with worse prognosis . Again, in LGG the protein level of EPCAM was associated with poor prognosis, which contradicts to the TCGA trends . In thyroid cancers, the presence of EPCAM cleavage product was associated with more aggressive disease progression, although gene expression was not measured in this report . Finally, in agreement with the TCGA dataset, high EPCAM expression was associated with better prognosis in kidney cancers . Overall, after comparison of TCGA data with the literature, it is possible to conclude that genes of the first cluster (top, ) are mostly predictors of poor prognosis, whereas genes of the second cluster (bottom, ) predict survival in a cancer type-specific manner . For interrogation of TCGA expression and survival data, we used standard analytic tools from the TCGA project portal GDC . The discrepancies between results of TCGA data analysis and the literature could originate from different experimental methods used to assess gene expression, or different cohorts of patients and different treatment regimens among others. Thus, results of positive or negative gene association with patient survival require independent verification to identify or to confirm reliable biomarkers of disease progression and potential targets for drug development.
Molecules involved in the CTC heterotypic interaction and known drug targets Analysis of the TCGA data and the literature revealed that high expression of molecules involved in the CTC heterotypic interactions predicts survival in many cancer types. Accordingly, as it is discussed in the previous sections, these molecules are pivotal for metastasis and therefore sometimes represent targets of clinically approved or experimental cancer drugs. Specifically, results of TCGA dataset analysis suggest poor prognosis for IL6 overexpressing low grade gliomas and glioblastomas. Indeed, pre-clinical data demonstrated that IL6 blockade combined with CD40 stimulation sensitized glioblastoma to immune checkpoint inhibitors and improved survival . Likewise, pre-clinical investigations revealed that targeting of the IL6 signaling might be beneficial for other cancers as well, where bad prognosis is associated with high IL6 level such as renal cell carcinoma , non-small cell lung cancer , and breast cancer . We found a single, currently suspended clinical trial of the IL6R antibody tocilizumab for gliomas and glioblastoma treatment (NCT04729959), trials for metastatic breast cancer (NCT03135171), non-small lung cancer among others (NCT04940299, ). Targeting of IL-6 improves immunotherapy outcome in mice models . However, IL6-specific antibody siltuximab demonstrated no efficiency against renal cell carcinoma and prostate cancer . IL1R1 expression predicts poor survival in nearly the same set of cancer types as IL6R. There are multiple clinical trials testing IL1R agonist an anti-rheumatoid arthritis drug anakinra against multiple myeloma , metastatic breast cancer, and colorectal cancer , listed in . However, we didn’t find any specific records for gliomas, lung or kidney cancers. Still, several preclinical investigations have shown that targeting of IL1 signaling in GBM , LGG kidney and lung cancer suggest its potential clinical usefulness. Expression of adhesion molecule ICAM1 also predicts poor prognosis for several cancer types, closely mimicking the effects observed for the IL1R1 and IL6R genes. Specifically, low ICAM1 expression corresponds to better survival in GBM. Indeed, bispecific CAR-T cells against EPCAM and ICAM1 elicited good response in GBM mice model , consistent with other preclinical studies . Similarly, CAR-T cells targeted against ICAM1 were successfully tested in mice models of gastric , thyroid , and triple negative breast cancer . ICAM-1 conjugated with a cytotoxic drug was extensively tested for multiple myeloma and another bispecific anti-CD38-ICAM-1 drug for multiple myeloma is under development . The vaccine targeting ICAM-1 is also at the early stage of clinical investigation against ICAM-1 overexpressing bladder cancers or lung cancer (NCT02043665). However, so far, we did not find reports on ICAM-1 targeted therapies clinically tested against gliomas and kidney cancers. A stem cell marker CD44 predicts poor prognosis in renal cancers and in gliomas. The CD44-specific antibody RG7356 in clinical trials showed moderate efficiency in solid tumors and in acute myeloid leukemia . There is also multiple evidence suggesting potential efficiency of CD44 targeting for the treatment of GBM, although additional clinical validation is clearly needed . Catumaxomab (genetically engineered bivalent anti-EPCAM and anti-CD3 antibody) is approved for the treatment of malignant ascites and it has been also used experimentally for the treatment of bladder and ovarian cancers. Bispecific CAR-T simultaneously targeting EPCAM and ICAM-1 demonstrated promising results in mice models of gastric and pancreatic cancers . Anti-CXCR4 antibody demonstrated efficiency in multiple myeloma in combination with lenalidomide or bortezomib plus dexamethasone , and several related clinical trials are ongoing. A CXCR4 inhibitor AMD3100/Plerixafor was approved by FDA as a hematopoietic stem cell mobilizer and it was recently tested in humans against pancreatic and colorectal cancers as the potential inducer of the immune response . Also, preclinical studies showed that inhibition of the CXCR4 might be potentially efficient against other cancers including GBM , and the first human clinical trial of plerixafor as an adjunct to combined chemoradiotherapy was conducted in newly diagnosed GBM patients achieving median overall survival of ~21 months. This is a significant improvement over ~17 months period characteristic for the standard chemoradiotherapy . Finally, gamma secretase inhibitors showed therapeutic effects only in CNS tumors and desmoids . Targeting of TGF-β receptor is also in development and in clinical trials . In turn, anti-VCAM antibodies dramatically reduced pancreatic ductal adenocarcinoma progression in mice models . In , we summarized drugs targeting molecules involved in the CTC heterotypic interactions.
Conclusions Analysis of the literature describing factors leading to formation of CTC clusters revealed three major features. First - the presence of either heterotypic or homotypic CTC aggregates often means unfavorable prognosis and predicts metastasis in many cancer types. Targeting the formation of such clusters is a valuable strategy for metastasis suppression . Second - cells carry mesenchymal (Vimentin) and epithelial (E-cad) markers together, which is a hallmark of intermediate epithelial associated with stemness of cancer cells . Third - in turn, intermediate Mesenchymal- Epithelial state frequently coincides with polyploidy as it was shown in lung and colorectal cancers . In lung cancer, polyploidy was accompanied by the interaction with WBCs, which were identified as neutrophils or PMN-MDSCs. It is well established that both polyploidy/mixed EMT phenotype and immunosuppressive PMN-MDSC and TAM contribute to cancer progression, however, how the interaction between them mediates metastatic advantage is yet to be investigated. Taken together, these findings highlight common mechanisms of metastasis with implication for drug development and cancer treatment.
JMR wrote the text, edited figures and performed data analysis, TM wrote the text and prepared figures, AAB conceptualization and text writing and editing, JMR, AAB and TM These authors contributed equally to this work and share first authorship. OAR, DAD, VVP, and IVA collected related literature, compiled data and wrote the manuscript. All authors contributed to the article and approved the submitted version.
|
Dental service sector and patient-reported oral health outcomes: Modification by trust in dentists | 56b49a0e-1806-483a-a2ae-5df09c96a2b9 | 10050452 | Dental[mh] | Health outcomes are based on objective clinical test results rather than on the patients' own measures ( ). However, the paradigm has shifted to the biopsychosocial model of health where subjective indicators—patient-reported outcomes (PROs) ( )—complementarily or primarily evaluate healthcare practices ( ). The rationale for adopting patients' perspectives on measuring health outcomes is in line with patient-centered care, one of the aims for the quality of care ( ). PROs refer to “any report coming directly from patients, without interpretation by physicians or others” ( ) sharing the core concept with “person-reported outcome” or “self-rated health” ( ). Dentistry has also developed and implemented context-/disease-specific PROs, as well as perceived oral health ( ). Patient-reported oral health outcomes are commonly assessed using self-rated dental health ( ) and oral health-related quality of life (OHRQoL) such as the Oral Health Impact Profile ( ). Clinical encounters remain an essential component in the healthcare system whether in terms of face-to-face practice or online distant consultation ( ). Provider–patient relationships are at the center of clinical healthcare ( ), which also applies to dentistry ( ). Given that normative patterns of patient-centeredness have led to the basis of “relationship-centered care” ( ), the assessment of health with PROs should incorporate variables of provider–patient relationships as a potential determinant. To provide more context on oral health outcomes, favorable dentist–patient relationships (DPRs) are empirically associated with better OHRQoL ( , ). Although it is difficult to operationalize the construct of DPR ( ), trust in dentists ( , ) has been acknowledged as a salient contribution factor to establish a therapeutic relationship along with satisfaction, dental fear, communication, and control at dental encounters ( ). Oral healthcare is generally provided through two dental service sectors: public and private care ( ). In Australia, based on the latest national survey, the majority of the adult population (81.8%) made their last dental visit to private practices for the limited eligibility with means tests and long waiting lists ( ). Despite the relatively small portion of dental healthcare, public services have been shown to be associated with unfavorable access to services and poor oral health outcomes in significant measures ( , ). The public dental sector is more likely to have problem-oriented services than preventive/maintenance care ( ), leading to a higher prevalence of dental caries and periodontal disease ( ). However, the inequity of oral health in different dental service sectors has been largely studied with a focus on clinical outcomes, setting aside patients' perceptions. Furthermore, the relationship between dentists and patients has not been sufficiently considered in this disparity. Derived from the gap in previous research findings, this study aimed to examine the association of patient-reported oral health outcomes with the dental service sector and trust in dentists, a representative variable of DPR. By extension, we aimed to assess whether trust in dentists has an interaction effect with the dental service sector on the association of oral health outcomes. The main hypothesis to test was that those in public dental services with lower trust in dentists were more likely to have poor patient-reported oral health outcomes. We compared the differences in PROs in oral health between private and public dental care, allowing for sociodemographic characteristics and possible modification of this DPR variable.
A total of 12,245 adults aged 18 years or older in South Australia were randomly drawn after stratification by sex and age from the Electoral Roll, a comprehensive sampling frame ( ). Data were collected by mailed self-completed questionnaires in 2015–2016, which implied that informed consent was obtained by voluntarily returning the survey forms. The cross-sectional data analyzed in this study were part of the baseline resource for a prospective cohort project for longitudinal changes in oral health outcomes by different determinants ( ). This study was approved by the Human Research Ethics Committee of the University of Adelaide (H-288-2011). All the procedures in this study were performed in accordance with the Declaration of Helsinki. The outcome variables were self-rated dental health (SRDH) and Oral Health Impact Profile (OHIP-14) to assess PROs of oral health. The SRDH is a single item of self-rating global oral health based on the question, “How would you rate your dental health?” with five response levels as follows: excellent, very good, good, poor, and very poor ( ). It has been commonly incorporated in population-based surveys ( ), with acceptable properties in the validation and predictive capability of clinical outcomes ( ). The OHIP-14 is a 14-item scale that captures the perceived oral health impact on a 5-point Likert scale ranging from never to very often ( ). The scale has been adequately validated and is widely accepted for assessing OHRQoL ( ). For the purpose of analysis in this study, “poor” oral health and oral health “impact” were defined as participants reporting the lowest two response options: either poor or very poor in SRDH and fairly often or very often in any single or multiple items of OHIP-14. The explanatory variables were the dental service sector and trust in dentists. The dental service sector was dichotomized from the question of where the last dental visit was made with choices of public or private services. Trust in dentists was measured using the Dentist Trust Scale (DTS), an 11-item psychometric scale, on a 5-point Likert scale (from 1 = strongly disagree to 5 = strongly agree) ( ). DTS was modified from the original “trust in physicians” scale, and both satisfied construct validity and reliability (Cronbach's α = 0.92 in the current study) ( , ). The response score for each DTS item was summed (ranging from 11 to 55; higher scores indicated higher trust), and the total score was classified into tertiles as a category variable (lower tertile ranging from 11 to < 38, middle from 38 to < 45, and upper from 45 to 55). Other covariates were included in the analysis to adjust for the demographic and socioeconomic characteristics. Demographic variables were age (categorized as “18–39,” “40–59,” or “≥60” years) and sex (“female” or “male”). Socioeconomic status (SES) was assessed using annual household income (“ < $80,000” or “≥$80,000” in AUD) and the highest level of education completed (“ ≤ year 12 or certificate” or “diploma/degree”). The collected data were prepared using data cleaning/screening before descriptive statistics and association analyses. Respondents with critical missing values (e.g., SRDH or any item of OHIP-14) and/or the number of missing items >20% in the DTS (≥3 items missing) were filtered out. To prevent acquiescence bias, those with identical responses for all items in the DTS were excluded, considering the inclusion of two reverse-coded items in the scale. Missing values of up to two items in the DTS were imputed using the expectation–maximization algorithm with an iterative maximum-likelihood estimation. Data were weighted by the distribution of age by sex to represent the population estimates of the variables. Descriptive statistics with a frequency table and unadjusted bivariate associations were analyzed for outcomes, explanatory variables, and covariates. Adjusted associations were calculated with prevalence ratios (PRs) using log-binomial regression. Interaction terms between the private dental service sector and levels of trust were included to test possible modifications. SPSS Statistics (version 25.0., IBM Corp., Chicago, IL) was used for all statistical analyses, and a p -value of < 0.05 was adopted as the threshold for statistical significance.
Response data were analyzed from 4,027 respondents after excluding 491 participants who were screened for missing values and unengaged data criteria. The adjusted valid response rate was 40.0%. The sociodemographic characteristics of the study participants were compared with those of the general population census data to check for possible response bias ( ). The respondents' profile had a close approximation of the population data, with minor differences, mainly in SES. A larger proportion of adults with better SES were sampled in this study with a higher education level of diploma/degree (42.2 vs. 30.0%) and income ranging ≥$80,000 (45.5 vs. 39.8%). Participants were of a slightly higher percentage from the younger age group and private dental sector. The descriptive statistics and unadjusted associations are presented in . The mean of the summed DTS scores was statistically different among the age groups and education levels. The older age group and those with lower education levels had higher trust in dentists ( p < 0.01). The DTS score was slightly higher in the private dental services group, but the difference was not statistically significant ( p = 0.060). Bivariate association analysis indicated that the prevalence of poor dental health from SRDH and oral health impact from OHIP-14 were associated with all the variables included in the model ( p < 0.01). The common pattern of both outcome variables was that of a higher prevalence in those with lower income and education, public dental sector, and lower trust in dentists. , of stratified percentages confirm the pattern of prevalence in accordance with the dental service sector and DTS tertile. Regarding the prevalence of SRDH, older adults and individuals of the male sex were more likely to report poor oral health. In contrast, the middle-aged group and female sex showed a higher prevalence of oral health impact in OHIP-14. The overall prevalence of poor dental health and oral health impacts was 10.9 and 19.2%, respectively. After adjusting for all relevant variables included in the multivariable regression, a similar pattern of association was observed in the adjusted PRs ( ). The direction of unadjusted associations was maintained but attenuated with the loss of statistical significance in the DTS middle tertile for SRDH, and age of ≥60 years, and DTS middle and lower tertiles for OHIP-14. There was a significant interaction effect between private dental services and DTS lower tertile on the prevalence of oral health impact ( p < 0.05). The higher PR (1.51, 95% CI 1.06–2.14) indicated that those with lower trust in dentists in the private sector had a much higher prevalence of oral health impact compared to those in the upper tertile of trust in the private sector. The other interaction terms showed adjusted PRs lower than 1.0 but did not reach statistical significance.
The findings of this study showed that poor dental health and oral health impacts were associated with the dental service sector and trust in dentists. In addition to the main effects of the explanatory variables, lower trust modified the relationship between private dental services and oral health impact as an interaction effect. Throughout the analysis, sociodemographic covariates were associated with PROs of oral health in both the unadjusted and adjusted models. This study reaffirms the social gradient and inequality in oral health ( ). Those in worse SES with lower income and education level were more likely to be consistently involved in unfavorable oral health outcomes in both bivariate and multivariable analyses. Regardless of clinical indicators or subjective self-ratings of oral health ( ), the pattern has been deeply rooted in the social determinant framework ( ). Although social inequalities in oral health have been highlighted over the past few decades, their root causes are still yet to be addressed properly and inveterately ( ). From the perspective of the oral healthcare system, inequality also depends on disparities in access to care ( ) and its relevant clinical outcomes ( ) between the private and public dental service sectors. This study reported a similar finding that public dental service users had a higher prevalence of poor dental health and oral health impacts as subjective oral health outcomes. In particular, the association between oral health outcomes and the dental service sector remained statistically significant after adjusting for SES variables—socioeconomic disadvantages ( ). This indicates the need to investigate the role of the dental care delivery system as an independent determinant of oral health outcomes. In addition, trust in dentists was included in the model as a psychosocial factor at the micro level of social dentistry ( ). The empirical results of the association between higher trust and better oral health outcomes can support the rationale for a favorable DPR beyond normative suggestions ( ). An incongruent pattern of reporting oral health was found for sex differences. Female participants self-rated their dental health on a better level (9.0 vs. 12.8% in SRDH) but felt more of an oral health impact (21.3 vs. 17.0% in OHIP-14) than their male counterparts. This inconsistency has been consistently presented in a series of population-based surveys (three national surveys conducted between 2004 and 2018) in Australia, where the current study was performed ( ). With no exception since 2004, sex differences in Australian adults indicated that women have better self-rated dental health but more complaints in specific dental conditions. Cognitive dissonance may occur from conceptual differences between SRDH and OHIP-14, despite their commonality as PROs. The former focused on self-rating global oral health, but the specific referents were taken differently by respondents ( ), which contrasts with the latter of the less equivocal multi-item scales based on seven dimensions ( ). More specifically, on demographics, the pattern may be derived from the finding that women were likely to perform better oral hygiene behaviors but report more concerns about dental complaints ( ). However, opposing results have also been reported, such as Asian American subgroups ( ) and Brazilian adolescents ( ). The sex difference in PROs of oral health needs to be investigated further in a rigorous systematic search ( ). The interpretation of the interaction effects with adjusted PRs requires caution. shows the PRs in a relative frame rather than a subgroup analysis. For example, aside from statistical significance, the PRs (< 1.0) of interaction terms should not be interpreted as those in the lower/middle tertile of trust having a lower prevalence of oral health outcomes in each sector. Instead, the significant PR of the interaction indicates that the negative effect of lower trust leading to an oral health impact is more pronounced in private dental services than in the public sector. As shown in , the increase in the prevalence of OHIP-14 by lower trust is far greater in private services, resulting in a much smaller relative difference from the prevalence in the public sector than in any other segment. The possibility of potential confounding or mediation should also be considered. The claim that trust in dentists may be a confounder was dismissed, as the distribution of DTS tertiles in the private and public sectors was not statistically different ( p = 0.201). For the mediation effect, in addition to the similar distribution, it appeared to be less likely that the association between outcomes and the dental service sectors would remain statistically significant after adjusting for DTS in the model ( ). , show the prevalence of PROs and the pattern of how participants report oral health outcomes by dental service sector and DTS tertile levels. If private dental patients have lower trust in dentists, they are likely to report disproportionally worse oral health than those with middle and upper levels of trust (prevalence from upper to lower tertile: 7.3%, 6.9%, and 13.3% for SRDH; 13.0%, 13.4%, and 24.5% for OHIP-14). Compared with the private sector, public dental service users are likely to detect higher trust in dentists in terms of oral health outcomes than those in the middle and lower DTS tertiles. This pattern in the public sector appears clearly for the prevalence of OHIP-14 (28.2, 32.5, and 33.3%), and SRDH also partially supports it with a difference of 3.5% from middle to lower tertiles vs. 6.3% from upper to middle DTS tertiles (14.7, 21.0, and 24.5%). This study has some limitations. The cross-sectional design can only purport the association of outcomes with explanatory variables, not necessarily causal inferences from the findings. Despite a similar profile to the aforementioned population, study participants might have different characteristics, causing selection bias. For example, the overall prevalence of poor oral health in SRDH was 10.9% in this study, which was considerably lower than that reported in national surveys [23.9% in 2017–2018 ( ) and 18.8% in 2010 ( )]. However, this discrepancy may be due to a measurement bias with different rating statements. The two national surveys adopted lower response levels with “fair” and “poor” rather than “poor” and “very poor” options in this study for the definition of poor oral health. Thus, the rating scale should be consistent across studies for comparability in future. Another limitation is the absence of important covariates due to the study topic. Except for sociodemographic variables, oral health behaviors (e.g., tooth brushing and smoking) and dental service variables (e.g., time since/purpose of the last dental visit and perceived dental needs) have been reported to be associated with the dental service sector ( , ) and oral health outcomes ( , ). Moreover, adult development may be associated with the acceleration of trust ( ), which needs to be considered as a potential covariate in further studies. Nevertheless, this study could provide a dental care system with a more comprehensive understanding of PROs, as multiple measures are recommended to assess different aspects of perceived oral health for dental service planning ( ). The findings of this study have practical implications. First, efforts to tackle inequality in oral health need to command attention in terms of the dental service sector. In addition to socioeconomic disadvantages, the quality of public dental services should also be considered an independent factor. Second, trust in dentists for better DPR may be a determinant of PROs of oral health. In particular, lower trust harshly impacts private dental patients—the majority of dental services provided in Australia—than those in the public sector. Finally, for female adult patients in Australia, a probing single question about global oral health asked by clinicians in dental encounters may lead to missing out on specific dental conditions inadvertently. Considering that women are more likely to experience communication problems with their dentists ( ), the clinical implication may be salutary to establish better DPR with female patients.
Patient-reported oral health outcomes were associated with sociodemographic characteristics, the dental service sector, and trust in dentists. Lower trust in dentists in private dental care had a disproportionately worse effect on oral health impact compared to those with higher trust in the private service sector. The disparity in oral health outcomes between dental service sectors needs to be addressed both independently and in association with covariates, including socioeconomic disadvantages. Trust in dentists should also be established to improve oral health outcomes, particularly for private dental service users with lower levels of trust.
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.
The studies involving human participants were reviewed and approved by the Human Research Ethics Committee of the University of Adelaide. The patients/participants provided their written informed consent to participate in this study.
YS contributed to the conception of the article, data analysis, interpretation of results, and drafting of the manuscript. PS, RN, and H-JC contributed to the interpretation of results and critical revision of the manuscript. DB contributed to the conception of the article and critical revision of the manuscript. All authors have read and approved the content of the manuscript.
|
Comment on: ‘Impact of the COVID-19 pandemic on the research activities of UK ophthalmologists’ | 7aa38e7c-8828-44c7-82ee-ae395a203790 | 10050802 | Ophthalmology[mh] | Hogg et al. have explored the effects of the COVID-19 pandemic on the research activities of UK ophthalmologists . Their survey, distributed to members of the Royal College of Ophthalmologists, revealed that the COVID-19 pandemic negatively impacted the research of 91.2% (104 out of 114) of research-active respondents. This included a loss of research time ( n = 69), research delays ( n = 96) and funding shortfalls ( n = 63). This is likely to have negatively impacted the opportunities for interested medical students and trainees to become involved in ophthalmic research. Undergraduate involvement in ophthalmic research has numerous benefits for academic institutions, supervisors, medical students and the wider scientific community. Through undertaking research projects, students can gain research skills, mentorship and a deeper insight into a career in Ophthalmology. The latter is increasingly important given that 72% of UK medical school Ophthalmology placements last one week or less in duration and the COVID-19 pandemic has reduced clinical exposure in Ophthalmology . Furthermore, research outputs enhance students’ and trainees’ portfolios, enabling them to make competitive applications for Ophthalmology training in the future. As medical students in our clinical years during the COVID-19 pandemic, we appreciate the impact of the pandemic on clinical exposure in Ophthalmology and consequent opportunities to meet potential supervisors. With fewer ophthalmologists conducting research during the pandemic, there are likely to be fewer opportunities for interested medical students. A potential solution to increase high-quality research outputs, whilst involving medical students in research, is to harness the power of collaborative learning and develop an undergraduate research network with appropriate supervision . Furthermore, there is currently a lack of research exploring the perceived barriers to undergraduate research involvement in the UK . Future studies investigating the enablers and barriers to medical student involvement in ophthalmic research, and the impact of the COVID-19 pandemic on the research activities of undergraduates, are welcomed. |
Rhizosphere Fungal Dynamics in Sugarcane during Different Growth Stages | 3d878338-6a6b-48a1-8013-747933be9012 | 10052501 | Microbiology[mh] | The microbial community in the rhizosphere environment is critical for the health of terrestrial plants and sustainable soil development . As more attention is paid to the relationship between plants and soil microbiomes, comprising the interactions between microbial and plant root systems and genetic components [ , , ], it is apparent that more exploration is needed of what constitutes normal temporal variations in rhizosphere microbial community structures and compositions. In particular, research related to rhizosphere fungal communities that may produce soil-borne diseases of plants is essential. Variations in the rhizosphere fungal group and between different plants and life-cycle stages have a lot of space for exploration. As we all know, after germinating in the soil, plant roots begin to recruit microbial communities closely related to their growth through the accumulation of root secretions, rhizosphere deposition, and nutrient uptake. Thus, the growth activities of plants during different periods cause changes in rhizosphere fungal communities and functions. For instance, model plants ( Arabidopsis thaliana and Medicago truncatula ) can retain certain fungal populations in the soil through certain chemicals secreted by their root systems . In addition, advances in sequencing technology provide a foundation for research on the dynamic changes of microbes in the rhizosphere of plants in space and time . Similarly, these types of studies are needed to understand the immigration and emigration patterns of microorganisms between different parts of the plant, between different points of time and between different plants and soil environments [ , , ], especially the related research on time series. For example, Zhang et al. (2018) identified biomarker taxa and established a model to correlate root microbiota with rice resident time in the field using a machine learning approach , and Li et al. (2020) analyzed the variation of the fungal community structure in the rhizosphere of Gardenia , which provided a new theoretical framework for further study on the rhizosphere mechanism . Furthermore, as an important economic crop of food and biofuel, sugarcane is constantly expanding its planting area under the condition of global warming . The study of sugarcane rhizosphere fungi temporal variation has great significance in better understanding the interaction between sugarcane and rhizosphere microbiota . For example, sugarcane growth is facing a huge threat due to Yellow Canopy Syndrome (YCS) in Australia. YCS is a kind of largely undiagnosed plant disease that is impacting sugarcane growth across Queensland, Australia, causing huge yield losses . Studying the temporal variation of sugarcane rhizosphere microorganisms provides a new way of thinking about the solution of YCS. Additionally, given the growing importance of sugarcane , identifying ways to promote the healthy growth of sugarcane and sustainably increase productivity is critical, and hence, the potential to harness microorganisms from the sugarcane rhizosphere has recently gained more attention [ , , ]. For instance, rhizosphere microorganisms are involved in the solubilization of phosphorus and potassium containing minerals . Sugarcane smut disease, caused by a fungus called Sporisorium scitamineum (Basidiomycota, Ustilaginales), is a limiting factor to cane production and a potential threat to the sugar industry. Juma and Musyimi’s research showed that the selected isolates from sugarcane rhizosphere microflora had an evidently antagonistic activity against the Sporisorium scitamineum , and is recommended as a potential biocontrol agent for this pathogen . In a related study, She et al. (2021) found maize rhizosphere microorganisms are critical for facilitating microbiome bioremediation for soil affected by neutral-alkaline mining . Moreover, understanding how the sugarcane rhizosphere fungal population change with stages would provide valuable data and could in the long-term result in the identification of potential management strategies for healthy cultivation of sugarcane, including an improved breeding process taking the microbe into account or better targeted biological prevention and treatments . Here, we used sugarcane as an experimental plant to fill critical knowledge gaps that will accelerate our exploration process to successfully harness rhizosphere fungal groups to sustainably enhance sugarcane productivity. This also has reference significance for other crops in nature. Our study had three main objectives: (i) to determine the drivers of the sugarcane fungal community under field conditions; (ii) to understand the seasonal trends of fungi in the rhizosphere of sugarcane; and (iii) to explore the assembly and interaction relationships of rhizosphere fungal communities during sugarcane development. To achieve this, we examined fungal community assemblages of sugarcane rhizosphere soil through Illumina amplicon sequencing of the samples collected from young to mature sugarcane (Seeding, Tillering, Elongation, and Maturity). Our results provide more foundational insight into the process of sugarcane root microbiota exploration.
2.1. Sugarcane Rhizosphere Nutrients Produce Differential Changes at Different Stages The results showed that the height and stem diameter of sugarcane plants increased continuously over time, with the rate of diseased plants reaching their highest (19.8%) in Elongation ( ). We also investigated the temporal variation of soil biochemical properties collected in Seeding, Tillering, Elongation, and Maturity during the different growth stages of sugarcane. Our analysis showed that soil biochemical properties varied with time. Soil properties such as soil pH and available nitrogen (AN) decreased considerably ( p < 0.05), with Elongation and Maturity recording the lowest soil pH, while Seeding had recorded the highest, followed by Tillering. In addition, the soil organic matter (OM) and total phosphorus (TP) were significantly higher in Elongation compared to Seeding, Tillering, and Maturity ( p < 0.05). We also observed that Seeding recorded a significant amount of soil total nitrogen (TN) relative to the other stages. Compared to Elongation, a considerable amount of soil available phosphorus (AP) was recorded in Seeding and Tillering, and the available phosphorus nutrient decreased by 20.07% and 25.37%, respectively. In comparison to Tillering and Maturity, soil total potassium (TK) revealed significant improvement ( p < 0.05) in Seeding, followed by Elongation. However, soil available potassium (AK) also showed significant differences among the different time points, with the lowest AK values in Maturity ( p < 0.05), which is significantly reduced by 24.68% compared with the highest value. Soil temperature peaked significantly in Tillering and Elongation than in Seeding and Maturity ( ). 2.2. Rhizosphere Fungal Diversity Differs between Different Stages and Correlates with Soil Properties The total number of sequences in the raw sequencing data is 4,592,374 with an average length of 401.21. The shortest sequence length is 221 and the longest sequence length is 453. For details, see Annexes-1. Fungal diversity (Shannon and Simpson), richness (ACE), and Sobs (number of observed species) were assessed during the different time points. It was observed that fungal diversity increased substantially ( p < 0.05) in Tillering compared with the other stages ( and ). Moreover, compared to Maturity, fungal richness and Sobs were significantly enriched ( p < 0.05) in Tillering, followed by Seeding and Elongation ( , ). In addition, the dilution curve showed a sufficient amount of sequencing data ( ). We carried out Principal Coordinates Analysis (PCoA) to explore and visualise similarities or dissimilarities in fungal community composition in all the samples collected during the different growth stages of sugarcane. Based on the growth changes of FN 41 sugarcane throughout its life cycle, simulation map with distinctive features for different stages were constructed ( A). The analysis revealed that fungal community composition was clustered together in the different samples, and the soil samples shifted with different stages in the first axis ( B). The relationship between soil environmental variables and fungal alpha diversity was investigated using Pearson’s correlation coefficients ( A). The analysis demonstrated that soil temperature (Tem) exhibited a strong and positive association with fungal Sobs and ACE, Shannon and PC1. However, soil Tem was negatively associated with PC2. Furthermore, soil pH revealed a significant positive correlation with PC1, Sobs, and ACE. Soil TN and AN were significantly and positively associated with PC1. Soil TN also exhibited a positive relationship with PC2, while TK had a strong positive association with PC1 and PC2. We also noticed that soil TP demonstrated a positive correlation with PC2. On the other hand, soil TN and TK were negatively related to Shannon. Meanwhile, regression analysis was conducted to further confirm the association among soil environmental variables, fungal ACE (richness), Shannon (diversity), PC1 and PC2 during the different time points ( ). The analysis demonstrated that soil pH and AN showed a positive relationship with fungal richness. Moreover, soil Tem exhibited a positive relationship with fungal richness, particularly in Tillering and Elongation ( ). Soil pH and C/N showed a positive correlation with fungal richness in Tillering, whereas AN showed a strong positive correlation with fungal richness in Seeding. In Tillering and Elongation, soil Tem revealed a strong positive relationship with fungal richness ( ). Furthermore, soil pH, TN, AP, AN, and AK had positive relationships with PC1 in Seeding, while in Tillering and Elongation, soil Tem was positively related to PC1 ( ), whereas in Seeding, TN and TK had a positive association with PC2, while in Elongation and Maturity soil TP and AK demonstrated a positive correlation with PC2, respectively ( ). 2.3. Variability in Fungal Community Composition at Different Growth Stages Fungal relative abundance during the stages was assessed at the phylum level. The analysis showed that Ascomycota accounted for more than 50% of the absolute dominance in the entire sample present in the rhizosphere soil of sugarcane, while Basidiomycota, Chytridiomycota, and Glomeromycota were present in smaller dominant groups in the hot zone of sugarcane roots. However, Chytridiomycota and Glomeromycota were significantly higher ( p < 0.05) in Tillering than the other stages ( ). Basidiomycota increased significantly in Tillering and Seeding compared to other phases and the mean sample abundance peaked in Tillering ( ). Venn diagram analysis demonstrated that 15, 23, 11, and 11 unique fungal OTUs were detected in Seeding, Tillering, Elongation, and Maturity, respectively ( B). Moreover, among the soil samples collected in the various stages, 451 OTUs were shared among the different time points. In Tillering, the highest OTUs were recorded compared with the other sampling times. 2.4. Fungal Community Composition, FUNGuild, and Soil Physicochemical Properties Are Interrelated The different environmental variables were significantly correlated with the variation of selected soil fungal genera ( A). RDA demonstrated the relationship between soil properties and fungal communities, which had an eigenvalue of 0.2846 for the first axis and 0.0802 for the second axis, respectively. The vectors indicated the following: total N, total P, total K, pH, available N and soil temperature played a greater role than available K, available P and organic matter for sugarcane rhizosphere fungi genera. The soil fungus community was dominated by Basidiomycota and Ascomycota, which showed stronger associations with higher pH, available N and soil temperature, while Chytridiomycota slightly revealed the opposite trend. We then used the Source Model of Plant Microbiome (SMPM) to estimate the proportion of sugarcane rhizosphere fungal communities from “adjacent period” and “unknown” sources and took the adjacent periods as the source and library in turn ( B). The results showed that the related fungal communities mainly came from the transmission of adjacent periods, and the transfer proportion decreased gradually with the migration of time. In addition, Tremellales and Saccharomycetales, as the dominating orders in fungi, were positively associated with pH and AN. However, Sordariomycetes and Pseudallescheria showed a stronger negative association with pH and AN ( D). To further explore the environmental driving factors of the sugarcane fungus community, we conducted a Mantel-test analysis and analyzed the correlation between the microbial matrix and the soil property matrix, and then we correlated distance-corrected dissimilarities of taxonomic and functional community composition with those of environmental factors. Overall, pH, TN and TK exhibited the strongest correlations with the taxonomic composition in the sugarcane rhizosphere soil ( C), while no significant correlation was found for other soil factors ( p < 0.05). In addition, soil temperature and pH were only weakly correlated with taxonomic and functional community composition. Besides, almost all the environmental variables exhibited no significant correlations, except for AN. 2.5. Fungal Communities Differ in Co-Occurrence Networks with Changes in Growth Periods We employed Manhattan plots to examine fungal community composition OTUs differences in sugarcane rhizosphere soil during the growth stages ( and ). The analysis revealed that Basidiomycota, Ascomycota, and Chytridiomycota were significantly more abundant in Tillering than in Seeding ( A). Additionally, Ascomycota increased considerably in Tillering relative to Seeding; however, Basidiomycota and Chytridiomycota decreased profoundly in Seeding ( B). In Elongation, Basidiomycota and Ascomycota increased significantly, while Chytridiomycota diminished considerably in Maturity ( C). Meanwhile, Venn diagram analysis was adopted to gain a deeper understanding of the genera of fungal OTUs from one time point to another. The analysis demonstrated that between Seeding and Tillering, the highest number of enriched fungal OTUs was detected, followed by Elongation to Maturity ( E). However, between Tillering and Elongation the highest amount of depleted fungal OTUs were identified, followed by Elongation to Maturity ( D). Meanwhile, the co-occurrence network showed the interactions of rhizosphere fungi in sugarcane during four critical periods, where the fungi belonging to mainly Talaromycetes , Talaromyces , Fusarium , Sordariales , and Pseudallescheria were dominated ( F). The role of each fungus in the network changed over time during different reproductive periods, with the degree and importance changing. Among the four critical fertility networks, Seeding had the largest average density and mean degree, followed by Maturity ( ). 2.6. Construction of Structural Equation Models Related to Pokkah Boeng Disease of Sugarcane We constructed an SEM to assess the direct and indirect effects of soil properties (pH, AN and TN), microbial genera ( Fusarium and Talaromyces ), and microbial diversity (ACE and Shannon) on the incidence of sugarcane ( and and ). We used a multigroup modelling method to assess which relationships exist between soil properties, microbial communities, and sugarcane disease during the growing process. The results of model-1 showed that soil pH was significantly and directly affected the abundance of Fusarium, sugarcane disease and fungal diversity. However, soil pH negatively regulated the disease rate of sugarcane (the lower the pH, the more disease occurs) ( A and ). Meanwhile, AN, TN and disease had similar significant regulatory relationships as pH and disease ( , ). In addition, soil temperature affected the diversity of rhizosphere fungi, and had both direct and indirect effects on the abundance of Fusarium and Talaromyces. 2.7. The Assembly Process of Fungal Communities Is Influenced by Changes in the Growth Period The results showed that the mean nearest-taxon index (NTI) was greater than 0, and the mean nearest taxon distance between samples (β-NTI) values among the samples from the critical fertility period of sugarcane were mainly concentrated in the interval of −2 to 2, indicating that the changes in microbial community structure during the critical fertility period were mainly influenced by stochastic factors. The existence of a certain number of samples with β-NTI values greater than 2 among the soils in different periods indicated that the changes in the rhizosphere microbial community structure of some soil samples were mainly influenced by deterministic factors ( B). In addition, the β-NTI values of samples between stages (Seeding–Tillering, Tillering–Elongation, and Elongation–Maturity) were between (−2, 2), indicating that the changes in the rhizosphere microbial community structure of sugarcane with time were mainly influenced by stochastic factors, but there were differences in the contribution of stochastic factors with time ( ). 2.8. Functional and Evolutionary Relationships among the Major Fungal Genera of Rhizosphere Soil As for functional classification, the fungal communities in rhizosphere soil were classified by using the trophic mode ( ). A maximum likelihood phylogenetic tree was conducted to further visualized the relationship among crucial fungal communities. Following the procedure, the top 28 genera, were classified into five guilds. They were divided into Undefined Saprotroph, Dung Saprotroph, Animal Pathogen, Plant Pathogen, and Endophyte, respectively. Based on the classification, the results revealed that 33.3% of these genera belonged to Undefined Saprotroph, 3.3 % belonged to Plant Pathogen, 6.6 % belonged to Animal Pathogen, 6% belonged to Dung Saprotroph, 3.3% belonged to Endophyte while 46.6% of these genera were unclassified. We also observed that these top 28 genera, were detected in 8 different phyla, of which five were identified as Ascomycota, Chytridiomycota, Glomeromycota, and Basidiomycota, respectively. Moreover, the plant pathogenic fungi observed in this study was Cochliobolus, and it is worth noting that the abundance of Fusarium genera was higher in Tillering and Maturity. In addition, according to the results of the FUNGuild classification difference ( ), in the comparison of the Seeding and Tillering of sugarcane growth ( ), there were more classifications with significant differences ( p < 0.05). They were Dung Saprotroph-Soil Saprotroph-Wood Saprotrop, Animal Pathogen–Endophyte–Lichen Parasite–Plant, Orchid Mycorrhizal–Plant Pathogen–Wood Saprotroph, Endophyte–Plant Pathogen, Animal Pathogen–Soil Saprotroph, respectively. However, with the shift in sugarcane stages, the number of significantly different FUNGuild classifications gradually decreased, accompanied by a decrease in the significance of differences between some FUNGuild classifications. ( ).
The results showed that the height and stem diameter of sugarcane plants increased continuously over time, with the rate of diseased plants reaching their highest (19.8%) in Elongation ( ). We also investigated the temporal variation of soil biochemical properties collected in Seeding, Tillering, Elongation, and Maturity during the different growth stages of sugarcane. Our analysis showed that soil biochemical properties varied with time. Soil properties such as soil pH and available nitrogen (AN) decreased considerably ( p < 0.05), with Elongation and Maturity recording the lowest soil pH, while Seeding had recorded the highest, followed by Tillering. In addition, the soil organic matter (OM) and total phosphorus (TP) were significantly higher in Elongation compared to Seeding, Tillering, and Maturity ( p < 0.05). We also observed that Seeding recorded a significant amount of soil total nitrogen (TN) relative to the other stages. Compared to Elongation, a considerable amount of soil available phosphorus (AP) was recorded in Seeding and Tillering, and the available phosphorus nutrient decreased by 20.07% and 25.37%, respectively. In comparison to Tillering and Maturity, soil total potassium (TK) revealed significant improvement ( p < 0.05) in Seeding, followed by Elongation. However, soil available potassium (AK) also showed significant differences among the different time points, with the lowest AK values in Maturity ( p < 0.05), which is significantly reduced by 24.68% compared with the highest value. Soil temperature peaked significantly in Tillering and Elongation than in Seeding and Maturity ( ).
The total number of sequences in the raw sequencing data is 4,592,374 with an average length of 401.21. The shortest sequence length is 221 and the longest sequence length is 453. For details, see Annexes-1. Fungal diversity (Shannon and Simpson), richness (ACE), and Sobs (number of observed species) were assessed during the different time points. It was observed that fungal diversity increased substantially ( p < 0.05) in Tillering compared with the other stages ( and ). Moreover, compared to Maturity, fungal richness and Sobs were significantly enriched ( p < 0.05) in Tillering, followed by Seeding and Elongation ( , ). In addition, the dilution curve showed a sufficient amount of sequencing data ( ). We carried out Principal Coordinates Analysis (PCoA) to explore and visualise similarities or dissimilarities in fungal community composition in all the samples collected during the different growth stages of sugarcane. Based on the growth changes of FN 41 sugarcane throughout its life cycle, simulation map with distinctive features for different stages were constructed ( A). The analysis revealed that fungal community composition was clustered together in the different samples, and the soil samples shifted with different stages in the first axis ( B). The relationship between soil environmental variables and fungal alpha diversity was investigated using Pearson’s correlation coefficients ( A). The analysis demonstrated that soil temperature (Tem) exhibited a strong and positive association with fungal Sobs and ACE, Shannon and PC1. However, soil Tem was negatively associated with PC2. Furthermore, soil pH revealed a significant positive correlation with PC1, Sobs, and ACE. Soil TN and AN were significantly and positively associated with PC1. Soil TN also exhibited a positive relationship with PC2, while TK had a strong positive association with PC1 and PC2. We also noticed that soil TP demonstrated a positive correlation with PC2. On the other hand, soil TN and TK were negatively related to Shannon. Meanwhile, regression analysis was conducted to further confirm the association among soil environmental variables, fungal ACE (richness), Shannon (diversity), PC1 and PC2 during the different time points ( ). The analysis demonstrated that soil pH and AN showed a positive relationship with fungal richness. Moreover, soil Tem exhibited a positive relationship with fungal richness, particularly in Tillering and Elongation ( ). Soil pH and C/N showed a positive correlation with fungal richness in Tillering, whereas AN showed a strong positive correlation with fungal richness in Seeding. In Tillering and Elongation, soil Tem revealed a strong positive relationship with fungal richness ( ). Furthermore, soil pH, TN, AP, AN, and AK had positive relationships with PC1 in Seeding, while in Tillering and Elongation, soil Tem was positively related to PC1 ( ), whereas in Seeding, TN and TK had a positive association with PC2, while in Elongation and Maturity soil TP and AK demonstrated a positive correlation with PC2, respectively ( ).
Fungal relative abundance during the stages was assessed at the phylum level. The analysis showed that Ascomycota accounted for more than 50% of the absolute dominance in the entire sample present in the rhizosphere soil of sugarcane, while Basidiomycota, Chytridiomycota, and Glomeromycota were present in smaller dominant groups in the hot zone of sugarcane roots. However, Chytridiomycota and Glomeromycota were significantly higher ( p < 0.05) in Tillering than the other stages ( ). Basidiomycota increased significantly in Tillering and Seeding compared to other phases and the mean sample abundance peaked in Tillering ( ). Venn diagram analysis demonstrated that 15, 23, 11, and 11 unique fungal OTUs were detected in Seeding, Tillering, Elongation, and Maturity, respectively ( B). Moreover, among the soil samples collected in the various stages, 451 OTUs were shared among the different time points. In Tillering, the highest OTUs were recorded compared with the other sampling times.
The different environmental variables were significantly correlated with the variation of selected soil fungal genera ( A). RDA demonstrated the relationship between soil properties and fungal communities, which had an eigenvalue of 0.2846 for the first axis and 0.0802 for the second axis, respectively. The vectors indicated the following: total N, total P, total K, pH, available N and soil temperature played a greater role than available K, available P and organic matter for sugarcane rhizosphere fungi genera. The soil fungus community was dominated by Basidiomycota and Ascomycota, which showed stronger associations with higher pH, available N and soil temperature, while Chytridiomycota slightly revealed the opposite trend. We then used the Source Model of Plant Microbiome (SMPM) to estimate the proportion of sugarcane rhizosphere fungal communities from “adjacent period” and “unknown” sources and took the adjacent periods as the source and library in turn ( B). The results showed that the related fungal communities mainly came from the transmission of adjacent periods, and the transfer proportion decreased gradually with the migration of time. In addition, Tremellales and Saccharomycetales, as the dominating orders in fungi, were positively associated with pH and AN. However, Sordariomycetes and Pseudallescheria showed a stronger negative association with pH and AN ( D). To further explore the environmental driving factors of the sugarcane fungus community, we conducted a Mantel-test analysis and analyzed the correlation between the microbial matrix and the soil property matrix, and then we correlated distance-corrected dissimilarities of taxonomic and functional community composition with those of environmental factors. Overall, pH, TN and TK exhibited the strongest correlations with the taxonomic composition in the sugarcane rhizosphere soil ( C), while no significant correlation was found for other soil factors ( p < 0.05). In addition, soil temperature and pH were only weakly correlated with taxonomic and functional community composition. Besides, almost all the environmental variables exhibited no significant correlations, except for AN.
We employed Manhattan plots to examine fungal community composition OTUs differences in sugarcane rhizosphere soil during the growth stages ( and ). The analysis revealed that Basidiomycota, Ascomycota, and Chytridiomycota were significantly more abundant in Tillering than in Seeding ( A). Additionally, Ascomycota increased considerably in Tillering relative to Seeding; however, Basidiomycota and Chytridiomycota decreased profoundly in Seeding ( B). In Elongation, Basidiomycota and Ascomycota increased significantly, while Chytridiomycota diminished considerably in Maturity ( C). Meanwhile, Venn diagram analysis was adopted to gain a deeper understanding of the genera of fungal OTUs from one time point to another. The analysis demonstrated that between Seeding and Tillering, the highest number of enriched fungal OTUs was detected, followed by Elongation to Maturity ( E). However, between Tillering and Elongation the highest amount of depleted fungal OTUs were identified, followed by Elongation to Maturity ( D). Meanwhile, the co-occurrence network showed the interactions of rhizosphere fungi in sugarcane during four critical periods, where the fungi belonging to mainly Talaromycetes , Talaromyces , Fusarium , Sordariales , and Pseudallescheria were dominated ( F). The role of each fungus in the network changed over time during different reproductive periods, with the degree and importance changing. Among the four critical fertility networks, Seeding had the largest average density and mean degree, followed by Maturity ( ).
We constructed an SEM to assess the direct and indirect effects of soil properties (pH, AN and TN), microbial genera ( Fusarium and Talaromyces ), and microbial diversity (ACE and Shannon) on the incidence of sugarcane ( and and ). We used a multigroup modelling method to assess which relationships exist between soil properties, microbial communities, and sugarcane disease during the growing process. The results of model-1 showed that soil pH was significantly and directly affected the abundance of Fusarium, sugarcane disease and fungal diversity. However, soil pH negatively regulated the disease rate of sugarcane (the lower the pH, the more disease occurs) ( A and ). Meanwhile, AN, TN and disease had similar significant regulatory relationships as pH and disease ( , ). In addition, soil temperature affected the diversity of rhizosphere fungi, and had both direct and indirect effects on the abundance of Fusarium and Talaromyces.
The results showed that the mean nearest-taxon index (NTI) was greater than 0, and the mean nearest taxon distance between samples (β-NTI) values among the samples from the critical fertility period of sugarcane were mainly concentrated in the interval of −2 to 2, indicating that the changes in microbial community structure during the critical fertility period were mainly influenced by stochastic factors. The existence of a certain number of samples with β-NTI values greater than 2 among the soils in different periods indicated that the changes in the rhizosphere microbial community structure of some soil samples were mainly influenced by deterministic factors ( B). In addition, the β-NTI values of samples between stages (Seeding–Tillering, Tillering–Elongation, and Elongation–Maturity) were between (−2, 2), indicating that the changes in the rhizosphere microbial community structure of sugarcane with time were mainly influenced by stochastic factors, but there were differences in the contribution of stochastic factors with time ( ).
As for functional classification, the fungal communities in rhizosphere soil were classified by using the trophic mode ( ). A maximum likelihood phylogenetic tree was conducted to further visualized the relationship among crucial fungal communities. Following the procedure, the top 28 genera, were classified into five guilds. They were divided into Undefined Saprotroph, Dung Saprotroph, Animal Pathogen, Plant Pathogen, and Endophyte, respectively. Based on the classification, the results revealed that 33.3% of these genera belonged to Undefined Saprotroph, 3.3 % belonged to Plant Pathogen, 6.6 % belonged to Animal Pathogen, 6% belonged to Dung Saprotroph, 3.3% belonged to Endophyte while 46.6% of these genera were unclassified. We also observed that these top 28 genera, were detected in 8 different phyla, of which five were identified as Ascomycota, Chytridiomycota, Glomeromycota, and Basidiomycota, respectively. Moreover, the plant pathogenic fungi observed in this study was Cochliobolus, and it is worth noting that the abundance of Fusarium genera was higher in Tillering and Maturity. In addition, according to the results of the FUNGuild classification difference ( ), in the comparison of the Seeding and Tillering of sugarcane growth ( ), there were more classifications with significant differences ( p < 0.05). They were Dung Saprotroph-Soil Saprotroph-Wood Saprotrop, Animal Pathogen–Endophyte–Lichen Parasite–Plant, Orchid Mycorrhizal–Plant Pathogen–Wood Saprotroph, Endophyte–Plant Pathogen, Animal Pathogen–Soil Saprotroph, respectively. However, with the shift in sugarcane stages, the number of significantly different FUNGuild classifications gradually decreased, accompanied by a decrease in the significance of differences between some FUNGuild classifications. ( ).
3.1. Changes in Sugarcane Growth Periods Alter the Composition and Diversity of Soil Properties and Fungal Communities Soil microbial communities associated with plants can have strong influences on plant growth as well as contribute to soil health and sustainable production [ , , ]. Therefore, understanding the temporal progression of rhizosphere microbiota is a prerequisite for plant and soil environmental improvement . Previous research showed that plants’ root microbiota composition varied with plant developmental stage , but these studies were carried out either with other model crops or under greenhouse conditions. Our findings provide a detailed description of the rhizosphere fungal population during the entire sugarcane growth period in the field, as well as insights into how sugarcane growth and the soil environment influence the development of the rhizosphere fungal population. We showed temporal shifts in fungal community composition during the life of sugarcane, and these results are missing in sugarcane-related research in the field ( A,B). Our findings showed that the nutrients in sugarcane rhizosphere soil exhibited significant changes during different growth periods ( p < 0.05), which we assumed was precipitated by the difference in sugarcane requirements for different nutrients at different growth stages . Moreover, although the alpha diversity of the sugarcane rhizosphere fungal community varies significantly between different growth periods, it has a gradually stable trend ( ). It showed that sugarcane has the ability to regulate its own rhizosphere fungal environment to maintain a stable state under the natural growth conditions. Whether this is applicable to other crop systems requires more extensive and in-depth studies. Additionally, the species composition showed that rhizosphere fungi were mainly Ascomycota and Basidiomycota during the growth period of sugarcane ( A), which was similar to the research data of Zeng et al. (2020) . Stursova et al. (2012) also found that compared with Basidiomycota, Ascomycota are more involved in cellulose decomposition . Meanwhile, Chytridiomycota and Glomeromycota, which had a lower relative abundance, also showed regular and significant changes over time ( p < 0.05). Determining the role, if any, of these low-abundance microbes in responding to changes in plant diseases, soil health, and so on will be a fascinating challenge for future studies. The Venn diagram showed that 451 OTUs were shared by sugarcane during the growth periods ( B). The number of OTU unique to each of these periods varies considerably. To further explore, we used the source model of microorganisms to analyse the fungal transmission ratio between stages. Since more than 90% (91–95%, B) of the fungal community were passed to the next period in each stage, it indicated that after the formation of the sugarcane rhizosphere fungal flora, although different microorganisms would be recruited or consumed at different periods, the overall structure would remain stable. Such a result also verifies the idea that plants can maintain resident soil fungal populations but not non-resident soil fungal populations, as has been previously verified in model plants . 3.2. Fungal Taxonomy, Function, and Soil Traits Were Interrelated, and the Transfer of Fungal Communities between Changes in Sugarcane Growth Periods Showed Regularity Additionally, there was a strong correlation between soil environmental factors and sugarcane rhizosphere fungi populations in different growth stages. RDA and network map showed that TN, TK, pH, and AN were the main factors driving sugarcane fungal communities ( A,D). These soil nutrients directly or indirectly affected the survival and growth of sugarcane fungal communities, which is in agreement with the results reported by Zhang et al. (2016). They indicated that organic matter, total N and total P significantly affected soil fungal community composition in the southeastern Tengger Desert . To further explore the influence of these nutrient factors on the fungal community during the growth period of sugarcane, we used the mantel test to calculate the correlation between three matrices and further validated that TN, TK, and pH were still the main environmental factors driving the OTU composition of the fungal community ( C). There were more genera that were decreasing in relative abundance in the rhizosphere over the life cycle of sugarcane, while fewer genera were increasing in relative abundance. It may imply the degree of sugarcane’s control over the rhizosphere fungal population under natural conditions. Additionally, the stabilization of the fungal population makes it difficult to re-enrich once it is lost from the rhizosphere environment, and the difficulty of plant recruitment to soil fungi also may cause this phenomenon ( D,E). This is similar to the results of tracking changes in the rice root microbial flora . These data reinforce the separation role that the growth period plays in distinguishing plant rhizosphere microbial flora as observed in other studies [ , , ], but it needs more research to illustrate. During the whole growth period of sugarcane, there are 10 fungal genera that have been decreasing, mainly including Ascomycota, Chytridiomycota, Eurotiales, Hypocreales, Sordariomycetes, and Tremellales. We speculate that the decline of these fungal genera is closely related to the genotype and developmental stage of sugarcane . This requires us to conduct further verification. Such results provide a list of sugarcane rhizosphere fungi, which should now be targeted to elucidate their potential functions in the roots of sugarcane (symptomless colonizers vs. plant growth-promoters vs. pathogens) by targeted separation and sequencing technology. Furthermore, according to our results, Pseudallescheria had been enriched during the first three growth periods of sugarcane growth (Seeding–Elongation). However, Nectriaceae were significantly ( p < 0.05) increased in the Tillering, Elongation and maturation stages ( A–C). Studies have reported that Pseudallescheria is a filamentous pathogenic fungus. Moreover, it is not only a potential human and animal pathogen, but also exists in the soil environment . Pseudallescheria is of special importance for biological health and whether its significant variation in the rhizosphere could lead to local and disseminated infection of the host is a matter of alarm . Studying the time trend of this pathogenic fungus in the sugarcane rhizosphere has great significance for the defense against certain new diseases that occur in the early growth of sugarcane in the future. Similarly, the ascomycete family Nectriaceae also includes numerous plant and human pathogens . The enrichment of Nectriaceae is presumed to be potentially related to certain diseases that occurred in the late stage of sugarcane growth. However, evidence of pathogenicity does not necessarily exist, which needs to be established through more in-depth and mechanistic studies. Nonetheless, our findings revealed the most basic information and provided the possibility of sugarcane disease research. Additionally, it was found in the co-occurrence network that some fungal interactions between genera disappeared and then reappeared with time, presumably because the “autonomous consciousness” of sugarcane and environmental factors adjusted the balance of fungal interactions. The specific regulatory pathways and the types and sources of metabolites involved in them are not known. Whether this phenomenon is associated with the development of pokkah boeng disease remains to be proven. 3.3. Structural Equation Modeling Demonstrates a Correlation between Key Fungi, Soil Properties, and Sugarcane Disease Rates Structural equation models (SEMs) indicated that soil temperature is a critical factor affecting the α-diversity of rhizosphere fungal communities of sugarcane, which showed significance in all models that we constructed ( and ). Soil temperature changes triggered by seasonal changes had a certain perturbing effect on the rhizosphere community . In addition, the strong direct negative effects of soil pH, AN and TN on the disease status of sugarcane suggest that soil infertile would increase the likelihood of sugarcane disease . At the same time, Fusarium and Talaromyces also responded to changes in the soil nutrient environment, thus interacting adversely with the plant. The high-quality genome sequence of Fusarium andiyazi in China, published by Bao et al. (2021), provides more clarity on the role of Fusarium in sugarcane pokkah boeng disease . 3.4. Rhizosphere Fungal Community Assembly Is Mainly Influenced by Stochastic Factors Furthermore, the results of the rhizosphere fungal community assembly process showed that the majority of the βNTI values among the samples were less than |2| ( B), indicating that the changes observed in the sugarcane rhizosphere fungal community over time were primarily influenced by stochastic factors , but the stochastic contribution increased and then decreased with sugarcane growth and development, eventually reaching the lowest value at the maturity stage ( ). This is due to stochastic changes in the probability distribution and the relative abundance of species of rhizosphere fungi in the early stages of sugarcane growth (ecological drift) , and the formation of an adaptive system of root and soil environment in the later stages of sugarcane growth, resulting in abiotic and biotic factors playing an increasingly pivotal role in the presence or absence and relative abundance of rhizosphere fungi. Furthermore, mutual inhibition between rhizosphere fungi has an effect on community assembly, but the specific mechanism of action and whether it leads to a deterministic process of rhizosphere fungal communities needs to be supported by more research evidence. Phylogenetic trees and FUNGuild functional prediction analysis were applied to the study of sugarcane rhizosphere fungal communities to simplify and visualize the complex community functions and evolutionary relationships ( and ). FUNGuild analysis revealed that the abundance of fungi showed differential responses across stages, for example, Fusarium was abundant in Seeding and Elongation and Talaromyces in Maturity. These results correlate with the work of Li et al. (2021), who reported that some potential biocontrol genera change with plant growth and tillage . While showing dominant species richness, it also provided new ideas for seasonal prediction of rhizosphere pathogens of sugarcane and disease control during the different growth periods, such as annotated phytopathogens ( Cochliobolus ) that directly or indirectly contribute to the disease of sugarcane in a given stage.
Soil microbial communities associated with plants can have strong influences on plant growth as well as contribute to soil health and sustainable production [ , , ]. Therefore, understanding the temporal progression of rhizosphere microbiota is a prerequisite for plant and soil environmental improvement . Previous research showed that plants’ root microbiota composition varied with plant developmental stage , but these studies were carried out either with other model crops or under greenhouse conditions. Our findings provide a detailed description of the rhizosphere fungal population during the entire sugarcane growth period in the field, as well as insights into how sugarcane growth and the soil environment influence the development of the rhizosphere fungal population. We showed temporal shifts in fungal community composition during the life of sugarcane, and these results are missing in sugarcane-related research in the field ( A,B). Our findings showed that the nutrients in sugarcane rhizosphere soil exhibited significant changes during different growth periods ( p < 0.05), which we assumed was precipitated by the difference in sugarcane requirements for different nutrients at different growth stages . Moreover, although the alpha diversity of the sugarcane rhizosphere fungal community varies significantly between different growth periods, it has a gradually stable trend ( ). It showed that sugarcane has the ability to regulate its own rhizosphere fungal environment to maintain a stable state under the natural growth conditions. Whether this is applicable to other crop systems requires more extensive and in-depth studies. Additionally, the species composition showed that rhizosphere fungi were mainly Ascomycota and Basidiomycota during the growth period of sugarcane ( A), which was similar to the research data of Zeng et al. (2020) . Stursova et al. (2012) also found that compared with Basidiomycota, Ascomycota are more involved in cellulose decomposition . Meanwhile, Chytridiomycota and Glomeromycota, which had a lower relative abundance, also showed regular and significant changes over time ( p < 0.05). Determining the role, if any, of these low-abundance microbes in responding to changes in plant diseases, soil health, and so on will be a fascinating challenge for future studies. The Venn diagram showed that 451 OTUs were shared by sugarcane during the growth periods ( B). The number of OTU unique to each of these periods varies considerably. To further explore, we used the source model of microorganisms to analyse the fungal transmission ratio between stages. Since more than 90% (91–95%, B) of the fungal community were passed to the next period in each stage, it indicated that after the formation of the sugarcane rhizosphere fungal flora, although different microorganisms would be recruited or consumed at different periods, the overall structure would remain stable. Such a result also verifies the idea that plants can maintain resident soil fungal populations but not non-resident soil fungal populations, as has been previously verified in model plants .
Additionally, there was a strong correlation between soil environmental factors and sugarcane rhizosphere fungi populations in different growth stages. RDA and network map showed that TN, TK, pH, and AN were the main factors driving sugarcane fungal communities ( A,D). These soil nutrients directly or indirectly affected the survival and growth of sugarcane fungal communities, which is in agreement with the results reported by Zhang et al. (2016). They indicated that organic matter, total N and total P significantly affected soil fungal community composition in the southeastern Tengger Desert . To further explore the influence of these nutrient factors on the fungal community during the growth period of sugarcane, we used the mantel test to calculate the correlation between three matrices and further validated that TN, TK, and pH were still the main environmental factors driving the OTU composition of the fungal community ( C). There were more genera that were decreasing in relative abundance in the rhizosphere over the life cycle of sugarcane, while fewer genera were increasing in relative abundance. It may imply the degree of sugarcane’s control over the rhizosphere fungal population under natural conditions. Additionally, the stabilization of the fungal population makes it difficult to re-enrich once it is lost from the rhizosphere environment, and the difficulty of plant recruitment to soil fungi also may cause this phenomenon ( D,E). This is similar to the results of tracking changes in the rice root microbial flora . These data reinforce the separation role that the growth period plays in distinguishing plant rhizosphere microbial flora as observed in other studies [ , , ], but it needs more research to illustrate. During the whole growth period of sugarcane, there are 10 fungal genera that have been decreasing, mainly including Ascomycota, Chytridiomycota, Eurotiales, Hypocreales, Sordariomycetes, and Tremellales. We speculate that the decline of these fungal genera is closely related to the genotype and developmental stage of sugarcane . This requires us to conduct further verification. Such results provide a list of sugarcane rhizosphere fungi, which should now be targeted to elucidate their potential functions in the roots of sugarcane (symptomless colonizers vs. plant growth-promoters vs. pathogens) by targeted separation and sequencing technology. Furthermore, according to our results, Pseudallescheria had been enriched during the first three growth periods of sugarcane growth (Seeding–Elongation). However, Nectriaceae were significantly ( p < 0.05) increased in the Tillering, Elongation and maturation stages ( A–C). Studies have reported that Pseudallescheria is a filamentous pathogenic fungus. Moreover, it is not only a potential human and animal pathogen, but also exists in the soil environment . Pseudallescheria is of special importance for biological health and whether its significant variation in the rhizosphere could lead to local and disseminated infection of the host is a matter of alarm . Studying the time trend of this pathogenic fungus in the sugarcane rhizosphere has great significance for the defense against certain new diseases that occur in the early growth of sugarcane in the future. Similarly, the ascomycete family Nectriaceae also includes numerous plant and human pathogens . The enrichment of Nectriaceae is presumed to be potentially related to certain diseases that occurred in the late stage of sugarcane growth. However, evidence of pathogenicity does not necessarily exist, which needs to be established through more in-depth and mechanistic studies. Nonetheless, our findings revealed the most basic information and provided the possibility of sugarcane disease research. Additionally, it was found in the co-occurrence network that some fungal interactions between genera disappeared and then reappeared with time, presumably because the “autonomous consciousness” of sugarcane and environmental factors adjusted the balance of fungal interactions. The specific regulatory pathways and the types and sources of metabolites involved in them are not known. Whether this phenomenon is associated with the development of pokkah boeng disease remains to be proven.
Structural equation models (SEMs) indicated that soil temperature is a critical factor affecting the α-diversity of rhizosphere fungal communities of sugarcane, which showed significance in all models that we constructed ( and ). Soil temperature changes triggered by seasonal changes had a certain perturbing effect on the rhizosphere community . In addition, the strong direct negative effects of soil pH, AN and TN on the disease status of sugarcane suggest that soil infertile would increase the likelihood of sugarcane disease . At the same time, Fusarium and Talaromyces also responded to changes in the soil nutrient environment, thus interacting adversely with the plant. The high-quality genome sequence of Fusarium andiyazi in China, published by Bao et al. (2021), provides more clarity on the role of Fusarium in sugarcane pokkah boeng disease .
Furthermore, the results of the rhizosphere fungal community assembly process showed that the majority of the βNTI values among the samples were less than |2| ( B), indicating that the changes observed in the sugarcane rhizosphere fungal community over time were primarily influenced by stochastic factors , but the stochastic contribution increased and then decreased with sugarcane growth and development, eventually reaching the lowest value at the maturity stage ( ). This is due to stochastic changes in the probability distribution and the relative abundance of species of rhizosphere fungi in the early stages of sugarcane growth (ecological drift) , and the formation of an adaptive system of root and soil environment in the later stages of sugarcane growth, resulting in abiotic and biotic factors playing an increasingly pivotal role in the presence or absence and relative abundance of rhizosphere fungi. Furthermore, mutual inhibition between rhizosphere fungi has an effect on community assembly, but the specific mechanism of action and whether it leads to a deterministic process of rhizosphere fungal communities needs to be supported by more research evidence. Phylogenetic trees and FUNGuild functional prediction analysis were applied to the study of sugarcane rhizosphere fungal communities to simplify and visualize the complex community functions and evolutionary relationships ( and ). FUNGuild analysis revealed that the abundance of fungi showed differential responses across stages, for example, Fusarium was abundant in Seeding and Elongation and Talaromyces in Maturity. These results correlate with the work of Li et al. (2021), who reported that some potential biocontrol genera change with plant growth and tillage . While showing dominant species richness, it also provided new ideas for seasonal prediction of rhizosphere pathogens of sugarcane and disease control during the different growth periods, such as annotated phytopathogens ( Cochliobolus ) that directly or indirectly contribute to the disease of sugarcane in a given stage.
4.1. Sugarcane Planting In the spring of 2017, sugarcane cultivar, FN 41 was grown in a separate sugarcane experimental field in China to track the rhizosphere fungal population change procedure during the entire sugarcane growth cycle. Before planting, sugarcane seed-canes were soaked in 0.1% carbendazim ( w / v ) for 10 min to avoid surface-associated microbes followed by soaking in water for 24 h . After soaking seed-canes, they were transferred to the fields at the Baisha experimental station (119°06′ E, 26°23′ N). Before the start of the experiment, the same variety of sugarcane crop was grown in the test plot for a period of three years. Abundant heat resources in the test area to supply sugarcane growth. The field growth of sugarcane was managed uniformly according to the farming conditions of local growers. A randomized group design was used, with an area of 1008 m 2 , 8 rows per zone, 5 m row length, 1.2 m row spacing, 48 m 2 per plot, and 3 replications, distributed in randomized groups in each plot. The planting height of sugarcane was 1/2 of the row height. Twenty-one sugarcane stalks were randomly selected in each field and measured with a measuring tape and vernier caliper to determine stalk height and stem diameter in each period. Meanwhile, the number of effective plants, diseased plants and total plants were also counted. 4.2. Sample Collection The experiment started on the 3 January 2017. The soil samples were collected at the Baisha Experimental Station of the National Sugarcane Research Center of Fujian Agriculture and Forestry University (subtropical monsoon climate, the annual average temperature is 19.5 °C, the annual average precipitation is 1673.9 mm). At different fertility stages, we used the “S-shaped sampling method” to select 21 sampling points in the sugarcane field . According to the shaking-off method of Riley and Barber , the soil adhered to the sugarcane roots was brushed with a small sterile brush, and soil samples were finally collected, sieved at 2.0 mm, and stored in a refrigerator at −20 °C . The samples were collected during the sugarcane seeding stage (Seeding 4, 2017), tillering period (Tillering 2, 2017), elongation period (Elongation 17, 2017), and maturity period (Maturity 29, 2017), with a total of 84 samples (4 periods, 21 repetitions). 4.3. Determination of Pokkah Boeng Disease of Sugarcane and Soil Physio-Chemical Properties The physicochemical indicators in the rhizosphere soil that had a strong influence on the sugarcane and varied greatly during the succession of growth periods were selected for measurement. Soil suspension with water (1:2.5 WV −1 ) was prepared to estimate soil pH using a pH meter (PHS-3C, INESA Scientific Instrument Co., Ltd., Shanghai, China) . The soil temperature (Tem) was measured with the Soil Temperature Detector (Model: JC-TW, Shandong, China). The available nitrogen was measured using the alkaline hydrolyzable diffusion method , Soil total nitrogen and organic matter were determined by Kjeldahl digestion and determined by the oil bath–K 2 CrO 7 titration method . C:N is the ratio of soil total N to organic matter. Soil total potassium and total phosphorus were determined by digestion with HF-HClO 4 , followed by flame photometry and molybdenum-blue colorimetry, respectively . Available potassium was extracted by ammonium acetate and determined by flame photometry . Available phosphorus was extracted by sodium bicarbonate and determined using the molybdenum blue method . Field judgment of pokkah boeng disease of sugarcane was mainly based on the symptoms described in previous studies and divided into three types: the chlorotic phase, the acute phase or top-rot phase and the knife-cut phase (associate with top rot phase). The effective number of stems/ha = Effective number of stems/m 2 × 666.67/average row spacing. 4.4. DNA Extraction and PCR Amplification For 84 soil samples, DNA was extracted using a Power Soil DNA Isolation Kit (MoBio Laboratories Inc., Carlsbad, CA, USA) according to the manufacturer’s instructions. A NanoDrop 2000 spectrophotometer (Thermo Scientific, Waltham, MA, USA) was employed to assess the concentration and quality of DNA. Amplification of 18S rDNA gene fragments was carried out using primers set SSU0817F (5′-TTAGCATGGAATAATRRAATAGGA-3′)/SSU1196R(5′-TCTGGACCTGGTGAGTTTCC-3′) . The reaction conditions used for DNA amplification were: 95 °C for 3 min, followed by 35 cycles of 95 °C for 30 s, 55 °C for 30 s, and 72 °C for 45 s, with a final extension at 72 °C for 10 min (GeneAmp 9700, ABI, Foster City, CA, USA). PCR reactions were carried out in triplicate in a 20 μL mixture containing 2 μL of 2.5 mM sNTPs, 4 μL of 5× Fast Pfu buffer, 0.4 μL of Fast Pfu polymerase, 0.4 μL of each primer (5 μM), and template DNA (10 ng). Extraction of amplicons was carried out using an AxyPrep DNA Gel Extraction Kit (Axygen Biosciences, Union City, CA, USA). Later, the DNA was quantified using QuantiFluor™-ST (Promega, Madison, WI, USA). Purified amplicons were pooled in equimolar and paired-end sequenced (2 × 250) on an Illumina MiSeq platform (Majorbio, Shanghai, China) following the standard procedures. 4.5. Data Quality Control and Filtering The sequencing extracted data were saved in fastq format, while the paired reads were spliced (merged) into one sequence based on the overlap between PE reads, and quality control was performed on the read quality and merging effect. Valid sequences were obtained from the barcodes and primers at the first and last ends of the sequences, and then the double-ended sequences were spliced (Flash, 1.2.11) to generate an abundance table for each taxonomy (QIIME, 1.9.1). The complete data sets generated in our study have been deposited in the NCBI Sequence Read Archive database under BioProject ID PRJNA721464. 4.6. Sequences and Statistical Analysis The UPARSE standard pipeline was used to analyze the sequence data . Briefly, sequences with short reads (<250 bp) were filtered out before for downstream analysis. Sequences with ≥97% similarity were clustered into OTUs. All sequences were assigned using the RDP classifier to identify taxa with a confidence threshold of 0.8 . We selected these OTU with 97% similarity, and then calculated the Alpha diversity index under different random sampling using Mothur . Later, we used R to draw the rarefaction curves. The DPS software was used to analyze the variance of the soil physical and chemical properties, and the significance was calculated based on the Bonferroni test (FDR adjusted p < 0.05) . RDA was used to visualize the relationship between fungal communities and soil environmental factors. Network analysis was performed using R to calculate the correlation between the factors (Spearman correlation), and Cytoscape (version 3.6.1) was used to adjust and visualize the results . Analysis of differential OTU abundance and taxa was performed using a DESeq2 of the R package, and then we used a Manhattan plot to visualize the results (R 3.6.0). Structural equation modeling (SEM) was performed using IBM SPSS Amos 26. Maximum likelihood estimation with standard errors was also used . The nearest-taxon index (NTI) and βNTI (999 random) were used to quantify changes in rhizosphere fungal phylogeny over time, and the two indices were calculated using the package “picante” . The heatmap and functional annotation and maximum likelihood trees were created using the majorbio platform ( http://cloud.majorbio.com , accessed on 12 September 2022). FUNGuild classification map was completed using STAMP (version 2.1.1), comparing the two stages, using Welch’s t-test, 95% confidence intervals, p < 0.05. Network diagramming and parameter were calculated using R, Cytoscape and UCINET 6 together .
In the spring of 2017, sugarcane cultivar, FN 41 was grown in a separate sugarcane experimental field in China to track the rhizosphere fungal population change procedure during the entire sugarcane growth cycle. Before planting, sugarcane seed-canes were soaked in 0.1% carbendazim ( w / v ) for 10 min to avoid surface-associated microbes followed by soaking in water for 24 h . After soaking seed-canes, they were transferred to the fields at the Baisha experimental station (119°06′ E, 26°23′ N). Before the start of the experiment, the same variety of sugarcane crop was grown in the test plot for a period of three years. Abundant heat resources in the test area to supply sugarcane growth. The field growth of sugarcane was managed uniformly according to the farming conditions of local growers. A randomized group design was used, with an area of 1008 m 2 , 8 rows per zone, 5 m row length, 1.2 m row spacing, 48 m 2 per plot, and 3 replications, distributed in randomized groups in each plot. The planting height of sugarcane was 1/2 of the row height. Twenty-one sugarcane stalks were randomly selected in each field and measured with a measuring tape and vernier caliper to determine stalk height and stem diameter in each period. Meanwhile, the number of effective plants, diseased plants and total plants were also counted.
The experiment started on the 3 January 2017. The soil samples were collected at the Baisha Experimental Station of the National Sugarcane Research Center of Fujian Agriculture and Forestry University (subtropical monsoon climate, the annual average temperature is 19.5 °C, the annual average precipitation is 1673.9 mm). At different fertility stages, we used the “S-shaped sampling method” to select 21 sampling points in the sugarcane field . According to the shaking-off method of Riley and Barber , the soil adhered to the sugarcane roots was brushed with a small sterile brush, and soil samples were finally collected, sieved at 2.0 mm, and stored in a refrigerator at −20 °C . The samples were collected during the sugarcane seeding stage (Seeding 4, 2017), tillering period (Tillering 2, 2017), elongation period (Elongation 17, 2017), and maturity period (Maturity 29, 2017), with a total of 84 samples (4 periods, 21 repetitions).
The physicochemical indicators in the rhizosphere soil that had a strong influence on the sugarcane and varied greatly during the succession of growth periods were selected for measurement. Soil suspension with water (1:2.5 WV −1 ) was prepared to estimate soil pH using a pH meter (PHS-3C, INESA Scientific Instrument Co., Ltd., Shanghai, China) . The soil temperature (Tem) was measured with the Soil Temperature Detector (Model: JC-TW, Shandong, China). The available nitrogen was measured using the alkaline hydrolyzable diffusion method , Soil total nitrogen and organic matter were determined by Kjeldahl digestion and determined by the oil bath–K 2 CrO 7 titration method . C:N is the ratio of soil total N to organic matter. Soil total potassium and total phosphorus were determined by digestion with HF-HClO 4 , followed by flame photometry and molybdenum-blue colorimetry, respectively . Available potassium was extracted by ammonium acetate and determined by flame photometry . Available phosphorus was extracted by sodium bicarbonate and determined using the molybdenum blue method . Field judgment of pokkah boeng disease of sugarcane was mainly based on the symptoms described in previous studies and divided into three types: the chlorotic phase, the acute phase or top-rot phase and the knife-cut phase (associate with top rot phase). The effective number of stems/ha = Effective number of stems/m 2 × 666.67/average row spacing.
For 84 soil samples, DNA was extracted using a Power Soil DNA Isolation Kit (MoBio Laboratories Inc., Carlsbad, CA, USA) according to the manufacturer’s instructions. A NanoDrop 2000 spectrophotometer (Thermo Scientific, Waltham, MA, USA) was employed to assess the concentration and quality of DNA. Amplification of 18S rDNA gene fragments was carried out using primers set SSU0817F (5′-TTAGCATGGAATAATRRAATAGGA-3′)/SSU1196R(5′-TCTGGACCTGGTGAGTTTCC-3′) . The reaction conditions used for DNA amplification were: 95 °C for 3 min, followed by 35 cycles of 95 °C for 30 s, 55 °C for 30 s, and 72 °C for 45 s, with a final extension at 72 °C for 10 min (GeneAmp 9700, ABI, Foster City, CA, USA). PCR reactions were carried out in triplicate in a 20 μL mixture containing 2 μL of 2.5 mM sNTPs, 4 μL of 5× Fast Pfu buffer, 0.4 μL of Fast Pfu polymerase, 0.4 μL of each primer (5 μM), and template DNA (10 ng). Extraction of amplicons was carried out using an AxyPrep DNA Gel Extraction Kit (Axygen Biosciences, Union City, CA, USA). Later, the DNA was quantified using QuantiFluor™-ST (Promega, Madison, WI, USA). Purified amplicons were pooled in equimolar and paired-end sequenced (2 × 250) on an Illumina MiSeq platform (Majorbio, Shanghai, China) following the standard procedures.
The sequencing extracted data were saved in fastq format, while the paired reads were spliced (merged) into one sequence based on the overlap between PE reads, and quality control was performed on the read quality and merging effect. Valid sequences were obtained from the barcodes and primers at the first and last ends of the sequences, and then the double-ended sequences were spliced (Flash, 1.2.11) to generate an abundance table for each taxonomy (QIIME, 1.9.1). The complete data sets generated in our study have been deposited in the NCBI Sequence Read Archive database under BioProject ID PRJNA721464.
The UPARSE standard pipeline was used to analyze the sequence data . Briefly, sequences with short reads (<250 bp) were filtered out before for downstream analysis. Sequences with ≥97% similarity were clustered into OTUs. All sequences were assigned using the RDP classifier to identify taxa with a confidence threshold of 0.8 . We selected these OTU with 97% similarity, and then calculated the Alpha diversity index under different random sampling using Mothur . Later, we used R to draw the rarefaction curves. The DPS software was used to analyze the variance of the soil physical and chemical properties, and the significance was calculated based on the Bonferroni test (FDR adjusted p < 0.05) . RDA was used to visualize the relationship between fungal communities and soil environmental factors. Network analysis was performed using R to calculate the correlation between the factors (Spearman correlation), and Cytoscape (version 3.6.1) was used to adjust and visualize the results . Analysis of differential OTU abundance and taxa was performed using a DESeq2 of the R package, and then we used a Manhattan plot to visualize the results (R 3.6.0). Structural equation modeling (SEM) was performed using IBM SPSS Amos 26. Maximum likelihood estimation with standard errors was also used . The nearest-taxon index (NTI) and βNTI (999 random) were used to quantify changes in rhizosphere fungal phylogeny over time, and the two indices were calculated using the package “picante” . The heatmap and functional annotation and maximum likelihood trees were created using the majorbio platform ( http://cloud.majorbio.com , accessed on 12 September 2022). FUNGuild classification map was completed using STAMP (version 2.1.1), comparing the two stages, using Welch’s t-test, 95% confidence intervals, p < 0.05. Network diagramming and parameter were calculated using R, Cytoscape and UCINET 6 together .
When focusing on soil productivity, rhizosphere microbiomes are a critical factor and closely related to plant health, especially the fungal community. Under the background conditions of this study, both the growth stages and soil characteristics of sugarcane had significant effects on the composition and function of the rhizosphere fungal community. The pH, TN, TK and AN decreased significantly during the growth of sugarcane. TN, TK, and AN were significantly decreased by 12.41%, 26.11%, and 31.32%, respectively, compared to the early growth stage. The dominant fungal phyla were Ascomycota, Basidiomycota, and Chytridiomycota, and the dominance of dominant fungal groups can be explained by both the temporal dynamics of sugarcane roots and the awareness of plant-autonomous regulation. We also discovered that diseases had significant and strong negative effects on selected soil traits (pH, Tem, and AN) using SEM, while soil temperature had a strong and direct positive effect on fungal α-diversity, and Fusarium and Talaromyces had some direct and indirect associations with sugarcane diseases. It indicates soil nutrients affected plant health and rhizosphere fungal interactions. The variation in rhizosphere fungal structure was mainly influenced by stochastic factors, and the decrease in stochastic contribution rate was hypothesized to be due to the coupling of sensitive fungal genera or modules in the rhizosphere fungal network with sugarcane growth dynamics to coordinate the growth process. These results contribute to the understanding of plant–rhizosphere fungal interactions and provide additional opportunities for the development of green agriculture and preventive tools for crop pests and diseases.
|
Common Variants Near ZIC1 and ZIC4 in Autopsy-Confirmed Multiple System Atrophy | bfdfe07c-2e41-4466-b2c8-84c440ada6a6 | 10052809 | Forensic Medicine[mh] | Patient Recruitment Ethical approval had been obtained from all responsible ethics committees. All participants had given written consent. Neuropathologists at each recruitment site ( ) based the definite neuropathological diagnosis of MSA on histopathological criteria, taking into account glial cytoplasmic inclusions immunoreactive for α-synuclein in characteristic anatomical distribution as a defining feature of MSA. Age, sex, disease history (including disease onset and duration), and neuropathological findings were recorded in a standardized manner for all cases. Controls were ethnically matched to cases and either derived from biobanks KORA-gen or popGen (Europe sites) or from a North American site (Alzheimer’s Disease Genetics Consortium). The Alzheimer’s Disease Genetics Consortium assembled and genotyped DNA from subjects enrolled in the 29 NIA-Alzheimer’s Disease Centers located across the United States. For this study, the Alzheimer’s Disease Genetics Consortium provided a subset of mostly clinical, cognitively normal controls. Patients and controls were of North-Western European and African American ancestry. DNA Extraction We isolated DNA from 30 mg frozen cerebellar cortex using QIAamp DNA Mini Kit (Qiagen, Venlo, the Netherlands). DNA extraction was performed at German Center for Neurodegenerative Diseases (DZNE, Munich, Germany). DNA was stored at −80°C until use. DNA concentration was measured using a NanoDrop Spectro-photometer. DNA quality was determined by gel electrophoresis. Genotyping All samples were genotyped on Infinium Global Screening Arrays (Illumina, San Diego, CA, USA). The cases were genotyped at the Institute of Clinical Molecular Biology, Kiel University, Germany. The samples were genotyped in one batch on array version 2.0 for cases and version 1.0 for controls. Genotypes were called using Illumina Genome studio according to the manufacturer’s instructions using in-house cluster files. Quality Control and Imputation We used PLINK (v. 1.9) [1] and R (v. 3.6.3) for all analyses. Only variants successfully genotyped in both the patient and the control populations were included in the subsequent analyses. Variants with multicharacter allele codes, insertions, deletions, duplicated markers, and all A/T and G/C variants were excluded. We excluded all samples discordant between reported and genotypic sex. Missing sex was imputed, and samples with ambiguously imputed sex were discarded. After a first step of filtering out samples and variants with call rate of less than 85%, we excluded variants with an individual call rate of less than 98% in a second filtering step. Next, we removed variants with a minor allele frequency <0.01, a significant deviation from Hardy–Weinberg equilibrium ( P < 1 × 10 −6 ) in controls, or informative missingness ( P < 1 × 10 −5 ). Subsequently, we excluded individuals with a variant call rate of <98% or an outlying heterozygosity rate (mean ± 3 standard deviations). We used a pruned dataset containing only markers in low linkage-disequilibrium regions (pairwise r 2 < 0.2) to test for duplicated individuals and cryptic relatedness (Pihat > 0.125) using pairwise genome-wide estimates of the proportion of identity by descent. For each detected sample pair we excluded the individual with a lower call rate. Ethnical outliers were identified by a principal-component analysis (PCA) together with the publically available 1000 Genomes data with known ethnicities. Because the study population has genetically a mainly European ancestry, as ascertained by the PCA, we determined a European center and excluded samples more than 1.5 times the maximal European Euclidean distance away from this center. After a first association analysis of genotyped single nucleotide polymorphisms (SNPs) only, we inspected visually the cluster plots of all variants with a P value <1 × 10 −5 and discarded variants without adequate cluster separation. Imputation was carried out on the quality-assured dataset using the TOPMed Imputation Server, which employs Eagle2 for phasing and minimac4 for the imputation of genotypes. , The most likely genotype is used in downstream analyses. Variants were again filtered for minor allele frequency and deviation from Hardy–Weinberg equilibrium in controls with the same thresholds as before. In addition, SNPs with an imputation quality score R 2 < 0.7 were excluded, leaving 8,131,900 variants for analyses. As a final step of the quality-control procedure, we used the R package PCAmatchR to ethnically match cases to controls with a 1:4 ratio to overcome possible difficulties with population stratification, leading to 3240 individuals for the analyses. Association Analysis We used logistic regression to test the additive genetic model of each marker for association with disease status. Following scree plot analysis, we incorporated the first two dimensions of the PCA and sex as covariates. We used a genome-wide significance threshold of P < 5 × 10 −8 and the threshold of P < 5 × 10 −6 for suggestive association. Conditional analyses, including, in turn, each SNP with a suggestive association as additional covariate, were conducted to identify adjacent independent signals. Furthermore, we tested for clumps of correlated SNPs, ie, to assess how many independent loci had been associated, and determined the number of variants supporting the lead SNP at each locus, ie, variants with P values less than the clumping threshold of 5 × 10 −5 are in linkage disequilibrium ( r 2 ≥ 0.4) and not farther than 250 kb away from the respective SNP. Visualization of the results was carried out with R and LocusZoom for regional plots. Variant positions in this article are reported on human genome version 38 (GRCh38/hg38). Immunohistochemistry on MSA Patients’ Brain Formalin-fixed and paraffin-embedded (FFPE) tissues from patients with MSA and controls without neurological or psychiatric diseases were obtained from the Neurobiobank Munich (Germany). All autopsy cases of the Neurobiobank Munich were collected on the basis of an informed consent according to the guidelines of the ethics commission of the Ludwig-Maximilians-University (Munich, Germany; #345–13). MSA cases had been diagnosed according to established histopathological diagnostic criteria. , For ZIC4 immunohistochemistry, 5-μm-thick sections of FFPE tissues of the frontal cortex and the cerebellar hemisphere, including the dentate nucleus, were prepared. After deparaffinization, heat-induced epitope retrieval was performed in Tris/EDTA, pH 9, at 95°C for 30 minutes. For blocking of endogenous peroxidase and unspecific protein binding, the sections were incubated with 5% H 2 O 2 in methanol for 20 minutes and I-Block reagent (Applied Biosystems, Waltham, MA, USA) for 15 minutes. Subsequently, ZIC4 primary antibody (rabbit, polyclonal; Merck/Sigma-Aldrich, Darmstadt, Germany) was applied overnight at 4°C at a dilution of 1:100. Signal detection was performed using the DCS Chromo Line DAB kit (DCS, Hamburg, Germany) according to the manufacturer’s instructions. Sections were counterstained for 1 minute with Mayer’s hemalum solution (Waldeck, Münster, Germany). To determine the fractions of ZIC4-positive neurons of all neurons in the dentate nucleus, we scanned stained slides using a slide scanner (Axio Scan. Z1; Zeiss, Oberkochen, Germany) and visualized using the free ZEN lite software (v. 3.3; Zeiss). For statistical evaluation of the data, Student t test was used, and statistical significance was defined as P < 0.05.
Ethical approval had been obtained from all responsible ethics committees. All participants had given written consent. Neuropathologists at each recruitment site ( ) based the definite neuropathological diagnosis of MSA on histopathological criteria, taking into account glial cytoplasmic inclusions immunoreactive for α-synuclein in characteristic anatomical distribution as a defining feature of MSA. Age, sex, disease history (including disease onset and duration), and neuropathological findings were recorded in a standardized manner for all cases. Controls were ethnically matched to cases and either derived from biobanks KORA-gen or popGen (Europe sites) or from a North American site (Alzheimer’s Disease Genetics Consortium). The Alzheimer’s Disease Genetics Consortium assembled and genotyped DNA from subjects enrolled in the 29 NIA-Alzheimer’s Disease Centers located across the United States. For this study, the Alzheimer’s Disease Genetics Consortium provided a subset of mostly clinical, cognitively normal controls. Patients and controls were of North-Western European and African American ancestry.
We isolated DNA from 30 mg frozen cerebellar cortex using QIAamp DNA Mini Kit (Qiagen, Venlo, the Netherlands). DNA extraction was performed at German Center for Neurodegenerative Diseases (DZNE, Munich, Germany). DNA was stored at −80°C until use. DNA concentration was measured using a NanoDrop Spectro-photometer. DNA quality was determined by gel electrophoresis.
All samples were genotyped on Infinium Global Screening Arrays (Illumina, San Diego, CA, USA). The cases were genotyped at the Institute of Clinical Molecular Biology, Kiel University, Germany. The samples were genotyped in one batch on array version 2.0 for cases and version 1.0 for controls. Genotypes were called using Illumina Genome studio according to the manufacturer’s instructions using in-house cluster files.
We used PLINK (v. 1.9) [1] and R (v. 3.6.3) for all analyses. Only variants successfully genotyped in both the patient and the control populations were included in the subsequent analyses. Variants with multicharacter allele codes, insertions, deletions, duplicated markers, and all A/T and G/C variants were excluded. We excluded all samples discordant between reported and genotypic sex. Missing sex was imputed, and samples with ambiguously imputed sex were discarded. After a first step of filtering out samples and variants with call rate of less than 85%, we excluded variants with an individual call rate of less than 98% in a second filtering step. Next, we removed variants with a minor allele frequency <0.01, a significant deviation from Hardy–Weinberg equilibrium ( P < 1 × 10 −6 ) in controls, or informative missingness ( P < 1 × 10 −5 ). Subsequently, we excluded individuals with a variant call rate of <98% or an outlying heterozygosity rate (mean ± 3 standard deviations). We used a pruned dataset containing only markers in low linkage-disequilibrium regions (pairwise r 2 < 0.2) to test for duplicated individuals and cryptic relatedness (Pihat > 0.125) using pairwise genome-wide estimates of the proportion of identity by descent. For each detected sample pair we excluded the individual with a lower call rate. Ethnical outliers were identified by a principal-component analysis (PCA) together with the publically available 1000 Genomes data with known ethnicities. Because the study population has genetically a mainly European ancestry, as ascertained by the PCA, we determined a European center and excluded samples more than 1.5 times the maximal European Euclidean distance away from this center. After a first association analysis of genotyped single nucleotide polymorphisms (SNPs) only, we inspected visually the cluster plots of all variants with a P value <1 × 10 −5 and discarded variants without adequate cluster separation. Imputation was carried out on the quality-assured dataset using the TOPMed Imputation Server, which employs Eagle2 for phasing and minimac4 for the imputation of genotypes. , The most likely genotype is used in downstream analyses. Variants were again filtered for minor allele frequency and deviation from Hardy–Weinberg equilibrium in controls with the same thresholds as before. In addition, SNPs with an imputation quality score R 2 < 0.7 were excluded, leaving 8,131,900 variants for analyses. As a final step of the quality-control procedure, we used the R package PCAmatchR to ethnically match cases to controls with a 1:4 ratio to overcome possible difficulties with population stratification, leading to 3240 individuals for the analyses.
We used logistic regression to test the additive genetic model of each marker for association with disease status. Following scree plot analysis, we incorporated the first two dimensions of the PCA and sex as covariates. We used a genome-wide significance threshold of P < 5 × 10 −8 and the threshold of P < 5 × 10 −6 for suggestive association. Conditional analyses, including, in turn, each SNP with a suggestive association as additional covariate, were conducted to identify adjacent independent signals. Furthermore, we tested for clumps of correlated SNPs, ie, to assess how many independent loci had been associated, and determined the number of variants supporting the lead SNP at each locus, ie, variants with P values less than the clumping threshold of 5 × 10 −5 are in linkage disequilibrium ( r 2 ≥ 0.4) and not farther than 250 kb away from the respective SNP. Visualization of the results was carried out with R and LocusZoom for regional plots. Variant positions in this article are reported on human genome version 38 (GRCh38/hg38).
Formalin-fixed and paraffin-embedded (FFPE) tissues from patients with MSA and controls without neurological or psychiatric diseases were obtained from the Neurobiobank Munich (Germany). All autopsy cases of the Neurobiobank Munich were collected on the basis of an informed consent according to the guidelines of the ethics commission of the Ludwig-Maximilians-University (Munich, Germany; #345–13). MSA cases had been diagnosed according to established histopathological diagnostic criteria. , For ZIC4 immunohistochemistry, 5-μm-thick sections of FFPE tissues of the frontal cortex and the cerebellar hemisphere, including the dentate nucleus, were prepared. After deparaffinization, heat-induced epitope retrieval was performed in Tris/EDTA, pH 9, at 95°C for 30 minutes. For blocking of endogenous peroxidase and unspecific protein binding, the sections were incubated with 5% H 2 O 2 in methanol for 20 minutes and I-Block reagent (Applied Biosystems, Waltham, MA, USA) for 15 minutes. Subsequently, ZIC4 primary antibody (rabbit, polyclonal; Merck/Sigma-Aldrich, Darmstadt, Germany) was applied overnight at 4°C at a dilution of 1:100. Signal detection was performed using the DCS Chromo Line DAB kit (DCS, Hamburg, Germany) according to the manufacturer’s instructions. Sections were counterstained for 1 minute with Mayer’s hemalum solution (Waldeck, Münster, Germany). To determine the fractions of ZIC4-positive neurons of all neurons in the dentate nucleus, we scanned stained slides using a slide scanner (Axio Scan. Z1; Zeiss, Oberkochen, Germany) and visualized using the free ZEN lite software (v. 3.3; Zeiss). For statistical evaluation of the data, Student t test was used, and statistical significance was defined as P < 0.05.
Patient Sample From the initial sample of 731 cases, 13 cases had to be excluded because of insufficient tissue quality. After thorough quality control and filtering, 648 cases and 2592 controls covering 8,131,900 variants were included in the association analysis ( ). The number of excluded samples and variants in each step of the quality-control procedure is shown in and . Association Results We performed logistic regression incorporating sex and determined the first two dimensions of PCA as covariates using the scree plot method. The genomic inflation factor of λ = 1.01 (unimputed λ = 1.01; ) indicates that no significant population stratification was present ( ). We did not identify any disease-associated variants with a P value less than the genome-wide significance threshold of P < 5 × 10 −8 , but suggestive associations with P < 5 × 10 −6 at 10 different loci ( ) with the leading SNP at each locus shown in . Conditional analyses, including, in turn, any SNP with P < 5 × 10 −6 , excluded the presence of multiple independent signals at each locus. All variants with suggestive associations are listed in . The most noteworthy hits were rs16859966 on chromosome 3 ( P = 8.6 × 10 −7 ; odds ratio [OR], 1.58; 95% confidence interval [CI]: 1.32–1.89), rs7013955 on chromosome 8 ( P = 3.7 × 10 −6 ; OR, 1.8; 95% CI: 1.40–2.31), and rs116607983 on chromosome 4 ( P = 4.0 × 10 −6 ; OR, 2.93; 95% CI: 1.86–4.63), which were supported by at least one additional genotype, as well as several imputed SNPs with P values less than the clumping threshold of 5 × 10 −5 as discovered in the clumping analysis ( ). The genes closest to the chromosome 3 locus are the Long Intergenic Non-Protein Coding RNA 2032 ( LINC02032 ) approximately 100 kb downstream and the zinc-finger proteins of cerebellum 1 and 4 genes ( ZIC1, ZIC4 ), located roughly 600 kb upstream ( ). The top SNP rs7013955 on chromosome 8 maps to the lysyl oxidaselike 2 gene ( LOXL2 ; ). The association signal around SNP rs116607983 on chromosome 4 is located in a region devoid of protein-coding genes approximately 2000 kb to either side ( ). A fourth locus on chromosome 5 (rs2279135) was also supported by multiple clumped SNPs, but all SNPs, including the lead SNP, were imputed ( ). Several variants clumped at the chromosome 5 locus were located in the ARHGEF37 gene, coding for Rho Guanine Nucleotide Exchange Factor 37 ( ). None of the identified SNPs is an expression quantitative trait locus in brain tissues according the Genotype Tissue Expression project. At four of the six remaining loci with variants exhibiting suggestive associations, at most two supporting SNPs were present, which were all imputed; in the other two loci, no supporting SNPs could be found in the clumping analysis ( , ). We did not investigate these loci further because it is unlikely that they represent valid associations. No significant associations with Bonferroni-adjusted P values were detected with previously reported Parkinson’s disease associations from a meta-analysis of 17 datasets from a Parkinson’s disease GWAS ( ). ZIC4 Immunohistochemistry on MSA Patients’ Brain ZIC4 and ZIC1 are known to play a critical role in the embryonal development of the cerebellum. Heterozygous deletions comprising the ZIC1 and ZIC4 locus have been associated with the Dandy–Walker malformation, a rare congenital condition characterized by a hypoplastic cerebellar vermis and an enlarged fourth ventricle. , In mice, deletions of ZIC1 and ZIC4 lead to a striking phenotype similar to the Dandy–Walker malformation with cerebellar hypoplasia and foliation defects. , In addition, paraneoplastic autoantibodies against ZIC4 protein are linked to severe cerebellar dys-function and degeneration. , Because cerebellar degeneration and corresponding symptoms are also a central hallmark of MSA, we decided to follow up on a potential role of ZIC4 in MSA patient brains by performing immunohistochemical stainings. For ZIC1, no primary antibody was appropriately sensitive and specific on human tissue in our hands. Thus, FFPE tissues of the cerebellum and, for comparison, the frontal cortex of patients with MSA (n = 10 SND, n = 14 OPCA/mixed phenotype) and healthy controls (b = 5) were stained with antibodies raised against ZIC4. Nuclear and cytoplasmic staining of frontal cortex neurons was observed in all brains examined without differences between healthy controls and patients with MSA ( – ). In the cerebellar dentate nucleus, we found strong expression of ZIC4 in a subset of neurons in healthy controls, as well as patients with MSA with predominant SND ( , , , ). In contrast, patients with MSA with mixed subtype or OPCA showed reduced numbers of ZIC4-positive neurons, which were furthermore only weakly stained ( , ). Quantification of the proportions of ZIC4-positive neurons among the total number of dentate nucleus neurons depicted relatively constant proportions in healthy controls and patients with MSA-SND (33.2% ± 0.0% vs 32.6% 0.0%), whereas in patients with MSA-OPCA or MSA-mixed phenotype, we found significantly lower percentages of ZIC4-positive neurons (15.5% ± 0.1%) ( ).
From the initial sample of 731 cases, 13 cases had to be excluded because of insufficient tissue quality. After thorough quality control and filtering, 648 cases and 2592 controls covering 8,131,900 variants were included in the association analysis ( ). The number of excluded samples and variants in each step of the quality-control procedure is shown in and .
We performed logistic regression incorporating sex and determined the first two dimensions of PCA as covariates using the scree plot method. The genomic inflation factor of λ = 1.01 (unimputed λ = 1.01; ) indicates that no significant population stratification was present ( ). We did not identify any disease-associated variants with a P value less than the genome-wide significance threshold of P < 5 × 10 −8 , but suggestive associations with P < 5 × 10 −6 at 10 different loci ( ) with the leading SNP at each locus shown in . Conditional analyses, including, in turn, any SNP with P < 5 × 10 −6 , excluded the presence of multiple independent signals at each locus. All variants with suggestive associations are listed in . The most noteworthy hits were rs16859966 on chromosome 3 ( P = 8.6 × 10 −7 ; odds ratio [OR], 1.58; 95% confidence interval [CI]: 1.32–1.89), rs7013955 on chromosome 8 ( P = 3.7 × 10 −6 ; OR, 1.8; 95% CI: 1.40–2.31), and rs116607983 on chromosome 4 ( P = 4.0 × 10 −6 ; OR, 2.93; 95% CI: 1.86–4.63), which were supported by at least one additional genotype, as well as several imputed SNPs with P values less than the clumping threshold of 5 × 10 −5 as discovered in the clumping analysis ( ). The genes closest to the chromosome 3 locus are the Long Intergenic Non-Protein Coding RNA 2032 ( LINC02032 ) approximately 100 kb downstream and the zinc-finger proteins of cerebellum 1 and 4 genes ( ZIC1, ZIC4 ), located roughly 600 kb upstream ( ). The top SNP rs7013955 on chromosome 8 maps to the lysyl oxidaselike 2 gene ( LOXL2 ; ). The association signal around SNP rs116607983 on chromosome 4 is located in a region devoid of protein-coding genes approximately 2000 kb to either side ( ). A fourth locus on chromosome 5 (rs2279135) was also supported by multiple clumped SNPs, but all SNPs, including the lead SNP, were imputed ( ). Several variants clumped at the chromosome 5 locus were located in the ARHGEF37 gene, coding for Rho Guanine Nucleotide Exchange Factor 37 ( ). None of the identified SNPs is an expression quantitative trait locus in brain tissues according the Genotype Tissue Expression project. At four of the six remaining loci with variants exhibiting suggestive associations, at most two supporting SNPs were present, which were all imputed; in the other two loci, no supporting SNPs could be found in the clumping analysis ( , ). We did not investigate these loci further because it is unlikely that they represent valid associations. No significant associations with Bonferroni-adjusted P values were detected with previously reported Parkinson’s disease associations from a meta-analysis of 17 datasets from a Parkinson’s disease GWAS ( ).
ZIC4 and ZIC1 are known to play a critical role in the embryonal development of the cerebellum. Heterozygous deletions comprising the ZIC1 and ZIC4 locus have been associated with the Dandy–Walker malformation, a rare congenital condition characterized by a hypoplastic cerebellar vermis and an enlarged fourth ventricle. , In mice, deletions of ZIC1 and ZIC4 lead to a striking phenotype similar to the Dandy–Walker malformation with cerebellar hypoplasia and foliation defects. , In addition, paraneoplastic autoantibodies against ZIC4 protein are linked to severe cerebellar dys-function and degeneration. , Because cerebellar degeneration and corresponding symptoms are also a central hallmark of MSA, we decided to follow up on a potential role of ZIC4 in MSA patient brains by performing immunohistochemical stainings. For ZIC1, no primary antibody was appropriately sensitive and specific on human tissue in our hands. Thus, FFPE tissues of the cerebellum and, for comparison, the frontal cortex of patients with MSA (n = 10 SND, n = 14 OPCA/mixed phenotype) and healthy controls (b = 5) were stained with antibodies raised against ZIC4. Nuclear and cytoplasmic staining of frontal cortex neurons was observed in all brains examined without differences between healthy controls and patients with MSA ( – ). In the cerebellar dentate nucleus, we found strong expression of ZIC4 in a subset of neurons in healthy controls, as well as patients with MSA with predominant SND ( , , , ). In contrast, patients with MSA with mixed subtype or OPCA showed reduced numbers of ZIC4-positive neurons, which were furthermore only weakly stained ( , ). Quantification of the proportions of ZIC4-positive neurons among the total number of dentate nucleus neurons depicted relatively constant proportions in healthy controls and patients with MSA-SND (33.2% ± 0.0% vs 32.6% 0.0%), whereas in patients with MSA-OPCA or MSA-mixed phenotype, we found significantly lower percentages of ZIC4-positive neurons (15.5% ± 0.1%) ( ).
As part of the study, brain banks were contacted worldwide, and all available white MSA brains were included. As in the prior GWAS with 918 predominantly clinically diagnosed MSA patients, our current GWAS of 648 patients with autopsy-confirmed MSA did not identify disease-associated common variants less than the genome-wide significance threshold. Previously, hypothesis-driven candidate gene studies found inconsistent results for genetic variants and genes potentially associated with MSA. An association of MSA with genetic variants in COQ2, SNCA, MAPT , and PRNP had been discussed. – , However, these genes have not been convincingly confirmed in other candidate gene studies and have not been associated in a previous MSA GWAS. This preceding GWAS analyzed 918 mostly clinical cases and 3864 controls. Overall, this GWAS did not identify any genome-wide significant hits. Because our prior GWAS of 219 patients with autopsy-confirmed corticobasal degeneration did identify significant disease-associated common variants, our current findings strongly suggest that the genetic contribution to disease risk is smaller in MSA. Nevertheless, our study demonstrates several suggestive associations at different loci, which may provide relevant hypotheses for follow-up investigations into the pathogenesis of MSA. Specifically, we identified a variant on chromosome 3 (rs16859966; P = 8.6 × 10 −7 ; OR, 1.58; 95% CI: 1.32–1.89) located upstream of ZIC1 and ZIC4 . ZIC1 and ZIC4 are located in close genomic proximity to each other and encode transcription factors highly expressed in different brain areas. , Proper function of these proteins is critical for the development of the CNS, particularly the cerebellum. Although no effect of rs16859966 on ZIC1 or ZIC4 expression is recorded in the Genotype Tissue Expression database, rare genetic variants or deletions in ZIC1 or ZIC4 result in congenital cerebellar defects. , , A heterozygous deletion of ZIC1 and ZIC4 causes the Dandy–Walker malformation, a developmental disorder of the cerebellum. , Remarkably, two recent epigenomic analyses in brain tissue of MSA point to ZIC4 . , Moreover, paraneoplastic autoantibodies against ZIC4 induce cerebellar degeneration. Due to the pronounced cerebellar degeneration in MSA, we followed up on a possible role of ZIC4 In MSA. Although we could detect a relatively constant proportion of approximately one-third ZIC4-positive neurons among all neurons in the cerebellar dentate nucleus in healthy controls and patients with MSASND, cases with MSA-OPCA or the mixed MSA phenotype showed significantly lower fractions of ZIC4-positive neurons. This finding suggests that ZIC4 may be involved in the neurodegeneration in MSA. The involvement of ZIC4 mutations in the Dandy–Walker cerebellar malformation and the paraneoplastic ZIC4 autoantibody–associated cerebellar degeneration could suggest a pathomechanism in MSA, by which altered ZIC4 expression could increase neuronal vulnerability. Further analyses of a potential functional interaction of α-synuclein and ZIC4 are currently ongoing. Explorative analysis of PD-related associations identified by GWAS yielded no significant association in MSA when adjusting for multiple testing. However, for unadjusted P values, five SNPs reached a significance threshold of P < 0.05,which might be interesting to study further. This study has a major limitation. Typically, a GWAS is conceptualized as a two-stage design with a discovery stage and a replication stage and supposedly achieving “genome-wide significance” in the discovery stage. The P values in the replication stage should remain significant after Bonferroni correction. Due to the limited number of autopsy-confirmed MSA cases worldwide, we could not conduct a two-stage procedure, let alone a further independent replication. In view of the aforementioned diagnostic uncertainty in clinical cases, a replication in predominantly clinically diagnosed MSA cases did not seem desirable. Therefore, we strongly encourage bringing MSA cases to autopsy and conducting a further independent replication study to confirm or refute the hypotheses provided by our study.
fS1 fS2 tS3 tS4 tS1 tS2
|
Archaeal and Extremophilic Bacteria from Different Archaeological Excavation Sites | bf93fc6b-ca44-4b9f-9d17-f0c5e9dca9cc | 10052888 | Microbiology[mh] | The character of the soil, its fertility and robustness are closely connected with the composition of soil bacterial communities . Soils not only contain a huge number of bacteria, but represent, in general, ecosystems of very high diversity. Typically, a few bacterial types are present in high cell numbers, whereas a vast number of different bacteria form a microbial background in a more or less dormant state . Therefore, most soils have an extensive reservoir of very different genetic resources, which can be activated when the environmental conditions are changed . This community background is the basis for the robustness of soil microbial communities. This allows them to adapt to the slow drifts in the character of an environment as well as in the case of immediate drastic changes. Each volume element of soil changes the quantitative and qualitative compositions of its bacterial community due to the shifts and breaks during the evolution of the environment, including macro flora and fauna as well as geological and other physical processes and the accompanying chemical changes. Bacteria which had been essential for physiological activity during an earlier phase disappear or might fall into a dormant state under new environmental conditions in a later phase. In this sense, the composition of “silent bacteria” stores ecological information about the past to some extent. The large number of less or non-active soil bacteria forms a kind of “ecological memory” of soil. In addition to natural factors, human impacts contribute to the development of soil microbial systems. Agriculture, cattle breeding, settlements and handcraft activities lead to changes in the physical and chemical parameters of soil, the exchange of gases, humidity, and so on. In particular, the exploration of natural resources, mining and industrial production has a huge impact on the local environmental conditions and is marked mainly by big changes in soil microbial communities [ , , ]. A typical, and drastic, example is the development of acid drainage from industrial mining sites [ , , ]. Since industrialization, the human impact on the environment is very strong and has achieved dangerous global dimensions during the last decades. However, it has to be seen that this impact starts much earlier with the beginning of the formation of culturally influenced habitats—occurring in middle Europe the latest with the introduction of agriculture about 7500 years before. It became enforced with the fabrication of metals and the development of cities. Therefore, not only the modern industrial society but also older cultures have impacted the soils and their bacterial communities . During the investigation of soil samples from areas with pre-industrial human impact, special bacterial types from the archaeal domain as well as other extremophilic types had been observed several times ; for example, Hadesarchaeota from the early copper mining places of the East Harz region . Hadesarchaeota was first found in deep gold mines in South Africa . This group and other special groups of bacteria such as Aenigmarchaeota, Asgardarchaea and Nanoarchaea have been reported to be detected in extreme environments, in the deep ground, or from deep sea hot spots typically. Remarkably, a considerable portion of bacterial DNA from near-surface soil indicating such extremophilic types was also found in soil samples from archaeological excavation places. On the one hand, an increasing number of investigations indicate that such bacteria are not exclusively present in the deep ground and the deep sea but are largely distributed. On the other hand, special types of areas used to promote the presence of the above-mentioned archaea. It is to be expected that natural conditions and recent human activities are only one part of the factors for the development of soil bacterial communities. In addition, their composition is modulated by traces of former events and human impacts. Soils and the patterns of their bacterial communities are embossed by their history which includes pre-industrial land use as well as soil-related human activities in the middle age and in prehistorical periods. Therefore, it is assumed that each soil bacterial community includes a certain “memory of soil history”. In addition to other bacteria, even Archaea should be regarded as possible indicators for such a memory effect. Here, the appearance of such microorganism groups in soil samples from a historical tannery area of the city of Jena and some other archaeological excavation sites is reported as an example of the early human impact on soil microbial communities and their archaeal components.
2.1. Presence of Archaeal Phyla in the Archaeological Samples The regarded soil samples show considerable differences in the presence of Archaea: in most samples between about 5 and 20% ( ). Whereas the samples from Kölleda (HB53-1 and HB53-2) contained less than 1%, the replaced topsoil from the ancient coal mine area of Bennstedt (HB59-1 and HB59-2) is marked by more than 40% of Archaea in the obtained 16S r-RNA reads. More than 30% Archaea content was found in samples HB55-1 and HB55-2 (Jena, Germany, depth 1.6 m). The dominating archaeal phylum in these samples as well as in most of the other samples is Crenarchaeota . Crenarchaeota are widely abundant in different environments, such as in oceans , forests, grasslands, permafrost, and in fresh water, whereby the typical abundance was found to be up to about 3% of bacterial 16S r-RNA genes . The red ash deposit from the saline of Bad Dürrenberg (HB61) contains a much lower portion of Archaea, but in this group Nanoarchaeota is the most abundant Archaea with about 80% of all Archaea. Nanoarchaeota are microorganisms with reduced genomes which are symbionts of other Archaea and have been mainly found in marine thermal vents and hot springs . Comparatively high amounts of Nanoarchaeota were observed in all samples from the ancient tannery area of the city of Jena. Their high presence in many of the investigated samples suggests the promotion of Nanoarchaeota by thermal processes—possibly in connection with application of mineral coal as it was used in the saline processes supplying the ashes of samples HB61-1, -2 and HB62-1, -2. In addition, most samples from the sample group from the ancient tannery area show a significant presence of Asgardarchaeota . This Archaea are identified as sharing several features with eukaryotic microorganisms and seem to be one of the closest prokaryotic relatives to the eukaryotes. Asgardarchaeota have been found in a wide range of microbiomes on Earth, preferably from sediment habitats . One special aspect of their adaptation to extreme environments is their ability to metabolize hydrocarbons ; another aspect is their tolerance against salt stress—probably related to the development of salt-adaptive strategies . The occurrence of Asgardarcheota in the soil samples from different sampling sites of the ancient tannery area might be related to the application of sodium chloride and other salts in the tannery and dying processes. Interestingly, Asgardarchaeota are present in the upper layers of the tannery area of Jena, from which all samples are marked by comparatively high pH values (see ). In contrast, the deeper sediment layers from Jena (HB55, HB56), which are marked by lower electrical conductivity values (lower salt content) and lower organic content, supplied no reads for Asgardarchaeota. In addition to the Crenarchaeota, Thermoplasmatota have been proven in nearly all samples. They represent the second highest abundant archaeal phylum in the samples of the grey ash deposit of the Saline of Bad Dürrenberg. Thermoplasmatota is one of the most abundant components in archaeal sea plankton and members of this phylum seem to be related to the metabolic oxidation of sulphur compounds . The appearance of Micrarchaeota in the samples from the ancient coal mine area near Bennstedt corresponds to the fact that this archaeal phylum is related to acidic environments and can be associated with Thermoplasmatales . Indeed, the three samples of Bennstedt showed the lowest pH values of the investigated soil samples (pH between 4.05 and 4.22 ). 2.2. Principle Component Analysis and Comparison of Samples by Different Archaea Taxons A more detailed comparison of samples can be achieved with a principal component analysis (PCA) on the OTU level. For the calculation, the PCA tool of Mathlab (R2015b) was used. The procedure is based on a minimization of the sum of square deviations of linear distances. The input data are the r-values (logarithmic values calculated following Equation (1)) of the abundances of all found OTUs of the domain of Archaea of all the regarded samples (74 different OTUs in total). These values are applied instead of the absolute read numbers in the PCA in order to include high-abundant and low-abundant types in a balanced way. The results are shown in two-dimensional plots for the first and the second as well as for the third and the fourth principal components ( ). It is obvious that the different sampling places are clearly distinguished by the PCA plots. Four-point clouds are displayed in the plot for first and second PCA: The central point cloud is formed by the samples from Großengottern, from Bad Dürrenberg and from Kölleda. Well separated from these points and clearly distinct from each other are the samples from Bennstedt, the samples from the deeper layers of Jena (1.60 m and 2 m depth) and the other samples of Jena. The 1.60 m samples and the 2 m sample from Jena are similar in PC3 but separated in PC4 (yellow dots and blue stars in ). In the PC3/PC4 plot, the samples from Kölleda (blue diamonds), Bad Dürrenberg (pink crosses) and Großengottern (red squares) are separated from each other too ( ). 2.3. Specificity of Soil Samples by Abundance of Different Archaea Among the highly abundant Crenarchaeota , the three families Nitrososphaeracea, Nitrosopumilaceae and Nitrosotalaceae of the class Nitrososphaeria (Nitrososphaera) are the most abundant in the investigated soil samples. Nitrososphaeria are known, in general, for their role in nitrogen cycling and are found in heavy metal ion-polluted soils—for example, in mining areas and in acid mine drainage . Such mine drainages are typically marked by low pH values, enhanced heavy metal ion concentration—whereby the low pH supports the mobilization and bioavailability of these metal ions—and enhanced salt content. The Nitrososphaeria are first isolated from garden soil and recognized for their ability to oxidize ammonia . Nitrososphaeraceae are abundant in the majority of the investigated samples ( a), but particularly high in HB54-1/-2 (Großengottern) and HB60-1/-2 (Bad Dürrenberg- top soil). The abundance in the samples from Großengottern corresponds well with the heavy metal contamination in the local soil, whereas the samples HB60-1/-2 are marked by very high electrical conductivity, indicating the high salt content of the place where the historical ashes had been deposed. In addition, this bacterial group indicates the former release of nitrogen-species as ammonia into the soil, which matches to the former input of nitrogen-riche organic material and waste into the soil. Remarkably, a dominance of Nitrososphaeraceae corresponds with the general high abundance of ammonia-oxidizing bacteria in the sample set of Jena. In these soils, a high input of organic material—and in particular animal residues as well as urea—could be expected from the former tannery activities in this area. It is interesting, too, with respect to a possible role of copper, in particular in the case of the soil samples HB54-1 and HB54-2 which were in direct contact with the bronze artefacts of the hoard of Großengottern. Nitrosopumilaceae ( b) are also highly abundant in the samples from the old tannery area of Jena but are found (above 0.1% of reads) in the sample set of Bad Dürrenberg only in the grey ash deposit (HB62-1 and HB62-2). Nitrosopumilaceae are known for their ability to oxidize ammonia . Similar to Nitrososphaeracea , they are related to low oxygen availability and are found typically in near-surface sediment layers and in the bathypelagic zone of oceans . The abundance of Nitrosotaleaceae shows a rather different distribution picture in the investigated samples ( c). In the sample set of the tannery area of Jena, they are present in medium or higher read numbers only in samples HB38-1/-2 and in the deep reference samples (HB55-1/-2 and HB56-1/-2). In contrast, they are comparatively highly abundant in the samples of Bennstedt (HB57-1/-2, HB58-1/-2 and HB59-1/-2), which are marked by the absence of Nitrosopumilaceae. This characteristic difference between the soil samples of Bennstedt and most other samples is probably due to the low pH of these samples, which corresponds with the described dominance of this family at low pH . However, it has to be remarked that the pH value is probably not the only criterion for the presence of Nitrosotaleaceae in the archaeologically taken soils because they are also present in HB38-1/-2, in HB55-1/-2 and HB56-1/-2, which have pH-values around 9. The newly defined phylum Woesearchaeota (DPANN superphylum) is represented by three OTUs of the order Woesearchaeales ( ). These bacteria have been identified as surprisingly highly abundant in very different habitats, among them oil reservoirs and sulphur springs. Obviously, they are part of a consortia with methanogenic bacteria which are involved in carbon cycling under anaerobic conditions. In addition, they are involved in the nitrogen fixation, denitrification and reduction of sulphate . Probably, they are symbionts of methanogenic bacteria in the degradation processes of synthetic polymers in wastewater . The type GW2011_GWC1_47_15 was found in all samples of the tannery area of Jena as well as in the samples from the saline ash deposits of Bad Dürrenberg ( a). In contrast, the types AR15 and AR20 have been proven in a part of the samples from Jena only, whereby the samples HB36-1/-2 and HB40-1/-2 show particular high abundances for both OTUs ( b,c). The phylum Aenigmarchaeota and the classes Heimdallarchaeia and Bathyarchaiea are mainly found in the samples of Jena too, but not, or very less, in the other samples ( ). The same is the case for some other less abundant archaeal OTUs as Methanosarcina, Cand. Iainarchaeum, Methanocellales—Rice Cluster and Thermoplasmata—DHVEG-1 ( ). Bathyarchaiea have been rather recently identified as one of the “most abundant microorganisms on earth” possessing an important spectrum of metabolic abilities including the degradation of proteins, lipids and aromatics such as benzoate . These bacteria are mainly present in anoxic sediments typically characterized by a high abundance of archaeal in the soil bacteria communities In addition to the samples from Großengottern (HB54/1 und HB54/2), they have been found in all samples of Jena, which obviously corresponds to the complex composition of the former input of waste in these soils. Comparing the different sampling places, certain specificity can be discerned concerning selected OTUs. For example, the euryarchaeotic type Methanofastidiosales , uncult. was found in samples HB39-1 and HB39-2 only ( a), which originated from the interior soil of one vat (Jena). Other OTUs seem to be characteristic for a sample group. This is the case for Candidatus Micrarchaeum , which was obtained from the three sampling sites of the early coal mining area of Bennstedt exclusively ( b). This is caused, obviously, by the general low pH of these soil samples and confirms the acidophilic character of Micrarchaeum (see ). In addition, Nitrososphaeria , Grp. 1.1c and Thermoplasmata , uncult. reflect the special conditions of the samples from Bennstedt. Aenigmarchaeales are typical for the samples from the tannery area of Jena. They are found there in all samples, but not in the other sample sets ( c). The deep-layer reference samples from Jena are marked by a comparatively high number of reads of an OTU named “ Nitrosphaerie, uncult ” ( d). This OTU was indicated by very few reads in HB34, HB38 (both Jena, outside vat) and additionally in HB54 (Großengottern). Finally, the OTU “ Cand. Methanoperedens ” was found in a part of the samples from Jena only—thereby most abundant in HB32 and HB38 ( e). This type of Archaea is described as to couple the anaerobic oxidation of methane with the reduction of nitrate , which would correspond well with human impact which caused high organic content in the historical and prehistorical soil development of settlement and handcraft areas. This matches to the finding of other nitrate-reducing and methanogenic bacteria in the old tannery area of Jena too. Investigations on sludges from recent tannery activities confirm the importance of Archaea in tannery-related soils. The addition of tannery waste in composting supplied a significant shift in the composition of soil bacterial communities , whereby the relative abundances of Archaea have been found higher in anaerobic than in aerobic tannery sludges . 2.4. Dadabacteria and Zixibacteria Most samples from the old tannery area of Jena are marked by a remarkably high number of reads for Dadabacteria ( a) and Zixibacteria ( b). Typically, they are represented by between several hundred and several thousand reads. Dadabacteria are nearly not found in the samples of Kölleda, Bennstedt and Bad Dürrenberg. A significant number of reads for Zixibacteria was observed in HB60-1/-2 (Bad Dürrenberg topsoil) and both types in the samples from Großengottern (HB54-1/-2) only. Recently, Zixibacteria has been identified as one of the prevalent phyla in a chemoautotrophical ecosystem driven by hydrogen sulphide-rich groundwater found in a cave in Romania ). Sulphur-rich residues from tannery processes such as fur, horn, and hair may have been a source of keratin and sulphur-based metabolic activity in the soil of the former tannery area in Jena. Dadabacteria are known from marine, terrestrial, sub-surface and hydrothermal environments. They preferentially metabolize microbial organic matter and can degrade peptidoglycans and phospholipids in particular. High contents of Dadabacteria have been found in coastal sediments too. The highest abundances seem to be correlated with human impact. Thus, Dadabacteria are an indicator for pollution and are probably related to the degradation of aromatic compounds . In addition to Archaea and other bacteria, an enrichment of Dadabacteria was reported to be found in recent tannery sludge . This observation can be confirmed by the results reported here which connect the abundance of Dadabacteria with the comparatively high abundance of Archaea in a historical tannery area. 2.5. Discussion of Abundance of Archaea Related to Prehistorical and Historical Burial, Settlement and Workplaces The composition of soil bacterial communities of archaeological places is of large interest. Despite that fact, up to now few studies on soil bacteria communities of prehistorical and historical settlement and workplaces are available. The results speak for an impact of former human activities which is still reflected in the recent bacterial communities. For example, it is assumed that burial places possess the “highest prokaryotic diversity of any environment” , but the knowledge about these bacterial communities is still low . In the soils of the burial sites of Jingzhou, China (about 5th to 3rd century BC), Crenarchaeota have been identified, but in a low content next to the more than 95% of Proteobacteria , Actinobacteria , Bacteriodetes and Firmicutes (, Figure 1). Archaea had been found up to a content of about 8% in some soil samples from the ancient copper mining areas of the East Harz region . NGS investigations on soil bacteria communities of historical rice fields speak to the fact that methanogenic Archaea could be used as an indicator for medieval rice cultivation . For comparison with the archaeal components in our investigations, the proof of Archaea in soils from a prehistoric settlement of Sicily are particularly interesting: Human impact layers from this iron age settlement (6th century BC) of Monte Iato (Sicily/Italy) showed in the NGS data of archaeal domain a strong dominance of Nitrososphaera , a lower content of Euryarchaeota , in particular Methanomassilococcus, and only low contents of other Archaea . Methanococcus was also found in low content in one sample from Großengottern, but in several samples from the tannery area of Jena. Different OTUs of the Nitrosospharea have been found in all samples investigated in this presented study. As in the case of the samples from Monte Iato, these ammonia-oxidizing bacteria seem to present a strong indicator of human impact on ancient soil.
The regarded soil samples show considerable differences in the presence of Archaea: in most samples between about 5 and 20% ( ). Whereas the samples from Kölleda (HB53-1 and HB53-2) contained less than 1%, the replaced topsoil from the ancient coal mine area of Bennstedt (HB59-1 and HB59-2) is marked by more than 40% of Archaea in the obtained 16S r-RNA reads. More than 30% Archaea content was found in samples HB55-1 and HB55-2 (Jena, Germany, depth 1.6 m). The dominating archaeal phylum in these samples as well as in most of the other samples is Crenarchaeota . Crenarchaeota are widely abundant in different environments, such as in oceans , forests, grasslands, permafrost, and in fresh water, whereby the typical abundance was found to be up to about 3% of bacterial 16S r-RNA genes . The red ash deposit from the saline of Bad Dürrenberg (HB61) contains a much lower portion of Archaea, but in this group Nanoarchaeota is the most abundant Archaea with about 80% of all Archaea. Nanoarchaeota are microorganisms with reduced genomes which are symbionts of other Archaea and have been mainly found in marine thermal vents and hot springs . Comparatively high amounts of Nanoarchaeota were observed in all samples from the ancient tannery area of the city of Jena. Their high presence in many of the investigated samples suggests the promotion of Nanoarchaeota by thermal processes—possibly in connection with application of mineral coal as it was used in the saline processes supplying the ashes of samples HB61-1, -2 and HB62-1, -2. In addition, most samples from the sample group from the ancient tannery area show a significant presence of Asgardarchaeota . This Archaea are identified as sharing several features with eukaryotic microorganisms and seem to be one of the closest prokaryotic relatives to the eukaryotes. Asgardarchaeota have been found in a wide range of microbiomes on Earth, preferably from sediment habitats . One special aspect of their adaptation to extreme environments is their ability to metabolize hydrocarbons ; another aspect is their tolerance against salt stress—probably related to the development of salt-adaptive strategies . The occurrence of Asgardarcheota in the soil samples from different sampling sites of the ancient tannery area might be related to the application of sodium chloride and other salts in the tannery and dying processes. Interestingly, Asgardarchaeota are present in the upper layers of the tannery area of Jena, from which all samples are marked by comparatively high pH values (see ). In contrast, the deeper sediment layers from Jena (HB55, HB56), which are marked by lower electrical conductivity values (lower salt content) and lower organic content, supplied no reads for Asgardarchaeota. In addition to the Crenarchaeota, Thermoplasmatota have been proven in nearly all samples. They represent the second highest abundant archaeal phylum in the samples of the grey ash deposit of the Saline of Bad Dürrenberg. Thermoplasmatota is one of the most abundant components in archaeal sea plankton and members of this phylum seem to be related to the metabolic oxidation of sulphur compounds . The appearance of Micrarchaeota in the samples from the ancient coal mine area near Bennstedt corresponds to the fact that this archaeal phylum is related to acidic environments and can be associated with Thermoplasmatales . Indeed, the three samples of Bennstedt showed the lowest pH values of the investigated soil samples (pH between 4.05 and 4.22 ).
A more detailed comparison of samples can be achieved with a principal component analysis (PCA) on the OTU level. For the calculation, the PCA tool of Mathlab (R2015b) was used. The procedure is based on a minimization of the sum of square deviations of linear distances. The input data are the r-values (logarithmic values calculated following Equation (1)) of the abundances of all found OTUs of the domain of Archaea of all the regarded samples (74 different OTUs in total). These values are applied instead of the absolute read numbers in the PCA in order to include high-abundant and low-abundant types in a balanced way. The results are shown in two-dimensional plots for the first and the second as well as for the third and the fourth principal components ( ). It is obvious that the different sampling places are clearly distinguished by the PCA plots. Four-point clouds are displayed in the plot for first and second PCA: The central point cloud is formed by the samples from Großengottern, from Bad Dürrenberg and from Kölleda. Well separated from these points and clearly distinct from each other are the samples from Bennstedt, the samples from the deeper layers of Jena (1.60 m and 2 m depth) and the other samples of Jena. The 1.60 m samples and the 2 m sample from Jena are similar in PC3 but separated in PC4 (yellow dots and blue stars in ). In the PC3/PC4 plot, the samples from Kölleda (blue diamonds), Bad Dürrenberg (pink crosses) and Großengottern (red squares) are separated from each other too ( ).
Among the highly abundant Crenarchaeota , the three families Nitrososphaeracea, Nitrosopumilaceae and Nitrosotalaceae of the class Nitrososphaeria (Nitrososphaera) are the most abundant in the investigated soil samples. Nitrososphaeria are known, in general, for their role in nitrogen cycling and are found in heavy metal ion-polluted soils—for example, in mining areas and in acid mine drainage . Such mine drainages are typically marked by low pH values, enhanced heavy metal ion concentration—whereby the low pH supports the mobilization and bioavailability of these metal ions—and enhanced salt content. The Nitrososphaeria are first isolated from garden soil and recognized for their ability to oxidize ammonia . Nitrososphaeraceae are abundant in the majority of the investigated samples ( a), but particularly high in HB54-1/-2 (Großengottern) and HB60-1/-2 (Bad Dürrenberg- top soil). The abundance in the samples from Großengottern corresponds well with the heavy metal contamination in the local soil, whereas the samples HB60-1/-2 are marked by very high electrical conductivity, indicating the high salt content of the place where the historical ashes had been deposed. In addition, this bacterial group indicates the former release of nitrogen-species as ammonia into the soil, which matches to the former input of nitrogen-riche organic material and waste into the soil. Remarkably, a dominance of Nitrososphaeraceae corresponds with the general high abundance of ammonia-oxidizing bacteria in the sample set of Jena. In these soils, a high input of organic material—and in particular animal residues as well as urea—could be expected from the former tannery activities in this area. It is interesting, too, with respect to a possible role of copper, in particular in the case of the soil samples HB54-1 and HB54-2 which were in direct contact with the bronze artefacts of the hoard of Großengottern. Nitrosopumilaceae ( b) are also highly abundant in the samples from the old tannery area of Jena but are found (above 0.1% of reads) in the sample set of Bad Dürrenberg only in the grey ash deposit (HB62-1 and HB62-2). Nitrosopumilaceae are known for their ability to oxidize ammonia . Similar to Nitrososphaeracea , they are related to low oxygen availability and are found typically in near-surface sediment layers and in the bathypelagic zone of oceans . The abundance of Nitrosotaleaceae shows a rather different distribution picture in the investigated samples ( c). In the sample set of the tannery area of Jena, they are present in medium or higher read numbers only in samples HB38-1/-2 and in the deep reference samples (HB55-1/-2 and HB56-1/-2). In contrast, they are comparatively highly abundant in the samples of Bennstedt (HB57-1/-2, HB58-1/-2 and HB59-1/-2), which are marked by the absence of Nitrosopumilaceae. This characteristic difference between the soil samples of Bennstedt and most other samples is probably due to the low pH of these samples, which corresponds with the described dominance of this family at low pH . However, it has to be remarked that the pH value is probably not the only criterion for the presence of Nitrosotaleaceae in the archaeologically taken soils because they are also present in HB38-1/-2, in HB55-1/-2 and HB56-1/-2, which have pH-values around 9. The newly defined phylum Woesearchaeota (DPANN superphylum) is represented by three OTUs of the order Woesearchaeales ( ). These bacteria have been identified as surprisingly highly abundant in very different habitats, among them oil reservoirs and sulphur springs. Obviously, they are part of a consortia with methanogenic bacteria which are involved in carbon cycling under anaerobic conditions. In addition, they are involved in the nitrogen fixation, denitrification and reduction of sulphate . Probably, they are symbionts of methanogenic bacteria in the degradation processes of synthetic polymers in wastewater . The type GW2011_GWC1_47_15 was found in all samples of the tannery area of Jena as well as in the samples from the saline ash deposits of Bad Dürrenberg ( a). In contrast, the types AR15 and AR20 have been proven in a part of the samples from Jena only, whereby the samples HB36-1/-2 and HB40-1/-2 show particular high abundances for both OTUs ( b,c). The phylum Aenigmarchaeota and the classes Heimdallarchaeia and Bathyarchaiea are mainly found in the samples of Jena too, but not, or very less, in the other samples ( ). The same is the case for some other less abundant archaeal OTUs as Methanosarcina, Cand. Iainarchaeum, Methanocellales—Rice Cluster and Thermoplasmata—DHVEG-1 ( ). Bathyarchaiea have been rather recently identified as one of the “most abundant microorganisms on earth” possessing an important spectrum of metabolic abilities including the degradation of proteins, lipids and aromatics such as benzoate . These bacteria are mainly present in anoxic sediments typically characterized by a high abundance of archaeal in the soil bacteria communities In addition to the samples from Großengottern (HB54/1 und HB54/2), they have been found in all samples of Jena, which obviously corresponds to the complex composition of the former input of waste in these soils. Comparing the different sampling places, certain specificity can be discerned concerning selected OTUs. For example, the euryarchaeotic type Methanofastidiosales , uncult. was found in samples HB39-1 and HB39-2 only ( a), which originated from the interior soil of one vat (Jena). Other OTUs seem to be characteristic for a sample group. This is the case for Candidatus Micrarchaeum , which was obtained from the three sampling sites of the early coal mining area of Bennstedt exclusively ( b). This is caused, obviously, by the general low pH of these soil samples and confirms the acidophilic character of Micrarchaeum (see ). In addition, Nitrososphaeria , Grp. 1.1c and Thermoplasmata , uncult. reflect the special conditions of the samples from Bennstedt. Aenigmarchaeales are typical for the samples from the tannery area of Jena. They are found there in all samples, but not in the other sample sets ( c). The deep-layer reference samples from Jena are marked by a comparatively high number of reads of an OTU named “ Nitrosphaerie, uncult ” ( d). This OTU was indicated by very few reads in HB34, HB38 (both Jena, outside vat) and additionally in HB54 (Großengottern). Finally, the OTU “ Cand. Methanoperedens ” was found in a part of the samples from Jena only—thereby most abundant in HB32 and HB38 ( e). This type of Archaea is described as to couple the anaerobic oxidation of methane with the reduction of nitrate , which would correspond well with human impact which caused high organic content in the historical and prehistorical soil development of settlement and handcraft areas. This matches to the finding of other nitrate-reducing and methanogenic bacteria in the old tannery area of Jena too. Investigations on sludges from recent tannery activities confirm the importance of Archaea in tannery-related soils. The addition of tannery waste in composting supplied a significant shift in the composition of soil bacterial communities , whereby the relative abundances of Archaea have been found higher in anaerobic than in aerobic tannery sludges .
Most samples from the old tannery area of Jena are marked by a remarkably high number of reads for Dadabacteria ( a) and Zixibacteria ( b). Typically, they are represented by between several hundred and several thousand reads. Dadabacteria are nearly not found in the samples of Kölleda, Bennstedt and Bad Dürrenberg. A significant number of reads for Zixibacteria was observed in HB60-1/-2 (Bad Dürrenberg topsoil) and both types in the samples from Großengottern (HB54-1/-2) only. Recently, Zixibacteria has been identified as one of the prevalent phyla in a chemoautotrophical ecosystem driven by hydrogen sulphide-rich groundwater found in a cave in Romania ). Sulphur-rich residues from tannery processes such as fur, horn, and hair may have been a source of keratin and sulphur-based metabolic activity in the soil of the former tannery area in Jena. Dadabacteria are known from marine, terrestrial, sub-surface and hydrothermal environments. They preferentially metabolize microbial organic matter and can degrade peptidoglycans and phospholipids in particular. High contents of Dadabacteria have been found in coastal sediments too. The highest abundances seem to be correlated with human impact. Thus, Dadabacteria are an indicator for pollution and are probably related to the degradation of aromatic compounds . In addition to Archaea and other bacteria, an enrichment of Dadabacteria was reported to be found in recent tannery sludge . This observation can be confirmed by the results reported here which connect the abundance of Dadabacteria with the comparatively high abundance of Archaea in a historical tannery area.
The composition of soil bacterial communities of archaeological places is of large interest. Despite that fact, up to now few studies on soil bacteria communities of prehistorical and historical settlement and workplaces are available. The results speak for an impact of former human activities which is still reflected in the recent bacterial communities. For example, it is assumed that burial places possess the “highest prokaryotic diversity of any environment” , but the knowledge about these bacterial communities is still low . In the soils of the burial sites of Jingzhou, China (about 5th to 3rd century BC), Crenarchaeota have been identified, but in a low content next to the more than 95% of Proteobacteria , Actinobacteria , Bacteriodetes and Firmicutes (, Figure 1). Archaea had been found up to a content of about 8% in some soil samples from the ancient copper mining areas of the East Harz region . NGS investigations on soil bacteria communities of historical rice fields speak to the fact that methanogenic Archaea could be used as an indicator for medieval rice cultivation . For comparison with the archaeal components in our investigations, the proof of Archaea in soils from a prehistoric settlement of Sicily are particularly interesting: Human impact layers from this iron age settlement (6th century BC) of Monte Iato (Sicily/Italy) showed in the NGS data of archaeal domain a strong dominance of Nitrososphaera , a lower content of Euryarchaeota , in particular Methanomassilococcus, and only low contents of other Archaea . Methanococcus was also found in low content in one sample from Großengottern, but in several samples from the tannery area of Jena. Different OTUs of the Nitrosospharea have been found in all samples investigated in this presented study. As in the case of the samples from Monte Iato, these ammonia-oxidizing bacteria seem to present a strong indicator of human impact on ancient soil.
3.1. Soil Samples In this study, the results of 16S r-RNA profilings using NGS data from 35 samples from five different archaeological investigated places ( ) were analysed for the presence of selected groups of Archaea. The starting point was the observation that in some of these samples astonishingly high portions of Archaea were present. All samples were taken during archaeological excavations but under different original motivations. In the case of samples HB53-1 and HB53-2 (Kölleda) and HB54-1 and HB54-2 (Großengottern), soil samples were taken from the direct neighbourhood of bronze artefacts with the original intention to search for heavy metal-tolerant bacterial strains. The samples of Bennstedt and Bad Dürrenberg were taken to compare three sites closely situated to each other with different soil characters: in Bennstedt from a small ancient shaft for a coal mining survey, in Bad Dürrenberg from ash deposits of the local old saline. All other samples originated from a historical tannery area in the city of Jena, where residues of vats of unknown functions had been detected. An overview of included samples is given in . The samples differ considerably in their chemical properties. General differences in the soil properties of the sampling sites are reflected in the soil pH and the electrical conductivities ( ). The samples of the ancient coal mining sites of Bennstedt (HB57, HB58 and HB59) are marked by the acidic character of the soil. All other samples show a nearly neutral or moderate alkaline character. The sampling sites of the historical saline in Bad Dürrenberg (HB60, HB61, HB62) are marked by high electrical conductivity values, which can be interpreted as a high salt content of the ash deposits and the related soil. The investigated soil samples originated from very different archeological situations related to former human activities which led to the expectation of memory effects in soil microbiomes. The samples from Kölleda (HB53-1/-2) and Großengottern (HB54-1/-2) are related to humus soil in immediate contact with deposed pre-historical bronze artefacts. For these samples, impacts could be expected from the temporal release of copper ions and ions of minority alloying metal—depending on changing humidity and pH of near-surface soil—on the one hand, and in general by the buried ancient surface soil, on the other hand. In contrast, the samples from Bennstedt are related to early industrial mining prospection from the time around 1800. Humus material from the historical surface was mainly present in the replaced topsoil (samples HB59-1/-2), whereas the refilling material of the prospection shaft (HB58-1/-2) was low-humus sediment and the material from the coal seam (HB59-1/-2) was first coming to the surface after deposition millions of years ago. All six samples are marked by considerably low pH-values (around 4.1) which speaks to the presence of acidophilic soil microorganisms. The samples from Bad Dürrenberg (HB60-1/-2, HB61-1/-2 and HB62-1/-2) showed higher pH-values (between 7.7 and 8.2) but are marked by high electrical conductivity (above 1 mS/cm), which hints to preferential abundances of halotolerant types. The high conductivity is related to the fact that the soil samples had been taken from ash deposits of a historical saline facility originating from the early industrial period (beginning of the 19th century). The other samples (from Jena, Germany, Inselplatz, HB32-1 to HB40-2) were taken during the excavation of a late medieval to early modern time tannery area. There, human impacts are expected in connection with the deposition of animal waste and feces as well as tannery and dyeing-related materials and chemicals, which speaks to the presence of metal-tolerant as well as of nitrate-, ammonia- and sulphur-compound-related bacteria. Thus, the investigated soil samples reflect very different archaeological situations and led to the expectation of significant differences in the soil bacterial communities. These differences are analysed in the following using the 16S r-RNA data of the archaeal part of the soil bacterial community. 3.2. Sample Processing DNeasy ® PowerSoil ® Pro Kits (Qiagen, Hilden, Germany) were applied for the isolation of DNA from soil material following supplier’s instructions. For selective DNA amplification with PCR an Edvocycler (Edvotek, Washington, DC, USA) was used. The quality of PCR products was checked after each PCR step with gel electrophoresis in 1% agarose gels. The primary amplificates as well as the completed pooled libraries were purified following the supplier’s standard protocol with application of the ProNex ® Size-Selective Purification System (Promega, Madison, WI, USA). Adaptor primers Amplicon PCR A519F-Ad (5′ TCGTCGG-CAGCGTCAGATGTGTATAAGAGACAGCAGCMGCCGCGGTAA 3′) and Bact_805R-Ad (5′-GTCTCGTGGGCTCGGAGATGTGTATAAGAGACAGGACTACHVGGGTATCTAATC 3′) were obtained from Eurofins Genomics (Ebersberg, Germany). They were applied in a concentration of 100 pmol/µL. The PCR mixtures (25 µL in total per reaction) were composed as follows: 0.5 µL of DNA isolation eluate, 2 mM MgCl 2 , 200 µM dNTP mix, 0.65 Units GoTaq ® G2 Hot Start DNA Polymerase, nuclease-free water (all reagents from Promega, Madison (USA)) and 1 µM of each primer. For PCR amplification, the following steps were performed: initial denaturation for 5 min at 94 °C, 30 amplification cycles involving 30 s denaturation at 94 °C, 30 s primer annealing at 50 °C and 30 s extension at 72 °C. After thermocycling, the process was finished with a final extension reaction at 72 °C for 5 min. The forward and reverse indexing primers for index PCR were applied in a concentration of 1.25 pmol/µL. They were also obtained with Eurofins Genomics (Ebersberg, Germany). The PCR for the index process was realized using 25 µL per reaction as a total volume composed of 2.5 µL of Amplicon PCR product, 2.5 mM MgCl 2 , 300 µM dNTP mix, 0.5 Units GoTaq ® Mdx Hot Start DNA Polymerase, nuclease-free water (all reagents from Promega Corp., Madison, WI 53711 Madison (USA)) and 125 nM of each of the two primers of the respective indexing–primer pairing. The index–primer PCR involved the following process steps: initial denaturation for 3 min at 95 °C; 30 amplification cycles involving 30 s denaturation at 95 °C; 30 s primer annealing at 55 °C; and 30 s extension at 72 °C. After the 30 cycles of thermocycling, the process was finished with a final extension reaction at 72 °C for 5 min. 3.3. Processing of NGS Data The Galaxy open-source platform ( https://usegalaxy.org/ ) was used for the conversion of the obtained sequence data (fastq files) to contig files (fasta) and quality files (mothur (version 1.39.5)). A high median quality score was obtained for all investigated data sets. The contig files were aligned to rRNA databases based on the NCBI cloud using the SILVAngs data analysis service ( https://ngs.arb-silva.de/silvangs ). It allowed for a detailed community analysis of previously obtained sequencing data [ , , ]. For all analyses, the pre-set parameter configurations of the SILVAngs database version 138.1 are applied . In many cases, the final obtained sequencing data allow an assignment of taxonomical groups down to the genus level. In other cases, it is only possible to identify higher taxonomical levels as families, orders, classes or phyla. The lowest identified level for each distinguished bacterial type is referenced as the “Operational Taxonomic Unit” (OTU). In addition to the absolute numbers of reads for each OTU and their percentages or all reads of a sample, a logarithmic value is used here for comparing the compositions of soil bacterial communities. Therefore, the logarithms-related values r is applied, which is defined by the read number N per single OTU and the total number of all reads for each sample N sum as follow: r = log10(1 + 1000000 × N/N sum ) (1) The method of Principal Component Analysis (PCA) was used for a general identification of the relations between the composition patterns of soil bacterial communities of the single samples. The PCA is a statistical method for characterization of large data sets in multidimensional parameter spaces. It allows the reduction of the dimensionality of such parameter spaces and to identify the most important components. It transforms all data from the original multidimensional coordinate system into a new coordinate system with the best fitting for the first coordinate (“First Principal Component” PC1), the second-best fitting for the second coordinate (“Second Principal Component” PC2), and so on. In many cases, the correlation plots of both first coordinates give a sufficient illustration for the clustering of data. Here, the PCA function of MATLAB (R2015b) was used for the calculation of the Principal Components in the abundances of Archaea for the investigated sample set.
In this study, the results of 16S r-RNA profilings using NGS data from 35 samples from five different archaeological investigated places ( ) were analysed for the presence of selected groups of Archaea. The starting point was the observation that in some of these samples astonishingly high portions of Archaea were present. All samples were taken during archaeological excavations but under different original motivations. In the case of samples HB53-1 and HB53-2 (Kölleda) and HB54-1 and HB54-2 (Großengottern), soil samples were taken from the direct neighbourhood of bronze artefacts with the original intention to search for heavy metal-tolerant bacterial strains. The samples of Bennstedt and Bad Dürrenberg were taken to compare three sites closely situated to each other with different soil characters: in Bennstedt from a small ancient shaft for a coal mining survey, in Bad Dürrenberg from ash deposits of the local old saline. All other samples originated from a historical tannery area in the city of Jena, where residues of vats of unknown functions had been detected. An overview of included samples is given in . The samples differ considerably in their chemical properties. General differences in the soil properties of the sampling sites are reflected in the soil pH and the electrical conductivities ( ). The samples of the ancient coal mining sites of Bennstedt (HB57, HB58 and HB59) are marked by the acidic character of the soil. All other samples show a nearly neutral or moderate alkaline character. The sampling sites of the historical saline in Bad Dürrenberg (HB60, HB61, HB62) are marked by high electrical conductivity values, which can be interpreted as a high salt content of the ash deposits and the related soil. The investigated soil samples originated from very different archeological situations related to former human activities which led to the expectation of memory effects in soil microbiomes. The samples from Kölleda (HB53-1/-2) and Großengottern (HB54-1/-2) are related to humus soil in immediate contact with deposed pre-historical bronze artefacts. For these samples, impacts could be expected from the temporal release of copper ions and ions of minority alloying metal—depending on changing humidity and pH of near-surface soil—on the one hand, and in general by the buried ancient surface soil, on the other hand. In contrast, the samples from Bennstedt are related to early industrial mining prospection from the time around 1800. Humus material from the historical surface was mainly present in the replaced topsoil (samples HB59-1/-2), whereas the refilling material of the prospection shaft (HB58-1/-2) was low-humus sediment and the material from the coal seam (HB59-1/-2) was first coming to the surface after deposition millions of years ago. All six samples are marked by considerably low pH-values (around 4.1) which speaks to the presence of acidophilic soil microorganisms. The samples from Bad Dürrenberg (HB60-1/-2, HB61-1/-2 and HB62-1/-2) showed higher pH-values (between 7.7 and 8.2) but are marked by high electrical conductivity (above 1 mS/cm), which hints to preferential abundances of halotolerant types. The high conductivity is related to the fact that the soil samples had been taken from ash deposits of a historical saline facility originating from the early industrial period (beginning of the 19th century). The other samples (from Jena, Germany, Inselplatz, HB32-1 to HB40-2) were taken during the excavation of a late medieval to early modern time tannery area. There, human impacts are expected in connection with the deposition of animal waste and feces as well as tannery and dyeing-related materials and chemicals, which speaks to the presence of metal-tolerant as well as of nitrate-, ammonia- and sulphur-compound-related bacteria. Thus, the investigated soil samples reflect very different archaeological situations and led to the expectation of significant differences in the soil bacterial communities. These differences are analysed in the following using the 16S r-RNA data of the archaeal part of the soil bacterial community.
DNeasy ® PowerSoil ® Pro Kits (Qiagen, Hilden, Germany) were applied for the isolation of DNA from soil material following supplier’s instructions. For selective DNA amplification with PCR an Edvocycler (Edvotek, Washington, DC, USA) was used. The quality of PCR products was checked after each PCR step with gel electrophoresis in 1% agarose gels. The primary amplificates as well as the completed pooled libraries were purified following the supplier’s standard protocol with application of the ProNex ® Size-Selective Purification System (Promega, Madison, WI, USA). Adaptor primers Amplicon PCR A519F-Ad (5′ TCGTCGG-CAGCGTCAGATGTGTATAAGAGACAGCAGCMGCCGCGGTAA 3′) and Bact_805R-Ad (5′-GTCTCGTGGGCTCGGAGATGTGTATAAGAGACAGGACTACHVGGGTATCTAATC 3′) were obtained from Eurofins Genomics (Ebersberg, Germany). They were applied in a concentration of 100 pmol/µL. The PCR mixtures (25 µL in total per reaction) were composed as follows: 0.5 µL of DNA isolation eluate, 2 mM MgCl 2 , 200 µM dNTP mix, 0.65 Units GoTaq ® G2 Hot Start DNA Polymerase, nuclease-free water (all reagents from Promega, Madison (USA)) and 1 µM of each primer. For PCR amplification, the following steps were performed: initial denaturation for 5 min at 94 °C, 30 amplification cycles involving 30 s denaturation at 94 °C, 30 s primer annealing at 50 °C and 30 s extension at 72 °C. After thermocycling, the process was finished with a final extension reaction at 72 °C for 5 min. The forward and reverse indexing primers for index PCR were applied in a concentration of 1.25 pmol/µL. They were also obtained with Eurofins Genomics (Ebersberg, Germany). The PCR for the index process was realized using 25 µL per reaction as a total volume composed of 2.5 µL of Amplicon PCR product, 2.5 mM MgCl 2 , 300 µM dNTP mix, 0.5 Units GoTaq ® Mdx Hot Start DNA Polymerase, nuclease-free water (all reagents from Promega Corp., Madison, WI 53711 Madison (USA)) and 125 nM of each of the two primers of the respective indexing–primer pairing. The index–primer PCR involved the following process steps: initial denaturation for 3 min at 95 °C; 30 amplification cycles involving 30 s denaturation at 95 °C; 30 s primer annealing at 55 °C; and 30 s extension at 72 °C. After the 30 cycles of thermocycling, the process was finished with a final extension reaction at 72 °C for 5 min.
The Galaxy open-source platform ( https://usegalaxy.org/ ) was used for the conversion of the obtained sequence data (fastq files) to contig files (fasta) and quality files (mothur (version 1.39.5)). A high median quality score was obtained for all investigated data sets. The contig files were aligned to rRNA databases based on the NCBI cloud using the SILVAngs data analysis service ( https://ngs.arb-silva.de/silvangs ). It allowed for a detailed community analysis of previously obtained sequencing data [ , , ]. For all analyses, the pre-set parameter configurations of the SILVAngs database version 138.1 are applied . In many cases, the final obtained sequencing data allow an assignment of taxonomical groups down to the genus level. In other cases, it is only possible to identify higher taxonomical levels as families, orders, classes or phyla. The lowest identified level for each distinguished bacterial type is referenced as the “Operational Taxonomic Unit” (OTU). In addition to the absolute numbers of reads for each OTU and their percentages or all reads of a sample, a logarithmic value is used here for comparing the compositions of soil bacterial communities. Therefore, the logarithms-related values r is applied, which is defined by the read number N per single OTU and the total number of all reads for each sample N sum as follow: r = log10(1 + 1000000 × N/N sum ) (1) The method of Principal Component Analysis (PCA) was used for a general identification of the relations between the composition patterns of soil bacterial communities of the single samples. The PCA is a statistical method for characterization of large data sets in multidimensional parameter spaces. It allows the reduction of the dimensionality of such parameter spaces and to identify the most important components. It transforms all data from the original multidimensional coordinate system into a new coordinate system with the best fitting for the first coordinate (“First Principal Component” PC1), the second-best fitting for the second coordinate (“Second Principal Component” PC2), and so on. In many cases, the correlation plots of both first coordinates give a sufficient illustration for the clustering of data. Here, the PCA function of MATLAB (R2015b) was used for the calculation of the Principal Components in the abundances of Archaea for the investigated sample set.
Soil samples from five archaeological excavations have been investigated by 16S r-RNA metagenomics for the presence of Archaea in soil bacterial communities. Thereby, considerable differences have been observed. Whereas Archaea in one place are nearly negligible, they are present in other samples with more than 10%, in one case more than 40%, of the bacterial population. Samples of the historical tannery area (Jena, Germany) are marked by a high abundance of Nanoarchaeota and a considerable abundance of Asgardarchaeota . The samples are also marked by high abundances of Ammonia-oxidizing Crenarchaeota and by a significant presence of Zixibacteria and Dadabacteria. The different excavation places can be distinguished by a Principal Component Analysis (PCA) of archaeal OTUs. This speaks for the fact—despite all differences between specific samples—that a typical pattern of Archaea marks each place. These patterns and the specific compositions of soil bacterial communities in the single samples probably reflect the recent ecological situation and represent the “ecological memory” of former local human impacts dating back centuries or millennia. In addition to the recognition of differences in archaeal communities from different early human-impacted soils and the consequences of the former human activities on recent local soil bacterial communities, the archaeal strains from these places are also interesting as a potential reservoir for the future search for usable microorganisms in biotechnology. The application of Archaea and their metabolic products is under discussion and soils with different kinds of ancient human impacts should be considered as an interesting special source for new technically usable bacterial strains in the future.
|
Anterior horn damage in brachial multisegmental amyotrophy with superficial siderosis and dural tear: an autopsy case report | 9dff6e5d-672a-431b-bcd9-87bc83beae45 | 10053131 | Forensic Medicine[mh] | Superficial siderosis (SS) is a disease characterized by hemosiderin deposition on the surface of the central nervous system (CNS) due to chronic bleeding in the subarachnoid space . Patients with SS usually show slow progressive cerebellar ataxia, deafness, spasticity, and/or dementia . Various disorders such as trauma, arteriovenous malformation, amyloid angiopathy, tumor, and surgery of the CNS can induce SS. However, recent studies have demonstrated that the most common etiology of SS is dural tear at the lower cervical and thoracic spine with or without trauma (i.e., duropathies) . A small proportion of these patients with ventral intraspinal fluid collection develop brachial multisegmental amyotrophy [ – ]. T2-weighted magnetic resonance imaging (MRI) in some patients with brachial multisegmental amyotrophy shows bilateral hyperintensity at the anterior horn of the cervical spinal cord, which produces a so-called “snake eyes” appearance in the images . However, pathophysiology of brachial multisegmental amyotrophy with ventral intraspinal fluid collection has not been elucidated. Herein, for the first time, we describe spinal cord pathology in an autopsied patient with brachial multisegmental amyotrophy accompanied with SS, dural tear, and snake-eyes appearance.
A man without any family history of neurological diseases had four traffic accidents at the ages of 13, 19, 35, and 39 years. The traffic accidents did not induce or exacerbate any neurological symptoms. At the age of 38 years, he gradually developed progressive muscle atrophy and weakness in the predominantly proximal bilateral upper limbs. He showed tinnitus and hearing impairment at the age of 50 years. At the age of 53 years, brain and spinal cord MRI detected extensive SS. He was transferred to our hospital. Upon admission, mental and cognitive state were normal. Except for moderate hearing loss, cranial nerve functions were preserved. Bilateral muscle atrophy was apparent in the predominantly proximal upper limbs. He showed upper limb weakness (i.e., supraspinatus, 3/3 (right/left); pectoralis major 3/3; deltoid, 2/3; biceps, 3/4; triceps, 4/4; wrist extensor, 5/4; wrist flexor, 5/5; finger extensor, 5/2; dorsal interossei, 3/4 on the Medical Research Council Scale (MRC, 0–5)), without lower limb weakness. There were no sensory or autonomic dysfunctions. Tendon reflex in the upper limbs was decreased or diminished. Tendon reflex in the lower limbs was brisk without any pathological reflexes. He showed wide-based gait and was not able to perform tandem gait. No abnormal blood test findings were detected. Appearance of the cerebrospinal fluid (CSF) was bloody. Motor nerve conduction studies (NCSs) of the right limbs detected decreased amplitude of compound muscle action potentials in the median (780 μV) and ulnar (4.5 mV) nerves. F-wave occurrence of the median nerve was also decreased (30%). However, delay of F-wave and distal latency was mild, and motor nerve conduction velocity was preserved. Results of sensory NCSs in the upper limbs and those of motor and sensory NCSs in the lower extremities were normal. On electromyography (EMG) of the right extremities, apparent chronic denervation and reinnervation changes with scarce active denervation changes were detected in the upper limbs. In the quadriceps femoris muscle, EMG findings were normal. Transcranial magnetic stimulation showed prolonged central conduction time in the upper and lower limbs. T2*-weighted imaging of the brain (Fig. ) and spinal (Fig. A) MRI detected extensive SS. Ventral intraspinal fluid collection ranging from the C3 to L1 spinal levels was also detected (Fig. A-D, indicated by arrow). T2-weighted imaging showed high-intensity lesions at the bilateral anterior horn with snake-eyes appearance, ranging from the C3 to C7 spinal levels without apparent cervical canal stenosis (Fig. B, D). MRI also revealed a discontinuous part of the dura at the Th1/2 spinal level, suggesting dural tear (Fig. D, indicated by arrow head). We diagnosed him with brachial multisegmental amyotrophy accompanied with SS and dural tear. At the age of 54, repair of dural tear at the spinal level of Th2 were performed. Although temporary deterioration of ataxia occurred following the operation, progression of neurological dysfunctions stopped thereafter. At the age of 56, clear appearance of the CSF was confirmed by lumbar puncture. The snake-eyes appearance remained in MR images without recurrence of ventral intraspinal fluid collection (Fig. E, F). At the age of 58 years, he accidentally died due to drowning in the bath. Autopsy was performed with permission from his family. Macroscopically, the brain and spinal cord were brown due to extensive hemosiderin deposits (Fig. A, B). Microscopically, the cerebellum showed severe loss of the Purkinje cells along with hemosiderin deposition. In the spinal cord, hemosiderin deposition was nearly even throughout the surface; however, severe tissue damage of the spinal gray matter such as the anterior horns and intermediate zone was observed from the C3 to Th5 segments (Fig. C-F). The damage was more accentuated in segments from the C5 to Th2. In the anterior horns at these segments, neuronal cells were severely depleted, and remaining neurons were atrophic (Fig. E). Chromatolysis, known as axonal reaction, was not apparent in the remaining neurons. In contrast to the anterior horns and intermediate zone, no obvious tissue damage was observed in the posterior horns (Fig. F). Anterior horn cells were relatively preserved in segments from the middle thoracic (Th7) to lumbar cord (Fig. G, H); however, a small amount of these neurons showed chromatolysis (Fig. I). Mild degeneration of the white matter due to superficial hemosiderin deposition was also observed in the superficial region of the entire spinal cord (Fig. C, D, G).
Brachial multisegmental amyotrophy is a very rare phenotype among patients with SS and ventral intraspinal fluid collection . In previous clinical and radiological case studies [ – ], two speculative theories for motor dysfunction in patients with brachial multisegmental amyotrophy have been propsed; 1) compressive damage to the anterior horns from ventral intraspinal fluid collection, and 2) stretching damage to the motor nerve roots induced by posterior shift of the spinal cord due to ventral intraspinal fluid collection [ – ]. In our patient, snake-eyes appearance was detected on cervical MRI, similar to that in some of reported cases of brachial multisegmental amyotrophy with SS and dural tear . Snake-eyes appearance on cervical spinal cord MRI had been described in patients with brachial mutisegmental amyotrophy caused by cervical spondylosis or ossification of the posterior longitudinal ligament (OPLL) . To our knowledge, patients with snake-eyes appearance who have sole damage to the nerves such as neuropathy, radiculopathy, and plexopathy have not been described previously. Pathological analysis of an autopsied patient with OPLL and snake-eyes appearance on MRI has shown intramedullary cystic necrosis around the central gray matter and the ventolateral posterior column, with loss of the anterior horn cells . Autopsied case series of cervical spondylotic myelopathy have demonstrated similar neuropathological findings, and atrophy and neural loss are considered to start at the anterior horns and intermediate zone of the spinal gray matter . Pathophysiology of compressive spinal cord damage has been postulated to be due to circulatory disturbance [ – ]. In our patient, snake-eyes appearance on MRI and distribution of spinal cord damage on pathological analysis were similar to those findings in patients with compressive myelopathy . The similarity of histological changes of the spinal gray matter between spondylotic myelopathy and the present case suggests the pivotal role of ventral intraspinal fluid collection in the development of the anterior horn damage. On the other hand, amount of ventral intraspinal fluid collection was too small to severely compress the cervical spinal cord on MRI. Compared with the cervical spinal cord level, the amount of ventral intraspinal fluid collection was large at the thoracic spinal cord level especially in the middle thoracic spine. Reasons for the discrepancy in the levels between conspicuous anterior horn cell loss at the middle cervical to upper thoracic cord and the largest amount of ventral intraspinal fluid collection at the middle thoracic cord are unclear. However, a larger range of motion in the cervical spine than in the thoracic spine could be related to this discrepancy. Indeed, we assessed the spinal cord on MRI in a neutral position alone. In Hirayama disease, MRI at a neck flexion position can detect compression of the cervical spinal cord induced by forward displacement of the cervical dural sac . Cervical MRI during neck extension and/or flexion might show dynamic compression of the cervical cord due to ventral intraspinal fluid collection. Further studies are needed on the dynamic change of ventral intraspinal fluid collection and spinal cord following posture change. As for the idea that ventral intraspinal fluid collection may stretch the motor nerve roots causing brachial multisegmental amyotrophy, extensive damage of the anterior horns and intermediate zone with snake-eyes appearance on MRI in our patient is unlikely to be caused by damage to the motor nerve roots alone. Diffuse superficial deposition of hemosiderin on the whole segment of the spinal cord on our pathological analysis does not support that brachial multisegmental amyotrophy is mainly induced by hemosiderin deposition. However, a small amount of motor neurons with chromatolysis were detected in the segment with preserved anterior horn cells, and this finding may be due to damage to the axons of motor neurons in the anterior horns induced by hemosiderin deposition. No apparent chromatolysis at the segments with severe anterior horn damage may be due to depletion of almost all large neurons corresponding to motor neurons at these segments. On the other hand, some patients with both brachial multisegmental amyotrophy and ventral intraspinal fluid collection do not show SS on brain MRI . Furthermore, Morishima et al., reported a case with neither SS on brain MRI nor red blood cells in the CSF . These reports also suggest that hemosiderin deposition is not essential for developing brachial multisegmental amyotrophy in patients with ventral intraspinal fluid collection. Further studies are needed to elucidate the pathophysiology of brachial multisegmental amyotrophy accompanied with SS and dural tear.
We pathologically confirmed selective neuronal loss of both anterior horns and intermediate zone at the upper cervical to middle thoracic spinal gray matter in a patient with brachial multisegmental amyotrophy accompanied with SS, dural tear, and snake-eyes appearance on MRI. Extensive anterior horn damage in our patient may be due to dynamic compression induced by ventral intraspinal fluid collection.
|
Pathogenic role of Twist-1 protein in hydatidiform molar pregnancies and investigation of its potential diagnostic utility in complete moles | 9ba2c04c-76fa-45bc-914c-34425269bcab | 10053139 | Anatomy[mh] | Gestational trophoblastic disease encompasses a spectrum of pregnancy-related disorders, ranging from premalignant disorders of complete and partial hydatidiform mole, and the malignant disorders of invasive mole, choriocarcinoma, and the rare placental-site trophoblastic tumor . Hydatidiform mole (HM) refers to an abnormal pregnancy characterized by varying degrees of trophoblastic proliferation (both cytotrophoblasts and syncytiotrophoblasts) and vesicular swelling of placental villi associated with an absent or abnormal fetus/embryo. Its incidence is mainly affected by geographical location. South-eastern Asia, the Middle East, and South America show the highest cases, whereas it is the lowest in Western Europe and North America . Two syndromes of HM have been described based on both morphologic and cytogenetic criteria. Complete hydatidiform mole is the category without an embryo/fetal tissue exhibiting diandric diploid karyotype . Partial hydatidiform moles on the other hand can demonstrate evidence of fetal development with a diandric triploid karyotype . Clinical symptoms of a molar pregnancy generally include vaginal bleeding, large uterine size, severe vomiting, premature pre-eclampsia, absent fetal heartbeat, and a significant increase in serum ß-hCG levels. CMs previously presented in the late first or early second trimester with characteristic snowstorm-like ultrasonography. However, today, symptomatic women with vaginal bleeding are more likely to refer for abortion in the early stages of pregnancy due to the widespread use of sonography and quantitative measurement of ß-hCG that allow clinicians to make an earlier diagnosis, with most evacuated at the gestational age of 8 to 12 weeks. Partial moles (PM) are mostly clinically presented as missed abortions with a small uterus . Sonography findings in the first trimester of molar pregnancy are less clear, with fair to moderate interobserver agreement . Histopathological examination remains the cornerstone of the diagnosis of HM. CM is known by hydropic villi with cistern formation, trophoblastic proliferation with abnormal distribution, and loss of polarity . PM microscopic diagnosis is based on the identification of a mixture of two villous populations including small fibrotic and enlarged irregularly shaped villi with mild to moderate circumferential trophoblastic proliferation . However, the diagnosis and classification of HM have become increasingly difficult because HMs are now commonly evacuated at an earlier stage and do not satisfy the well-established classic morphological features. The diagnosis of HM based on morphology alone is susceptible to inter-observer variability and therefore suboptimal diagnostic reproducibility . Differentiating a molar pregnancy from non-molar specimens (NMS) and the classification of HM as CM (including early CM), PM, or hydropic miscarriage is important for clinical practice and the outcome ; trophoblastic neoplasia (invasive mole or choriocarcinoma) follows CM in 15–20% of cases ; while less than 5% of PMs will develop postmolar gestational trophoblastic neoplasia (GTN); metastases occur rarely and the diagnosis of choriocarcinoma has not been confirmed after a PM . The histological, clinical, and ultrasonographical manifestations in different molar gestations usually overlap, leading to a demanding process of making a final diagnosis. It is crucial to investigate new biomarkers and molecular techniques in clinical trials to establish a better routine practice of differentiating these subtypes of molar gestation . The immunohistochemical study can play an important role in diagnosis. P57 (a paternally imprinted, maternally expressed gene), if absent can make a diagnosis in favor of CM, versus its positivity in hydropic abortions and PMs . Flow cytometry, cytogenetic study, and short tandem genomic imprinting can also assist in distinguishing diploid complete from triploid PMs . Twist-1 is an essential protein in epithelial-mesenchymal transition (EMT), especially in cancer formation and progression to invasive and metastatic tumors, and is notably expressed in carcinosarcomas. It is also required for trophoblastic differentiation, placental formation, gastrulation, mesoderm formation, and neural crest migration . A recent study has shown that Twist-1 can be the marker of choice in the CM/PM differentiation .
In this retrospective cross-sectional study, 87 cases were chosen from uterine curettage specimens with a diagnosis of molar pregnancy; 47 cases of CM and 40 cases of PM were randomly selected based on histopathological criteria, and by electronic search in the hospital information system (HIS) of pathology department of cancer institute of Imam Khomeini Hospital Complex (IKHC), Tehran, Iran from 2014 to 2017. Clinical characteristics of patients such as age, gestational age, the number of previous pregnancies, and serum ß-hCG level are determined based on patients’ clinical records. The study was approved by the local ethics committee of our university (IR.TUMS.IKHC.REC.1400.089). Patients with the diagnosis of complete or partial mole which had been confirmed by P57 immunohistochemistry (IHC) entered the study; while the cases with insufficient and inappropriate pathology samples, cases with non-diagnostic IHC results, and the patients with unavailable clinical information were excluded. The cases were reviewed by two expert gynecopathologists (Dr. Soheila Sarmadi and Dr. Fatemeh Nili). Strict morphological criteria for the selection of HMs were applied. For the distinction of PM and CM, in addition to morphological findings, an IHC study for P57 was done. After the selection of the appropriate block and preparing 3- m -thick unstained slides, overnight drying at 60 ºC, deparaffinization, rehydration, and heat-induced epitope retrieval were done. After blockage of endogenous peroxidase, the specimens were incubated with primary antibody P57 (MAD-000721 QD P57 9KP10, Master diagnostic: Spain) and Twist-1 (Mouse monoclonal antibody, 10E4E6, dilution: 1/100, Boston: USA) and finally Master Polymer Detection kit (HRP). The p57 immunoreactivity was interpreted as satisfactorily negative when villous stromal cells and cytotrophoblasts were entirely negative or demonstrated only limited expression (nuclear staining < 10% of these cell types) with the presence of internal positive control (maternal decidua and/or intermediate trophoblastic cells exhibiting nuclear expression of p57). Positive p57 immunoreactivity was interpreted when the extent of staining in these cell types was extensive or diffuse. Nuclear staining of Twist-1 in villous stromal cells and syncytiotrophoblasts was evaluated and analyzed as the following variables: the percentage of positive cells (PS), intensity of nuclear staining (IS): score 0 (no staining); score 33% (weak nuclear staining), score 66% (moderate nuclear staining) and score 100% (strong nuclear staining). A comprehensive score (CS) was calculated by multiplying IS and PS as described above. Using Receiver-Operating Characteristic (ROC) curves, the best cut-off values for differentiation of molar pregnancies, the highest sensitivity, specificity, and positive and negative predictive values were calculated. The quantitative and qualitative variables were compared by Chi-square and independent sample T-tests, respectively. P-values less than 0.05 were considered significant.
Demographic and clinical findings 87 women affected by molar pregnancy were included in the study. The mean age of the subjects was 28.22 ± 6.98 years ranged 17 to 52 years with a mean gestational age of 73.75 ± 13.66 days. The mean serum ß-hCG was 99935.84 ± 10,684,057 (median: 57,810, range: 1395-452614) IU/mL. Regarding gravidity, 38 (43.2%) were null gravid, 23 (26.1%) were primigravid and others were multigravid. Four patients (4.5%) had a previous history of molar pregnancy. Also, 26 (31.7%) expressed experiencing abortion. The mean age of patients and gestational age were not statistically different in CM and PM. Serum ß-hCG level was significantly higher in CM. Ultrasonographic examination diagnosed CMs with a higher prevalence in comparison to PMs. Past history of abortion and molar pregnancy was not significantly different in CM and PMs (Table ). Twist-1 immunoreactivity in villous stromal cells The mean percentage of positive villous stromal cells in CM and PM pregnancies was 69.04 ± 18.98 and 32.75 ± 22.55, respectively (p < 0.001). The weak intensity of stromal cells staining was found in 10.6% and 75.0%, moderate in 59.6% and 12.5%, and strong in 29.8% and 12.5% of CM and PMs, respectively indicating a significantly higher intensity in CMs (p < 0.001). Similarly, the mean of stromal cell comprehensive score (CS) in complete and partial mole was 52.97 ± 23.81 and 19.16 ± 24.88, respectively which demonstrates a statistically significant difference (p < 0.001). According to ROC curve analysis, both percentages of stromal (AUC = 0.858, 95%CI: 0.770 to 0.946) and CS stromal cells (AUC = 0.846, 95% CI: 0.751 to 0.940) could differentiate complete from partial molar pregnancy (Fig. ). Here, Twist-1 positive staining in more than 50% of stromal cells can differentiate CM and PM with 89.5% sensitivity, 75% specificity, 80.7% positive predictive value (PPV), and 85.7% negative predictive value (NPV) (Table ). Twist-1 immunoreactivity in syncytiotrophoblasts The mean percentage of syncytiotrophoblast staining with Twist-1 in CM and PM pregnancies was 7.87 ± 9.31 and 27.75 ± 21.48, respectively (p < 0.001). Thirty-four (34%) of CMs and 7.5% of PMs were negative. A weak intensity in 59.6% and 42.5% and moderate intensity in 6.4% and 50.0% of CM and PMs were identified, respectively (p < 0.001). None of the cases reveal strong staining. Similarly, the mean CS was 3.11 ± 5.14 and 15.87 ± 14.91 in syncytiotrophoblasts of CM and PM, respectively (p < 0.001). According to ROC curve analysis, percentage (AUC = 0.782, 95%CI: 0.684 to 0.881) and CS of syncytiotrophoblast cells (AUC = 0.794, 95%CI: 0.697 to 0.891) could differentiate complete from partial molar pregnancy (Fig. ). Negative or weak nuclear staining in less than 10% of syncytiotrophoblasts can differentiate CM and PM with 82.9% sensitivity, 60% specificity, 70.9% PPV, and 75% NPV (Table ).
87 women affected by molar pregnancy were included in the study. The mean age of the subjects was 28.22 ± 6.98 years ranged 17 to 52 years with a mean gestational age of 73.75 ± 13.66 days. The mean serum ß-hCG was 99935.84 ± 10,684,057 (median: 57,810, range: 1395-452614) IU/mL. Regarding gravidity, 38 (43.2%) were null gravid, 23 (26.1%) were primigravid and others were multigravid. Four patients (4.5%) had a previous history of molar pregnancy. Also, 26 (31.7%) expressed experiencing abortion. The mean age of patients and gestational age were not statistically different in CM and PM. Serum ß-hCG level was significantly higher in CM. Ultrasonographic examination diagnosed CMs with a higher prevalence in comparison to PMs. Past history of abortion and molar pregnancy was not significantly different in CM and PMs (Table ).
The mean percentage of positive villous stromal cells in CM and PM pregnancies was 69.04 ± 18.98 and 32.75 ± 22.55, respectively (p < 0.001). The weak intensity of stromal cells staining was found in 10.6% and 75.0%, moderate in 59.6% and 12.5%, and strong in 29.8% and 12.5% of CM and PMs, respectively indicating a significantly higher intensity in CMs (p < 0.001). Similarly, the mean of stromal cell comprehensive score (CS) in complete and partial mole was 52.97 ± 23.81 and 19.16 ± 24.88, respectively which demonstrates a statistically significant difference (p < 0.001). According to ROC curve analysis, both percentages of stromal (AUC = 0.858, 95%CI: 0.770 to 0.946) and CS stromal cells (AUC = 0.846, 95% CI: 0.751 to 0.940) could differentiate complete from partial molar pregnancy (Fig. ). Here, Twist-1 positive staining in more than 50% of stromal cells can differentiate CM and PM with 89.5% sensitivity, 75% specificity, 80.7% positive predictive value (PPV), and 85.7% negative predictive value (NPV) (Table ).
The mean percentage of syncytiotrophoblast staining with Twist-1 in CM and PM pregnancies was 7.87 ± 9.31 and 27.75 ± 21.48, respectively (p < 0.001). Thirty-four (34%) of CMs and 7.5% of PMs were negative. A weak intensity in 59.6% and 42.5% and moderate intensity in 6.4% and 50.0% of CM and PMs were identified, respectively (p < 0.001). None of the cases reveal strong staining. Similarly, the mean CS was 3.11 ± 5.14 and 15.87 ± 14.91 in syncytiotrophoblasts of CM and PM, respectively (p < 0.001). According to ROC curve analysis, percentage (AUC = 0.782, 95%CI: 0.684 to 0.881) and CS of syncytiotrophoblast cells (AUC = 0.794, 95%CI: 0.697 to 0.891) could differentiate complete from partial molar pregnancy (Fig. ). Negative or weak nuclear staining in less than 10% of syncytiotrophoblasts can differentiate CM and PM with 82.9% sensitivity, 60% specificity, 70.9% PPV, and 75% NPV (Table ).
Complete and partial mole, as the most common types of GTDS, are genetically different disorders. Despite the different morphological findings, there are some overlapping histopathological features. The tumors are treated similarly, but the behavior is different with a higher probability of invasion in the CM. In this way, the patients with CMs are followed more rigorously than the patients with PMs. The agreement between pathologists in differentiating the subtypes of molar pregnancies is fair . In some cases, the use of additional diagnostic tests such as IHC will be required. The high sensitivity and specificity of the P57 marker for differentiating complete and PMs have been confirmed in different studies . However, based on the recent 2020 WHO classification of female genital tract tumors, molecular genotyping is essential to confirm the diagnosis of PMs . Unfortunately, genetic testing of PM samples is expensive and not available in all centers. In this study, the expression level of the Twist-1 marker in stromal cells of villi as well as syncytiotrophoblasts was evaluated quantitatively (percentage of positive cells) and qualitatively (staining intensity) separately and as a total comprehensive score. The percentage, intensity, and overall comprehensive scores of Twist-1 expressions were significantly higher in the villi stromal cells in the CM and these values were lower in the syncytiotrophoblasts (p < 0.001) (Figs. , ). Analysis of ROC curves revealed moderate to strong staining in more than 50% of villous stromal cells as the best cut-off value, which can differentiate CM and PM with 89.5% sensitivity and 75% specificity. In syncytiotrophoblasts of CM, negative or weak staining in less than 10% of syncytiotrophoblasts, can distinguish CM and PM with 83% sensitivity and 60% specificity. There are limited studies about the diagnostic value of Twist-1 for the diagnosis of molar pregnancies. In a study by Rabab A Moussa (2018), they assessed whether the expression of Twist-1, Ki-67, and E-cadherin can guide the differential diagnosis of CM, PM, and hydropic abortion (HA). Differential expression of Twist-1, Ki-67, and E-cadherin was analyzed in gestational products from 55 cases of CM, PM, and HA using immunohistochemistry. Prior to analysis, the studied cases were confirmed by flow cytometric assessment of DNA ploidy and p57 immunostaining. They suggested that a positive stromal score of more than 73 can distinguish CM from PM and HA with 100% sensitivity, 100% specificity, 100% positive predictive value (PPV), and 100% negative predictive value (NPV). In their study, syncytiotrophoblasts in none of the CM cases showed nuclear staining . Although the results of both studies show high diagnostic accuracy of the Twist-1 marker in differentiating the subtypes of molar pregnancy, the cut-off values are different. The sensitivity and specificity of diagnosis in our study are lower than the study of Moussa et al. Since the qualitative values of Twist-1 expression intensity have been used to analyze the results and the scoring of this variable can be different between different observers, this can affect the numerical values of the data and the overall cut-off. The sample size examined in our study was more than the previous study. The potential differences in the efficacy of the antibodies and IHC staining protocols may also influence the results. Unfortunately, in our study, molecular evaluation was not available to confirm the cases with PM diagnosis. Strict morphological criteria and IHC staining for the P57 marker were used to detect molar pregnancies and differentiate complete from PMs. Only those cases that were agreed upon by two expert gynecological pathologists were included. There may be a possibility of Hydropic abortion misdiagnosis with PMs. In spite of that, by using strict morphological criteria and confirmatory p57 IHC staining, we are undoubted about the diagnosis of CM cases. The results of both studies, in addition to the introduction of a new IHC marker (Twist-1), raise the hypotheses regarding the different pathogenesis of molar diseases. Twist-1 is a transcriptional regulator that plays a role in mesodermal-derived tissues such as the uterus in stem cell differentiation . The role of this molecule in the formation of decidual tissue in the uterus has been suggested in previous studies . On the other hand, the molecule is a negative regulator of cytokine expression that is involved in cell-to-cell adhesion proteins such as E-cadherin and epithelial-mesenchymal transition (EMT). The reduction of E-cadherin expression and the EMT process plays an important role in creating the aggressiveness of trophoblastic cells to penetrate the uterine wall and form the placental tissue . Aberrant expression of E-cadherin in invasive moles compared to non-invasive moles has been suggested in previous studies . Increased expression of Twist-1 in the process of carcinogenesis and the development of EMT properties, resulting in an increased invasive capacity, metastasis, and poor prognosis in various tumors, have been investigated . In our study and the study of Moussa et al., a strong and significant increase in the expression of the Twist-1 marker was observed in the stromal cells of villi in CMs compared to the cases of PMs. Risk of invasion and developing choriocarcinoma in CMs is three or four times higher than PMs . Excessive proliferation, atypia and mitosis of trophoblastic cells in CMs is correlated with their more aggressive behavior . The results of our study suggest an additional pathogenic mechanism indicating more invasive nature of the villous stromal cells in CMs. The opposite result was observed in the expression of Twist-1 in the syncytiotrophoblasts. In CMs, this expression was significantly lower than in PMs. The Twist-1 molecule is involved in the fusion of cytotrophoblasts and the formation of syncytiotrophoblasts . Syncytiotrophoblasts play an effective role in the transfer of nutrients, gases, and waste products between the mother and the fetus. Dysregulation of this process has been suggested in pregnancy complications such as pre-eclampsia and IUGR and recurrent pregnancy loss . The results of our study are suggestive of disturbance in this process in CMs and a more effective event in PMs. Despite high sensitivity and specifity of Twist-1, there is no superiority for diagnostic utility, compared with the previously well-known P57 IHC marker. Banet et al. demonstrated high correlation of P57 expression with molecular genotyping results in CMs. P57 is almost always positive in CMs (with more than 99% accuracy). So it’s an extremely reliable, easy to perform or interpret method for the diagnosis of CMs in routine practice . Distinction of PMs and hydropic abortions is a more challenging concern in daily practice of the pathologists. Due to our limitation in selection and confirmation of PMs, we couldn’t address this issue. In the study of Moussa et al. villous stromal cells in PMs expressed more significant percentage of Twist-1. But expression in syncytiotrophoblats was almost similar. Nowadays, molecular genotyping which is an essential factor for the diagnosis of PMs is the most reliable diagnostic method.
A higher expression of Twist-1 in villous stromal cells of hydatidiform moles is a sensitive and specific marker for the diagnosis of CMs. An elevated expression of this marker in villous stromal cells suggests another pathogenic mechanism for more aggressiveness of CMs in addition to the characteristics of trophoblast cells. The opposite result was obtained in the expression of Twist-1 in the syncytiotrophoblasts, compatible with defects in the process of formation of these supportive cells in CMs.
|
From Biology to Diagnosis and Treatment: The Ariadne’s Thread in Cancer of Unknown Primary | 1a5ea54d-faf5-4116-bec5-e70dee8fb140 | 10053301 | Anatomy[mh] | Cancer of unknown primary origin (CUP) is a diverse category of cancers with varying clinical and histological characteristics for which no original tumour site has been identified despite a thorough diagnostic workup . It is the seventh to the eighth most frequent malignant disease for both sexes, whilst it has risen to become the fourth most common cause of cancer-related death, accounting for 1–3% of all human cancers . Recently, a consensus was reached on the first diagnostic layer for CUP. History and physical examination; full blood count; along with serum markers, a CT scan of the chest/abdomen/pelvis, and a biopsy of the most accessible lesion, followed by immunohistochemical testing should be the starting point. On the other hand, the symptom-guided magnetic resonance imaging (MRI) or ultrasound, a positron emission tomography/computerised tomography (PET/CT) scan, targeted gene panels, immunohistochemical markers, and whole genome sequencing remain debatable . With regard to organisational recommendations, namely, the National Comprehensive Cancer Network (NCCN), American Society of Clinical Oncology (ASCO), European Society for Medical Oncology (ESMO), and Spanish Society of Medical Oncology (SEOM), the histological evaluation is standard and is based on morphology and algorithmic immunohistochemistry (IHC) . Traditionally, patients with CUP can be categorised into two prognostic groups according to their clinical and pathologic presentations. Those with a constellation of manifestations that can be assigned to a primary account for around 15–20% of CUP and are treated accordingly. The remaining patients have an unfavourable prognosis and are commonly treated with empirical chemotherapy . The research in the field of CUP is mainly directed towards the development of molecular diagnostics to facilitate accurate prediction of the primary site, rather than to further investigate the already existing chemotherapeutics. This review aims to highlight different aspects around CUP such as the biology, clinicopathological subsets, diagnostic work-up, and therapeutic strategies.
From the epidemiological perspective, around 8600 new CUPs are diagnosed each year in the United Kingdom, making this the fifteenth most prevalent cancer . In 2012, Denmark had around 338 new cases per 100,000 people, compared to 284 in Germany, 296 in Canada, and 318 in the United States . Since the early 1980s, CUP incidence rates in the United States have been falling at a pace of 3.6% per year over the previous two decades. Since 1973, the rate of non-microscopically confirmed CUP has decreased by 2.6% every year . CUP incidence rates in Scotland climbed from 7 to 8 per 100,000 in the 1960s to a peak of 14–18 per 100,000 in the early to mid-1990s, before falling precipitously to 8 per 100,000 in 2009 . Between 1999 and 2017, the incidence of CUP decreased in Korea. CUP incidence probably decreased due to improved diagnostics, which have led to better identification of the primary culprit. Genomic profiling testing may help in identifying molecular signatures in CUP patients and enable targeted treatment . Patients aged 80 and up had the highest risk of occurrence. The survival rate climbed from 14.2% in 1999–2002 to 27.3% in 2013–2017 . In Sweden, it is estimated that the age-standardised incidence increased from 10 per 100,000 in the early 1960s to 16 per 100,000 around the year 2000 for both males and females .
Any degree of smoking raises the chance of CUP with respiratory system metastases to 4.9%. In a study, smoking was strongly linked to an increased risk of CUP, with a relative risk of 3.66% for ongoing, heavy smokers (>25 cigarettes/day) compared to never smokers (standardised for other risk factors), and a relative risk of 5.12% for patients with CUP who died within less 12 months since the diagnosis . Other risk factors include alcohol intake, body mass index (BMI), waist circumference, diabetes, and a poor educational level or socioeconomic position . The pathophysiology of familial CUP is characterised by the existence of a genetic vulnerability that puts family members of a CUP patient at a higher risk of CUP and other tumours . Relatives of CUP patients are more likely to develop CUP as well as other malignant cancers, especially of the lung, pancreas, and colon .
Generally, the ongoing improved knowledge of the biology of several cancers has enabled a more-accurate classification, diagnosis, and prognosis, as well as providing guidance in the tailoring of specific therapies. However, in CUP, the data are not very mature. The early dissemination of tumour cells implies subsequent independent progression of the primary tumour and metastases, under the selection pressure of the immune system. The clinical observation of the high systemic relapse rate in CUP patients with localised disease treated with curative intention with surgery and/or radiotherapy supports that model. Still, molecular platforms represent a key component of the diagnostic work-up and clinical management . 4.1. Chromosomal Abnormalities Many tumours have an abnormal number of chromosomes, a condition known as aneuploidy. Chromosomal instability (CIN), a mechanism of continual chromosomal missegregation, is a common cause of aneuploidy . Oncogenesis necessitates huge numbers of genetic changes that cannot be caused by the regular rate of mutation, necessitating some sort of innate genomic instability to produce a mutator phenotype . In a survey of 152 individuals with metastatic adenocarcinoma or undifferentiated CUP, 106 (70%) had clusters of cells with aberrant cellular DNA composition, while the rest were diploid. The prevalence of aneuploidy was comparable between sexes and had no discernible link with the various forms of metastatic invasion . CIN is a known driver of early dissemination and aggressive behaviour of CUP. In a recently published study, researchers investigated the genomic information of CUP samples analysed using a hybridisation-capture-based next-generation sequencing (NGS) assay in 410 cancer-associated genes . CUP samples presented mainly a very low aneuploidy score (AS) (63.0%; n = 92), followed by intermediate and low AS (16.4%; n = 24 each) and high AS (4.1%; n = 6). The high-intermediate AS groups lacked an enriched genetic alteration, and the presence of a TP53 or KRAS mutation did not correlate with a high AS. This differs from what was previously known about these mutations in aneuploid cancers. The researchers concluded that CUP patients present individual gene alterations implicated in immune evasion and resistance to ICI, but further clinical investigations are needed to clarify the interplay between CIN, point mutations, and the immune system. 4.2. Oncogenes and Proteins Oncogenes play a critical role in cancer formation by either overexpression or amplification. The occurrence of protein overexpression or oncogene gene alterations in CUP is comparable to rates observed in metastatic cancers of known primary origin, with the expected variability . PI3K , Ras , p53 , PTEN , Rb , and p16INK4a are among the oncogenes and tumour suppressors that are often mutated in cancerous cells . The mutation trends of the tumour suppressor TP53 (which encodes p53), ataxia telangiectasia mutated ( ATM ), and cyclin-dependent kinase inhibitor 2A ( p16INK4A and p14ARF encoded by CDKN2A) corroborate the oncogene-induced DNA replication stress concept . Studies that assessed archival tumour tissues with NGS revealed alterations of TP53 (38–55%), Ras (18–20%), CDKN2A (19%), MYC (12%), ARID1A (11%), and PIK3CA (9–14%) . EGFR is widely expressed in CUP (74–75%), according to immunohistochemical studies, but c-KIT and HER2/neu are seldom active (overexpression in 4–27%) . No meaningful link has been found between EGFR expression level and patient outcomes . However, the expression of EGFR is associated with sensitivity to platinum-based regimens. CUP patients with overexpression of HER2/neu have mostly supradiaphragmatic disease, whereas histologically they are predominantly poorly differentiated adenocarcinomas. Given that HER2/neu amplifications have not been identified as driver mutations, very little response data have been published in HER2/neu-altered CUP. Regardless of the fact that RAS-pathway-activating mutations are described in approximately 20% of patients with CUP, there is not any prognostic significance . Generally, RAS-driven cancers are considered to be among the most difficult to treat; however, they are potentially targetable with MEK inhibitors . To assess whether sensitivity to trametinib could be predicted in CUP cases, as well as to provide a tool to stratify patients for trials, an original “trametinib response signature” has been described in the literature . This signature anticipated the experimentally assessed response to trametinib in agnospheres and was retrieved also in the matched patients’ tissues. It finally predicted the response in a retrospective cohort of CUP cases. Interestingly, CUP sensitivity predicted by the trametinib signature approximates that of BRAF-mutated melanoma. Although less frequent, BRAF V600E mutations were found in 1.6% (7 out of 442) in a large series of patients with CUP . Circulating tumour DNA revealed that 80% of CUP patients (353 out of 442) had detectable alterations and 66% (290 out of 442) had at least one characterised alteration in the above-mentioned case series. Among these patients, alterations in MAPK and PI3K signalling were identified in 31.2% and 18.1%, respectively . Genomic DNA is continuously confronted with a large number of DNA lesions. It is required for the cells to counteract DNA damage by activating the DNA damage response (DDR) in view of keeping the genome stable and securing cellular homeostasis . Several DDR pathways have evolved in cells to repair different types of damage. BRCA1 and BRCA2 tumour suppressor genes play an important role in DDR, and mutations in these genes confer a high risk of breast and ovarian cancers . BRCA1 mutation carriers are at high risk of CUP (relative risk (RR) 3.45, 95% CI 2.35–5.07, p < 0.001) . 4.3. Angiogenesis A cancer cell must be able to split from the main tumour; penetrate through surrounding tissues and basement membranes; and then enter and survive in the circulation, lymphatics, or peritoneal space to colonise a secondary location. This is followed by extravasation into surrounding tissue, survival in the alien milieu, proliferation, and angiogenesis activation, all while avoiding apoptosis or an immune response . Angiogenesis includes several stages, such as proteolytic degradation of the basement membrane and surrounding extracellular matrix, endothelial cell proliferation and migration, and finally tube formation . Cancer cells rely heavily on this pathway for development, survival, and invasion. The activation of an angiogenic switch is essential for a lesion to expand above a certain length . It seems that CUP presents an angiogenic incompetence at the primary site that limits the development of the primary tumour. Endogenous positive angiogenic factors include vascular endothelial growth factor (VEGF), platelet-derived growth factor (PDGF), fibroblast growth factors (FGFs), epidermal growth factor (EGF), transforming growth factor (TGF), matrix metalloproteinases (MMPs), tumour necrosis factor (TNF), and angiopoietins, whereas endogenous negative angiogenic factors are interleukins, interferon, tissue inhibitors of metalloproteinases (TIMP), angiostatin, and endostatins . The role of the angiogenesis within the biology of CUP is supported by the observation of its absence in primary tumours inducing dormancy, whilst it is present at metastatic sites. However, VEGF expression is not associated with prognosis, excluding the positive association between VEGF and the density of micro-vessels. A study reported that regardless of the overexpression of VEGF in 26% of a CUP case series, there was not any prognostic impact of CD34 and VEGF on the survival . Similarly, the comparison between 39 liver metastases from patients with CUP versus 30 liver metastases from colon and breast cancer did not reveal differences in the density of micro-vessels; both groups exhibited high angiogenic activity . Finally, a study demonstrated low expression of VEGF protein in patients with CUP. Fifty patients with squamous carcinomas metastatic to the cervical lymph nodes were compared with 52 patients with metastases from a known primary. The authors proposed a pattern independent of angiogenesis of metastatic spread for squamous CUP metastasising to the cervical lymph nodes . 4.4. Evasion of Immune Destruction Tumours avoid immune surveillance by generating immunosuppressive cytokines, including TGF-β. TGF-β has been shown to selectively block the production of five cytolytic gene products, namely, perforin, granzyme A, granzyme B, Fas ligand, and interferon-γ, which are together involved in cytotoxic T-lymphocytes (CTL)-mediated tumour cytotoxicity. TGF-β-activated Smad and ATF1 transcription factors bind to their promoter regions, repressing granzyme B and interferon-γ . In a study, programmed cell death-1 (PD-1) expression was detected in the tumour-infiltrating lymphocytes of 58.7% of patients with CUP, whereas programmed death-ligand 1 (PD-L1) expression was found in 22.5% of the CUP specimens . Within the context of the immune microenvironment markers, tumour mutation load and microsatellite instability were high in 11.8% and 1.8% of CUP patients, respectively . Microsatellite instability was associated with a high tumour mutational burden and represented a predictive biomarker of response to immune checkpoint inhibitors in several malignancies . Plasma-based circulating cell-free DNA (cfDNA) assays identified mutations in the DDR protein MLH1 (mutL homologue 1) in 1.6% of CUP patients .
Many tumours have an abnormal number of chromosomes, a condition known as aneuploidy. Chromosomal instability (CIN), a mechanism of continual chromosomal missegregation, is a common cause of aneuploidy . Oncogenesis necessitates huge numbers of genetic changes that cannot be caused by the regular rate of mutation, necessitating some sort of innate genomic instability to produce a mutator phenotype . In a survey of 152 individuals with metastatic adenocarcinoma or undifferentiated CUP, 106 (70%) had clusters of cells with aberrant cellular DNA composition, while the rest were diploid. The prevalence of aneuploidy was comparable between sexes and had no discernible link with the various forms of metastatic invasion . CIN is a known driver of early dissemination and aggressive behaviour of CUP. In a recently published study, researchers investigated the genomic information of CUP samples analysed using a hybridisation-capture-based next-generation sequencing (NGS) assay in 410 cancer-associated genes . CUP samples presented mainly a very low aneuploidy score (AS) (63.0%; n = 92), followed by intermediate and low AS (16.4%; n = 24 each) and high AS (4.1%; n = 6). The high-intermediate AS groups lacked an enriched genetic alteration, and the presence of a TP53 or KRAS mutation did not correlate with a high AS. This differs from what was previously known about these mutations in aneuploid cancers. The researchers concluded that CUP patients present individual gene alterations implicated in immune evasion and resistance to ICI, but further clinical investigations are needed to clarify the interplay between CIN, point mutations, and the immune system.
Oncogenes play a critical role in cancer formation by either overexpression or amplification. The occurrence of protein overexpression or oncogene gene alterations in CUP is comparable to rates observed in metastatic cancers of known primary origin, with the expected variability . PI3K , Ras , p53 , PTEN , Rb , and p16INK4a are among the oncogenes and tumour suppressors that are often mutated in cancerous cells . The mutation trends of the tumour suppressor TP53 (which encodes p53), ataxia telangiectasia mutated ( ATM ), and cyclin-dependent kinase inhibitor 2A ( p16INK4A and p14ARF encoded by CDKN2A) corroborate the oncogene-induced DNA replication stress concept . Studies that assessed archival tumour tissues with NGS revealed alterations of TP53 (38–55%), Ras (18–20%), CDKN2A (19%), MYC (12%), ARID1A (11%), and PIK3CA (9–14%) . EGFR is widely expressed in CUP (74–75%), according to immunohistochemical studies, but c-KIT and HER2/neu are seldom active (overexpression in 4–27%) . No meaningful link has been found between EGFR expression level and patient outcomes . However, the expression of EGFR is associated with sensitivity to platinum-based regimens. CUP patients with overexpression of HER2/neu have mostly supradiaphragmatic disease, whereas histologically they are predominantly poorly differentiated adenocarcinomas. Given that HER2/neu amplifications have not been identified as driver mutations, very little response data have been published in HER2/neu-altered CUP. Regardless of the fact that RAS-pathway-activating mutations are described in approximately 20% of patients with CUP, there is not any prognostic significance . Generally, RAS-driven cancers are considered to be among the most difficult to treat; however, they are potentially targetable with MEK inhibitors . To assess whether sensitivity to trametinib could be predicted in CUP cases, as well as to provide a tool to stratify patients for trials, an original “trametinib response signature” has been described in the literature . This signature anticipated the experimentally assessed response to trametinib in agnospheres and was retrieved also in the matched patients’ tissues. It finally predicted the response in a retrospective cohort of CUP cases. Interestingly, CUP sensitivity predicted by the trametinib signature approximates that of BRAF-mutated melanoma. Although less frequent, BRAF V600E mutations were found in 1.6% (7 out of 442) in a large series of patients with CUP . Circulating tumour DNA revealed that 80% of CUP patients (353 out of 442) had detectable alterations and 66% (290 out of 442) had at least one characterised alteration in the above-mentioned case series. Among these patients, alterations in MAPK and PI3K signalling were identified in 31.2% and 18.1%, respectively . Genomic DNA is continuously confronted with a large number of DNA lesions. It is required for the cells to counteract DNA damage by activating the DNA damage response (DDR) in view of keeping the genome stable and securing cellular homeostasis . Several DDR pathways have evolved in cells to repair different types of damage. BRCA1 and BRCA2 tumour suppressor genes play an important role in DDR, and mutations in these genes confer a high risk of breast and ovarian cancers . BRCA1 mutation carriers are at high risk of CUP (relative risk (RR) 3.45, 95% CI 2.35–5.07, p < 0.001) .
A cancer cell must be able to split from the main tumour; penetrate through surrounding tissues and basement membranes; and then enter and survive in the circulation, lymphatics, or peritoneal space to colonise a secondary location. This is followed by extravasation into surrounding tissue, survival in the alien milieu, proliferation, and angiogenesis activation, all while avoiding apoptosis or an immune response . Angiogenesis includes several stages, such as proteolytic degradation of the basement membrane and surrounding extracellular matrix, endothelial cell proliferation and migration, and finally tube formation . Cancer cells rely heavily on this pathway for development, survival, and invasion. The activation of an angiogenic switch is essential for a lesion to expand above a certain length . It seems that CUP presents an angiogenic incompetence at the primary site that limits the development of the primary tumour. Endogenous positive angiogenic factors include vascular endothelial growth factor (VEGF), platelet-derived growth factor (PDGF), fibroblast growth factors (FGFs), epidermal growth factor (EGF), transforming growth factor (TGF), matrix metalloproteinases (MMPs), tumour necrosis factor (TNF), and angiopoietins, whereas endogenous negative angiogenic factors are interleukins, interferon, tissue inhibitors of metalloproteinases (TIMP), angiostatin, and endostatins . The role of the angiogenesis within the biology of CUP is supported by the observation of its absence in primary tumours inducing dormancy, whilst it is present at metastatic sites. However, VEGF expression is not associated with prognosis, excluding the positive association between VEGF and the density of micro-vessels. A study reported that regardless of the overexpression of VEGF in 26% of a CUP case series, there was not any prognostic impact of CD34 and VEGF on the survival . Similarly, the comparison between 39 liver metastases from patients with CUP versus 30 liver metastases from colon and breast cancer did not reveal differences in the density of micro-vessels; both groups exhibited high angiogenic activity . Finally, a study demonstrated low expression of VEGF protein in patients with CUP. Fifty patients with squamous carcinomas metastatic to the cervical lymph nodes were compared with 52 patients with metastases from a known primary. The authors proposed a pattern independent of angiogenesis of metastatic spread for squamous CUP metastasising to the cervical lymph nodes .
Tumours avoid immune surveillance by generating immunosuppressive cytokines, including TGF-β. TGF-β has been shown to selectively block the production of five cytolytic gene products, namely, perforin, granzyme A, granzyme B, Fas ligand, and interferon-γ, which are together involved in cytotoxic T-lymphocytes (CTL)-mediated tumour cytotoxicity. TGF-β-activated Smad and ATF1 transcription factors bind to their promoter regions, repressing granzyme B and interferon-γ . In a study, programmed cell death-1 (PD-1) expression was detected in the tumour-infiltrating lymphocytes of 58.7% of patients with CUP, whereas programmed death-ligand 1 (PD-L1) expression was found in 22.5% of the CUP specimens . Within the context of the immune microenvironment markers, tumour mutation load and microsatellite instability were high in 11.8% and 1.8% of CUP patients, respectively . Microsatellite instability was associated with a high tumour mutational burden and represented a predictive biomarker of response to immune checkpoint inhibitors in several malignancies . Plasma-based circulating cell-free DNA (cfDNA) assays identified mutations in the DDR protein MLH1 (mutL homologue 1) in 1.6% of CUP patients .
The minority of patients with CUP (15–20%) present with clinical and pathological features that can be attributed to a primary culprit . The favourable risk cancer subgroup comprises peritoneal adenocarcinomatosis of a serous papillary subtype, isolated axillary nodal metastases in females, squamous cell carcinoma involving nonsupraclavicular cervical lymph nodes, single metastatic deposit from unknown primary, neuroendocrine carcinomas of unknown primary, and men with blastic bone metastases and elevated prostate-specific antigen (PSA). The treatment of these patients is compatible with the corresponding primary guidelines for metastatic disease. Currently, new favourable subsets of CUP have emerged, including colorectal, lung, and renal CUP, which underly specific treatments . These patients generally harbour a chemosensitive disease and, as such, longer life expectancy. The remaining 80–85% of CUP patients are assigned to the unfavourable subset comprise two prognostic groups, according to the performance status (0 or 1) and lactate dehydrogenase (LDH) level . These patients do not respond well to the empiric broad-spectrum chemotherapy and therefore the median overall survival is approximately 6–10 months. As far as the unfavourable subset is concerned, the one-year survival rates in good- and poor-risk patients are 53% and 23%, respectively.
The diagnosis of CUP is established when a metastatic cancer is histologically confirmed in the absence of identifiable primary tumour site, despite the extensive diagnostic evaluation. Recent research has focused on using genomics and transcriptomics to identify the origin of the primary tumour, but it is still not always performed, especially in low-resource environments . The development of tissue of origin classifiers for the analysis and diagnostics of CUP using a whole genome sequencing dataset of both primary and metastatic tumours is still an effort in progress . 6.1. Pathology and Immunohistochemistry From the histological perspective, CUP is defined as well- or moderately differentiated adenocarcinomas, accounting for 50–70% of all cases, with poorly differentiated carcinomas and adenocarcinomas making up another 20–30%, and the remaining being squamous-cell carcinomas (5–8%) and undifferentiated malignant neoplasms (2–3%) with inability of light microscopy to distinguish among carcinomas, lymphomas, melanomas, and sarcomas . The diagnoses of neuroendocrine tumours, melanomas, and sarcomas can be based on immunoperoxidase staining. Indeed, standard morphology and IHC remain the main strategy for the identification of the primary tumour in patients with CUP. The technique involves the analysis of tissue sections with antibodies against particular tumour-specific antigens, structural tissue components, hormonal receptors, hormones, or antigens . In the first instance, IHC differentiates well- and moderately-differentiated adenocarcinomas, squamous cell carcinomas, carcinomas with neuroendocrine differentiation, poorly differentiated carcinomas, and undifferentiated neoplasms. In squamous cell carcinomas and neuroendocrine carcinomas, the use of cell differentiation markers is advised, especially when tumour morphology is heterogenous or poorly differentiated. The IHC detection of markers, such as vimentin, S100 family proteins, HMB45 antigen, or CD45, may classify part of CUPs as non-carcinomas—sarcomas, melanomas, or lymphomas—that can be treated appropriately . Nevertheless, in the CUPISCO trial, the misdiagnosis of CUP due to sarcomas and melanomas represented 1.6% and 5.6%, respectively, of the failure cases . The most commonly used markers for the staining of CUP are the keratin family members, CK7 and CK20, with CK7+/CK20− being the most common in CUP . In CK7+/CK20− cases, ER positivity, and GATA3 positivity in ER-negative cases primarily direct clinicians to the breast as a possible site of cancer origin, especially in patients with axillary lymph node metastases . Although TTF1 expression in a metastatic setting does not unquestionably prove a primary origin in the lung, all TTF1/napsin A-positive cases should be radiologically investigated to rule out lung primitivity. The expression of PAX8/WT1 is considered in order to investigate a possible gynaecological origin. Overall, in approximately one-third of CUP cases, the primary site is identifiable through the IHC staining panels . However, a consensus panel of IHC markers has not yet been established, whilst no single pathognomonic marker exists for a conclusive diagnosis. Moreover, IHC has limited value in the diagnosis of poorly differentiated cancers. When immunoperoxidase stains are inconclusive in young patients with poorly differentiated tumours, electron microscopy should be considered in their evaluation. The presence of pleomorphic neoplastic cells with cytoplasmic vacuolations and/or cytoplasmic, non-membrane bound, electron-dense deposits detected by electron microscopy may suggest the diagnosis of a poorly differentiated carcinomas. Cytogenetic studies may be useful for the evaluation of young patients with poorly differentiated carcinomas or undifferentiated lesions that are responsive to chemotherapy. Finally, neurosecretory granules are specifically detected by electron microscopy in neuroendocrine tumours. Limitations of the IHC are the lack of reproducibility, due to the IHC-generated preparations, along with methodological issues. Failure to ensure that samples are of a high quality can hinder subsequent image analysis processes, negatively impact on data quality, and in some cases prevent an image analysis study from proceeding. Research using digitised histopathology slides for the development of artificial intelligence algorithms has increased markedly over recent years. 6.2. Diagnostic Radiology Image-assisted technologies has revolutionised the diagnosis of CUP. CT and conventional MRI have both been used to locate lesions, considering the clinical manifestation of CUP. The diagnostic accuracy of CT scans is around 55% (36–74%), mainly in pancreatic, colorectal, and lung cancer, while MRI has a sensitivity of 70% in detecting primary breast cancers . However, the diagnosis can be challenging if the primary tumour is small in size or has regressed, hindering the diagnosis. In particular, these cases may be successfully facilitated with the 2-[ 18 F] fluoro-2-deoxy-d-glucose (FDG) PET/CT, but still the detection rate is around 40% . The most frequent primary sites identified by PET are lung (33%) and head and neck (27%), followed by pancreas, breast, and colon (4–5%). Finally, 68 Ga-DOTA-NOC receptor PET/CT is recommended for the identification of primary neuroendocrine tumours, along with their metastases . Mammography is recommended for female patients with metastatic adenocarcinomas involving axillary lymph nodes. In patients with mammographically occult breast cancer, breast MRI may be considered. 6.3. Endoscopy Endoscopy should be directed towards investigating specific symptoms and signs or when specific histopathological findings are available. Fiberoptic bronchoscopy is reasonable for patients with respiratory symptoms and/or expression of CK7 and TTF1, whereas colonoscopy should be requested for those with abdominal symptoms or occult blood in the stool and/or expression of CK7, CK20, and CDX2. The sensitivity and specificity of the endoscopies are generally low. 6.4. Serum Tumour Markers In almost 70% of CUP patients, more than one marker can be concomitantly elevated in a non-specific way. Routine request of cancer antigen 125 (CA 125), cancer antigen 15.3 (CA 15-3), carbohydrate antigen 19-9 (CA 19-9), and carcinoembryonic antigen (CEA) is not recommended due to lack of prognostic and/or predictive value . However, there are some clinical scenarios in which serum tumour markers may have some diagnostic value. Indeed, serum PSA should be evaluated in men with osteoblastic bone metastases, CA 125 in women with primary serous papillary peritoneal adenocarcinoma, and CA 15-3 in females with isolated axillary adenocarcinoma. Finally, a high level of thyroglobulin in patients with CUP and bone metastasis may indicate occult thyroid cancer . 6.5. Liquid Biopsy Improvements in nucleic acid sequencing technologies have enabled the detection of low quantities of tumour genetic material within the blood and show the potential to be both sensitive and specific to an individual’s tumour. These blood-based biomarkers include cfDNA, tumour microRNAs (miRNAs), and platelet-derived tumour mRNA, as well as analysis of DNA, RNAs, and protein expression from individual circulating tumour cells. Within this context, the use of liquid biopsies reduces the need for intrusive diagnostic biopsies and provides enough material to perform the diagnostic procedures. For instance, even though the presence of aberrant hypermethylation of tumour-suppressor genes in serum DNA has been detectable before the millennium, more sensitive and quantitative techniques for analysis of DNA methylation are required to expedite its incorporation in the clinical setting . In diffuse large-B-cell lymphoma, detection of aberrant DAPK1 methylation in cfDNA at the time of diagnosis is a positive prognostic biomarker, whilst in hepatocellular carcinoma, methylation of VIM is an early detection biomarker . Overall, there is evidence that the tissue of origin can be determined using cfDNA . 6.6. Molecular Profiling for the Tissue of Origin Molecular profiling technologies including microarray-based gene expression profiling, reverse transcriptase polymerase chain reaction, RNA sequencing, somatic gene mutation profiling with NGS, and DNA methylation profiling were used to define the primary culprit among patients with CUP. However, the implementation of tissue-of-origin classifiers in CUP is limited due to the absence of primary tumour. Some studies were conducted to validate predictions of the primary origin, on the basis of autopsy data, latent primary emergence, or IHC. Gene expression profiling has been directly compared to IHC, within known metastatic tumour types. Accuracy of gene expression profiling was 89%, compared with 83% for IHC when only one round of IHC determined the diagnosis, whereas in poorly differentiated cancers, such as CUP, the percentages were 83% and 67%, respectively . Nevertheless, only a limited number of studies have investigated the clinical outcomes of CUP patients, treated on the basis of gene expression predictions . Predicting the tissue-of-origin via molecular profiling is a debated topic within CUP, given that it is difficult to be molecularly classified in the absence of histological definition. However, the incorporation of molecular classifiers to the standard diagnostic workup may potentially identify atypical presentations of patients for whom site-specific therapies would be effective . PlexinB2 (PlxnB2) is a semaphorin receptor implicated in the regulation of cancer cell proliferation, invasiveness, and metastatic spreading . The G842C-PlxnB2 variant has been investigated in an effort to establish a proof of principle about the relevance of axon guidance genes in CUP. This mutation affected the conserved fold of an IPT domain, a moiety also found in Met and Ron oncogenic receptors . Notably, the large intracellular portion of the plexins does not contain a kinase domain or other classical signalling domains; nevertheless, it regulates the activity of monomeric GTPases, especially R-Ras, Rap-1, and RhoA. Moreover, plexins have been shown to couple with transmembrane tyrosine kinases such as ErbB2 and Met, triggering alternative noncanonical signalling cascades, especially in cancer cells . A recent study demonstrated that G842C-mutated PlxnB2 was competent for signalling, even in the absence of semaphorin stimulation . Moreover, although knocking down PlxnB2 expression in CUP cells bearing a wild-type receptor had no any functional impact, the metastatic cells carrying the G842C mutation were found to be dependent on this variant PlxnB2 to sustain self-renewal and proliferation in culture, along with tumorigenesis in mice. These data indicated that G842C-PlxnB2 may be considered a gain-of-function mutation. Members of the tyrosine kinase receptor family have been associated with plexin signalling in cancer cells. According to the study, PlxnB2 was found in complex with EGFR, and EGFR phosphorylation was enhanced in the presence of G842C-PlxnB2. Moreover, the greater invasiveness of CUP cells driven by the expression of the mutated plexin was abrogated by selective EGFR inhibitors, namely, cetuximab and erlotinib. These data provide evidence of the functional involvement of an unexpected aberrant signalling pathway in CUP development and prompt for the characterisation of additional axon guidance mutated genes in CUP.
From the histological perspective, CUP is defined as well- or moderately differentiated adenocarcinomas, accounting for 50–70% of all cases, with poorly differentiated carcinomas and adenocarcinomas making up another 20–30%, and the remaining being squamous-cell carcinomas (5–8%) and undifferentiated malignant neoplasms (2–3%) with inability of light microscopy to distinguish among carcinomas, lymphomas, melanomas, and sarcomas . The diagnoses of neuroendocrine tumours, melanomas, and sarcomas can be based on immunoperoxidase staining. Indeed, standard morphology and IHC remain the main strategy for the identification of the primary tumour in patients with CUP. The technique involves the analysis of tissue sections with antibodies against particular tumour-specific antigens, structural tissue components, hormonal receptors, hormones, or antigens . In the first instance, IHC differentiates well- and moderately-differentiated adenocarcinomas, squamous cell carcinomas, carcinomas with neuroendocrine differentiation, poorly differentiated carcinomas, and undifferentiated neoplasms. In squamous cell carcinomas and neuroendocrine carcinomas, the use of cell differentiation markers is advised, especially when tumour morphology is heterogenous or poorly differentiated. The IHC detection of markers, such as vimentin, S100 family proteins, HMB45 antigen, or CD45, may classify part of CUPs as non-carcinomas—sarcomas, melanomas, or lymphomas—that can be treated appropriately . Nevertheless, in the CUPISCO trial, the misdiagnosis of CUP due to sarcomas and melanomas represented 1.6% and 5.6%, respectively, of the failure cases . The most commonly used markers for the staining of CUP are the keratin family members, CK7 and CK20, with CK7+/CK20− being the most common in CUP . In CK7+/CK20− cases, ER positivity, and GATA3 positivity in ER-negative cases primarily direct clinicians to the breast as a possible site of cancer origin, especially in patients with axillary lymph node metastases . Although TTF1 expression in a metastatic setting does not unquestionably prove a primary origin in the lung, all TTF1/napsin A-positive cases should be radiologically investigated to rule out lung primitivity. The expression of PAX8/WT1 is considered in order to investigate a possible gynaecological origin. Overall, in approximately one-third of CUP cases, the primary site is identifiable through the IHC staining panels . However, a consensus panel of IHC markers has not yet been established, whilst no single pathognomonic marker exists for a conclusive diagnosis. Moreover, IHC has limited value in the diagnosis of poorly differentiated cancers. When immunoperoxidase stains are inconclusive in young patients with poorly differentiated tumours, electron microscopy should be considered in their evaluation. The presence of pleomorphic neoplastic cells with cytoplasmic vacuolations and/or cytoplasmic, non-membrane bound, electron-dense deposits detected by electron microscopy may suggest the diagnosis of a poorly differentiated carcinomas. Cytogenetic studies may be useful for the evaluation of young patients with poorly differentiated carcinomas or undifferentiated lesions that are responsive to chemotherapy. Finally, neurosecretory granules are specifically detected by electron microscopy in neuroendocrine tumours. Limitations of the IHC are the lack of reproducibility, due to the IHC-generated preparations, along with methodological issues. Failure to ensure that samples are of a high quality can hinder subsequent image analysis processes, negatively impact on data quality, and in some cases prevent an image analysis study from proceeding. Research using digitised histopathology slides for the development of artificial intelligence algorithms has increased markedly over recent years.
Image-assisted technologies has revolutionised the diagnosis of CUP. CT and conventional MRI have both been used to locate lesions, considering the clinical manifestation of CUP. The diagnostic accuracy of CT scans is around 55% (36–74%), mainly in pancreatic, colorectal, and lung cancer, while MRI has a sensitivity of 70% in detecting primary breast cancers . However, the diagnosis can be challenging if the primary tumour is small in size or has regressed, hindering the diagnosis. In particular, these cases may be successfully facilitated with the 2-[ 18 F] fluoro-2-deoxy-d-glucose (FDG) PET/CT, but still the detection rate is around 40% . The most frequent primary sites identified by PET are lung (33%) and head and neck (27%), followed by pancreas, breast, and colon (4–5%). Finally, 68 Ga-DOTA-NOC receptor PET/CT is recommended for the identification of primary neuroendocrine tumours, along with their metastases . Mammography is recommended for female patients with metastatic adenocarcinomas involving axillary lymph nodes. In patients with mammographically occult breast cancer, breast MRI may be considered.
Endoscopy should be directed towards investigating specific symptoms and signs or when specific histopathological findings are available. Fiberoptic bronchoscopy is reasonable for patients with respiratory symptoms and/or expression of CK7 and TTF1, whereas colonoscopy should be requested for those with abdominal symptoms or occult blood in the stool and/or expression of CK7, CK20, and CDX2. The sensitivity and specificity of the endoscopies are generally low.
In almost 70% of CUP patients, more than one marker can be concomitantly elevated in a non-specific way. Routine request of cancer antigen 125 (CA 125), cancer antigen 15.3 (CA 15-3), carbohydrate antigen 19-9 (CA 19-9), and carcinoembryonic antigen (CEA) is not recommended due to lack of prognostic and/or predictive value . However, there are some clinical scenarios in which serum tumour markers may have some diagnostic value. Indeed, serum PSA should be evaluated in men with osteoblastic bone metastases, CA 125 in women with primary serous papillary peritoneal adenocarcinoma, and CA 15-3 in females with isolated axillary adenocarcinoma. Finally, a high level of thyroglobulin in patients with CUP and bone metastasis may indicate occult thyroid cancer .
Improvements in nucleic acid sequencing technologies have enabled the detection of low quantities of tumour genetic material within the blood and show the potential to be both sensitive and specific to an individual’s tumour. These blood-based biomarkers include cfDNA, tumour microRNAs (miRNAs), and platelet-derived tumour mRNA, as well as analysis of DNA, RNAs, and protein expression from individual circulating tumour cells. Within this context, the use of liquid biopsies reduces the need for intrusive diagnostic biopsies and provides enough material to perform the diagnostic procedures. For instance, even though the presence of aberrant hypermethylation of tumour-suppressor genes in serum DNA has been detectable before the millennium, more sensitive and quantitative techniques for analysis of DNA methylation are required to expedite its incorporation in the clinical setting . In diffuse large-B-cell lymphoma, detection of aberrant DAPK1 methylation in cfDNA at the time of diagnosis is a positive prognostic biomarker, whilst in hepatocellular carcinoma, methylation of VIM is an early detection biomarker . Overall, there is evidence that the tissue of origin can be determined using cfDNA .
Molecular profiling technologies including microarray-based gene expression profiling, reverse transcriptase polymerase chain reaction, RNA sequencing, somatic gene mutation profiling with NGS, and DNA methylation profiling were used to define the primary culprit among patients with CUP. However, the implementation of tissue-of-origin classifiers in CUP is limited due to the absence of primary tumour. Some studies were conducted to validate predictions of the primary origin, on the basis of autopsy data, latent primary emergence, or IHC. Gene expression profiling has been directly compared to IHC, within known metastatic tumour types. Accuracy of gene expression profiling was 89%, compared with 83% for IHC when only one round of IHC determined the diagnosis, whereas in poorly differentiated cancers, such as CUP, the percentages were 83% and 67%, respectively . Nevertheless, only a limited number of studies have investigated the clinical outcomes of CUP patients, treated on the basis of gene expression predictions . Predicting the tissue-of-origin via molecular profiling is a debated topic within CUP, given that it is difficult to be molecularly classified in the absence of histological definition. However, the incorporation of molecular classifiers to the standard diagnostic workup may potentially identify atypical presentations of patients for whom site-specific therapies would be effective . PlexinB2 (PlxnB2) is a semaphorin receptor implicated in the regulation of cancer cell proliferation, invasiveness, and metastatic spreading . The G842C-PlxnB2 variant has been investigated in an effort to establish a proof of principle about the relevance of axon guidance genes in CUP. This mutation affected the conserved fold of an IPT domain, a moiety also found in Met and Ron oncogenic receptors . Notably, the large intracellular portion of the plexins does not contain a kinase domain or other classical signalling domains; nevertheless, it regulates the activity of monomeric GTPases, especially R-Ras, Rap-1, and RhoA. Moreover, plexins have been shown to couple with transmembrane tyrosine kinases such as ErbB2 and Met, triggering alternative noncanonical signalling cascades, especially in cancer cells . A recent study demonstrated that G842C-mutated PlxnB2 was competent for signalling, even in the absence of semaphorin stimulation . Moreover, although knocking down PlxnB2 expression in CUP cells bearing a wild-type receptor had no any functional impact, the metastatic cells carrying the G842C mutation were found to be dependent on this variant PlxnB2 to sustain self-renewal and proliferation in culture, along with tumorigenesis in mice. These data indicated that G842C-PlxnB2 may be considered a gain-of-function mutation. Members of the tyrosine kinase receptor family have been associated with plexin signalling in cancer cells. According to the study, PlxnB2 was found in complex with EGFR, and EGFR phosphorylation was enhanced in the presence of G842C-PlxnB2. Moreover, the greater invasiveness of CUP cells driven by the expression of the mutated plexin was abrogated by selective EGFR inhibitors, namely, cetuximab and erlotinib. These data provide evidence of the functional involvement of an unexpected aberrant signalling pathway in CUP development and prompt for the characterisation of additional axon guidance mutated genes in CUP.
Traditionally, CUP patients who are classified into one of the favourable subsets are treated according to their corresponding primary guidelines for metastatic disease. CUP patients with poorly differentiated carcinoma with midline distribution (extragonadal germ cell syndrome) should be managed like poor prognosis germ cell tumours with platinum-based combination chemotherapy. More than 50% response has been reported, with 15–25% complete responders and 10–15% long-term disease-free survivors. Women with papillary adenocarcinoma of the peritoneal cavity are optimally treated as FIGO stage III ovarian cancer. The recommended strategy includes aggressive surgical cytoreduction, followed by platinum-based postoperative chemotherapy. The median response rate is 80%, whilst 30–40% of the patients are complete responders. Similarly to FIGO stage III ovarian cancer patients, the median survival is 36 months . For the subgroup of women with adenocarcinoma involving only axillary lymph nodes, locoregional treatment with or without systemic therapy is suggested. The management is compatible with stage II/III breast cancer, resulting in 5- and 10-year overall survival rates of 75 and 60%, respectively. The patients with squamous cell carcinoma involving cervical lymph nodes are treated with locoregional management, according to the guidelines for locally advanced head and neck cancer. The 5-year survival rates range from 35 to 50% with documented long-term disease-free survivors. Surgery alone is inferior and only recommended in selected patients, particularly those with pN1 neck disease with no extracapsular extension. Radiotherapy to the ipsilateral cervical nodes alone is still inferior to extensive irradiation to both sides of the neck and the mucosa in the entire pharyngeal axis and larynx. Whether such intensive irradiation prolongs survival is still uncertain. Although the role of systemic chemotherapy remains undefined, concurrent chemoradiotherapy seems to be beneficial, particularly in patients with an N2 or N3 lymph node disease. The group of CUP patients with poorly differentiated neuroendocrine carcinomas should be treated with empirical platinum-based or platinum-taxane chemotherapy. The reported response rates are as high as 50–70% with 25% complete responders and 10–15% long-term survivors. Men with blastic bone metastases and elevated PSA are considered as having advanced/metastatic prostate cancer and treated accordingly. The appropriate approach of CUP patients with a single small metastasis is the local treatment with either resection and/or radiotherapy. A considerable number of these patients have a long disease-free survival . Finally, the treatment of Merkel cell cancer (MCC) of unknown origin is largely multimodal in nature and includes surgery, radiotherapy, and chemotherapy. For primary MCC that is associated with clinically positive nodal disease or with positive sentinel node, complete dissection of the involved regional nodal basin is recommended . MCC is radiosensitive, and as such, radiotherapy may be an alternative definitive treatment for medically ineligible surgical resection patients. In contrast, adjuvant chemotherapy has a limited role in MCC. The treatment of patients with unfavourable CUP subsets is usually empirical chemotherapy, consisting of either taxanes or platinum-based regimens, on the basis of randomised trials showing dismal survival improvements . The biomarker-based approach has been considered using targeted-therapy; nevertheless, the available evidence is limited to anecdotal cases . Site-specific therapy guided by molecular classifiers was evaluated in this context. A meta-analysis of two retrospective and two prospective trials evaluating site-specific treatments in CUP was performed . A trend towards improved overall survival was noted with site-specific versus empiric treatment for CUP (hazard ratio (HR) = 0.73, 95% confidence interval (CI) 0.52–1.02). The results of this meta-analysis highlighted the significant heterogeneity between the prospective studies comparing molecularly tailored to empiric therapy for CUP. In the most up-to-date meta-analysis of five studies that included 1114 patients, site-specific therapy was not significantly associated with improved overall survival (HR 0.75, 95% CI 0.55–1.03, p = 0.069) compared with empiric therapy . CUPISCO (NCT03498521) is an ongoing prospective, phase II, randomised study designed to elucidate the potential benefit of treatment following genomic profiling, as compared to standard chemotherapy of CUP patients . The study includes an atezolizumab monotherapy arm for the tumour mutational burden-high patients and a combination chemotherapy/atezolizumab arm for patients with tumour mutational burden-low or unknown tumours. The study experienced severe issues in patients’ accrual, along with screen failures. Molecular analyses, such as the identification of currently non-targetable alterations that may affect disease dynamics or be correlated with resistance, should be performed. Since CUP is clinically and molecularly heterogeneous, it would be reasonable to establish master protocols for enhancing the clinical trial strategy and direct patients to individually tailored treatment. I-PREDICT is an ongoing study that recruits patients with treatment-refractory solid tumours, including CUP, managed with individualised treatment, on the basis of genomic profiling (NCT02534675) . Patients treated with matched therapy that impacted more than half of their genomic alterations achieved significantly better outcomes than those from the lower match group. Within this context, immunotherapies have the potential to improve outcomes in this population, due to the PD-L1 expression and high tumour mutational burden in 22.5% and 11.8% of CUP patients, respectively . Overall, the genomic mutation correlates of response and resistance to immune checkpoint inhibitors do not differ between CUP and tumours that are immune checkpoint inhibitor eligible . Tumour mutational burden >10 mutations per megabase trended towards better outcomes in CUP patients treated with immune checkpoint inhibitors. Furthermore, MDM2 amplification, which is associated with lack of response to immune checkpoint inhibitors, has been detected in 2% of CUP patients . Initially, some anecdotal cases showed clinical activity of immune checkpoint inhibitors in CUP, irrespective of the presumed tissue of origin . Throughout time, we understood that the immune profiling of CUP is similar to that of malignancies responsive to immune checkpoint inhibitors and as such several trials investigate their efficacy in CUP . The phase II NCT03391973 and NCT03752333 trials of pembrolizumab are currently in progress, whereas NivoCUP, an open-label phase II study, has already demonstrated a clinical benefit of nivolumab in CUP patients . The reported objective response rate (ORR) was 22.2% in 45 previously treated patients, which met the primary endpoint. In 5 out of 12 patients who achieved a partial or complete response, the duration of the response was longer than 6 months. In the same subset of previously treated patients, the median overall survival was 15.9 months, whilst in the entire population of chemotherapy-naïve and previously treated patients, the ORR and the median overall survival were 21.4% and 16.2 months, respectively. These data provide evidence that nivolumab should be further investigated and may be incorporated in the therapeutic strategy of CUP. In the same study, a very low number of patients were treated upfront with nivolumab with 18.2% ORR. Overall, there is strong evidence that identification of predictive biomarkers is crucial in order to identify this one-fifth of CUP patients who may potentially respond to immune checkpoint inhibitors.
CUP is a heterogeneous group of metastatic tumours with a distinct natural history that mainly depends on clinicopathological criteria. While favourable groups are treated according to their corresponding primary tumour, unfavourable groups are treated with empirical chemotherapy, usually having a dismal prognosis. Several tissue-of-origin classifiers have been developed, collecting evidence that supported their translational potential in the clinical management of CUP patients. Several studies focused on genomic analysis of ctDNA and included some CUP cases among other tumour types, showing high sensitivity rates in the identification of oncogenic and actionable alterations in CUP. Small non-coding RNAs and epigenetic modifications are particularly appealing. Such biomarkers could potentially endorse the access to more specific therapies. The knowledge of the primary site remains fundamental because specific driver mutations could be predictive of responses in some tumour types but not in others. Immunotherapy is emerging as a potentially winning therapeutic strategy in several cancer types. Reasonably, it has gained interest even in the subset of CUP patients. Liquid biopsy could help in unveiling druggable alterations using a non-invasive approach. Therefore, molecular diagnostics, combined with genetic profiling, might become the standard of care for future CUP management. There is a significant limitation of the research on therapeutic strategies in CUP; this is the non-inclusion of many patients within the expanded favourable CUP subsets in the randomised trials who may be treated by their oncologists according to a potential primary tumour. New comprehensive clinical trial designs have been proposed to overcome the methodological issues encountered in CUP research implementing the latest diagnostics and therapeutic advances of CUP research.
|
Antibody–Biopolymer Conjugates in Oncology: A Review | ce8da1fc-c620-44c5-a797-70dacbe2e367 | 10053780 | Internal Medicine[mh] | Cancer is the leading cause of death worldwide with the latest statistics indicating that 1 in 3 people will develop cancer in their lifetime . To address this issue, the development of innovative therapies is extremely needed. World Cancer Day (celebrated annually on 4 February) is an initiative to unite the entire world together in the fight against this global epidemic. Interestingly, in 1910, Paul Ehrlich proposed the ‘magic bullet’ concept in the context of direct accessing desired targets by effective therapeutic molecules with minimal effects to normal cells . One solution is to identify overexpressed antigens to differentiate healthy cells form cancer cells, with some examples being human epidermal growth factor receptor (HER2) and MUC1 related to the breast cancer or cluster of differentiate 20 (CD20) related to the B cell lymphoma . Monoclonal antibodies (mAbs) were developed to precisely target antigens or receptors to cancer cells and as such, are referred to as targeted therapies. This strategy was advanced following the progress of the hybridoma technology in 1975 , and numerous mAbs have been approved. One such example is Herceptin (trastuzumab), a mAb used to primarily treat breast cancer by targeting the HER2 receptor; avastin, rituximab, or cetuximab should not be overlooked. Nevertheless, therapies based on mAbs alone are insufficient, mainly due to low lethality of cancer cells . Therefore, a new concept called antibody biopolymer conjugates (ABC), referring to the product of coupling mAbs with biologic polymers via molecular engineering, has been conceived to develop higher affinity, selectivity, and effectiveness of drugs to be delivered to a target. As such, a recombinant, full length, humanized mAbs is conjugated to a biopolymer. Monomeric units are covalently bonded to larger molecules–polymers. Biopolymers are natural polymers–bio-polymeric molecules derived from cells or extracellular compounds. They predominantly include three types, (i) polynucleotides, (ii) polypeptides, and (iii) polysaccharides and consist of repeating long chains. Biopolymers have many uses in medicine, such as in biomedical engineering, tissue engineering, regenerative medicine, used in medical devices, and are widely used in pharmaceutics and drug delivery. As this approach is a targeted approach, the effectiveness, the half-life, and tissue bioavailability are considerably increased. Improved biocompatibility and fast systemic clearing cannot be neglected. Indeed, a conjugate of a humanized mAb against vascular endothelial growth factor (anti-VEGF) and phosphorylcholine-based biopolymer, results in extended half-life to 10–12 days from 3–4 days of non-biopolymer conjugate . A schematic diagram of ABC is shown in . With their biological effects and proven therapeutic applications of ABC in the field of ophthalmic disease, it is a major step forward for the healthcare system. In the conventional method, the frequency of dosage and the continuous updating of injections along with the economic burden disturbs the life of patients. Therefore, a safer and more effective alternative is required which has a longer duration of action along with patient compliance. Research corresponding to effectiveness of ABCs in the field of oncology has been flourishing tremendously in the last decade. Herein, we emphasize the chemistry involved in the current development of ABCs and its advantages over traditional methods of drug delivery systems, with special emphasis in oncology. Their clinical applications in the field of cancer treatment are also highlighted.
Antibody–polymer/biopolymer conjugations, directed against tumor-associated antigens, have been an area of much curiosity for some time . ABCs have been developed to improve efficacy, minimize toxicity, and to attain site-specificity and excellent safety . The novel drug delivery system, which is composed of polymers or biopolymers and their conjugates with antibodies or fragments of antibodies are suitable as carriers and as anti-cancer and anti-inflammatory agents . Antibodies are usually coupled to biopolymers using bioconjugation method. Bioconjugation is a chemical reaction between two molecules (at least one of which should be a biomolecule) to form a stable covalent linkage . The advantage of applying conjugates lies in the selective delivery to the target site and in the possible protection of antibodies against fast enzymatic degradation and excretion, thereby leading to a higher antibody concentration against the tumorigenic antigen . Moving forward, bioconjugation methods are highly site-specific and cause minimal perturbation to the active form of the biomolecules . Classical bioconjugation reactions include second order reactions which rely on the reactivity of functional groups present in protein, peptide, or sugar-like biomolecules. Cysteine and lysine residues of proteins are commonly encountered for site-specific bioconjugation which contains thiol and amino functional groups, respectively . Because of its high chemo selectivity, click chemistry has become an efficient strategy for modifying functionality of biomolecules . By virtue of the admirable stability of amide linkages, they become attractive for bioconjugation . A biopolymer containing carboxylic acid group can be treated with coupling agent such as ethyl(dimethylaminopropyl) carbodiimide (EDC) and N-hydroxysuccinimide (NHS) to activate carboxylic acid into corresponding N-hydroxysuccinimidyl ester, which will be treated with amino groups of lysine side chains and N-terminus of peptide to form amide bonds ( A). Thiols are more potent nucleophiles than amino groups in aqueous solutions. As cysteine is the second least common amino acid in natural proteins, the derivatization of a cysteine residue is a popular method of bioconjugation . The thiol-reactive functional groups include maleimides, bisulfides, and haloacetamides. Amongst them, maleimides are widely used electrophiles for thiol-mediated bioconjugation. Thiol-containing antibodies undergo Michael addition with maleimide-bearing biopolymers to afford succinimidyl thioethers ( B) . Biologic oligomers and biopolymers, such as carbohydrates, peptides, and nucleic acids, have been modified by using the copper-catalyzed azide-alkyne cycloaddition click reaction ( C) . Antibody–biopolymer conjugates are also known as biopolymeric prodrugs, which are anticipated as novel drug delivery systems formulated for the incorporation of therapeutic agents into biopolymers of choice using selected functionalities . Antibodies serve a dual purpose here—as therapeutic agent and as targeting moieties to attain the site specificity . These biopolymeric prodrugs fit in with current trends oriented towards natural sources. Biopolymers are natural polymers produced by the cells of living organisms which can be classified according to the monomers used, such as polynucleotides, polypeptides, and polysaccharides. Chitosan, hyaluronic acid, dextran, heparin, silk fibroin, pullulan, or polysaccharides from Auricularia auricula are good examples of natural polymers employed in biopolymeric prodrugs . They have widely been exploited for the delivery of therapeutic agents due to their outstanding biocompatible and biodegradable characteristics , representing either a valuable or renewable source . Thus, therapeutic agents such as antibodies are chemically conjugated with various biopolymers and the resulting conjugates could slowly elute their active ingredients following cleavage of the biopolymer–antibody linkages under physiological conditions. These conjugates offer several advantages over their therapeutic precursors such as enhancement of water solubility and bioavailability, improvement of pharmacokinetic and biodistribution profiles of antibodies, protecting them from deactivation, and facilitating their transport to the targeted sites . It can be mentioned that the selective activity is provided via key characteristics of neoplastic pathology, including inter alia specific enzymes, hypoxia, and extracellular pH. Interestingly, the strategic concept of utilizing specific triggers for the activation of prodrugs, in terms of antibody-directed prodrug therapy, gene-directed enzyme prodrug therapy, and virus-directed enzyme prodrug therapy, was introduced in the late 1980s . Next, a new mechanism to activate drugs, depending on the stimuli, known as polymer-directed enzyme drug therapy was developed. Here, the polymer–enzyme conjugates increase the selective release of drug from a polymer conjugate. More specifically, a new polymer enzyme-loaded nanoreactor to release the active drug from prodrug was reported . This technology was used in liposomal formulations, known as polymer enzyme liposome therapy, in which the drug molecule is released from the liposome into the tumor site . summarizes some examples of antibody–biopolymer conjugates and their efficacy towards several cancers. Poly (β-L-malic acid) and polyethylene glycol (PEG)–antibody conjugates by covalently incorporating anti-HER2/neu peptide (AHNP) (trastuzumab-mimetic 12-merpeptide) for the treatment breast cancer . Jun Xiao et al. synthesized glycol chitosan (GC) and gemcitabine conjugate using NHS/EDC method, which was further conjugated with chitosan antibody and anti-EGFR antibody to afford ABC-GC-Gemcitabine nano bioconjugates. It was reported that these conjugates were able to reduce pancreatic cancer cell proliferation and colony formation, and also inhibited the migration and invasion of SW1990 cells. Refs. proposed a microfluidic-assisted approach using a polydimethylsiloxane (PDMS) Y-shaped microreactor for the covalent conjugation of Trastuzumab (TZB), a recombinant antibody targeting HER2 (human epidermal growth factor receptor 2), to doxorubicin-loaded PLGA/Chitosan NPs (PLGA/DOX/Ch NPs) using EDC and N-hydroxysulfosuccinimide (sNHS) mediated bioconjugation reactions. The conjugate showed promising results when checked in vitro against HER2+ breast cancer cells . Rong Zhu et al. carried out a study where CD147 monoclonal antibody was coupled with a complex of α-hederin (α-hed) and chitosan (CS) nanoparticles (NPs) using NHS and EDC. The researchers reported that the half-maximum inhibiting concentration (IC 50 ) of α-Hed-CS-CD147-NPs in human liver cancer cell lines HepG2 and SMMC-7721 was lower than that of free α-Hed and α-Hed-CS-NPs .
Monoclonal antibodies are widely used in the treatment and management of cancer patients. However, due to lack of selectivity, it leads to tremendous toxicity. The conjugation with biopolymers enables the selective delivery of antibodies by having a specific affinity towards some over-expressed or lacking receptors on tumor cells . Polymers can form monomolecular structures or micelles to perform their tasks . Vladimir P. Torchilin and co-workers stated that 2C5 antibody conjugated with taxol micelles successfully reduced the tumor mass in the heart in both in vivo and in vitro studies . Micellar complex recognizes the surface of the tumor but not the normal cell, hence imparting selectivity of antibody delivery. It also facilitates the delivery of aquaphobic pharmaceuticals . In addition, mAb-ch735 was shown to specifically target cancer cell membranes. This targeted approach can allow antibodies to be rapidly internalized into endosomes and lysosomes . Further, the development of nanotechnology has provided an added advantage to ABC. As such, the biodegradable tamoxifen antibody-conjugated polymeric nanoparticles have a targeted delivery of antibodies against breast cancer . α-hederin chitosan conjugated with CD147 monoclonal antibody escalate the action on liver cells malignancy . Due to the lipophilic profile of α-hederin, it holds low bioavailability and poor oral absorption which can be overcome by entrapping it in biopolymer (here, in chitosan). It is also noted that addition of methyl group in CS increases the anti-tumor property . The hepatocellular cell line CC531 responds extremely well to the mAb CC52-liposome conjugates. In mice, it was shown that even 24 h later, 20–30% liposomes were still present in the bloodstream, showing persistent release to maintain the ideal antibody concentration. The control had a 2-fold greater splenic uptake than the conjugated ones, which lengthens the duration of the effect . The effectiveness of drug when conjugated with ABC is increased many-fold. It also helps with the pharmacokinetic aspects of the molecule. Liposomes associated with doxorubicin antibody showed a significant effect on murine Lewis lung carcinoma (LLC) and human mammary adenocarcinoma BT-20 cell lines. Within 24 h, the mAb 2C5 lysis was 90% of LLC and 80% of BT-20 cells . Stomach carcinogenesis is on account of overexpression of epidermal growth factor receptor (EGFR) in 27–44% of the initial tumor patient. Cetuximab conjugate with docetaxel loaded poly (γ-glutamic acid) nanoparticle targets on EGRF which arrests the cell division and eventually leads to cell death . The capacity to initiate or regulate drug release in polymeric-based nanoparticles gives them certain benefits over other nanoparticles such as liposomes, such as the possibility for a more powerful medicinal payload and improved stability (liposomal phospholipids are susceptible to oxidation) . Further, in vitro in human fibrosarcoma cells, the mAb 19–24 was efficient when combined with daunomycin using dextran as opposed to being unconjugated . In another study drug–prodrug nanoparticle (PDNP) binding to HepG2 cells was much lower compared to mAb-PDNP. Moreover, poly (butyl cyanoacrylate) (PBCA) nanoparticles conjugated with mAb showed that cytotoxicity was improved by 40% as compared to carboplatin alone. In rats with glioblastoma, the survival was longer when compared to free carboplatin treatment and side effects were significantly reduced in the liver, kidney, and brain compared to carboplatin alone . 3.1. Clinical Studies Clinical trials are of utmost importance to determine the efficacy and toxicity of drug molecules, and other effects on human health outcomes. KSI-301 is the first-in-class of ABC, consisting of anti-VEGF IgG1 monoclonal antibody and phosphorylcholine-based biopolymer, designed to increase either ocular half-life or tissue bioavailability. It has a high binding affinity. Indeed, in preclinical studies, the half-life was extended (~11 days) in terms of ranibizumab (~3 days) and aflibercept (~4 days). Moreover, KSI-301 can be cleared rapidly. It is currently in human clinical trials against retinal vascular diseases . Nevertheless, it should be highlighted that ABC is a novel approach to enhance clinical effects and, despite being ABC an astounding molecule, human clinical trials are still limited in the field of cancer . As such, further studies are required in humans. Alternative Diagnostic Uses ABCs also possess imaging qualities with one example being in ovarian cancer. In fact, ovarian cancer can be easily identified using polymeric fluorescent nanoparticles (PFNPs) attributed to their distinctive optical characteristics. Excellent photosensitivity makes PFNPs an innovative tool for the detection and treatment of ovarian cancer cells . In in vitro studies, overexpression of prostate stem cell antigen (PSCA) was identified in 90% of samples where the ability of mAbs to identify PSCA was significantly increased by the use of conjugate–dextran antibodies with superparamagnetic iron oxide . As such, the labeled cell produces intense blue color upon binding with anti-PSCA-dextran-superparamagnetic iron oxide . In non-Hodgkin’s lymphoma (NHL), anti-CD20 polyclonal antibody-radioisotopes/fluorophores conjugated with chitosan has been used for its detection .
Clinical trials are of utmost importance to determine the efficacy and toxicity of drug molecules, and other effects on human health outcomes. KSI-301 is the first-in-class of ABC, consisting of anti-VEGF IgG1 monoclonal antibody and phosphorylcholine-based biopolymer, designed to increase either ocular half-life or tissue bioavailability. It has a high binding affinity. Indeed, in preclinical studies, the half-life was extended (~11 days) in terms of ranibizumab (~3 days) and aflibercept (~4 days). Moreover, KSI-301 can be cleared rapidly. It is currently in human clinical trials against retinal vascular diseases . Nevertheless, it should be highlighted that ABC is a novel approach to enhance clinical effects and, despite being ABC an astounding molecule, human clinical trials are still limited in the field of cancer . As such, further studies are required in humans. Alternative Diagnostic Uses ABCs also possess imaging qualities with one example being in ovarian cancer. In fact, ovarian cancer can be easily identified using polymeric fluorescent nanoparticles (PFNPs) attributed to their distinctive optical characteristics. Excellent photosensitivity makes PFNPs an innovative tool for the detection and treatment of ovarian cancer cells . In in vitro studies, overexpression of prostate stem cell antigen (PSCA) was identified in 90% of samples where the ability of mAbs to identify PSCA was significantly increased by the use of conjugate–dextran antibodies with superparamagnetic iron oxide . As such, the labeled cell produces intense blue color upon binding with anti-PSCA-dextran-superparamagnetic iron oxide . In non-Hodgkin’s lymphoma (NHL), anti-CD20 polyclonal antibody-radioisotopes/fluorophores conjugated with chitosan has been used for its detection .
ABCs also possess imaging qualities with one example being in ovarian cancer. In fact, ovarian cancer can be easily identified using polymeric fluorescent nanoparticles (PFNPs) attributed to their distinctive optical characteristics. Excellent photosensitivity makes PFNPs an innovative tool for the detection and treatment of ovarian cancer cells . In in vitro studies, overexpression of prostate stem cell antigen (PSCA) was identified in 90% of samples where the ability of mAbs to identify PSCA was significantly increased by the use of conjugate–dextran antibodies with superparamagnetic iron oxide . As such, the labeled cell produces intense blue color upon binding with anti-PSCA-dextran-superparamagnetic iron oxide . In non-Hodgkin’s lymphoma (NHL), anti-CD20 polyclonal antibody-radioisotopes/fluorophores conjugated with chitosan has been used for its detection .
Antibody conjugations provide a novel strategy to produce effective diagnostic and therapeutic (or theranostic) as well as imaging systems. The antibody–biopolymer conjugates are rapidly becoming important compounds and no doubt a revolution in oncology that offers enormous potential and new prospects in ‘drugging the undruggable’ targets. Pros and Cons of ABC The main advantages of ABCs are as follows: extended half-life/treatment durability, increased tissue bioavailability, a high binding affinity, deeper potency, fast systemic clearance, enhanced tissue penetration, improved stability, efficacy, biocompatibility, and safety . Conjugations with polymer nano-delivery systems lead to better solubility and immunological profile . They can deliver poorly soluble chemotherapeutic agents to suppress multiple drug resistance phenomena in tumor cells . Furthermore, ABCs can detect cancer at its earliest stage and induce significant cancer cell death . Bioconjugates provide synergistic anti-tumor effects and potency to overcome the complications resulting from chemotherapeutics. Many biopolymers have more complex structures, like the human body; thus, they are better with bodily integration. Moreover, biopolymers, as natural polymers can mimic body parts to sustain normal biological functions. Therefore, they are ideal in biomedical engineering . Diverse efforts have been invested into designing suitable ABC anticancer strategies and several ABCs have been developed in several clinical trials. Nanotechnology is a powerful approach in terms of ABC-based diagnosis and therapy of cancers due to its ease in synthesis, tuneability, and bio-functionalization . The unique feature of bio-conjugates is the selective delivery of drugs to pathological sites and the improvement of molecular retention in the blood circulation system . ABCs are stimuli-sensitive and thus, more effective compared to other delivery systems . Importantly, the ABC platform stabilizes the structure of antibodies in terms of favorable clinical features (e.g., low nonspecific interactions and antibody recycling) . Overall, ABCs represent a big step towards perfect diagnosis techniques and therapeutic options–effective, affordable, and with less treatment burden. Nevertheless, they are still at the proof-of-concept stage. In vitro efficacy is not ideally translated into clinical effects and the production cost of ABCs is currently high. Thus, ABC therapy is yet not available as a treatment option, especially in underdeveloped countries. In this regard, further studies are required to improve its stability, efficacy, and effectiveness in human clinical trials .
The main advantages of ABCs are as follows: extended half-life/treatment durability, increased tissue bioavailability, a high binding affinity, deeper potency, fast systemic clearance, enhanced tissue penetration, improved stability, efficacy, biocompatibility, and safety . Conjugations with polymer nano-delivery systems lead to better solubility and immunological profile . They can deliver poorly soluble chemotherapeutic agents to suppress multiple drug resistance phenomena in tumor cells . Furthermore, ABCs can detect cancer at its earliest stage and induce significant cancer cell death . Bioconjugates provide synergistic anti-tumor effects and potency to overcome the complications resulting from chemotherapeutics. Many biopolymers have more complex structures, like the human body; thus, they are better with bodily integration. Moreover, biopolymers, as natural polymers can mimic body parts to sustain normal biological functions. Therefore, they are ideal in biomedical engineering . Diverse efforts have been invested into designing suitable ABC anticancer strategies and several ABCs have been developed in several clinical trials. Nanotechnology is a powerful approach in terms of ABC-based diagnosis and therapy of cancers due to its ease in synthesis, tuneability, and bio-functionalization . The unique feature of bio-conjugates is the selective delivery of drugs to pathological sites and the improvement of molecular retention in the blood circulation system . ABCs are stimuli-sensitive and thus, more effective compared to other delivery systems . Importantly, the ABC platform stabilizes the structure of antibodies in terms of favorable clinical features (e.g., low nonspecific interactions and antibody recycling) . Overall, ABCs represent a big step towards perfect diagnosis techniques and therapeutic options–effective, affordable, and with less treatment burden. Nevertheless, they are still at the proof-of-concept stage. In vitro efficacy is not ideally translated into clinical effects and the production cost of ABCs is currently high. Thus, ABC therapy is yet not available as a treatment option, especially in underdeveloped countries. In this regard, further studies are required to improve its stability, efficacy, and effectiveness in human clinical trials .
According to the WHO in 2022, cancer has a significant impact in a large population and causes close to 10 million fatalities per year worldwide. The traditional approach of treating cancer has its own disadvantages. Immunization is a crucial component of treating several cancers, yet it has pitfalls such a short half-life and poor pharmacokinetics. However, ABC is an innovative, complex medicinal chemical that enhances the sanative effects of antibodies. Utilizing antibodies makes it easier to target specific cells and effectively reduces illness. Clinical studies on several compounds that treat cancer are being monitored. ABC has targeted breast cancer in a clinical study since it is the primary cause of mortality from cancer, and the healthcare system is hoping for a favorable outcome. The remarkable properties of ABCs will need to be established through further clinical trials. ABCs have the potential to alter the mode of how drugs are administered, with outstanding therapeutic results and with fewer negative effects.
|
Diagnostic performance of Idylla MSI test in colorectal cancer biopsies | 372dbbd4-fa6d-452a-aafa-1fc90e987b4e | 10053848 | Anatomy[mh] | Microsatellites are small elements of repeated DNA, which are prone to form mismatches during DNA replication. DNA mismatch repair (MMR) system encoded by the MMR genes ( MLH1, MSH2, MSH6 , and PMS2 ) normally functions to correct these replication errors. Deficient MMR (dMMR) system leads to microsatellite instability (MSI) and hypermutation phenomenon causing increased cancer susceptibility . The MMR system can be compromised by epigenetic mechanism, usually by acquired MLH1 promoter hypermethylation, or by genetic inactivation characteristic to Lynch syndrome (LS). MSI has been identified in approximately 15% of colorectal cancer (CRC) cases, of which approximately 80% are sporadic . Universal testing for dMMR/MSI has been recommended in CRC to screen for LS and to guide optimal follow-up and treatment for the patients . It can be critical to analyze dMMR/MSI status from pre-treatment rectal cancer biopsy specimens, considering the loss of tumor burden due to successful neoadjuvant treatment and possible post-neoadjuvant absence of MSH6 protein expression . Since immuno-oncological treatments have shown notable effectiveness in treating advanced dMMR/MSI CRC patients , and recently also in the neoadjuvant setting with excellent responses , identification of dMMR/MSI status at initial diagnosis using biopsy samples has been proposed to be necessary diagnostic procedure . The two most general methods to detect dMMR/MSI in CRC are MMR IHC and polymerase chain reaction (PCR)-based microsatellite tests, of which IHC is widely used as gold standard in pathology laboratories . Both methods can also be used to complement each other in diagnostically challenging cases and to confirm the MSI status before starting immuno-oncological treatments . IHC is the most affordable method but requires trained personnel and laborious hands-on time and is also prone to pitfalls in pre-analytical and analytical phases. It can, however, be used on biopsy samples with low tumor cell content. PCR-based microsatellite tests, on the other hand, may have sensitivity issues with low tumor cell percentage or low DNA purity/yield. Idylla MSI test is a novel real-time PCR based assay which analyzes a set of seven microsatellite marker specific probes ( ACVR2A , BTBD7 , DIDO1 , MRE11 , RYR3 , SEC31A , and SULF2 ) using fluorescent-labeled molecular beacons and melting curve analysis. Idylla test is a rapid way to assess MSI/MSS status from formalin-fixed paraffin-embedded (FFPE) tissue sections, comprising automatic DNA extraction, PCR amplification, software interpretation and reporting. The specificity and sensitivity of Idylla MSI test to detect MSI status has been shown to be as high as 98–100% and 94–100% respectively evaluated mainly from surgical CRC specimens and with tumor cell content over 20% . Since the diagnostic performance of Idylla MSI test has previously been shown to be optimal in CRC surgical resection specimens, especially giving no false positives , we here wanted to scrutinize the usability of Idylla MSI test in CRC biopsies. For that, we compared the performance of the Idylla MSI test to the gold standard MMR IHC in 117 CRC biopsy samples with known dMMR status.
Sample selection We analyzed 117 colonoscopy biopsies with known dMMR status from CRC patients, who had biopsies taken in The Hospital District of Helsinki and Uusimaa between October 2019 and December 2021. The patients had not undergone any neoadjuvant treatment prior the colonoscopy and the tumors were routinely screened for MMR proteins MLH1, MSH2, MSH6, and PMS2 using IHC. The set of CRC biopsies included 84 adenocarcinomas not otherwise specified (NOS), 13 partim mucinosum adenocarcinomas (less than 50% mucinous component), two mucinous adenocarcinomas (50% or more mucinous component), five signet ring cell carcinomas (50% or more signet ring cells), four adenocarcinomas with signet ring differentiation (less than 50% signet ring cells), eight adenomas with minimal invasive component and one large adenoma with high-grade dysplasia. The study was approved by the Ethics Committee of the HUH. Immunohistochemistry All 117 CRC biopsies had undergone diagnostic IHC to detect the loss of MMR protein expression as a gold standard and the method was performed as described previously . The absence of one or more MMR protein expression with positive external and internal controls was considered as dMMR. The dMMR status of the biopsies was re-analyzed in a blinded manner by AR and IU, and all cases were confirmed to be dMMR. Idylla MSI test We analyzed the 117 dMMR CRC biopsies with automated Idylla MSI test, according to the manufacturer’s protocol. The recommended tumor cell percentage for Idylla MSI test is ≥ 20% for CRC samples and the required total tissue area is 25–300 mm 2 using 10 μm thick sections. For the analysis, one to four 10 μm sections were cut from the FFPE biopsy tissue blocks with a Leica SM2000R microtome (Leica Microsystems GmbH, Wetzlar, Germany). The tumor cell percentage was estimated from the biopsy HE slides by two independent observers (AR and IU) both before and after cutting the tissue sections for Idylla. Macrodissection was performed only for two biopsy samples where the non-neoplastic colon areas were clearly separate from the cancerous areas. The tumor cell percentages varied between 5 and 80% of which 21 samples had < 20% tumor content. Detection of two or more mutant microsatellite markers ( ACVR2A , BTBD7 , DIDO1 , MRE11 , RYR3 , SEC31A , and/or SULF2 ) using the Idylla MSI test is classified as MSI, whereas less than two mutant markers lead to MSS result. MSI score cutoff value 0.5 is used to judge the marker as mutated.
We analyzed 117 colonoscopy biopsies with known dMMR status from CRC patients, who had biopsies taken in The Hospital District of Helsinki and Uusimaa between October 2019 and December 2021. The patients had not undergone any neoadjuvant treatment prior the colonoscopy and the tumors were routinely screened for MMR proteins MLH1, MSH2, MSH6, and PMS2 using IHC. The set of CRC biopsies included 84 adenocarcinomas not otherwise specified (NOS), 13 partim mucinosum adenocarcinomas (less than 50% mucinous component), two mucinous adenocarcinomas (50% or more mucinous component), five signet ring cell carcinomas (50% or more signet ring cells), four adenocarcinomas with signet ring differentiation (less than 50% signet ring cells), eight adenomas with minimal invasive component and one large adenoma with high-grade dysplasia. The study was approved by the Ethics Committee of the HUH.
All 117 CRC biopsies had undergone diagnostic IHC to detect the loss of MMR protein expression as a gold standard and the method was performed as described previously . The absence of one or more MMR protein expression with positive external and internal controls was considered as dMMR. The dMMR status of the biopsies was re-analyzed in a blinded manner by AR and IU, and all cases were confirmed to be dMMR.
We analyzed the 117 dMMR CRC biopsies with automated Idylla MSI test, according to the manufacturer’s protocol. The recommended tumor cell percentage for Idylla MSI test is ≥ 20% for CRC samples and the required total tissue area is 25–300 mm 2 using 10 μm thick sections. For the analysis, one to four 10 μm sections were cut from the FFPE biopsy tissue blocks with a Leica SM2000R microtome (Leica Microsystems GmbH, Wetzlar, Germany). The tumor cell percentage was estimated from the biopsy HE slides by two independent observers (AR and IU) both before and after cutting the tissue sections for Idylla. Macrodissection was performed only for two biopsy samples where the non-neoplastic colon areas were clearly separate from the cancerous areas. The tumor cell percentages varied between 5 and 80% of which 21 samples had < 20% tumor content. Detection of two or more mutant microsatellite markers ( ACVR2A , BTBD7 , DIDO1 , MRE11 , RYR3 , SEC31A , and/or SULF2 ) using the Idylla MSI test is classified as MSI, whereas less than two mutant markers lead to MSS result. MSI score cutoff value 0.5 is used to judge the marker as mutated.
Our dMMR CRC biopsy set (n = 117) consisted of patients with median age of 76 years (range from 27 to 99) with slight female predominance (56.4%). Tumors were mainly low-grade (72.6%) and localized to the right colon (69.2%), the proportion of rectum tumors being 10.3%. Majority of the cases represented dMLH1 (85.5%) with minority of other dMMR subclasses (dMSH2 9.4%, dMSH6 2.6% and dPMS2 2.6%). We analyzed the 117 CRC biopsies by Idylla MSI test and the concordance between Idylla and MMR IHC was 96.6% (113/117) among all the cases (5–80% tumor cell percentage) and 99.0% (95/96) among the cases with the recommended ≥ 20% tumor cell content. Of the four discrepant cases, three contained tumor cell percentage from 5 to 15% and the remaining fourth case (C40) had tumor cell percentage 40–50%, estimated from pre- and/or post-Idylla HE slides (Table ). There were no invalid results, but one cassette failure occurred in the study set and a re-analysis was needed for that sample. We next re-analysed all the four discrepant biopsies with Idylla using increased number of sections (from two sections to three to four sections) and the three suboptimal cases remained discrepant (MSS), whereas the result for the case C40 changed to MSI (Fig. ; Table ). All discrepant cases had 0/7 microsatellite markers mutated (Supplementary Figure ). We further analyzed the surgical resection CRC samples of the discrepant cases with tumor cell content of > 20% by Idylla and all the results were confirmed to be MSI. In cases classified MSI by Idylla, the most commonly mutated markers were DIDO1 and ACVR2A with frequencies of 98% and 92%, respectively (Fig. ).
We here evaluated the diagnostic performance of Idylla MSI test to detect MSI in CRC biopsy specimens as compared to MMR IHC. For that we analyzed 117 dMMR CRC biopsies. The concordance between Idylla and IHC was 99.0% (95/96) with the cases having recommended ≥ 20% tumor cell content. However, it was 96.6% (113/117) in the whole set including also the 21 less optimal cases with < 20% tumor cell content. The weakness of this study was that we did not include any proficient MMR CRC biopsies to the study and therefore could not evaluate the specificity of the Idylla MSI test in CRC biopsy material. The study focused on evaluation of the performance of Idylla test only in IHC dMMR CRC biopsies, because the optimal specificity (98–100%) of the Idylla MSI test has been repeatedly confirmed in CRC and the sensitivity (94–100%) with false negative results has been more often compromised with Idylla when compared to IHC . We identified four discrepant cases from which 3/4 had pre- and/or post Idylla tumor cell content less than 20%, explaining the repeated MSS result in CRC biopsy specimens. One initially discrepant case (C40) with an adequate tumor cell percentage was detected MSI only after re-analysis with increased number of sections. This case represented signet ring cell carcinoma from which the tumor cell content can be challenging to evaluate (Fig. ), and this histological type of the tumor may require increased number of tissue flakes for successful analysis. Defective molecular testing has also previously been reported in cancer samples with mucinous component leading to poor DNA quality . Our findings of the excellent diagnostic performance of the Idylla MSI test in CRC biopsy material is in line with previous studies where the MSI/MSS status has been evaluated mainly from surgical CRC specimens. The concordance across Idylla and IHC has been previously reported to be 96–100% with 100% specificity in CRC with tumor cell content ≥ 20% . In our MSI CRC biopsy cases detected by the Idylla, the least mutated biomarker (61%) was SEC31A concordant with previous studies . Differing from the previous reports the most often mutated marker was DIDO1 (98%) and ACVR2A (92%) was only the second most mutated. Good accuracy of Idylla MSI test has also been confirmed in biopsy material of gastric adenocarcinoma by Farmkiss et al. scoring 96% concordance and 100% specificity between Idylla and IHC (n = 50) . To our knowledge this is the first study to evaluate the diagnostic performance of the Idylla MSI test in CRC biopsy material. Our study confirms that the Idylla MSI test is an accurate diagnostic method to identify MSI status also in CRC biopsy samples with ≥ 20% tumor cell content. Biopsies of signet ring cell carcinomas may need increased number of tissue flakes for the analysis even with instructed tumor cell percentage. We can conclude that the Idylla MSI test offers a competent tool for MSI screening in CRC biopsies.
Below is the link to the electronic supplementary material. Supplementary Material 1
|
Functional and Phenotypic Characterisations of Common Syngeneic Tumour Cell Lines as Estrogen Receptor-Positive Breast Cancer Models | db0c8208-979c-4be6-8704-fd45681ac12d | 10053941 | Anatomy[mh] | Breast cancer (BCa) incidence has been increasing steadily for decades and now affects 1 in 7 women in the western world, making it the most prevalent cancer in women. The estrogen receptor-positive (ER + ) BCa subtype is driving this increase in incidence. Endocrine therapies such as tamoxifen effectively treat ER + BCa and are also used in the preventative setting . Unfortunately, some ER + BCa tumours in advanced ER + BCa patients acquire resistance to tamoxifen , and due to the real or perceived side effects of this treatment, uptake and adherence rates of tamoxifen regimens in the preventative setting are poor . This underscores the importance of developing additional endocrine therapies for ER + BCa. Pre-clinical studies often rely on patient-derived xenograft models, donor tissues or cells implanted in immunocompromised mice, to test next-generation endocrine therapies. There are also several genetically engineered mouse strains that develop ER + mammary tumours and can be used for pre-clinical testing, including AIB1-overexpressing mice, STAT1 knockout mice and MMTV Pik3ca H1047R transgenic mice . However, as the tumours that form in these transgenic models exhibit a long latency, it renders pre-clinical in vivo experimentation time-consuming and costly. Alternatively, transplantable syngeneic breast cancers in mice are of great value as they allow assessment of tumours in normal stromal and immune microenvironments and can generate tumours and metastatic disease with shorter latency . Two mouse mammary tumour cell lines are considered ER + ; SSM3 and J110 . However, there are conflicting reports of whether either of these cell lines or others (67NR, E0771, D2.0R, D2A1, 4T1.2) are genuinely ER + in vitro and reliably produce ER + tumours in vivo. A model of ER+ BCa should be dependent on ER signalling for growth. Here we have tested available syngeneic cell lines for ER activity using immunohistochemical staining for Erα protein, flow cytometry, gene expression profiles and in vitro tamoxifen sensitivity. The data demonstrate that the SSM3 line is the only luminal ER + and tamoxifen-sensitive cell line. SSM3 represents a reliable model for assessing the efficacy of new therapies designed at targeting ER + BCa development. Our study also highlights the need for additional syngeneic models of ER + BCa to represent the different molecular subcategories of ER + BCa that exist in humans.
2.1. ERα Staining in Syngeneic Tumours Formalin-fixed paraffin-embedded tissue sections from SSM3, 4T1.2, 67NR, J110, EO771, D2.0R and D2A1 tumours grown in mice were stained for Erα using a clinically validated antibody . The SSM3 tumours showed strong and extensive nuclear staining (>90%) in epithelial tumour cells . These tumours were also assessed pathologically and were classified as adenocarcinomas which are also an accurate representation of common ER + tumours . Similarly, 67NR tumours had extensive nuclear staining but weaker staining. Scattered nuclear staining was observed in D2A1, 4T1.2 and J110 tumours. However, this staining overlapped with areas of stromal and immune infiltration, as demonstrated by CD45 immunohistochemical staining on consecutive sections . D2.0R and EO771 were negative for nuclear Erα expression but displayed some regions of cytoplasmic staining . In almost all the different tumours, cytoplasmic Erα staining was observed in regions of the tumours that had immune infiltration identified through CD45 staining, indicating, as above, cytoplasmic staining of immune cells rather than epithelial tumour cells . 2.2. SSM3 Tumours Are the Only Syngeneic Model That Responds to Tamoxifen In Vitro and In Vivo Using the well-characterised human ER + MCF7 cells, we conducted a dose–response to tamoxifen and showed that whilst high concentrations (10 µM) were cytotoxic, lower doses of tamoxifen could inhibit the growth of MCF7 cells . As a negative control, we included a dose–response in ER - MDA-MB-231 cells and showed that, apart from the cytotoxic dose (10 µM), these cells were insensitive to tamoxifen treatment. All mouse tumour cell lines were tested at 1 µM, which resulted in reduced proliferation of SSM3 cells. Previously we have shown that 67NR tumours are mildly sensitive to tamoxifen and that 4T1.2 tumours are insensitive . Again, we found that 67NR cells are mildly sensitive. J110 cells, D20R and D2A1 cells were all insensitive to growth inhibitory effects of tamoxifen . The EO771 cell line response to tamoxifen was not assessed, as it was already revealed it had no sensitivity . To validate our results in vivo we treated mice injected with SSM3 cells in their fourth mammary fat pads with 1 mg of tamoxifen daily. Tamoxifen significantly slowed tumour growth compared to vehicle controls . We did not observe any toxicity issues with this dosing regimen. 2.3. SSM3 Cells Are the Only Syngeneic Cell Line to Express Cell Surface Markers of Luminal Mammary Epithelial Cells It is appreciated now that breast cancer arises from a dysregulation of the normal breast epithelium. The molecular definition of the cell lineages in the normal breast has been instrumental in allowing these relationships to be identified. For example, Receptor activator for nuclear factor κB (RANK)+ luminal progenitors are known to drive basal BCa development in breast cancer type 1 susceptibility protein (BRCA1) mutation carriers . As ER + BCa are a luminal tumour type that display luminal differentiation, we assessed which of the existing syngeneic lines were luminal in nature. Using normal mouse mammary epithelial cells as a standard plot, cell lines were assessed for the proportion of luminal, basal and stromal cells using flow cytometry . Lineage gates representing endothelial (CD31), leukocyte (CD45) and erythroid (TER119) populations were used, but, as anticipated, these cell lines were all lineage-negative. Lineage-negative cells were then assessed for expression of the epithelial marker EpCAM and the luminal/basal discriminator CD49f (α6 integrin) . The addition of CD49b (α2β1 integrin) is used as a progenitor marker and Sca-1 (Ly6a) as a marker of ER + cells . SSM3 cells were luminal in nature (CD49f+/EpCAM+) and found to have high expression of Sca-1, yet low expression of CD49b, which classifies them as a mature luminal cell type . 67NR and 4T1 cells showed a hybrid phenotype between stromal and basal, whilst J110 cells appeared basal. EO771 cells expressed stromal markers. D20R cells were again a hybrid between stromal and basal, but the populations were slightly shifted compared to the other cell lines with this phenotype. D2A1 cells were also stromal . Thus, the only luminal breast cancer cell line in this collection is the SSM3 line. 2.4. Molecular Characterisation of the Syngeneic Lines to Identify the Cell Lineage Gene expression array data exist for 4T1.2, 67NR, EO771 whole tumours (GSE42272) and for D2.0R (GSE112904) and D2A1 cells (GSE12882) , and to complete the gene expression analysis, we performed RNAseq on J110 and SSM3 cells. Using the gene expression signatures developed by Lim and colleagues , we assessed whether the J110 and SSM3 cells had higher expression of certain cell epithelial cell lineages. J110 cells had high expression of stromal, mammary stem cell (MaSC) and luminal progenitor signatures, and lower expression of luminal mature genes ( A,B). SSM3 cells had high expression of the luminal mature and to a lesser extent, the luminal progenitor signature, but low expression of the stromal and MaSC signatures ( A,B). The existing gene expression array data could not be included bioinformatically in the same analysis due to the differences in the platforms used. However, when we assessed the cell lineages in 4T1.2, 67NR and EO771 tumours, they showed the highest enrichment of stromal and luminal progenitor signatures and also MaSC gene signatures, with a very low expression of luminal mature genes ( C). Gene expression data from D2.0R and D2A1 cells grown in 3D in vitro had high expression of stromal, MaSC and luminal progenitor genes and low expression of luminal mature genes ( D). Finally, to show that the ER pathway is intact in SSM3 cells and thus mediating the effects of Tamoxifen in vitro and in vivo, we assessed whether they expressed genes, previously shown to be induced by estrogen in MCF7 cells ( PGR, CDH1, CTSD and TRIM25 ). shows that they are indeed increased in SSM3 compared J110 cells.
Formalin-fixed paraffin-embedded tissue sections from SSM3, 4T1.2, 67NR, J110, EO771, D2.0R and D2A1 tumours grown in mice were stained for Erα using a clinically validated antibody . The SSM3 tumours showed strong and extensive nuclear staining (>90%) in epithelial tumour cells . These tumours were also assessed pathologically and were classified as adenocarcinomas which are also an accurate representation of common ER + tumours . Similarly, 67NR tumours had extensive nuclear staining but weaker staining. Scattered nuclear staining was observed in D2A1, 4T1.2 and J110 tumours. However, this staining overlapped with areas of stromal and immune infiltration, as demonstrated by CD45 immunohistochemical staining on consecutive sections . D2.0R and EO771 were negative for nuclear Erα expression but displayed some regions of cytoplasmic staining . In almost all the different tumours, cytoplasmic Erα staining was observed in regions of the tumours that had immune infiltration identified through CD45 staining, indicating, as above, cytoplasmic staining of immune cells rather than epithelial tumour cells .
Using the well-characterised human ER + MCF7 cells, we conducted a dose–response to tamoxifen and showed that whilst high concentrations (10 µM) were cytotoxic, lower doses of tamoxifen could inhibit the growth of MCF7 cells . As a negative control, we included a dose–response in ER - MDA-MB-231 cells and showed that, apart from the cytotoxic dose (10 µM), these cells were insensitive to tamoxifen treatment. All mouse tumour cell lines were tested at 1 µM, which resulted in reduced proliferation of SSM3 cells. Previously we have shown that 67NR tumours are mildly sensitive to tamoxifen and that 4T1.2 tumours are insensitive . Again, we found that 67NR cells are mildly sensitive. J110 cells, D20R and D2A1 cells were all insensitive to growth inhibitory effects of tamoxifen . The EO771 cell line response to tamoxifen was not assessed, as it was already revealed it had no sensitivity . To validate our results in vivo we treated mice injected with SSM3 cells in their fourth mammary fat pads with 1 mg of tamoxifen daily. Tamoxifen significantly slowed tumour growth compared to vehicle controls . We did not observe any toxicity issues with this dosing regimen.
It is appreciated now that breast cancer arises from a dysregulation of the normal breast epithelium. The molecular definition of the cell lineages in the normal breast has been instrumental in allowing these relationships to be identified. For example, Receptor activator for nuclear factor κB (RANK)+ luminal progenitors are known to drive basal BCa development in breast cancer type 1 susceptibility protein (BRCA1) mutation carriers . As ER + BCa are a luminal tumour type that display luminal differentiation, we assessed which of the existing syngeneic lines were luminal in nature. Using normal mouse mammary epithelial cells as a standard plot, cell lines were assessed for the proportion of luminal, basal and stromal cells using flow cytometry . Lineage gates representing endothelial (CD31), leukocyte (CD45) and erythroid (TER119) populations were used, but, as anticipated, these cell lines were all lineage-negative. Lineage-negative cells were then assessed for expression of the epithelial marker EpCAM and the luminal/basal discriminator CD49f (α6 integrin) . The addition of CD49b (α2β1 integrin) is used as a progenitor marker and Sca-1 (Ly6a) as a marker of ER + cells . SSM3 cells were luminal in nature (CD49f+/EpCAM+) and found to have high expression of Sca-1, yet low expression of CD49b, which classifies them as a mature luminal cell type . 67NR and 4T1 cells showed a hybrid phenotype between stromal and basal, whilst J110 cells appeared basal. EO771 cells expressed stromal markers. D20R cells were again a hybrid between stromal and basal, but the populations were slightly shifted compared to the other cell lines with this phenotype. D2A1 cells were also stromal . Thus, the only luminal breast cancer cell line in this collection is the SSM3 line.
Gene expression array data exist for 4T1.2, 67NR, EO771 whole tumours (GSE42272) and for D2.0R (GSE112904) and D2A1 cells (GSE12882) , and to complete the gene expression analysis, we performed RNAseq on J110 and SSM3 cells. Using the gene expression signatures developed by Lim and colleagues , we assessed whether the J110 and SSM3 cells had higher expression of certain cell epithelial cell lineages. J110 cells had high expression of stromal, mammary stem cell (MaSC) and luminal progenitor signatures, and lower expression of luminal mature genes ( A,B). SSM3 cells had high expression of the luminal mature and to a lesser extent, the luminal progenitor signature, but low expression of the stromal and MaSC signatures ( A,B). The existing gene expression array data could not be included bioinformatically in the same analysis due to the differences in the platforms used. However, when we assessed the cell lineages in 4T1.2, 67NR and EO771 tumours, they showed the highest enrichment of stromal and luminal progenitor signatures and also MaSC gene signatures, with a very low expression of luminal mature genes ( C). Gene expression data from D2.0R and D2A1 cells grown in 3D in vitro had high expression of stromal, MaSC and luminal progenitor genes and low expression of luminal mature genes ( D). Finally, to show that the ER pathway is intact in SSM3 cells and thus mediating the effects of Tamoxifen in vitro and in vivo, we assessed whether they expressed genes, previously shown to be induced by estrogen in MCF7 cells ( PGR, CDH1, CTSD and TRIM25 ). shows that they are indeed increased in SSM3 compared J110 cells.
Syngeneic mouse models are a unique preclinical tool that is now heavily used to assess the impact of emerging therapies on cancer growth rates in animals with intact immune systems. This is particularly important now that the critical role of the immune system in cancer has been revealed and immune checkpoint inhibitors have been added to the arsenal of anti-cancer therapies . Previously it was reported that EO771 cells are luminal in nature due to the expression of ERβ, PR and ErbB2 in RT PCR data and that they are tamoxifen-sensitive . However, we show that neither the cell lines nor the tumours produced by EO771 cells are luminal or positive for ER + . In agreement with Johnstone et al. 2015 , we have also demonstrated their highly mammary stem cell and stromal genomic signatures, but no luminal signature. In the 67NR cells that have been reported previously as a non-metastatic ER + BCa cell line , we found weak ER staining and a partial sensitivity to tamoxifen, consistent with the conclusions reported in Johnstone et al. 2015 . 4T1.2 cells are generally considered to be TNBC, but we (and others) have shown that whilst insensitive to tamoxifen, they do express ER mRNA , indicating that transcript expression alone cannot be used to define preclinical ER + models. J110 cells were developed from Amplified In Breast 1 (AIB1)-overexpressing mice and were initially deemed ER + cells, as per IHC staining . However, several studies have subsequently shown that they are non-responsive to tamoxifen both in vitro and in vivo . We note that our studies show that J110 cells alone do not have a luminal phenotype using flow cytometry. In addition, nuclear and cytoplasmic staining of ERα in some tumours generated from J110 cells may be confusing the field regarding ER classification. However, the positive staining correlates with areas of immune infiltration, as delineated by CD45 + . It is known that some immune cells in the mammary gland, such as macrophages, are ER + , but are not routinely accounted for in the pathological classification of hormone receptor status in breast tumours . Our research shows that apart from SSM3 cells, none of the other mammary tumour cell lines had all the features of truly ER + cells; which are nuclear ER positivity by IHC in epithelial cells, a luminal cell surface phenotype (Sca1+), a luminal molecular subtype and sensitivity to tamoxifen exposure. SSM3 cells were derived from signal transducer and activator of transcription 1 ( STAT1) -deficient mice . STAT1 is a transcription factor required for interferon signalling . It is lost or significantly diminished in 45% of human ER + /PR+ breast cancers compared to normal breast tissue . Mice lacking STAT1 developed ER + breast tumours with a long latency (~23 months) . Three tumour cell lines have been established from these mice (SSM1, SSM2 and SSM3), with SSM3 cells showing ERα positivity and responsiveness in mice to oophorectomy (removing their source of endogenous estrogen) . In line with these findings, we confirm that the SSM3 cell line is a model of ER+ breast cancer. Whilst the SSM3 model is robust, it does not recapitulate the heterogenous nature of ER + BCa, as not all BCa are STAT1-deficient. The most common mutations in ER + BCa are PIK3CA , MLL3 , MAP3K1 , GATA3 , MAPK24 (and ER itself) . Further studies need to focus on developing additional syngeneic mouse models of ER + BCa that recapitulate the clinical disease. Such models need careful definition of their phenotype, including an extensive analysis of their luminal cell properties and their responsiveness to endocrine therapy. The presence of transcripts for ERα is insufficient for defining luminal ER+ tumours.
4.1. Cell Lines The EO771 cell line was derived from a spontaneous mammary tumour in a C57BL/6 mouse . The 67NR and 4T1 cell lines were derived from different subpopulations of a single mammary tumour that arose in a BALB/c/C3H mouse . The 4T1.2 variant was derived from 4T1 cells as described previously . The SSM3 cells (kindly gifted by Robert Schreiber, Washington University) were generated from mammary tumours that arose in 129SvEv STAT1-deficient mice . J110 cells (kindly gifted by Myles Brown, Dana Farber Cancer Institute) were derived from transgenic mice that expressed AIB1 cDNA under the control of the MMTV LTR . D2.0R and D2A1 cells (kindly gifted by Dalit Barkan, University of Haifer, Haifer, Israel) were derived from clones of a BALB/C mouse with a D2 hyperplastic alveolar nodule (HAN) . 67NR and 4T1.2 mammary tumour cells were maintained in Eagle’s minimum essential medium (alpha modification) supplemented with 5% ( v/v ) foetal bovine serum (FBS) (SAFC Biosciences, Brooklyn, Victoria, Australia) and 1% ( v/v ) penicillin-streptomycin. EO771 cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM) containing HEPES (20 mM) (Gibco, Billings, MT, USA) supplemented with 10% ( v/v ) FBS, penicillin (100 IU/mL) and streptomycin (100 μg/mL) (Gibco). SSM3 cells were maintained in DMEM/F12 + Hepes 15 mM, 10% FBS, 2% L-glutamine, 0.05 mM β-mercaptoethanol, 0.3 μM hydrocortisone (Sigma-Aldrich, Saint Louis, MO, USA), 5 μg/mL insulin, 10 ng/mL holo-transferrin. J110 cells were cultured in DMEM/F-12 (Corning, Corning, NY, USA) containing 5% FBS, 5 µg/mL insulin and 0.1 µg/mL hydrocortisone. MCF7 cells were maintained in RPMI supplemented with 10% ( v/v ) FBS, 1% ( v/v ) penicillin-streptomycin and 10 μg/mL insulin. MDA-MB-231 cells were maintained in L-glutamine Dulbecco’s modified Eagles’ medium (DMEM) (Gibco, Billings, MT, USA) supplemented with 10% FBS and 1% penicillin/streptomycin. All cells were cultured at 37 °C in 5% CO 2 ( v/v ) in air and were maintained in culture for a maximum of 4–5 weeks. The only exception to this was the SSM3 cells that are maintained in 10% CO 2 . 4.2. Isolation, Staining and Flow Cytometric Analysis of Normal and Transformed Mouse Mammary Cells Primary mouse mammary gland cells were used as controls for flow cytometry and were isolated using mechanical and enzymatic disaggregation as described previously . The epithelial subpopulations were isolated from the disaggregated samples by flow cytometry as detailed previously . Hematopoietic lineage cells were stained with anti-CD45-PE-Cy7, anti-TER119-PE-Cy7 and anti-CD31-PE-Cy7 (BD Biosciences, Franklin Lakes, NJ, USA). Cells were simultaneously stained with mammary-specific lineage markers, anti-Epcam BV421(BD Biosciences), anti-CD49f-APC (Biolegend, San Diego, CA, USA), anti-Sca-1-PE (BD Biosciences), anti-CD49b-FITC (Biolegend, San Diego, CA, USA) and propidium iodide (Sigma). Cells were analysed on a Fortessa-X20 flow cytometer (Becton Dickinson, Franklin Lakes, NJ, USA). Linear density contour plots were used to describe flow cytometry gates. Fluorescence-minus-one control gates defined marker-negative populations. 4.3. Immunohistochemical Analysis Tissue sections were stained using a standard protocol. In brief, slides were heated for antigen retrieval by pressure cooker treatment in 0.01 M sodium citrate buffer, pH 6.0 (125 °C for 3 min, 90 °C for 10 s). Sections were then blocked in 3% H 2 O 2 before blocking in 5% normal goat serum or CAS-Block Histochemical Reagent (Thermofisher, Waltham, MA, USA). Slides were incubated with primary antibodies and were incubated overnight at 4 °C. Biotin-conjugated goat anti-rabbit or anti-mouse secondary antibodies (Dako, Carpinteria, CA, USA) were used at a 1:250 or 1:300 dilution for 1 h at room temperature. Specific primary–secondary antibody complexes were detected using ABC reagent (Vector Laboratories, CA, USA) and visualised using a 3,3′-diaminobenzidine peroxidase substrate kit (Vector Laboratories, Burlingame, CA, USA). Sections were counterstained with hematoxylin, dehydrated and mounted. The primary antibodies were as follows: mouse monoclonal anti-human ERα (1:50, clone 6F11 Abcam, Cambridge, UK) and rat anti-mouse CD45 (1:300, CM5p, Novocastra, Newcastle upon Tyme, UK). Staining was viewed on a BX-53 light microscope (Olympus, Macquarie Park, NSW, Australia). Images were acquired using Spot software version 5.0 (Spot Imaging Solutions, Diagnostic Instruments Inc., Bentleigh East, VIC, Australia). 4.4. In Vitro Assessment of Tamoxifen Sensitivity Proliferation assays were completed using the sulforhodamine B (SRB) colorimetric assay as described previously . Cells were seeded into 96-well plates at relevant cell densities suited for each cell line. Proliferation was assessed in the presence of 0.1, 1 and 10 µM (MCF7, MDA-MB-231) or 1 µM (SSM3, 67NR, J110, D2.0R, D2A1) 4-hydroxytamoxifen (4-OHT) dissolved in ethanol and diluted to a final ethanol concentration of 1% ( v / v ). Statistical analysis was completed on GraphPad using a two-way ANOVA to determine significant differences. 4.5. In Vivo Assessment of Tamoxifen Sensitivity in the SSM3 Mouse Model All animal work was completed with approval from the Peter MacCallum Cancer Centre Animal Ethics Committee (E594). 129SvEv mice were injected with 100,000 SSM3 BCa cells in the right fourth inguinal mammary fat pad. 17β-estradiol pellets (0.3 mg) were implanted subcutaneously in the dorsal flank of each mouse. Tumour volume was monitored using digital callipers and once tumours reached a volume of 200 mm 3 (palpable), mice were administered either tamoxifen citrate (Sigma) at 1 mg/day or vehicle control via daily subcutaneous injection. 4.6. RNAseq Analysis J110 and SSM3 cells were grown to 70–80% confluency, detached and washed in PBS. RNA was extracted using Trizol lysis following the manufacturer’s instructions (Invitrogen) and then a phenol, chloroform extraction was performed. RNA was precipitated using isopropanol, cleaned with ethanol and then solubilised in RNase-free water. The quality and quantity of RNA was assessed on the Agilent 2200 Tapestation (Agilent Technologies, Santa Clara, CA, USA). The Molecular Genomics Core at the Peter MacCallum Cancer Centre used the QuantSeq 3′ mRNA-seq Library Prep Kit for Illumina (Lexogen 015) according to the manufacturer’s instructions to generate the sequencing libraries from 500 ng of purified RNA ( n = 3 biological replicates of each cell line). We then generated 75 bp single end reads with a depth of ~6 million reads per sample using the Illumina NextSeq500. Sequencing reads were demultiplexed using bcl2fastq (v 2.17.1.14), low quality (Q < 30) reads removed, and trimmed at the 5′ and 3′ ends using cutadapt (v 2.1) to remove adapter sequences and poly-A-tail-derived reads, respectively. Sequencing reads were mapped to the mouse reference genome (mm10) using HISAT2 (v 2.0.4) and counted using the featureCounts command of the Subread package (v 2.0.0) . Read normalisation and differential gene expression analysis were performed in R (v 4.1.0) using R packages limma (v 3.48.3) and EdgeR (v 3.34.0) . The data from SSM3 and J110 cells have been deposited on Gene Expression Omnibus GSE226910. GSE112094 and GSE172882 were mapped using STAR (v 2.5.3a) , counted using Subread (v 2.0.0) and voom normalised using limma (v 3.48.3) . Normalised expression from GSE42272 was collapsed to the median gene level. Signature scores using MaSC, luminal progenitor and luminal mature gene signatures and stromal gene signature were calculated as previously defined by Lim et al. 2009 . Briefly, the signature score is calculated for each sample and is the average log expression of the genes in the signature weighted by the direction and magnitude (logFC) . R packages pheatmap (v 1.0.12), ggplot2 (v 3.3.5) and ggrepel (v 0.9.1) were used to generate figures.
The EO771 cell line was derived from a spontaneous mammary tumour in a C57BL/6 mouse . The 67NR and 4T1 cell lines were derived from different subpopulations of a single mammary tumour that arose in a BALB/c/C3H mouse . The 4T1.2 variant was derived from 4T1 cells as described previously . The SSM3 cells (kindly gifted by Robert Schreiber, Washington University) were generated from mammary tumours that arose in 129SvEv STAT1-deficient mice . J110 cells (kindly gifted by Myles Brown, Dana Farber Cancer Institute) were derived from transgenic mice that expressed AIB1 cDNA under the control of the MMTV LTR . D2.0R and D2A1 cells (kindly gifted by Dalit Barkan, University of Haifer, Haifer, Israel) were derived from clones of a BALB/C mouse with a D2 hyperplastic alveolar nodule (HAN) . 67NR and 4T1.2 mammary tumour cells were maintained in Eagle’s minimum essential medium (alpha modification) supplemented with 5% ( v/v ) foetal bovine serum (FBS) (SAFC Biosciences, Brooklyn, Victoria, Australia) and 1% ( v/v ) penicillin-streptomycin. EO771 cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM) containing HEPES (20 mM) (Gibco, Billings, MT, USA) supplemented with 10% ( v/v ) FBS, penicillin (100 IU/mL) and streptomycin (100 μg/mL) (Gibco). SSM3 cells were maintained in DMEM/F12 + Hepes 15 mM, 10% FBS, 2% L-glutamine, 0.05 mM β-mercaptoethanol, 0.3 μM hydrocortisone (Sigma-Aldrich, Saint Louis, MO, USA), 5 μg/mL insulin, 10 ng/mL holo-transferrin. J110 cells were cultured in DMEM/F-12 (Corning, Corning, NY, USA) containing 5% FBS, 5 µg/mL insulin and 0.1 µg/mL hydrocortisone. MCF7 cells were maintained in RPMI supplemented with 10% ( v/v ) FBS, 1% ( v/v ) penicillin-streptomycin and 10 μg/mL insulin. MDA-MB-231 cells were maintained in L-glutamine Dulbecco’s modified Eagles’ medium (DMEM) (Gibco, Billings, MT, USA) supplemented with 10% FBS and 1% penicillin/streptomycin. All cells were cultured at 37 °C in 5% CO 2 ( v/v ) in air and were maintained in culture for a maximum of 4–5 weeks. The only exception to this was the SSM3 cells that are maintained in 10% CO 2 .
Primary mouse mammary gland cells were used as controls for flow cytometry and were isolated using mechanical and enzymatic disaggregation as described previously . The epithelial subpopulations were isolated from the disaggregated samples by flow cytometry as detailed previously . Hematopoietic lineage cells were stained with anti-CD45-PE-Cy7, anti-TER119-PE-Cy7 and anti-CD31-PE-Cy7 (BD Biosciences, Franklin Lakes, NJ, USA). Cells were simultaneously stained with mammary-specific lineage markers, anti-Epcam BV421(BD Biosciences), anti-CD49f-APC (Biolegend, San Diego, CA, USA), anti-Sca-1-PE (BD Biosciences), anti-CD49b-FITC (Biolegend, San Diego, CA, USA) and propidium iodide (Sigma). Cells were analysed on a Fortessa-X20 flow cytometer (Becton Dickinson, Franklin Lakes, NJ, USA). Linear density contour plots were used to describe flow cytometry gates. Fluorescence-minus-one control gates defined marker-negative populations.
Tissue sections were stained using a standard protocol. In brief, slides were heated for antigen retrieval by pressure cooker treatment in 0.01 M sodium citrate buffer, pH 6.0 (125 °C for 3 min, 90 °C for 10 s). Sections were then blocked in 3% H 2 O 2 before blocking in 5% normal goat serum or CAS-Block Histochemical Reagent (Thermofisher, Waltham, MA, USA). Slides were incubated with primary antibodies and were incubated overnight at 4 °C. Biotin-conjugated goat anti-rabbit or anti-mouse secondary antibodies (Dako, Carpinteria, CA, USA) were used at a 1:250 or 1:300 dilution for 1 h at room temperature. Specific primary–secondary antibody complexes were detected using ABC reagent (Vector Laboratories, CA, USA) and visualised using a 3,3′-diaminobenzidine peroxidase substrate kit (Vector Laboratories, Burlingame, CA, USA). Sections were counterstained with hematoxylin, dehydrated and mounted. The primary antibodies were as follows: mouse monoclonal anti-human ERα (1:50, clone 6F11 Abcam, Cambridge, UK) and rat anti-mouse CD45 (1:300, CM5p, Novocastra, Newcastle upon Tyme, UK). Staining was viewed on a BX-53 light microscope (Olympus, Macquarie Park, NSW, Australia). Images were acquired using Spot software version 5.0 (Spot Imaging Solutions, Diagnostic Instruments Inc., Bentleigh East, VIC, Australia).
Proliferation assays were completed using the sulforhodamine B (SRB) colorimetric assay as described previously . Cells were seeded into 96-well plates at relevant cell densities suited for each cell line. Proliferation was assessed in the presence of 0.1, 1 and 10 µM (MCF7, MDA-MB-231) or 1 µM (SSM3, 67NR, J110, D2.0R, D2A1) 4-hydroxytamoxifen (4-OHT) dissolved in ethanol and diluted to a final ethanol concentration of 1% ( v / v ). Statistical analysis was completed on GraphPad using a two-way ANOVA to determine significant differences.
All animal work was completed with approval from the Peter MacCallum Cancer Centre Animal Ethics Committee (E594). 129SvEv mice were injected with 100,000 SSM3 BCa cells in the right fourth inguinal mammary fat pad. 17β-estradiol pellets (0.3 mg) were implanted subcutaneously in the dorsal flank of each mouse. Tumour volume was monitored using digital callipers and once tumours reached a volume of 200 mm 3 (palpable), mice were administered either tamoxifen citrate (Sigma) at 1 mg/day or vehicle control via daily subcutaneous injection.
J110 and SSM3 cells were grown to 70–80% confluency, detached and washed in PBS. RNA was extracted using Trizol lysis following the manufacturer’s instructions (Invitrogen) and then a phenol, chloroform extraction was performed. RNA was precipitated using isopropanol, cleaned with ethanol and then solubilised in RNase-free water. The quality and quantity of RNA was assessed on the Agilent 2200 Tapestation (Agilent Technologies, Santa Clara, CA, USA). The Molecular Genomics Core at the Peter MacCallum Cancer Centre used the QuantSeq 3′ mRNA-seq Library Prep Kit for Illumina (Lexogen 015) according to the manufacturer’s instructions to generate the sequencing libraries from 500 ng of purified RNA ( n = 3 biological replicates of each cell line). We then generated 75 bp single end reads with a depth of ~6 million reads per sample using the Illumina NextSeq500. Sequencing reads were demultiplexed using bcl2fastq (v 2.17.1.14), low quality (Q < 30) reads removed, and trimmed at the 5′ and 3′ ends using cutadapt (v 2.1) to remove adapter sequences and poly-A-tail-derived reads, respectively. Sequencing reads were mapped to the mouse reference genome (mm10) using HISAT2 (v 2.0.4) and counted using the featureCounts command of the Subread package (v 2.0.0) . Read normalisation and differential gene expression analysis were performed in R (v 4.1.0) using R packages limma (v 3.48.3) and EdgeR (v 3.34.0) . The data from SSM3 and J110 cells have been deposited on Gene Expression Omnibus GSE226910. GSE112094 and GSE172882 were mapped using STAR (v 2.5.3a) , counted using Subread (v 2.0.0) and voom normalised using limma (v 3.48.3) . Normalised expression from GSE42272 was collapsed to the median gene level. Signature scores using MaSC, luminal progenitor and luminal mature gene signatures and stromal gene signature were calculated as previously defined by Lim et al. 2009 . Briefly, the signature score is calculated for each sample and is the average log expression of the genes in the signature weighted by the direction and magnitude (logFC) . R packages pheatmap (v 1.0.12), ggplot2 (v 3.3.5) and ggrepel (v 0.9.1) were used to generate figures.
|
Menstrual, fertility and psychological impacts after uterine compression sutures for postpartum hemorrhage: a prospective cohort study | 601e8396-eadb-4ab9-82e7-193473a210b4 | 10053948 | Suturing[mh] | Uterine atony is the commonest etiology of primary postpartum hemorrhage (PPH), ranging from 30 to 80% worldwide . In the event of failed hemostasis with uterotonic agents, early recourse to conservative surgical management is advisable to prevent maternal morbidity and mortality . Since 1997, various techniques of uterine compression sutures were described to treat atony, such as B-lynch suture, Hayman suture and Cho suture [ – ]. Being relatively easy and quick to perform, these sutures were efficacious in preventing 97% of hemostatic hysterectomies . In our previous publication, 75% of hysterectomies were prevented in women treated with uterine compression sutures with additional second-line hemostatic procedures . Overall, the short-term complication rate was low. While uterine compression sutures were considered effective and safe, literatures on long-term menstruation and fertility outcomes were scarce. The fertility rate after different sutures ranged from 11 to 75% . Limiting by small number of women included, these studies suggested most pregnancies were uncomplicated and carried till term [ – ]. Nevertheless, rare complications of fundal uterine rupture at third trimester and placenta accreta spectrum overlying the rupture site were reported which raised the concern of whether localized uterine necrosis after compression sutures might increase the risk of myometrial weakening and abnormal placental implantation [ – ]. Menstrual outcome is another important indicator of fertility preservation, especially in women with no desire for future pregnancy. Limited available studies suggested no significant change in menstrual pattern was observed after compression sutures with 91.5 to 100% of these women reported return of menstruation within eight months after delivery . Nevertheless, these data might subject to recall bias as most studies were retrospective. Psychological impact among women with PPH without hysterectomy is often neglected. Women described the experience as ‘a feeling of powerlessness’ and ‘a fear of bleeding until death takes over’ . Sentilhes reported two-third of women with pervasive negative memory and fear of death after PPH . In addition, 5.9% of women reported sexual problems and 60% suffered from intense anxiety during their next pregnancy . Up till now, there has been no research investigating the psychological impacts of both women and their partners after uterine compression sutures for PPH. In the present study, we aim to determine any change in menstrual pattern, adverse pregnancy outcomes and psychological impact in women after uterine compression sutures.
This was a prospective cohort study conducted in Tuen Mun Hospital, Hong Kong SAR over a 13-year period. The inclusion criteria were women delivered with primary PPH (blood loss > = 500ml) successfully treated with uterine compression sutures between January 2009 to June 2022 who consented to this study. We performed uterine compression suture as the first line surgical treatment for woman undergoing caesarean section with uterine atony after failing standard uterotonic agents (oxytocin, ergometrine, carbetocin, carboprost and misoprostol). The types of compression sutures performed would be at obstetrician’s discretion. In case of unsuccessful hemostasis, uterine artery ligation was performed for further devascularization, followed by intraoperative uterine artery embolization by radiologist if woman was hemodynamically stable, or hysterectomy if she was unstable. In case of hemorrhage from placenta previa, we performed uterine artery ligation as the first line surgical treatment for devascularization, followed by compression suture if there remained localized placental site bleeding. As for atonic PPH after completion of caesarean section or vaginal delivery, more conservative management including intrauterine balloon tamponade and uterine artery embolization performed in interventional radiology suite were considered. Principles outlined in the Declaration of Helsinki were followed and this study was approved by the New Territories West Cluster Research Ethics Committee, Hospital Authority, Hong Kong SAR. All compression sutures were applied by specialists who had regular training on obstetric emergencies. B-lynch suture, Hayman suture and Cho suture were included in our study design and surgical techniques were strictly followed [ – ]. Monocryl-1 (polyglecaprone 25) absorbable suture was used for B-Lynch and Hayman suture while Vicryl-1 (polyglactin 910) was used for Cho suture. Uterine artery ligations were performed with Vicryl-1 (polyglactin 910) suture. Bakri balloon (Cook Medical, US) was used as intrauterine balloon tamponade. Uterine artery embolization was performed by radiologist intraoperatively in theatre or in interventional radiology suite, involving injection of absorbable gelatin sponges to extravasating uterine vessels for devascularization. Women were excluded from the present study if hemostatic hysterectomies were performed. Baseline characteristics and clinical details of women were identified through both clinical notes and electronic patient record system. The baseline psychological status of women was assessed by the Edinburgh Postnatal Depression Scale (EPDS), which is a questionnaire conducted by adding together the scores of 10 items, indicating the severity of depressive symptoms. Menstrual and psychological assessment Women were assessed in postnatal clinic at six weeks, four months, one year and two years after delivery. Informed consent was obtained. During each visit, they were enquired about the mode of feeding, contraception, cessation of lochia, return of menstruation and a full menstrual history. Primary endpoints were defined as regular cycle, no change in menstrual flow, no change in menstrual days, no dysmenorrhea or no change in the severity of dysmenorrhea as compared to before. In case of menstrual abnormalities, gynecological examination, pregnancy test, cervical smear, ultrasound of pelvis (GE Healthcare Voluson) and endometrial biopsy with or without diagnostic hysteroscopy (KARL STORZ) were performed to rule out structural causes. Genital swabs and serum hormone level (estradiol, follicle-stimulating hormone, prolactin and thyroid hormone) were performed if clinically indicated. In the last session of the follow-up, women were assessed on their psychological impact using a standardized questionnaire designed by Sentilhes . Subsequent pregnancy outcomes In June 2022, we identified all subsequent pregnancies of these women delivered in any public hospitals in Hong Kong SAR through territory-wide electronic registry system. To identify any subsequent pregnancies delivered in private sector or outside our territory, all women received telephone interviews and a list of questions relating to their pregnancies outcomes were asked. A standardized script was available for obtaining consent over telephone. Control group was identified by the next five consecutive women who suffered from atonic PPH, successfully treated with uterotonic agents and had subsequent pregnancies. Women in both groups were matched by their order of pregnancies, number and mode of deliveries. Women with any other previous surgeries were excluded. The demographics and clinical details of these women were retrieved in the same manner as study group. Baseline characteristics of subsequent pregnancies including age of women, interpregnancy interval and order of pregnancy were collected. Number of subsequent ectopic pregnancies, miscarriages, terminations, pregnancies beyond 24 weeks gestation, hypertensive disease, placenta previa, placenta accreta, preeclampsia and uterine rupture were compared. Gestation at delivery, birth weight, numbers of small-for-gestational age fetuses, operative findings, recurrence of atonic PPH and number of repeated compression sutures were compared between the two groups. Statistical analysis SPSS Statistics version 21 (IBM, Armonk, NY) was used. Categorical and continuous variables were presented as n (%) and median (range) respectively. Pearson chi-square test and Fisher’s exact test were used for comparing dichotomous data while Student’s t -test and Mann-Whitney U test were used for comparing continuous data. A two-sided P value less than 0.05 was considered statistically significant.
Women were assessed in postnatal clinic at six weeks, four months, one year and two years after delivery. Informed consent was obtained. During each visit, they were enquired about the mode of feeding, contraception, cessation of lochia, return of menstruation and a full menstrual history. Primary endpoints were defined as regular cycle, no change in menstrual flow, no change in menstrual days, no dysmenorrhea or no change in the severity of dysmenorrhea as compared to before. In case of menstrual abnormalities, gynecological examination, pregnancy test, cervical smear, ultrasound of pelvis (GE Healthcare Voluson) and endometrial biopsy with or without diagnostic hysteroscopy (KARL STORZ) were performed to rule out structural causes. Genital swabs and serum hormone level (estradiol, follicle-stimulating hormone, prolactin and thyroid hormone) were performed if clinically indicated. In the last session of the follow-up, women were assessed on their psychological impact using a standardized questionnaire designed by Sentilhes .
In June 2022, we identified all subsequent pregnancies of these women delivered in any public hospitals in Hong Kong SAR through territory-wide electronic registry system. To identify any subsequent pregnancies delivered in private sector or outside our territory, all women received telephone interviews and a list of questions relating to their pregnancies outcomes were asked. A standardized script was available for obtaining consent over telephone. Control group was identified by the next five consecutive women who suffered from atonic PPH, successfully treated with uterotonic agents and had subsequent pregnancies. Women in both groups were matched by their order of pregnancies, number and mode of deliveries. Women with any other previous surgeries were excluded. The demographics and clinical details of these women were retrieved in the same manner as study group. Baseline characteristics of subsequent pregnancies including age of women, interpregnancy interval and order of pregnancy were collected. Number of subsequent ectopic pregnancies, miscarriages, terminations, pregnancies beyond 24 weeks gestation, hypertensive disease, placenta previa, placenta accreta, preeclampsia and uterine rupture were compared. Gestation at delivery, birth weight, numbers of small-for-gestational age fetuses, operative findings, recurrence of atonic PPH and number of repeated compression sutures were compared between the two groups.
SPSS Statistics version 21 (IBM, Armonk, NY) was used. Categorical and continuous variables were presented as n (%) and median (range) respectively. Pearson chi-square test and Fisher’s exact test were used for comparing dichotomous data while Student’s t -test and Mann-Whitney U test were used for comparing continuous data. A two-sided P value less than 0.05 was considered statistically significant.
Study participants Over the 13-year study period, 90 uterine compression sutures (0.1%) were performed in the background of 80,087 deliveries. After excluding 10 women who had hysterectomies performed due to failed hemostasis, 80 women were eligible for this study (Figure ). The baseline characteristics of women with uterine compression sutures were shown in Table . Most participants were primiparous young women with singleton pregnancy. All women delivered by cesarean sections with all compression sutures performed at the same operation. 86.3% of uterine compression sutures were applied for uterine atony. There were nine postpartum short-term complications identified, including four hematometra requiring drainage, four endometritis treated with antibiotics and one retained product of conception requiring surgical evacuation. Majority (97.5%) of women had no antecedent psychiatric disorders with a median Edinburgh Postnatal Depression Scale (EPDS) score of three at their sixth week appointment. Seventy-six (95%) of them had EPDS score fewer than 10. Menstrual outcome and psychological impact Sixty-eight (85%) out of 80 women were compliant to our prospective assessment. The median follow-up period was 12 months (range 6–20 months). Of those who were not on exclusive breastfeeding, 87.9% of women reported return of menses within six months after delivery at a median of nine weeks (Table ). Of note, all participants had return of menses by the end of their last visits. Regular monthly cycle was observed in 95.6% of women with over three quarters of them reporting similar menstrual flow (75%), menstrual days (85.3%) and no change in dysmenorrhea status (88.2%) when compared to before. Eight (11.8%) women reported hypomenorrhea after uterine compression sutures. Among them, two (25%) were diagnosed with Asherman’s syndrome (Figure ), one (12.5%) with mild synechiae and one (12.5%) delivered at age 53 was perimenopausal. None of them had future fertility wish. Among other women experiencing changes in menstrual pattern, workup did not reveal any structural causes. Regarding assessment on psychological impact, all women completed questionnaires at the end of their last visits (Table ). Among them, 72.1% declined any more fertility wish with 26.5% did so due to fear of recurrence of PPH. Importantly, 38.2% of women recalled unpleasant memories especially fear of pain and death; and 22.1% of them reported life-long adverse impact especially tokophobia (73.3%). All participants denied any long-term effects on sexual function or marital relationship. When women were asked about their partners’ views on the delivery episodes, 54.4% of them recalled unpleasant memories and 51.5% declined any wish for future pregnancy. Subsequent pregnancy outcomes All women responded to our telephone interviews. None of them reported infertility. Among the 19 (27.9%) women who expressed wishes of getting pregnant again in postnatal clinic, 17 (89.5%) of them conceived. Among all 23 subsequent pregnancies, there were 16 livebirths, four miscarriages, two terminations due to fetal anomaly and one ectopic pregnancy (Table ). All pregnancies were conceived naturally. Of the 16 singleton livebirths, there were no cases of preeclampsia, placenta accreta spectrum or uterine rupture. The median gestational age, term delivery, preterm delivery, stillbirth, birth weight and the proportion of small-for-gestational-age fetuses were comparable between both study and control groups. All women with previous uterine compression sutures decided to have repeated cesarean sections in their next pregnancies. Concerning operative finding, there were significant higher risk of omental or bowel adhesions to the uterus in women with previous uterine compression sutures (37.5% vs. 8.8%, p = 0.007). Eleven out of 16 women suffered from recurrence of atonic PPH (68.8% vs. 7.5%, p < 0.001) with a median blood loss of 800mL compared with 250mL in the control group ( p < 0.001). Two women in the study group failed medical treatment requiring repeated uterine compression sutures for hemostasis (12.5% vs. 0%, p = 0.024). No peripartum hysterectomies were performed in our cohort.
Over the 13-year study period, 90 uterine compression sutures (0.1%) were performed in the background of 80,087 deliveries. After excluding 10 women who had hysterectomies performed due to failed hemostasis, 80 women were eligible for this study (Figure ). The baseline characteristics of women with uterine compression sutures were shown in Table . Most participants were primiparous young women with singleton pregnancy. All women delivered by cesarean sections with all compression sutures performed at the same operation. 86.3% of uterine compression sutures were applied for uterine atony. There were nine postpartum short-term complications identified, including four hematometra requiring drainage, four endometritis treated with antibiotics and one retained product of conception requiring surgical evacuation. Majority (97.5%) of women had no antecedent psychiatric disorders with a median Edinburgh Postnatal Depression Scale (EPDS) score of three at their sixth week appointment. Seventy-six (95%) of them had EPDS score fewer than 10.
Sixty-eight (85%) out of 80 women were compliant to our prospective assessment. The median follow-up period was 12 months (range 6–20 months). Of those who were not on exclusive breastfeeding, 87.9% of women reported return of menses within six months after delivery at a median of nine weeks (Table ). Of note, all participants had return of menses by the end of their last visits. Regular monthly cycle was observed in 95.6% of women with over three quarters of them reporting similar menstrual flow (75%), menstrual days (85.3%) and no change in dysmenorrhea status (88.2%) when compared to before. Eight (11.8%) women reported hypomenorrhea after uterine compression sutures. Among them, two (25%) were diagnosed with Asherman’s syndrome (Figure ), one (12.5%) with mild synechiae and one (12.5%) delivered at age 53 was perimenopausal. None of them had future fertility wish. Among other women experiencing changes in menstrual pattern, workup did not reveal any structural causes. Regarding assessment on psychological impact, all women completed questionnaires at the end of their last visits (Table ). Among them, 72.1% declined any more fertility wish with 26.5% did so due to fear of recurrence of PPH. Importantly, 38.2% of women recalled unpleasant memories especially fear of pain and death; and 22.1% of them reported life-long adverse impact especially tokophobia (73.3%). All participants denied any long-term effects on sexual function or marital relationship. When women were asked about their partners’ views on the delivery episodes, 54.4% of them recalled unpleasant memories and 51.5% declined any wish for future pregnancy.
All women responded to our telephone interviews. None of them reported infertility. Among the 19 (27.9%) women who expressed wishes of getting pregnant again in postnatal clinic, 17 (89.5%) of them conceived. Among all 23 subsequent pregnancies, there were 16 livebirths, four miscarriages, two terminations due to fetal anomaly and one ectopic pregnancy (Table ). All pregnancies were conceived naturally. Of the 16 singleton livebirths, there were no cases of preeclampsia, placenta accreta spectrum or uterine rupture. The median gestational age, term delivery, preterm delivery, stillbirth, birth weight and the proportion of small-for-gestational-age fetuses were comparable between both study and control groups. All women with previous uterine compression sutures decided to have repeated cesarean sections in their next pregnancies. Concerning operative finding, there were significant higher risk of omental or bowel adhesions to the uterus in women with previous uterine compression sutures (37.5% vs. 8.8%, p = 0.007). Eleven out of 16 women suffered from recurrence of atonic PPH (68.8% vs. 7.5%, p < 0.001) with a median blood loss of 800mL compared with 250mL in the control group ( p < 0.001). Two women in the study group failed medical treatment requiring repeated uterine compression sutures for hemostasis (12.5% vs. 0%, p = 0.024). No peripartum hysterectomies were performed in our cohort.
Main findings Our study demonstrated majority of women had return of menstruation with similar patterns after uterine compression sutures. Two cases of Asherman’s syndrome and one mild uterine synechiae were diagnosed among eight women with hypomenorrhea. None of our subjects contemplating next pregnancy reported infertility. Among the 23 subsequent pregnancies, despite most outcome parameters were comparable between women in both study and control groups, we observed a higher incidence of omental or bowel adhesion, recurrence of atonic PPH and repeated compression sutures in the study group. A significant proportion of women were reluctant for future pregnancy after compression sutures with more than one-fifth of them admitting unpleasant memories and long-term adverse impact. Strengths and limitations The main strength in our study is the two-year longitudinal clinical follow-up with high patients’ compliance. Thus, clinical information was recorded prospectively minimizing recall bias. In addition, by retrieving pregnancy outcomes from both electronic system and telephone interviews, we improved our data quality in terms of accuracy and exhaustiveness by including early pregnancy loss and terminations which might not had been managed in public sectors. Apart from pregnancies, we also investigated other indicators of subsequent fertility including couple’s fertility wish, workup of any subfertility and return of menstruation. Most of the previous publications included women with B-lynch sutures while there have been reports suggesting probable higher rate of complications after Cho sutures or when compression suture was combined with pelvic vessel ligation . Thus, another distinct advantage is that we included a variety of compression sutures and a combination with other uterine sparing techniques. Our study is also the first which reported the psychological impact of women after uterine compression sutures. One of our limitations is all of our women with compression sutures had it performed during caesarean section and none at relaparotomy or after vaginal delivery. Furthermore, we did not record the decision time of compression sutures from the start of operation which could have affected the subsequent outcomes. Another pitfalls in our study is the lack of comparison of menstrual and psychological outcomes with those in control group. One of our limitations is the potential underestimation of women with mild asymptomatic uterine synechiae as the indication of hysteroscopy was symptom-driven. In addition, since the last assessment ranged from six months to 20 months postpartum, the results of women’s psychological assessment could be influenced by a change in psychological status with time. Moreover, we did not take into account of women’s social support, socioeconomic status and newborn’s condition which could have affected the psychological outcomes of them. Finally, since we did not invite the partners for direct interview, their responses were indirectly collected from our participants. Interpretation Uterine compression sutures are easier, quicker and cost-effective to perform comparing with other uterine conservation techniques. Overall, data on long-term pregnancy outcomes has been reassuring [ , – ]. Nevertheless, most of the literatures were case reports or small case series. Poujade O et al. reported uterine synechiae in one-fourth of their cohort which is higher than our result . It is likely that our study might have underestimated women with mild asymptomatic uterine synechiae as hysteroscopy was offered only in symptomatic women. In our study, there were two women with Asherman’s syndrome and one mild uterine synechiae. Both of our women with Asherman’s syndrome had Hayman sutures applied during their elective cesarean sections. Their immediate post-operative course was unremarkable. The first woman with uterine atony secondary to amniotic fluid embolism also had uterine artery ligations and later pelvic vessel embolization six hours after operation due to persistent bleeding. The procedure was performed by radiologist with gelatin sponge particles in interventional radiologist suite. We postulated the severe hemodynamic shock and the combination of three devascularization techniques might have contributed to uterine ischemia and scarring over the endometrium. The second woman had caesarean sections performed due to cephalopelvic disproportion. Hayman sutures and uterine artery ligations were performed due to uterine atony. Another woman with mild uterine synechiae had elective caesarean sections performed complicating with uterine atony requiring Hayman’s sutures. Presenting with pelvic pain at her second week after delivery, hematometra was diagnosed requiring drainage twice. We agree with other authors that compression sutures might potentiate the formation of endometrial scarring and ischemic damage by excessive suture tension on the myometrium . Moreover, hematometra and uterine synechiae might be more prevalent after Hayman and Cho sutures as they directly obliterate the anterior and posterior uterine walls. In order to prevent endometrial scarring, application of appropriate tension on uterine walls by suture material which is firm, monofilament and quickly absorbed is of paramount importance. Recently, there were reports of novel compression sutures which were removed in early postoperative period . Nevertheless, the sample size was small and further evaluation on long-term outcome is needed before clinical application. Our subsequent pregnancy rate was 29% which corresponds well with previous studies . Pregnancies from three months to ten years after uterine compression sutures have been reported . Similarly, in the present study, the interpregnancy interval was between five months to nine years. Comparing with control group with no compression sutures performed, we did not identify any differences in the rate of preterm birth and small-for-gestational-age fetus. Regarding mode of delivery, literatures reported successful vaginal deliveries in up to one-third of the cases . Interestingly, none of our women chose vaginal delivery. Taking into account of their response in the psychological interviews, their negative memories and fear of recurrence of PPH might explain their wish of having the deliveries conducted in a relatively more controlled manner. We observed significantly more atonic PPH (68.8%) and repeated compression sutures (12.5%) in our cohort. Similar findings were reported by Fuglsang who observed 15% of subsequent deliveries required repeated B-Lynch sutures or hysterectomies . In addition, significantly more omental or bowel adhesions were noted in one-third of our cohort. Hence, our results suggest a higher level of anticipation of PPH and surgical difficulty should be considered in subsequent deliveries. Previous studies reported severe complications of uterine rupture in next pregnancies [ – ]. In our cohort, we reported none but one woman with evidence of grooves over the uterus which was likely an ischemic defect suggestive of excessive pressure on the myometrium. The number of women with uterine grooves is likely to be underreported as these features could only be diagnosed during operation. Therefore, high vigilance of uterine dehiscence or rupture should be maintained in any women presenting with abdominal pain antenatally. Overall, with the small sample size, it is difficult to draw a conclusion about the effect of compression sutures on subsequent pregnancy outcomes. Remarkably, 72.1% of women and 51.5% of their partners were reluctant for future pregnancy. Nearly one-fourth of them and over half of their partners reported negative memories afterwards. Significantly, among those with long-term adverse impact after the delivery episodes, 73.3% had tokophobia. With our result, we suggest involving professional input in debriefing and providing psychological support sensitively to couple would be beneficial in early postnatal period. Conducting a longer period of research is required to assess the impact to their long-term quality of life.
Our study demonstrated majority of women had return of menstruation with similar patterns after uterine compression sutures. Two cases of Asherman’s syndrome and one mild uterine synechiae were diagnosed among eight women with hypomenorrhea. None of our subjects contemplating next pregnancy reported infertility. Among the 23 subsequent pregnancies, despite most outcome parameters were comparable between women in both study and control groups, we observed a higher incidence of omental or bowel adhesion, recurrence of atonic PPH and repeated compression sutures in the study group. A significant proportion of women were reluctant for future pregnancy after compression sutures with more than one-fifth of them admitting unpleasant memories and long-term adverse impact.
The main strength in our study is the two-year longitudinal clinical follow-up with high patients’ compliance. Thus, clinical information was recorded prospectively minimizing recall bias. In addition, by retrieving pregnancy outcomes from both electronic system and telephone interviews, we improved our data quality in terms of accuracy and exhaustiveness by including early pregnancy loss and terminations which might not had been managed in public sectors. Apart from pregnancies, we also investigated other indicators of subsequent fertility including couple’s fertility wish, workup of any subfertility and return of menstruation. Most of the previous publications included women with B-lynch sutures while there have been reports suggesting probable higher rate of complications after Cho sutures or when compression suture was combined with pelvic vessel ligation . Thus, another distinct advantage is that we included a variety of compression sutures and a combination with other uterine sparing techniques. Our study is also the first which reported the psychological impact of women after uterine compression sutures. One of our limitations is all of our women with compression sutures had it performed during caesarean section and none at relaparotomy or after vaginal delivery. Furthermore, we did not record the decision time of compression sutures from the start of operation which could have affected the subsequent outcomes. Another pitfalls in our study is the lack of comparison of menstrual and psychological outcomes with those in control group. One of our limitations is the potential underestimation of women with mild asymptomatic uterine synechiae as the indication of hysteroscopy was symptom-driven. In addition, since the last assessment ranged from six months to 20 months postpartum, the results of women’s psychological assessment could be influenced by a change in psychological status with time. Moreover, we did not take into account of women’s social support, socioeconomic status and newborn’s condition which could have affected the psychological outcomes of them. Finally, since we did not invite the partners for direct interview, their responses were indirectly collected from our participants.
Uterine compression sutures are easier, quicker and cost-effective to perform comparing with other uterine conservation techniques. Overall, data on long-term pregnancy outcomes has been reassuring [ , – ]. Nevertheless, most of the literatures were case reports or small case series. Poujade O et al. reported uterine synechiae in one-fourth of their cohort which is higher than our result . It is likely that our study might have underestimated women with mild asymptomatic uterine synechiae as hysteroscopy was offered only in symptomatic women. In our study, there were two women with Asherman’s syndrome and one mild uterine synechiae. Both of our women with Asherman’s syndrome had Hayman sutures applied during their elective cesarean sections. Their immediate post-operative course was unremarkable. The first woman with uterine atony secondary to amniotic fluid embolism also had uterine artery ligations and later pelvic vessel embolization six hours after operation due to persistent bleeding. The procedure was performed by radiologist with gelatin sponge particles in interventional radiologist suite. We postulated the severe hemodynamic shock and the combination of three devascularization techniques might have contributed to uterine ischemia and scarring over the endometrium. The second woman had caesarean sections performed due to cephalopelvic disproportion. Hayman sutures and uterine artery ligations were performed due to uterine atony. Another woman with mild uterine synechiae had elective caesarean sections performed complicating with uterine atony requiring Hayman’s sutures. Presenting with pelvic pain at her second week after delivery, hematometra was diagnosed requiring drainage twice. We agree with other authors that compression sutures might potentiate the formation of endometrial scarring and ischemic damage by excessive suture tension on the myometrium . Moreover, hematometra and uterine synechiae might be more prevalent after Hayman and Cho sutures as they directly obliterate the anterior and posterior uterine walls. In order to prevent endometrial scarring, application of appropriate tension on uterine walls by suture material which is firm, monofilament and quickly absorbed is of paramount importance. Recently, there were reports of novel compression sutures which were removed in early postoperative period . Nevertheless, the sample size was small and further evaluation on long-term outcome is needed before clinical application. Our subsequent pregnancy rate was 29% which corresponds well with previous studies . Pregnancies from three months to ten years after uterine compression sutures have been reported . Similarly, in the present study, the interpregnancy interval was between five months to nine years. Comparing with control group with no compression sutures performed, we did not identify any differences in the rate of preterm birth and small-for-gestational-age fetus. Regarding mode of delivery, literatures reported successful vaginal deliveries in up to one-third of the cases . Interestingly, none of our women chose vaginal delivery. Taking into account of their response in the psychological interviews, their negative memories and fear of recurrence of PPH might explain their wish of having the deliveries conducted in a relatively more controlled manner. We observed significantly more atonic PPH (68.8%) and repeated compression sutures (12.5%) in our cohort. Similar findings were reported by Fuglsang who observed 15% of subsequent deliveries required repeated B-Lynch sutures or hysterectomies . In addition, significantly more omental or bowel adhesions were noted in one-third of our cohort. Hence, our results suggest a higher level of anticipation of PPH and surgical difficulty should be considered in subsequent deliveries. Previous studies reported severe complications of uterine rupture in next pregnancies [ – ]. In our cohort, we reported none but one woman with evidence of grooves over the uterus which was likely an ischemic defect suggestive of excessive pressure on the myometrium. The number of women with uterine grooves is likely to be underreported as these features could only be diagnosed during operation. Therefore, high vigilance of uterine dehiscence or rupture should be maintained in any women presenting with abdominal pain antenatally. Overall, with the small sample size, it is difficult to draw a conclusion about the effect of compression sutures on subsequent pregnancy outcomes. Remarkably, 72.1% of women and 51.5% of their partners were reluctant for future pregnancy. Nearly one-fourth of them and over half of their partners reported negative memories afterwards. Significantly, among those with long-term adverse impact after the delivery episodes, 73.3% had tokophobia. With our result, we suggest involving professional input in debriefing and providing psychological support sensitively to couple would be beneficial in early postnatal period. Conducting a longer period of research is required to assess the impact to their long-term quality of life.
Majority of women with history of uterine compression sutures had similar menstruation and pregnancy outcomes as compared to those who did not have sutures. In their future pregnancies, despite most pregnancies were carried till term uneventfully, a higher risk of women with atonic PPH, repeated compression sutures, omental and bowel adhesions were observed. Professional input may be beneficial to address couple’s traumatic memory and emotional recovery after delivery episodes. Large-scale study is needed to narrow this risk estimate.
Below is the link to the electronic supplementary material. Additional file 1: Figure S1. Study population Additional file 2: Figure S2. Asherman’s syndrome
|
Hot Melt Extrusion as an Effective Process in the Development of Mucoadhesive Tablets Containing | a6ddf614-ad36-4101-a9e5-9f97c380ad56 | 10054152 | Pharmacology[mh] | Periodontal disease and other oral infections are a significant global burden on oral health, with severe periodontitis responsible for losing many teeth in the adult population worldwide. The WHO Global Oral Health Status Report (2022) estimated that oral diseases affect nearly 3.5 billion people worldwide. At the same time, severe periodontal disease is estimated to affect approximately 19% of the world’s adult population, equivalent to more than 1 billion cases worldwide . Over the past 50 years, both systemic and local administration have been crucial methods for drug delivery to treat oral infections. The use of systemic administration in treating oral infections has yielded some positive results and has been widely used for antibiotic therapy in the treatment of periodontitis . However, systemic administration may result in issues including dysbacteriosis and inadequate biodistribution . Because of these apparent disadvantages of systemic administration, there is a great need for local drug delivery systems to improve the prevention and treatment of periodontitis and other oral infections. Local drug delivery systems that are placed directly on the oral mucosa can provide a sufficiently high concentration of an active substance for a reasonably long period of time. Other significant advantages of local drug delivery systems include avoidance of gastrointestinal problems and first-pass metabolism due to direct application to a specific site; higher efficacy and fewer side effects due to controlled drug release; and improved patient compliance due to reduced dosing frequency and easier oral application and the ability to quickly remove the drug in the event of irritability . Moreover, local drug delivery systems exert therapeutic effects mainly through the content of three types of active compounds, including antibacterial agents, modulators of inflammation, and alveolar bone and tissue repair agents for treating periodontitis. One of the medicinal raw materials for which much has been described and evidence of medical use exists is Scutellariae baicalensis radix (Baikal Skullcap Root). The healing effect of S. baicalensis root is due to the presence of bioactive compounds, mainly flavones such as baicalin and wogonoside and their aglycones, baicalein and wogonin . Thanks to the content of bioactive compounds, primarily baicalin, the plant material has anti-inflammatory properties by inhibiting the expression of proinflammatory mediators such as IL-1, IL-6, IL-8, and TNF in gingival tissues, antioxidant properties, and antibacterial properties against Streptococcus mutans , Fusobacterium nucleatum, Aggregatibacter actinomycetemcomitans , and Porphyromonas gingivalis for treating and preventing oral disease . In addition, it has been shown that baicalein can increase the expression of osteogenic markers in human periodontal ligament cells, which is valuable in treating periodontitis . Despite the many valuable health-promoting properties of S. baicalensis root, the solubility of baicalin is limited, which classifies it as IV BCS . The low solubility limits the application of compounds also in local drug delivery systems. Therefore, one of the biggest challenges for pharmaceutical researchers has been increasing the solubility of the insoluble compound with pharmacological potency. Various methods have been developed, including particle size reduction, solubilization, and solid dispersion, with the latter, produced by hot melt extrusion (HME) becoming increasingly desirable . Compared to traditional techniques, HME can offer numerous advantages, both economic benefits due to the shorter time to manufacture the final product and environmental benefits due to the elimination of solvents in the processing process . From a pharmaceutical process point of view, HME involves pumping polymeric materials with a rotating screw at temperatures above their glass transition temperature to achieve molecular-level mixing of active compounds and thermoplastic binders, polymers, or both. The components are changed by this molecular mixing into an amorphous product with a homogeneous shape and density, which improves the dissolution profile of the poorly water-soluble substance . As mentioned above, the low solubility of baicalin limits its application. Therefore, this work aimed to use HME technology to improve the physicochemical properties of baicalin as concluded in lyophilized S. baicalensis root extract. Nevertheless, the use of the extrudate itself in medicine is limited; hence, the optimization of the tableting process was done. Thus, the impact of the HME process on the rheological properties, such as tableting and compressibility properties, of the obtained extrudate-based blends was additionally assessed. Receiving appropriate baicalin release profiles as well as mucoadhesive functionality was indicated as being necessary for the development of this form of the drug. In the first stage, a lyophilized extract of Scutellariae baicalensis radix was obtained according to the procedure described previously . In an earlier study, the phytochemical and biological properties (including antioxidant and anti-inflammatory activity) of the obtained extract were confirmed. At the same time, the work aimed to use HME to improve the parameters for the release of active compounds from the obtained tablets and also to improve the tableting process itself. For a better understanding of all experimental work, all steps were collected in . As the first task, preparing three types of solid dispersions were possible using the hot melt extrusion technique. Importantly, in all cases, the torque measured during extrusion was similar (around 0.72 Nm), not causing any difficulty during processing. The process temperature (150 °C) did not decompose the active ingredient, i.e., baicalin (melting point 202–205 °C). This was also checked during the evaluation of the content of active compounds by the HPLC method (chromatogram of standards presented in ), which was validated according to ICH guidelines and whose validation parameters are collected in . The phytochemical profile remained at the same level as in the original lyophilized extract: baicalin—2.61 mg per 100 mg of extract; baicalein—323.40 µg per 100 mg of extract; and wogonin—40.30 µg per 100 mg of extract. With the preserved phytochemical profile of bioactive compounds, there was no need to re-examine the biological properties because it is the content of active compounds that determines those effects. shows the macroscopy pictures of HPMC-based extrudates. As seen, in the case of system-HPMC 75:25 w / w the inner structure appeared to be non-homogeneous, with a slightly rough surface and somehow “granular” inner structure with variable color. The internal structure changed as the amount of HPMC in the extrudate increased. So in the case of system HPMC 25:75 w / w outer structure appeared to be rather homogeneous and the surface relatively smooth. The obtained extrudates were characterized by their structure (XRPD) and possible intermolecular chemical bond formation (FTIR-ATR). The X-ray diffractograms of the lyophilized extract and its system with chitosan show a large broadening of the diffraction peaks, which at low intensity indicates their amorphous structure, which was described previously . HPMC is also amorphous in nature. So, hot melt extrudates are based on HPMC, which can be named an amorphous matrix former, transformed into an amorphous state, or molecularly dissolved in the carrier . It is shown that the relationship in the 75/25 system shows the lowest degree of order (the lowest intensity and visibility of reflections). The increase in the amount of HPMC in the relationship with chitosan results in a clearer structural response of the extrudate system (an increase in the intensity of reflections and their visibility)—indirectly, this indicates a better reaction of the extract with chitosan to obtain an amorphous system. However, the addition of HPMC is necessary for processing reasons. The addition of the extract affects the position of reflections in the obtained extrudates. From the analysis carried out for the systems, the averaged result based on the chitosan base gives a reduction in the interplanar distances. However, no linear relationship was noted in all analyzed cases . The obtained extrudates were characterized by their possibility to form intermolecular chemical bonds (FTIR-ATR) . Bands of S. baicalensis lyophilized extract at 3330 cm −1 , 1720 cm −1 , and 1660 cm −1 are characteristic for vibration of the O–H, –COOH, and C=O groups, while signals at 1600 cm −1 and 1580 cm −1 for the C=C vibration of the aromatic rings in the structure of flavones. The broad bands in the range 1200–900 cm −1 are characteristic of vibrations of C–O bonds of saccharides . For the HPMC spectrum, a wide band was observed at 3300 cm −1 , associated with the presence of -OH groups. While the complex band between 1200 and 950 cm −1 is related to numerous C–O vibrations, including glycosidic C-O-C, C-OH, C-OCH 3 , C-OCH 2 CH 2 OH . In the case of extrudates, it can be observed that the bands at 3300 and 1600 cm −1 changed, broadened, and decreased in intensity, which means intermolecular hydrogen bonds between the extract and carrier, which has also been observed for solid dispersions of pure baicalin . shows the release profiles of baicalin from ground hot melt extrudates based on HPMC in three different ratios. For comparison, the baicalin release from the lyophilized extract as well as the dissolution rate of pure baicalin are also shown. The dissolution rate of pure baicalin was very low; only 50% of pure baicalin was dissolved in 4 h due to its poor wettability and agglomeration. An increased dissolution rate of baicalin from the freeze-dried extract was observed, reaching 80% dissolved baicalin within 15 min, which is related to the change from crystalline to amorphous form. The HME process additionally improved the dissolution. Despite the slower solubility of baicalin, 80% over 90 min due to the presence of HPMC, the HME process improves wettability, reduces the size of baicalin dispersion, and prevents agglomeration of particles . Differences in baicalin release from HPMC extrudates depend on the amount of carrier in the system. Firstly, differences in dissolution rates were statistically significant among the three extrudates (in all cases, f 1 was below 20 and f 2 was below 50). Secondly, with the increase in the amount of HPMC, baicalin dissolves to a lesser extent due to the hydration of the outer layer of the system, which causes the formation of a gel layer on its surface. This reduces the amount of water that enters the system’s core, which can hinder the movement of the active compounds and cause them to dissolve slowly . Additionally, permeability coefficients using the PAMPA test were established. While the permeation test is not critical when talking about local application, it is intended to more extensively check the material’s properties after extrusion. The permeability coefficient for pure baicalin, calculated from equation no. 1 ( n = 6), was 0.02 ± 0.01 × 10 −6 cm/s, which is in line with previous research , and also confirms its low permeability, classifying baicalin as BCS IV . Due to the improved solubility associated with the amorphization of baicalin, the permeability of the compound also increased. Permeability coefficients for extrudates were 0.96 ± 0.02, 0.74 ± 0.02 and 0.58 ± 0.02 × 10 −6 cm/s, respectively, for extrudates 75:25, 50:50 and 25:75. The decrease in the permeation coefficient with the increase in the HPMC content in the system can also be explained by the formation of a gel layer, which makes it difficult for the active ingredients to reach the biological barrier. However, thanks to the amorphization of the system, multiple increases in the permeation of baicalin can be observed, which in turn is associated with an increase in its dissolution rate from extrudates. Nevertheless, penetration above 1 × 10 −6 cm/s was still not achieved, so it cannot be said that the system penetrates well. What is intended from the point of view of topical application within the oral cavity, a very well-constructed system has been achieved where baicalin appears at the application site in a higher dose due to the increase in release rate while not penetrating the systemic circulation, staying at the place of application. In the presented study, the antimicrobial activity of the prepared extrudates against microorganisms colonizing the oral cavity (e.g., S. mutans ) and bacteria whose presence in the oral cavity causes the development of infection (e.g., P. aeruginosa. S. aureus, E. aerogenes ) was investigated. The antimicrobial activity of binary systems was evaluated according to their inhibition zone diameter against six species of bacteria . presents the results of the impact of those tested on the ability to increase micro- organisms in the medium. Apart from the obvious fact that the lyophilized extract has antibacterial activity, which was described earlier, it is worth noting that chitosan has an equally important effect on the activity of the extrudates. So, the results revealed the highest antimicrobial activity for extrudate 25:75. In contrast, the most significant increase in activity following the combination of S. baicalensis radix extract with chitosan was observed against S. mutans, the most sensitive strain and, at the same time, one of the pathogens causing periodontitis. In liquid cultures, three different concentrations of extrudates were used. The results in shows that only a concentration of 100 mg/mL exhibited antimicrobial activity. In the next step, tablets containing all three types of extrudates were successfully prepared. A total of 6 formulations were prepared: three had extrudates (formulations F1, F3, and F5) and three contained the identical amounts of ingredients but in the form of powders, which were controls and comparative formulations (formulations F2, F4, and F6). HPMC was a carrier for hot melt extrusion in all formulations. Still, it also imparted mucoadhesive properties to the prepared systems. Firstly, tablets containing formulations F1–F6 were initially characterized in terms of tabletability, compressibility, and compactability . The tabletability of the tablets decreased in the following order: F1~F3 > F5 > F6 > F4 > F2 ( a,b). Such an order is related to the composition of the individual components. In general, it can be said that extrudate-based tablets showed better tabletability properties than those containing powders. However, a smaller amount of HPMC in the tablet increases its tabletability. According to the overall trend of the compressibility profile, the porosity level or the solid fraction value decreases as the pressure load applied to the powder samples grows ( b). The compactibility profiles for all six formulations are generally comparable, with little benefit for powder-based tablets. Finally, a powder’s compactability is defined as its ability to form coherent, strong tablets. Low-density tablets are obviously more porous because they have more pores, leading to poorer interparticle bonding. As a result, less power is needed to break down those tablets. The order of decreasing compactability appears to be as follows: F1 > F3 > F6 > F5 > F4 > F2 ( d,e). It can also be seen from c that tablet tensile strength decreases exponentially with increasing porosity, which fits the Ryshkewitch equation (equations no. 5–6) where T S0 is the extrapolated tensile strength at zero porosity and T S0 is often used to compare bond strength. The T S0 of F2, F4, and F6 was, respectively, 1.5, 2.4, and 3.2 Mpa, the bonding strength of powders was poor. The T S0 of extrudates were enhanced to 4.7 and 4.2 for F1 and F3, and obtained a surprisingly low value for F6–1.7 Mpa. Bond strength increased, illustrating that the HME process can significantly improve bond strength. This may be due to better uniformity and closer contact during melting and extrusion. It can also be the result of the transformation from the crystalline to the amorphous form of the active compounds . Based on the above parameters, the best tablet properties were obtained for formulations F1 (extrudates of system-HPMC 75:25) and F3 (extrudates of system-HPMC 50:50). In the next step, the dissolution rate of baicalin from the F1–F6 formulation was assessed . As described above, an increased dissolution rate of baicalin from the lyophilized extract and extrudates was observed, which is related to the change from crystalline to amorphous form . The dissolution profiles of baicalin from extrudates and formulations prepared from them (extrudates 75:25 and F1, extrudates 50:50 and F3, and extrudates 25:75 and F5) differing in the pressure used to prepare the tablets were compared. In each case, it was noticed that the dissolution rate decreased with the increase in compression pressure, but the differences were not statistically significant (in all cases, f 1 was below 20 and f 2 was above 50). Importantly, baicalin release from extrudate-based formulations was relatively fast, even faster than release from powder-based tablets. In the case of powder-based tablets, a slow and controlled release was observed ( d–f). This substantial difference can be explained by the swelling behavior of these extrudates as well as HPMC, calculated from equation no. 7 . While powder-based systems rapidly absorb water upon contact with the release medium, extrudate-based tablets remained almost intact. The HPMC powder swelled significantly, and the active compounds had to pass through the polymer network, and a more extensive layer of gel formed around the powder tablets, which made it difficult for baicalin to dissolve and enter the release medium . HME changes the behavior of HPMC, and the reprocessed carrier has less water absorption and sticky layer properties. In addition, changes in the release of baicalin can be observed depending on the amount of HPMC in the system. When the percentage of HPMC in the formulation increased, the baicalin release rate decreased simultaneously, both from extrudate- and powder-based tablets. Differences in the release of baicalin from extrudate- and powder-based tablets are also visible in the kinetics of its release. Mathematical models describing the release kinetics of baicalin from formulations F1–F6 are collected in . As indicated above, the release of baicalin from powder-based tablets (formulations F2, F4, and F6) is considerably slowed down and controlled, and the release of baicalin follows zero-order kinetics. It means that the release rate of baicalin is constant over a period of time. Such controlled release systems are indicated and developed in therapeutic drug delivery systems, but if mucosal application within the oral cavity is desired, complete release of the substance should occur within 2 h because a longer stay of the tablet stuck to the mucous membrane may be uncomfortable for the patient. In this regard, baicalin release is preferred from an extrudate-based tablet (formulations F1, F3, and F5). For these formulations, as the most probable, the Higuchi model was shown, which best describes the release from the matrix system and suggests that the baicalin was primarily released by diffusion and that its release was from a homogeneous flat matrix that did not degrade . Additionally, a good fit to Korsmeyer-Peppas with ‘n’ values in the range 0.45–0.89 indicated the release approximated the non-Fickian diffusion release mechanism . The relative complexity of the prepared formulations may indicate that the active compound release is controlled by more than one mechanism; a coupling of polymer erosion, swelling, and dissolution, which were all involved in the release process, which is consistent with the previous literature data . Finally, the mucoadhesive properties of formulations F1–F6 were evaluated by rheological measurements . The basis of the blends is a lyophilized extract with chitosan, to which HPMC was added as a carrier. The mucoadhesive properties of chitosan are widely known, and chitosan–mucin interact mainly electrostatically, supported by other types of interactions (e.g., hydrogen bonds and hydrophobic association) . In this case, to demonstrate the mucoadhesive effect, an appropriate pH is necessary (pH < 6), so the pH of the oral cavity is on the limit . However, HPMC is a non-ionic polymer; the medium’s pH had no effect on how well it stuck to the mucosa. Thus, in the case of the described blends, HPMC is the primary mucoadhesive agent. It has a lot of hydroxyl groups, which allow it to form intermolecular bonds (including hydrogen interactions) with the components of mucus . Formulations containing unprocessed HPMC, i.e., powder-based blends (formulations F2, F4, and F6) had more vital adhesion forces than their corresponding extrudate-based blends, possibly due to their elasticity, hydrogen bonding, molecular weight, and cross-linking. Internal forces are represented by viscosity, whereas the force needed to separate a polymer from a surface is known as adhesion force . Moreover, the adhesive force of all blends decreased with a decrease in the HPMC content, which aligns with previous outcomes . Tablets were additionally tested for their residence time to elaborate on their mucoadhesive behavior upon continuous contact with the medium-simulating saliva . All of the formulations that were tested attached to the tissue right away, swelled progressively when they came into contact with the acceptor medium, and showed no evidence of disintegration at any point during the test. Despite the continuous movement of the cylindrical probe, the contact time of tablets F4 and F6 with the mucosal surface was preserved within 240 min of the test. In contrast, formulations F2 and F5 separated from the tissue after 220 min, F3 after 200 min, and F1 after 180 min . This behavior of the tablets may be due to the higher viscosity and greater mucoadhesive strength of the unprocessed HPMC, as described above. 3.1. Plant Material Plant raw material, Scutellariae baicalensis radix , was purchased from NANGA (Zlotow, Poland), the country of origin: China (Lot No. 243042021). 3.2. Chemicals and Reagents Baicalin (≥95%, HPLC) was obtained from Sigma-Aldrich (Poznan, Poland). Excipients, such as chitosan with a degree of acetylation of 90% with a viscosity range of 500 mPas (marked as 90/500), was supplied from Heppe Medical Chitosan GmbH (Halle, Germany), (hydroxypropyl)methyl cellulose (HPMC) with an average Mn~90.000 (~15.000 cP), and magnesium stearate, were supplied by Sigma-Aldrich (Poznan, Poland). Microcrystalline cellulose (MCC) VIVAPUR 102 was supplied by JRS PHARMA (Rosenberg, Germany). Reagent for mucoadhesive tests: mucin from porcine stomach was obtained from Sigma-Aldrich (Poznan, Poland). HPLC grade acetonitrile and water were obtained from Merck. High-quality pure water and ultra-high-quality pure water were prepared using an Direct-Q 3 UV Merck Millipore purification system. 3.3. Preparation of Solid Dispersion Systems 3.3.1. Preparation of Extract System 5.0 g of the dried root of Scutellariae baicalensis radix was extracted four times with an ethanol–water mixture (8:2 v / v ) for 60 min at 70 °C on an ultrasound-assisted water bath. The obtained extracts were collected and concentrated on a vacuum evaporator at a temperature 50°C to a volume of 20.0 mL (BÜCHI Rotavapor R-210) obtaining at that time DER 1:4. Then the extract was frozen and lyophilized (CHRIST 1–4 LSC, Osterode am Harz, Germany). The temperature on the freeze dryer shelf was heated and ranged from +15 °C to +20 °C; the temperature inside the product was estimated −4 °C; and the condensation temperature was set at −48 °C. The freeze-drying was conducted at reduced pressure (1.030 mbar) for 48 h. So obtained lyophilized extract was combined with chitosan 90:500 in a weight ratio of 2:1 and named as ‘system’ for further tests . 3.3.2. Hot Melt Extrusion (HME) Extrusion was performed on a HAAKE MiniCTW micro-conical twin screw extruder (Thermo Scientific, Karlsruhe, Germany). The above-described system of lyophilized extract and chitosan in ratio 1:2 and carrier (HPMC) in three different ratios were mixed with a mortar and pestle and subsequently fed manually into the hopper of the extruder at barrel temperature of 150 °C and screw speed of 150 rpm. The extrudates were collected, ground softly manually with a pestle and mortar, passed through an 80 mesh sieve, and kept in a desiccator at room temperature for further analysis. 3.3.3. Extrudate Characterization Powder X-ray Diffraction (PXRD) The crystallographic structure of the samples was analyzed by X-ray diffraction (XRD, Panalytical Empyrean, Almelo, The Netherlands) equipment with the copper anode (CuKα—1.54 Å) in a Brag-Brentano reflection mode configuration with 45 kV and 40 mA parameters. The measurement parameters were set up for 3–60° with a 45 s per step 0.05° in all cases. Fourier Transform Infrared Spectroscopy with Attenuated Total Reflectance (FTIR-ATR) The FTIR-ATR spectra were measured between 400 cm −1 and 4000 cm −1 , with a resolution set to 1 cm −1 , with a Shimadzu IRTracer-100 spectrometer equipped with a QATR-10 single bounce—diamond extended range—and LabSolution IR software. 3.3.4. Determinations of Active Components Content The contents of the main active compounds (baicalin, baicalein, and wogonin) were determined by using the HPLC-Diode-Array Detection method described previously by Paczkowska-Walendowska et al. . Briefly, separations were performed on a Kinetex ® C18 column, 5 μm particle size, 100 mm × 2.1 mm (Phenomenex, Poland). The mobile phase was composed of phosphoric acid 0.1% (A) and acetonitrile (B), with a gradient elution: 0–20 min, 10–40% B; 20–22 min, 10% B. The detection was performed at a wavelength (λ max ) of 280 nm. The flow rate of the mobile phase was set at 1.0 mL/min, and the column temperature was set at 30 °C. Injection volume was 10 µL. The test was repeated three times. 3.3.5. In Vitro Release Studies An Agilent 708-DS apparatus was used for the dissolution studies. At 37 ± 0.5 °C, a typical paddle method was employed, with 50 rpm for stirring. Extrudates samples (~100 mg) were dissolved in 300 mL of an artificial saliva solution with the following ingredients: potassium chloride (1.20 g), sodium chloride (0.85 g), dipotassium hydrogen orthophosphate (0.35 g), magnesium chloride (0.05 g), calcium chloride (0.20 g), xylitol (20.0 g), and water up to 1L; the pH was adjusted to 6.8 by 1 M HCl. At certain intervals (15, 30, 60, 120, and 240 min), liquid samples were taken, and an equal volume of temperature-stabilized medium was substituted. A nylon membrane filter with a mesh size of 0.45 was used to filter the samples. The previously published HPLC method was used to ascertain the levels of baicalin in the filtered acceptor solutions. Sink conditions were preserved in the studies. The test was repeated for six samples of each substance. The release profiles were compared by means of the model proposed by Moore and Flanner, which is based on two-factor values, f 1 and f 2 . 3.3.6. Permeability Studies The permeability of an active compound (baicalin) enclosed in systems through artificial biological membranes was investigated by using the PAMPA™ (parallel artificial membrane permeability assay) gastrointestinal tract (GIT) assay (Pion Inc., Billerica, MA, USA) according to the protocol supplied with the kit. Extrudate- and powder-based systems (concentration 10 mg/mL) were dissolved in donor solutions (artificial saliva solution at pH 6.8). The acceptor plates were loaded with acceptor Prisma buffer at pH 7.4. The plates were put together and incubated under the following conditions: temperature set at 37 °C for 15 min with continuous stirring at 50 rpm. Each experiment was repeated at least three times. The amount of permeated baicalin was determined using the HPLC method described above. The test was repeated six samples of each substance. The apparent permeability coefficients (P app ) were calculated from the following equation: (1) P a p p = − l n ( 1 − C A C e q u i l i b r i u m ) S × ( 1 V D + 1 V A ) × t where V D is the donor volume, V A is the acceptor volume, C equilibrium is the equilibrium concentration C e q u i l i b r i u m = C D × V D + C A × V A V D + V A , C D is the donor concentration, C A is the acceptor concentration, S is the membrane area, and t is the incubation time (in seconds). 3.3.7. Microbiological Activity Assay Well Diffusion Method All microorganism strains were inoculated in Müeller-Hinton broth (pH 7.4) for approximately 16 h. The concentration of the suspensions was adjusted to 0.5 (optical density) by means of a spectrophotometer. Antimicrobial activity of the S. baicalensis radix lyophilized extract and extrudates were determined by the Agar well diffusion method against reference strains and clinical isolates of bacteria that colonize the oral cavity ( Escherichia coli, Pseudomonas aeruginosa, Streptococcus mutans, Staphylococcus aureus, Staphylococcus epidermidis, and Enterobacter aerogenes ). The 20 mL of sterilized Nutrient Agar was poured into sterile petri plates. Following solidification, 100 μL of standardized inoculate from each isolate was inoculated on Nutrient agar plates using sterilized spreaders. The wells were punched over the agar plates using a sterile gel puncher of 6 mm diameter. A measure of 100 μL of the lyophilized extract and extrudates was poured into separate wells. Samples were dissolved in 1% ( v / v ) dimethylsulphoxide (DMSO), which was used as a negative control. Plates were incubated at 37 °C for 24 h. Triplets of the experiment were maintained for each bacterial strain to ensure reliability. Following incubation, the diameter of the circular inhibitory zones formed around each well was measured in mm and recorded. Liquid Culture Method In the first stage of the research, strains of microorganisms were prepared. For this purpose, 0.1 g of bacterial lyophilisate was suspended in 10 mL of Müeller-Hinton liquid propagation medium. The samples were incubated at 37 °C for 18 h in order to activate and multiply the biomass. After incubation, the biomass was centrifuged from the substrate (14.000 rpm for 10 min). The supernatant was discarded, and the pellet was resuspended in 10 mL of 0.9% NaCl and centrifuged again. This procedure was performed three times. Then, the biomass was diluted in 0.9% NaCl, so that the concentration of microorganisms was 1.0 × 10 2 cfu/mL. At the same time, three solutions were prepared (the solvent was 0.9% NaCl) of the test samples at a concentration of 10, 50, and 100 mg/mL. Then, the dilutions prepared in this way were inoculated with the prepared suspension of microorganisms. The samples were mixed and incubated at 37 °C for 18 h. The number of microorganisms was analyzed before and after incubation using media intended for a given group of microorganisms. 3.4. Tableting Process A laboratory scale, single-punch tableting equipment called the NP-RD10A Tablet Press was used to compressed tablets that were flat-faced and 8 mm in diameter (Natoli, Saint Charles, MO, USA). Utilizing a variety of compaction forces between 1000 and 3000 N, the compaction characteristics of tablets were evaluated (corresponding to compression pressures in a range from 20 to 60 MPa). When the desired compaction force was reached, the pressure was let go. Two types of formulations were prepared, containing extrudates or powder systems in appropriate proportions of ingredients. lists the ingredients of the formulations. 3.4.1. Tablet Characterization Immediately following the tablets compacting, the newly created tablets were weighed. A procedure outlined in Ph.Eur. 9th was used to control the uniformity of the tablet mass. A manual vernier caliper was also used to measure the diameter and thickness of 20 tablets that were chosen at random. Standard deviations and mean values were computed following all measurements (SD). The tablet hardness was determined using the procedures outlined in Ph.Eur. 9th and was evaluated using the PTB-M manual tablet hardness testing device (Natoli, Saint Charles, MO, USA). Each hardness number is a mean with a standard deviation that represents the average of six measurements. Tensile strength ( σ ) values were calculated on the basis of the breaking force ( F ) values (N), where d is the diameter of the tablet (mm) and h is the thickness of the tablet (mm) . (2) σ = 2 F π d h Solid fraction ( SF ) was calculated by the equation, where W t is the weight of the tablet (mg), v is the tablet volume, and ρ true is the powder true density (g/cm 3 ). (3) F = W t ρ t r u e v The tablet porosity ( ε ) was calculated from the SF using the following equation: (4) ε = 1 − S F Compactibility of the powders were analysed with the Ryshkewitch equation: (5) ε = ε 0 × exp ( − b × P ) (6) T S = T 0 × exp ( − k × ε ) where the porosity of powder when p = 0; b is a constant that is inversely proportional to the yield strength of the materials; T S and T 0 are the tablet tensile strength and the limiting tablet tensile strength at zero porosity, respectively, and k is an empirical constant . 3.4.2. In Vitro Release Studies In vitro release studies were performed according to the methodology described in . The test was repeated 6 times for each formulation. The resulting active compound release profiles were fitted to the following mathematical models in order to study the release kinetics: : zero-order equation: F = k × t , first-order equation: n F = k × t , Higuchi equation: F = k t 1 / 2 , Korsmeyer-Peppas equation: F = k t n , where F —the fraction of released drug, k —the constant associated with the release, and t— the time. 3.4.3. Swelling Index Each tablet was individually weighted and placed in a 25 mL beaker that contained 10 mL of an artificial saliva solution at pH of 6.8 and at 37 ± 0.5 °C. Tablets were taken out, cleaned with filter paper, and reweighted at the preset intervals (15, 30, 60, 120, and 240 min). The swelling index was calculated by using the following formula: (7) S I = W 2 − W 1 W 1 where SI is the swelling index, W 1 is the initial weight of the tablet, W 2 is the weight of the tablet after the particular swelling time interval. Each experiment was performed in triplicate. 3.4.4. In Vitro Assessment of Mucin-Biopolymer Bioadhesive Bond Strength A viscometric method was used to quantify mucin-polymers’ bioadhesive bond strength. The assessment was carried out according to Hassan and Gallo’s procedure . Each experiment was performed in triplicate. 3.4.5. Determination of the Residence Time The residence time of tablets on regenerated cellulose membranes imitating porcine buccal mucosa was evaluated on an adjusted apparatus for the disintegration time test according to previous tests described by Paczkowska-Walendowska et al. . Briefly, the medium was an artificial saliva solution at pH 6.8 maintained at 37 ± 0.5 °C. Each tablet was brought into contact with foil by putting on a finger force for 5 s. The time necessary to detach the formulation from the foil simulating mucosal tissue was measured within 4h of the performed test. Studies were carried out in triplicate. 3.5. Statistical Analysis Software called Statistica 13.3 was used for the statistical analysis. The Shapiro-Wilk test was used to determine whether the results were normal. The ANOVA test, together with the post hoc Tukey’s range test for multiple comparisons, was used to examine the variances between the mean values. At p < 0.05, differences between groups were deemed significant. Plant raw material, Scutellariae baicalensis radix , was purchased from NANGA (Zlotow, Poland), the country of origin: China (Lot No. 243042021). Baicalin (≥95%, HPLC) was obtained from Sigma-Aldrich (Poznan, Poland). Excipients, such as chitosan with a degree of acetylation of 90% with a viscosity range of 500 mPas (marked as 90/500), was supplied from Heppe Medical Chitosan GmbH (Halle, Germany), (hydroxypropyl)methyl cellulose (HPMC) with an average Mn~90.000 (~15.000 cP), and magnesium stearate, were supplied by Sigma-Aldrich (Poznan, Poland). Microcrystalline cellulose (MCC) VIVAPUR 102 was supplied by JRS PHARMA (Rosenberg, Germany). Reagent for mucoadhesive tests: mucin from porcine stomach was obtained from Sigma-Aldrich (Poznan, Poland). HPLC grade acetonitrile and water were obtained from Merck. High-quality pure water and ultra-high-quality pure water were prepared using an Direct-Q 3 UV Merck Millipore purification system. 3.3.1. Preparation of Extract System 5.0 g of the dried root of Scutellariae baicalensis radix was extracted four times with an ethanol–water mixture (8:2 v / v ) for 60 min at 70 °C on an ultrasound-assisted water bath. The obtained extracts were collected and concentrated on a vacuum evaporator at a temperature 50°C to a volume of 20.0 mL (BÜCHI Rotavapor R-210) obtaining at that time DER 1:4. Then the extract was frozen and lyophilized (CHRIST 1–4 LSC, Osterode am Harz, Germany). The temperature on the freeze dryer shelf was heated and ranged from +15 °C to +20 °C; the temperature inside the product was estimated −4 °C; and the condensation temperature was set at −48 °C. The freeze-drying was conducted at reduced pressure (1.030 mbar) for 48 h. So obtained lyophilized extract was combined with chitosan 90:500 in a weight ratio of 2:1 and named as ‘system’ for further tests . 3.3.2. Hot Melt Extrusion (HME) Extrusion was performed on a HAAKE MiniCTW micro-conical twin screw extruder (Thermo Scientific, Karlsruhe, Germany). The above-described system of lyophilized extract and chitosan in ratio 1:2 and carrier (HPMC) in three different ratios were mixed with a mortar and pestle and subsequently fed manually into the hopper of the extruder at barrel temperature of 150 °C and screw speed of 150 rpm. The extrudates were collected, ground softly manually with a pestle and mortar, passed through an 80 mesh sieve, and kept in a desiccator at room temperature for further analysis. 3.3.3. Extrudate Characterization Powder X-ray Diffraction (PXRD) The crystallographic structure of the samples was analyzed by X-ray diffraction (XRD, Panalytical Empyrean, Almelo, The Netherlands) equipment with the copper anode (CuKα—1.54 Å) in a Brag-Brentano reflection mode configuration with 45 kV and 40 mA parameters. The measurement parameters were set up for 3–60° with a 45 s per step 0.05° in all cases. Fourier Transform Infrared Spectroscopy with Attenuated Total Reflectance (FTIR-ATR) The FTIR-ATR spectra were measured between 400 cm −1 and 4000 cm −1 , with a resolution set to 1 cm −1 , with a Shimadzu IRTracer-100 spectrometer equipped with a QATR-10 single bounce—diamond extended range—and LabSolution IR software. 3.3.4. Determinations of Active Components Content The contents of the main active compounds (baicalin, baicalein, and wogonin) were determined by using the HPLC-Diode-Array Detection method described previously by Paczkowska-Walendowska et al. . Briefly, separations were performed on a Kinetex ® C18 column, 5 μm particle size, 100 mm × 2.1 mm (Phenomenex, Poland). The mobile phase was composed of phosphoric acid 0.1% (A) and acetonitrile (B), with a gradient elution: 0–20 min, 10–40% B; 20–22 min, 10% B. The detection was performed at a wavelength (λ max ) of 280 nm. The flow rate of the mobile phase was set at 1.0 mL/min, and the column temperature was set at 30 °C. Injection volume was 10 µL. The test was repeated three times. 3.3.5. In Vitro Release Studies An Agilent 708-DS apparatus was used for the dissolution studies. At 37 ± 0.5 °C, a typical paddle method was employed, with 50 rpm for stirring. Extrudates samples (~100 mg) were dissolved in 300 mL of an artificial saliva solution with the following ingredients: potassium chloride (1.20 g), sodium chloride (0.85 g), dipotassium hydrogen orthophosphate (0.35 g), magnesium chloride (0.05 g), calcium chloride (0.20 g), xylitol (20.0 g), and water up to 1L; the pH was adjusted to 6.8 by 1 M HCl. At certain intervals (15, 30, 60, 120, and 240 min), liquid samples were taken, and an equal volume of temperature-stabilized medium was substituted. A nylon membrane filter with a mesh size of 0.45 was used to filter the samples. The previously published HPLC method was used to ascertain the levels of baicalin in the filtered acceptor solutions. Sink conditions were preserved in the studies. The test was repeated for six samples of each substance. The release profiles were compared by means of the model proposed by Moore and Flanner, which is based on two-factor values, f 1 and f 2 . 3.3.6. Permeability Studies The permeability of an active compound (baicalin) enclosed in systems through artificial biological membranes was investigated by using the PAMPA™ (parallel artificial membrane permeability assay) gastrointestinal tract (GIT) assay (Pion Inc., Billerica, MA, USA) according to the protocol supplied with the kit. Extrudate- and powder-based systems (concentration 10 mg/mL) were dissolved in donor solutions (artificial saliva solution at pH 6.8). The acceptor plates were loaded with acceptor Prisma buffer at pH 7.4. The plates were put together and incubated under the following conditions: temperature set at 37 °C for 15 min with continuous stirring at 50 rpm. Each experiment was repeated at least three times. The amount of permeated baicalin was determined using the HPLC method described above. The test was repeated six samples of each substance. The apparent permeability coefficients (P app ) were calculated from the following equation: (1) P a p p = − l n ( 1 − C A C e q u i l i b r i u m ) S × ( 1 V D + 1 V A ) × t where V D is the donor volume, V A is the acceptor volume, C equilibrium is the equilibrium concentration C e q u i l i b r i u m = C D × V D + C A × V A V D + V A , C D is the donor concentration, C A is the acceptor concentration, S is the membrane area, and t is the incubation time (in seconds). 3.3.7. Microbiological Activity Assay Well Diffusion Method All microorganism strains were inoculated in Müeller-Hinton broth (pH 7.4) for approximately 16 h. The concentration of the suspensions was adjusted to 0.5 (optical density) by means of a spectrophotometer. Antimicrobial activity of the S. baicalensis radix lyophilized extract and extrudates were determined by the Agar well diffusion method against reference strains and clinical isolates of bacteria that colonize the oral cavity ( Escherichia coli, Pseudomonas aeruginosa, Streptococcus mutans, Staphylococcus aureus, Staphylococcus epidermidis, and Enterobacter aerogenes ). The 20 mL of sterilized Nutrient Agar was poured into sterile petri plates. Following solidification, 100 μL of standardized inoculate from each isolate was inoculated on Nutrient agar plates using sterilized spreaders. The wells were punched over the agar plates using a sterile gel puncher of 6 mm diameter. A measure of 100 μL of the lyophilized extract and extrudates was poured into separate wells. Samples were dissolved in 1% ( v / v ) dimethylsulphoxide (DMSO), which was used as a negative control. Plates were incubated at 37 °C for 24 h. Triplets of the experiment were maintained for each bacterial strain to ensure reliability. Following incubation, the diameter of the circular inhibitory zones formed around each well was measured in mm and recorded. Liquid Culture Method In the first stage of the research, strains of microorganisms were prepared. For this purpose, 0.1 g of bacterial lyophilisate was suspended in 10 mL of Müeller-Hinton liquid propagation medium. The samples were incubated at 37 °C for 18 h in order to activate and multiply the biomass. After incubation, the biomass was centrifuged from the substrate (14.000 rpm for 10 min). The supernatant was discarded, and the pellet was resuspended in 10 mL of 0.9% NaCl and centrifuged again. This procedure was performed three times. Then, the biomass was diluted in 0.9% NaCl, so that the concentration of microorganisms was 1.0 × 10 2 cfu/mL. At the same time, three solutions were prepared (the solvent was 0.9% NaCl) of the test samples at a concentration of 10, 50, and 100 mg/mL. Then, the dilutions prepared in this way were inoculated with the prepared suspension of microorganisms. The samples were mixed and incubated at 37 °C for 18 h. The number of microorganisms was analyzed before and after incubation using media intended for a given group of microorganisms. 5.0 g of the dried root of Scutellariae baicalensis radix was extracted four times with an ethanol–water mixture (8:2 v / v ) for 60 min at 70 °C on an ultrasound-assisted water bath. The obtained extracts were collected and concentrated on a vacuum evaporator at a temperature 50°C to a volume of 20.0 mL (BÜCHI Rotavapor R-210) obtaining at that time DER 1:4. Then the extract was frozen and lyophilized (CHRIST 1–4 LSC, Osterode am Harz, Germany). The temperature on the freeze dryer shelf was heated and ranged from +15 °C to +20 °C; the temperature inside the product was estimated −4 °C; and the condensation temperature was set at −48 °C. The freeze-drying was conducted at reduced pressure (1.030 mbar) for 48 h. So obtained lyophilized extract was combined with chitosan 90:500 in a weight ratio of 2:1 and named as ‘system’ for further tests . Extrusion was performed on a HAAKE MiniCTW micro-conical twin screw extruder (Thermo Scientific, Karlsruhe, Germany). The above-described system of lyophilized extract and chitosan in ratio 1:2 and carrier (HPMC) in three different ratios were mixed with a mortar and pestle and subsequently fed manually into the hopper of the extruder at barrel temperature of 150 °C and screw speed of 150 rpm. The extrudates were collected, ground softly manually with a pestle and mortar, passed through an 80 mesh sieve, and kept in a desiccator at room temperature for further analysis. Powder X-ray Diffraction (PXRD) The crystallographic structure of the samples was analyzed by X-ray diffraction (XRD, Panalytical Empyrean, Almelo, The Netherlands) equipment with the copper anode (CuKα—1.54 Å) in a Brag-Brentano reflection mode configuration with 45 kV and 40 mA parameters. The measurement parameters were set up for 3–60° with a 45 s per step 0.05° in all cases. Fourier Transform Infrared Spectroscopy with Attenuated Total Reflectance (FTIR-ATR) The FTIR-ATR spectra were measured between 400 cm −1 and 4000 cm −1 , with a resolution set to 1 cm −1 , with a Shimadzu IRTracer-100 spectrometer equipped with a QATR-10 single bounce—diamond extended range—and LabSolution IR software. The crystallographic structure of the samples was analyzed by X-ray diffraction (XRD, Panalytical Empyrean, Almelo, The Netherlands) equipment with the copper anode (CuKα—1.54 Å) in a Brag-Brentano reflection mode configuration with 45 kV and 40 mA parameters. The measurement parameters were set up for 3–60° with a 45 s per step 0.05° in all cases. The FTIR-ATR spectra were measured between 400 cm −1 and 4000 cm −1 , with a resolution set to 1 cm −1 , with a Shimadzu IRTracer-100 spectrometer equipped with a QATR-10 single bounce—diamond extended range—and LabSolution IR software. The contents of the main active compounds (baicalin, baicalein, and wogonin) were determined by using the HPLC-Diode-Array Detection method described previously by Paczkowska-Walendowska et al. . Briefly, separations were performed on a Kinetex ® C18 column, 5 μm particle size, 100 mm × 2.1 mm (Phenomenex, Poland). The mobile phase was composed of phosphoric acid 0.1% (A) and acetonitrile (B), with a gradient elution: 0–20 min, 10–40% B; 20–22 min, 10% B. The detection was performed at a wavelength (λ max ) of 280 nm. The flow rate of the mobile phase was set at 1.0 mL/min, and the column temperature was set at 30 °C. Injection volume was 10 µL. The test was repeated three times. An Agilent 708-DS apparatus was used for the dissolution studies. At 37 ± 0.5 °C, a typical paddle method was employed, with 50 rpm for stirring. Extrudates samples (~100 mg) were dissolved in 300 mL of an artificial saliva solution with the following ingredients: potassium chloride (1.20 g), sodium chloride (0.85 g), dipotassium hydrogen orthophosphate (0.35 g), magnesium chloride (0.05 g), calcium chloride (0.20 g), xylitol (20.0 g), and water up to 1L; the pH was adjusted to 6.8 by 1 M HCl. At certain intervals (15, 30, 60, 120, and 240 min), liquid samples were taken, and an equal volume of temperature-stabilized medium was substituted. A nylon membrane filter with a mesh size of 0.45 was used to filter the samples. The previously published HPLC method was used to ascertain the levels of baicalin in the filtered acceptor solutions. Sink conditions were preserved in the studies. The test was repeated for six samples of each substance. The release profiles were compared by means of the model proposed by Moore and Flanner, which is based on two-factor values, f 1 and f 2 . The permeability of an active compound (baicalin) enclosed in systems through artificial biological membranes was investigated by using the PAMPA™ (parallel artificial membrane permeability assay) gastrointestinal tract (GIT) assay (Pion Inc., Billerica, MA, USA) according to the protocol supplied with the kit. Extrudate- and powder-based systems (concentration 10 mg/mL) were dissolved in donor solutions (artificial saliva solution at pH 6.8). The acceptor plates were loaded with acceptor Prisma buffer at pH 7.4. The plates were put together and incubated under the following conditions: temperature set at 37 °C for 15 min with continuous stirring at 50 rpm. Each experiment was repeated at least three times. The amount of permeated baicalin was determined using the HPLC method described above. The test was repeated six samples of each substance. The apparent permeability coefficients (P app ) were calculated from the following equation: (1) P a p p = − l n ( 1 − C A C e q u i l i b r i u m ) S × ( 1 V D + 1 V A ) × t where V D is the donor volume, V A is the acceptor volume, C equilibrium is the equilibrium concentration C e q u i l i b r i u m = C D × V D + C A × V A V D + V A , C D is the donor concentration, C A is the acceptor concentration, S is the membrane area, and t is the incubation time (in seconds). Well Diffusion Method All microorganism strains were inoculated in Müeller-Hinton broth (pH 7.4) for approximately 16 h. The concentration of the suspensions was adjusted to 0.5 (optical density) by means of a spectrophotometer. Antimicrobial activity of the S. baicalensis radix lyophilized extract and extrudates were determined by the Agar well diffusion method against reference strains and clinical isolates of bacteria that colonize the oral cavity ( Escherichia coli, Pseudomonas aeruginosa, Streptococcus mutans, Staphylococcus aureus, Staphylococcus epidermidis, and Enterobacter aerogenes ). The 20 mL of sterilized Nutrient Agar was poured into sterile petri plates. Following solidification, 100 μL of standardized inoculate from each isolate was inoculated on Nutrient agar plates using sterilized spreaders. The wells were punched over the agar plates using a sterile gel puncher of 6 mm diameter. A measure of 100 μL of the lyophilized extract and extrudates was poured into separate wells. Samples were dissolved in 1% ( v / v ) dimethylsulphoxide (DMSO), which was used as a negative control. Plates were incubated at 37 °C for 24 h. Triplets of the experiment were maintained for each bacterial strain to ensure reliability. Following incubation, the diameter of the circular inhibitory zones formed around each well was measured in mm and recorded. Liquid Culture Method In the first stage of the research, strains of microorganisms were prepared. For this purpose, 0.1 g of bacterial lyophilisate was suspended in 10 mL of Müeller-Hinton liquid propagation medium. The samples were incubated at 37 °C for 18 h in order to activate and multiply the biomass. After incubation, the biomass was centrifuged from the substrate (14.000 rpm for 10 min). The supernatant was discarded, and the pellet was resuspended in 10 mL of 0.9% NaCl and centrifuged again. This procedure was performed three times. Then, the biomass was diluted in 0.9% NaCl, so that the concentration of microorganisms was 1.0 × 10 2 cfu/mL. At the same time, three solutions were prepared (the solvent was 0.9% NaCl) of the test samples at a concentration of 10, 50, and 100 mg/mL. Then, the dilutions prepared in this way were inoculated with the prepared suspension of microorganisms. The samples were mixed and incubated at 37 °C for 18 h. The number of microorganisms was analyzed before and after incubation using media intended for a given group of microorganisms. All microorganism strains were inoculated in Müeller-Hinton broth (pH 7.4) for approximately 16 h. The concentration of the suspensions was adjusted to 0.5 (optical density) by means of a spectrophotometer. Antimicrobial activity of the S. baicalensis radix lyophilized extract and extrudates were determined by the Agar well diffusion method against reference strains and clinical isolates of bacteria that colonize the oral cavity ( Escherichia coli, Pseudomonas aeruginosa, Streptococcus mutans, Staphylococcus aureus, Staphylococcus epidermidis, and Enterobacter aerogenes ). The 20 mL of sterilized Nutrient Agar was poured into sterile petri plates. Following solidification, 100 μL of standardized inoculate from each isolate was inoculated on Nutrient agar plates using sterilized spreaders. The wells were punched over the agar plates using a sterile gel puncher of 6 mm diameter. A measure of 100 μL of the lyophilized extract and extrudates was poured into separate wells. Samples were dissolved in 1% ( v / v ) dimethylsulphoxide (DMSO), which was used as a negative control. Plates were incubated at 37 °C for 24 h. Triplets of the experiment were maintained for each bacterial strain to ensure reliability. Following incubation, the diameter of the circular inhibitory zones formed around each well was measured in mm and recorded. In the first stage of the research, strains of microorganisms were prepared. For this purpose, 0.1 g of bacterial lyophilisate was suspended in 10 mL of Müeller-Hinton liquid propagation medium. The samples were incubated at 37 °C for 18 h in order to activate and multiply the biomass. After incubation, the biomass was centrifuged from the substrate (14.000 rpm for 10 min). The supernatant was discarded, and the pellet was resuspended in 10 mL of 0.9% NaCl and centrifuged again. This procedure was performed three times. Then, the biomass was diluted in 0.9% NaCl, so that the concentration of microorganisms was 1.0 × 10 2 cfu/mL. At the same time, three solutions were prepared (the solvent was 0.9% NaCl) of the test samples at a concentration of 10, 50, and 100 mg/mL. Then, the dilutions prepared in this way were inoculated with the prepared suspension of microorganisms. The samples were mixed and incubated at 37 °C for 18 h. The number of microorganisms was analyzed before and after incubation using media intended for a given group of microorganisms. A laboratory scale, single-punch tableting equipment called the NP-RD10A Tablet Press was used to compressed tablets that were flat-faced and 8 mm in diameter (Natoli, Saint Charles, MO, USA). Utilizing a variety of compaction forces between 1000 and 3000 N, the compaction characteristics of tablets were evaluated (corresponding to compression pressures in a range from 20 to 60 MPa). When the desired compaction force was reached, the pressure was let go. Two types of formulations were prepared, containing extrudates or powder systems in appropriate proportions of ingredients. lists the ingredients of the formulations. 3.4.1. Tablet Characterization Immediately following the tablets compacting, the newly created tablets were weighed. A procedure outlined in Ph.Eur. 9th was used to control the uniformity of the tablet mass. A manual vernier caliper was also used to measure the diameter and thickness of 20 tablets that were chosen at random. Standard deviations and mean values were computed following all measurements (SD). The tablet hardness was determined using the procedures outlined in Ph.Eur. 9th and was evaluated using the PTB-M manual tablet hardness testing device (Natoli, Saint Charles, MO, USA). Each hardness number is a mean with a standard deviation that represents the average of six measurements. Tensile strength ( σ ) values were calculated on the basis of the breaking force ( F ) values (N), where d is the diameter of the tablet (mm) and h is the thickness of the tablet (mm) . (2) σ = 2 F π d h Solid fraction ( SF ) was calculated by the equation, where W t is the weight of the tablet (mg), v is the tablet volume, and ρ true is the powder true density (g/cm 3 ). (3) F = W t ρ t r u e v The tablet porosity ( ε ) was calculated from the SF using the following equation: (4) ε = 1 − S F Compactibility of the powders were analysed with the Ryshkewitch equation: (5) ε = ε 0 × exp ( − b × P ) (6) T S = T 0 × exp ( − k × ε ) where the porosity of powder when p = 0; b is a constant that is inversely proportional to the yield strength of the materials; T S and T 0 are the tablet tensile strength and the limiting tablet tensile strength at zero porosity, respectively, and k is an empirical constant . 3.4.2. In Vitro Release Studies In vitro release studies were performed according to the methodology described in . The test was repeated 6 times for each formulation. The resulting active compound release profiles were fitted to the following mathematical models in order to study the release kinetics: : zero-order equation: F = k × t , first-order equation: n F = k × t , Higuchi equation: F = k t 1 / 2 , Korsmeyer-Peppas equation: F = k t n , where F —the fraction of released drug, k —the constant associated with the release, and t— the time. 3.4.3. Swelling Index Each tablet was individually weighted and placed in a 25 mL beaker that contained 10 mL of an artificial saliva solution at pH of 6.8 and at 37 ± 0.5 °C. Tablets were taken out, cleaned with filter paper, and reweighted at the preset intervals (15, 30, 60, 120, and 240 min). The swelling index was calculated by using the following formula: (7) S I = W 2 − W 1 W 1 where SI is the swelling index, W 1 is the initial weight of the tablet, W 2 is the weight of the tablet after the particular swelling time interval. Each experiment was performed in triplicate. 3.4.4. In Vitro Assessment of Mucin-Biopolymer Bioadhesive Bond Strength A viscometric method was used to quantify mucin-polymers’ bioadhesive bond strength. The assessment was carried out according to Hassan and Gallo’s procedure . Each experiment was performed in triplicate. 3.4.5. Determination of the Residence Time The residence time of tablets on regenerated cellulose membranes imitating porcine buccal mucosa was evaluated on an adjusted apparatus for the disintegration time test according to previous tests described by Paczkowska-Walendowska et al. . Briefly, the medium was an artificial saliva solution at pH 6.8 maintained at 37 ± 0.5 °C. Each tablet was brought into contact with foil by putting on a finger force for 5 s. The time necessary to detach the formulation from the foil simulating mucosal tissue was measured within 4h of the performed test. Studies were carried out in triplicate. Immediately following the tablets compacting, the newly created tablets were weighed. A procedure outlined in Ph.Eur. 9th was used to control the uniformity of the tablet mass. A manual vernier caliper was also used to measure the diameter and thickness of 20 tablets that were chosen at random. Standard deviations and mean values were computed following all measurements (SD). The tablet hardness was determined using the procedures outlined in Ph.Eur. 9th and was evaluated using the PTB-M manual tablet hardness testing device (Natoli, Saint Charles, MO, USA). Each hardness number is a mean with a standard deviation that represents the average of six measurements. Tensile strength ( σ ) values were calculated on the basis of the breaking force ( F ) values (N), where d is the diameter of the tablet (mm) and h is the thickness of the tablet (mm) . (2) σ = 2 F π d h Solid fraction ( SF ) was calculated by the equation, where W t is the weight of the tablet (mg), v is the tablet volume, and ρ true is the powder true density (g/cm 3 ). (3) F = W t ρ t r u e v The tablet porosity ( ε ) was calculated from the SF using the following equation: (4) ε = 1 − S F Compactibility of the powders were analysed with the Ryshkewitch equation: (5) ε = ε 0 × exp ( − b × P ) (6) T S = T 0 × exp ( − k × ε ) where the porosity of powder when p = 0; b is a constant that is inversely proportional to the yield strength of the materials; T S and T 0 are the tablet tensile strength and the limiting tablet tensile strength at zero porosity, respectively, and k is an empirical constant . In vitro release studies were performed according to the methodology described in . The test was repeated 6 times for each formulation. The resulting active compound release profiles were fitted to the following mathematical models in order to study the release kinetics: : zero-order equation: F = k × t , first-order equation: n F = k × t , Higuchi equation: F = k t 1 / 2 , Korsmeyer-Peppas equation: F = k t n , where F —the fraction of released drug, k —the constant associated with the release, and t— the time. Each tablet was individually weighted and placed in a 25 mL beaker that contained 10 mL of an artificial saliva solution at pH of 6.8 and at 37 ± 0.5 °C. Tablets were taken out, cleaned with filter paper, and reweighted at the preset intervals (15, 30, 60, 120, and 240 min). The swelling index was calculated by using the following formula: (7) S I = W 2 − W 1 W 1 where SI is the swelling index, W 1 is the initial weight of the tablet, W 2 is the weight of the tablet after the particular swelling time interval. Each experiment was performed in triplicate. A viscometric method was used to quantify mucin-polymers’ bioadhesive bond strength. The assessment was carried out according to Hassan and Gallo’s procedure . Each experiment was performed in triplicate. The residence time of tablets on regenerated cellulose membranes imitating porcine buccal mucosa was evaluated on an adjusted apparatus for the disintegration time test according to previous tests described by Paczkowska-Walendowska et al. . Briefly, the medium was an artificial saliva solution at pH 6.8 maintained at 37 ± 0.5 °C. Each tablet was brought into contact with foil by putting on a finger force for 5 s. The time necessary to detach the formulation from the foil simulating mucosal tissue was measured within 4h of the performed test. Studies were carried out in triplicate. Software called Statistica 13.3 was used for the statistical analysis. The Shapiro-Wilk test was used to determine whether the results were normal. The ANOVA test, together with the post hoc Tukey’s range test for multiple comparisons, was used to examine the variances between the mean values. At p < 0.05, differences between groups were deemed significant. Extrudates containing Scutellariae baicalensis radix extract can be prepared at temperatures of about 150 °C, which does not decompose the active compounds. The proposed ground hot-melt extrudates based on HPMC show an interesting potential for improving the solubility of the poorly water-soluble active substance—baicalin. So, hot-melt extrusion is a good technique to improve the physicochemical properties of baicalin. Further, in order to obtain a suitable pharmaceutical form, the production process of mucoadhesive tablets containing extrudates was optimized. The prepared extrudates, differing in HPMC content, showed different tabletability, compressibility, and compactibility properties. As expected, the different content of the carrier influenced the release profile of baicalin from the tablets and the mucoadhesive properties. Higher HPMC content resulted in prolonged release of the substance, resulting from the diffusion of the substance through the polymer network. At the same time, the same carrier ensured that the tablets were kept in the affected area for a sufficiently long time. Importantly, the process did not reduce the biological, including microbiological, activity of the obtained extrudates. Considering the complex matrix, both the tabletability/compactibility properties of the blends and the degree of release of the active substance, as well as mucoadhesive properties that give functionality to the developed tablets, should be considered. The best tabletability properties, a valuable baicalin release profile while maintaining sufficient mucoadhesive properties to condition the tablet’s retention in the application site and the effectiveness of therapy, are provided by the F3 formulation, which contains the extrudate with lyophilized extract-HPMC 50:50 w / w . |
MALDI MSI Separation of Same Donor’s Fingermarks Based on Time of Deposition—A Proof-of-Concept Study | 56085531-4e7f-49e5-938e-f3dd0ebfdfbb | 10054356 | Forensic Medicine[mh] | Despite the advent of DNA profiling, fingerprints remain a primary form of biometric identification informing both investigations and judicial debates. A suspect identification is typically made by comparing a crime scene mark to a fingerprint record held in national databases (provided that the suspect has had a previous arrest) or to a fingerprint taken from a person of interest. Whilst three levels of details are available to fingerprint experts to compare a crime scene mark to a fingerprint, the composition and distribution of local characteristics of the ridge pattern called minutiae (level 2 details) are particularly important for identification, after having ascertained that the less specific level 1 details are present (e.g., arch, loop, delta etc.). Whilst generally successful, fingerprinting becomes difficult for smudged, faint, and very partial marks. Additionally, overlapping fingermarks pose a further challenge given by the difficulty to distinguish two or more ridge patterns to assign these impressions to the individuals that have generated them. The ability to separate overlapping fingermarks would be very important for both eliminating irrelevant marks and distinguishing, for example, between the victim’s and the offender’s fingermarks. In the last 15 years, the analytical community has offered spectroscopic imaging [ , , ] and mass spectrometric imaging solutions to this problem (in combination with various chemometric approaches) [ , , , , , , ]. All the approaches proposed are based on the principle that, as hundreds to thousands of molecules can be detected in fingermarks and as this composition varies all the time, it is possible to pick out those compounds that are uniquely present in each of the marks analysed and can be selected to reconstruct the fingermark ridge pattern. In terms of the mass spectrometric approach, a few techniques have been reported to successfully separate fingermark impressions, namely Desorption Electrospray Ionisation Mass Spectrometry imaging (DESI MSI) , Matrix-Sssisted Laser Desorption Ionisation Mass Spectrometry Imaging (MALDI MSI) [ , , ], (nanoparticle-assisted) Laser Desorption Ionisation and time-of-flight Secondary Ion Mass Spectrometry (ToF-SIMS) . Whilst these techniques have been used to separate ridge patterns from two or more individuals, to the authors’ knowledge, no study has been reported to date attempting separation of fingermarks left on a surface by the same individual. The only studies that are relevant to this subject are from Lauzon et al. and Gorka et al. . In their report , the former group suggests the possibility to use the chemical profile of the fingermark composition to connect fingermarks of the same individual left in different areas of a crime scene. For example, should a smudged mark be unidentifiable in one area of the crime scene (for example the kitchen), it could be assigned to a certain individual, should their identifiable mark be found there or in a different area (for example the living room). This link would provide the opportunity to reconstruct the movements of the offender during the crime based on the principle that the molecular profiles of the smudged and identifiable marks are the same. This assumption would be much easier to accept for marks that have been deposited at the same time. However, despite the long-standing common understanding and literature-supported knowledge that the fingermark composition varies according to metabolism, pharmacological and pathological states, stress, physical activity, diet, etc., Becue’s group has been challenging the fingerprint community with the idea that the intra-donor fingermark chemical composition remains fairly constant over time; as a result, it is potentially distinctive of an individual (some sort of biological passport) or permits categorisation of a group of individuals. In their recent MALDI MS-based work, Gorka et al. analysed natural fingerprints from 13 individuals over 12 months, concluding that overall “the composition of the fingermarks provided by the 13 donors is quite consistent over the months and the year”. Specifically, they report that (i) the compositions at different times of the year (for example January vs. July) are closely linked in the heatmaps and (ii) the compositional consistency observed over the year is also reproducible between the thirteen donors given that percentage of the m / z features retained between each selection is similar for every participant to the study. Whilst the heatmaps are sound, donors’ habits, medications, diet, etc. at the time of each fingermark deposition does not appear to have been recorded. Additionally, donors were only asked to not wash their hands 45 min prior to each deposition and thus reducing the chance of picking up different contaminants. Importantly, whilst the composition may be constant, the abundance of those components may be different, and the resulting ion intensity should be considered a discriminant of the intra-donor chemical profile. Finally, due to lower mass range chosen (100–2000 Da) and the fact that only some species may ionise in positive mode or are not masked by the presence of the matrix selected, the vast majority of the anti-microbial peptides and small proteins as well as amino acids are excluded from detection. Therefore, whilst certainly a very interesting paper, these uncertainties still justify the investigation into the separation of overlapping fingermarks deposited by the same individual at different times by MALDI MSI. The possibility to separate these types of marks is extremely relevant in a forensic context as it may inform on the legitimate access or lack thereof of the suspect to the scene. If same-day fingermark compositions cannot be distinguished, their separation must mean they were deposited at different times, whilst linking the suspect to the scene (through the provision of biometric identification). There remains the issue of dating a crime fingermark with some notable efforts from the mass spectrometry community [ , , , ]. The issue of establishing the age of a fingermark is not extensively addressed here as it needs, in the authors’ opinion, a much larger, concerted, and systematic approach to be appropriately tackled. However, a proof-of-concept verification of the hypothesis that it is possible to separate the same person’s fingermarks deposited at different times will both contribute to add complementary knowledge to that provided by Gorka et al. and to the global fingermark-dating problem. Results from the present study have shown it is possible to separate overlapping fingermarks from the same donor in most cases, although it becomes more difficult to observe minutiae for some marks older than 15 days (often due to sub-optimal contact pressure resulting in ridge merging). As only one donor was used, capabilities and limitations will need to be verified with a larger cohort of donors.
The separation of overlapping fingermarks and fingermark age determination by mass spectrometry-based approaches are topics that have been previously investigated and reported in the scientific literature [ , , , , , , ]. However, to date, no study has been published on the separation of overlapping fingermarks from the same donor and deposited at different times . Preliminarily, groomed fingermarks were analysed to address the hypothesis that the molecular composition of fingermarks from the same donor, but of different ages, is different enough to permit separation of the ridge pattern. Partially overlapping groomed fingermarks of different ages were therefore generated by the same donor in a total of eight pairs, as shown in . Only a small section of the mark was imaged to reduce the run time and ensure all the intended analyses could be completed in the time available for this study. shows representative m / z ions responsible for the separation of the marks. As it can be observed, it was generally possible to achieve separation of the marks albeit with some exceptions. These exceptions were mainly due to occasional lack of clarity of the ridge detail, likely resulting from a combination of sub-optimal matrix application and occasional sub-optimal fingermark deposition leading to ridge merging due to excessive contact pressure. The latter circumstance was particularly deleterious for pairs 5 and 6, which did not generate any useful images and were not reported in . reports the optical images of the groomed set showing instances of sub-optimal fingermark deposition. In other circumstances, the image of the younger fingermark could be separated/recovered but not that of the older mark (pair 1- 1 d vs. 1 h, pair 2- 3 d vs. 1 h, pair 3- 8 d vs. 3 h). However, this circumstance does not mean fingermarks as old as 1 day and older cannot be imaged; in fact, pairs 4 and 8 have yielded images of 14- and 46-day-old marks. In one case only, it was the younger (37 d) fingermark image that could not be separated/recovered (pair 8- 46 d vs. 37 d). It was reasonable to hypothesise that for groomed marks, endogenous species would be the main separating ions, given that marks were generated by rubbing the fingertips on the forehead and cheeks prior to subsequent deposition of the marks. However, it is important to note that for the scope of this paper, which is the attainment of biometric information by separating two impressions, knowing the identity of the ions that separate the two fingermarks is not essential. Whichever is the ion fit for this purpose, in any given instance, can be used without further investigating its identity. In any case, we have attempted a putative identification of the separating ions. It is likely the ions at m / z 340.394 and 368.426 are exogenous quaternary ammonium ions (C 23 H 50 N + , mass accuracy −2.3 ppm and C 25 H 54 N + , mass accuracy −3.8 ppm, respectively). While the ion m / z 327.337 (separating the younger mark of pair 3) could not be putatively identified, the ion at m / z 327.382 (partially separating the younger mark of pair 7) may be the C-13 isotope of the didecyldimethylammonium ion discussed later in this paper. As for the other ions separating the fingermarks, a quick lipid map search ( https://www.lipidmaps.org , accessed on 17 March 2023) using a mass tolerance of +/−0.001 Th yielded a putative identification only for the ion at m / z 387.099 as a potassiated phospholipid derivative, lysophosphatidic acid LPA 12:3 of formula C 15 H 25 O 7 PK (mass accuracy −5.7 ppm), reported as involved to train the skin to repair and strengthen the epidermis . To investigate applicability in real case scenarios, these analyses were repeated using natural marks, which are impressions made without grooming or any prior preparation of the fingertip. However, the marks were aged for slightly different duration with respect to the groomed marks ( ), generating another set of eight pairs of partially overlapping fingermarks ( ), though employing the same donor and environmental aging conditions. The slight difference in the age of the marks was dictated by instrumental availability. This natural mark set was then analysed by MALDI MSI upon optimisation of the matrix deposition conditions that are reported in the Methods . In this case too, shows representative m / z ions responsible for the separation of the marks. A higher image resolution and clarity of the fingermark ridge detail was achieved in this instance. However, it was clear that for natural marks too, in some cases, excessive pressure had occasionally been applied, or too little chemical content was deposited (pair 1 younger mark), resulting in marks that were not recovered or that do not show clear ridge detail. reports the optical images of the natural sets showing instances of sub-optimal fingermark deposition. Generally, the separation of the ridge pattern was successful, and the number of older marks that could not be separated was lower than the groomed set. However, some additional observations need to be made. Older fingermark ridge patterns could not be separated for pair 4 (14 d vs. 3 h) and pair 5 (46 d vs. 3 h). However, as in the case of groomed pairs, the ridge detail for the marks of 26, 37, and 46 days of age could be separated/recovered in other instances (pairs 7 and 8). In the case of pair 1 (1 d vs. 3 h), it was the younger fingermark’s ridge pattern (3 h) that could not be recovered. However, 3 hr old fingermarks were successfully imaged for the other pairs in this set of impressions (pairs 2–5). For pair 6, the squared frame highlights an area of the separated older mark, which possibly also exhibits some level of overlapping with the younger mark. Pairs 7 and 8 are interesting in that the squared frames highlight the fact that “separating ions” are not necessarily molecules unique to one fingermark, but they can be present in both, though in very different abundances such that separation can still occur. For pairs 7 and 8, the marks at 26 and 37 days are still visible, and whilst separated from the 46-day-old mark, the 26-day-old mark exhibits an even better clarity of the ridge detail. As for the groomed fingermarks ( ), the ions m / z 340.399 and 368.429 could be putatively assigned the formulas of C 23 H 50 N + (mass accuracy −2.35 ppm) and C 25 H 54 N + (mass accuracy −3.8 ppm), respectively. The ion at m / z 326.379 is also a contaminant and has previously been detected and identified in fingermarks as the didecyldimethylammonium ion . The ions at nominal m / z 523 were suspected to be the external contaminant previously detected of formula C 36 H 76 N + . However, the relative error is too high to permit this assignment (>20 ppm). A lipid maps search permitted the ion at m / z 381.247 to be putatively identified as a sterol lipid (M+H-H 2 O]+) of formula C 22 H 38 O 4 S (mass accuracy −2.8 ppm). The ion at m / z 415.15 could be assigned to another sodiated lysophosphatidic acid derivative (LPA 14:3;O) of formula C 17 H 29 O 8 PNa; however, the relative error of −12.5 ppm casts stronger doubts on this ion’s identity. The lipid maps search did not allow any other univocal putative identification for any of the remaining separating ions shown in , and MS/MS experiments are necessary to establish their identity. Regions of interest were drawn from the natural set of marks from the non-overlapping areas; spectral variability reflecting variable molecular composition could be observed in the mass range investigated even by manual inspection ( ), either in terms of absence/presence of signals or in terms of the relative intensity. As expected, the spectral profiles change dramatically with the age of the mark. Visually, the spectral profiles of marks of 3 h, 3 days, and 8 days of age differ significant from those with ages of 15 days, 26 days, and 46 days. In particular, the lower molecular range exhibits signals of higher intensity, as highlighted by the zoomed in spectra (for example for the 3h and 3 days vs. 8 days), possibly reflecting the breakdown/degradation of complex lipids with time. Though more donors and replicates would strengthen the reliability of this observation in a follow up study, this spectral variability explains the fingermark image separation achieved. The zoomed in spectra for the marks at 15, 26, and 46 days have a different (higher relative intensity) y scale compared to the other marks to better highlight any differences between these older marks; the ions at nominal m / z 304, 326, and 332 decrease in intensity as the age increases. The ions at m / z 304 and 332 are exogenous species and have been putatively identified as dodecylbenzyldimethylammonium (nominal m / z 304) with two extra CH 2 units, respectively, also previously detected in fingermarks . As reported earlier in this paper, the ion at nominal m/z 326 is also a contaminant and has previously been detected and/or identified in fingermarks as didecyldimethylammonium ion . These ions are common species found in toiletries (e.g., hair gels) and antibacterial products, such as hand sanitizers. They are observed for all marks up to 46 days of age. Their constant presence in all the marks examined indicate these are persistent substances and can be used to obtain MALDI molecular images of fingermarks recovered at crime scenes that are not accessed immediately (up to 46 days of age in this study) due to their high ionisation efficiency. The mass spectral profiles of the 26- and 46-day-old marks appear to exhibit a higher ion population in the higher mass range ( m / z 690–900) compared to the 15-day-old mark. Again, this is possibly reflecting the breakdown/degradation of high molecular weight compounds (beyond the mass range measured in this study) with time. Based on the findings reported in and , another experiment was conducted whereby the same donor deposited non-overlapping natural fingermarks, which were aged for 0, 7, and 14 days, matrix-spray-coated under the same conditions as for the overlapping fingermarks and subjected to manual firing of the laser in discrete locations (MALDI MS Profiling (MALDI MSP)). The 7- and 14-days-aging points for the chemical profiling were selected based on the more dramatic difference in spectral profiles observed for the overlapping fingermarks of 8 and 15 days of age ( ). Mass spectra were subsequently processed for submission to multivariate statistical analysis. Principle component analysis (PCA) shows clear separation between fresh marks (0 days) and marks of older age points (7 and 14 days) ( ). In contrast, group overlap was observed between the 7- and 14-days-age points indicating that, overall, the compositional change of marks between 7 and 14 days old is not enough to permit differentiation between 1-week- and 2-weeks-old marks. Importantly, however, the inability under these experimental conditions to differentiate between 7- and 14-day-old marks does not mean that unique molecules could not be found, allowing a separation of the ridge pattern for these two age points. A subsequent linear discriminant analysis was performed using OPLS-DA with a seven-fold cross validation on the marks of 0 and 14 days of age. The two 0-day old outliers observed in the PCA score plot ( ) were excluded from data before performing the analysis. The OPLS-DA model could describe all included data (R 2 X(cum) = 0.829), showing clear separation of the two age groups (R 2 Y(cum) = 0.833) as well as robustness in the ability to classify and distinguish fresh from 14-days-old marks (Q 2 (cum) = 0.785) ( A). An S-plot based on the OPLS-DA model was used to reveal the most discriminant ions for the two age groups ( B). The most discriminating ions are m / z 304.301, 550.629, 332.332, and 522.598 with VIP scores of 6.3, 5.6, 3.9, and 3.7, respectively. Interestingly, despite the ion at m / z 326.379 displaying a high VIP score (4.5), suggesting a high contribution to the separation of the two groups of marks in the OPLS-DA model, the S-plot indicates instead a low reliability in the model, i.e., high risk of spurious correlation. Similarly to the ions at nominal m / z 304, 326 and 332, the ion at nominal m / z 551 is also an exogenous species and is putatively identified as the dimethyloctadecylammonium ion from previous published work [ , , ]. In addition, the discriminating ion at m / z 499.566 may also putatively be identified as ditallowdimethylammonium ion, likely resulting from the use of personal and household products, as reported by Manier et al. . This time the ion at nominal m / z 523 could be putatively identified as another ditallowdimethylammonium ion (C 36 H 76 N + , mass accuracy −1.9 ppm) belonging to the same family as the ion at m / z 494.566. Therefore, within this profiling experiment and based on the statistical analysis, the most age-discriminant ions are exogenous contaminants. However, as shown in , endogenous compounds may also contribute to the separation between overlapping fingermarks. In conclusion, this study indicates it is possible for this donor to separate overlapping fingermarks deposited at different times as a result of a different enough molecular composition. This result is supported especially by the use of natural marks (not artificially enriched). However, a multi-donor study is required to extend this observation to the general population. The output from this study is not in contrast with the work published by Gorka et al. ; intra-donor variability may still, as they conclude from their study, be very low. However, to separate overlapping fingermarks, even just one molecule per mark out of the thousands detected is needed to be unique to obtain a separate image of the two impressions. In the future, this capability will be very useful when supported by a method that definitively established the age of a fingermark to establish (i) one person has deposited both overlapping marks, (ii) the identity of the individual (through the separated ridge detail obtained), and (iii) the time at which the scene was accessed on both occasions (in the case of two overlapping fingermarks) to establish legitimate access.
3.1. Materials Trifluoroacetic acid (TFA), α-cyano-4-hydroxycinnamic acid (α-CHCA) was purchased from Sigma Aldrich (Poole, UK). Acetonitrile (ACN), acetone, and methanol were purchased from Fisher Scientific (Loughborough, UK). The double-sided conductive carbon tape was obtained from TAAB (Aldermaston, UK). TLC sheets were purchased from (Merck, UK). 3.2. Methods 3.2.1. Instrument and Instrumental Conditions All MALDI MS spectrometric analyses were carried out using the Waters MALDI QTOF Synapt G2 HDMS instrument (Waters Corporation, Manchester, UK). Data acquisition was performed within the m / z range of 100–1000 in positive sensitivity mode with a scan time of 1 s after calibrating the instrument with a saturated solution of red phosphorus in ACN. The Nd:YAG laser repetition rate was set to 1 kHz and laser power to 250 a.u. for all analyses. MALDI MS Images were acquired at a spatial resolution of 100 µm × 100 µm. For chemical analysis, data were acquired from three to four (random) areas for each fingermark using a circular shutting pattern consisting of approximately 100 shots. 3.2.2. Data Processing of MALDI MS Data MALDI MS fingermark images of the 4500 most intense ion signals were generated using the HDI software (v. 1.6, Waters Corp., UK) using a mass window of 0.02 Da. Visual inspection of those images led to the selection of the m / z ions that produced the best separation of the ridge pattern. MS spectra from each of the fingermark pairs were obtained through the selection of regions of interest in the outermost left and right sides of the ion images, respectively, where the fingermarks do not overlap. 3.2.3. Preparation of Fingermarks Samples for MALDI MSI TLC sheets were preliminarily prepared, removing the silica by soaking them in methanol and sonicating for 15 min. Upon wiping the residual silica with acetone, they were cut into glass slide sizes and used as the fingermark deposition surfaces. On eight of these resulting separate aluminium sheets, a groomed fingermark (deposited without washing hands at any particular time before the deposition but after rubbing the fingertip on forehead and cheeks) was deposited by the same donor and allowed to age for a definite time at ambient office conditions. Subsequently, the same donor deposited a second groomed fingermark partially overlapping the first, and the two impressions were allowed to age further for a definite time period. Specifically, eight pairs of different age overlapping fingermarks were deposited, as summarised in , where sets 1–5 intended to investigate the separation between old and fresh fingermarks and sets 6–8 the separation between old fingermarks (but of different age). An additional set of eight pairs of overlapping natural fingermarks (deposited without prior preparation of the fingertip, i.e., without washing hands or touching skin at any particular time before the deposition) were also prepared and aged in the same way. Matrix deposition was optimised, as described in , and the aluminum sheets were then mounted onto the Synapt MALDI target plate using double-sided conductive carbon tape. All samples were imaged by MALDI MS immediately after deposition of the matrix. The time delay between the deposition of the youngest fingermark and the start of MALDI MS analysis was 20–30 min (this time delay is included in the indicated ages of the fingermarks). 3.2.4. Preparation of Fingermark Samples for Chemical Analysis On seven separate aluminium slides, natural smudged fingermarks were deposited by the same person. Three of the fingermarks were allowed to age for seven days, while the remaining four were aged for fourteen days, all at ambient office conditions. Another fresh (25–107 min) smudged fingermark (natural) was then deposited by the person on the same aluminium sheet (one per slide for a total of seven) but without overlapping the first fingermark. The matrix was then sprayed, as described in , and the aluminium slide was mounted onto the MALDI sample plate using double-sided conductive carbon tape. The samples were analysed by MALDI MS immediately after deposition of the matrix. Three to four technical replicates were measured for each fingermark. 3.2.5. Matrix Application α-CHCA matrix was prepared at a concentration of 5 mg/mL in 70:30 ACN:0.5% TFA aq . The matrix was sprayed using the HTX M3+ automated sprayer (HTX Technologies, Chapel Hill, NC, US) and deposited onto the sample in four layers (no dry time between passes) with an alternating vertical and horizontal spraying pattern and a track spacing of 4 mm. Flow rate, sprayer velocity, pressure, and temperature were set at 0.1 mL/min, 1200 mm/min, 10 psi, and 75 °C, respectively. 3.2.6. Multivariate Analysis The MALDI MS data from the chemical analysis of the aged (non-overlapping) fingermarks were transformed from continuum to centroid spectra in MassLynx (Waters Corp.) using automatic peak detection with subtraction of background (polynomial order 15, below curve 10%, tolerance 0.010). All spectra were then imported into SpecAlign (v. 2.4.1; Cartwright Group, PTCL, University of Oxford) for peak alignment. Baseline was subtracted (window size 5), and the data were de-noised (threshold 0.5) and normalized to TIC ( m / z range 100–1000). The spectra were then aligned to an average spectrum using the PAFFT algorithm (minimum segment size: 449 points; max shift: 20; scale: 1; reference: 0). Peaks were selected automatically (baseline cutoff = 0.5; window = 1; height ratio = 1.5). Peaks with a maximum intensity less than 10,000 were manually removed. Peaks corresponding to isotopes and matrix adducts were also removed. The processed data, containing a total of 194 ion peaks, was imported into SIMCA (v. 16.0.1; Sartorius Stedim Data Analytics AB, Göttingen, Germany) for multivariate statistical analysis. Pareto scaling was applied before principal component analysis (PCA) and orthogonal partial least-squares discriminant analysis (OPLS-DA). The unsupervised PCA served as a quick visualization of data, while the supervised OPLS-DA was used to differentiate between 0- and 14-days-old fingermarks and to identify ions important for the group separation. Hotelling’s T 2 with a significance level of 0.05 were used to identify potential outliers. Three components were used for the OPLS-DA model. The parameters Q 2 , R 2 X, R 2 Y (X being the matrix of ion features, and Y the matrix of the 0- and 14-day groups) were used to evaluate the performance of the model. Q 2 was obtained by seven-fold cross validation and explains the predictability of the model, whereas R 2 explains how well the model fit the data. A Q 2 score >0.4 and an R 2 > 0.5 is considered to indicate a robust model.
Trifluoroacetic acid (TFA), α-cyano-4-hydroxycinnamic acid (α-CHCA) was purchased from Sigma Aldrich (Poole, UK). Acetonitrile (ACN), acetone, and methanol were purchased from Fisher Scientific (Loughborough, UK). The double-sided conductive carbon tape was obtained from TAAB (Aldermaston, UK). TLC sheets were purchased from (Merck, UK).
3.2.1. Instrument and Instrumental Conditions All MALDI MS spectrometric analyses were carried out using the Waters MALDI QTOF Synapt G2 HDMS instrument (Waters Corporation, Manchester, UK). Data acquisition was performed within the m / z range of 100–1000 in positive sensitivity mode with a scan time of 1 s after calibrating the instrument with a saturated solution of red phosphorus in ACN. The Nd:YAG laser repetition rate was set to 1 kHz and laser power to 250 a.u. for all analyses. MALDI MS Images were acquired at a spatial resolution of 100 µm × 100 µm. For chemical analysis, data were acquired from three to four (random) areas for each fingermark using a circular shutting pattern consisting of approximately 100 shots. 3.2.2. Data Processing of MALDI MS Data MALDI MS fingermark images of the 4500 most intense ion signals were generated using the HDI software (v. 1.6, Waters Corp., UK) using a mass window of 0.02 Da. Visual inspection of those images led to the selection of the m / z ions that produced the best separation of the ridge pattern. MS spectra from each of the fingermark pairs were obtained through the selection of regions of interest in the outermost left and right sides of the ion images, respectively, where the fingermarks do not overlap. 3.2.3. Preparation of Fingermarks Samples for MALDI MSI TLC sheets were preliminarily prepared, removing the silica by soaking them in methanol and sonicating for 15 min. Upon wiping the residual silica with acetone, they were cut into glass slide sizes and used as the fingermark deposition surfaces. On eight of these resulting separate aluminium sheets, a groomed fingermark (deposited without washing hands at any particular time before the deposition but after rubbing the fingertip on forehead and cheeks) was deposited by the same donor and allowed to age for a definite time at ambient office conditions. Subsequently, the same donor deposited a second groomed fingermark partially overlapping the first, and the two impressions were allowed to age further for a definite time period. Specifically, eight pairs of different age overlapping fingermarks were deposited, as summarised in , where sets 1–5 intended to investigate the separation between old and fresh fingermarks and sets 6–8 the separation between old fingermarks (but of different age). An additional set of eight pairs of overlapping natural fingermarks (deposited without prior preparation of the fingertip, i.e., without washing hands or touching skin at any particular time before the deposition) were also prepared and aged in the same way. Matrix deposition was optimised, as described in , and the aluminum sheets were then mounted onto the Synapt MALDI target plate using double-sided conductive carbon tape. All samples were imaged by MALDI MS immediately after deposition of the matrix. The time delay between the deposition of the youngest fingermark and the start of MALDI MS analysis was 20–30 min (this time delay is included in the indicated ages of the fingermarks). 3.2.4. Preparation of Fingermark Samples for Chemical Analysis On seven separate aluminium slides, natural smudged fingermarks were deposited by the same person. Three of the fingermarks were allowed to age for seven days, while the remaining four were aged for fourteen days, all at ambient office conditions. Another fresh (25–107 min) smudged fingermark (natural) was then deposited by the person on the same aluminium sheet (one per slide for a total of seven) but without overlapping the first fingermark. The matrix was then sprayed, as described in , and the aluminium slide was mounted onto the MALDI sample plate using double-sided conductive carbon tape. The samples were analysed by MALDI MS immediately after deposition of the matrix. Three to four technical replicates were measured for each fingermark. 3.2.5. Matrix Application α-CHCA matrix was prepared at a concentration of 5 mg/mL in 70:30 ACN:0.5% TFA aq . The matrix was sprayed using the HTX M3+ automated sprayer (HTX Technologies, Chapel Hill, NC, US) and deposited onto the sample in four layers (no dry time between passes) with an alternating vertical and horizontal spraying pattern and a track spacing of 4 mm. Flow rate, sprayer velocity, pressure, and temperature were set at 0.1 mL/min, 1200 mm/min, 10 psi, and 75 °C, respectively. 3.2.6. Multivariate Analysis The MALDI MS data from the chemical analysis of the aged (non-overlapping) fingermarks were transformed from continuum to centroid spectra in MassLynx (Waters Corp.) using automatic peak detection with subtraction of background (polynomial order 15, below curve 10%, tolerance 0.010). All spectra were then imported into SpecAlign (v. 2.4.1; Cartwright Group, PTCL, University of Oxford) for peak alignment. Baseline was subtracted (window size 5), and the data were de-noised (threshold 0.5) and normalized to TIC ( m / z range 100–1000). The spectra were then aligned to an average spectrum using the PAFFT algorithm (minimum segment size: 449 points; max shift: 20; scale: 1; reference: 0). Peaks were selected automatically (baseline cutoff = 0.5; window = 1; height ratio = 1.5). Peaks with a maximum intensity less than 10,000 were manually removed. Peaks corresponding to isotopes and matrix adducts were also removed. The processed data, containing a total of 194 ion peaks, was imported into SIMCA (v. 16.0.1; Sartorius Stedim Data Analytics AB, Göttingen, Germany) for multivariate statistical analysis. Pareto scaling was applied before principal component analysis (PCA) and orthogonal partial least-squares discriminant analysis (OPLS-DA). The unsupervised PCA served as a quick visualization of data, while the supervised OPLS-DA was used to differentiate between 0- and 14-days-old fingermarks and to identify ions important for the group separation. Hotelling’s T 2 with a significance level of 0.05 were used to identify potential outliers. Three components were used for the OPLS-DA model. The parameters Q 2 , R 2 X, R 2 Y (X being the matrix of ion features, and Y the matrix of the 0- and 14-day groups) were used to evaluate the performance of the model. Q 2 was obtained by seven-fold cross validation and explains the predictability of the model, whereas R 2 explains how well the model fit the data. A Q 2 score >0.4 and an R 2 > 0.5 is considered to indicate a robust model.
All MALDI MS spectrometric analyses were carried out using the Waters MALDI QTOF Synapt G2 HDMS instrument (Waters Corporation, Manchester, UK). Data acquisition was performed within the m / z range of 100–1000 in positive sensitivity mode with a scan time of 1 s after calibrating the instrument with a saturated solution of red phosphorus in ACN. The Nd:YAG laser repetition rate was set to 1 kHz and laser power to 250 a.u. for all analyses. MALDI MS Images were acquired at a spatial resolution of 100 µm × 100 µm. For chemical analysis, data were acquired from three to four (random) areas for each fingermark using a circular shutting pattern consisting of approximately 100 shots.
MALDI MS fingermark images of the 4500 most intense ion signals were generated using the HDI software (v. 1.6, Waters Corp., UK) using a mass window of 0.02 Da. Visual inspection of those images led to the selection of the m / z ions that produced the best separation of the ridge pattern. MS spectra from each of the fingermark pairs were obtained through the selection of regions of interest in the outermost left and right sides of the ion images, respectively, where the fingermarks do not overlap.
TLC sheets were preliminarily prepared, removing the silica by soaking them in methanol and sonicating for 15 min. Upon wiping the residual silica with acetone, they were cut into glass slide sizes and used as the fingermark deposition surfaces. On eight of these resulting separate aluminium sheets, a groomed fingermark (deposited without washing hands at any particular time before the deposition but after rubbing the fingertip on forehead and cheeks) was deposited by the same donor and allowed to age for a definite time at ambient office conditions. Subsequently, the same donor deposited a second groomed fingermark partially overlapping the first, and the two impressions were allowed to age further for a definite time period. Specifically, eight pairs of different age overlapping fingermarks were deposited, as summarised in , where sets 1–5 intended to investigate the separation between old and fresh fingermarks and sets 6–8 the separation between old fingermarks (but of different age). An additional set of eight pairs of overlapping natural fingermarks (deposited without prior preparation of the fingertip, i.e., without washing hands or touching skin at any particular time before the deposition) were also prepared and aged in the same way. Matrix deposition was optimised, as described in , and the aluminum sheets were then mounted onto the Synapt MALDI target plate using double-sided conductive carbon tape. All samples were imaged by MALDI MS immediately after deposition of the matrix. The time delay between the deposition of the youngest fingermark and the start of MALDI MS analysis was 20–30 min (this time delay is included in the indicated ages of the fingermarks).
On seven separate aluminium slides, natural smudged fingermarks were deposited by the same person. Three of the fingermarks were allowed to age for seven days, while the remaining four were aged for fourteen days, all at ambient office conditions. Another fresh (25–107 min) smudged fingermark (natural) was then deposited by the person on the same aluminium sheet (one per slide for a total of seven) but without overlapping the first fingermark. The matrix was then sprayed, as described in , and the aluminium slide was mounted onto the MALDI sample plate using double-sided conductive carbon tape. The samples were analysed by MALDI MS immediately after deposition of the matrix. Three to four technical replicates were measured for each fingermark.
α-CHCA matrix was prepared at a concentration of 5 mg/mL in 70:30 ACN:0.5% TFA aq . The matrix was sprayed using the HTX M3+ automated sprayer (HTX Technologies, Chapel Hill, NC, US) and deposited onto the sample in four layers (no dry time between passes) with an alternating vertical and horizontal spraying pattern and a track spacing of 4 mm. Flow rate, sprayer velocity, pressure, and temperature were set at 0.1 mL/min, 1200 mm/min, 10 psi, and 75 °C, respectively.
The MALDI MS data from the chemical analysis of the aged (non-overlapping) fingermarks were transformed from continuum to centroid spectra in MassLynx (Waters Corp.) using automatic peak detection with subtraction of background (polynomial order 15, below curve 10%, tolerance 0.010). All spectra were then imported into SpecAlign (v. 2.4.1; Cartwright Group, PTCL, University of Oxford) for peak alignment. Baseline was subtracted (window size 5), and the data were de-noised (threshold 0.5) and normalized to TIC ( m / z range 100–1000). The spectra were then aligned to an average spectrum using the PAFFT algorithm (minimum segment size: 449 points; max shift: 20; scale: 1; reference: 0). Peaks were selected automatically (baseline cutoff = 0.5; window = 1; height ratio = 1.5). Peaks with a maximum intensity less than 10,000 were manually removed. Peaks corresponding to isotopes and matrix adducts were also removed. The processed data, containing a total of 194 ion peaks, was imported into SIMCA (v. 16.0.1; Sartorius Stedim Data Analytics AB, Göttingen, Germany) for multivariate statistical analysis. Pareto scaling was applied before principal component analysis (PCA) and orthogonal partial least-squares discriminant analysis (OPLS-DA). The unsupervised PCA served as a quick visualization of data, while the supervised OPLS-DA was used to differentiate between 0- and 14-days-old fingermarks and to identify ions important for the group separation. Hotelling’s T 2 with a significance level of 0.05 were used to identify potential outliers. Three components were used for the OPLS-DA model. The parameters Q 2 , R 2 X, R 2 Y (X being the matrix of ion features, and Y the matrix of the 0- and 14-day groups) were used to evaluate the performance of the model. Q 2 was obtained by seven-fold cross validation and explains the predictability of the model, whereas R 2 explains how well the model fit the data. A Q 2 score >0.4 and an R 2 > 0.5 is considered to indicate a robust model.
|
Psychophysiological Parameters Predict the Performance of Naive Subjects in Sport Shooting Training | f3179a29-56e2-435a-a66d-ad20ea7645e8 | 10054378 | Physiology[mh] | Sport shooting represents a complex sensorimotor process requiring a high level of visuospatial work. Shooting sports demand athletes maintain a good psychological state , stress control ability, and the ability to efficiently allocate cognitive resources (e.g., attention) during the shooting and aiming period . As a consequence, training in sport shooting is a non-trivial challenge that often requires an individualized approach, especially in a sport with such high achievements . The identification of the psychological and psychophysiological profile of a successful shooter is associated with superior performance, and the building of a training strategy focused on achieving the quickest achievement of this state could help in solving this problem. The development of modern, compact, and mobile devices for multimodal monitoring of human physiological parameters and the rapid progress in neuroimaging technologies makes it possible to monitor the current state of an athlete concurrently with their behavioral performance to form representations of a successful profile. Currently, research in this direction is mainly focused on identifying biomarkers of the cardiovascular and respiratory systems operation , gaze behavior , as well as EEG biomarkers of successful shooters ; this research is generally based on the comparison of novice shooters with professional athletes . However, with this approach, it is impossible to obtain information about the “trajectory” of the transformation from a novice shooter to a professional. A promising experimental paradigm from this point of view is the paradigm aimed at comparing successful and unsuccessful attempts at sport shooting training sessions in a naïve group. This approach makes it possible to identify what distinguishes successful attempts in novice athletes at the level of physiological parameters and EEG characteristics and to investigate the effect of training in detail. Recently, a trend for research in this direction has emerged. Note the study that revealed EEG and kinematic biomarkers of precision motor control and changes in the neurophysiological substrates in naïve participants that may underlie motor learning during simulated marksmanship in immersive virtual reality. However, many issues still remain unexplored. In particular, it is unknown exactly how and which physiological parameters and EEG characteristics change during sports shooting training; for example, which parameters correlate with shooting success and can thus claim to be biomarkers that are components of a professional athlete’s profile. Moreover, most studies generally examine the dynamics of one or two physiological parameters during shooting training (e.g., a study utilized synchronized monitoring of EEG and electrocardiogram (ECG) to understand the mechanism of dual activation of the brain and heart in pistol athletes during shooting performances). At the same time, a deeper understanding of the relationship between physiological and psychological processes and training success can only be achieved by simultaneously considering as many physiological parameters as possible. Biomarkers of successful sport shooting should be searched not at the level of operation of individual subsystems of the human body but at the level of their joint operation and interaction; therefore, it is necessary to use multimodal registration of physiological parameters to solve this problem . The present study takes a step toward solving the problems formulated. Here, we analyze multimodal data of subjects (EEG, ECG, electrooculogram (EOG), respiration activity (R), and fatigue tests) naïve to sport shooting training and study correlations between the psychophysiological parameters and shooting performance of the subjects. The special aspect of this study is the analysis of changes in fatigue levels during training and its effect on shooting success. From a fundamental point of view, sport provides an ideal model for understanding neural adaptations associated with intensive training over time. We believe that the increased knowledge of links between physiological parameters, brain activity, and behavior characteristics will help to improve the effect of sport shooting training and thus enhance sports performance. 2.1. Participants Experimental study included 21 healthy volunteers (all male, age 19–25, with an average age of 21 and a standard deviation of ∼1.5, right-handed). All subjects had no diseases that affected sight or locomotor functions. A healthy lifestyle was advised for the subjects prior to the experiment, which included sufficient night rest, no alcohol or drug consumption, and moderate physical activity. All subjects were volunteers; they were informed about the details of the study prior to participation, were able to ask related questions, and after that, provided informed consent. All participants were naïive to sport shooting, so before the experiment, a trained coach explained to them the basic principles and safety regulations. This study was conducted according to the guidelines of the Declaration of Helsinki and approved by the Ethics Committee of Lobachevsky University (Protocol №3 from 8 April 2021). 2.2. Experimental Setup During the experiment, we recorded multimodal data from a subject: EEG, EOG, ECG, respiration activity (R). The placement of all sensors is shown in A. All these signals were recorded by a wearable EEG recorder “Encephalan-EEGR-19/26” (Medicom MTD, Russia). The sampling rate for all types of data was 250 Hz. For EEG recording, we used 31 Ag/AgCl electrodes placed on the scalp according to the international scheme “10-10” ( A, grey circles). Other biological signals, besides EEG, were acquired through additional POLY channels of “Encephalan”. To record EOG, we used 2 electrodes (“EOG+” and “EOG-”) above and below the right eye ( A, green circles). The resulting EOG signal was calculated as the difference between these two signals. The right eye was chosen as it is usually the one used while aiming the shot. To record ECG, we placed 1 electrode on the subject’s back near the left scapula ( A, blue circle). Respiration activity was collected via a belt-shaped sensor wrapped around the subject’s chest ( A, white stripe). The stretching and contraction of the belt are associated with the expansion and compression of the thorax during respiration. When choosing the sensors’ placement, we tried not to restrict the subject’s movement and, at the same time, tried to minimize the influence of this movement on the recorded signals. The “Encephalan” device was placed on the small of the back with a special belt, and all wires from the device to the sensors were tightly packed together and fixed on the back. The “Encephalan” was connected to the PC through Bluetooth, so this connection provided no additional restriction on the subject’s movement. 2.3. Experimental Procedure The shooting was performed from an upright position, as illustrated in B. For the experiment, we chose an air rifle with characteristics close to the real rifle used by sportsmen in biathlons. The rifle’s dimensions are 1010 / 270 / 85 mm (length/height/width), and its weight is ∼4 kg. The rifle uses a 4.5 mm caliber with a 5-round magazine and open sights. Since this was an air rifle, the recoil was not significant. Protective gear included shooting glasses but not headphones. The subjects shot at 5 separate targets at a distance of 10 m. The targets mimicked the ones used in biathlons at a distance of 50 m, so the targets in the experiment were properly scaled in size. The subject had visual and audial feedback after each shot—the successfully struck target changed color and provided distinct sound. The experimental session included 21 series of shootings with Multidimensional Fatigue Inventory (MFI-20) , and the NASA Task Load Index (NASA-TLX) tests before the first and after the last series correspondingly (see C). The first series was treated as a test, so these results were excluded from further analysis. Each series included the following steps (see D): Preparation —the subject received the rifle loaded with 5 bullets from the assistant and assumed shooting stance; Shooting —the subject performed 5 shots at 5 targets in any order; Completion —the subject quit shooting stance and handed the rifle back to the assistant for reloading; VAS —the subject passed a visual analog scale (VAS) test for fatigue estimation; Rest —the subject rested for 60 s before the next series. To assess changes in some behavioral and physiological characteristics throughout the experiment, we turned 20 series of shootings into 4 blocks. This was done by averaging results of 5 consecutive series, i.e., 1–5, 6–10, 11–15, 16–20. MFI-20 is the test aimed at assessing a subject’s fatigue through self-report. This test includes 20 questions covering 5 dimensions of fatigue: Physical, Mental, and General Fatigue, as well as Reduced Activity and Motivation. NASA-TLX is another instrument to measure fatigue, but in this case, task-induced fatigue. The test includes several scales and their paired comparisons that help to assess 6 factors: Physical, Mental, and Temporal Demand, as well as Effort, Frustration, and Performance. VAS is used to subjectively measure the fatigue of the subject in his current state. Self-report is performed with the help of a continuous scale, on which the subject chooses the value of his current fatigue. The scale varies between “the lowest” and “the highest fatigue”. For all fatigue-assessment tests, we used a tablet computer. We considered several factors during statistical analysis: “block”—reflects the course of the experiment, includes blocks 1–4; “phase”—reflects the subject’s type of activity in the experiment, including rest and shooting; “result”—reflects successfulness on each shot, including hits and misses. 2.4. Data Processing The goals of preprocessing procedure were the following: for EEG data—to obtain clear signals without noises and artifacts for further time-frequency analysis, for respiration, EOG, and ECG—to obtain signals clear enough for extracting desired features such as blink rate or heart rate. For EEG preprocessing, we used Fieldtrip toolbox for MATLAB . EEG signals were filtered with a band-pass filter (cut-off frequencies—1 and 70 Hz) and 50 Hz notch filter in preparation for further time-frequency analysis. To remove eye- and heart-related activity artifacts from EEG, we used a method based on Independent Component Analysis (ICA). For this, we applied ft_componentanalysis with the method runica . We decomposed EEG data into a set of independent components, searched components with artifacts, removed them, and then restored EEG signals with the remaining components. To ensure data quality, we performed additional visual data analysis with ft_rejectvisual . We rejected trials of data and/or EEG channels with severe artifacts remaining after the ICA-based procedure. Most of these artifacts were related to the subject’s active movement. We removed “bad” trials from the dataset, while for “bad” channels, we performed a repairing procedure with ft_channelrepair . We performed a time-frequency analysis of EEG signals using continuous wavelet transform (CWT) with Morlet mother wavelet function . We considered wavelet power (WP) as W n ( f , t ) , where n = 1 , 2 , … , N is the number of EEG channel ( N = 31 for the considered dataset), f and t are the frequency and time point. WP is one of the common CWT-based characteristics to describe the time-frequency structure of a signal . To reduce the data dimensionality, we considered averaged CWT spectra. Firstly, we averaged WP over several areas in the cortex: frontal (F), central (C), parietal (P), occipital (O), left temporal (LT), and right temporal (RT) (see E). Secondly, we averaged WP over commonly used frequency bands: delta (1–4 Hz), theta (4–8 Hz), alpha (8–13 Hz), and beta (13–30 Hz). In our research, we considered a 2-s time interval just before the subject pulled the trigger. So we additionally averaged WP over this time interval. We used the NeuroKit2 software package to process signals obtained from the respiratory sensor. NeuroKit2 is an open-source Python package designed to process neurophysiological signals . For primary processing and filtering of the incoming signal, we used a linear detrending method with subsequent application of a low-pass fifth-order IIR Butterworth filter at the frequency of 2 Hz. The procedure is based on the zero-crossing algorithm with the amplitude threshold described in . Then, we determined peaks (beginning of exhalation) and valleys (beginning of inhalation) using different sets of parameters described in . Next, we determined the breathing phase defined between “1” for inspiration (inhalation) and “0” for expiration (exhalation). Then, we calculated the instantaneous frequency of the signal (in “1/min”) from a series of peaks. It is calculated as “60/period”, where the period is the time between peaks. To interpolate the frequency over the entire duration of the signal, the monotone cubic interpolation method was used. We also calculated the average values of frequencies at different stages of the experiment. For this purpose, the instantaneous respiration rate was calculated for each session at the moments of shooting and rest; further, the obtained rate values were averaged and added up for each subject. We analyzed EOG to detect eye movement and blinking using the methods of the software package MNE , which turned out to be the most effective for this problem. We used a default set of parameters for this method. Additionally, we obtained the values of the signal peaks, which correspond to the moments of the subject’s blinks. Next, we calculated the blink rate (in minutes) from the series of peaks as “60/period”. Monotone cubic interpolation method was used to interpolate the frequency for the entire duration of the signal. Then, the average values of blink rates at the moments of shooting and rest were obtained for each subject. To process the ECG signal, we filtered the data using high-pass and low-pass filters in the 1–6 Hz range. Further, R-peaks, which are distinguished by high amplitude and frequency, were selected from the prepared signal. We calculated heart rate as the inverse of the R-R interval ( 1 / t R − R ). All heart rate values for each individual step were averaged for each subject. We have considered different time window scales for the analysis of heart rate, respiration rate, and blink rate. To find a difference between stages of the experiment (rest vs. shooting), we averaged heart rate, respiration rate, and blink rate in windows length equal to respective stages. The time length of windows for the resting stage is 60 s, but windows for the shooting stage have different lengths (average length of 22.5 s) because of different rates of shooting across the subjects and shooting stages. Additionally, we analyzed the influence of instantaneous (right at the moment of shot) RR on shooting results. The main effects at the group level were evaluated via Repeated Measures Analysis of Variance (RM ANOVA). We considered “block”, “result”, “phase”, and cortical area as within-subject factors in those statistical tests where the influence of these factors was considered. The post hoc analysis used either paired samples t -test or Wilcoxon signed-rank test, depending on the samples’ normality. Normality was tested via the Shapiro–Wilk test. The group-level correlation analysis between all pairs of characteristic changes during the experiment, such as heart rate, respiration rate, characteristic of the brain activity, hit rate, and subjective fatigue, was performed using repeated measures correlation. Correlations between subjective tests (MFI-20, NASA-TLX) and shooting accuracy were searched using Spearman’s rank correlation coefficient. We used several open-source statistical packages in Python, such as Pingouin, SciPy, statsmodels, and a package called JASP for statistical analysis and results visualization. Experimental study included 21 healthy volunteers (all male, age 19–25, with an average age of 21 and a standard deviation of ∼1.5, right-handed). All subjects had no diseases that affected sight or locomotor functions. A healthy lifestyle was advised for the subjects prior to the experiment, which included sufficient night rest, no alcohol or drug consumption, and moderate physical activity. All subjects were volunteers; they were informed about the details of the study prior to participation, were able to ask related questions, and after that, provided informed consent. All participants were naïive to sport shooting, so before the experiment, a trained coach explained to them the basic principles and safety regulations. This study was conducted according to the guidelines of the Declaration of Helsinki and approved by the Ethics Committee of Lobachevsky University (Protocol №3 from 8 April 2021). During the experiment, we recorded multimodal data from a subject: EEG, EOG, ECG, respiration activity (R). The placement of all sensors is shown in A. All these signals were recorded by a wearable EEG recorder “Encephalan-EEGR-19/26” (Medicom MTD, Russia). The sampling rate for all types of data was 250 Hz. For EEG recording, we used 31 Ag/AgCl electrodes placed on the scalp according to the international scheme “10-10” ( A, grey circles). Other biological signals, besides EEG, were acquired through additional POLY channels of “Encephalan”. To record EOG, we used 2 electrodes (“EOG+” and “EOG-”) above and below the right eye ( A, green circles). The resulting EOG signal was calculated as the difference between these two signals. The right eye was chosen as it is usually the one used while aiming the shot. To record ECG, we placed 1 electrode on the subject’s back near the left scapula ( A, blue circle). Respiration activity was collected via a belt-shaped sensor wrapped around the subject’s chest ( A, white stripe). The stretching and contraction of the belt are associated with the expansion and compression of the thorax during respiration. When choosing the sensors’ placement, we tried not to restrict the subject’s movement and, at the same time, tried to minimize the influence of this movement on the recorded signals. The “Encephalan” device was placed on the small of the back with a special belt, and all wires from the device to the sensors were tightly packed together and fixed on the back. The “Encephalan” was connected to the PC through Bluetooth, so this connection provided no additional restriction on the subject’s movement. The shooting was performed from an upright position, as illustrated in B. For the experiment, we chose an air rifle with characteristics close to the real rifle used by sportsmen in biathlons. The rifle’s dimensions are 1010 / 270 / 85 mm (length/height/width), and its weight is ∼4 kg. The rifle uses a 4.5 mm caliber with a 5-round magazine and open sights. Since this was an air rifle, the recoil was not significant. Protective gear included shooting glasses but not headphones. The subjects shot at 5 separate targets at a distance of 10 m. The targets mimicked the ones used in biathlons at a distance of 50 m, so the targets in the experiment were properly scaled in size. The subject had visual and audial feedback after each shot—the successfully struck target changed color and provided distinct sound. The experimental session included 21 series of shootings with Multidimensional Fatigue Inventory (MFI-20) , and the NASA Task Load Index (NASA-TLX) tests before the first and after the last series correspondingly (see C). The first series was treated as a test, so these results were excluded from further analysis. Each series included the following steps (see D): Preparation —the subject received the rifle loaded with 5 bullets from the assistant and assumed shooting stance; Shooting —the subject performed 5 shots at 5 targets in any order; Completion —the subject quit shooting stance and handed the rifle back to the assistant for reloading; VAS —the subject passed a visual analog scale (VAS) test for fatigue estimation; Rest —the subject rested for 60 s before the next series. To assess changes in some behavioral and physiological characteristics throughout the experiment, we turned 20 series of shootings into 4 blocks. This was done by averaging results of 5 consecutive series, i.e., 1–5, 6–10, 11–15, 16–20. MFI-20 is the test aimed at assessing a subject’s fatigue through self-report. This test includes 20 questions covering 5 dimensions of fatigue: Physical, Mental, and General Fatigue, as well as Reduced Activity and Motivation. NASA-TLX is another instrument to measure fatigue, but in this case, task-induced fatigue. The test includes several scales and their paired comparisons that help to assess 6 factors: Physical, Mental, and Temporal Demand, as well as Effort, Frustration, and Performance. VAS is used to subjectively measure the fatigue of the subject in his current state. Self-report is performed with the help of a continuous scale, on which the subject chooses the value of his current fatigue. The scale varies between “the lowest” and “the highest fatigue”. For all fatigue-assessment tests, we used a tablet computer. We considered several factors during statistical analysis: “block”—reflects the course of the experiment, includes blocks 1–4; “phase”—reflects the subject’s type of activity in the experiment, including rest and shooting; “result”—reflects successfulness on each shot, including hits and misses. The goals of preprocessing procedure were the following: for EEG data—to obtain clear signals without noises and artifacts for further time-frequency analysis, for respiration, EOG, and ECG—to obtain signals clear enough for extracting desired features such as blink rate or heart rate. For EEG preprocessing, we used Fieldtrip toolbox for MATLAB . EEG signals were filtered with a band-pass filter (cut-off frequencies—1 and 70 Hz) and 50 Hz notch filter in preparation for further time-frequency analysis. To remove eye- and heart-related activity artifacts from EEG, we used a method based on Independent Component Analysis (ICA). For this, we applied ft_componentanalysis with the method runica . We decomposed EEG data into a set of independent components, searched components with artifacts, removed them, and then restored EEG signals with the remaining components. To ensure data quality, we performed additional visual data analysis with ft_rejectvisual . We rejected trials of data and/or EEG channels with severe artifacts remaining after the ICA-based procedure. Most of these artifacts were related to the subject’s active movement. We removed “bad” trials from the dataset, while for “bad” channels, we performed a repairing procedure with ft_channelrepair . We performed a time-frequency analysis of EEG signals using continuous wavelet transform (CWT) with Morlet mother wavelet function . We considered wavelet power (WP) as W n ( f , t ) , where n = 1 , 2 , … , N is the number of EEG channel ( N = 31 for the considered dataset), f and t are the frequency and time point. WP is one of the common CWT-based characteristics to describe the time-frequency structure of a signal . To reduce the data dimensionality, we considered averaged CWT spectra. Firstly, we averaged WP over several areas in the cortex: frontal (F), central (C), parietal (P), occipital (O), left temporal (LT), and right temporal (RT) (see E). Secondly, we averaged WP over commonly used frequency bands: delta (1–4 Hz), theta (4–8 Hz), alpha (8–13 Hz), and beta (13–30 Hz). In our research, we considered a 2-s time interval just before the subject pulled the trigger. So we additionally averaged WP over this time interval. We used the NeuroKit2 software package to process signals obtained from the respiratory sensor. NeuroKit2 is an open-source Python package designed to process neurophysiological signals . For primary processing and filtering of the incoming signal, we used a linear detrending method with subsequent application of a low-pass fifth-order IIR Butterworth filter at the frequency of 2 Hz. The procedure is based on the zero-crossing algorithm with the amplitude threshold described in . Then, we determined peaks (beginning of exhalation) and valleys (beginning of inhalation) using different sets of parameters described in . Next, we determined the breathing phase defined between “1” for inspiration (inhalation) and “0” for expiration (exhalation). Then, we calculated the instantaneous frequency of the signal (in “1/min”) from a series of peaks. It is calculated as “60/period”, where the period is the time between peaks. To interpolate the frequency over the entire duration of the signal, the monotone cubic interpolation method was used. We also calculated the average values of frequencies at different stages of the experiment. For this purpose, the instantaneous respiration rate was calculated for each session at the moments of shooting and rest; further, the obtained rate values were averaged and added up for each subject. We analyzed EOG to detect eye movement and blinking using the methods of the software package MNE , which turned out to be the most effective for this problem. We used a default set of parameters for this method. Additionally, we obtained the values of the signal peaks, which correspond to the moments of the subject’s blinks. Next, we calculated the blink rate (in minutes) from the series of peaks as “60/period”. Monotone cubic interpolation method was used to interpolate the frequency for the entire duration of the signal. Then, the average values of blink rates at the moments of shooting and rest were obtained for each subject. To process the ECG signal, we filtered the data using high-pass and low-pass filters in the 1–6 Hz range. Further, R-peaks, which are distinguished by high amplitude and frequency, were selected from the prepared signal. We calculated heart rate as the inverse of the R-R interval ( 1 / t R − R ). All heart rate values for each individual step were averaged for each subject. We have considered different time window scales for the analysis of heart rate, respiration rate, and blink rate. To find a difference between stages of the experiment (rest vs. shooting), we averaged heart rate, respiration rate, and blink rate in windows length equal to respective stages. The time length of windows for the resting stage is 60 s, but windows for the shooting stage have different lengths (average length of 22.5 s) because of different rates of shooting across the subjects and shooting stages. Additionally, we analyzed the influence of instantaneous (right at the moment of shot) RR on shooting results. The main effects at the group level were evaluated via Repeated Measures Analysis of Variance (RM ANOVA). We considered “block”, “result”, “phase”, and cortical area as within-subject factors in those statistical tests where the influence of these factors was considered. The post hoc analysis used either paired samples t -test or Wilcoxon signed-rank test, depending on the samples’ normality. Normality was tested via the Shapiro–Wilk test. The group-level correlation analysis between all pairs of characteristic changes during the experiment, such as heart rate, respiration rate, characteristic of the brain activity, hit rate, and subjective fatigue, was performed using repeated measures correlation. Correlations between subjective tests (MFI-20, NASA-TLX) and shooting accuracy were searched using Spearman’s rank correlation coefficient. We used several open-source statistical packages in Python, such as Pingouin, SciPy, statsmodels, and a package called JASP for statistical analysis and results visualization. 3.1. The Behavioral Data Analysis The results of the assessing subject’s state before the experimental task with the MFI-20 test are shown in A. The median values are low (less than 8 out of a possible 20) across all scales of MFI-20, which confirms that none of the subjects has asthenia of any type. To assess the task-induced load, we used a NASA-TLX test, and the results are shown in B. We found that the experimental task induces low temporal and mental loads, while the main load is caused by the effort to preserve a certain level of performance. The results of the change in fatigue level during the task assessed with VAS after each series of shootings are shown in C. We considered z-scored results of VAS for a more universal data presentation. We found a significant increase in fatigue from block to block, and post hoc analysis showed significant differences between all blocks of the experiment. However, absolute values for the induced increase in fatigue (i.e., the difference between fatigue at the beginning and at the end of the experiment) are close to 30 out of 100 (maximal value in the scale). We suggest that this result indicates a low overall increase in fatigue during the experiment. We used the hit rate as a parameter for evaluating the success of performance. The subjects coped well with the task: ∼65% of the shots hit the target on average. We analyzed changes in hit rate over the course of the experiment and found a significant increase in hit rate (RM ANOVA: p < 0.001 ). Post hoc analysis showed significant differences between the first and fourth blocks, as well as between the third and fourth blocks. 3.2. The Physiological Data Analysis We analyzed changes in physiological characteristics during the experimental task, both in the resting and shooting phases. 3.2.1. Heart Rate We did not find significant changes in the heart rate during the experiment, as well as no significant differences between heart rates at rest and shooting phases. However, we found an interaction effect between factors “block” and “phase” ( p = 0.000531). Post hoc analysis showed that there are significant differences in heart rate between blocks 1–3 and 1–4 in the rest phase (see A). Additionally, we considered heart rate variability as another characteristic of heart activity but did not find significant changes. 3.2.2. Respiration Rate Then, we analyzed the dynamics of respiration rate and found the interaction effect between “block” and “phase” ( p = 0.045) factors, while no changes were detected in respiration rate during the experiment and between the phases. In the post hoc analysis, we found a decrease in respiration rate during the shooting phase, but the statistical significance of these changes is near the accepted threshold (see B). Further, we studied the effect of instantaneous respiration rate on shooting success and found a significant difference in the instantaneous respiration rate between misses and hits ( p = 0.043, see C). 3.2.3. Blinking Rate We have not found significant changes in the blinking rate during the experiment or any relationship between the blinking rate and the hit rate. 3.2.4. Brain Electrical Activity We analyzed changes in the electrical activity of the brain directly before each shot, both for the “block” and “result” factors. We did not find significant changes during the experiment. However, for energy in the delta range, we found a main effect of shooting results ( p = 0.042) and cortex areas ( p = 0.013) (see D). For energy in the alpha range, we did not reveal the main effects. Nevertheless, we found an interaction effect between cortex areas and shooting results ( p = 0.016). In the post hoc analysis, we found significant changes in the right temporal lobe in the alpha range ( p = 0.049507); however, a p -value was not adjusted for multiple comparisons. Finally, we revealed that energies in the delta range and the alpha range in the right temporal lobe were significantly less before a hit compared to a miss. 3.3. Correlation Analysis To identify the relationships between the characteristics under study, we performed a correlation analysis. The results of correlation analysis are shown in . We discovered that changes in subjective fatigue positively correlate with average heart rate in the rest phase ( r = 0.42 ). Simultaneously, the hit rate correlates with the following parameters: respiration rate in the resting phase ( r = 0.33 ), respiration rate in the shooting phase ( r = − 0.35 ), and energies before the shot in the theta range in the frontal and central regions ( r = − 0.33 and r = − 0.33 , respectively). Additionally, we identified the correlation between the NASA-TLX and the hit rate ( ρ = − 0.532 ). The results of the assessing subject’s state before the experimental task with the MFI-20 test are shown in A. The median values are low (less than 8 out of a possible 20) across all scales of MFI-20, which confirms that none of the subjects has asthenia of any type. To assess the task-induced load, we used a NASA-TLX test, and the results are shown in B. We found that the experimental task induces low temporal and mental loads, while the main load is caused by the effort to preserve a certain level of performance. The results of the change in fatigue level during the task assessed with VAS after each series of shootings are shown in C. We considered z-scored results of VAS for a more universal data presentation. We found a significant increase in fatigue from block to block, and post hoc analysis showed significant differences between all blocks of the experiment. However, absolute values for the induced increase in fatigue (i.e., the difference between fatigue at the beginning and at the end of the experiment) are close to 30 out of 100 (maximal value in the scale). We suggest that this result indicates a low overall increase in fatigue during the experiment. We used the hit rate as a parameter for evaluating the success of performance. The subjects coped well with the task: ∼65% of the shots hit the target on average. We analyzed changes in hit rate over the course of the experiment and found a significant increase in hit rate (RM ANOVA: p < 0.001 ). Post hoc analysis showed significant differences between the first and fourth blocks, as well as between the third and fourth blocks. We analyzed changes in physiological characteristics during the experimental task, both in the resting and shooting phases. 3.2.1. Heart Rate We did not find significant changes in the heart rate during the experiment, as well as no significant differences between heart rates at rest and shooting phases. However, we found an interaction effect between factors “block” and “phase” ( p = 0.000531). Post hoc analysis showed that there are significant differences in heart rate between blocks 1–3 and 1–4 in the rest phase (see A). Additionally, we considered heart rate variability as another characteristic of heart activity but did not find significant changes. 3.2.2. Respiration Rate Then, we analyzed the dynamics of respiration rate and found the interaction effect between “block” and “phase” ( p = 0.045) factors, while no changes were detected in respiration rate during the experiment and between the phases. In the post hoc analysis, we found a decrease in respiration rate during the shooting phase, but the statistical significance of these changes is near the accepted threshold (see B). Further, we studied the effect of instantaneous respiration rate on shooting success and found a significant difference in the instantaneous respiration rate between misses and hits ( p = 0.043, see C). 3.2.3. Blinking Rate We have not found significant changes in the blinking rate during the experiment or any relationship between the blinking rate and the hit rate. 3.2.4. Brain Electrical Activity We analyzed changes in the electrical activity of the brain directly before each shot, both for the “block” and “result” factors. We did not find significant changes during the experiment. However, for energy in the delta range, we found a main effect of shooting results ( p = 0.042) and cortex areas ( p = 0.013) (see D). For energy in the alpha range, we did not reveal the main effects. Nevertheless, we found an interaction effect between cortex areas and shooting results ( p = 0.016). In the post hoc analysis, we found significant changes in the right temporal lobe in the alpha range ( p = 0.049507); however, a p -value was not adjusted for multiple comparisons. Finally, we revealed that energies in the delta range and the alpha range in the right temporal lobe were significantly less before a hit compared to a miss. We did not find significant changes in the heart rate during the experiment, as well as no significant differences between heart rates at rest and shooting phases. However, we found an interaction effect between factors “block” and “phase” ( p = 0.000531). Post hoc analysis showed that there are significant differences in heart rate between blocks 1–3 and 1–4 in the rest phase (see A). Additionally, we considered heart rate variability as another characteristic of heart activity but did not find significant changes. Then, we analyzed the dynamics of respiration rate and found the interaction effect between “block” and “phase” ( p = 0.045) factors, while no changes were detected in respiration rate during the experiment and between the phases. In the post hoc analysis, we found a decrease in respiration rate during the shooting phase, but the statistical significance of these changes is near the accepted threshold (see B). Further, we studied the effect of instantaneous respiration rate on shooting success and found a significant difference in the instantaneous respiration rate between misses and hits ( p = 0.043, see C). We have not found significant changes in the blinking rate during the experiment or any relationship between the blinking rate and the hit rate. We analyzed changes in the electrical activity of the brain directly before each shot, both for the “block” and “result” factors. We did not find significant changes during the experiment. However, for energy in the delta range, we found a main effect of shooting results ( p = 0.042) and cortex areas ( p = 0.013) (see D). For energy in the alpha range, we did not reveal the main effects. Nevertheless, we found an interaction effect between cortex areas and shooting results ( p = 0.016). In the post hoc analysis, we found significant changes in the right temporal lobe in the alpha range ( p = 0.049507); however, a p -value was not adjusted for multiple comparisons. Finally, we revealed that energies in the delta range and the alpha range in the right temporal lobe were significantly less before a hit compared to a miss. To identify the relationships between the characteristics under study, we performed a correlation analysis. The results of correlation analysis are shown in . We discovered that changes in subjective fatigue positively correlate with average heart rate in the rest phase ( r = 0.42 ). Simultaneously, the hit rate correlates with the following parameters: respiration rate in the resting phase ( r = 0.33 ), respiration rate in the shooting phase ( r = − 0.35 ), and energies before the shot in the theta range in the frontal and central regions ( r = − 0.33 and r = − 0.33 , respectively). Additionally, we identified the correlation between the NASA-TLX and the hit rate ( ρ = − 0.532 ). We analyzed multimodal psychophysiological data (EEG, ECG, EOG, respiration activity, and fatigue) to explore the neural and behavioral mechanisms underlying precision visual-motor control learning during sports shooting tasks. We systematically studied the relationship between physiological parameters, brain activity, and shooting performance over the course of learning to identify biomarkers that can be used to infer complex motor behavior. As expected, naive subjects significantly increased their hit rate during practice. Participants became, on average, ≈30% more accurate at shooting targets. Analysis of the physiological activity showed that performance improvements during the course of learning were accompanied by an increase in subjective fatigue and heart rate, wherein the average breathing rate remained unchanged. Respiration rate during shooting negatively correlates with marksmanship performance. We also found that the instantaneous respiration rate before a hit is higher than before a miss. Note that the work did not reveal the influence of the instantaneous respiration rate on the shooting results. However, in study , the authors showed that respiration rate is related to the mental load, with high and medium load characterized by a significantly higher rate. In this regard, we hypothesize that our results may reflect a connection between mental load, instantaneous respiration rate, and shooting results. We suggest that in the case of a hit, the subjects were more deeply immersed and concentrated on the task and, accordingly, experienced a higher mental load than in the case of a miss. Analysis of brain activity reveals several markers associated with shooting success. We found that average energy values in the delta range and the alpha range in the right temporal lobe were significantly less before a successful shot than before a miss. We identified that the hit rate negatively correlates with energies in the theta range in the frontal and central regions during the aiming period before shot execution. The sport marksmanship task used in this study is one of the most convenient examples of tasks that can be used for investigating neurophysiological mechanisms underlying precise visual–motor coordination in a complex naturalistic context. Usually, studies addressing visual–motor integration by analyzing noninvasive recordings of cortical activity, such as EEG, involve laboratory tasks with minimal mobility to reduce artifact-producing muscle activity. Real-world tasks in natural environments require unrestricted full-body movements arising from full engagement of perception, decision-making, error recognition, and motor control. The shooting task is a controlled, easily replicated natural exercise that is particularly useful for investigating psychophysiological markers of visual-motor skill learning because it produces discrete measures of performance, which can be compared with electrophysiological activity recorded in real-time. One of the main goals of this study was to identify EEG biomarkers of visual-motor skill learning during sport shooting tasks. Biomarkers are often referred to as quantitative indicators of a biological organism’s state and can be used to describe behavior-related psychophysiological processes. In recent years, the relationship of biomarkers with certain skills has been actively investigated . The identified associations of skills with biomarkers are a promising tool for training process optimization. In this study, we analyzed the relationship between EEG power in different frequency bands during the aiming period and shooting performance. We found that novices demonstrated delta and right temporal alpha EEG power increase before missing shots. Our results are in line with other studies reporting an overall reduction of alpha activity for experienced shooters . This effect is interpreted as a greater engagement of task-relevant attentional processes. Janelle et al. showed that shooting task expertise interacted with hemispheric activation levels. They demonstrated stronger alpha activity in the left hemisphere accompanied by its reduction in the right hemisphere for experts as compared to novices during the preparatory period before shot execution. Since shooting places high demands on visuospatial processing, the elevation of alpha power in the left temporal area may indicate a decrease of non-relevant to task cognitive activity (cognitive thinking, self-talk, or language analysis) and show that marksmen focused their attention on the visuospatial work dominated by right-brain areas . Our results show the existence of a negative correlation between theta-band energies in the frontal and central brain regions during the preparation period and shooting success. Frontal midline (Fm) theta activation has often been observed in tasks that required consistent attention to a stimulus . Recent studies reported Fm theta power as an indicator of sustained and internalized attention found in the preparation period in motor performance. Fm theta activity is linked to various kinds of attentional or working memory processes, such as working memory , learning , concentration , and action monitoring . Sauseng et al. associated Fm theta power with the number of cognitive resources allocated to attentional processes during a complex finger movement task learning. They clearly showed that Fm theta increased with increasing mental efforts and task demands. The results of our study are in line with these findings, demonstrating weaker theta activation with increasing correct acquisitions of the task and experience by the novice. Sport shooting task highly demands focused attention and precision visual-motor control. Shooting learning requires from the naive subjects a lot of cognitive resources and mental engagement. Therefore, the shooting training process is accompanied by a high level of mental effort reflected by increasing theta energy in the frontal and central brain regions. This explanation is confirmed by comparing the perceived workload level evaluated by NASA-TLX with the hit rate (see ). Subjects with high hit rates reported greater confidence by feeling less workload level (lower levels of stress and pressure). In line with these results, Borghini et al. demonstrated that the variation of the EEG power spectra in frontal areas in the theta band could be used as a measure for the training improvements of novices in flight simulation tasks. Their results showed that behavioral and task performance improvement was accompanied by a significant decrease in the theta band power over the frontal areas. Interestingly, the comparison of the time course of Fm theta during the aiming period in rifle shooting between experts and novices reveals that the theta power increased during the aiming process before the shot only for experts but not for novices . The authors assume that elite marksmen are better able to allocate cortical resources in time while novices are unable to focus attention exactly on the shooting time point. Note that this study has several limitations. First is the small number of participants (21). The second limitation is that only males participated in this study. Another limitation is using only EEG for the brain activity analysis since EEG has low spatial resolution compared to other techniques such as fMRI (functional magnetic resonance imaging). The last limitation is especially significant in the case of a possible investigation of visual-motor connection. For instance, in a recent paper , the usage of fMRI allowed researchers to discover a disrupted visual-motor connection in psychiatric disorders. In this study, however, fMRI is very difficult to use without substantial changes in the experimental paradigm. In conclusion, our study sheds light on the neural and behavioral mechanisms underlying precision visual-motor control learning during sport shooting. We found that performance improvements were accompanied by an increase in subjective fatigue and heart rate and that the respiration rate before a hit was higher than before a miss, potentially reflecting a connection between the mental load and shooting results. Additionally, we identified several EEG biomarkers of visual-motor skill learning, including head-averaged delta and right temporal alpha EEG power increase before missing shots and a negative correlation between theta-band energies in the frontal and central brain regions and shooting success. The results of this study highlight the importance of considering both neural and behavioral factors in precision visual-motor control learning and the potential for using psychophysiological parameters to improve shooting performance. These findings provide valuable insights into the neurophysiological mechanisms underlying visual-motor skill learning and have potential implications for the optimization of training processes. |
null | e046db02-08d7-42b3-8024-c7c831b442ed | 10054384 | Pharmacology[mh] | The world is enriched with a treasure trove of traditional medicinal herbs that are of global importance for health security. India harbours four mega-biodiversity hotspots and is highly enriched with 17,500 species of medicinal plants used effectively against multiple disorders . These medicinal herbs have been time-tested and recommended by saints, maharishis, vaidyas, and ayurvedic acharyas and have strong credence in different traditional medicinal systems such as ayurveda, unani, siddha, homeopathy, naturopathy, allopathy, and traditional Chinese medicine for treating ailments related to both humans and animals. Moreover, these medicines are safe, heal the cause of the ailment, and have less or no side effects compared to allopathic drugs . For centuries, plant essential oils have played a provocative role for mankind. The people of Egypt were known to be skilled perfumers and taught the art of perfumery to Hebrews around 5000 years ago . Earlier naturalized or wild plants provided social security to people in the form of supplements, fuel, fodder, raw material for companies, and an additional good income source. According to the WHO, approximately 80% of people are reliant on herbal remedies in developing nations. A total of 90% of herbal species used in India are brought from the western Himalayas, which is well known for its rich plant diversity, with 1748 medicinal species used in various fields such as pharmacological research, chemistry, clinical therapeutic studies, and pharmacognosy. Unfortunately, the traditional knowledge of herbal medicine is declining due to dependence on allopathy, which is associated with side-effects or ill effects on non-target organs. The synergistic effect of phytochemicals has multiple modes of actions that heal the disease and have immunomodulatory effects. In this post-COVID-19 era, people have become more conscious about their health and peace of mind. Thus, there is a drastic increase in interest and demand towards herbal medicines for improved quality of life . The volatile aromatic oils find applications and are used commercially in cosmetics, soap, perfumery, the spice industry, flavoured tea, drinks, traditional foods, pesticides, and pharmaceutical industries . The use of herbal-origin essential oils has increased greatly over the last few years, which has consequently increased the international market demand and decisively switched the trade. Sharma et al. reported that India produces 15 essential oils on a pilot/experimental scale and approx. 20 herbal essential oils at a commercial level that have a unique capability, and India holds an economic advantage due to its rich diversity in soil and favourable climate. The oil content and quality depend on the type of soil and the climatic conditions. Artemisia is a medicinally important genus belonging to the Asteraceae family which is also known as the Compositae family, thistle family, daisy family, and sunflower family . This genus is gaining much attention due to its remarkable medicinal properties, phytochemical diversity, and scientifically proven health benefits . The Chinese knew the therapeutic value of Artemisia 2000 years ago. In 1596, for the treatment of malaria symptoms, Li Shizhen suggested tea made from Artemisia (qinghao). The name ‘Artemisia’ was eventually derived from the great Greek goddess (Roman Diana) Artemis, the name of the Greek queens Artemisia I and II , and specifically named after the Caria Queen Artemisia II, who was a medical researcher and botanist by profession in the 4th century BC. The genus Artemisia , commonly known as wormwood, mugwort, or sagebrush, is distributed in the temperate zones of Europe, North America, and India . Species of this genus are used as folklore medicines and in pharmaceutical sectors their demand has increased because of their medicinal value and high commercial importance . The genus Artemisia comprises 500 species . The greatest number of species seems to occur in Asia, with 174 spp. in the ex-USSR, 150 spp. in China, 50 spp. in Japan, 35 spp. in Iran, and 35 spp. in India, mostly from the northwestern Himalayas . In India, the eminently recognized species that have been used traditionally are Artemisia vestita , Artemisia dracunculus , Artemisia brevifolia , Artemisia roxburghiana , Artemisia dubia , Artemisia herba-alba , Artemisia japonica , Artemisia santolinifolia , Artemisia maritima , Artemisia scoparia , Artemisia absinthium , Artemisia verlotiorum , Artemisia maritima , Artemisia annua, and Artemisia vulgaris. Mucciarelli and Maffei reported that Artemisia possess antioxidant, antimicrobial, anticoagulant, antispasmodic, antidiabetic, anti-helminthic, anticancer, anti-ulcer, anticonvulsant, stomachic, cardiac stimulant, insecticidal, febrifuge, and cytotoxic properties and are also used for the treatment of coughs, colds, dyspepsia, headaches, malaria, and inflammation . It is quite difficult to differentiate among the Artemisia species due to their morphological similarity , which leads to misinterpretation and misidentification of the products that are of economic and commercial medicinal value . Several Artemisia-based health care products (tablets, syrups, oils, creams) have been commercialized. The World Flora Online reported A. vestita Wall. ex Besser as an accepted botanical name, whereas A. vestita Wall. ex DC. and A. vestita var. vestita are synonyms of A. vestita. In India, one of the major aromatic plants, i.e., A. vestita, is under commerce, which is obtained by distillation, hydro diffusion, expression, solvent extraction, and the natural carriers–host organisms . Traditionally, local farmers and tribal communities refer to A. vestita as an anti-helminthic agent because of the unavailability and high cost of synthetic drugs . It is widely used by tribal people in the Kashmir Valley (‘Tethe-Ven’) for treating parasitic infections . In Tibet, it is commonly known by the name ‘Maolianhao’ (alias: Wannianpeng), a folk medicine , whereas in Kashmir it is known as ‘Roosi Tyethven’ . It is also commonly known as ‘Russian Wormwood’, ‘Ganga Tulsi’, ‘Buer’, ‘Drubsha’, ‘Seski, Kubsha’, ‘Chamariya’, ‘Kundja’, and ‘Kundiyaa’. To date, there have been fragmentary reports on A. vestita extracts, their mode of action, standardization, dose optimization, and toxicity. Through this article, we have tried to bridge the gap and provide explicit information on the distribution, botanical description, phytochemistry, and pharmacological activities of Russian wormwood. Artemisia vestita is widely distributed in East Asia including the Himalayas of Nepal, Pakistan, India, and Tibet to southern and central China , including hills, rocky slopes, grasslands, shrublands, and outer forest margins of various regions of Gansu, northwest Guangxi, north Hubei, Guizhou, Liaoning, west Sichuan, Qinghai, Xizang, Xinjiang, Yunnan, north India, Nepal, and north Pakistan. Due to geographical variation and seasonal factors, drastic variations have been reported in the chemical composition of the essential oil. In India, Drubsha is found in Himachal Pradesh, Kashmir, and Uttarakhand at an altitude of 2100–3000 m. The distribution of A. vestita in the Pooh region of Kinnaur district at different elevations noted a density of 1.27/ha, frequency of 10%, abundance of 12.67, ratio of abundance to frequency (A/F) of 1.27, and importance value index of 5.12 at an elevation of 2700–3200 m; a density of 2.25/ha, frequency of 15%, abundance of 15, A/F of 1, and IVI of 7.28 at an elevation of 3200–3700 m; and a density of 1.17/ha, frequency of 11.67%, abundance of 10, A/F of 0.86, and IVI of 53.53 at an elevation of 3700–4200 m . Important value indexes are the sum of the frequency, density, and dominance of the individual species. 3.1. Morphology A. vestita Wall . ex Besser is an aromatic, erect, perennial shrub that may attain a height of 2 m (5–120 cm) . The leaves are fern-like, soft, and hairy on the upper surface white and hairy towards the lower side, and pinnately cut. The flowers are small and appear creamy yellow, arranged in racemes (6–10); the heads of flowers are long, hairy, and compound, hanging gracefully on their slender nodding stalks. The fruits are shiny and smooth and the bracts are oblong and membranous . The leaves have clear abaxial and adaxial surfaces. Adaxial surface-elongated epidermal cells are partitioned and surface grooved with prominently ridged margins; the tertiary sculpture is aggregated; stomata are depressed, surrounded by thick, flat peristomal rims, transverse striata, thick inner ledges, and a gradually concave surface of the guard cells and transverse polar folds. The lobule tips are cap-like and swollen; stomata are also present in this particular region, with cells greatly undulated (V-shaped undulations) and elongated. The abaxial surface–cell outline is similar, somehow deeply undulate, with loose ‘V’-shaped undulations; it is tertiary sculptured and coarsely granular; the guard cells have oblique folds . 3.2. Vegetation Details A. vestita can dominate the grassland ecosystem . The A. vestita plants are hermaphrodite and pollinated by insects, and the seeds ripen in the months of August–October. Habitat : Woodland edge garden, sunny position, cultivated beds, hills, rocky slopes, grasslands, shrublands, and exterior forest margins at an altitude of 2000–4300 m above sea level . Cytology : Gupta et al. reported 2 n = 2x = 36 meiotic chromosome count, ploidy level (4×), pollen fertility 78–82%, and pollen grain size 22–24 μm in A. vestita collected from Haripurdhar and Churdhar (HP) at an altitude of 2400 and 3650 m, respectively, above sea level. Two other cytotypes, i.e., hexaploid (2 n = 54) and diploid (2 n = 18) , have also been reported. Chlorophyll content : Variations in the chlorophyll and anthocyanin content and the chlorophyll/carotenoid ratio have been reported in the temperate species of A. vestita found at an altitude of 550 and 3600 m above sea level in the Garhwal Himalayas. The total chlorophyll content observed in the lower leaves of A. vestita was 1.716 and 1.470, middle leaves 0.902 and 1.650, and top leaves 0.863 and 1.205 at two different altitudes (550 and 3600 m, respectively) . The molar chlorophyll/carotenoid ratio was observed to be lower (1) in the temperate species as compared to the tropical (1.7) and subtropical species (1.3) . At higher altitudes, A. vestita plants have relatively broader adaptability potential compared to lowland species, which is due to a higher osmotic concentration, greater lignification, and the tendency of osmoregulation in tissues, due to the conversion of starch into sugar content . A palynological study of A. vestita stated the quantitative characteristics of the plant, such as a polar axis of 19.38 ± 1.52µm, equatorial axis of 18.09 ± 1.51 µm, P/E (sphericity) of 1.07, thickness of exine of 2.13 ± 0.67 µm, and colpus length of 11.81 ± 1.69 µm, and spinules are prominent in the plant . A. vestita Wall . ex Besser is an aromatic, erect, perennial shrub that may attain a height of 2 m (5–120 cm) . The leaves are fern-like, soft, and hairy on the upper surface white and hairy towards the lower side, and pinnately cut. The flowers are small and appear creamy yellow, arranged in racemes (6–10); the heads of flowers are long, hairy, and compound, hanging gracefully on their slender nodding stalks. The fruits are shiny and smooth and the bracts are oblong and membranous . The leaves have clear abaxial and adaxial surfaces. Adaxial surface-elongated epidermal cells are partitioned and surface grooved with prominently ridged margins; the tertiary sculpture is aggregated; stomata are depressed, surrounded by thick, flat peristomal rims, transverse striata, thick inner ledges, and a gradually concave surface of the guard cells and transverse polar folds. The lobule tips are cap-like and swollen; stomata are also present in this particular region, with cells greatly undulated (V-shaped undulations) and elongated. The abaxial surface–cell outline is similar, somehow deeply undulate, with loose ‘V’-shaped undulations; it is tertiary sculptured and coarsely granular; the guard cells have oblique folds . A. vestita can dominate the grassland ecosystem . The A. vestita plants are hermaphrodite and pollinated by insects, and the seeds ripen in the months of August–October. Habitat : Woodland edge garden, sunny position, cultivated beds, hills, rocky slopes, grasslands, shrublands, and exterior forest margins at an altitude of 2000–4300 m above sea level . Cytology : Gupta et al. reported 2 n = 2x = 36 meiotic chromosome count, ploidy level (4×), pollen fertility 78–82%, and pollen grain size 22–24 μm in A. vestita collected from Haripurdhar and Churdhar (HP) at an altitude of 2400 and 3650 m, respectively, above sea level. Two other cytotypes, i.e., hexaploid (2 n = 54) and diploid (2 n = 18) , have also been reported. Chlorophyll content : Variations in the chlorophyll and anthocyanin content and the chlorophyll/carotenoid ratio have been reported in the temperate species of A. vestita found at an altitude of 550 and 3600 m above sea level in the Garhwal Himalayas. The total chlorophyll content observed in the lower leaves of A. vestita was 1.716 and 1.470, middle leaves 0.902 and 1.650, and top leaves 0.863 and 1.205 at two different altitudes (550 and 3600 m, respectively) . The molar chlorophyll/carotenoid ratio was observed to be lower (1) in the temperate species as compared to the tropical (1.7) and subtropical species (1.3) . At higher altitudes, A. vestita plants have relatively broader adaptability potential compared to lowland species, which is due to a higher osmotic concentration, greater lignification, and the tendency of osmoregulation in tissues, due to the conversion of starch into sugar content . A palynological study of A. vestita stated the quantitative characteristics of the plant, such as a polar axis of 19.38 ± 1.52µm, equatorial axis of 18.09 ± 1.51 µm, P/E (sphericity) of 1.07, thickness of exine of 2.13 ± 0.67 µm, and colpus length of 11.81 ± 1.69 µm, and spinules are prominent in the plant . A. vestita has been used as a folkoric medicine and was harvested from wild forests for use in anti-inflammatory and antifebrile medicines. Both aqueous and alcoholic solvents give a maximum amount of medicinal extract from the plant compared to other solvents . It is widely used for treating numerous inflammatory diseases in Tibet and China, such as contact dermatitis, rheumatoid arthritis, and sepsis . The leaves are crushed and applied externally on the skin as hemostatic . The plant is also used in treating stomach-aches . The essential oils of Kubsha are volatile and complex mixtures of sesquiterpenes providing a strong odour to the herbal plant. Extraction is performed using steam distillation or hydro-distillation methods. Leaves, stems, barks, aerial parts, inflorescences, whole plants, fruits, seeds, flowers, and roots are used for the extraction of essential oil and further used to combat human ailments, but the composition varies due to altitudinal variation. The isolated compounds have been identified using various techniques such as GC-MS, GC-FID, HRMS, UV, IR FTIR, HPLC-MS, GLC-MS, HPLC, UPLC-ESI-QqQLIT-MS/MS, 1D and 2D NMR, X-ray crystallography, and silica gel and polyacrylamide chromatography. On the basis of a literature survey, it has been reported and evidently showed that Russian wormwood essential oil composition is greatly influenced by the climate or geographical region and exhibits remarkable chemodiversity. Phytochemical studies revealed that A. vestita contains several monoterpenes, flavones, and sesquiterpenoids, among which the camphor/eucalyptol chemotype appears predominantly in most of the Artemisia species . A Chinese research group reported the isolation of 15 chemical compounds from A. vestita and identified them as taurin, isoferulic acid, 8-dimethoxy flavone, yomogin, friedelin, beta-sitosterol, α and β-amyrin, daucosterol, 7-hydroxy-6,8-dimethoxy coumarin, scoplatin, caffeic acid . Camphene, 1,8-Cineole, thujone, camphor, artemisia ketone, caryophyllene, and germacrene D were reported as major components of Artemisia species essential oil . Zhengming et al. conducted preliminary research on the chemical compounds, which were identified as saponins, organic acids, tannins, phenols, anthraquinones, flavonoids, lactones, coumarin, alkaloids, volatile oil, triterpenes, or steroids. Another study found 12 chemical constituents from essential oil including six monoterpenic derivatives, three monoterpenes, and three sesquiterpenes, among them 1, 8-cineol, camphor, and borneol, which were 39.01%, 26.92%, and 19.23%, respectively . In addition, daucosterol, stigmasterol, scopolin, scoparone, umbelliferone, and isoscopoletin-O-glucoside were yielded from the plant . A sesquiterpenoid Allohimachalol was identified in ref. ; α-, β- and γ-himachalene, germacrene D, caryophyllene, himachalol, α- and γ-atlantone, allo -himachalol, 1,8-cineole, santolina alcohols and their acetates, yomogi alcohol, thujanols and thujones were reported in ref. ; yomogi alcohol, (E)-2,5,5-Trimethylhepta-3,6-dien-2-ol, alpha-Atlantone, Himachalol, and gamma-Himachalene were found in ref. ; alpha-terpinene, terpenyl acetate, thujyl alcohol, α- and β-phellandrene, nerol, cineol, thujyl acetate, neral, artemisol, and beta-thujone were reported in ref. ; and arvestonol, arvestolides D-J, and arvestonates A-C occurred in ref. . Numerous sesquiterpenoids exhibiting biological activities were isolated from A. vestita and had a great influence on the plant’s defense against phytopathogenic fungi and pests, so they can be further utilized as antifungal agents and natural insecticides (scopoletin, ruin, luteolin, salicylic acid, naringenin, eugenol, kaempferol, dihydroartemisinin, isoeugenol, artemether, chrysin, and artemisinin) . Rutin compound was highest in the ethanolic extract of A. vestita collected from Jammu and Kashmirregion. In total, 27 compounds were identified, out of which the main components were eucalyptol, 1,8-cineol, grandisol, camphor, and germacrene D ; allohimachalol, himachalene, germacrene D, caryophyllene, and himachalol ; eucalyptol camphor and borneol ; 1,8-cineole, (E)-citral (13.7%), limonene, α-phellandrene, camphor, and (Z) and (E)-thujones ; β-caryophyllene, artemisia alcohol, artemisia ketone, 1,8-cineol, and α-phellandrene ; and pectolinarigenin, apigenin, cirsilineol, 5,7,3′,4′-tetrahydroxy-6,5′-Dimethoxyflavone, 7-methoxycoumarin, patuletin, annphenone, umbelliferone, scopoletin, 2,4-dihydroxy-6-methoxyacetophenone, and quercetin . Flavones such as pectolinarigenin, cirsilineol, jaceosidin, cirsimaritin, quercetin, hispidulin, 6-methoxytricin, apigenin, and acacetin have reported from Lhasa, Tibet . n-carpryaldehyde, a-phellandrene, 1,8-cineol, a-terpinene, thujone, thujyl alcohol, citronellol, citral, geraniol, aromadendrene, cadinene, and chamazulene have also been reported . A total of 18 components was found in Daksum, Kokerrnag, Kashmir, out of which the principal components were (E)-citral, 1,8-cineole, limonene, camphor, α-phellandrene, and (Z) and (E)-thujones. The dominant group were oxygenated monoterpenes, comprising 73.1% of terpenes in the plant essential oil composition, followed by monoterpene hydrocarbons (17.3%) . A higher content of oxygenated monoterpenes in plants imparts a strong characteristic aroma to the Artemisia species. α-amyrin, daucosterol, stigmasterol, β-sitosterol, scoparone, scopolin, umbelliferone, and isoscopoletin-o-glycoside have been extracted from the aerial parts of A. vestita . Sesqueterpenoids found in the plant are known for cytotoxic, antiviral, and antiphlogistic activities . β-caryophyllene, artemisia alcohol, artemisia ketone, 1, 8-cineol and α-phellandrene were found to be major active compounds in accessions from Nainital, Uttarakhand, whereas α-Pinene, Camphene, Artemiseole, α-Pinene, α-Myrcene, 1,8-Cineol (Eucalyptol), δ-Terpinene, Camphor, trans-Pinocamphone, a-Pinocarvone, Terpinen-4-ol, Grandisol, γ-Pyronene, Copaene, β-Cubebene, α-Caryophyllene, Caryophyllene, α-Amorphene, γ-Himachalene, Germacrene D, Aromadendrene, α-Zingiberene, γ-Elemene, δ-Cadinene, Caryophyllene oxide, α-Bisabolol oxide B, and (Z)-α–Santalolare active compounds were found in accessions from Mentougou District, Beijing. Pectolinarigenin, apigenin, cirsiliol, 7-methoxycoumarin, patuletin, annphenone, umbelliferone, scopoletin, 2,4-dihydroxy-6-methoxyacetophenone, quercetin, cirsilineol, jaceosidin, cirsimaritin, hispidulin, 6-methoxytricin, and acacetin were found in accessions from Lhasa, Tibet. Scopoletin, rutin, luteolin, salicylic acid, naringenin, eugenol, kaempferol, dihydroartemisinin, isoeugenol, artemether, chrysin, artemisinin, Himachalene, germacrene-D, caryophyllene, allohimachalol, himachalol, atlantone, yomogi alcohol, 1, 8-cineole, santolina alcohol, thujones, and thujanols were found in accessions from Srinagar, Kashmir. α-amyrin, daucosterol, stigmasterol, β-sitosterol, scoparone, scopolin, umbelliferone, isoscopoletin-o-glycoside, arvestonol, arvestolides D–J, and arvestonates A–C were found in accessions from China. Essential Oil A total of 202 biochemical compounds have been reported from different parts (stem, leaves, roots) of A. vestita . There are mainly flavonoids, terpenoids, oxygenated monoterpenes, sesquiterpene lactones, oxygenated sesquiterpenes, sesquiterpenoids, hydroxyl cinnamic acids, monoterpene hydrocarbons, azulenes, sesquiterpene hydrocarbons, sterols, phenylpropanoids, monoterpenoids, hydroxycoumarins, coumarins, flavonoid glycosides, organosulfonic acids, oxygenated triterpenes, and aromatic aldehydes. The chemical structures of the bioactive compounds present in Russian wormwood are shown in . The characteristic oil odour is determined by the constituents of fresh-smelling thujone and eucalyptol (1,8-cineole) and by the woody bark and the sweet note of atlantone and himachalol compounds. The odour of the plant’s essential oil is mainly woody, herbaceous, fresh, slightly sweet, and reminiscent of the sage and balsamic odour and is reported to be effective against dermatophytes . The oil can be used safely in perfumery, scented soaps, and cosmetics. Camphor, borneol, and eucalyptol are responsible for the pleasant aroma and are also known for their antifungal and antibacterial activities. The flowering tops of the plants when subjected to steam distillation yield yellow-, orange-, or brown-coloured oil with an aromatic, sweet-woody odour . The essential oil was produced from A. vestita from the Nainital hills, Shimla hills, and Kashmir valley . The physiochemical characteristics observed in the essential oil of A. vestita are a refractive index D of 1.4915, ester value of 55.45, acid value of 1.3, carbonyl percentage of 20.80, ester value of 124.4 after acetylation, 1:1 solubility (in 95% alcohol), and specific gravity of 0.910 . A total of 202 biochemical compounds have been reported from different parts (stem, leaves, roots) of A. vestita . There are mainly flavonoids, terpenoids, oxygenated monoterpenes, sesquiterpene lactones, oxygenated sesquiterpenes, sesquiterpenoids, hydroxyl cinnamic acids, monoterpene hydrocarbons, azulenes, sesquiterpene hydrocarbons, sterols, phenylpropanoids, monoterpenoids, hydroxycoumarins, coumarins, flavonoid glycosides, organosulfonic acids, oxygenated triterpenes, and aromatic aldehydes. The chemical structures of the bioactive compounds present in Russian wormwood are shown in . The characteristic oil odour is determined by the constituents of fresh-smelling thujone and eucalyptol (1,8-cineole) and by the woody bark and the sweet note of atlantone and himachalol compounds. The odour of the plant’s essential oil is mainly woody, herbaceous, fresh, slightly sweet, and reminiscent of the sage and balsamic odour and is reported to be effective against dermatophytes . The oil can be used safely in perfumery, scented soaps, and cosmetics. Camphor, borneol, and eucalyptol are responsible for the pleasant aroma and are also known for their antifungal and antibacterial activities. The flowering tops of the plants when subjected to steam distillation yield yellow-, orange-, or brown-coloured oil with an aromatic, sweet-woody odour . The essential oil was produced from A. vestita from the Nainital hills, Shimla hills, and Kashmir valley . The physiochemical characteristics observed in the essential oil of A. vestita are a refractive index D of 1.4915, ester value of 55.45, acid value of 1.3, carbonyl percentage of 20.80, ester value of 124.4 after acetylation, 1:1 solubility (in 95% alcohol), and specific gravity of 0.910 . Kubsha is highly enriched with various phytochemicals responsible for enormous pharmacological activities such as antiphlogistic, antifebrile, antifeedant, anti-helminthic, antibacterial, antifungal, antiviral, antitumor, antiproliferation, antidote, immunosuppressive activity, diuretic, hypoglycemic, antiepileptic, antioxidative, wound-healing, clearing away itching, ringworms, skin infections, and respiratory tract infections, ethnic therapy for colds, sinus drainage, maintaining ventilation, reducing inflammation and asthma, and many more. The anti-adipogenic activity of A. vestita requires in-depth studies . 6.1. Wound-Healing Fresh leaf paste is applied on wounds or cuts to stop the bleeding and inflammation . Chinese medicines include A. vestita as an alternative and complementary medicine for the treatment of skin diseases. A. vestita has a cold nature and is capable of treating skin eruptions, heat, and itching . Pastes of the leaves are applied for the treatment of skin infections, inflammation, ringworm, wounds, respiratory tract infections, ethnic therapy for colds, sinus drainage, and asthma . 6.2. Antidote A. vestita has been used for the treatment of snake bite due to its high content of monoterpenes, flavones, and sesquiterpenoids in the leaf extracts. 6.3. Antimicrobial The A. vestita essential oils are mainly composed of an odoriferous mixture of sesquiterpenes, monoterpenes, and aromatic compounds, which are used in naturopathy are very well known for their antimicrobial properties. Two major compounds, namely, grandisol and 1,8-cineol, have shown in vivo and in vitro antibacterial activity against respiratory-infection-causing bacteria. The oil exhibited MIC (minimum inhibitory concentration) values between 20 and 80 µg/mL, whereas the constituents exhibited between 130 and 200 µg/mL. The in vivo studies showed significant results of the oil and its component grandisol, which did not produce any toxic effects in mice . Eight components that have been reported to exhibit antibacterial activity are α- and β-thujone, terpinen-4-ol, linalool, nerol, geraniol, α-pinene, and 1,8-cineole; their percentages in the oil were determined in ref. . The plant extract and formulated gel have also shown significant results, as the plant extract exhibited MIC values between 100 and 240 µg/mL while the formulated gel (extract + natural polymer) exhibited MIC values between 30 and 85 µg/mL (unpublished work) against skin-infection-causing bacterial and fungal species. 6.4. Immunosuppressive Activity Plant constituents are useful in healing immunological disorders such as autoimmune disorders and also in organ transplantation, as they produce immune-suppressive agents . It has been reported that the essential oil of plants with higher amounts of α- and β-thujone have lesser or trace amounts of eucalyptol and camphor . Jaceosidin exerts immunosuppressive effects both in vitro and in vivo through the activation and inhibition of T-cell proliferation, which is associated with the down-regulation of interferon (IFN)- gamma signal transducers and activators of the (STAT1) transcription1 and transcription factor T-box TBX-21 signaling pathway . In addition, flavones such as apigenin, cirsilineol, and 6-methoxytricin from A. vestita have shown immunosuppressive and anti-inflammatory effects . These flavones specifically inhibit PCA (passive cutaneous anaphylaxis), which induces contact hypersensitivity, whereas lymphocyte proliferation is induced by Con-A and CD-25 expression in T-cells, which shows immunosuppressive effects. 6.5. Anti-Inflammatory Activity A. vestita extracts exhibit anti-inflammatory activity such as degranulation inhibition in mast cells and inflammatory cytokine production . The plant extracts inhibited the proliferation of mouse splenocytes and mixed lymphocytes while reducing the IL-2 interleukin level and the level of metallo-proteinase-9 in vivo and in vitro . The LD 50 of less than 1000 mg/Kg was devoid of antiprotozoal, antibacterial, antifungal, antiviral, anthelmintic, diuretic, hypoglycaemic and anticancer, which showed its effect on isolated guinea pig respiration, ileum, nictitating membrane, cardiovascular system, and central nervous system . A. vestita extract cross-linked with tragacanth gum (crosslinked polyacrylic-acid-based hydrogel) showed anti-inflammatory activity. The inhibition exhibited by a plant extract and plant extract formulated gel with a natural polymer of COX-1 (Cyclooxygenase-1) was found to be 97.962 ± 0.892% and 69.812 ± 0.911%, respectively, at 500 µg/mL concentration. Similarly, our group also explored a significant inhibition of cyclooxygenase-2. The inhibition percentage by the plant extract and plant extract formulated gel were 89.47 ± 1.401% and 52.76 ± 1.110%, respectively (unpublished work). 6.6. Anti-Epileptic Activity Hispidulin is a flavonoid naturally occurring in A. vestita with powerful anti-epileptic activity. Hispidulin showed a 21.1893 Kcal/mol in silico docking score while targeting the human enzyme glycogen phosphorylase-b/chrysin, which showed the greater potential of this inhibitor molecule to become an effective antidiabetic drug to control hyperglycemia in type-2 diabetes. However, the work demands thorough in vivo and in vitro studies for the molecules to be used as anti-hyperglycemic drugs . 6.7. Antifeedant Activity The compound artemivestinolide showed antifeedant activities against the third-instar larvae of Plutella xylostella with EC50 values of 25.3–42 and against the phytopathogenic fungi F. oxysporum (MIC-256 mg/L), P. oryzae (MIC-128 mg/L), and B. cinerea (MIC-256 mg/L) . A. vestita ethanolic extract exhibited anti-inflammatory, anti-helminthic, and insecticidal activity against Haemonchus contortus and Sitophilus zeamais, respectively . The whole-plant extract containing vegetative shoot exhibited a 87.2% reduction in the faecal egg count at 100 mg/kg, which showed significant activity against adult worms and larvae after 28 days post-treatment . In a study, essential oil of A. vestita showed potential fumigant activity with an LC 50 value of 13.42 mg/L and LD 50 value of 50.62 mg against adult Sitophilus zeamais in the fumigant bioassay and in the contact bioassay, respectively . Several monoterpenes, sesquiterpenoids, and flavones have been isolated from A. vestita and the essential oil chemical composition has been well-studied . 1,8-cineol has a cold-relieving effect with mucolytic and expectorant properties . 1,8-cineol and camphor present in plant essential oil act as fumigants with a broad insecticidal activity and possess the potential to expand as a novel natural fumigant for insect control in stored products . They are advantageous over conventional fumigants because they are non-persistent, biodegradable, and easily procurable and exhibit low toxicity to mammals . Fresh leaf paste is applied on wounds or cuts to stop the bleeding and inflammation . Chinese medicines include A. vestita as an alternative and complementary medicine for the treatment of skin diseases. A. vestita has a cold nature and is capable of treating skin eruptions, heat, and itching . Pastes of the leaves are applied for the treatment of skin infections, inflammation, ringworm, wounds, respiratory tract infections, ethnic therapy for colds, sinus drainage, and asthma . A. vestita has been used for the treatment of snake bite due to its high content of monoterpenes, flavones, and sesquiterpenoids in the leaf extracts. The A. vestita essential oils are mainly composed of an odoriferous mixture of sesquiterpenes, monoterpenes, and aromatic compounds, which are used in naturopathy are very well known for their antimicrobial properties. Two major compounds, namely, grandisol and 1,8-cineol, have shown in vivo and in vitro antibacterial activity against respiratory-infection-causing bacteria. The oil exhibited MIC (minimum inhibitory concentration) values between 20 and 80 µg/mL, whereas the constituents exhibited between 130 and 200 µg/mL. The in vivo studies showed significant results of the oil and its component grandisol, which did not produce any toxic effects in mice . Eight components that have been reported to exhibit antibacterial activity are α- and β-thujone, terpinen-4-ol, linalool, nerol, geraniol, α-pinene, and 1,8-cineole; their percentages in the oil were determined in ref. . The plant extract and formulated gel have also shown significant results, as the plant extract exhibited MIC values between 100 and 240 µg/mL while the formulated gel (extract + natural polymer) exhibited MIC values between 30 and 85 µg/mL (unpublished work) against skin-infection-causing bacterial and fungal species. Plant constituents are useful in healing immunological disorders such as autoimmune disorders and also in organ transplantation, as they produce immune-suppressive agents . It has been reported that the essential oil of plants with higher amounts of α- and β-thujone have lesser or trace amounts of eucalyptol and camphor . Jaceosidin exerts immunosuppressive effects both in vitro and in vivo through the activation and inhibition of T-cell proliferation, which is associated with the down-regulation of interferon (IFN)- gamma signal transducers and activators of the (STAT1) transcription1 and transcription factor T-box TBX-21 signaling pathway . In addition, flavones such as apigenin, cirsilineol, and 6-methoxytricin from A. vestita have shown immunosuppressive and anti-inflammatory effects . These flavones specifically inhibit PCA (passive cutaneous anaphylaxis), which induces contact hypersensitivity, whereas lymphocyte proliferation is induced by Con-A and CD-25 expression in T-cells, which shows immunosuppressive effects. A. vestita extracts exhibit anti-inflammatory activity such as degranulation inhibition in mast cells and inflammatory cytokine production . The plant extracts inhibited the proliferation of mouse splenocytes and mixed lymphocytes while reducing the IL-2 interleukin level and the level of metallo-proteinase-9 in vivo and in vitro . The LD 50 of less than 1000 mg/Kg was devoid of antiprotozoal, antibacterial, antifungal, antiviral, anthelmintic, diuretic, hypoglycaemic and anticancer, which showed its effect on isolated guinea pig respiration, ileum, nictitating membrane, cardiovascular system, and central nervous system . A. vestita extract cross-linked with tragacanth gum (crosslinked polyacrylic-acid-based hydrogel) showed anti-inflammatory activity. The inhibition exhibited by a plant extract and plant extract formulated gel with a natural polymer of COX-1 (Cyclooxygenase-1) was found to be 97.962 ± 0.892% and 69.812 ± 0.911%, respectively, at 500 µg/mL concentration. Similarly, our group also explored a significant inhibition of cyclooxygenase-2. The inhibition percentage by the plant extract and plant extract formulated gel were 89.47 ± 1.401% and 52.76 ± 1.110%, respectively (unpublished work). Hispidulin is a flavonoid naturally occurring in A. vestita with powerful anti-epileptic activity. Hispidulin showed a 21.1893 Kcal/mol in silico docking score while targeting the human enzyme glycogen phosphorylase-b/chrysin, which showed the greater potential of this inhibitor molecule to become an effective antidiabetic drug to control hyperglycemia in type-2 diabetes. However, the work demands thorough in vivo and in vitro studies for the molecules to be used as anti-hyperglycemic drugs . The compound artemivestinolide showed antifeedant activities against the third-instar larvae of Plutella xylostella with EC50 values of 25.3–42 and against the phytopathogenic fungi F. oxysporum (MIC-256 mg/L), P. oryzae (MIC-128 mg/L), and B. cinerea (MIC-256 mg/L) . A. vestita ethanolic extract exhibited anti-inflammatory, anti-helminthic, and insecticidal activity against Haemonchus contortus and Sitophilus zeamais, respectively . The whole-plant extract containing vegetative shoot exhibited a 87.2% reduction in the faecal egg count at 100 mg/kg, which showed significant activity against adult worms and larvae after 28 days post-treatment . In a study, essential oil of A. vestita showed potential fumigant activity with an LC 50 value of 13.42 mg/L and LD 50 value of 50.62 mg against adult Sitophilus zeamais in the fumigant bioassay and in the contact bioassay, respectively . Several monoterpenes, sesquiterpenoids, and flavones have been isolated from A. vestita and the essential oil chemical composition has been well-studied . 1,8-cineol has a cold-relieving effect with mucolytic and expectorant properties . 1,8-cineol and camphor present in plant essential oil act as fumigants with a broad insecticidal activity and possess the potential to expand as a novel natural fumigant for insect control in stored products . They are advantageous over conventional fumigants because they are non-persistent, biodegradable, and easily procurable and exhibit low toxicity to mammals . Although no specific reports on toxicity have been recorded for A. vestita extracts, the genus Artemisia contains allergenic sesquiterpenoid lactones that have the potential to cause skin reactions or dermatitis . The presence of volatile terpenoids and monoterpenes, i.e., pinene, eugenol, 1,8-cineole, limonene, citronellol, terpinolene, citronellal, thymol, and camphor, in A. vestita essential oil constituents provides repellent or toxic activity . Compounds named Arvestolides H and I showed inhibitory effects in BV-2 cells on nitric oxide production, which was induced by lipopolysaccharide with an IC 50 value of 43.2 μM for Arvestolides H and 39.9 μM for Arvestolides I. , whereas cirsilineol, apigenin, and 6-methoxytricin are the active components that inhibited the proliferation of T-cells and the activation of in vitro bioassays. Immune-suppressive compounds extracted from A. vestita will be an effective remedy for T-cell-mediated inflammation . Sesquiterpenes, coumarins, and flavones were reported in wormwood . The aqueous leaf extract alleviates picryl chloride (PCl)-induced contact hypersensitivity by blocking the T lymphocyte activation . In a study, flow cytometric and MTT assays were used for determining the CD 25 expression in T-cells and the proliferation of Con A induced lymphocytes . For cytotoxic effects, both S2 (extract) and S4 (extract + polymer) showed higher cell viability. At a 1000 µg/mL concentration of S2, the cytotoxicity to HaCat cells was 18.2 ± 0.35% compared to that of S4, which was 19.7 ± 0.29%. The results showed promising anti-inflammatory effects along with significant anti-cancer effects on HaCat cell lines (unpublished work). The aforementioned results provide A. vestita -based folklore medicine a rationale to be used in wound-healing and anti-cancer therapy. 8.1. Biological Activity of Annphenone Annphenone showed specific and potent antiproliferative activity against HepG2 cells and the IC50 value was 2.0 ± 0.4 μg/mL. During the cell cycle analysis in the G0/G1 phase, annphenone compound arrested the HepG2 cells when detecting the immunocytochemistry. It is suggested that the annphenone compound inhibits the catenin expression induced by the localization transfer, reducing the cyclin D1 protein expression. Furthermore, annphenone’s interaction as a possible ligand of the ASGP-R asialoglycoprotein receptor using a molecular docking simulation revealed its selectivity for hepatocellular carcinoma cells and potentially specific for antiproliferative activity . Annphenone is a promising anti-tumour agent present in the aqueous extract that reduced the contact sensitivity via down-regulating the adhesion, activation, and production of metalloproteinase T-lymphocytes in mice , whereas the ethanolic extract exerted anti-sepsis activity through down-regulation of the NF-kB and MAPK pathways . 8.2. Biological Activity of Cirsilineol Cirsilineol (4′,5-dihydroxy-3′,6,7-trimethoxyflavone) found in A. vestita extracts possesses potent anti-tumour and immune suppressive properties . Cirsilineol significantly inhibited the proliferation of multiple types of cancer cells (Skov-3, PC3, Caov-3, and Hela cells) in a concentration-dependent manner. It induced apoptosis in Caov-3 cells in a dose-dependent manner, which was determined with annexin V/propidium iodide double staining. To promote apoptosis, cirsilineol activates caspase-9, caspase-3, and PARP (poly ADP-ribose polymerase). Cirsilineol-induced loss of the mitochondrial membrane potential (MMP) brings a remarkable change and releases cytochrome-c to the cytosol. The induction of apoptosis via the mitochondrial pathway is the mode of action for the anti-proliferative activity of cirsilineol against cancerous cells. Moreover, cirsilineol is effective in ameliorating TNBS (tri-nitrobenzene sulfonic acid) -induced experimental colitis in mice, possibly because of its novel immunoregulatory activities with selective inhibitor IFN-γ/STAT1/T-bet signaling in the colonic lamina propria CD4 + T-cells particular for Crohn’s disease . 8.3. Biological Activity of Jaceosidin Regulation of the transcription activator and signal transducer (STAT 1) is being explored for the treatment of bowel diseases. However, few chemicals have been reported for the inhibition of STAT1/ IFN-g signaling for the treatment of Crohn’s disease. A natural compound, cirsilineol, isolated from the A. vestita plant significantly ameliorated TNBS (trinitro-benzene sulphonic acid)-induced T-cell-mediated mice colitis. It is closely associated with reduced auto-reactive T-cell activation and proliferation. Moreover, the action of anti-inflammatory and pro-inflammatory cytokines with cirsilineol therapy was found to increase the regulatory T-cell activity and decrease the effector Th-1 cell activity, as characterized by the up-regulation of TGF-b and IL-10 and down-regulation of IFN-g. Importantly, in the presence of a higher level of IFN-g the inhibition by compound cirsilineol of STAT1/IFN-g signalling seems reversible, suggesting that compound cirsilineol might be a potential candidate for the treatment of human inflammatory T-cell-mediated bowel diseases . The flavone jaceosidin isolated from A. vestita showed an antiproliferation effect on several human cancer cell lines. It significantly reduced SKOV-3, PC3, HeLa, and CAOV-3 cell proliferation in a concentration-dependent manner, whereas CAOV-3 showed time-dependent inhibition, and apoptosis increased in CAOV-3 cells. Flavone induced the cleavage of PARP (poly ADP-ribose polymerase) and caspase-3 and increased the cleaved caspase-9 levels. It also elevated the cytochrome c level in the cytosol, which shows the antitumor property of jaceosidin . Annphenone showed specific and potent antiproliferative activity against HepG2 cells and the IC50 value was 2.0 ± 0.4 μg/mL. During the cell cycle analysis in the G0/G1 phase, annphenone compound arrested the HepG2 cells when detecting the immunocytochemistry. It is suggested that the annphenone compound inhibits the catenin expression induced by the localization transfer, reducing the cyclin D1 protein expression. Furthermore, annphenone’s interaction as a possible ligand of the ASGP-R asialoglycoprotein receptor using a molecular docking simulation revealed its selectivity for hepatocellular carcinoma cells and potentially specific for antiproliferative activity . Annphenone is a promising anti-tumour agent present in the aqueous extract that reduced the contact sensitivity via down-regulating the adhesion, activation, and production of metalloproteinase T-lymphocytes in mice , whereas the ethanolic extract exerted anti-sepsis activity through down-regulation of the NF-kB and MAPK pathways . Cirsilineol (4′,5-dihydroxy-3′,6,7-trimethoxyflavone) found in A. vestita extracts possesses potent anti-tumour and immune suppressive properties . Cirsilineol significantly inhibited the proliferation of multiple types of cancer cells (Skov-3, PC3, Caov-3, and Hela cells) in a concentration-dependent manner. It induced apoptosis in Caov-3 cells in a dose-dependent manner, which was determined with annexin V/propidium iodide double staining. To promote apoptosis, cirsilineol activates caspase-9, caspase-3, and PARP (poly ADP-ribose polymerase). Cirsilineol-induced loss of the mitochondrial membrane potential (MMP) brings a remarkable change and releases cytochrome-c to the cytosol. The induction of apoptosis via the mitochondrial pathway is the mode of action for the anti-proliferative activity of cirsilineol against cancerous cells. Moreover, cirsilineol is effective in ameliorating TNBS (tri-nitrobenzene sulfonic acid) -induced experimental colitis in mice, possibly because of its novel immunoregulatory activities with selective inhibitor IFN-γ/STAT1/T-bet signaling in the colonic lamina propria CD4 + T-cells particular for Crohn’s disease . Regulation of the transcription activator and signal transducer (STAT 1) is being explored for the treatment of bowel diseases. However, few chemicals have been reported for the inhibition of STAT1/ IFN-g signaling for the treatment of Crohn’s disease. A natural compound, cirsilineol, isolated from the A. vestita plant significantly ameliorated TNBS (trinitro-benzene sulphonic acid)-induced T-cell-mediated mice colitis. It is closely associated with reduced auto-reactive T-cell activation and proliferation. Moreover, the action of anti-inflammatory and pro-inflammatory cytokines with cirsilineol therapy was found to increase the regulatory T-cell activity and decrease the effector Th-1 cell activity, as characterized by the up-regulation of TGF-b and IL-10 and down-regulation of IFN-g. Importantly, in the presence of a higher level of IFN-g the inhibition by compound cirsilineol of STAT1/IFN-g signalling seems reversible, suggesting that compound cirsilineol might be a potential candidate for the treatment of human inflammatory T-cell-mediated bowel diseases . The flavone jaceosidin isolated from A. vestita showed an antiproliferation effect on several human cancer cell lines. It significantly reduced SKOV-3, PC3, HeLa, and CAOV-3 cell proliferation in a concentration-dependent manner, whereas CAOV-3 showed time-dependent inhibition, and apoptosis increased in CAOV-3 cells. Flavone induced the cleavage of PARP (poly ADP-ribose polymerase) and caspase-3 and increased the cleaved caspase-9 levels. It also elevated the cytochrome c level in the cytosol, which shows the antitumor property of jaceosidin . Herbal flora is gaining much attention from scientists for the development of strategies and to know the therapeutic potential of novel herbal constituents to treat various health disorders. The information regarding the use of A. vestita as a folkoric medicine was mostly confined to the native inhabitants. The phytochemicals reported in A. vestita belong to the chemical classes of flavonoids, terpenoids, oxygenated monoterpenes, triterpenes, sesquiterpenes, hydroxyl cinnamic acids, mono- and diterpene hydrocarbons, aromatic aldehydes, azulenes, sesquiterpene hydrocarbons, sterols, phenylpropanoids, monoterpenoids, coumarins, and organosulfonic acids. The phenolic compounds present in the oil are responsible for the antioxidant activity, whereas flavones and sesquiterpenes exhibit anti-tumor and anti-inflammatory activity, respectively. The anti-inflammatory activity of A. vestita leaf extracts hhasave been time-tested. A. vestita can also be a promising source of anti-COVID-19 remedies. The inverse correlation between the antiviral activity of artemisinin contents and total flavonoid contents is reported. Artemisinin singly or in combination with other components acts synergistically to block the post-entry viral infection. Moreover, essential oil and extract from the pre- and post-flowering stages of A. vestita have been reported to possess antifungal activity against phytopathogenic fungi and can be an effective drug against dermatophytes. Although A. vestita has been used for the treatment of numerous ailments, standardization and extensive clinical studies shall facilitate optimizing the dose. Moreover, the plant is not yet extensively explored by the scientific community, as the successful reports related to the efficacy of its extracts are fragmentary and show variable results. Research related to the development of A. vestita -based value-added products and the enhancement of the efficacy of the extracts by blending with other natural extracts is also required. Being a reservoir of phytochemicals, it is necessary to conserve this species for further research so as to gain the associated medicinal benefits for health security. |
Allii Macrostemonis Bulbus: A Comprehensive Review of Ethnopharmacology, Phytochemistry and Pharmacology | 8d2af4be-f6ff-4eae-979b-ec497ab1babe | 10054501 | Pharmacology[mh] | AMB is a traditional Chinese herb with homology of medicine and food, named “薤白” in China. The 2020 edition of the ChP includes two basal plants of AMB, A. macrostemon and A. chinense , with Chinese herb names of “XiaoGenSuan” and “Xie”, respectively . The effect of AMB in traditional Chinese medicine (TCM) is to activate Yang and remove stasis, regulate Qi and eliminate stagnation. It is used to treat chest stuffiness and pains, distention and fullness, stomachache, diarrhea with rectal heaviness, headache, toothache and blood stasis. The mainly components in AMB include steroidal saponins, flavonoids, phenylpropanoids, alkaloids, amino acids, volatile oils, polysaccharides, organic acids and inorganic elements. Modern pharmacological studies show that AMB has effects including anti-platelet aggregation, hypolipidemic, hypoglycemic, antioxidant, cough and asthma, antibacterial, antitumor, antidepressant, etc. . At present, there are many studies on the effects of crude extracts or components of AMB on the treatment of chest pain and diarrhea, but there are few studies on its monomeric activity, quality and safety evaluation, which need to be further studied. Therefore, the literature on AMB should be reviewed and summarized to provide a theoretical basis for further research, expand its application and give full play to its therapeutic effects, so as to better serve human health. We conducted literature retrieval of AMB using electronic databases, including PubMed, CNKI, Web of Science, SpecialSciDBS, GBIF, Elsevier, and used national pharmaceutical standards, ancient Chinese medical classics, monographs on TCM, and academic papers to conduct a comprehensive analysis and summary. We used Allii Macrostemonis Bulbus, Allium macrostemon Bunge, Allium chinense G. Don, phytochemistry, steroidal saponins, pharmacological activity, anti-platelet aggregation, anti-atherosclerosis, cardiomyocytes, CHD, anti-cancer, antioxidant and antibacterial as keywords to review the information about botany, ethnopharmacology, phytochemistry, pharmacology, quality control and toxicology studies of AMB. Most of the Allium spp. of the family Liliaceae are distributed in the Northern Hemisphere, mainly in the Asia region, with about 660 species. Of these, 138 species grow in China, including 50 endemic varieties and 5 introduced varieties. Most varieties grow in arid areas, but a few species grow in ditch-side forests or watery meadows. A. macrostemon is distributed in all provinces and regions of China (except Xinjiang and Qinghai), mainly on mountain slopes, hills, valleys or grasslands at altitudes below 1500 m, and a few on mountain slopes at altitudes up to 3000 m (Yunnan and Tibet), and also in Russia, Korea and Japan. A. chinense is widely cultivated in the Yangtze River basin and provinces and regions south of China. It is also cultivated in Japan, Vietnam, Laos, Norway, and the United States . A. macrostemon and A. chinense are seasonal wild vegetables; their leaves are typically eaten in late spring and early summer, while the bulbs hidden underground are savored in late summer and early autumn. They are very similar in appearance and morphology. Both of them usually have 2–5 hollow leaves and cylindrical scapes; the involucre is 2-lobed, with umbels. Both have depressed nectaries and are covered at the base by cap-like projections; the styles extend beyond the perianth. The differences between the two in terms of plant appearance and morphology are shown in and . AMB was first recorded as a medicinal herb to treat weapon injuries and anti-fatigue in Shennong Bencao Jing compiled in the Eastern Han Dynasty; in the Tang Dynasty’s Qianjin Yi Fang and Bencao Shiyi , AMB was used to treat chest paralysis and heart pain, and to stop diarrhea and remove dysentery. With the development of the times and the advancement of science, AMB is also considered to be useful in the treatment of CHD, sudden death, nodules, stroke, burns, diarrhea, dysentery, cough and asthma, calming the fetus, and detoxification. From ancient times to the present, the concoction of AMB has also undergone a process from simple to complex . Since ancient times, AMB has been used in several formulas, as shown in , the most famous of which are the classical formulas mentioned in Zhang Zhongjing’s “ Jin Gui Yao Lue ” during the Eastern Han Dynasty, including Gualuo Xiebai Baijiu Decoction, Gualou Xiebai Banxia Decoction and Zhishi Xiebai Guizhi Decoction. Nowadays, these classical formulas of AMB as the “monarch drug” have been developed into proprietary Chinese patent medicines for clinical application. In addition, many proprietary Chinese patent medicines containing AMB have been developed in different dosage forms, such as Xuezhitong-capsules (XZT), Xiebai-powder, Tongxiening -granules, Dan-Lou-tablets, and Zhenxintong-oral liquid. AMB is extremely rich in phytochemicals and has been shown to contain steroidal saponins, flavonoids, phenylpropanoids, alkaloids, volatile oils, polysaccharides, organic acids, amino acids, etc. Different methods of preparation and extraction have a great influence on the content of active ingredients in AMB, and can even change its physicochemical properties, thus affecting the therapeutic effect . 5.1. Steroids and Steroidal Saponins Steroids are a general term for compounds with a steroid parent nucleus, i.e., a cyclopentanoperhydrophenanthrene carbon skeleton. The physiological function of steroid compounds depends on the type and number of functional groups attached to the core ring and the configuration of the positions . Steroid saponins are one of the main active substances in AMB; the parent nucleus is mainly of two types, spirostanol and furostanol, and the sugar part is mainly glucose, galactose, xylose, arabinose, and other monosaccharides. The sugar chain is usually attached to the C-3 position of spirostanol saponins, C-3 and C-6 positions of furostanol saponins, and to the C-1, C-6, C-12, and C-24 positions of steroid saponins; their structural diversity contributes to their wide range of pharmacological activities. Since the isolation of furostanoside and chinenoside I ( 54 ) from A. chinense in 1989, a total of 89 steroidal saponins have been isolated and obtained, including spirostanosides ( 1 – 28 ) and furostanosides ( 29 – 89 ), in addition to pregnane glycoside ( 90 ) and cholestane glycosides ( 91 – 92 ), sitosterol ( 93 ), stigmasterol ( 94 ), daucosterol ( 95 ), sitosteryl-6’-O-undecane-β-D-glucoside ( 96 ), etc. The structures are shown in and . 5.2. Volatile Oils and Sulfur-Containing Components The special odor of AMB originates from the sulfur-containing compounds in the volatile oil, which constitute over 50% . Most of the sulfur-containing compounds contain 1–5 S atoms in their molecules, characterized by the combination of different aliphatic side chains or rings on the sulfur skeleton. Some scholars used GC-MS to analyze the volatile oil of AMB and identified 14 chemical components, of which sulfur-containing compounds accounted for 93.46% . Interestingly, the composition and proportion of sulfur-containing compounds identified in the volatile oil of AMB from different origins varied considerably, which may be related to the origin of AMB, but all contained methyl allyl trisulfide ( 139 ) . In addition, there were differences in the chemical composition of volatile oils and their relative contents before and after AMB concoction. A total of 13 and 20 compounds were identified in the bulbs and leaves of fresh AMB, accounting for 62.5% and 59.63% of the total volatile oils, respectively; a total of 9 and 13 compounds were identified in the bulbs and leaves of AMB dried in an oven at 50 °C after steaming, accounting for 74.89% and 87.66% of the total, respectively . The structures of the sulfur-containing compounds are shown in and . 5.3. Nitrogen-Containing Components Nitrogen-containing compounds are also one of the main active substances in AMB. Adenosine ( 155 ) has been developed as an antiarrhythmic drug and was approved for use by the FDA in 1989. Adenosine ( 155 ) is present in large amounts in AMB and has strong platelet inhibitory activity ; therefore, the development of anti-arrhythmic drugs that are associated with AMB can be considered. In addition, endogenous nucleosides similar to adenosine ( 155 ) were identified, including thymidine ( 156 ) and guanosine ( 157 ), and other active ingredients were N- trans -feruloyltyramine ( 161 ), N-( p - cis -coumaroyl)-tyramine ( 163 ) and its trans -enantiomer ( 162 ), 2,3,4,9-tetrahydro-1H-pyrido [3, 4-b]indole-3-carboxylic acid ( 158 ) and its 1-methylated product ( 159 ) and tryptophan ( 160 ), etc. . In addition, AMB is rich in many free amino acids, including 19 common protein amino acids such as arginine, threonine, serine, and 4 non-protein amino acids . The structures of the nitrogen-containing compounds are shown in and . 5.4. Phenylpropanoids Phenylpropanoids are a naturally occurring class of compounds consisting of a benzene ring linked to three straight chain carbons (C6–C3 groups). In biosynthesis, most of these compounds are formed from anthranilic acid through a series of reactions such as deamination and hydroxylation by aromatic amino acids such as phenylalanine and tyrosine. Phenylpropanoids found in AMB include acanthoside D ( 164 ) , syringin ( 165 ) , In the leaves of AMB allimacronoid A ( 166 ) allimacronoid B ( 167 ), allimacronoid C ( 168 ), allimacronoid D ( 169 ), tuberonoid A ( 170 ), 1-O-(E)-feruloyl-β-D gentiobioside ( 171 ), 1-O-(E)-feruloyl-β-D-glucopyranoside ( 172 ), and trans -ferulic acid ( 173 ) . The structures of phenylpropanoid compounds are shown in and . 5.5. Flavonoids Flavonoids are a general term for a class of compounds derived from 2-phenylchromone as a backbone. Flavonoids in AMB are mainly flavonol glycosides and chalcones, including kaempferol-3-O-β-D-glucoside ( 174 ), kaempferol-3,7-O-β-D-diglucoside ( 175 ), kaempferol-3,4’-O-β-D-diglucoside ( 176 ), quercetin-3-O-β-D-glucoside ( 177 ), isorhamnetin-3-O-β-D-glucoside ( 178 ), isoliquiritigenin ( 179 ) and isoliquiritigenin-4-O-glucoside ( 180 ) . The structures of the flavonoids are shown in and . 5.6. Polysaccharides Polysaccharides are polymers of multiple monosaccharides linked by glycosidic bonds and are classified as homopolysaccharides and heteropolysaccharides. AMB contains a large number of polysaccharides. One study conducted acid hydrolysis tests on the three refined polysaccharides PAM-Ib, PAM-IIa and PAM-III’ from AMB and showed that all three polysaccharides contained galactose and glucose . Another study used enzymatic hydrolysis of AMB polysaccharides, and the results showed that the monosaccharides included arabinose, glucose, rhamnose, and galactose . Both AMP40N and AMP40S are polysaccharides isolated from AMB; AMP40N consists of arabinose and glucose, while AMP40S consists of rhamnose, arabinose, glucose and galactose and a certain amount of uridine monophosphate . It can be seen that there are great differences in the monosaccharide composition, glycosidic bond type, uronic acid content and properties of AMB polysaccharides obtained by different extraction methods, but most of them are polymerized with glucose, galactose, rhamnose and arabinose. Due to the complexity of polysaccharide structure and the limitation of research means, the research into polysaccharides lags far behind other types of compounds, and only some of the fungus polysaccharides are used in clinical practice. Therefore, the research on polysaccharide components in AMB should be increased, and the relationship between structure and function of AMB polysaccharides and their mechanism of action in vivo should be dissected. 5.7. Other Components Other compounds isolated from AMB include (3β, 4α)-Olean-12-en-28-oic acid-3-O-β-D-galactopyranosyloxy-23-hydroxy-6-O-β-D-xylopyranosyl-β-D-galactopyranosyl ester ( 181 ), prostaglandin A1 ( 182 ), prostaglandin B1 ( 183 ), 2-ene-butanol ( 184 ), ethyl acetate ( 185 ), limonene ( 186 ) and several fatty acid analogues, including succinic acid ( 187 ), tetradecanoic acid ( 188 ), oleic acid ( 189 ), palmitoleic acid ( 190 ), palmitic acid ( 191 ) and linoleic acid ( 192 ) , whose structures are shown in and . Steroids are a general term for compounds with a steroid parent nucleus, i.e., a cyclopentanoperhydrophenanthrene carbon skeleton. The physiological function of steroid compounds depends on the type and number of functional groups attached to the core ring and the configuration of the positions . Steroid saponins are one of the main active substances in AMB; the parent nucleus is mainly of two types, spirostanol and furostanol, and the sugar part is mainly glucose, galactose, xylose, arabinose, and other monosaccharides. The sugar chain is usually attached to the C-3 position of spirostanol saponins, C-3 and C-6 positions of furostanol saponins, and to the C-1, C-6, C-12, and C-24 positions of steroid saponins; their structural diversity contributes to their wide range of pharmacological activities. Since the isolation of furostanoside and chinenoside I ( 54 ) from A. chinense in 1989, a total of 89 steroidal saponins have been isolated and obtained, including spirostanosides ( 1 – 28 ) and furostanosides ( 29 – 89 ), in addition to pregnane glycoside ( 90 ) and cholestane glycosides ( 91 – 92 ), sitosterol ( 93 ), stigmasterol ( 94 ), daucosterol ( 95 ), sitosteryl-6’-O-undecane-β-D-glucoside ( 96 ), etc. The structures are shown in and . The special odor of AMB originates from the sulfur-containing compounds in the volatile oil, which constitute over 50% . Most of the sulfur-containing compounds contain 1–5 S atoms in their molecules, characterized by the combination of different aliphatic side chains or rings on the sulfur skeleton. Some scholars used GC-MS to analyze the volatile oil of AMB and identified 14 chemical components, of which sulfur-containing compounds accounted for 93.46% . Interestingly, the composition and proportion of sulfur-containing compounds identified in the volatile oil of AMB from different origins varied considerably, which may be related to the origin of AMB, but all contained methyl allyl trisulfide ( 139 ) . In addition, there were differences in the chemical composition of volatile oils and their relative contents before and after AMB concoction. A total of 13 and 20 compounds were identified in the bulbs and leaves of fresh AMB, accounting for 62.5% and 59.63% of the total volatile oils, respectively; a total of 9 and 13 compounds were identified in the bulbs and leaves of AMB dried in an oven at 50 °C after steaming, accounting for 74.89% and 87.66% of the total, respectively . The structures of the sulfur-containing compounds are shown in and . Nitrogen-containing compounds are also one of the main active substances in AMB. Adenosine ( 155 ) has been developed as an antiarrhythmic drug and was approved for use by the FDA in 1989. Adenosine ( 155 ) is present in large amounts in AMB and has strong platelet inhibitory activity ; therefore, the development of anti-arrhythmic drugs that are associated with AMB can be considered. In addition, endogenous nucleosides similar to adenosine ( 155 ) were identified, including thymidine ( 156 ) and guanosine ( 157 ), and other active ingredients were N- trans -feruloyltyramine ( 161 ), N-( p - cis -coumaroyl)-tyramine ( 163 ) and its trans -enantiomer ( 162 ), 2,3,4,9-tetrahydro-1H-pyrido [3, 4-b]indole-3-carboxylic acid ( 158 ) and its 1-methylated product ( 159 ) and tryptophan ( 160 ), etc. . In addition, AMB is rich in many free amino acids, including 19 common protein amino acids such as arginine, threonine, serine, and 4 non-protein amino acids . The structures of the nitrogen-containing compounds are shown in and . Phenylpropanoids are a naturally occurring class of compounds consisting of a benzene ring linked to three straight chain carbons (C6–C3 groups). In biosynthesis, most of these compounds are formed from anthranilic acid through a series of reactions such as deamination and hydroxylation by aromatic amino acids such as phenylalanine and tyrosine. Phenylpropanoids found in AMB include acanthoside D ( 164 ) , syringin ( 165 ) , In the leaves of AMB allimacronoid A ( 166 ) allimacronoid B ( 167 ), allimacronoid C ( 168 ), allimacronoid D ( 169 ), tuberonoid A ( 170 ), 1-O-(E)-feruloyl-β-D gentiobioside ( 171 ), 1-O-(E)-feruloyl-β-D-glucopyranoside ( 172 ), and trans -ferulic acid ( 173 ) . The structures of phenylpropanoid compounds are shown in and . Flavonoids are a general term for a class of compounds derived from 2-phenylchromone as a backbone. Flavonoids in AMB are mainly flavonol glycosides and chalcones, including kaempferol-3-O-β-D-glucoside ( 174 ), kaempferol-3,7-O-β-D-diglucoside ( 175 ), kaempferol-3,4’-O-β-D-diglucoside ( 176 ), quercetin-3-O-β-D-glucoside ( 177 ), isorhamnetin-3-O-β-D-glucoside ( 178 ), isoliquiritigenin ( 179 ) and isoliquiritigenin-4-O-glucoside ( 180 ) . The structures of the flavonoids are shown in and . Polysaccharides are polymers of multiple monosaccharides linked by glycosidic bonds and are classified as homopolysaccharides and heteropolysaccharides. AMB contains a large number of polysaccharides. One study conducted acid hydrolysis tests on the three refined polysaccharides PAM-Ib, PAM-IIa and PAM-III’ from AMB and showed that all three polysaccharides contained galactose and glucose . Another study used enzymatic hydrolysis of AMB polysaccharides, and the results showed that the monosaccharides included arabinose, glucose, rhamnose, and galactose . Both AMP40N and AMP40S are polysaccharides isolated from AMB; AMP40N consists of arabinose and glucose, while AMP40S consists of rhamnose, arabinose, glucose and galactose and a certain amount of uridine monophosphate . It can be seen that there are great differences in the monosaccharide composition, glycosidic bond type, uronic acid content and properties of AMB polysaccharides obtained by different extraction methods, but most of them are polymerized with glucose, galactose, rhamnose and arabinose. Due to the complexity of polysaccharide structure and the limitation of research means, the research into polysaccharides lags far behind other types of compounds, and only some of the fungus polysaccharides are used in clinical practice. Therefore, the research on polysaccharide components in AMB should be increased, and the relationship between structure and function of AMB polysaccharides and their mechanism of action in vivo should be dissected. Other compounds isolated from AMB include (3β, 4α)-Olean-12-en-28-oic acid-3-O-β-D-galactopyranosyloxy-23-hydroxy-6-O-β-D-xylopyranosyl-β-D-galactopyranosyl ester ( 181 ), prostaglandin A1 ( 182 ), prostaglandin B1 ( 183 ), 2-ene-butanol ( 184 ), ethyl acetate ( 185 ), limonene ( 186 ) and several fatty acid analogues, including succinic acid ( 187 ), tetradecanoic acid ( 188 ), oleic acid ( 189 ), palmitoleic acid ( 190 ), palmitic acid ( 191 ) and linoleic acid ( 192 ) , whose structures are shown in and . Studies have shown that crude extracts of AMB, monomeric components (e.g., macrostemonosides), and their compound preparations exert various pharmacological activities. Some of the pharmacological mechanisms are shown in . 6.1. Anti-Platelet Aggregation Effect Adhesion, aggregation and secretion are the three basic functions of platelets. Excessive platelet activation caused by pathological factors can promote platelet aggregation, which can cause thrombotic disease . In recent years, much attention has been paid to the role of platelet-associated inflammation in the pathogenesis of coronary artery disease. The release of CD40L after platelet activation and adhesion between platelets and neutrophils is one of the initiating links of thrombosis . Recent studies have suggested that platelets are involved in hemostasis and thrombosis, but also secrete various inflammatory factors such as adhesion molecules (Intercellular adhesion molecule-2), P-selectin and its ligand (P-selectin glycoprotein ligand-1), which have a direct chemotactic effect on leukocytes in blood vessels and regulate the development of inflammation . Inflammation contributes to vulnerable plaque thrombosis and plays an important role in the pathogenesis of acute coronary syndrome (ACS). It was found that steroidal saponins in AMB inhibit platelet CD40L expression and platelet neutrophil adhesion . AMB saponins inhibit arachidonic acid (AA), adenosine diphosphate (ADP) and platelet activation factor (PAF) induced platelet aggregation in a concentration-dependent manner in vitro and in vivo, reduce intra-platelet calcium ion concentration and adhesion between neutrophils and thrombin-activated platelets, and inhibit platelet aggregation induced by neutrophil supernatant . N- trans -feruloyltyramine ( 158 ), isolated from AMB, showed significant inhibition of both the first and second phases of ADP-induced human platelet aggregation, whereas N-( p-cis -coumaroyl)-tyramine ( 160 ) inhibited only the first phase of aggregation . Furosterosides in AMB reduce cardiomyocyte injury in SD rats both in vitro and in vivo by inhibiting platelet phosphatidylinositol 3-kinase/proteinserine-threonine kinase (PI3K/Akt) signaling pathway and thereby inhibiting ADP-induced platelet aggregation . Methyl allyl trisulfide ( 136 ), a sulfur-containing compound in AMB, showed strong inhibition of platelet aggregation activity . Given the relationship between platelets and inflammatory factors, it is suggested that the relationship between the pharmacological effects of AMB and inflammation is also one of the directions worth investigating. 6.2. Hypolipidemic and Anti-Atherosclerotic Effects Atherosclerosis (AS) is a chronic inflammatory disease caused by impaired lipid metabolism, usually forming plaques in medium and large arteries , and is a major cause of the development of CHD and cerebral vascular accident (CVA) . The accumulation of macrophages under the endothelium is thought to be the first step in the formation of AS, and over time, atherosclerotic plaques become more fibrotic and cause calcium deposits, which can eventually invade the lumen and lead to the development of ischemic disease . Mammalian target of rapamycin (m TOR) is a serine/threonine protein kinase found in mammals and has important roles in cell proliferation, survival, metabolism, autophagy, apoptosis, migration, and other biological processes. Several studies have shown that m TOR activation triggers endothelial dysfunction, foam cell formation, and vascular smooth muscle cell proliferation, thereby promoting the development and progression of AS . Furthermore, in the early stages of atherosclerosis, low-density lipoprotein (LDL) is retained in the intima and is modified to form multiple danger-associated molecular patterns (DAMP), mediated by oxidases, lipolytic enzymes, protein hydrolases, and reactive oxygen species, thereby acquiring immunogenicity , and immunogenic LDL activates vascular endothelial cells. Vascular endothelial cells regulate the structure and function of blood vessels by releasing biochemical factors such as nitric oxide (NO) and prostaglandin I2 (PGI2) . It was found that AMB total saponin and volatile oil extract could significantly reduce serum and liver total cholesterol (TC), triglyceride (TG), and LDL levels, and increase serum high-density lipoprotein (HDL) levels in rats on a high-fat diet, thus exerting a hypolipidemic effect . One of the possible mechanisms for AMB to lower lipids and prevent atherosclerosis is to increase the levels of PGI2 and PGE1 on the one hand and to interfere with AA metabolism and inhibit thromboxane A 2 (TXA 2 ) synthesis, on the other hand, thus changing the PGI2/TXA2 ratio and relieving the hypercoagulable state of blood . Another study showed that 10% AMB powder added to the high-fat diet of an animal with hyperlipidemia could upregulate the mRNA expression levels of low-density lipoprotein receptor (LDLR) and liver X receptor alpha (LXRα) in liver tissue, thus exerting its hypolipidemic effect . Macrostemonoside A ( 1 ) is a steroidal saponin isolated from AMB, which can reduce TC, TG, and LDL levels in mice serum and blood glucose levels in mice, and increase visfatin protein expression in 3T3-L1 cells . XZT is a proprietary Chinese patent medicine made from AMB extract. Studies have shown that XZT reduces fatty acid synthase (FAS) and LDL levels in the serum of ApoE −/− mice by activating reverse cholesterol transport (RCT) and increasing HDL levels, and that XZT reduces TG levels in patients with hyperlipidemia . The m TOR signaling pathway plays an important role in the progression and treatment of CHD. m TOR is mostly associated with cellular autophagy and apoptosis, and previous studies have demonstrated that autophagy has a dual role in atherosclerosis. The body needs moderate autophagy to stabilize plaque and inhibit excessive autophagy during cardiac I/R injury to reduce myocardial infarct size. Most of the monomeric components of TCM for the treatment of CHD are purified from blood-stasis-activating and qi-supplementing drugs, but the mechanisms of pharmacological effects of qi-activating drugs (e.g., AMB) and expectorants (e.g., Fructus Trichosanthis and Pinellia Tuber), which are commonly used in the clinical treatment of CHD, have been less studied, and research on the mechanisms of active components of these herbs should be strengthened. 6.3. Protection of Cardiomyocytes and Vascular Endothelial Cells Myocardial ischemia is the result of an imbalance in oxygen supply and demand to myocardial cells, and early hemodialysis is the most effective way to reduce post-ischemic myocardial injury . With the development of the application of interventions such as percutaneous coronary intervention, coronary artery bypass grafting, and thrombolysis, the myocardium can be resupplied with blood after ischemia, but the ensuing myocardial ischemia-reperfusion injury is a complex pathophysiological process involving multiple factors. The mechanism is currently believed to be closely related to inflammation, oxidative stress, vascular endothelial cell damage, platelet aggregation, and other factors, which can eventually lead to irreversible apoptosis or necrosis . Early reperfusion therapy can aggravate the myocardial injury and become an important factor affecting the outcome of ischemic therapy. The assessment and treatment of reperfusion injury remain a clinical challenge, and the causal mechanism is still unclear. One mechanism that has been identified is that ischemia-reperfusion triggers endothelial cell dysfunction and disrupts the endothelial structure of the blood vessels, thereby impeding blood circulation within the microvasculature. Endothelial cells are not only found in the lining of blood vessels but also cover the heart and lymphatic lumen longitudinally in a single layer, playing an important role in normal cardiac physiology and cardiac response to injury. Endothelial cells also act as secretory cells, secreting vasoactive substances, such as the vasoconstrictors endothelin (ET) and angiotensin, and vasodilators such as NO and endothelial-dependent hyperpolarizing factor (EDHF). They play an important role in regulating the tone of blood vessels, especially microcirculatory vessels; they can also synthesize and secrete relevant coagulation factors and fibrinolytic substances to maintain a dynamic balance between coagulation and fibrinolysis and influence the coagulation and fibrinolysis process, thus maintaining normal blood flow and circulation . It was found that AMB extract reduced the gene expression of inflammation-related cyclooxygenase-2 (COX-2), cyclooxygenase-1 (COX-1), inducible nitric oxide synthase (iNOS), and vasodilation-related endothelin-converting enzyme (ECE), and endothelial nitric oxide synthase (eNOS), but increased the gene expression of antioxidant superoxide dismutase (SOD) in a model of air-stressed vascular endothelial injury, thereby reducing endothelial vascular damage in model rats . At the same time, AMB extract also significantly reduced plasma ET level, increased serum NO level, and inhibited glucose-regulated protein 78 (GRP78) protein expression in aortic tissue to improve vascular endothelial function in model rats by suppressing endoplasmic reticulum stress . In a rat model of acute myocardial ischemia caused by open-chest ligation of the anterior descending branch of the rats’ left coronary artery, ethanolic extract of AMB can regulate the balance of lipid and protein metabolism and reduce the damage caused by acute myocardial ischemia in the rat organism . AMB extract also significantly increased serum glutathione peroxidase (GSH-Px) activity; it decreased acetylcholinesterase (TChE) activity, non-esterified fatty acid (NEFA), and malondialdehyde (MDA) content, and reduced the extent of myocardial injury in rats . In addition, AMB extracts could protect vascular endothelial function in depressed rats by enhancing 5-hydroxytryptamine 1D (5-HT 1D ) mRNA and protein expression, which mediates the diastolic effect, and inhibiting 5-hydroxytryptamine 2A (5-HT 2A ) mRNA and protein expression, which mediates the vasoconstrictive effect . 6.4. Anti-Cancer Effect In medicine, cancer is defined as a malignant tumor often originating from epithelial tissue, which is the most common type of malignancy. Globally, cancer has become the leading cause of human death and a serious obstacle to increasing human life expectancy . Today, global cancer incidence and mortality rates are increasing every year, with 28.4 million cancer cases expected in 2040 . The anti-cancer activity of AMB is mainly related to the water-soluble saponins, polysaccharides, and fat-soluble volatile oils contained in it. Reports have illustrated that the active components in AMB have been effective against human non-small cell lung cancer A549 , human lung cancer cells PC-9 , mice sarcoma cells S180 , mice liver cancer cells H22 , human gastric cancer cell SGC-7901 , human breast cancer MCF-7 , human neural cancer cell SF-268 , human lung cancer cells NCI-H460 , human cervical cancer HeLa cells , human colon cancer cells SW-480 , mice melanoma cells B16 , mice breast cancer cells 4T1 , human hepatoma cells Hep-3B , human hepatoma cells HepG2 , human lung adenocarcinoma cell SPC-A-1 , human gastric cancer cell MGC80-3 , human breast cancer cell MDA-MB-231 , human colon cancer cell SW620 and human nasopharyngeal carcinoma cells CNE-1 , which were inhibited in vivo or in vitro. Possible mechanisms of action include: regulation of EGFR/PI3K/m TOR and RAF/MAPK signaling pathways ; inhibition of tumor cell membrane phospholipid synthesis ; enhancement of immune function in mice, especially cellular immune function, which is dominant in tumor immunity, and thus suppression of tumor cells ; directly killing tumor cells by destroying nuclei and organelles ; altering the G 2 /M cell cycle of tumor cells ; promoting the expression of P53 protein to induce apoptosis ; decreasing mitochondrial membrane potential; up-regulating Bax mRNA expression, down-regulating Bcl-2 mRNA expression, and Bcl-2/Bax ratio; enhanceing Caspase-9 and Caspase-3 activity; inducing reactive oxygen species (ROS) production, and promoting apoptosis of tumor cells . 6.5. Antibacterial Effect The extracts of AMB have inhibitory effects on a variety of bacteria and fungi. It was found that the aqueous extract of AMB had a wide range of antibacterial abilities, and the antibacterial ability varied at different dilutions of the extracts, with a more desirable effect at higher concentrations, and weaker effect at higher dilutions . In addition, the ethanol extract of AMB also has an inhibitory effect on most bacteria, and the inhibition ability is influenced by temperature and pH. The strongest inhibition activity is at 50–60 °C and the activity decreases when the temperature is greater than 100 °C. The inhibition activity is stronger when the pH is neutral or nearly neutral, and the inhibition activity gradually decreases with the enhancement of acidity or alkalinity . AMB may exert its bacterial inhibitory effect by inhibiting the synthesis of bacterial-associated proteins, inhibiting the activity of related enzymes, or changing their cell structure . The material basis of these mechanisms may be related to the sulfur-containing compounds in AMB, and the specific mechanism of action needs to be investigated in depth. 6.6. Anti-Asthmatic Effect Asthma, as a chronic inflammatory disease of the respiratory tract, is one of the most common non-communicable diseases of the respiratory system in children and adults, often caused by allergic reactions. Stimuli such as histamine, acetylcholine, or cold air can cause airway hyperreactivity and produce airway obstruction, which can clinically cause recurrent episodes of wheezing, chest tightness, or coughing . Typical asthma pathology is characterized by airway inflammation, smooth muscle contraction, epithelial cell shedding, excessive mucus secretion, bronchial hyperresponsiveness, and mucosal edema . Standard therapies for asthma are mainly based on bronchodilators and immunosuppressive drugs, which provide short-term relief but not a cure. Chinese medicine has played an important role in the treatment of various respiratory diseases, including asthma, and has a history of more than 2000 years in the treatment of asthma. In recent years, more and more researchers have focused on the effects of Chinese medicine on asthma, and have achieved remarkable results in clinical trials or basic experimental models . Clinically, AMB can be used alone for the treatment of asthma, and in recent years, many studies have been conducted on the pharmacodynamic material basis of AMB for the treatment of asthma. It has been reported that in animal experiments, IL-6 mRNA content in the bronchial tissues of asthmatic guinea pigs was significantly increased . In clinical practice, serum levels of IL-6 are also significantly higher in asthmatics than in normal subjects . In addition, the balance of TXA 2 and PGI2 is an important regulatory mechanism in the pathophysiological mechanism of asthma, and if the ratio of TXA 2 /PGI2 is increased, it causes bronchial smooth muscle contraction leading to asthma; however, because of the instability of TXA2 and PGI2, the corresponding metabolites of both, thromboxane B 2 (TXB 2 ) and 6-keto-prostaglandin F 1α (6-Keto-PGF 1α ) are often measured . Studies have shown that AMB extract can reduce the expression levels of IL-6 and TXB 2 and up-regulate the expression level of 6-Keto-PGF 1α in the serum of asthmatic guinea pigs, thus achieving a panting effect . In vivo and in vitro, the active ingredients in AMB effectively diastole bronchial smooth muscle in a guinea pig model of histamine-induced asthma . In summary, we deduce that the mechanism by which AMB exerts its effect on wheezing may be through the inhibition of inflammatory response, alleviating chronic inflammation and thus relieving the spastic state of bronchial smooth muscle. 6.7. Antioxidant Effect ROS are oxygen-containing radicals with high oxidative capacity and high activity generated during metabolism, mainly including superoxide anion radical (O 2 − ), hydrogen peroxide (H 2 O 2 ), hydroxyl radical (·OH), etc. ROS are a double-edged sword for cellular life activities: on the one hand, ROS are important tools or signaling molecules in specific cells (such as macrophages, etc.) and play an important role in removing pathogenic microorganisms, maintaining the normal vascular function, and regulating intracellular homeostasis. On the other hand, when the excessive production of intracellular ROS exceeds the scavenging capacity of the antioxidant system in the body, they will attack proteins, DNA and lipids, causing oxidative stress, which is one of the important factors in the occurrence of cell damage, inflammation, and metabolic disorders . Antioxidant enzymes in the body mainly include SOD, GSH-Px, glutathione S-transferase (GST), catalase (CAT), etc. Non-enzyme antioxidants include glutathione, vitamin E, vitamin C, etc. SOD can effectively scavenge O 2 − , protect cells from oxidative damage, and also provide hydrogen atom ligands for the reduction of ROS to produce hydrogen peroxide, which in turn can be catalyzed by GSH-Px and CAT to produce water and oxygen to reduce oxidative stress damage . Oxidative stress is associated with multiple signaling pathway molecules. Nuclear factor erythroid 2-related factor 2 (Nrf2) is a basic leucine zipper transcription factor, and cytoplasmic Nrf2 is normally bound to Kelch-like ECH-associated protein-1. The free Nrf2 is able to translocate from the cytoplasm to the nucleus, where it forms a heterodimer with Maf family proteins and then binds to antioxidant response element sequences to induce the expression of downstream antioxidant enzymes, thereby scavenging ROS, inhibiting oxidative stress, maintaining the structural integrity and normal metabolic function of the cell, and exerting its transcriptional regulatory role . Nuclear factor kappa-B (NF-κB) is a dimeric protein of the Rel family. The heterodimer composed of p65 and p50 is a common activated form of NF-κB. NF-κB can promote the infiltration of neutrophils and macrophages and the release of cytokines, chemokines, adhesion molecules, etc., stimulate the expression and secretion of matrix metalloproteinases, activate nicotinamide adenine dinucleotide phosphate oxidase to produce large amounts of ROS, and trigger oxidative stress-related inflammatory diseases . Silent information regulator 1 (Sirt1) is a nicotinamide adenine dinucleotide-dependent deacetylase. Activated Sirt1 inhibits p66shc expression and reduces mitochondrial ROS production by regulating p66shc, which deacetylates histone H3 bound to the p66shc promoter . It is found that AMB extract alleviates liquor-induced oxidative stress in rats by increasing serum SOD and CAT activities and protecting T lymphocytes, and significantly inhibiting serum lipid peroxide formation . AMB polysaccharide, AMB saponin, and some sulfur-containing compounds can effectively scavenge DPPH, O 2 − and ·OH in vitro and inhibit the oxidation of Fe 2+ to a certain extent, and their antioxidant ability can be enhanced after modification with chlorosulfate-pyridine or α-amylase for AMB polysaccharide . Although there are many experimental studies on the antioxidant activity of various extracts of AMB, most of them are limited to in vitro experiments and the specific mechanism is not yet clear. The research efforts on oxidative stress signaling molecules should be deepened to elucidate the antioxidant mechanism of AMB at the molecular level. 6.8. Antidepressant Effect Depression is an affective disorder characterized by persistent mood abnormalities, mainly manifested as depressed mood, lack of pleasure, difficulty concentrating, fatigue, physical pain, and other symptoms, with a high disability rate and high patient suicide rate, which brings a serious economic burden to the patient’s family and society . The pathogenesis of depression has not yet been fully investigated and researchers have proposed various hypotheses, among which the monoamine transmitter theory suggests that the development of depression is mainly due to the reduction of 5-hydroxytryptamine (5-HT) and norepinephrine (NE) in the brain; therefore, inhibiting the degradation and reuptake of these two monoamines is beneficial to improve depressive symptoms . The neurotrophic factor hypothesis focuses on the brain-derived neurotrophic factor (BDNF) and suggests that an imbalance of brain derived neurotrophic factor precursor (proBDNF) and mature form of brain-derived neurotrophic factor (mBDNF) is closely related to the development of depression . The neurogenesis hypothesis suggests that downregulation of hippocampal neurogenesis is the cause of depression and that antidepressants work based on promoting neurogenesis . In addition, possible mechanisms such as the hypothalamic-pituitary-adrenal (HPA) axis dysregulation hypothesis, inflammation hypothesis, and genetic hypothesis have also been proposed to explain the development of depression . Depression is gradually becoming an important health problem faced by all human beings today, and its pathogenesis is complex. Although antidepressant western drugs are effective for patients with critical symptoms, they have more side effects in terms of mental and emotional effects when taken for a long time. Therefore, people gradually turn their horizons to Chinese medicine, but the composition of Chinese medicines is complex. It can be difficult to find the best component with significant efficacy among the complex and numerous components of compound medicines and single component treatments. The mechanism by which AMB exerts antidepressant effects on various animal models of depression (including rats and mice) may be through regulating the balance of the internal environment of depression model animals, promoting neurogenesis and BDNF production; at the same time, AMB can significantly improve the pathological changes of organ tissues in the relevant animal models . In addition, the analysis of lipids and acylcarnitine in the plasma of depressed rats by liquid chromatography/ion trap time of flight mass spectrometry and ultra-performance liquid chromatography/triple quadrupole mass spectrometry, respectively, showed that the AMB aqueous extract was able to restore the normal levels of these abnormally altered indicators . Although there are numerous studies on depression, the relevant mechanisms are still under-explained, and more rigorous experimental design is needed in the future, together with modern technology to reduce complex Chinese medicine into simpler groupings, purify components, or increase the study of mechanisms at the cellular-molecular level. It cannot be ignored, however, that Chinese medicine mostly follows a certain idea of combination, and it is necessary to maintain a cautious attitude whether the antidepressant components derived from the reductionist ideas of modern medicine can stand up to clinical tests. 6.9. Other Pharmacological Effects In addition to the above pharmacological effects, AMB and its compounds exhibit other activities such as analgesia, hypoxia tolerance, immunomodulation, promotion of osteogenesis, inhibition of hepatic drug enzymes, and mosquito control. Studies have shown that both the raw aqueous decoction of AMB and its fried aqueous decoction have strong analgesic effects and prolong the duration of hypoxia tolerance in mice with enhanced oxygen consumption induced by NaNO 2 intoxication and isoproterenol (ISO) under normoxic conditions. The mechanism of analgesia of AMB may be through the inhibition of voltage-sensitive Nav1.7 channels, thus reducing the excitability of peripheral neurons and exerting analgesic effects . AMB can increase the weight of mice’s immune organs, the spleen and thymus, and can increase carbon particle contouring index K and phagocytosis index α; that is, it can promote the phagocytosis of mononuclear macrophages and improve the specific immune function of the body. AMB volatile oil can increase the spleen index, macrophage phagocytosis rate and splenocyte proliferation index. The regulatory ability of AMB on the immune system may be one of the mechanisms of its anti-tumor effect . AMB alcohol extract can increase the expression of insulin-like growth factor-1 and bone morphogenetic protein-2, thus regulating the formation and resorption of bones and achieving the purpose of promoting bone growth . AMB aqueous extract can significantly reduce the content of cytochrome P450 in mice and has a significant inhibitory effect on hepatic drug enzymes . In addition, the volatile oil of AMB and its two main components (compounds 113 and 135 ) exhibited strong larvicidal effects against Aedes albopictus larvae, suggesting the existence of a basis for the development of mosquito control agents . The modern pharmacological studies on AMB are summarized in . Adhesion, aggregation and secretion are the three basic functions of platelets. Excessive platelet activation caused by pathological factors can promote platelet aggregation, which can cause thrombotic disease . In recent years, much attention has been paid to the role of platelet-associated inflammation in the pathogenesis of coronary artery disease. The release of CD40L after platelet activation and adhesion between platelets and neutrophils is one of the initiating links of thrombosis . Recent studies have suggested that platelets are involved in hemostasis and thrombosis, but also secrete various inflammatory factors such as adhesion molecules (Intercellular adhesion molecule-2), P-selectin and its ligand (P-selectin glycoprotein ligand-1), which have a direct chemotactic effect on leukocytes in blood vessels and regulate the development of inflammation . Inflammation contributes to vulnerable plaque thrombosis and plays an important role in the pathogenesis of acute coronary syndrome (ACS). It was found that steroidal saponins in AMB inhibit platelet CD40L expression and platelet neutrophil adhesion . AMB saponins inhibit arachidonic acid (AA), adenosine diphosphate (ADP) and platelet activation factor (PAF) induced platelet aggregation in a concentration-dependent manner in vitro and in vivo, reduce intra-platelet calcium ion concentration and adhesion between neutrophils and thrombin-activated platelets, and inhibit platelet aggregation induced by neutrophil supernatant . N- trans -feruloyltyramine ( 158 ), isolated from AMB, showed significant inhibition of both the first and second phases of ADP-induced human platelet aggregation, whereas N-( p-cis -coumaroyl)-tyramine ( 160 ) inhibited only the first phase of aggregation . Furosterosides in AMB reduce cardiomyocyte injury in SD rats both in vitro and in vivo by inhibiting platelet phosphatidylinositol 3-kinase/proteinserine-threonine kinase (PI3K/Akt) signaling pathway and thereby inhibiting ADP-induced platelet aggregation . Methyl allyl trisulfide ( 136 ), a sulfur-containing compound in AMB, showed strong inhibition of platelet aggregation activity . Given the relationship between platelets and inflammatory factors, it is suggested that the relationship between the pharmacological effects of AMB and inflammation is also one of the directions worth investigating. Atherosclerosis (AS) is a chronic inflammatory disease caused by impaired lipid metabolism, usually forming plaques in medium and large arteries , and is a major cause of the development of CHD and cerebral vascular accident (CVA) . The accumulation of macrophages under the endothelium is thought to be the first step in the formation of AS, and over time, atherosclerotic plaques become more fibrotic and cause calcium deposits, which can eventually invade the lumen and lead to the development of ischemic disease . Mammalian target of rapamycin (m TOR) is a serine/threonine protein kinase found in mammals and has important roles in cell proliferation, survival, metabolism, autophagy, apoptosis, migration, and other biological processes. Several studies have shown that m TOR activation triggers endothelial dysfunction, foam cell formation, and vascular smooth muscle cell proliferation, thereby promoting the development and progression of AS . Furthermore, in the early stages of atherosclerosis, low-density lipoprotein (LDL) is retained in the intima and is modified to form multiple danger-associated molecular patterns (DAMP), mediated by oxidases, lipolytic enzymes, protein hydrolases, and reactive oxygen species, thereby acquiring immunogenicity , and immunogenic LDL activates vascular endothelial cells. Vascular endothelial cells regulate the structure and function of blood vessels by releasing biochemical factors such as nitric oxide (NO) and prostaglandin I2 (PGI2) . It was found that AMB total saponin and volatile oil extract could significantly reduce serum and liver total cholesterol (TC), triglyceride (TG), and LDL levels, and increase serum high-density lipoprotein (HDL) levels in rats on a high-fat diet, thus exerting a hypolipidemic effect . One of the possible mechanisms for AMB to lower lipids and prevent atherosclerosis is to increase the levels of PGI2 and PGE1 on the one hand and to interfere with AA metabolism and inhibit thromboxane A 2 (TXA 2 ) synthesis, on the other hand, thus changing the PGI2/TXA2 ratio and relieving the hypercoagulable state of blood . Another study showed that 10% AMB powder added to the high-fat diet of an animal with hyperlipidemia could upregulate the mRNA expression levels of low-density lipoprotein receptor (LDLR) and liver X receptor alpha (LXRα) in liver tissue, thus exerting its hypolipidemic effect . Macrostemonoside A ( 1 ) is a steroidal saponin isolated from AMB, which can reduce TC, TG, and LDL levels in mice serum and blood glucose levels in mice, and increase visfatin protein expression in 3T3-L1 cells . XZT is a proprietary Chinese patent medicine made from AMB extract. Studies have shown that XZT reduces fatty acid synthase (FAS) and LDL levels in the serum of ApoE −/− mice by activating reverse cholesterol transport (RCT) and increasing HDL levels, and that XZT reduces TG levels in patients with hyperlipidemia . The m TOR signaling pathway plays an important role in the progression and treatment of CHD. m TOR is mostly associated with cellular autophagy and apoptosis, and previous studies have demonstrated that autophagy has a dual role in atherosclerosis. The body needs moderate autophagy to stabilize plaque and inhibit excessive autophagy during cardiac I/R injury to reduce myocardial infarct size. Most of the monomeric components of TCM for the treatment of CHD are purified from blood-stasis-activating and qi-supplementing drugs, but the mechanisms of pharmacological effects of qi-activating drugs (e.g., AMB) and expectorants (e.g., Fructus Trichosanthis and Pinellia Tuber), which are commonly used in the clinical treatment of CHD, have been less studied, and research on the mechanisms of active components of these herbs should be strengthened. Myocardial ischemia is the result of an imbalance in oxygen supply and demand to myocardial cells, and early hemodialysis is the most effective way to reduce post-ischemic myocardial injury . With the development of the application of interventions such as percutaneous coronary intervention, coronary artery bypass grafting, and thrombolysis, the myocardium can be resupplied with blood after ischemia, but the ensuing myocardial ischemia-reperfusion injury is a complex pathophysiological process involving multiple factors. The mechanism is currently believed to be closely related to inflammation, oxidative stress, vascular endothelial cell damage, platelet aggregation, and other factors, which can eventually lead to irreversible apoptosis or necrosis . Early reperfusion therapy can aggravate the myocardial injury and become an important factor affecting the outcome of ischemic therapy. The assessment and treatment of reperfusion injury remain a clinical challenge, and the causal mechanism is still unclear. One mechanism that has been identified is that ischemia-reperfusion triggers endothelial cell dysfunction and disrupts the endothelial structure of the blood vessels, thereby impeding blood circulation within the microvasculature. Endothelial cells are not only found in the lining of blood vessels but also cover the heart and lymphatic lumen longitudinally in a single layer, playing an important role in normal cardiac physiology and cardiac response to injury. Endothelial cells also act as secretory cells, secreting vasoactive substances, such as the vasoconstrictors endothelin (ET) and angiotensin, and vasodilators such as NO and endothelial-dependent hyperpolarizing factor (EDHF). They play an important role in regulating the tone of blood vessels, especially microcirculatory vessels; they can also synthesize and secrete relevant coagulation factors and fibrinolytic substances to maintain a dynamic balance between coagulation and fibrinolysis and influence the coagulation and fibrinolysis process, thus maintaining normal blood flow and circulation . It was found that AMB extract reduced the gene expression of inflammation-related cyclooxygenase-2 (COX-2), cyclooxygenase-1 (COX-1), inducible nitric oxide synthase (iNOS), and vasodilation-related endothelin-converting enzyme (ECE), and endothelial nitric oxide synthase (eNOS), but increased the gene expression of antioxidant superoxide dismutase (SOD) in a model of air-stressed vascular endothelial injury, thereby reducing endothelial vascular damage in model rats . At the same time, AMB extract also significantly reduced plasma ET level, increased serum NO level, and inhibited glucose-regulated protein 78 (GRP78) protein expression in aortic tissue to improve vascular endothelial function in model rats by suppressing endoplasmic reticulum stress . In a rat model of acute myocardial ischemia caused by open-chest ligation of the anterior descending branch of the rats’ left coronary artery, ethanolic extract of AMB can regulate the balance of lipid and protein metabolism and reduce the damage caused by acute myocardial ischemia in the rat organism . AMB extract also significantly increased serum glutathione peroxidase (GSH-Px) activity; it decreased acetylcholinesterase (TChE) activity, non-esterified fatty acid (NEFA), and malondialdehyde (MDA) content, and reduced the extent of myocardial injury in rats . In addition, AMB extracts could protect vascular endothelial function in depressed rats by enhancing 5-hydroxytryptamine 1D (5-HT 1D ) mRNA and protein expression, which mediates the diastolic effect, and inhibiting 5-hydroxytryptamine 2A (5-HT 2A ) mRNA and protein expression, which mediates the vasoconstrictive effect . In medicine, cancer is defined as a malignant tumor often originating from epithelial tissue, which is the most common type of malignancy. Globally, cancer has become the leading cause of human death and a serious obstacle to increasing human life expectancy . Today, global cancer incidence and mortality rates are increasing every year, with 28.4 million cancer cases expected in 2040 . The anti-cancer activity of AMB is mainly related to the water-soluble saponins, polysaccharides, and fat-soluble volatile oils contained in it. Reports have illustrated that the active components in AMB have been effective against human non-small cell lung cancer A549 , human lung cancer cells PC-9 , mice sarcoma cells S180 , mice liver cancer cells H22 , human gastric cancer cell SGC-7901 , human breast cancer MCF-7 , human neural cancer cell SF-268 , human lung cancer cells NCI-H460 , human cervical cancer HeLa cells , human colon cancer cells SW-480 , mice melanoma cells B16 , mice breast cancer cells 4T1 , human hepatoma cells Hep-3B , human hepatoma cells HepG2 , human lung adenocarcinoma cell SPC-A-1 , human gastric cancer cell MGC80-3 , human breast cancer cell MDA-MB-231 , human colon cancer cell SW620 and human nasopharyngeal carcinoma cells CNE-1 , which were inhibited in vivo or in vitro. Possible mechanisms of action include: regulation of EGFR/PI3K/m TOR and RAF/MAPK signaling pathways ; inhibition of tumor cell membrane phospholipid synthesis ; enhancement of immune function in mice, especially cellular immune function, which is dominant in tumor immunity, and thus suppression of tumor cells ; directly killing tumor cells by destroying nuclei and organelles ; altering the G 2 /M cell cycle of tumor cells ; promoting the expression of P53 protein to induce apoptosis ; decreasing mitochondrial membrane potential; up-regulating Bax mRNA expression, down-regulating Bcl-2 mRNA expression, and Bcl-2/Bax ratio; enhanceing Caspase-9 and Caspase-3 activity; inducing reactive oxygen species (ROS) production, and promoting apoptosis of tumor cells . The extracts of AMB have inhibitory effects on a variety of bacteria and fungi. It was found that the aqueous extract of AMB had a wide range of antibacterial abilities, and the antibacterial ability varied at different dilutions of the extracts, with a more desirable effect at higher concentrations, and weaker effect at higher dilutions . In addition, the ethanol extract of AMB also has an inhibitory effect on most bacteria, and the inhibition ability is influenced by temperature and pH. The strongest inhibition activity is at 50–60 °C and the activity decreases when the temperature is greater than 100 °C. The inhibition activity is stronger when the pH is neutral or nearly neutral, and the inhibition activity gradually decreases with the enhancement of acidity or alkalinity . AMB may exert its bacterial inhibitory effect by inhibiting the synthesis of bacterial-associated proteins, inhibiting the activity of related enzymes, or changing their cell structure . The material basis of these mechanisms may be related to the sulfur-containing compounds in AMB, and the specific mechanism of action needs to be investigated in depth. Asthma, as a chronic inflammatory disease of the respiratory tract, is one of the most common non-communicable diseases of the respiratory system in children and adults, often caused by allergic reactions. Stimuli such as histamine, acetylcholine, or cold air can cause airway hyperreactivity and produce airway obstruction, which can clinically cause recurrent episodes of wheezing, chest tightness, or coughing . Typical asthma pathology is characterized by airway inflammation, smooth muscle contraction, epithelial cell shedding, excessive mucus secretion, bronchial hyperresponsiveness, and mucosal edema . Standard therapies for asthma are mainly based on bronchodilators and immunosuppressive drugs, which provide short-term relief but not a cure. Chinese medicine has played an important role in the treatment of various respiratory diseases, including asthma, and has a history of more than 2000 years in the treatment of asthma. In recent years, more and more researchers have focused on the effects of Chinese medicine on asthma, and have achieved remarkable results in clinical trials or basic experimental models . Clinically, AMB can be used alone for the treatment of asthma, and in recent years, many studies have been conducted on the pharmacodynamic material basis of AMB for the treatment of asthma. It has been reported that in animal experiments, IL-6 mRNA content in the bronchial tissues of asthmatic guinea pigs was significantly increased . In clinical practice, serum levels of IL-6 are also significantly higher in asthmatics than in normal subjects . In addition, the balance of TXA 2 and PGI2 is an important regulatory mechanism in the pathophysiological mechanism of asthma, and if the ratio of TXA 2 /PGI2 is increased, it causes bronchial smooth muscle contraction leading to asthma; however, because of the instability of TXA2 and PGI2, the corresponding metabolites of both, thromboxane B 2 (TXB 2 ) and 6-keto-prostaglandin F 1α (6-Keto-PGF 1α ) are often measured . Studies have shown that AMB extract can reduce the expression levels of IL-6 and TXB 2 and up-regulate the expression level of 6-Keto-PGF 1α in the serum of asthmatic guinea pigs, thus achieving a panting effect . In vivo and in vitro, the active ingredients in AMB effectively diastole bronchial smooth muscle in a guinea pig model of histamine-induced asthma . In summary, we deduce that the mechanism by which AMB exerts its effect on wheezing may be through the inhibition of inflammatory response, alleviating chronic inflammation and thus relieving the spastic state of bronchial smooth muscle. ROS are oxygen-containing radicals with high oxidative capacity and high activity generated during metabolism, mainly including superoxide anion radical (O 2 − ), hydrogen peroxide (H 2 O 2 ), hydroxyl radical (·OH), etc. ROS are a double-edged sword for cellular life activities: on the one hand, ROS are important tools or signaling molecules in specific cells (such as macrophages, etc.) and play an important role in removing pathogenic microorganisms, maintaining the normal vascular function, and regulating intracellular homeostasis. On the other hand, when the excessive production of intracellular ROS exceeds the scavenging capacity of the antioxidant system in the body, they will attack proteins, DNA and lipids, causing oxidative stress, which is one of the important factors in the occurrence of cell damage, inflammation, and metabolic disorders . Antioxidant enzymes in the body mainly include SOD, GSH-Px, glutathione S-transferase (GST), catalase (CAT), etc. Non-enzyme antioxidants include glutathione, vitamin E, vitamin C, etc. SOD can effectively scavenge O 2 − , protect cells from oxidative damage, and also provide hydrogen atom ligands for the reduction of ROS to produce hydrogen peroxide, which in turn can be catalyzed by GSH-Px and CAT to produce water and oxygen to reduce oxidative stress damage . Oxidative stress is associated with multiple signaling pathway molecules. Nuclear factor erythroid 2-related factor 2 (Nrf2) is a basic leucine zipper transcription factor, and cytoplasmic Nrf2 is normally bound to Kelch-like ECH-associated protein-1. The free Nrf2 is able to translocate from the cytoplasm to the nucleus, where it forms a heterodimer with Maf family proteins and then binds to antioxidant response element sequences to induce the expression of downstream antioxidant enzymes, thereby scavenging ROS, inhibiting oxidative stress, maintaining the structural integrity and normal metabolic function of the cell, and exerting its transcriptional regulatory role . Nuclear factor kappa-B (NF-κB) is a dimeric protein of the Rel family. The heterodimer composed of p65 and p50 is a common activated form of NF-κB. NF-κB can promote the infiltration of neutrophils and macrophages and the release of cytokines, chemokines, adhesion molecules, etc., stimulate the expression and secretion of matrix metalloproteinases, activate nicotinamide adenine dinucleotide phosphate oxidase to produce large amounts of ROS, and trigger oxidative stress-related inflammatory diseases . Silent information regulator 1 (Sirt1) is a nicotinamide adenine dinucleotide-dependent deacetylase. Activated Sirt1 inhibits p66shc expression and reduces mitochondrial ROS production by regulating p66shc, which deacetylates histone H3 bound to the p66shc promoter . It is found that AMB extract alleviates liquor-induced oxidative stress in rats by increasing serum SOD and CAT activities and protecting T lymphocytes, and significantly inhibiting serum lipid peroxide formation . AMB polysaccharide, AMB saponin, and some sulfur-containing compounds can effectively scavenge DPPH, O 2 − and ·OH in vitro and inhibit the oxidation of Fe 2+ to a certain extent, and their antioxidant ability can be enhanced after modification with chlorosulfate-pyridine or α-amylase for AMB polysaccharide . Although there are many experimental studies on the antioxidant activity of various extracts of AMB, most of them are limited to in vitro experiments and the specific mechanism is not yet clear. The research efforts on oxidative stress signaling molecules should be deepened to elucidate the antioxidant mechanism of AMB at the molecular level. Depression is an affective disorder characterized by persistent mood abnormalities, mainly manifested as depressed mood, lack of pleasure, difficulty concentrating, fatigue, physical pain, and other symptoms, with a high disability rate and high patient suicide rate, which brings a serious economic burden to the patient’s family and society . The pathogenesis of depression has not yet been fully investigated and researchers have proposed various hypotheses, among which the monoamine transmitter theory suggests that the development of depression is mainly due to the reduction of 5-hydroxytryptamine (5-HT) and norepinephrine (NE) in the brain; therefore, inhibiting the degradation and reuptake of these two monoamines is beneficial to improve depressive symptoms . The neurotrophic factor hypothesis focuses on the brain-derived neurotrophic factor (BDNF) and suggests that an imbalance of brain derived neurotrophic factor precursor (proBDNF) and mature form of brain-derived neurotrophic factor (mBDNF) is closely related to the development of depression . The neurogenesis hypothesis suggests that downregulation of hippocampal neurogenesis is the cause of depression and that antidepressants work based on promoting neurogenesis . In addition, possible mechanisms such as the hypothalamic-pituitary-adrenal (HPA) axis dysregulation hypothesis, inflammation hypothesis, and genetic hypothesis have also been proposed to explain the development of depression . Depression is gradually becoming an important health problem faced by all human beings today, and its pathogenesis is complex. Although antidepressant western drugs are effective for patients with critical symptoms, they have more side effects in terms of mental and emotional effects when taken for a long time. Therefore, people gradually turn their horizons to Chinese medicine, but the composition of Chinese medicines is complex. It can be difficult to find the best component with significant efficacy among the complex and numerous components of compound medicines and single component treatments. The mechanism by which AMB exerts antidepressant effects on various animal models of depression (including rats and mice) may be through regulating the balance of the internal environment of depression model animals, promoting neurogenesis and BDNF production; at the same time, AMB can significantly improve the pathological changes of organ tissues in the relevant animal models . In addition, the analysis of lipids and acylcarnitine in the plasma of depressed rats by liquid chromatography/ion trap time of flight mass spectrometry and ultra-performance liquid chromatography/triple quadrupole mass spectrometry, respectively, showed that the AMB aqueous extract was able to restore the normal levels of these abnormally altered indicators . Although there are numerous studies on depression, the relevant mechanisms are still under-explained, and more rigorous experimental design is needed in the future, together with modern technology to reduce complex Chinese medicine into simpler groupings, purify components, or increase the study of mechanisms at the cellular-molecular level. It cannot be ignored, however, that Chinese medicine mostly follows a certain idea of combination, and it is necessary to maintain a cautious attitude whether the antidepressant components derived from the reductionist ideas of modern medicine can stand up to clinical tests. In addition to the above pharmacological effects, AMB and its compounds exhibit other activities such as analgesia, hypoxia tolerance, immunomodulation, promotion of osteogenesis, inhibition of hepatic drug enzymes, and mosquito control. Studies have shown that both the raw aqueous decoction of AMB and its fried aqueous decoction have strong analgesic effects and prolong the duration of hypoxia tolerance in mice with enhanced oxygen consumption induced by NaNO 2 intoxication and isoproterenol (ISO) under normoxic conditions. The mechanism of analgesia of AMB may be through the inhibition of voltage-sensitive Nav1.7 channels, thus reducing the excitability of peripheral neurons and exerting analgesic effects . AMB can increase the weight of mice’s immune organs, the spleen and thymus, and can increase carbon particle contouring index K and phagocytosis index α; that is, it can promote the phagocytosis of mononuclear macrophages and improve the specific immune function of the body. AMB volatile oil can increase the spleen index, macrophage phagocytosis rate and splenocyte proliferation index. The regulatory ability of AMB on the immune system may be one of the mechanisms of its anti-tumor effect . AMB alcohol extract can increase the expression of insulin-like growth factor-1 and bone morphogenetic protein-2, thus regulating the formation and resorption of bones and achieving the purpose of promoting bone growth . AMB aqueous extract can significantly reduce the content of cytochrome P450 in mice and has a significant inhibitory effect on hepatic drug enzymes . In addition, the volatile oil of AMB and its two main components (compounds 113 and 135 ) exhibited strong larvicidal effects against Aedes albopictus larvae, suggesting the existence of a basis for the development of mosquito control agents . The modern pharmacological studies on AMB are summarized in . The quality control of Chinese medicine is a prerequisite to ensuring the safe and effective clinical application of Chinese medicine. Standardized research on the quality of Chinese medicine is the top priority to achieve the sustainable development of Chinese medicine in recent years, and strengthening the quality control of Chinese medicine is of great significance to ensure the safety of people’s medicine and promote the development of the Chinese medicine industry. In the 2020 Edition of the ChP, the quality control of AMB mainly includes microscopic identification, thin-layer chromatography (TLC), moisture, total ash, and ethanol leachate detection, and states that the moisture content of AMB shall not exceed 10.0% by the toluene method, the total ash content should not exceed 5.0% by constant weight method, and the leachate content obtained by heating extraction with 75% ethanol shall not be less than 30.0% . It was reported that the surface-enhanced Raman scattering (SERS) spectra of AMB volatiles of different species from different production areas were tested with nano-silver sol as the substrate. The results showed that the SERS spectra of these batches of AMB volatiles were very similar; the intensity of the characteristic peaks varied somewhat, but the peak positions were basically unchanged, and the reproducibility was good, indicating that nano-silver sol could be used as the substrate of SERS for the determination of AMB volatiles . Other scholars have used chromatographic methods to study the content of each component in AMB. This includes the quantitative analysis of furostanol saponins in AMB using high performance liquid chromatography and determination of adenosine ( 155 ) in AMB by reversed-phase high performance liquid chromatography . Gas chromatography-mass spectrometry was used to qualitatively and quantitatively analyze the volatile oil of AMB, and the main components were identified as sulfur-containing compounds and their mass fractions . High performance liquid chromatography-mass spectrometry was used to determine the concentration of AMB saponins in rat plasma and tissues; the experimental results showed that AMB saponins were high in rat liver and kidney, and no such components were detected in brain and lung tissues. This method provides theoretical guidance for AMB quality control and drug use, but there are shortcomings, since the experimental detection of AMB saponins monomers only selected the highest plasma exposure monomers, leaving a future need to study other monomers with high relative exposure . Another study used chemometric methods to select the main components and major absorbed components in rats as their representative components. It then established ultra performance liquid chromatography coupled with quadrupole time-of-flight tandem mass spectrometry for the simultaneous determination of 54 components (15 components were quantitative and 39 components were semi-quantitative), which facilitated the screening of AMB quality markers . In addition, the determination of furostanol saponins content in AMB by colorimetric method with Ehrlich reagent can also be used as one of the methods to evaluate the quality standard of AMB . The main active ingredients in AMB are steroidal saponins, sulfur-containing, and nitrogen-containing compounds. So far, the quality markers of AMB are still unclear, and the current ChP does not have its quantitative standards, so deepening the screening of AMB quality markers is one of the efforts to optimize the quality control of AMB. Further research and development by scholars in this industry are needed to ensure its quality assurance and medication safety. The ancient Chinese medical classics, “ Mingyi Bielu ”, states that AMB is “bitter in taste and non-toxic”, and the same is true of “ Bencao Gangmu ”, which also states that it is non-toxic. In the 2020 Edition of the ChP, the recommended daily dose of AMB for adults is 5–10 g. To date, there have been very few reports of toxicity or side effects of AMB. After reviewing the relevant literature, only one case of intestinal rumbling and diarrhea with yellow watery stools after taking AMB was found, but the specific mechanism is unclear. It is speculated that the components contained in AMB may act as antigens or semi-antigens when they enter the body to cause metabolic diseases in the body, or the components in AMB may directly stimulate mast cells or basophils to release allergic mediators (such as histamine, 5-HT, etc.) or there could be direct activation of the complement system, direct or indirect action on target organs or organs in shock . The oral median lethal dose (LD 50 ) of AMB and its compounds were greater be more than 100 times of their respective clinical doses, and the toxicity was very low. The LD 50 of AMB (70.12 ± 3.49 g/kg) and the compounds (48.72 ± 1.79 g/kg) were administered intraperitoneally to mice, and the symptoms of toxicity were similar, including reduced activity, weakness of limbs, flaccidity, and convulsions . In addition, AMB should be used with caution in patients with Yin deficiency and fever and Qi deficiency, and it is said that AMB should not be consumed with beef. In summary, AMB can be considered non-toxic, with the possibility of toxic reactions only in rare cases or in very large doses for long-term use. AMB has a long history of use. As a special herbal medicine for the treatment of “obstruction of Qi in the chest”, AMB has the efficacy of activating Yang and removing stasis, regulating Qi and eliminating stagnation, and is abundant, inexpensive, and of high medicinal value. This review systematically summarizes botany, ethnopharmacology, phytochemistry, pharmacological effects, quality control, and toxicology of AMB. Botanically, AMB has two sources, A. macrostemon and A. chinense , which are very similar and can be distinguished in the intact plant by the shape and color of the bulb, the length of the scape and pedicel. However, the dried product is difficult to distinguish from its appearance and can be distinguished by microscopic identification. The origin of the two is also different (see ). In traditional applications, AMB is often used in combination with Fructus Trichosanthis, Pinellia Tuber, Cassia Twig, etc., and is clinically effective in the treatment of CHD, AP, and other diseases. However, the efficacy of AMB in ancient Chinese medical books is not limited to this, but also includes anti-fatigue, promotion of wound healing, treatment of CVA, etc. The research into how AMB achieves such effects should be broadened, to expand its medicinal scope and give greater play to its medicinal value. In addition to its medicinal use, AMB is also included as food in the Health Law of the People’s Republic of China, and this medicinal food homologation also provides a favorable condition for further development of AMB in the future. In recent years, the research results on AMB in phytochemistry and pharmacological effects have been remarkable. In phytochemistry, so far, more than 190 kinds of compounds have been extracted and isolated from AMB, with as many as 96 steroidal components, including 89 steroidal saponins, and also some sulfur-containing compounds, nitrogen-containing compounds, phenylpropanoids, and flavonoids. Modern pharmacological studies have shown that AMB has pharmacological activities in areas such as anti-platelet aggregation, hypolipidemia, anti-atherosclerosis, protection of cardiomyocytes and vascular endothelial cells, anticancer, antibacterial, anti-asthma, antioxidants, and antidepressant effects. According to the previous review, its importance may be summarized as follows. AMB may be used in the treatment of atherosclerosis, thrombosis, and hypertension caused by vascular endothelial cell injury and apoptosis. AMB exhibits protective effects on vascular endothelial cells along with antithrombotic and antihypertensive effects; endothelial cell injury is closely related to inflammatory response invasion and antioxidant effects, so the mechanism of endothelial cell protection by AMB may also be closely related to its anti-inflammatory and antioxidant effects. In addition, AMB can inhibit the invasion and migration of tumor cells to varying degrees, thus exerting its anti-tumor effects, and the mechanism is also related to the inhibition of platelet aggregation by AMB. Behind the above research results, there are still deficiencies in the research on AMB: (1) Despite the large number of compounds isolated from AMB, the 2020 Edition of the ChP still only has microscopic identification and TLC, and no quality markers for AMB have been identified; therefore, there is a need to strengthen the screening of quality markers for AMB in combination with relevant studies on chemical composition and pharmacological activity, so as to ensure herb quality and drug safety. (2) Due to the still large technical difficulties in the isolation and purification of a large number of monomeric compounds, most of the current pharmacodynamic studies on AMB saponins have used the total extracts of AMB saponins, while lacking in-depth molecular mechanism studies. Therefore, obtaining sufficient monomeric compounds of AMB saponins and their modification products by chemical synthesis can provide in-depth studies on the pharmacodynamic effects and molecular mechanisms of the monomeric components and provide a theoretical basis for clinical exploration of potential precursor drugs. (3) Using histological and other techniques, and linking the material reflecting the diversity of chemical components with transcriptomics, proteomics or metabolomics reflecting the pharmacological mechanisms, can further elucidate the modern pharmacological mechanism of TCM by modern scientific means under the premise of multi-component drug incorporation and provide new scientific ideas for the modernization of Chinese medicine. Therefore, it is urgent to further investigate and confirm the various activities of AMB using new pharmacological models, and to clarify the corresponding active sites and active components. (4) The toxicological studies of AMB are relatively few, and such studies should be deepened. The corresponding toxicological studies should be conducted under the guidance of TCM theory. In general, despite the many research findings on AMB, there are still many gaps. The top priority is the study of pharmacological activity of the monomeric components of AMB and the screening of quality markers. The information provided in this paper can help set targets for future research directions and commercial development of AMB. |
Sarcoma: Molecular Pathology, Diagnostics, and Therapeutics | ab878e88-024f-437f-9176-8c5baa58cecf | 10055792 | Pathology[mh] | |
Above-Standard Survival of Hepatocellular Carcinoma as the Final Outcome of Comprehensive Hepatology Care Programs in a Remote HCV-Endemic Area | 61d5cb67-b29f-47eb-8b80-64090b9c6e88 | 10056201 | Internal Medicine[mh] | Hepatocellular carcinoma (HCC) is the fifth most common cancer worldwide and the second most frequent cause of cancer-related deaths . HCC and its complications place a significant burden on public health, the healthcare system, and the economy globally. In 2020, the crude incidence rate of HCC was 46.61 per 100,000 person-years, the fifth highest among all cancers in Taiwan and the crude mortality rate was the second highest among all cancers, 32.99 per 100,000 person-years. Factors such as hepatitis B virus (HBV) or hepatitis C virus (HCV) infections, alcohol abuse, metabolic syndrome (particularly non-alcoholic fatty liver disease) and exposure to dietary toxins such as aflatoxins are considered to be related to the development of hepatocellular carcinoma . The above risk factors are potentially preventable, highlighting that risk prevention can decrease the burden of HCC . Furthermore, regular surveillance is associated with early HCC detection and might increase the chance of potentially curative treatment . In Taiwan, the primary cause of HCC is HBV infection, followed by HCV infection, with similar observations reported in most areas of Asia and Sub-Saharan Africa . Due to chronic carriers of hepatitis viruses usually being asymptomatic, most people remain unaware of their HBV or HCV infection and their infection will develop into more advanced diseases after several decades . As a result, the incidence and mortality of HCC in the Taiwanese population remain high and are worthy of attention, especially in rural and remote communities. With this in mind, screening and the early identification of asymptomatic people with chronic HBV or HCV infection became an important issue in Taiwan. This important health problem caught the attention of the government and hepatologists. Health scientists and doctors have become devoted to finding those who need to receive antiviral treatment for HBV or HCV to improve their health status and to also provide these patients with the chance for early detection and efficient treatment of HCC for a favorable outcome . Some major advances in HCC management strategies have facilitated the experts to update of the Barcelona Clinic Liver Cancer (BCLC) guideline in 2022 to guide clinical decision-making for HCC worldwide . Despite the great progress in the treatment of HCC, the overall prognosis is still unsatisfactory, and efforts are needed to overcome this dilemma. For remote and rural communities that typically face difficulties accessing medical resources, screening strategies to identify asymptomatic people living with hepatitis B or C, to diagnose patients with HCC early, and efficaciously refer individuals eligible for treatment, constitute a crucial component of HCC care . To achieve an ideal outcome of HCC for remote and rural communities, several challenges exist, including sustainable financing, practical and efficient screening programs, the continuum of treatment and care, and improving accessibility . Under the feedback funding programs of a local petrochemical corporation, Chang Gung Memorial Hospital Yunlin Branch (YLCGMH) cooperated with the Nursing Department of Chiayi Chang Gung University of Science and Technology to execute health checkups and provided post-examination care. The YLCGMH is a 100-bed local hospital with facilities for almost all laboratory tests, and radiological items, but it has only limited facilities for HCC treatment. Therefore, patients with newly diagnosed or recurrent HCC by either health checkups (screening group) or routine clinical services (non-screening group) in YLCGMH were transferred to the CGMH Chiayi branch (CYCGMH), a tertiary referral hospital with full facilities for HCC treatment. We conducted this retrospective study to evaluate the differences of HCC survival between screening and non-screening patient populations. We also compared the HCC survival result of our comprehensive hepatology care program for rural communities with the expectations of the updated BCLC guideline 2022 .
2.1. Background of Study Area Mailiao and Taihsi are two neighboring rural townships located in the central west coastal area of Taiwan, with a combined population of 77,110 registered residents in 2022. A previous study revealed the prevalence of anti-HCV reached as high as 55% among adult residents in the Taihsi Township . Furthermore, the prevalence of anti-HCV for male HCC patients in the Mailiao Township was more than 50% in a large cohort study . These two townships were reported to have the highest liver cancer mortality age-standardized rates (ASRs) up to 80.43 and 83.47 per 100,000 person-years in the Yunlin county . The YLCGMH is the closest local hospital to these two townships, both with an exceptionally high prevalence of chronic viral hepatitis B and C patients . As a local hospital, the routine daily medical work of YLCGMH includes multidisciplinary outpatient, inpatient and emergency services. This local hospital works to improve their patients’ health from nearby townships in coordination with the Nursing Department of Chiayi Chang Gung University of Science and Technology . In addition to annual checkups for Mailiao and Taihsi, outreach health checkups, including liver disease screening, were also conducted in five surrounding townships (Sihhu, Baojhong, Dacheng, Dongshih and Lunbei Townships), as depicted in The outreach health checkups were conducted in up to sixty-five villages between 2018 and 2022. Villagers positive for anti-HCV or HBsAg were called back for further evaluation and treatment . Regular shuttle services between YLCGMH and CYCGMH and certain townships were also provided to improve medical accessibility. 2.2. Patients Sponsored since 2012 by a local petrochemical corporation, all the registered residents in the Mailiao and Taihsi Townships were eligible for, and were encouraged to participate in, the annual health checkups without any restriction. Outreach health checkups for all inhabitants in another five surrounding townships were initiated in 2018. Items concerning hepatitis and HCC screening were included in these health checkup programs, i.e., hepatitis B virus surface antigen s(HBsAg), antibodies against HCV (anti-HCV), alpha-fetoproteins (AFP), and abdominal ultrasonography performed by hepatologists in YLCGMH. Patients with HBV and HCV were treated in this local hospital according to international guidelines. Participants of the health checkups and patients of routine clinical services diagnosed with HCC in YLCGMH were referred to CYCGMH for further treatment. About 10,000 people joined the health checkup program every year, with screening coverage rates above 50% of the total population in these nearby townships. Until 2022, an estimated 94.7% of participants who were positive for anti-HCV antibodies came back for HCV RNA testing, and 95.2% of viremic HCV patients in these townships have been cured with either peginterferon-based therapy or direct-acting antiviral drugs and they were encouraged to undergo regular follow-up after treatment. Our study included all patients with newly diagnosed HCC in YLCGMH from October 2017 to October 2022. This study was approved by the Institutional Review Board of Chang Gung Memorial Hospital. 2.3. Definition The diagnosis of HCC was based on the criteria of the American Association for the Study of Liver Disease (AASLD) and the European Association for the Study of the Liver (EASL) and confirmed by histological analysis, when available. The diagnosis of cirrhosis was based on any of the following modalities: liver histology; fibrosis-4 (FIB-4) index (>6.5); or the presence of clinical, radiological, endoscopic, or laboratory evidence of cirrhosis and/or portal hypertension. 2.4. Calculation of ALBI Grades The albumin–bilirubin (ALBI) grade, taking into account the levels of albumin and bilirubin, was calculated as follows: linear predictor = (log10 bilirubin × 0.66) + (albumin × −0.085), where the units of bilirubin and albumin are μmol/l and g/l, respectively. The ALBI grades were stratified into the following three grades: grade I, ≤−2.60; grade II, −2.60 to ≤−1.39; and grade III, >−1.39, as reported previously . 2.5. Screening and Linkage to Accessible Care The health checkup programs included screening tests for liver diseases, such as HBsAg, anti-HCV, aspartate aminotransferase (AST), alanine aminotransferase (ALT), AFP, and abdominal ultrasonography that was performed by hepatology specialists in YLCGMH. Individuals with highly suspected liver tumors detected in the health checkups were referred to the hepatology clinics of YLCGMH. A biweekly hepatology clinic composed of physicians and surgeons from CYCGMH established in October, 2017 accepted internal and external referrals of patients with uncommon or severe liver diseases, mostly hepatocellular carcinoma. CT and MRI scans of livers can be performed in YLCGMH for further liver tumor evaluation. Patients with newly diagnosed or recurrent HCC who needed specific HCC treatment were transferred to the CYCGMH, a tertiary referral center with full facilities for HCC treatment. There are regular shuttles between YLCGMH and CYCGMH to shorten the gap in medical access. After treatment, the patients can receive post-treatment care and follow-up in YLCGMH. 2.6. Statistical Analysis Statistical analyses were performed using the SPSS 23.0 statistical package (SPSS, Inc., Chicago, IL, USA). Quantitative variables were expressed as mean ± standard deviation (SD) or medians with interquartile ranges (IQR). The chi-square test and Fisher’s exact test were used to compare categorical variables; the t -test or Mann–Whitney U-test were used to compare continuous variables. The overall survival (OS) relationships between groups were analyzed using Kaplan–Meier survival curves and the log-rank test. A two-tailed p -value of < 0.05 was considered as statistically significant.
Mailiao and Taihsi are two neighboring rural townships located in the central west coastal area of Taiwan, with a combined population of 77,110 registered residents in 2022. A previous study revealed the prevalence of anti-HCV reached as high as 55% among adult residents in the Taihsi Township . Furthermore, the prevalence of anti-HCV for male HCC patients in the Mailiao Township was more than 50% in a large cohort study . These two townships were reported to have the highest liver cancer mortality age-standardized rates (ASRs) up to 80.43 and 83.47 per 100,000 person-years in the Yunlin county . The YLCGMH is the closest local hospital to these two townships, both with an exceptionally high prevalence of chronic viral hepatitis B and C patients . As a local hospital, the routine daily medical work of YLCGMH includes multidisciplinary outpatient, inpatient and emergency services. This local hospital works to improve their patients’ health from nearby townships in coordination with the Nursing Department of Chiayi Chang Gung University of Science and Technology . In addition to annual checkups for Mailiao and Taihsi, outreach health checkups, including liver disease screening, were also conducted in five surrounding townships (Sihhu, Baojhong, Dacheng, Dongshih and Lunbei Townships), as depicted in The outreach health checkups were conducted in up to sixty-five villages between 2018 and 2022. Villagers positive for anti-HCV or HBsAg were called back for further evaluation and treatment . Regular shuttle services between YLCGMH and CYCGMH and certain townships were also provided to improve medical accessibility.
Sponsored since 2012 by a local petrochemical corporation, all the registered residents in the Mailiao and Taihsi Townships were eligible for, and were encouraged to participate in, the annual health checkups without any restriction. Outreach health checkups for all inhabitants in another five surrounding townships were initiated in 2018. Items concerning hepatitis and HCC screening were included in these health checkup programs, i.e., hepatitis B virus surface antigen s(HBsAg), antibodies against HCV (anti-HCV), alpha-fetoproteins (AFP), and abdominal ultrasonography performed by hepatologists in YLCGMH. Patients with HBV and HCV were treated in this local hospital according to international guidelines. Participants of the health checkups and patients of routine clinical services diagnosed with HCC in YLCGMH were referred to CYCGMH for further treatment. About 10,000 people joined the health checkup program every year, with screening coverage rates above 50% of the total population in these nearby townships. Until 2022, an estimated 94.7% of participants who were positive for anti-HCV antibodies came back for HCV RNA testing, and 95.2% of viremic HCV patients in these townships have been cured with either peginterferon-based therapy or direct-acting antiviral drugs and they were encouraged to undergo regular follow-up after treatment. Our study included all patients with newly diagnosed HCC in YLCGMH from October 2017 to October 2022. This study was approved by the Institutional Review Board of Chang Gung Memorial Hospital.
The diagnosis of HCC was based on the criteria of the American Association for the Study of Liver Disease (AASLD) and the European Association for the Study of the Liver (EASL) and confirmed by histological analysis, when available. The diagnosis of cirrhosis was based on any of the following modalities: liver histology; fibrosis-4 (FIB-4) index (>6.5); or the presence of clinical, radiological, endoscopic, or laboratory evidence of cirrhosis and/or portal hypertension.
The albumin–bilirubin (ALBI) grade, taking into account the levels of albumin and bilirubin, was calculated as follows: linear predictor = (log10 bilirubin × 0.66) + (albumin × −0.085), where the units of bilirubin and albumin are μmol/l and g/l, respectively. The ALBI grades were stratified into the following three grades: grade I, ≤−2.60; grade II, −2.60 to ≤−1.39; and grade III, >−1.39, as reported previously .
The health checkup programs included screening tests for liver diseases, such as HBsAg, anti-HCV, aspartate aminotransferase (AST), alanine aminotransferase (ALT), AFP, and abdominal ultrasonography that was performed by hepatology specialists in YLCGMH. Individuals with highly suspected liver tumors detected in the health checkups were referred to the hepatology clinics of YLCGMH. A biweekly hepatology clinic composed of physicians and surgeons from CYCGMH established in October, 2017 accepted internal and external referrals of patients with uncommon or severe liver diseases, mostly hepatocellular carcinoma. CT and MRI scans of livers can be performed in YLCGMH for further liver tumor evaluation. Patients with newly diagnosed or recurrent HCC who needed specific HCC treatment were transferred to the CYCGMH, a tertiary referral center with full facilities for HCC treatment. There are regular shuttles between YLCGMH and CYCGMH to shorten the gap in medical access. After treatment, the patients can receive post-treatment care and follow-up in YLCGMH.
Statistical analyses were performed using the SPSS 23.0 statistical package (SPSS, Inc., Chicago, IL, USA). Quantitative variables were expressed as mean ± standard deviation (SD) or medians with interquartile ranges (IQR). The chi-square test and Fisher’s exact test were used to compare categorical variables; the t -test or Mann–Whitney U-test were used to compare continuous variables. The overall survival (OS) relationships between groups were analyzed using Kaplan–Meier survival curves and the log-rank test. A two-tailed p -value of < 0.05 was considered as statistically significant.
3.1. Baseline Characteristics of HCC Patients in the Screening and Control Groups One hundred and two patients with liver tumors were referred from YLCGMH to CYCGMH during the study period. Of them, 25 patients were excluded, as 15 patients had received previous HCC treatment, 4 patients were lost to follow-up, 3 patients refused further management due to old age, and 3 patients were proved to be non-HCC after surgery. Finally, 77 patients with newly diagnosed HCC were enrolled in this study. They were further categorized into two groups based on whether the initial diagnosis of the liver tumor was detected in the annual health checkup. The screening group included 53 (68.9%) patients with their HCC detected in the annual health checkup, while the remaining 24 (31.2%) patients with their HCC detected in the routine clinical service served as the control group . presents the baseline characteristics of the 77 HCC patients in this study cohort. There were 61 (79.2%) men and 16 (20.8%) women, with a mean age of 65.7 ± 11.1 years at enrollment. Of them, 66 patients were villagers in the Mailiao and Taihsi Townships. The risk factors of HCC were HBV (15/77; 19.4%), HCV (41/77; 53.2%), dual HBV and HCV (13/77; 16.8%), and non-HBV, non-HCV (NBNC) (8/77; 10.3%). The median level was 0.9 (IQR, 0.6–1.2) g/dL for total bilirubin, 4.2 (IQR, 3.9–4.6) g/dL for albumin, 45.9 (IQR, 26.0–55.5) U/L for AST, 43.5 (IQR, 21.5–59.0) U/L for ALT, and 8.4 (IQR, 3.6–54.3) ng/mL for AFP. Moreover, 73 patients belonged to Child–Pugh grade A, and 4 belonged to Child–Pugh grade B cirrhosis. In addition, there were 53, 23, and 1 patient with ALBI grade I, II, and III liver reserve, respectively. As for tumor stage, there were 21, 40 and 14 patients with BCLC stage 0, A and B, respectively, and 2 patients with BCLC stage C. Seventeen patients had more than one liver tumor. There were no statistically significant differences between the 53 patients in the screening group and the 24 patients in the control group regarding age (66 ± 11.0 vs. 65.3 ± 115, p = 0.809), gender (42:11 vs. 19:5, p = 0.994), distribution of viral hepatitis (NBNC:HBV:HCV: HBV + HCV 6:10:26:11 vs. 2:5:15:2, p = 0.523), Child–Pugh grade (A; B 52:1 vs. 21:3, p = 0.052) and tumor size in cm (2:2–5: > 5 18:33:2 vs. 9:11:4, p = 0.113). Patients in the screening group were 100% villagers in the Mailiao and Taihsi Townships; the control group included 11 patients (45.8%) who lived in other townships. When compared to the control group, the screening group had a higher median level of albumin (4.3, 4.1–4.6 vs. 3.9, 3.4–4.6 g/dL, p = 0.040), lower ALT level (37.2, 24.5–44.5 vs. 65.2, 29–77.2 U/L, p = 0.044), more cases diagnosed in the early stage (BCLC stage 0 + A 86.8% vs. 62.5%, p = 0.028), better liver function reserve (ALBI grade I 77.3% vs. 50%, p = 0.031), and a higher percentage of single tumors (86.7% vs. 58.3%, p = 0.001). 3.2. Antiviral Treaments for the Study Cohort Patients with HBV and HCV were treated in this local hospital according to international guidelines. There were 21 patients with detectable HBV DNA and 45 patients with detectable HVC RNA. All the patients with detectable HBV DNA received nucelos(t)ide analogues. Thirsty-three patients were treated with direct-acting antiviral agents and nine patients with peginterferon-based therapy among the patients with detectable HCV RNA. The SVR of the screening and control groups was as high as 86.2% and 92.3%, respectively. 3.3. Survival Outcome of the Study Cohort As shown in , the screening group had better overall survival (OS) than the control group, with a p -value of 0.048. More than half the patients among those two groups were still alive at the end of our study. Of the 77 patients, there were 21, 40, 14, and 2 patients with BCLC stage 0, A, B, and C, respectively. The Kaplan–Meier analysis revealed that the BCLC 0/A patients had a higher OS than the BCLC B and BCLC C patients ( p < 0.001) . In this 5-year follow-up study, more than half of the BCLC 0/A patients were still alive at the end of the follow-up. All BCLC stage 0 patients were alive at the end of the data analysis. The median survival values for the BCLC B and C patients were 40.7 months and 1.1 months, respectively, in our cohort.
One hundred and two patients with liver tumors were referred from YLCGMH to CYCGMH during the study period. Of them, 25 patients were excluded, as 15 patients had received previous HCC treatment, 4 patients were lost to follow-up, 3 patients refused further management due to old age, and 3 patients were proved to be non-HCC after surgery. Finally, 77 patients with newly diagnosed HCC were enrolled in this study. They were further categorized into two groups based on whether the initial diagnosis of the liver tumor was detected in the annual health checkup. The screening group included 53 (68.9%) patients with their HCC detected in the annual health checkup, while the remaining 24 (31.2%) patients with their HCC detected in the routine clinical service served as the control group . presents the baseline characteristics of the 77 HCC patients in this study cohort. There were 61 (79.2%) men and 16 (20.8%) women, with a mean age of 65.7 ± 11.1 years at enrollment. Of them, 66 patients were villagers in the Mailiao and Taihsi Townships. The risk factors of HCC were HBV (15/77; 19.4%), HCV (41/77; 53.2%), dual HBV and HCV (13/77; 16.8%), and non-HBV, non-HCV (NBNC) (8/77; 10.3%). The median level was 0.9 (IQR, 0.6–1.2) g/dL for total bilirubin, 4.2 (IQR, 3.9–4.6) g/dL for albumin, 45.9 (IQR, 26.0–55.5) U/L for AST, 43.5 (IQR, 21.5–59.0) U/L for ALT, and 8.4 (IQR, 3.6–54.3) ng/mL for AFP. Moreover, 73 patients belonged to Child–Pugh grade A, and 4 belonged to Child–Pugh grade B cirrhosis. In addition, there were 53, 23, and 1 patient with ALBI grade I, II, and III liver reserve, respectively. As for tumor stage, there were 21, 40 and 14 patients with BCLC stage 0, A and B, respectively, and 2 patients with BCLC stage C. Seventeen patients had more than one liver tumor. There were no statistically significant differences between the 53 patients in the screening group and the 24 patients in the control group regarding age (66 ± 11.0 vs. 65.3 ± 115, p = 0.809), gender (42:11 vs. 19:5, p = 0.994), distribution of viral hepatitis (NBNC:HBV:HCV: HBV + HCV 6:10:26:11 vs. 2:5:15:2, p = 0.523), Child–Pugh grade (A; B 52:1 vs. 21:3, p = 0.052) and tumor size in cm (2:2–5: > 5 18:33:2 vs. 9:11:4, p = 0.113). Patients in the screening group were 100% villagers in the Mailiao and Taihsi Townships; the control group included 11 patients (45.8%) who lived in other townships. When compared to the control group, the screening group had a higher median level of albumin (4.3, 4.1–4.6 vs. 3.9, 3.4–4.6 g/dL, p = 0.040), lower ALT level (37.2, 24.5–44.5 vs. 65.2, 29–77.2 U/L, p = 0.044), more cases diagnosed in the early stage (BCLC stage 0 + A 86.8% vs. 62.5%, p = 0.028), better liver function reserve (ALBI grade I 77.3% vs. 50%, p = 0.031), and a higher percentage of single tumors (86.7% vs. 58.3%, p = 0.001).
Patients with HBV and HCV were treated in this local hospital according to international guidelines. There were 21 patients with detectable HBV DNA and 45 patients with detectable HVC RNA. All the patients with detectable HBV DNA received nucelos(t)ide analogues. Thirsty-three patients were treated with direct-acting antiviral agents and nine patients with peginterferon-based therapy among the patients with detectable HCV RNA. The SVR of the screening and control groups was as high as 86.2% and 92.3%, respectively.
As shown in , the screening group had better overall survival (OS) than the control group, with a p -value of 0.048. More than half the patients among those two groups were still alive at the end of our study. Of the 77 patients, there were 21, 40, 14, and 2 patients with BCLC stage 0, A, B, and C, respectively. The Kaplan–Meier analysis revealed that the BCLC 0/A patients had a higher OS than the BCLC B and BCLC C patients ( p < 0.001) . In this 5-year follow-up study, more than half of the BCLC 0/A patients were still alive at the end of the follow-up. All BCLC stage 0 patients were alive at the end of the data analysis. The median survival values for the BCLC B and C patients were 40.7 months and 1.1 months, respectively, in our cohort.
Chronic HBV and HCV infections are endemic and considered to be the most important risk factor for HCC in Taiwan . HBV carriers accounted for up to 20% of the general population in previous decades in Taiwan, which brought upon an increased risk of HCC incidence . However, the prevalence of chronic HBV carriers and HBV-related HCC incidence declined dramatically after a nationwide HBV immunization program for newborns in Taiwan was launched on 1 July 1984 . Subsequently, the role of HCV in the etiology of HCC in Taiwan has increased in the last twenty years . The previous case–control study also found that chronic HCV infection was the primary cause of excess mortality from HCC in an HBV–HCV-endemic area . HCV infection has become a major public health issue in Taiwan. To uncover the asymptomatic hepatitis carriers and to eliminate hepatitis C in this population deserve more attention and efforts. There is still no effective vaccine to control HCV infection. The estimated prevalence of anti-HCV was 3.28% (1.8–5.5%) in the general population, whereas it is as high as 6% to 30% in some townships . The prevalence is much higher than the global prevalence . In a modeling study in 2015, the global prevalence of active HCV was estimated to be 1% . The availability of short and easily tolerable treatment courses that involve direct-acting antiviral drugs has made HCV elimination more feasible than interferon-based treatment. Subsequently, this has led to viral eradication in more than 98% of patients infected with hepatitis C virus. The treatment course usually only lasts for 8–12 weeks and with no or only minor adverse effects . However, there is still an extensive gap between the numbers of hepatitis patients infected and those diagnosed, especially in remote or rural townships . National Health Insurance (NHI) in Taiwan has put emphasis on the public health issue. To implement a cost-effective strategy to eradicate HCV, it is important to know the trends and gain a more detailed understanding of the possible high-HCV-endemic areas. A series of geographical distributions of HCV in Taiwan were analyzed . In addition, outreach screening programs were conducted to increase the accessibility of patients and to improve the health status for residents in rural communities . Tien et al. conducted a program of village-by-village screening tests in the Laiyi and Mudan Townships for hepatitis, linking outreach hepatology care at the two indigenous townships in Pingtung county localized at the most southern part of Taiwan . Up to 95.7% of residents with HCV in this program achieved a sustained virological response (SVR), defined as undetectable HCV viremia, for at least 12 weeks after the end of treatment . Since 2018, YLCGMH has coordinated with the Nursing Department of Chiayi Chang Gung University of Science and Technology to improve villagers’ health in nearby townships . Outreach health checkups, including liver disease screening, have been conducted in five surrounding townships. The staff of primary health care centers called back villagers when they were positive for anti-HCV or HBsAg, linking to further care . Until 2022, an estimated 94.7% of patients positive for anti-HCV came back for HCV RNA testing, and 95.2% of HCV patients in these townships have been cured either with peginterferon-based therapy or direct-acting antivirals drugs. Outreach screening programs and linkage to care would be a feasible model to decrease the burden of chronic virial hepatitis for residents in rural communities. The staff at primary health care centers are an important cadre of the primary health care workforce in many rural or low-income communities. They serve as liaisons between community members and health care providers, delivering appropriate health services to the community. The primary health care center links the villagers to medical care, but also provides appropriate health information. Many rural residents are unaware if they have chronic hepatitis C infection. In addition, very few are aware of the new drugs with high cure rates and minimal side effects that can greatly reduce their health burden . Staff at the primary health care center call back villagers for HCV RNA confirmation and free DAAs treatment when they are positive for anti-HCV, providing information on the benefits of DAA and follow-up medication compliance. Furthermore, they educate and empower residents in rural settings to raise awareness about the transmission routes of hepatitis C virus infection. The services offered by the primary health care center are very essential not only for the villagers, but also for the public health system, especially to reach poorer populations and those living in remote areas with limited access to quality medical care. Hepatocellular carcinoma has been one of the most common cancers in Taiwan for the past four decades. Nevertheless, the incidence of HCC decreased after the reimbursement of costs from the National Health Insurance (NHI) scheme for nationwide nucleos(t)ide analogues and interferon-based treatment for chronic HBV and HCV patients in 2003 and successful antiviral therapy treatment . Furthermore, survival improved after the improvement in tumor detection and treatment instrumentation in recent decades. Curative treatments became more widely available to HCC patients, generally improving prognosis . In Taiwan, the 5-year survival rates of BCLC stages 0, A, B, C, and D were 70%, 58%, 34%, 11%, and 4%, respectively . The median survival years of BCLC stages 0, A, B, C, and D were 9.7, 6.3, 2.7, 0.6, and 0.2 years, respectively . The proportion of BCLC stage 0 patients increased from 6.2% to 11.3%. The survival rate significantly increased year-by-year from 2011 to 2019 . It was interesting to note that even though the overall survival rates improved, the survival rates of patients with HCC who lived in rural areas remained lower than those in urban areas . The phenomenon was not only found among US adults, but also among patients with hepatocellular carcinoma residing in rural locations of Australia and Taiwan . The survival differences might be related to the fact that urban areas in Taiwan have more medical resources and populations with higher socioeconomic statuses. On the other hand, patients from rural households may have suboptimal access to liver disease care, which may translate into worse HCC outcomes in the rural areas of Taiwan . Mailiao and Taihsi are rural indigenous townships with a high risk of HCC mortality in the Yunlin county, with high priorities for HCC detection and treatment programs . The two townships were endemic for HCV infection due to limited medical recourses, low socioeconomic status, poor hygiene practices and inadequate sanitary conditions . It is served by the Chang Gung Memorial Hospital Yunlin Branch (YLCGMH), a 100-bed local hospital located near Mailiao and Taihsi. The YLCGMH is committed to providing HBV and HCV treatment for villagers living in rural areas. More than 1000 patients with HCV viremia have been detected by the annual health checkups and most of them have been treated by either interferon-based or direct-acting antiviral agent (DAA) regimens at YLCGMH. Among our study cohort, all the patients with detectable HBV DNA received nucelos(t)ide analogues according to international guidelines. Nowadays, the effectiveness of DAAs against HCV, following successful treatment of early hepatocellular carcinoma (HCC), has been extensively studied. The benefit of DAA against HCV following successful treatment of HCC remains controversial. One meta-analysis that assessed the HCC recurrence risk following DAA administration revealed inconclusive results . A large number of studies revealed that DAA therapy was associated with a significant reduction in the risk of death, improved overall survival and reduced the risk of hepatic decompensation . To achieve better overall survival rates of HCV-related HCC, most patients with detectable HCV RNA received peginterferon-based therapy or DAA for HCV eradication. The SVR of the screening and control groups was as high as 86.2% and 92.3%, respectively. The program of annual health checkups enabled the early detection of HCC. Patients in the screening group had better liver function reserve, demonstrated early-stage HCC and single tumors compared with the control group. In addition, patients in the screening group demonstrates more prolonged survival than the control group, with p = 0.048. Annual health checkups for the HCV-endemic townships can indeed identify HCC at early stages and improve patient survival in rural indigenous villages. Medical accessibility and continuum of care are essential in the outcome of HCC treatment. However, while YLCGMH can perform almost all types of laboratory tests, in addition to CT, and MRI scans for HCC detection, it has minimal facilities for HCC treatment. To compensate for this limitation and to close the medical resources gap between rural and urban townships, a comprehensive hepatology care program was set up in a remote HCV-endemic area. Patients with advanced liver diseases, mostly HCC, were referred to a biweekly hepatology clinic established in October 2017. Patients with newly diagnosed or recurrent HCC were transferred to the CGMH Chiayi branch (CYCGMH) for further management. Furthermore, medical shuttles were provided between YLCGMH and CYCGMH. Post-HCC treatment patients were under regular surveillance in YLCGMH. More than half of the patients with BCLC 0/A were alive at the end of the follow-up in the rural communities that were endemic for HCV. All BCLC stage 0 patients were alive at the end of the data analysis. Our cohorts’ median survival rate for patients in the B stage was 3.34 years. In Taiwan, the 5-year survival rates of BCLC stages 0 and A were 70% and the median survival rates of BCLC stage B patients were 2.7 years . The expectations of the BCLC guideline 2022 for patients in stages 0/A and B are >5 and >2.5 years . The outcome of BCLC 0/A and B in the rural cohort study was not inferior to the results of Taiwan and were also above the expectations of the BCLC guideline 2022 for stages 0/A and B. In consequence, the program shortened the health gap between the indigenous and the general population in Taiwan and improved patient outcomes. Chang Gung Memorial Hospital Yunlin Branch and the Nursing Department of Chiayi Chang Gung University of Science and Technology played an important role in improving the health status of residents living in central west coastal Taiwan, endemic for hepatitis C virus infection. Outreach screening programs and linkage to care systems for HBV treatment and HCV elimination can not only avoid the development of more advanced liver diseases, but also reduce the incidence of HCC. Furthermore, the annual health checkups and abdominal ultrasonography conducted by hepatologists resulted in the early detection of HCC. A biweekly hepatology clinic, referral system and medical shuttles provided optimal access to HCC treatment, as well as health care accessibility. The comprehensive liver disease care program provided early detection of liver tumors and improved the survival of HCC patients in a remote HCV-endemic area. Based on the above results, we proposed a practical model to improve the health status of rural citizens. However, there are still some limitations in our study. First, this was a retrospective study conducted in a local hospital and the sample size was small. This may limit the generalizability of our findings and the results may be affected by extreme values due to the small sample size. In addition, the screening coverage was not comprehensive because we only recruited volunteered participants instead of using a systematic screening approach. Third, the follow-up duration may not be long enough to present the real effects of the comprehensive liver disease care program. Large prospective cohort studies should be performed.
HCV eradication and HCC treatment are essential issues for rural indigenous townships in central west coastal Taiwan, due to limited medical resources or settings with inadequate access to medical resources. This successful comprehensive hepatology care model in this remote endemic area showed that the screening of hepatitis and HCC and linkage to the residents’ access to high-quality care result in favorable HCC survival rates.
|
Effects of a Comprehensive Person-Centered Care Education Program for Nursing Students | 91c92cc9-befd-4642-9dd0-5ecf2416dd0c | 10056725 | Patient-Centered Care[mh] | The paradigm of global healthcare is shifting from a disease-centered to a person-centered perspective. Person-centered care (PCC) refers to a type of care that considers individuals’ values and preferences, informs the patient about all aspects of health management, and supports realistic health and life goals . In other words, it is a type of care that focuses on the patient, respects the patient, actively responds to the patient, and ensures the patient’s rights and autonomy in treatment decisions . Essential elements of PCC are individual focus, empathy, communication, respect, and coordinated care . Patients who receive PCC feel more respected through individualized care, develop more confidence in improving their own health, and feel that they are given more support . PCC enhances patients’ health-related quality of life by reducing chronic disease morbidity and improving lifestyle . It also improves nursing care providers’ job satisfaction and quality of care . Thus, PCC is a core competency that must be continually advanced. Nursing students become partially involved in patient care during their clinical practicum ; therefore, it is imperative that PCC competence is cultivated through effective education programs so that nursing students who will be prospective nurses can provide quality patient care. Existing studies on nursing students conducted abroad have focused on PCC learning experiences for a semester and introduction to teaching approaches ; studies developing PCC education programs are scarce. In Korea, one study showed that a PCC education program based on a design-thinking approach improved the perceptions of individualized care , but most of the published studies investigated the predictors of PCC competence in nursing students [ , , ]. Although not PCC education programs for nursing students, a prior study implemented a person-centered dementia education program with nursing care providers in long-term care facilities , while another applied a PCC education program among clinical nurses . However, research on developing a PCC education program with practical applications for nursing students and evaluating its effectiveness has been insufficient. An effective educational approach is critical to enhance nursing students’ levels of caring . Furthermore, as PCC education focused on case studies helps nursing students to gain an understanding of PCC fundamentals and broadens their view of PCC , case study-based education is as important as theoretical education in cultivating knowledge and attitudes to promote PCC practice. This study utilizes case studies that would help students apply their learned knowledge in practice as well as theoretical education for PCC. Nursing students must also be taught effective communication skills to build an empathetic relationship with their patients to provide PCC [ , , ]. In other words, carefully listening to and communicating with patients is crucial to accurately understand their motivations, priorities, and preferences . Empathy is also an important factor for practicing PCC . It is important in providing meaningful care, which enhances patients’ well-being . Thus, PCC education for nursing students must also address empathy as well as communication skills. No study has yet reported the effects of a comprehensive PCC education program that encompasses a theoretical approach, PCC cases that help apply knowledge in practice, and education to boost communication competence and empathy. Therefore, this study developed a comprehensive PCC education program for nursing students undergoing clinical practicum and investigated the effects of the program on the key elements of PCC—individualized care, empathy, and communication competence—and clinical practice stress.
2.1. Study Design This study used a non-equivalent control group non-synchronized design. Conducting the study simultaneously with the experimental and control groups would have elevated the risk of diffusion effect, as participants were undergoing clinical practicum at the same hospital. Therefore, this study used a non-synchronized design, wherein the control group underwent the study first, followed by the experimental group. 2.2. Participants The inclusion criteria were (1) third-year nursing students who had experienced clinical training for more than one semester at a general hospital, (2) scheduled for clinical practicum during the study period, and (3) no prior participation in a similar PCC education program. The exclusion criteria were (1) unavailable to undergo at least two weeks of on-campus practicum and clinical practicum in the hospital after the completion of the education program and (2) unable to participate in all sessions of the education program. Sample size was calculated using the G*Power 3.1 software. For independent t-tests with a significance (α) of 0.05, power of 0.8, and effect size of 0.8, the minimum sample size was 26 for each group. The dropout criteria were missing any of the four sessions of the program or failure to complete the pre-test or post-test. 2.3. Enrollment, Allocation, and Blinding Participants were recruited among third-year nursing students at a nursing school in C city in South Korea through social network service and bulletin board. The researcher anticipated a high dropout rate as the study was conducted during clinical practicum. Considering about 30% dropout, the researcher intended to recruit 68 participants, but 73 participants were enrolled. The recruiting process, educational intervention, and surveys were conducted from 18 May 2021 to 10 December 2021. A third person not involved in the study arbitrarily assigned participants into the experimental group and control group. Participants were given a detailed explanation about the study procedure during the informed consent process; they were aware of which program they were participating in. The researcher administered the education program; therefore, the researcher was also aware of the provided content. To reduce the consequent bias by these problems, this study used a non-synchronized design to ensure that the control group and experimental group did not have any contact with each other, and the researcher followed a manual when providing education to ensure consistency in each session. The experimental group completed the pre-test immediately before beginning the education program and completed the post-test after two weeks of on-campus and clinical practicum following program completion. The control group also completed the post-test after two weeks of on-campus and clinical practicum following the pre-test. The researcher and research assistants administered the pre-test and post-test. Of 73 participants, 6 participants who could not undergo at least 2 weeks of on-campus and clinical practicum following the completion of the education program were excluded. Of 67 participants meeting the criteria, the control group was expected to have a higher dropout rate because they only complete the pre-test and post-test without intervention; therefore, 36 participants were assigned to the control group and 31 were assigned to the experimental group. A total of 7 participants in the control group failed to complete the post-test, resulting in a total of 29 participants, and all 31 participants in the experimental group were included. Thus, a total of 60 participants were included in the final analysis ( ). 2.4. Development Process of Comprehensive PCC Education Program Existing studies were reviewed to identify the essential contents for the comprehensive PCC education program. Nursing students need to know the concept and principles of PCC to perform PCC during clinical practicum. Thus, this study reviewed previous studies [ , , , ]. Detailed PCC education that provides practical assistance is needed for nursing students, and case studies help students understand PCC fundamentals . Thus, this study included PCC cases proposed by Price that can be applied in clinical practicum. Moreover, empathy, communication, respect, relationship, and individualized focus are important factors in PCC . This study examined previous research on improving empathy and communication training in nursing students [ , , ]. The program was structured to include the definition of PCC, core premises, and effects of PCC to help students gain an understanding of PCC, broaden their view, and cultivate a positive attitude toward PCC. Case studies of PCC were taken from the book written by Price and translated into Korean. Three cases were included: helping a patient understand their situation (patient with multiple sclerosis), helping a patient learn about the treatment, care, and recovery (pediatric patient with diabetes), and helping a patient cope with changes and anxiety (patient with colorectal cancer). Empathy training included contents about understanding the concept of empathy, knowing one’s emotions, expressing emotions, and listening. The goal of the empathy training was to ensure that the students had an accurate understanding of empathy, knew how to express empathy, recognized their own emotions, and listened correctly to identify others’ emotions. Communication training comprised open-ended questions, affirmations, I-message, and reflective listening. The goal of communication training was to help nursing students to build a relationship of trust with the patient, identify the patient’s values, needs, and preferences, deliver information individually, and help patients actively improve their own lifestyle. This study set the program to 4 sessions with reference to the 2–6 sessions used by a study providing a person-centered dementia care education program to nursing care providers in long-term care facilities and a study providing a PCC education program to clinical nurses . Each session was set to last 65 min based on the 60min session used previously , and each session comprised an introduction (5 min), education (50 min), and hands-on practice and sharing of thoughts (10 min). The educational approaches used were lectures, role play, hands-on practice, and sharing of thoughts based on previous studies . The study parameters were assessed before and after the program using self-report questionnaires. The questionnaires were administered at baseline and after two weeks of on-campus and clinical practicum following the intervention. The educational materials were written in 12-point font size to provide to the participants at each session. Moreover, a training manual for each session was developed to ensure consistency in the education program. The lectures consisted of 20 min of understanding of PCC and case study analysis, 15 min of empathy training, and 15 min of communication training for each session ( ). The content validity of the developed comprehensive PCC education program was evaluated. The content on understanding of PCC and the case studies were reviewed by a professor with experience in PCC research and two nursing students, and the empathy and communication training contents were reviewed by three professors with experience in administering relevant education and two nursing students. 2.5. Intervention The developed comprehensive PCC education program consisted of a total of four sessions (two sessions per week) for 65 min per session. The researcher conducted the program in a quiet lecture room. The students were divided into three-person or four-person groups to promote interaction. In the initial part of the program, the researcher introduced the outlines for the corresponding session and had students share their thoughts on the changes they had during clinical practicum since the preceding session. During hands-on training, the students could practice what they learned during the education, and during sharing of thoughts, they shared their thoughts after each session. Both lecture and practice were included in each session to enhance students’ understanding and utilization of educational contents. In addition, the researcher tried to identify and solve students’ difficulties during practice. The researcher asked students to think about and express what appropriate empathy and communication should be provided to patients and how to address situations through a case study. However, whether the students understood the contents of each session well was not objectively evaluated in each session. The lectures were given according to the manual, and the students were given educational materials and a book containing PCC cases. The education program was administered by the researcher, who has experience in research on communication-enhancing programs and PCC and has translated a book on PCC. 2.6. Measurements Participants’ sex, age, religion, health status, satisfaction with college, satisfaction with nursing major, and academic performance in the preceding semester were surveyed. Individualized care, empathy, communication competence, and clinical practice stress were measured immediately before the intervention and two weeks after the intervention using the self-report questionnaires. The pre-test was written based on students’ experience of providing care to patients during clinical practice before the comprehensive PCC education program, and the post-test was written based on the experience of providing care to patients during the on-campus practicum and the clinical practicum in the hospital after the completion of the education program. Individualized care was assessed using the Individualized Care Scale-Nurse A version (ICS-A-Nurse) developed by Suhonen et al. and adapted to Korean for use among nursing students by Park . The Individualized Care Scale measures individualized patient care based on nurses’ perceptions and includes the following: (1) assessment of patients’ needs, preferences, and perceptions; (2) patients’ participation in their care; and (3) care based on patient’s individualized information , which are key elements of person-centered care. This 17-item instrument uses a 5-point Likert scale, and a higher score indicates greater individualized care. The Cronbach’s α was 0.89 in the study by Park and 0.87 in this study. Empathy was assessed using the Jefferson Scale of Empathy-Health Profession Students’ version (JSE-HPS) developed by Hojat et al. and adapted to Korean and validated by Hong . Permission to use the Korean version of the instrument was obtained from Thomas Jefferson University. This instrument contains 20 items (10 negatively worded, 10 positively worded) and uses a 7-point Likert scale, wherein a higher score indicates greater empathy. The Cronbach’s α was 0.88 in the study by Hong and 0.85 in this study. Communication competence was measured using the Global Interpersonal Communication Competence (GICC) scale developed by Hur . This 15-item instrument uses a 5-point Likert scale, and a higher score indicates better communication competence. The Cronbach’s α was 0.72 in the study by Hur and 0.80 in this study. Clinical practice stress was assessed using the instrument developed by Beck and Srivastava and adapted to Korean and modified by Lee and Kim . This 24-item instrument uses a 5-point Likert scale, and a higher score indicates greater clinical practice stress. The Cronbach’s α was 0.91 in the study by Lee and Kim and 0.89 in this study. 2.7. Data Analysis The collected data were analyzed using SPSS version 25.0 (IBM Corp., Armonk, NY, USA). The baseline homogeneity of general characteristics between the experimental and control groups was tested using the chi-squared test, Fisher’s exact test, and independent t-test, and the normality of the dependent variables was analyzed with the Shapiro–Wilk test. The baseline homogeneity of the dependent variables and the effects of the comprehensive PCC education program were analyzed using the Mann–Whitney test or independent t-test depending on the normality of the data. 2.8. Ethical Considerations This study was approved by the Ethics Committee of Kangwon National University (IRB No.: KWNUIRB-2021-04-011-001) prior to data collection and intervention administration. All participants volunteered to participate in the study and were provided detailed explanations about the purpose and procedure of study, guarantee of anonymity of collected data, benefits and risks, confidentiality, and withdrawal from the study both verbally and in writing. The study began after obtaining written informed consent.
This study used a non-equivalent control group non-synchronized design. Conducting the study simultaneously with the experimental and control groups would have elevated the risk of diffusion effect, as participants were undergoing clinical practicum at the same hospital. Therefore, this study used a non-synchronized design, wherein the control group underwent the study first, followed by the experimental group.
The inclusion criteria were (1) third-year nursing students who had experienced clinical training for more than one semester at a general hospital, (2) scheduled for clinical practicum during the study period, and (3) no prior participation in a similar PCC education program. The exclusion criteria were (1) unavailable to undergo at least two weeks of on-campus practicum and clinical practicum in the hospital after the completion of the education program and (2) unable to participate in all sessions of the education program. Sample size was calculated using the G*Power 3.1 software. For independent t-tests with a significance (α) of 0.05, power of 0.8, and effect size of 0.8, the minimum sample size was 26 for each group. The dropout criteria were missing any of the four sessions of the program or failure to complete the pre-test or post-test.
Participants were recruited among third-year nursing students at a nursing school in C city in South Korea through social network service and bulletin board. The researcher anticipated a high dropout rate as the study was conducted during clinical practicum. Considering about 30% dropout, the researcher intended to recruit 68 participants, but 73 participants were enrolled. The recruiting process, educational intervention, and surveys were conducted from 18 May 2021 to 10 December 2021. A third person not involved in the study arbitrarily assigned participants into the experimental group and control group. Participants were given a detailed explanation about the study procedure during the informed consent process; they were aware of which program they were participating in. The researcher administered the education program; therefore, the researcher was also aware of the provided content. To reduce the consequent bias by these problems, this study used a non-synchronized design to ensure that the control group and experimental group did not have any contact with each other, and the researcher followed a manual when providing education to ensure consistency in each session. The experimental group completed the pre-test immediately before beginning the education program and completed the post-test after two weeks of on-campus and clinical practicum following program completion. The control group also completed the post-test after two weeks of on-campus and clinical practicum following the pre-test. The researcher and research assistants administered the pre-test and post-test. Of 73 participants, 6 participants who could not undergo at least 2 weeks of on-campus and clinical practicum following the completion of the education program were excluded. Of 67 participants meeting the criteria, the control group was expected to have a higher dropout rate because they only complete the pre-test and post-test without intervention; therefore, 36 participants were assigned to the control group and 31 were assigned to the experimental group. A total of 7 participants in the control group failed to complete the post-test, resulting in a total of 29 participants, and all 31 participants in the experimental group were included. Thus, a total of 60 participants were included in the final analysis ( ).
Existing studies were reviewed to identify the essential contents for the comprehensive PCC education program. Nursing students need to know the concept and principles of PCC to perform PCC during clinical practicum. Thus, this study reviewed previous studies [ , , , ]. Detailed PCC education that provides practical assistance is needed for nursing students, and case studies help students understand PCC fundamentals . Thus, this study included PCC cases proposed by Price that can be applied in clinical practicum. Moreover, empathy, communication, respect, relationship, and individualized focus are important factors in PCC . This study examined previous research on improving empathy and communication training in nursing students [ , , ]. The program was structured to include the definition of PCC, core premises, and effects of PCC to help students gain an understanding of PCC, broaden their view, and cultivate a positive attitude toward PCC. Case studies of PCC were taken from the book written by Price and translated into Korean. Three cases were included: helping a patient understand their situation (patient with multiple sclerosis), helping a patient learn about the treatment, care, and recovery (pediatric patient with diabetes), and helping a patient cope with changes and anxiety (patient with colorectal cancer). Empathy training included contents about understanding the concept of empathy, knowing one’s emotions, expressing emotions, and listening. The goal of the empathy training was to ensure that the students had an accurate understanding of empathy, knew how to express empathy, recognized their own emotions, and listened correctly to identify others’ emotions. Communication training comprised open-ended questions, affirmations, I-message, and reflective listening. The goal of communication training was to help nursing students to build a relationship of trust with the patient, identify the patient’s values, needs, and preferences, deliver information individually, and help patients actively improve their own lifestyle. This study set the program to 4 sessions with reference to the 2–6 sessions used by a study providing a person-centered dementia care education program to nursing care providers in long-term care facilities and a study providing a PCC education program to clinical nurses . Each session was set to last 65 min based on the 60min session used previously , and each session comprised an introduction (5 min), education (50 min), and hands-on practice and sharing of thoughts (10 min). The educational approaches used were lectures, role play, hands-on practice, and sharing of thoughts based on previous studies . The study parameters were assessed before and after the program using self-report questionnaires. The questionnaires were administered at baseline and after two weeks of on-campus and clinical practicum following the intervention. The educational materials were written in 12-point font size to provide to the participants at each session. Moreover, a training manual for each session was developed to ensure consistency in the education program. The lectures consisted of 20 min of understanding of PCC and case study analysis, 15 min of empathy training, and 15 min of communication training for each session ( ). The content validity of the developed comprehensive PCC education program was evaluated. The content on understanding of PCC and the case studies were reviewed by a professor with experience in PCC research and two nursing students, and the empathy and communication training contents were reviewed by three professors with experience in administering relevant education and two nursing students.
The developed comprehensive PCC education program consisted of a total of four sessions (two sessions per week) for 65 min per session. The researcher conducted the program in a quiet lecture room. The students were divided into three-person or four-person groups to promote interaction. In the initial part of the program, the researcher introduced the outlines for the corresponding session and had students share their thoughts on the changes they had during clinical practicum since the preceding session. During hands-on training, the students could practice what they learned during the education, and during sharing of thoughts, they shared their thoughts after each session. Both lecture and practice were included in each session to enhance students’ understanding and utilization of educational contents. In addition, the researcher tried to identify and solve students’ difficulties during practice. The researcher asked students to think about and express what appropriate empathy and communication should be provided to patients and how to address situations through a case study. However, whether the students understood the contents of each session well was not objectively evaluated in each session. The lectures were given according to the manual, and the students were given educational materials and a book containing PCC cases. The education program was administered by the researcher, who has experience in research on communication-enhancing programs and PCC and has translated a book on PCC.
Participants’ sex, age, religion, health status, satisfaction with college, satisfaction with nursing major, and academic performance in the preceding semester were surveyed. Individualized care, empathy, communication competence, and clinical practice stress were measured immediately before the intervention and two weeks after the intervention using the self-report questionnaires. The pre-test was written based on students’ experience of providing care to patients during clinical practice before the comprehensive PCC education program, and the post-test was written based on the experience of providing care to patients during the on-campus practicum and the clinical practicum in the hospital after the completion of the education program. Individualized care was assessed using the Individualized Care Scale-Nurse A version (ICS-A-Nurse) developed by Suhonen et al. and adapted to Korean for use among nursing students by Park . The Individualized Care Scale measures individualized patient care based on nurses’ perceptions and includes the following: (1) assessment of patients’ needs, preferences, and perceptions; (2) patients’ participation in their care; and (3) care based on patient’s individualized information , which are key elements of person-centered care. This 17-item instrument uses a 5-point Likert scale, and a higher score indicates greater individualized care. The Cronbach’s α was 0.89 in the study by Park and 0.87 in this study. Empathy was assessed using the Jefferson Scale of Empathy-Health Profession Students’ version (JSE-HPS) developed by Hojat et al. and adapted to Korean and validated by Hong . Permission to use the Korean version of the instrument was obtained from Thomas Jefferson University. This instrument contains 20 items (10 negatively worded, 10 positively worded) and uses a 7-point Likert scale, wherein a higher score indicates greater empathy. The Cronbach’s α was 0.88 in the study by Hong and 0.85 in this study. Communication competence was measured using the Global Interpersonal Communication Competence (GICC) scale developed by Hur . This 15-item instrument uses a 5-point Likert scale, and a higher score indicates better communication competence. The Cronbach’s α was 0.72 in the study by Hur and 0.80 in this study. Clinical practice stress was assessed using the instrument developed by Beck and Srivastava and adapted to Korean and modified by Lee and Kim . This 24-item instrument uses a 5-point Likert scale, and a higher score indicates greater clinical practice stress. The Cronbach’s α was 0.91 in the study by Lee and Kim and 0.89 in this study.
The collected data were analyzed using SPSS version 25.0 (IBM Corp., Armonk, NY, USA). The baseline homogeneity of general characteristics between the experimental and control groups was tested using the chi-squared test, Fisher’s exact test, and independent t-test, and the normality of the dependent variables was analyzed with the Shapiro–Wilk test. The baseline homogeneity of the dependent variables and the effects of the comprehensive PCC education program were analyzed using the Mann–Whitney test or independent t-test depending on the normality of the data.
This study was approved by the Ethics Committee of Kangwon National University (IRB No.: KWNUIRB-2021-04-011-001) prior to data collection and intervention administration. All participants volunteered to participate in the study and were provided detailed explanations about the purpose and procedure of study, guarantee of anonymity of collected data, benefits and risks, confidentiality, and withdrawal from the study both verbally and in writing. The study began after obtaining written informed consent.
3.1. Test of Homogeneity for the General Characteristics of Subjects There were no significant differences in sex, age, religion, subjective health status, satisfaction with college life, satisfaction with nursing major, and academic performance in the preceding semester between the two groups, confirming baseline homogeneity in the general characteristics between the two groups ( ). 3.2. Baseline Homogeneity of the Dependent Variables With the exception of individualized care (experimental group W = 0.964, p = 0.363, control group W = 0.919, p = 0.029), all dependent variables, namely empathy (experimental group W = 0.945, p = 0.116, control group W = 0.933, p = 0.067), communication competence (experimental group W = 0.969, p = 0.500, control group W = 0.987, p = 0.965), and clinical practice stress (experimental group W = 0.948, p = 0.139, control group W = 0.971, p = 0.579), were normally distributed. There were no significant differences in these dependent variables between the two groups, confirming baseline homogeneity ( ). 3.3. Effects of Comprehensive PCC Education Program shows the changes after the comprehensive PCC education program. The changes in individualized care, empathy, communication competence, and clinical practice stress scores after the education program significantly differed between the experimental group and the control group. Compared to the control group, individualized care (Z = −2.44, p = 0.024), empathy (t = −2.76, p =0.008), and communication competence (t = −3.25, p = 0.002) were significantly improved. Individualized care scores increased in the experimental group (5.61 ± 11.38) but decreased in the control group (−0.24 ± 6.26). Empathy scores increased in the experimental group (3.29 ± 6.90) but decreased in the control group (−2.96 ± 4.48). Additionally, communication competence scores in the experimental group increased (0.77 ± 4.42) but decreased in the control group (−2.96 ± 4.48). Finally, clinical practice stress (t = 2.90, p = 0.005) was significantly reduced in the experimental group compared to the control group. Clinical practice stress scores decreased in the experimental group (−1.41 ± 7.93) but increased in the control group (4.48 ± 7.81) ( ).
There were no significant differences in sex, age, religion, subjective health status, satisfaction with college life, satisfaction with nursing major, and academic performance in the preceding semester between the two groups, confirming baseline homogeneity in the general characteristics between the two groups ( ).
With the exception of individualized care (experimental group W = 0.964, p = 0.363, control group W = 0.919, p = 0.029), all dependent variables, namely empathy (experimental group W = 0.945, p = 0.116, control group W = 0.933, p = 0.067), communication competence (experimental group W = 0.969, p = 0.500, control group W = 0.987, p = 0.965), and clinical practice stress (experimental group W = 0.948, p = 0.139, control group W = 0.971, p = 0.579), were normally distributed. There were no significant differences in these dependent variables between the two groups, confirming baseline homogeneity ( ).
shows the changes after the comprehensive PCC education program. The changes in individualized care, empathy, communication competence, and clinical practice stress scores after the education program significantly differed between the experimental group and the control group. Compared to the control group, individualized care (Z = −2.44, p = 0.024), empathy (t = −2.76, p =0.008), and communication competence (t = −3.25, p = 0.002) were significantly improved. Individualized care scores increased in the experimental group (5.61 ± 11.38) but decreased in the control group (−0.24 ± 6.26). Empathy scores increased in the experimental group (3.29 ± 6.90) but decreased in the control group (−2.96 ± 4.48). Additionally, communication competence scores in the experimental group increased (0.77 ± 4.42) but decreased in the control group (−2.96 ± 4.48). Finally, clinical practice stress (t = 2.90, p = 0.005) was significantly reduced in the experimental group compared to the control group. Clinical practice stress scores decreased in the experimental group (−1.41 ± 7.93) but increased in the control group (4.48 ± 7.81) ( ).
This study aimed to evaluate the effects of a comprehensive PCC education program on nursing students undergoing clinical practicum in the hospital. No study has yet reported the effectiveness of a comprehensive PCC education program for nursing students in Korea and abroad. Therefore, this study is significant as the first study investigating the effects of a comprehensive PCC education program on individualized care, empathy, communication competence, and clinical practice stress among nursing students undergoing clinical practicum. First, the comprehensive PCC education program significantly improved individualized care compared to the control group. We could not compare our results directly with those in the literature since the study that applied flipped and simulated learning to enhance nursing students’ understanding of PCC . However, it is consistent with the study that showed individualized care improved after an education program based on a design-thinking approach in nursing students . In our study, there are several reasons why the comprehensive PCC education program was effective. First, the fact that the comprehensive PCC education program in this study was effective in enhancing individualized care may be attributable to case study-based education. A theoretical approach to PCC alone does not adequately help nursing students to understand PCC in nursing practice and consider their personal views toward PCC . Thus, we speculate that including case-based learning in PCC education for nursing students was effective. Furthermore, the program simultaneously improved empathy and communication competence, which are the essential factors of PCC , and this is presumed to have contributed to boosting individualized care in nursing students. Finally, the students were given an opportunity to apply what they had learned in the education program during on-campus and clinical practicum for two weeks, which also would have contributed to enhancing individualized care. Second, the comprehensive PCC education program also significantly enhanced empathy compared to the control group, and empathy is an essential attribute for nursing students to promote PCC, owing to its strong association with PCC competence . Understanding the patient’s stance and expressing empathy are key features of PCC . Korean nursing curricula include one semester of a communication course in the first year, and although students learn about empathy as part of this course, it is difficult to enhance empathy only through a short class. Thus, various scenarios of empathy applicable to clinical practicum should be developed to train students. In a systematic review of the literature on empathy education for nursing students , the most effective empathy education was indicated to be involved immersive and experiential simulation-based interventions. The results of this study highlight the importance of not only theoretical education about empathy but also its practice directly through case studies. Third, the comprehensive PCC education program significantly improved communication competence compared to the control group. Effective communication between patients and nurses is an essential requirement for PCC . Among the five aspects of professional nursing competence in nursing students, communication was most strongly associated with PCC competence . In this sense, an improvement of communication skills after completing a comprehensive PCC education program is meaningful. As previously mentioned, nursing students in Korea take a communication course only in their first year of school. However, first-year students lack a deep understanding of the nursing discipline and person-centered care; therefore, it would be more effective to provide intensive communication education during the third year, in which students begin their clinicals, such that they can effectively communicate with patients with confidence. Moreover, students should be continuously trained for communication skills applicable in future clinical settings. Finally, the comprehensive PCC education program significantly reduced nursing students’ clinical practice stress compared to the control group. Nursing students have been reported to be under heavy stress during clinical practicum, primarily due to fear of unknown situations, lack of competence, and lack of control in their relationship with patients . The program seemed to help students provide patient care with more confidence and effectively communicate with patients, thereby reducing their clinical practice stress. Previous findings that PCC increases job satisfaction while reducing stress in healthcare providers partially support our result. This study is significant as the first study in Korea and abroad to implement and evaluate a comprehensive PCC education program in nursing students. Moreover, the fact that we developed a feasible comprehensive PCC education program for nursing students undergoing clinical practicum adds to the significance of this study. Nursing students who completed the education program applied what they had learned into practice during on-campus and clinical practicum and evaluated the effectiveness of the program themselves, thereby improving the accuracy of the assessment. Finally, this study showed that a program that simultaneously boosts empathy and communication competence as well as PCC through theoretical and case-based education is effective, thereby establishing evidence supporting the implementation of a comprehensive intervention. This study has a few limitations. We could not use a randomized experimental design and used a non-synchronized design to prevent any diffusion effects between the experimental group and control group. Hence, bias caused by various exogenous variables that might have occurred during this period could not be controlled. Moreover, we conducted the comprehensive PCC education program at one university in Korea; therefore, these results have limitations in generalizing the findings. Lastly, although we administered the post-test after the students had about two weeks of on-campus and clinical practicum following the completion of the intervention, we did not use a longitudinal design and thus could not evaluate whether the effects are retained in the long term. In the future, longitudinal randomized controlled trials are needed to substantiate the findings of this study. Moreover, replication studies are needed with nursing students from diverse universities. Finally, studies should develop more patient scenarios for case-based learning, an essential component of comprehensive PCC education, and assess the effects of such education.
This study developed, implemented, and evaluated the effects of a comprehensive PCC education program in nursing students undergoing clinical practicum. The program improved nursing students’ PCC competence, empathy, and communication competence while reducing their clinical practice stress. Thus, a comprehensive PCC education program could be utilized as a pre-practicum intervention for nursing students to boost their confidence and promote PCC practice during clinical practicum.
|
Viral and Bacterial Communities Collaborate through Complementary Assembly Processes in Soil to Survive Organochlorine Contamination | 6b5fcad0-163f-4ad4-b9c7-84c9dbd162e7 | 10056961 | Microbiology[mh] | Bacterial viruses (here referred to as viruses), which are the most abundant and diverse entities on earth, play a key role in the ecological and evolutionary processes of microbial communities by directly killing hosts or integrating into host genomes without lysing them ( ). During lytic infection, viruses directly lyse host bacterial cells in a short period, exerting “top-down” regulation on bacterial communities ( ). However, viruses can also integrate into their host’s bacterial genome and then redirect host bacterial metabolic activities through virus-encoded auxiliary metabolic genes (AMGs) ( ). These virus-encoded AMGs have been widely recognized as impacting diverse metabolic activities in the soil, such as nutrient element transformation, energy metabolism, and contaminant biodegradation ( ). For instance, the organochlorine pesticide-degrading gene L-DEX, atrazine-degrading gene trzN , and various chromium (Cr)-detoxifying genes have been identified as virus-encoded AMGs in contaminated soils ( , , ). Moreover, a greater diversity of virus-encoded AMGs has been shown to help bacterial hosts survive in contaminated environments ( ). Therefore, it is essential to explore the role of viral communities as ecological drivers that shape microbial communities in contaminated soils. Microbial community assembly is the process of species colonization and interaction to establish and maintain local communities. This process, which occurs through continuous migration from the regional species pool, is driven by stochastic and deterministic processes ( ). The deterministic factors include environmental filtering of abiotic conditions and biotic interactions (e.g., competition, exploitation, mutualism, predation, and host filtering), while the stochastic factors consist of unpredictable birth, death, reproduction, and diversification ( ). The process of bacterial community assembly varies according to the types and extent of pollution in soils. A previous study found that the assembly of rare and abundant bacterial communities was dominated by different processes in oil-contaminated soils. Specifically, while the abundant bacterial subcommunity was influenced mainly by edaphic factors (i.e., deterministic processes), the assembly of the rare subcommunity was dominated by stochastic processes ( ). In another study, both deterministic and stochastic processes drove the succession of a bacterial community under polychlorinated biphenyls stress, indicating the dynamic nature of bacterial community assembly ( ). Recent studies have explored the assembly of viral communities to understand the comprehensive response of microbial communities to environmental stresses. Danczak et al. ( ) revealed that the viral community was influenced by different assembly processes with time in fractured shale ecosystems and that the processes of viral and bacterial community assembly were interrelated. However, compared with bacterial community assembly processes, the factors that drive the assembly of viral communities and their interactions with bacterial communities in contaminated soils are less well understood. Organochlorine pesticides (OCPs) are synthetic pesticides with broad applications in agricultural and chemical industries, which have been detected frequently worldwide in soils ( , ). Due to their high toxicity and recalcitrance, soils contaminated by OCPs have been of great concern to human health and environmental security ( , ). Organochlorine pesticide exposure exerts direct selection on soil bacterial communities. While bacterial taxa that are less resistant to OCPs become less abundant or disappear under heavy pollutant exposure, OCP-resistant bacterial taxa increase their abundance due to a competitive advantage ( ). Our previous study suggested that viral communities played crucial roles in the bacterial survival of heavy OCP contamination (total pesticide content varying from 1,083.7 ± 40.4 to 4,595.8 ± 344.0 mg · kg −1 ) and the degradation of OCPs through virus-encoded AMGs in soils ( ). However, the presence of genes is organism dependent, and viral communities are regulated by host population dynamics or phage defense systems. Therefore, we hypothesize that the assembly process at the taxonomic and functional levels of bacteria and virus in OCP-exposed soils was consistent and that bacterial community assembly was driven by a deterministic process, while viral communities were dominated by stochastic assembly under OCP exposure. Moreover, viral communities could enable the deterministic assembly of bacterial communities due to the potential benefits for the host bacteria. In this study, we used a combination of metagenomics/viromics and bioinformatics approaches to explore how OCP exposure affects the composition of both bacterial and viral communities in a closed OCP-contaminated site in the Yangtze River Delta, China. Variance in the assembly processes of bacterial taxa and functional genes, as well as viral taxa and AMGs, was investigated. We applied the normalized stochasticity ratio to investigate the bacterial and viral community assembly process at the taxon and gene levels in OCP-contaminated soil. The dominant processes that drove the assembly of bacterial and viral communities in heavily OCP-stressed soils were deterministic and stochastic, respectively. Moreover, the stochastic assembly of viruses and AMGs was found to help maintain functional redundancy of the bacterial community. Taken together, our findings suggest the need to study assembly processes at multiple trophic levels to comprehensively understand the ecological drivers of microbial communities under contaminant exposure.
Bacterial profiles in clean and OCP-contaminated soils. To determine the influence of OCP exposure on bacterial communities at different levels, we compared the relative abundance, alpha- and beta-diversity of bacterial taxa, and genes between clean and OCP-contaminated soil. Metagenomics analyses revealed that bacterial taxa and genes had distinct characteristics related to their relative abundances across the soils. We found distinct bacterial families in OCP-contaminated soils, such as unclassified Actinobacteria and unclassified Proteobacteria and Comamonadaceae , and OCP exposure was associated with increased relative abundances of Nocardioidaceae (clean, 0.9%; light, 5.1%; and heavy, 4.9%), Streptomycetaceae (clean, 1.3%; light, 4.6%; and heavy, 4.6%), Microbacteriaceae (clean, 0.2%; light, 3.4%; and heavy, 3.0%), Burkholderiaceae (clean, 0.7%; light, 3.0%; and heavy, 3.0%), Micromonosporaceae (clean, 0.8%; light, 2.7%; and heavy, 2.6%), and Conexibacteraceae (clean, 0.2%; light, 2.8%; and heavy, 2.8%) ( ; see Table S1_taxonomy in the supplemental material). In addition, the relative abundances of organohalide-respiring bacteria (OHRBs), such as Enterobacteriaceae , Desulfovibrio , Pseudomonas ( ), and Dehalococcoides ( ), in OCP-contaminated soils were significantly higher than those in clean soils (one-way analysis of variance [ANOVA], P < 0.05). In contrast, the relative abundance of bacterial functional genes annotated by the Kyoto Encyclopedia of Genes and Genomes (KEGG) database remained stable across soils ( ; Table S1_KEGG annotation). Lower richness (clean, 2,896.3; light, 1,475.7; and heavy, 1,468.3 for bacterial taxa; clean, 6,053.3; light, 4,948.0; and heavy, 4,991.3 for bacterial genes) and ACE indexes (clean, 15.2; light, 14.0; and heavy, 13.9 for bacterial taxa; clean, 40.9; light, 37.3; and heavy, 37.5 for bacterial genes) but higher Simpson’s (clean, 0.9; light, 1.0; and heavy, 1.0 for bacterial taxa; clean, 1.0; light, 1.0; and heavy, 1.0 for bacterial genes) and Pielou indexes (clean, 0.6; light, 0.7; and heavy, 0.7 for bacterial taxa; clean, 0.9; light, 0.9; and heavy, 0.9 for bacterial genes) of bacterial taxa and genes were detected in OCP-contaminated soils ( ), suggesting decreased richness but increased evenness of bacterial communities due to OCP exposure. We also used nonmetric multidimensional scaling (NMDS) to study the Bray-Curtis distance of bacterial taxa and functional gene composition among treatments. Bacterial taxonomic and genetic composition clearly differed between clean and OCP-contaminated soils but not between light and heavy contaminated soils (Adonis analysis, P < 0.05) ( ). Mantel analysis indicated a significant association between the composition of bacterial taxa and functional genes (Mantel test, P < 0.05). Viral profiles in clean and OCP-contaminated soils. Siphoviridae was the most abundant family in clean (92.0%) and OCP-contaminated soils (62.2% on average). OCP exposure was associated with increased relative abundances of Podoviridae (clean, 0.3%; light, 15.0%; and heavy, 13.5%), Myoviridae (clean, 1.4%; light, 6.7%; and heavy, 7.4%), and Autographiviridae (clean, 0.1%; light, 1.1%; and heavy, 1.0%) but decreased relative abundances of Siphoviridae (clean, 92.0%; light, 61.7%; and heavy, 62.7%) (see Table S2 in the supplemental material). Notably, Schitoviridae and Demerecviridae were detected only in OCP-contaminated soils ( ). Specifically, the relative abundance of virulent viruses (average 22.6%) in OCP-contaminated soil was higher than that in clean soil (average, 14.5%) (one-way ANOVA, P < 0.05) ( ), and the relative abundance of virulent viruses did not clearly differ between light and heavy contamination (one-way ANOVA, P > 0.05) ( ). The alpha diversity of viral taxa indicated higher richness, ACE, Simpson, and Pielou indexes in OCP-contaminated soil ( ). The viral taxa clustered together according to the OCP contamination gradient, but the viral taxa in clean and OCP-contaminated soils separated into two subgroups (see Fig. S1 in the supplemental material). We annotated 3,277 viral AMG sequences in total to explore the effects of viruses on bacterial metabolism (see Table S3 in the supplemental material). We used both KEGG and CAZy databases for viral AMG annotation, and the best hits (defined by bit score, default minimum threshold of 60) reported for each database were provided in a single output file to avoid overlap. According to the KEGG database, most AMGs were annotated as “nucleotide metabolism,” “amino acid metabolism,” and “carbohydrate metabolism” ( ). In addition, virus-encoded AMGs in OCP-contaminated soil (55 gene classes) linked to nutrient cycling (carbon [C], nitrogen [N], phosphorus [P], and sulfur [S]) and pesticide degradation are more diverse and exclusive than those in clean soil (33 gene classes) (Table S3). Specifically, the cysH gene associated with hydrogen sulfide metabolism and two genes associated with pesticide degradation (aldehyde dehydrogenase [ALDH] and l -2-haluric acid dehalogenase [EC: 3.8.1.2], which are responsible for the transformation of chlorobenzene and chloroalkene, respectively) were found only in OCP-contaminated soils. The virus-encoded AMGs cover a range of bacterial metabolic activities and pesticide degradation pathways, indicating that the AMGs can enhance the bacterial community nutrient metabolism and degradation of OCPs, thereby benefiting bacterial survival in OCP-contaminated soils. Similar to viral taxa, alpha diversity (richness, ACE, Simpson, and Pielou indexes) of viral AMGs was higher in OCP-contaminated soil than that in clean soil. Beta diversity analysis indicated that clean and OCP-contaminated soils had different AMG compositions (Fig. S1). Environmental factors drove the variation in bacterial and viral community structure. To investigate the impact of abiotic factors on microbial communities, variance partitioning analysis (VPA) was used to determine the explanatory degree of different abiotic factors on bacterial and viral communities at the taxon and gene levels. Single abiotic factors and their combinations that can best explain the variance in bacterial and viral compositions were screened out (see Table S4 in the supplemental material). The explanatory degree of single abiotic factors was <20% on average, with pH making the greatest contribution to the variance in viral taxonomic and viral genetic composition (19.4% and 13.7%, respectively). The combination of pH, cation exchange capacity (CEC), and m -nitrochlorobenzene had the highest explanatory degree (97.3%) for bacterial taxonomic variance. Combinations of pH, CEC, and total phosphorus (TP) were associated most closely with viral taxa (70.7%); pH, CEC, and total nitrogen (TN) were associated most closely with bacterial genes (81.5%); and pH, TP, and m -nitrochlorobenzene were associated most closely with viral AMGs (78.6%; Fig. S2A). Together, the compositions of bacterial taxa, viral taxa, and their genes were simultaneously under the impact of multiple abiotic factors across soils. Community assemblies of bacterial taxa and genes. The normalized stochasticity ratio (NST) was determined to investigate the community assembly processes affecting both bacterial and viral communities ( ). The normalized stochasticity ratio reflects the relative importance of stochasticity in community assembly based on magnitude rather than on the significance of the difference between observed and null expectation as a quantitative measure of stochasticity. In this work, the NST was used to quantify the relative importance of deterministic (<50%) and stochastic processes (>50%) in community assembly ( ). For bacterial community assembly, the relative importance of stochastic processes was 74.1%, 7.3%, and 6.8% in clean, light, and heavy contaminated soils, respectively, indicating that the relative importance of stochastic process decreased with elevated OCP contamination gradient in soils. Specifically, bacterial community assembly was dominated by stochastic processes in clean soil but driven by deterministic processes in OCP-contaminated soils. Similar to bacterial community assembly, the relative importance of stochastic processes on bacterial gene assembly decreased with increasing contamination levels, which were 48.5%, 13.7%, and 8.9% in clean, light, and heavy contaminated soils, respectively. Because the NST values were all lower than 50%, the bacterial gene assembly was dominated by deterministic processes in clean and OCP-contaminated soils, and the significance of deterministic processes increased with elevated OCP contamination. Additionally, the migration rate of bacterial taxa in OCP-contaminated soils was significantly lower than that in clean soil (one-way ANOVA, P < 0.05) ( ) but did not vary significantly between light and heavy contaminated soils (one-way ANOVA, P > 0.05) ( ). Conversely, the migration rate of bacterial genes in OCP-contaminated soil was significantly higher than that in clean soil (one-way ANOVA, P < 0.05) ( ) and could not be distinguished between light and heavy contaminated soils (one-way ANOVA, P > 0.05) ( ). The niche breadth of bacterial taxa in OCP-contaminated soil was significantly higher than that in clean soil (16.2 ± 0.6) but did not differ significantly between light (58.8 ± 0.7) and heavy contaminated soils (59.2 ± 1.2) (one-way ANOVA, P > 0.05) ( ). The niche breadth of bacterial genes in OCP-contaminated soil was higher than that in clean soil, but the differences in the niche breadth of bacterial functional genes in different contamination gradients were not significant (one-way ANOVA, P > 0.05) ( ). OCP exposure exerted strong selection that decreased bacterial community mobility and increased the niche breadth of surviving bacteria on the bacterial community assembly process. Community assemblies of viral taxa and virus-encoded AMGs. To investigate the impact of OCP exposure on the assembly of viral taxa and AMGs in soils, we also determined the NST, migration rate, and niche breadth of viral communities between clean and OCP-contaminated soils. The relative importance of stochastic process to viral taxa assembly was 26.8%, 87.8%, and 78.3% in clean, light, and heavy contaminated soils, respectively, indicating that OCP exposure enhanced the relative importance of stochastic processes in the assembly of viral communities. A similar trend was also observed for the assembly of viral AMGs. As a result, the assembly of viral taxa and AMGs was dominated by deterministic processes in clean soil but by stochastic processes in OCP-contaminated soils ( ). Stochastic processes had strong effects on the viral community and AMG composition. The migration rate of viral taxa followed the order of clean soil < light contamination < heavy contamination (one-way ANOVA, P < 0.05) ( ), indicating that OCP pollution significantly increased the migration rate of viral taxa in soil. Similarly, OCP exposure increased the migration rate of viral AMGs from 0.62 in clean soil to 0.66 in OCP-contaminated soils. The niche breadth of viral taxa and AMGs in OCP-contaminated soils was higher than in clean soil (one-way ANOVA, P < 0.05) ( ). OCP exposure significantly increased the migration rate and niche breadth of viruses and AMG, while also inducing stochastic-dominated assembly processes. Putative virus-host linkage in clean and OCP-contaminated soils. We used three approaches, including tRNA sequence alignment, comparison of clustered regularly interspaced short palindromic repeats (CRISPR) spacers, and JGI database matching to explore virus-host linkage. Overall, 4,041 viral sequences were linked with 10,932 putative bacterial host sequences (belonging to 28 bacterial phyla) across the soils. Siphoviridae had the broadest host range, which was linked to 21 bacterial phyla, including the 10 most abundant bacterial phyla ( ). The remaining virus families consisted of Podoviridae (15 bacterial phyla), Myoviridae (13 bacterial phyla), Autographiviridae (3 bacterial phyla), Phycodnaviridae (2 bacterial phyla), Herelleviridae (2 bacterial phyla), Schitoviridae (1 bacterial phyla), Chaseviridae (1 bacterial phyla), and Ackermannviridae (1 bacterial phyla). Phycodnaviridae , Herelleviridae , Schitoviridae , Ackermannviridae , and Chaseviridae were detected only in OCP-contaminated soil. The relative abundance of the 10 most abundant bacterial phylum hosts were associated with 4,017 viral sequences (about 99.4% of all viral sequences matched the host). Of these hosts, the most abundant ( Proteobacteria ) was associated with 1,409 viral sequences (~34.9% of the overall viral sequences). Together, viruses exerted great impact on all dominant bacteria except for “ Candidatus Rokubacteria,” which was infected only by one viral sequence. We defined viruses linking 10 bacterial phyla as broad-host viruses. Overall, Siphoviridae were classified as broad-host viruses in clean and OCP-contaminated soils, while Podoviridae and Myoviridae were considered broad-host viruses only in OCP-contaminated soils (see Table S5_Virus-host linkages in the supplemental material). Therefore, viruses in OCP-contaminated soil were linked to more diverse bacterial hosts than those in clean soil.
To determine the influence of OCP exposure on bacterial communities at different levels, we compared the relative abundance, alpha- and beta-diversity of bacterial taxa, and genes between clean and OCP-contaminated soil. Metagenomics analyses revealed that bacterial taxa and genes had distinct characteristics related to their relative abundances across the soils. We found distinct bacterial families in OCP-contaminated soils, such as unclassified Actinobacteria and unclassified Proteobacteria and Comamonadaceae , and OCP exposure was associated with increased relative abundances of Nocardioidaceae (clean, 0.9%; light, 5.1%; and heavy, 4.9%), Streptomycetaceae (clean, 1.3%; light, 4.6%; and heavy, 4.6%), Microbacteriaceae (clean, 0.2%; light, 3.4%; and heavy, 3.0%), Burkholderiaceae (clean, 0.7%; light, 3.0%; and heavy, 3.0%), Micromonosporaceae (clean, 0.8%; light, 2.7%; and heavy, 2.6%), and Conexibacteraceae (clean, 0.2%; light, 2.8%; and heavy, 2.8%) ( ; see Table S1_taxonomy in the supplemental material). In addition, the relative abundances of organohalide-respiring bacteria (OHRBs), such as Enterobacteriaceae , Desulfovibrio , Pseudomonas ( ), and Dehalococcoides ( ), in OCP-contaminated soils were significantly higher than those in clean soils (one-way analysis of variance [ANOVA], P < 0.05). In contrast, the relative abundance of bacterial functional genes annotated by the Kyoto Encyclopedia of Genes and Genomes (KEGG) database remained stable across soils ( ; Table S1_KEGG annotation). Lower richness (clean, 2,896.3; light, 1,475.7; and heavy, 1,468.3 for bacterial taxa; clean, 6,053.3; light, 4,948.0; and heavy, 4,991.3 for bacterial genes) and ACE indexes (clean, 15.2; light, 14.0; and heavy, 13.9 for bacterial taxa; clean, 40.9; light, 37.3; and heavy, 37.5 for bacterial genes) but higher Simpson’s (clean, 0.9; light, 1.0; and heavy, 1.0 for bacterial taxa; clean, 1.0; light, 1.0; and heavy, 1.0 for bacterial genes) and Pielou indexes (clean, 0.6; light, 0.7; and heavy, 0.7 for bacterial taxa; clean, 0.9; light, 0.9; and heavy, 0.9 for bacterial genes) of bacterial taxa and genes were detected in OCP-contaminated soils ( ), suggesting decreased richness but increased evenness of bacterial communities due to OCP exposure. We also used nonmetric multidimensional scaling (NMDS) to study the Bray-Curtis distance of bacterial taxa and functional gene composition among treatments. Bacterial taxonomic and genetic composition clearly differed between clean and OCP-contaminated soils but not between light and heavy contaminated soils (Adonis analysis, P < 0.05) ( ). Mantel analysis indicated a significant association between the composition of bacterial taxa and functional genes (Mantel test, P < 0.05).
Siphoviridae was the most abundant family in clean (92.0%) and OCP-contaminated soils (62.2% on average). OCP exposure was associated with increased relative abundances of Podoviridae (clean, 0.3%; light, 15.0%; and heavy, 13.5%), Myoviridae (clean, 1.4%; light, 6.7%; and heavy, 7.4%), and Autographiviridae (clean, 0.1%; light, 1.1%; and heavy, 1.0%) but decreased relative abundances of Siphoviridae (clean, 92.0%; light, 61.7%; and heavy, 62.7%) (see Table S2 in the supplemental material). Notably, Schitoviridae and Demerecviridae were detected only in OCP-contaminated soils ( ). Specifically, the relative abundance of virulent viruses (average 22.6%) in OCP-contaminated soil was higher than that in clean soil (average, 14.5%) (one-way ANOVA, P < 0.05) ( ), and the relative abundance of virulent viruses did not clearly differ between light and heavy contamination (one-way ANOVA, P > 0.05) ( ). The alpha diversity of viral taxa indicated higher richness, ACE, Simpson, and Pielou indexes in OCP-contaminated soil ( ). The viral taxa clustered together according to the OCP contamination gradient, but the viral taxa in clean and OCP-contaminated soils separated into two subgroups (see Fig. S1 in the supplemental material). We annotated 3,277 viral AMG sequences in total to explore the effects of viruses on bacterial metabolism (see Table S3 in the supplemental material). We used both KEGG and CAZy databases for viral AMG annotation, and the best hits (defined by bit score, default minimum threshold of 60) reported for each database were provided in a single output file to avoid overlap. According to the KEGG database, most AMGs were annotated as “nucleotide metabolism,” “amino acid metabolism,” and “carbohydrate metabolism” ( ). In addition, virus-encoded AMGs in OCP-contaminated soil (55 gene classes) linked to nutrient cycling (carbon [C], nitrogen [N], phosphorus [P], and sulfur [S]) and pesticide degradation are more diverse and exclusive than those in clean soil (33 gene classes) (Table S3). Specifically, the cysH gene associated with hydrogen sulfide metabolism and two genes associated with pesticide degradation (aldehyde dehydrogenase [ALDH] and l -2-haluric acid dehalogenase [EC: 3.8.1.2], which are responsible for the transformation of chlorobenzene and chloroalkene, respectively) were found only in OCP-contaminated soils. The virus-encoded AMGs cover a range of bacterial metabolic activities and pesticide degradation pathways, indicating that the AMGs can enhance the bacterial community nutrient metabolism and degradation of OCPs, thereby benefiting bacterial survival in OCP-contaminated soils. Similar to viral taxa, alpha diversity (richness, ACE, Simpson, and Pielou indexes) of viral AMGs was higher in OCP-contaminated soil than that in clean soil. Beta diversity analysis indicated that clean and OCP-contaminated soils had different AMG compositions (Fig. S1).
To investigate the impact of abiotic factors on microbial communities, variance partitioning analysis (VPA) was used to determine the explanatory degree of different abiotic factors on bacterial and viral communities at the taxon and gene levels. Single abiotic factors and their combinations that can best explain the variance in bacterial and viral compositions were screened out (see Table S4 in the supplemental material). The explanatory degree of single abiotic factors was <20% on average, with pH making the greatest contribution to the variance in viral taxonomic and viral genetic composition (19.4% and 13.7%, respectively). The combination of pH, cation exchange capacity (CEC), and m -nitrochlorobenzene had the highest explanatory degree (97.3%) for bacterial taxonomic variance. Combinations of pH, CEC, and total phosphorus (TP) were associated most closely with viral taxa (70.7%); pH, CEC, and total nitrogen (TN) were associated most closely with bacterial genes (81.5%); and pH, TP, and m -nitrochlorobenzene were associated most closely with viral AMGs (78.6%; Fig. S2A). Together, the compositions of bacterial taxa, viral taxa, and their genes were simultaneously under the impact of multiple abiotic factors across soils.
The normalized stochasticity ratio (NST) was determined to investigate the community assembly processes affecting both bacterial and viral communities ( ). The normalized stochasticity ratio reflects the relative importance of stochasticity in community assembly based on magnitude rather than on the significance of the difference between observed and null expectation as a quantitative measure of stochasticity. In this work, the NST was used to quantify the relative importance of deterministic (<50%) and stochastic processes (>50%) in community assembly ( ). For bacterial community assembly, the relative importance of stochastic processes was 74.1%, 7.3%, and 6.8% in clean, light, and heavy contaminated soils, respectively, indicating that the relative importance of stochastic process decreased with elevated OCP contamination gradient in soils. Specifically, bacterial community assembly was dominated by stochastic processes in clean soil but driven by deterministic processes in OCP-contaminated soils. Similar to bacterial community assembly, the relative importance of stochastic processes on bacterial gene assembly decreased with increasing contamination levels, which were 48.5%, 13.7%, and 8.9% in clean, light, and heavy contaminated soils, respectively. Because the NST values were all lower than 50%, the bacterial gene assembly was dominated by deterministic processes in clean and OCP-contaminated soils, and the significance of deterministic processes increased with elevated OCP contamination. Additionally, the migration rate of bacterial taxa in OCP-contaminated soils was significantly lower than that in clean soil (one-way ANOVA, P < 0.05) ( ) but did not vary significantly between light and heavy contaminated soils (one-way ANOVA, P > 0.05) ( ). Conversely, the migration rate of bacterial genes in OCP-contaminated soil was significantly higher than that in clean soil (one-way ANOVA, P < 0.05) ( ) and could not be distinguished between light and heavy contaminated soils (one-way ANOVA, P > 0.05) ( ). The niche breadth of bacterial taxa in OCP-contaminated soil was significantly higher than that in clean soil (16.2 ± 0.6) but did not differ significantly between light (58.8 ± 0.7) and heavy contaminated soils (59.2 ± 1.2) (one-way ANOVA, P > 0.05) ( ). The niche breadth of bacterial genes in OCP-contaminated soil was higher than that in clean soil, but the differences in the niche breadth of bacterial functional genes in different contamination gradients were not significant (one-way ANOVA, P > 0.05) ( ). OCP exposure exerted strong selection that decreased bacterial community mobility and increased the niche breadth of surviving bacteria on the bacterial community assembly process.
To investigate the impact of OCP exposure on the assembly of viral taxa and AMGs in soils, we also determined the NST, migration rate, and niche breadth of viral communities between clean and OCP-contaminated soils. The relative importance of stochastic process to viral taxa assembly was 26.8%, 87.8%, and 78.3% in clean, light, and heavy contaminated soils, respectively, indicating that OCP exposure enhanced the relative importance of stochastic processes in the assembly of viral communities. A similar trend was also observed for the assembly of viral AMGs. As a result, the assembly of viral taxa and AMGs was dominated by deterministic processes in clean soil but by stochastic processes in OCP-contaminated soils ( ). Stochastic processes had strong effects on the viral community and AMG composition. The migration rate of viral taxa followed the order of clean soil < light contamination < heavy contamination (one-way ANOVA, P < 0.05) ( ), indicating that OCP pollution significantly increased the migration rate of viral taxa in soil. Similarly, OCP exposure increased the migration rate of viral AMGs from 0.62 in clean soil to 0.66 in OCP-contaminated soils. The niche breadth of viral taxa and AMGs in OCP-contaminated soils was higher than in clean soil (one-way ANOVA, P < 0.05) ( ). OCP exposure significantly increased the migration rate and niche breadth of viruses and AMG, while also inducing stochastic-dominated assembly processes.
We used three approaches, including tRNA sequence alignment, comparison of clustered regularly interspaced short palindromic repeats (CRISPR) spacers, and JGI database matching to explore virus-host linkage. Overall, 4,041 viral sequences were linked with 10,932 putative bacterial host sequences (belonging to 28 bacterial phyla) across the soils. Siphoviridae had the broadest host range, which was linked to 21 bacterial phyla, including the 10 most abundant bacterial phyla ( ). The remaining virus families consisted of Podoviridae (15 bacterial phyla), Myoviridae (13 bacterial phyla), Autographiviridae (3 bacterial phyla), Phycodnaviridae (2 bacterial phyla), Herelleviridae (2 bacterial phyla), Schitoviridae (1 bacterial phyla), Chaseviridae (1 bacterial phyla), and Ackermannviridae (1 bacterial phyla). Phycodnaviridae , Herelleviridae , Schitoviridae , Ackermannviridae , and Chaseviridae were detected only in OCP-contaminated soil. The relative abundance of the 10 most abundant bacterial phylum hosts were associated with 4,017 viral sequences (about 99.4% of all viral sequences matched the host). Of these hosts, the most abundant ( Proteobacteria ) was associated with 1,409 viral sequences (~34.9% of the overall viral sequences). Together, viruses exerted great impact on all dominant bacteria except for “ Candidatus Rokubacteria,” which was infected only by one viral sequence. We defined viruses linking 10 bacterial phyla as broad-host viruses. Overall, Siphoviridae were classified as broad-host viruses in clean and OCP-contaminated soils, while Podoviridae and Myoviridae were considered broad-host viruses only in OCP-contaminated soils (see Table S5_Virus-host linkages in the supplemental material). Therefore, viruses in OCP-contaminated soil were linked to more diverse bacterial hosts than those in clean soil.
Bacterial genetic functions remained stable under OCP exposure. In this work, although OCP exposure clearly changed the composition of bacterial taxa, it did not alter the composition of bacterial genes in soils (analysis of similarity [ANOSIM], P > 0.05) ( ). The higher fluctuation of bacterial taxa could have occurred because the diversity of bacterial taxa in the soils was lower than that of bacterial genes ( ). Functional redundancy of the bacterial community contributed to the stable composition of bacterial functional genes, regardless of OCP exposure. The decrease in the diversity of bacterial taxa could be associated with the disappearance of certain species that are sensitive to OCP stress in the soil ( ), which could further impact the functional genes harbored by bacteria. Because of functional redundancy and frequent horizontal gene transfer (HGT) between bacteria, the impact of OCP exposure on bacterium-carried genes is commonly not as great as the impact on bacterial taxa ( , ). In addition, the functional genes are often substitutable, and multiple genes perform the same function ( ). New bases are generated at random sites of genes affected by gene mutation. Because of the degeneracy of codons (that is, multiple codons correspond to the same amino acid), codon changes caused by random gene mutation do not always affect the types of amino acids produced in the translation process; therefore, the function of the protein corresponding to the gene remains the same ( ). Gene mutation (stochastic process) decreased the relative importance of the effects of OCP contamination stress on functional gene composition but did not completely change the gene functional classification. Overall, the gene functional classification remained stable under different OCP contamination stresses. Viral life cycling also facilitated the functional redundancy of host bacteria. Viruses depend on bacterial hosts to complete their own reproduction, either by infecting and lysing host cells (lytic viruses) or by integrating their genes into the hosts’ genome without killing host cells (lysogenic viruses), allowing the virus to act as a vector for horizontal gene transfer between hosts ( , , ). In this study, OCP exposure increased the proportion of lysogenic viruses, suggesting that a higher pool of viruses was involved in the horizontal transfer of functional genes. Moreover, OCP exposure increased the host range of viruses in soils, suggesting higher flexibility of the virus-host interaction and a greater probability of horizontal gene transfer between bacterial communities ( ). The KEGG and CAZy annotations revealed the dominant AMG functional categories, including nucleotide metabolism, amino acid metabolism, and carbohydrate metabolism ( ), which indicate that the viruses were capable of mediating the horizontal transfer of these functions between hosts, thereby contributing to the functional redundancy of bacterial communities under OCP-contaminated soils. Deterministic processes governed the assembly of bacterial communities in OCP-contaminated soils. While community assembly of bacterial taxa has been studied extensively in soil ecosystems, bacterial genes remain less explored. Community assembly is influenced by both deterministic and stochastic processes. Deterministic processes include abiotic conditions (environmental filtering) and biotic interactions (such as competition, exploitation, mutualism, predation, and host filtering), while stochastic processes generally consist of drift, dispersal, and diversification ( , ). We found that bacterial community assembly was dominated by stochastic processes in clean soils and dominated by deterministic processes in OCP-contaminated soils ( ), which is consistent with the results of a previous study that revealed deterministic process drove the assembly of microbial communities in oil-contaminated soils ( ). OHRB-mediated reductive dehalogenation removes halogens from various organohalogens and obtains the energy required for growth ( ). OHRBs are also dominated by deterministic processes in OCP-contaminated soils, indicating the strong selective effect of OCPs on OHRBs. The intense OCP stress is associated with reduced bacterial taxa diversity ( ), suggesting that selection is of major importance in community assembly. The migration rate is the movement of species from a hypothetical species pool to a specific environment ( ). The gradual decrease in bacterial taxa migration in response to OCP exposure indicated that the dispersal limitation decreased the significance of the stochastic process in the microbial assembly pattern in OCP-contaminated soil ( ), which was consistent with the NST results. Additionally, niche breadth is related to species adaption to the environment, indicating the diversity and abundance of resources that species can use and their dispersal ability ( ). In clean soils, weak environmental selective pressures and high taxa diversity resulted in great competition and small niche breadth of bacterial taxa. Conversely, OCP stress led to the disappearance of sensitive taxa, induced the community structure to be more homogeneous, and increased the number of niches available for surviving bacteria in the soil, thereby causing bacterial taxa in OCP-contaminated soils to have a broad niche breadth. The assembly pattern of bacterial taxa and genes is consistent with the “competitive lottery mode” derived from the lottery hypothesis ( ). This hypothesis holds that a specific niche in an ecosystem needs to be occupied by species with corresponding functional genes (the selection of functional genes is a deterministic process), but the niche will be occupied by bacterial species that arrive first (determination of the order of species occupying the ecological niche is stochastic). The functional gene migration rate is the horizontal transfer of genes between bacteria, and the increased gene migration rate in OCP-contaminated soil may be due to increased virus-mediated transduction as described above. Stochastic processes drove the assembly of viral communities in OCP-contaminated soils. Currently, the assembly of viral communities under pollutant exposure is not well understood. Here, viromics analyses revealed that viral community assembly in the OCP-contaminated soil was dominated by stochastic processes ( ). These findings are similar to those of a previous study that showed viral community assembly in salinity-stressed fractured shale ecosystems was driven by nondominant processes, including stochastic processes, such as weak selection, weak dispersal, diversification, and drift ( ). The migration rate of the viral community increased with the degree of contamination, suggesting that the dispersal limitation was constrained and the relative importance of stochastic processes increased in OCP-contaminated soils ( ). Because OCP exposure affected primarily viral communities through its effects on hosts, the hosts play an important role in the assembly of viral communities. In community ecology, the host can be regarded as a resource necessary for the virus survival, which is a special “niche” of viruses. Compared with viruses in clean soil, those in the OCP-contaminated soil had a broader host range and niche breadth ( ). This finding indicates that there are more resources available for viruses and weaker competition within their community in OCP-contaminated soil, so the stochasticity of virus-host encounter rates becomes more important for virus assembly, further reducing the impact of deterministic process on viral community assembly. If most hosts are infected by various viruses (i.e., more niches overlap), the competition between viruses is pretty strong. Based on results of viral-host predictions, we screened 242 specialized host that were infected by only one viral sequence (45 in clean soil and 197 in OCP-contaminated soil), indicating that the virus in the OCP-contaminated soil had more specialized hosts and less niche overlap, which also reduce determinism in the viral community assembly (Table S5_Specialized host). Drift is more important when selection is weak and community size is small ( ). Additionally, lytic viruses cannot be ignored when studying the assembly of viral communities. During the lysis process, the virus releases progeny, which results in a random change in the relative abundance of different virus species within the community. The increase in the proportion of lytic viruses in the OCP-contaminated soil may also facilitate the increased stochasticity of virus community assembly ( ). The ecological drivers that direct the assembly of viral and host bacterial communities are largely unknown, even though viral-encoded accessory genes help host bacteria to survive in polluted environments, such as arsenic-resistance gene ars C/ ars M, organochlorine pesticide degradation gene L-DEX, and atrazine degradation gene trz N ( , , ). Virus-encoded AMGs are a major pathway through which viruses become involved with and redirect host metabolic activities, and studying the assembly process of AMGs is of great importance for understanding how viruses affect host metabolism. The stochastic process of AMG assembly increased with the degree of OCP contamination, which is supported by the variations in AMG migration rate across different contamination gradients, as AMGs had a higher migration rate and lower dispersal limitation in OCP-contaminated soils. Genes are closely related to organisms, and generalist viruses provide greater possibilities for gene replication and expression than experts, so we examined the proportion of generalists or experts in AMG-containing viruses. The relative abundance of generalist AMG-containing phages is significantly higher than that of specialist AMG-containing phages in OCP-contaminated soils (Table S5_AMG-containing phages), and these generalist AMG-containing phages were dominated by stochastic assembly (Table S5_NST of generalists). Therefore, most AMGs in OCP-contaminated soils are stochastically assembled under the influence of the generalist AMG-containing phage assembly process. Considering that diversification is an evolutionary process that generates new genetic variation, higher AMG diversity in OCP-contaminated soils may represent a strong diversification effect ( ). This increased diversification strengthened the effect of drift, further increasing the relative importance of the stochastic assembly process of AMGs in OCP-contaminated soils. Stochasticity of viral community assisted the bacterial community to relieve OCP stress. With the increase in OCP contents, the niche breadth of host bacteria and viruses that can infect the host and the relative abundance of host bacteria both increased (see Fig. S2B in the supplemental material), indicating that the host bacteria have a competitive advantage in OCP-contaminated soil. As described above, viruses facilitated the functional redundancy of host bacteria under OCP contamination. Nevertheless, the role of stochastic processes that dominate the assembly of viral taxa and AMGs needs further discussion. Our previous study revealed that the virus-carried L-2-haloacid dehalogenase gene (L-DEX) can be successfully expressed in Escherichia coli and can degrade l -2-haloacid pesticide precursors, thereby increasing the host ability to tolerate pesticide stress ( ). Other virus-encoded stress-resistant AMGs were also detected and found to play promising roles in host adaptation to contaminant stress in soils, including chromium-resistant genes in chromium-contaminated slag sites and chlorohydrolase gene trz N in atrazine-contaminated soils ( , ). In addition to antistress genes in the virus, the virus can also induce the acquisition of antistress genes by the bacterial community in many ways, and the stochastic assembly process of the virus and AMG also has a direct or indirect impact on bacterial communities in contaminated soil ( ) ( ). The stochastic assembly process of viral communities, which consists of dispersal, diversification, and drift, might facilitate AMG dissemination between hosts. Drift represents changes in the relative abundance of taxa within a community caused by the inherent stochastic processes of unpredictable birth, death, reproduction, and diversification ( ). The increased migration rate of viruses accelerates virus-mediated horizontal gene transfer (HGT) between bacterial hosts, which increases the dissemination of resistant genes within bacterial communities. Virus-mediated HGT plays an important role in bacterial ecology and evolution, which allows bacteria to evolve rapidly and to adapt to changing environmental conditions. Viruses can transfer genetic material between bacteria through generalized, specialized, and lateral transduction, and the widely detected viral AMGs appear to be the result of phages acquiring host metabolic genes through HGT events ( ). Viruses can promote the occurrence of HGT in soil microbial communities under anthropogenic contaminated conditions, such as subinhibitory antibiotic contamination ( ). In our previous work, a high number of polyvalent viruses carrying AMGs, which had more infectible bacterial hosts that could promote HGT of AMGs in bacterial communities, has been detected in OCP-contaminated soil compared with those in clean soil ( ). We also detected the unique organochlorine pesticide degradation AMGs encoding L-DEX and aldehyde dehydrogenase (ALDH) and an increased abundance of viral AMGs linked to carbon, nitrogen, and sulfur metabolism in the OCP-contaminated soils ( ). These AMGs involved in virus-mediated HGT were more diverse and transferred more frequently in OCP-contaminated soil than those in clean soil, suggesting that virus-mediated HGT could help the transfer of AMGs between hosts and expand the adaptation of bacterial communities to adverse conditions. Moreover, the diversification of viral AMGs can be considered a random mutation of genes. As a result, the stochastic dominated AMG assembly process in OCP-contaminated soil indicates a greater effect of diversification on AMG composition, thereby increasing the likelihood of variation in stress-resistant genes. Because of the coevolving and reciprocal interactions between organisms and their environment, viruses reprogram host metabolic activities through virus-carried AMGs to sustain a stable and favorable living environment. Taken together, the stochastic-process-dominated assembly of viral communities facilitates AMG dissemination among host bacteria, which helps to maintain the deterministic assembly of bacterial communities ( ). We suggest that viruses and their assembly processes provide a novel avenue for understanding the bioremediation of contaminated soils. The key functional character of viruses is that they can positively alter host stress resistance in OCP-contaminated soils, which may have a tremendous impact on microbial community dynamics within relatively short time periods, while viruses can stably produce progeny within the host cell. Collectively, our results indicate that the viral community is a crucial biotic factor that cannot be ignored when studying the adaptation of bacterial communities to environmental stress. Conclusions. We investigated the impact of OCP exposure on the taxonomic and genetic composition of viral and bacterial communities in soil using metagenomics and viromics approaches. The bacterial community composition varied significantly under different OCP-contaminated gradients, while the bacterial functional gene composition remained stable. As OCP contamination gradients increased, bacterial taxa and functional gene diversity decreased, while viral taxonomic and AMG diversity increased, resulting in the shift of bacterial community assembly from stochastic-dominant to deterministic-driven processes. In contrast, the assembly of both viral taxa and AMGs changed from deterministic-driven to stochastic-dominant processes in response to OCP exposure in soils. The stochastic assembly of viral communities might contribute to the functional redundancy and increased resistance of bacterial hosts in OCP-contaminated soils. By analyzing viromics data, we identified AMGs in the viral genome, but the expression of AMGs within host cells needs to be further explored. This study is the first one that aimed to understand the assembly process of bacterial and viral community under OCP stress. These findings provide information regarding microbial community responses to OCP stress and reveal the collaborative interaction between viral and bacterial communities to resist pollutant stress, thereby revealing a novel avenue for the application of microbiomes to soil remediation.
In this work, although OCP exposure clearly changed the composition of bacterial taxa, it did not alter the composition of bacterial genes in soils (analysis of similarity [ANOSIM], P > 0.05) ( ). The higher fluctuation of bacterial taxa could have occurred because the diversity of bacterial taxa in the soils was lower than that of bacterial genes ( ). Functional redundancy of the bacterial community contributed to the stable composition of bacterial functional genes, regardless of OCP exposure. The decrease in the diversity of bacterial taxa could be associated with the disappearance of certain species that are sensitive to OCP stress in the soil ( ), which could further impact the functional genes harbored by bacteria. Because of functional redundancy and frequent horizontal gene transfer (HGT) between bacteria, the impact of OCP exposure on bacterium-carried genes is commonly not as great as the impact on bacterial taxa ( , ). In addition, the functional genes are often substitutable, and multiple genes perform the same function ( ). New bases are generated at random sites of genes affected by gene mutation. Because of the degeneracy of codons (that is, multiple codons correspond to the same amino acid), codon changes caused by random gene mutation do not always affect the types of amino acids produced in the translation process; therefore, the function of the protein corresponding to the gene remains the same ( ). Gene mutation (stochastic process) decreased the relative importance of the effects of OCP contamination stress on functional gene composition but did not completely change the gene functional classification. Overall, the gene functional classification remained stable under different OCP contamination stresses. Viral life cycling also facilitated the functional redundancy of host bacteria. Viruses depend on bacterial hosts to complete their own reproduction, either by infecting and lysing host cells (lytic viruses) or by integrating their genes into the hosts’ genome without killing host cells (lysogenic viruses), allowing the virus to act as a vector for horizontal gene transfer between hosts ( , , ). In this study, OCP exposure increased the proportion of lysogenic viruses, suggesting that a higher pool of viruses was involved in the horizontal transfer of functional genes. Moreover, OCP exposure increased the host range of viruses in soils, suggesting higher flexibility of the virus-host interaction and a greater probability of horizontal gene transfer between bacterial communities ( ). The KEGG and CAZy annotations revealed the dominant AMG functional categories, including nucleotide metabolism, amino acid metabolism, and carbohydrate metabolism ( ), which indicate that the viruses were capable of mediating the horizontal transfer of these functions between hosts, thereby contributing to the functional redundancy of bacterial communities under OCP-contaminated soils.
While community assembly of bacterial taxa has been studied extensively in soil ecosystems, bacterial genes remain less explored. Community assembly is influenced by both deterministic and stochastic processes. Deterministic processes include abiotic conditions (environmental filtering) and biotic interactions (such as competition, exploitation, mutualism, predation, and host filtering), while stochastic processes generally consist of drift, dispersal, and diversification ( , ). We found that bacterial community assembly was dominated by stochastic processes in clean soils and dominated by deterministic processes in OCP-contaminated soils ( ), which is consistent with the results of a previous study that revealed deterministic process drove the assembly of microbial communities in oil-contaminated soils ( ). OHRB-mediated reductive dehalogenation removes halogens from various organohalogens and obtains the energy required for growth ( ). OHRBs are also dominated by deterministic processes in OCP-contaminated soils, indicating the strong selective effect of OCPs on OHRBs. The intense OCP stress is associated with reduced bacterial taxa diversity ( ), suggesting that selection is of major importance in community assembly. The migration rate is the movement of species from a hypothetical species pool to a specific environment ( ). The gradual decrease in bacterial taxa migration in response to OCP exposure indicated that the dispersal limitation decreased the significance of the stochastic process in the microbial assembly pattern in OCP-contaminated soil ( ), which was consistent with the NST results. Additionally, niche breadth is related to species adaption to the environment, indicating the diversity and abundance of resources that species can use and their dispersal ability ( ). In clean soils, weak environmental selective pressures and high taxa diversity resulted in great competition and small niche breadth of bacterial taxa. Conversely, OCP stress led to the disappearance of sensitive taxa, induced the community structure to be more homogeneous, and increased the number of niches available for surviving bacteria in the soil, thereby causing bacterial taxa in OCP-contaminated soils to have a broad niche breadth. The assembly pattern of bacterial taxa and genes is consistent with the “competitive lottery mode” derived from the lottery hypothesis ( ). This hypothesis holds that a specific niche in an ecosystem needs to be occupied by species with corresponding functional genes (the selection of functional genes is a deterministic process), but the niche will be occupied by bacterial species that arrive first (determination of the order of species occupying the ecological niche is stochastic). The functional gene migration rate is the horizontal transfer of genes between bacteria, and the increased gene migration rate in OCP-contaminated soil may be due to increased virus-mediated transduction as described above.
Currently, the assembly of viral communities under pollutant exposure is not well understood. Here, viromics analyses revealed that viral community assembly in the OCP-contaminated soil was dominated by stochastic processes ( ). These findings are similar to those of a previous study that showed viral community assembly in salinity-stressed fractured shale ecosystems was driven by nondominant processes, including stochastic processes, such as weak selection, weak dispersal, diversification, and drift ( ). The migration rate of the viral community increased with the degree of contamination, suggesting that the dispersal limitation was constrained and the relative importance of stochastic processes increased in OCP-contaminated soils ( ). Because OCP exposure affected primarily viral communities through its effects on hosts, the hosts play an important role in the assembly of viral communities. In community ecology, the host can be regarded as a resource necessary for the virus survival, which is a special “niche” of viruses. Compared with viruses in clean soil, those in the OCP-contaminated soil had a broader host range and niche breadth ( ). This finding indicates that there are more resources available for viruses and weaker competition within their community in OCP-contaminated soil, so the stochasticity of virus-host encounter rates becomes more important for virus assembly, further reducing the impact of deterministic process on viral community assembly. If most hosts are infected by various viruses (i.e., more niches overlap), the competition between viruses is pretty strong. Based on results of viral-host predictions, we screened 242 specialized host that were infected by only one viral sequence (45 in clean soil and 197 in OCP-contaminated soil), indicating that the virus in the OCP-contaminated soil had more specialized hosts and less niche overlap, which also reduce determinism in the viral community assembly (Table S5_Specialized host). Drift is more important when selection is weak and community size is small ( ). Additionally, lytic viruses cannot be ignored when studying the assembly of viral communities. During the lysis process, the virus releases progeny, which results in a random change in the relative abundance of different virus species within the community. The increase in the proportion of lytic viruses in the OCP-contaminated soil may also facilitate the increased stochasticity of virus community assembly ( ). The ecological drivers that direct the assembly of viral and host bacterial communities are largely unknown, even though viral-encoded accessory genes help host bacteria to survive in polluted environments, such as arsenic-resistance gene ars C/ ars M, organochlorine pesticide degradation gene L-DEX, and atrazine degradation gene trz N ( , , ). Virus-encoded AMGs are a major pathway through which viruses become involved with and redirect host metabolic activities, and studying the assembly process of AMGs is of great importance for understanding how viruses affect host metabolism. The stochastic process of AMG assembly increased with the degree of OCP contamination, which is supported by the variations in AMG migration rate across different contamination gradients, as AMGs had a higher migration rate and lower dispersal limitation in OCP-contaminated soils. Genes are closely related to organisms, and generalist viruses provide greater possibilities for gene replication and expression than experts, so we examined the proportion of generalists or experts in AMG-containing viruses. The relative abundance of generalist AMG-containing phages is significantly higher than that of specialist AMG-containing phages in OCP-contaminated soils (Table S5_AMG-containing phages), and these generalist AMG-containing phages were dominated by stochastic assembly (Table S5_NST of generalists). Therefore, most AMGs in OCP-contaminated soils are stochastically assembled under the influence of the generalist AMG-containing phage assembly process. Considering that diversification is an evolutionary process that generates new genetic variation, higher AMG diversity in OCP-contaminated soils may represent a strong diversification effect ( ). This increased diversification strengthened the effect of drift, further increasing the relative importance of the stochastic assembly process of AMGs in OCP-contaminated soils.
With the increase in OCP contents, the niche breadth of host bacteria and viruses that can infect the host and the relative abundance of host bacteria both increased (see Fig. S2B in the supplemental material), indicating that the host bacteria have a competitive advantage in OCP-contaminated soil. As described above, viruses facilitated the functional redundancy of host bacteria under OCP contamination. Nevertheless, the role of stochastic processes that dominate the assembly of viral taxa and AMGs needs further discussion. Our previous study revealed that the virus-carried L-2-haloacid dehalogenase gene (L-DEX) can be successfully expressed in Escherichia coli and can degrade l -2-haloacid pesticide precursors, thereby increasing the host ability to tolerate pesticide stress ( ). Other virus-encoded stress-resistant AMGs were also detected and found to play promising roles in host adaptation to contaminant stress in soils, including chromium-resistant genes in chromium-contaminated slag sites and chlorohydrolase gene trz N in atrazine-contaminated soils ( , ). In addition to antistress genes in the virus, the virus can also induce the acquisition of antistress genes by the bacterial community in many ways, and the stochastic assembly process of the virus and AMG also has a direct or indirect impact on bacterial communities in contaminated soil ( ) ( ). The stochastic assembly process of viral communities, which consists of dispersal, diversification, and drift, might facilitate AMG dissemination between hosts. Drift represents changes in the relative abundance of taxa within a community caused by the inherent stochastic processes of unpredictable birth, death, reproduction, and diversification ( ). The increased migration rate of viruses accelerates virus-mediated horizontal gene transfer (HGT) between bacterial hosts, which increases the dissemination of resistant genes within bacterial communities. Virus-mediated HGT plays an important role in bacterial ecology and evolution, which allows bacteria to evolve rapidly and to adapt to changing environmental conditions. Viruses can transfer genetic material between bacteria through generalized, specialized, and lateral transduction, and the widely detected viral AMGs appear to be the result of phages acquiring host metabolic genes through HGT events ( ). Viruses can promote the occurrence of HGT in soil microbial communities under anthropogenic contaminated conditions, such as subinhibitory antibiotic contamination ( ). In our previous work, a high number of polyvalent viruses carrying AMGs, which had more infectible bacterial hosts that could promote HGT of AMGs in bacterial communities, has been detected in OCP-contaminated soil compared with those in clean soil ( ). We also detected the unique organochlorine pesticide degradation AMGs encoding L-DEX and aldehyde dehydrogenase (ALDH) and an increased abundance of viral AMGs linked to carbon, nitrogen, and sulfur metabolism in the OCP-contaminated soils ( ). These AMGs involved in virus-mediated HGT were more diverse and transferred more frequently in OCP-contaminated soil than those in clean soil, suggesting that virus-mediated HGT could help the transfer of AMGs between hosts and expand the adaptation of bacterial communities to adverse conditions. Moreover, the diversification of viral AMGs can be considered a random mutation of genes. As a result, the stochastic dominated AMG assembly process in OCP-contaminated soil indicates a greater effect of diversification on AMG composition, thereby increasing the likelihood of variation in stress-resistant genes. Because of the coevolving and reciprocal interactions between organisms and their environment, viruses reprogram host metabolic activities through virus-carried AMGs to sustain a stable and favorable living environment. Taken together, the stochastic-process-dominated assembly of viral communities facilitates AMG dissemination among host bacteria, which helps to maintain the deterministic assembly of bacterial communities ( ). We suggest that viruses and their assembly processes provide a novel avenue for understanding the bioremediation of contaminated soils. The key functional character of viruses is that they can positively alter host stress resistance in OCP-contaminated soils, which may have a tremendous impact on microbial community dynamics within relatively short time periods, while viruses can stably produce progeny within the host cell. Collectively, our results indicate that the viral community is a crucial biotic factor that cannot be ignored when studying the adaptation of bacterial communities to environmental stress.
We investigated the impact of OCP exposure on the taxonomic and genetic composition of viral and bacterial communities in soil using metagenomics and viromics approaches. The bacterial community composition varied significantly under different OCP-contaminated gradients, while the bacterial functional gene composition remained stable. As OCP contamination gradients increased, bacterial taxa and functional gene diversity decreased, while viral taxonomic and AMG diversity increased, resulting in the shift of bacterial community assembly from stochastic-dominant to deterministic-driven processes. In contrast, the assembly of both viral taxa and AMGs changed from deterministic-driven to stochastic-dominant processes in response to OCP exposure in soils. The stochastic assembly of viral communities might contribute to the functional redundancy and increased resistance of bacterial hosts in OCP-contaminated soils. By analyzing viromics data, we identified AMGs in the viral genome, but the expression of AMGs within host cells needs to be further explored. This study is the first one that aimed to understand the assembly process of bacterial and viral community under OCP stress. These findings provide information regarding microbial community responses to OCP stress and reveal the collaborative interaction between viral and bacterial communities to resist pollutant stress, thereby revealing a novel avenue for the application of microbiomes to soil remediation.
Sample collection. Soil samples were collected from a closed organochlorine pesticide factory in Jiangsu Province, China (120.228193′N, 31.758075′E), in 2019. The soil around the organochlorine pesticide factory was exposed continuously to organochlorine pesticides (OCPs), such as chlorobenzene, dichlorobenzene, and nitrochlorobenzene, from 1975 to 2007. Based on the unweighted pair group method using average linkages (UPGMA) clustering result, the nine soils were classified into three distinct groups with no (C1 to C3, no pesticides were detected), light (S1 to S3, the total pesticide content various form 281.3 ± 21.4 to 509.8 ± 28.7 mg · kg −1 ), and heavy (S4 to S6, the total pesticide content various from 1,083.7 ± 40.4 to 4,595.8 ± 344.0 mg · kg −1 ) pesticide stress ( ). The surface soil (0 to 20 cm) was collected from each sampling area, and 2.0 kg of soil was selected randomly using the five-point sampling method. Soil samples were briefly stored at 4°C in 1-L sterile polypropylene Falcon tubes and then stored at −80°C until analysis. Soil physicochemical properties and pesticide content determination. Soil samples were ground through a 2.0-mm sieve and then used for the determination of the soil pH, soil organic matter (SOM), cation exchange capacity (CEC), total nitrogen (TN), total phosphorus (TP), and available sulfur ( ). The pesticide contents were determined by extracting the pesticides with dichloromethane using an accelerated solvent extraction system (ASE-200; Dionex, USA) followed by gas chromatography-mass spectrometry (GC-MS) analysis (GCMS 6890N-5973 N; Agilent, USA) as described in our previous publications ( , ). Bacterial and viral DNA extraction, sequencing, and analysis. Bacterial and viral DNA extraction, sequencing, and analysis were performed according to Zheng et al. ( ). Briefly, a FastDNA spin kit for soil (MP Bio) and TaKaRa MiniBEST viral RNA/DNA extraction kit 5.0 were used to extract bacterial and viral DNA, respectively. The viral DNA extraction methods are provided in Supplementary Methods in the Supplemental Material. After quality screening was conducted by Cutadapt (v1.2.1), a total of ~8.8 billion clean reads of bacterial metagenomes (~0.8 billion per clean soil samples and ~1.06 billion per pesticide-contaminated soil samples) and ~9.6 billion clean reads (~1.06 billion per sample) of viromes were obtained and used for de novo assembly (see Table S6 in the supplemental material) ( , ). All the raw reads and assembled metagenomics data are publicly accessible and can be download online from https://ngdc.cncb.ac.cn/bioproject/browse/PRJCA003886 . Viral protein clustering and distribution. Viral protein clustering and distribution were performed according to Zheng et al. ( ). Briefly, identified viral contigs were clustered into virus populations (vOTUs) using ClusterGenomes (v1.1.3; 95% identity and 80% coverage). vOTUs larger than 10 kb were subjected to protein clustering using vConTACT (v2.0; default parameters) based on the NCBI bacterial and archaeal viral RefSeq v85 database ( ). All protein sequence alignments were performed using DIAMOND 0.9.10 to group proteins ( ). Virus taxonomic annotation was performed using vConTACT (v2.0), and viral proteins were matched to the RefSeq virus database using BLASTp (bitcore, ≥50). Viral host prediction. The following three methods were used for virus host prediction: (i) the tRNA sequences were recovered from the viral genome with ARAGORN (v1.2.38) and then aligned with bacterial sequences by BLAST searches (100% coverage and 100% sequence identity) ( ); (ii) the sequence similarity of bacterial and viral CRISPR spacers was used to predict viral hosts. CRISPR spacers were recovered using CRASS from bacterial metagenomic paired-end (PE) reads ( ). BLASTn (E value, 10 −10 ; 100% nucleotide identity) was used to compare bacterial CRISPR spacer sequences and viral contigs; (iii) viral sequences were submitted to the JGI Virus Sequence Database to predict bacterial hosts (E value, 10 −5 ; sequence identity, ≥95%) ( ). The tRNA sequence alignment, CRISPR spacers comparison, and JGI database matching linked viral contigs with 35, 28, and 10,869 putative bacterial host sequences, respectively. Influence of abiotic factors on taxa and gene composition. Variance partitioning analysis (VPA) was used to determine the impact of abiotic factors on microbial communities ( ). Abiotic factors include soil physicochemical properties (e.g., pH, soil organic matter, and total nitrogen) and pesticide content (e.g., benzene, chlorobenzene, and meta-nitrochlorobenzene) (see Table S7 in the supplemental material). Prior to VPA, bioenv analysis was performed to sort out the effects of combinations of abiotic factors on taxa and genes composition. VPA and bioenv analysis were conducted using the relative abundance of taxa obtained from metagenomic sequencing and genes using the KEGG ontology (KO) group abundance annotated by the KEGG database. The contributions of physicochemical properties and pesticide content to community change were analyzed quantitatively by VPA and bioenv analysis with the “varpart” function and “bioenv” function in the “vegan” package, respectively. Community assembly analysis. The normalized stochasticity ratio (NST) reflects the relative importance of stochasticity in community assembly based on the magnitude of the difference between observed and null expectation as a quantitative measure of stochasticity. The NST value is between 0 and 1. When the NST value of the community is above 50%, stochastic processes are considered dominant in the community, while a value lower than 50% indicates deterministic processes that drive community assembly. The NST can be calculated as follows: NSS = ∑ ij ξ ( C ij , E ij ¯ ) − min k [ ∑ ij ξ ( E ij ( k ) , E ij ¯ ) ] ∑ ij ξ ( C D ij , E ij ¯ ) − min k [ ∑ ij ξ ( E ij ( k ) , E ij ¯ ) ] , C D ij = { 1 C ij ≥ E ij ¯ 0 C ij < E ij ¯ , ξ ( x , y ) = x − y x − δ δ = { 0 x ≥ y 1 x < y , NST = 1 − NSS , where D ij and C ij are the dissimilarity and actual similarity values between the i th community and the j th community, respectively; D C ij represents the similarity between community i and j under extremely deterministic assembly; and E ij (k) indicates one of the null expected values of similarity between community i and j under stochastic assembly. The “tNST” and “nst.boot” functions of the “NST” package were used to calculate NST values. Migration rate data were calculated by an analysis of observed operational taxonomic unit (OTU) mean relative abundances using the likelihood formula developed by Tetame with Etienne ( , ). With this method, the value of the migration rate is between 0 and 1. Higher migration rate values indicate that microbial communities are less limited by dispersal. In other words, when the value of migration approaches 1, the dispersal limitation decreases (i.e., all species migrate from the regional species pool) ( ). Niche breadth. We calculated niche breadth to represent the fitness and diversity of available resources of communities. The Levins niche breadth index was calculated using the following function: B j = 1 / ∑ i = 1 N P i j 2 , where B j is the niche breadth of taxon j in the metacommunity, N represents the count of communities in the metacommunity, and P ij refers to the proportion of taxon j in community i . The niche breadth was calculated using the “nst.boot” functions of the “SPAA” package in R ( ). Data statistical analysis. Data were analyzed using R 4.1.2 and visualized with GraphPad Prism 8.0.2. Alpha and beta diversity were calculated using the “vegan” package ( ). One-way ANOVA was used to identify significant differences between samples. An NMDS stress value of <0.05 based on the Bray-Curtis distance was considered to represent a good fit of NMDS analysis to the data. ADONIS was performed to determine whether the difference identified by NMDS was significant. Virus and host link networks were visualized with Cytoscape 3.9.1.
Soil samples were collected from a closed organochlorine pesticide factory in Jiangsu Province, China (120.228193′N, 31.758075′E), in 2019. The soil around the organochlorine pesticide factory was exposed continuously to organochlorine pesticides (OCPs), such as chlorobenzene, dichlorobenzene, and nitrochlorobenzene, from 1975 to 2007. Based on the unweighted pair group method using average linkages (UPGMA) clustering result, the nine soils were classified into three distinct groups with no (C1 to C3, no pesticides were detected), light (S1 to S3, the total pesticide content various form 281.3 ± 21.4 to 509.8 ± 28.7 mg · kg −1 ), and heavy (S4 to S6, the total pesticide content various from 1,083.7 ± 40.4 to 4,595.8 ± 344.0 mg · kg −1 ) pesticide stress ( ). The surface soil (0 to 20 cm) was collected from each sampling area, and 2.0 kg of soil was selected randomly using the five-point sampling method. Soil samples were briefly stored at 4°C in 1-L sterile polypropylene Falcon tubes and then stored at −80°C until analysis.
Soil samples were ground through a 2.0-mm sieve and then used for the determination of the soil pH, soil organic matter (SOM), cation exchange capacity (CEC), total nitrogen (TN), total phosphorus (TP), and available sulfur ( ). The pesticide contents were determined by extracting the pesticides with dichloromethane using an accelerated solvent extraction system (ASE-200; Dionex, USA) followed by gas chromatography-mass spectrometry (GC-MS) analysis (GCMS 6890N-5973 N; Agilent, USA) as described in our previous publications ( , ).
Bacterial and viral DNA extraction, sequencing, and analysis were performed according to Zheng et al. ( ). Briefly, a FastDNA spin kit for soil (MP Bio) and TaKaRa MiniBEST viral RNA/DNA extraction kit 5.0 were used to extract bacterial and viral DNA, respectively. The viral DNA extraction methods are provided in Supplementary Methods in the Supplemental Material. After quality screening was conducted by Cutadapt (v1.2.1), a total of ~8.8 billion clean reads of bacterial metagenomes (~0.8 billion per clean soil samples and ~1.06 billion per pesticide-contaminated soil samples) and ~9.6 billion clean reads (~1.06 billion per sample) of viromes were obtained and used for de novo assembly (see Table S6 in the supplemental material) ( , ). All the raw reads and assembled metagenomics data are publicly accessible and can be download online from https://ngdc.cncb.ac.cn/bioproject/browse/PRJCA003886 .
Viral protein clustering and distribution were performed according to Zheng et al. ( ). Briefly, identified viral contigs were clustered into virus populations (vOTUs) using ClusterGenomes (v1.1.3; 95% identity and 80% coverage). vOTUs larger than 10 kb were subjected to protein clustering using vConTACT (v2.0; default parameters) based on the NCBI bacterial and archaeal viral RefSeq v85 database ( ). All protein sequence alignments were performed using DIAMOND 0.9.10 to group proteins ( ). Virus taxonomic annotation was performed using vConTACT (v2.0), and viral proteins were matched to the RefSeq virus database using BLASTp (bitcore, ≥50).
The following three methods were used for virus host prediction: (i) the tRNA sequences were recovered from the viral genome with ARAGORN (v1.2.38) and then aligned with bacterial sequences by BLAST searches (100% coverage and 100% sequence identity) ( ); (ii) the sequence similarity of bacterial and viral CRISPR spacers was used to predict viral hosts. CRISPR spacers were recovered using CRASS from bacterial metagenomic paired-end (PE) reads ( ). BLASTn (E value, 10 −10 ; 100% nucleotide identity) was used to compare bacterial CRISPR spacer sequences and viral contigs; (iii) viral sequences were submitted to the JGI Virus Sequence Database to predict bacterial hosts (E value, 10 −5 ; sequence identity, ≥95%) ( ). The tRNA sequence alignment, CRISPR spacers comparison, and JGI database matching linked viral contigs with 35, 28, and 10,869 putative bacterial host sequences, respectively.
Variance partitioning analysis (VPA) was used to determine the impact of abiotic factors on microbial communities ( ). Abiotic factors include soil physicochemical properties (e.g., pH, soil organic matter, and total nitrogen) and pesticide content (e.g., benzene, chlorobenzene, and meta-nitrochlorobenzene) (see Table S7 in the supplemental material). Prior to VPA, bioenv analysis was performed to sort out the effects of combinations of abiotic factors on taxa and genes composition. VPA and bioenv analysis were conducted using the relative abundance of taxa obtained from metagenomic sequencing and genes using the KEGG ontology (KO) group abundance annotated by the KEGG database. The contributions of physicochemical properties and pesticide content to community change were analyzed quantitatively by VPA and bioenv analysis with the “varpart” function and “bioenv” function in the “vegan” package, respectively.
The normalized stochasticity ratio (NST) reflects the relative importance of stochasticity in community assembly based on the magnitude of the difference between observed and null expectation as a quantitative measure of stochasticity. The NST value is between 0 and 1. When the NST value of the community is above 50%, stochastic processes are considered dominant in the community, while a value lower than 50% indicates deterministic processes that drive community assembly. The NST can be calculated as follows: NSS = ∑ ij ξ ( C ij , E ij ¯ ) − min k [ ∑ ij ξ ( E ij ( k ) , E ij ¯ ) ] ∑ ij ξ ( C D ij , E ij ¯ ) − min k [ ∑ ij ξ ( E ij ( k ) , E ij ¯ ) ] , C D ij = { 1 C ij ≥ E ij ¯ 0 C ij < E ij ¯ , ξ ( x , y ) = x − y x − δ δ = { 0 x ≥ y 1 x < y , NST = 1 − NSS , where D ij and C ij are the dissimilarity and actual similarity values between the i th community and the j th community, respectively; D C ij represents the similarity between community i and j under extremely deterministic assembly; and E ij (k) indicates one of the null expected values of similarity between community i and j under stochastic assembly. The “tNST” and “nst.boot” functions of the “NST” package were used to calculate NST values. Migration rate data were calculated by an analysis of observed operational taxonomic unit (OTU) mean relative abundances using the likelihood formula developed by Tetame with Etienne ( , ). With this method, the value of the migration rate is between 0 and 1. Higher migration rate values indicate that microbial communities are less limited by dispersal. In other words, when the value of migration approaches 1, the dispersal limitation decreases (i.e., all species migrate from the regional species pool) ( ).
We calculated niche breadth to represent the fitness and diversity of available resources of communities. The Levins niche breadth index was calculated using the following function: B j = 1 / ∑ i = 1 N P i j 2 , where B j is the niche breadth of taxon j in the metacommunity, N represents the count of communities in the metacommunity, and P ij refers to the proportion of taxon j in community i . The niche breadth was calculated using the “nst.boot” functions of the “SPAA” package in R ( ).
Data were analyzed using R 4.1.2 and visualized with GraphPad Prism 8.0.2. Alpha and beta diversity were calculated using the “vegan” package ( ). One-way ANOVA was used to identify significant differences between samples. An NMDS stress value of <0.05 based on the Bray-Curtis distance was considered to represent a good fit of NMDS analysis to the data. ADONIS was performed to determine whether the difference identified by NMDS was significant. Virus and host link networks were visualized with Cytoscape 3.9.1.
|
COntrolling NUTritional Status (CONUT) as Predictive Score of Hospital Length of Stay (LOS) and Mortality: A Prospective Cohort Study in an Internal Medicine and Gastroenterology Unit in Italy | a670e669-cd23-4aee-8d31-a41322ca858f | 10057573 | Internal Medicine[mh] | Hospital malnutrition represents an acknowledged risk factor for many adverse clinical outcomes , and the clinical management of malnourished patients is affected by higher in-hospital morbidity, mortality, and healthcare costs. A recent study demonstrated additional costs for hospital malnutrition of over $58 billion dollars in Western countries . It is estimated that about 30% of hospitalized patients both in the United States and Europe present with malnutrition or risk of malnutrition at admission . In Italy, a recent hospital report found over half of the patients at risk of malnutrition and over a third already malnourished at hospital admission . Malnutrition is also an independent risk factor of poor postoperative outcomes in surgical patients and has been linked to an increased risk of infections , significantly higher mortality for sepsis , a higher risk of pressure ulcers, and a worse outcome of wound healing . In critically ill patients, major outcomes such as the duration of mechanical ventilation, the length of stay (LOS) in intensive care units (ICU), or infections are influenced by pre-existing malnutrition . Hospital malnutrition maybe more evident in a Gastroenterology Department due to the role of the gastrointestinal tract in nutrients absorption . However, despite these known associations, in daily clinical practice, hospital malnutrition remains often unrecognized, and the assessment of clinical nutrition of hospitalized patients is still underrated, probably due to a lack of awareness among clinicians, while focusing on diagnosis or treatment . Several tools have been released by international societies for the screening—Nutrition Risk Screening 2002 (NRS-2002), Malnutrition Universal Screening Tool (MUST)—for the diagnosis of malnutrition, the most recent being the Global Leadership Initiative on Malnutrition (GLIM) criteria . Despite a large diffusion among scientific sessions, the real application of such validated tools appears not sufficient in hospital settings, perhaps due to a lack of training, staff, and time . The COntrolling NUTritional status (CONUT) score, a simple index calculated using serum routine analysis (albumin, total lymphocyte count, and total cholesterol) is associated with short- and long-term prognosis in several diseases . The CONUT score has been proven not only to correlate with malnutrition grade but also to have a high predictive value concerning clinical outcomes and morbidity. For example, in patients with cancer, a higher CONUT score predicts a lower overall survival, a lower progress/recurrence-free survival, and a lower cancer-specific survival after surgery , and a similar predictive value has also been observed for non-solid tumors and other hematologic disorders . However, the CONUT score has also been investigated as a predictor of morbidity or mortality in various conditions other than malignancies, for example, in patients undergoing liver transplant or heart bypass surgery , in patients with acute heart failure , or in patients with pulmonary embolism . To date, fewer studies have been produced about in-hospital short-term outcomes such as the LOS or 30-day re-admission rates in medical units. A recent monocentric Chinese study performed by Hao in 2022 demonstrated that a higher CONUT score predicts a higher LOS and in-hospital mortality, specifically in patients with ischemic stroke ; another recent, large multicenter retrospective study performed in China in older adults, collecting data from more than eleven thousand patients, demonstrated that a higher CONUT score predicts a longer LOS and in-hospital mortality in elderly patients . However, similar studies concerning LOS or in-hospital mortality in more heterogeneous cohorts of patients or in Western countries are still lacking. Thus, we aimed to test CONUT at admission as a predictive score of hospital outcomes, such as LOS, in-hospital mortality, and 30-day re-admission rate in an Internal Medicine and Gastroenterology Department of an Italian Tertiary Care University hospital.
2.1. Study Design and Ethical Committee Approval We performed a single-center, observational, prospective, cohort study. The study conformed to the Declaration of Helsinki and the norms of Good Clinical Practice. The Ethical Committee of Fondazione Policlinico A. Gemelli IRCCS, Catholic University of the Sacred Heart approved the protocol (code 2638/22). The STROBE guidelines for cohort studies have been followed . 2.2. Patients Included patients were all adults (>18 years old) admitted to the Internal Medicine and Gastroenterology ward at the Fondazione Policlinico Agostino Gemelli IRCCS, Rome, Italy, from March 2021 to February 2022. All participants received information about the procedures to be performed in the study. Consent forms recording the agreement of patients to participate in the study were collected. Patients unable or refusing to give their consent to the study were excluded. 2.3. Protocol Description Patients were assessed by the hospital staff (B.E.A. and M.I.) upon admission and then referred to internal medicine residents (R.B., M.D., and T.G.). Residents explained the protocol to the patients, requested informed consent, and collected data. Then, they collected demographic characteristics, primary diagnoses, and comorbidities; the registered date of hospital admission and discharge (or death, if any); clinical data; laboratory values; anthropometric—weight, height, and body mass index (BMI) —and other nutritional variables (i.e., NRS-2002, MUST, and nutritional supplementation). Due to the simultaneous presence of more diseases in this category of patients, the Charlson comorbidity index (CCI) was calculated for each patient and preferred as a synthetic item instead of the single admission diagnoses. CONUT classes were defined based on serum albumin (g/dL), total lymphocyte count (count/mm 3 ), and total cholesterol (mg/dL) as reported in . The primary outcome measure for the present analysis was LOS and the secondary one was mortality during hospitalization. The re-admission rate within 30 days was also evaluated. 2.4. Data Collection and Statistical Analysis Data were collected using a specific Excel© spreadsheet and shown using descriptive statistical methods. The Kolmogorov–Smirnov test was used to assess the normality of variables. Categorical variables were expressed as numbers (percentage) and continuous variables as mean ± standard deviation or median (interquartile range). Patients were categorized according to total CONUT score into four classes (normal, mild, moderate, severe) and then grouped into two main classes (“normal-mild” and “moderate-severe”) for the inferential analyses. To estimate the risk of moderate-severe CONUT relative to normal-mild CONUT for the primary and secondary outcome measures, we used a multivariable logistic regression model. Kaplan–Meier curves were drawn, and the log-rank test was adopted to compare the obtained LOS intervals according to CONUT main classes. A receiver operating curve (ROC) was constructed to provide the sensibility and specificity of CONUT to predict mortality. The optimal cut-off value of CONUT was calculated by applying the Youden Method to the constructed ROC. A previous study reported an incidence of CONUT of more than 4 of 53.1% . With a margin of error of 7% and a confidence interval (CI) of 95%, we estimated 196 patients to be enrolled to intercept the above-mentioned incidences (percentages). We used the STATA ® Software (Version 14.0, Stata Corporation; College Station, TX, USA) to perform statistical analyses.
We performed a single-center, observational, prospective, cohort study. The study conformed to the Declaration of Helsinki and the norms of Good Clinical Practice. The Ethical Committee of Fondazione Policlinico A. Gemelli IRCCS, Catholic University of the Sacred Heart approved the protocol (code 2638/22). The STROBE guidelines for cohort studies have been followed .
Included patients were all adults (>18 years old) admitted to the Internal Medicine and Gastroenterology ward at the Fondazione Policlinico Agostino Gemelli IRCCS, Rome, Italy, from March 2021 to February 2022. All participants received information about the procedures to be performed in the study. Consent forms recording the agreement of patients to participate in the study were collected. Patients unable or refusing to give their consent to the study were excluded.
Patients were assessed by the hospital staff (B.E.A. and M.I.) upon admission and then referred to internal medicine residents (R.B., M.D., and T.G.). Residents explained the protocol to the patients, requested informed consent, and collected data. Then, they collected demographic characteristics, primary diagnoses, and comorbidities; the registered date of hospital admission and discharge (or death, if any); clinical data; laboratory values; anthropometric—weight, height, and body mass index (BMI) —and other nutritional variables (i.e., NRS-2002, MUST, and nutritional supplementation). Due to the simultaneous presence of more diseases in this category of patients, the Charlson comorbidity index (CCI) was calculated for each patient and preferred as a synthetic item instead of the single admission diagnoses. CONUT classes were defined based on serum albumin (g/dL), total lymphocyte count (count/mm 3 ), and total cholesterol (mg/dL) as reported in . The primary outcome measure for the present analysis was LOS and the secondary one was mortality during hospitalization. The re-admission rate within 30 days was also evaluated.
Data were collected using a specific Excel© spreadsheet and shown using descriptive statistical methods. The Kolmogorov–Smirnov test was used to assess the normality of variables. Categorical variables were expressed as numbers (percentage) and continuous variables as mean ± standard deviation or median (interquartile range). Patients were categorized according to total CONUT score into four classes (normal, mild, moderate, severe) and then grouped into two main classes (“normal-mild” and “moderate-severe”) for the inferential analyses. To estimate the risk of moderate-severe CONUT relative to normal-mild CONUT for the primary and secondary outcome measures, we used a multivariable logistic regression model. Kaplan–Meier curves were drawn, and the log-rank test was adopted to compare the obtained LOS intervals according to CONUT main classes. A receiver operating curve (ROC) was constructed to provide the sensibility and specificity of CONUT to predict mortality. The optimal cut-off value of CONUT was calculated by applying the Youden Method to the constructed ROC. A previous study reported an incidence of CONUT of more than 4 of 53.1% . With a margin of error of 7% and a confidence interval (CI) of 95%, we estimated 196 patients to be enrolled to intercept the above-mentioned incidences (percentages). We used the STATA ® Software (Version 14.0, Stata Corporation; College Station, TX, USA) to perform statistical analyses.
3.1. Baseline Characteristics of Patients Two hundred and three patients were evaluated, of which 127 (62.6%) were males and 76 (37.4%) females; the mean age was 66.05 ± 14.1 years. Most patients (68.5%) were admitted from the emergency department. The mean BMI (kg/m 2 ) was 25.02 (SD ± 4.88) and the mean CCI was 3.02 (SD ± 2.43). According to NRS-2002, 70 patients (34.5%) were at risk of malnutrition. Conversely, according to MUST, 31 patients (15.3%) were at medium risk whereas almost half of the entire sample (48.7%) were at high risk of malnutrition. According to CONUT, 44 (21.7%) patients had a normal nutritional status (CONUT 0–1), 66 (32.5%) had a mild (CONUT 2–4), 68 (33.5%) had a moderate (CONUT 5–8), and 25 (12.3%) had a severe impairment of nutritional status (CONUT 9–12). The mean LOS in days was 8.24 ± 5.75; 38 (18.7%) patients developed a refeeding syndrome (RS); 9 patients (4.4%) died during hospitalization. All baseline data are shown in . The CONUT classes (normal-mild vs. moderate-severe) correlated with age, admission type (elective or emergency), NRS-2002, MUST, the risk and occurrence of RS, the need for nutritional supplementation within 48 h from admission—either high-calorie and high-protein oral nutritional supplements (ONS) or artificial (enteral or parenteral) nutrition. As regards the main outcome measures, CONUT correlated with LOS and in-hospital mortality; re-admission within 30 days was not statistically different in the two groups . 3.2. Associations of Risk Factors with LOS Patients admitted with a CONUT score ≤ 4 had a lower mean LOS than those with a CONUT score ≥ 5 (6.5 ± 4.0 vs. 9.9 ± 6.4 days; p < 0.0001). At the univariate analysis, the ER admission, NRS-2002 > 3, MUST ≥ 2, a moderate/severe CONUT class, refeeding syndrome (RS) risk, and RS confirmed diagnosis were found to be risk factors for longer LOS. On the contrary, a normal-mild CONUT class was shown as a protective factor. In the multivariate analysis, ER admission, a moderate-severe CONUT score, and RS diagnosis were confirmed as independent risk factors of delayed LOS . The Kaplan–Meier method confirmed different LOS curves between normal-mild and moderate-severe CONUT classes ( p < 0.0001) as shown in . 3.3. Associations of Risk Factors with Hospital Mortality Nine patients (4.4%) died during hospitalization. Higher CONUT scores and RS diagnosis were shown as potential risk factors for mortality in the univariate analysis. On the other side, a higher BMI was associated with lower mortality risk, as well as nutritional supplementation received within 48 h from admission . Due to the limited number of death events in our study population (9), a multivariate analysis was not feasible. However, as reported at the ROC curve, the CONUT score was a reliable predictor of mortality, with an area under the ROC curve (AUC) of 0.831 (95% CI 0.680–0.982); the optimal cut-off obtained was 8.5 .
Two hundred and three patients were evaluated, of which 127 (62.6%) were males and 76 (37.4%) females; the mean age was 66.05 ± 14.1 years. Most patients (68.5%) were admitted from the emergency department. The mean BMI (kg/m 2 ) was 25.02 (SD ± 4.88) and the mean CCI was 3.02 (SD ± 2.43). According to NRS-2002, 70 patients (34.5%) were at risk of malnutrition. Conversely, according to MUST, 31 patients (15.3%) were at medium risk whereas almost half of the entire sample (48.7%) were at high risk of malnutrition. According to CONUT, 44 (21.7%) patients had a normal nutritional status (CONUT 0–1), 66 (32.5%) had a mild (CONUT 2–4), 68 (33.5%) had a moderate (CONUT 5–8), and 25 (12.3%) had a severe impairment of nutritional status (CONUT 9–12). The mean LOS in days was 8.24 ± 5.75; 38 (18.7%) patients developed a refeeding syndrome (RS); 9 patients (4.4%) died during hospitalization. All baseline data are shown in . The CONUT classes (normal-mild vs. moderate-severe) correlated with age, admission type (elective or emergency), NRS-2002, MUST, the risk and occurrence of RS, the need for nutritional supplementation within 48 h from admission—either high-calorie and high-protein oral nutritional supplements (ONS) or artificial (enteral or parenteral) nutrition. As regards the main outcome measures, CONUT correlated with LOS and in-hospital mortality; re-admission within 30 days was not statistically different in the two groups .
Patients admitted with a CONUT score ≤ 4 had a lower mean LOS than those with a CONUT score ≥ 5 (6.5 ± 4.0 vs. 9.9 ± 6.4 days; p < 0.0001). At the univariate analysis, the ER admission, NRS-2002 > 3, MUST ≥ 2, a moderate/severe CONUT class, refeeding syndrome (RS) risk, and RS confirmed diagnosis were found to be risk factors for longer LOS. On the contrary, a normal-mild CONUT class was shown as a protective factor. In the multivariate analysis, ER admission, a moderate-severe CONUT score, and RS diagnosis were confirmed as independent risk factors of delayed LOS . The Kaplan–Meier method confirmed different LOS curves between normal-mild and moderate-severe CONUT classes ( p < 0.0001) as shown in .
Nine patients (4.4%) died during hospitalization. Higher CONUT scores and RS diagnosis were shown as potential risk factors for mortality in the univariate analysis. On the other side, a higher BMI was associated with lower mortality risk, as well as nutritional supplementation received within 48 h from admission . Due to the limited number of death events in our study population (9), a multivariate analysis was not feasible. However, as reported at the ROC curve, the CONUT score was a reliable predictor of mortality, with an area under the ROC curve (AUC) of 0.831 (95% CI 0.680–0.982); the optimal cut-off obtained was 8.5 .
After evaluating 203 patients admitted to an Internal Medicine and Gastroenterology Department, we demonstrated that the CONUT score can be a reliable predictor of higher LOS and in-hospital mortality. Indeed, at admission, patients reporting a CONUT score ≥ 5 points have nearly 90% probability of a longer LOS than those with a lower score. The predictive value of the CONUT score in assessing LOS was confirmed in the multivariate analysis. Interestingly, an NRS-2002 score > 3 (risk of malnutrition) and MUST ≥ 2 (high risk of malnutrition) showed an association with a higher LOS only in the univariate analysis. This is of interest, due to the objective nature of the CONUT score, based only on simple laboratory tests easily obtained in almost all clinical settings. Even if mortality events were only nine during hospitalization, univariate analysis confirmed a high CONUT score as a predictive risk factor of mortality, as also shown in the ROC curve. Thus, we can argue that a baseline CONUT value of 9 (or more) at admission predicts mortality during the hospital stay. These results align with those of other reports investigating the role of CONUT in predicting LOS and mortality in several hospital settings, especially in elderly patients and in Eastern countries . In details, Nishi et al. performed a retrospective analysis of a multicenter Japanese registry involving 838 patients (mean age 72 years) admitted for heart failure (HF): high CONUT scores were correlated with increased risk of in-hospital death in unadjusted and adjusted models and LOS . Kato et al., analyzing data from a similar registry of patients admitted for acute decompensated heart failure (ADHF) (2466 patients, mean age 80 years), concluded that high CONUT scores were associated with higher in-hospital mortality and infection even when adjusting for other clinical covariates . More recently, a Chinese study including patients admitted for acute ischemic stroke (AIS) (1079 patients, mean age 81 years) found a linear association between CONUT scores and LOS, and a significant association with hospital mortality . Another retrospective study, analyzing data from 11,795 older adult Chinese patients found a higher LOS in higher CONUT classes and recognized CONUT (at the score ≥ 6) as the best predictor of in-hospital mortality among other five nutrition-related tools (including NRS-2002) . Despite the lesser study population, we confirmed such evidence in a prospective cohort study, in Italy, in a different clinical setting (Internal Medicine and Gastroenterology department) and enrolling patients with a younger mean age (66 years). This confirms the reliability of the CONUT score as a predictive marker of short-term clinical outcomes irrespective of the geographical area and the population’s age. Indeed, the clinical value of CONUT resides in its simple laboratory data (albumin, cholesterol, lymphocytes count), reflecting the patients’ immunonutritional status. As regards albumin, it has been questioned as a proxy measure of nutritional status or total muscle mass, and rather indicated as a negative acute phase protein . However, low albumin serum concentrations still have a predictive role in adverse outcomes in different clinical contexts of disease-related malnutrition, as demonstrated in recent studies . Moreover, low serum albumin levels are associated with increased short- and long-term mortality in hospitalized patients, and serum albumin levels are an important predictor of in-hospital mortality or hospital complications in elderly patients . On the other hand, the total lymphocyte count (≤1500 cells/mm 3 ) may have a few limits due to other possible biasing conditions (i.e., hematological or infective diseases); however, recent studies on COVID have associated the total lymphocyte count with worse hospital outcomes and mortality in a context of severe inflammation . Regarding total cholesterol, previous studies have associated low plasmatic levels with poor nutritional intake, systemic inflammation, and a worse prognosis in hospitalized patients, thus demonstrating a potential predictive value . We did not find any difference between “normal-mild” and “moderate-severe” CONUT classes in terms of hospital re-admission within 30 days. This could be explained by the small number of re-admission events (8 vs. 5, respectively). Moreover, we do not register data about the re-admission type (if elective or by the emergency department), so we cannot make an inference about whether the re-admission could be related to malnutrition itself or other causes. Our results highlighted the role of nutritional supplementation (received within 48 h from admission) in reducing mortality risk by nearly 90%. The nutritional supplementation included both high-calorie and high-protein ONS and artificial enteral or parenteral nutrition, according to the prescriptions of clinical nutrition team. This confirms the results of the EFFORT study, a multicentric randomized controlled trial, which demonstrated, in a large number of patients at nutritional risk, that an individualized nutritional support in medical inpatients could reduce adverse events and in-hospital mortality . Moreover, in this study, we collected data about the occurrence of RS, since this work shared the same registry used for another of our study focusing on this topic, to which we remand for further details . RS may occur when malnourished patients receive a prompt normocaloric artificial (enteral or parenteral) refeeding; it consists in a rapid shift in fluids and electrolytes in the intracellular space resulting in electrolytes abnormalities and cellular edema. It may have a dramatic impact in terms of morbidity and mortality, even if it is still underestimated and, as regards this study, it is significantly higher in the moderate-severe CONUT class. This confirms the efficacy of CONUT as a nutritional predictive score. The strengths of this study are homogeneous data collection and the prospective nature of the study. Moreover, to the best of our knowledge, this is the first Italian study on this topic. The main limitations are the monocentric design and the small number of deaths which does not allow us to perform a multivariable analysis, even if this demonstrated the efficiency of the department care. Thus, we think that CONUT value in predicting in-hospital mortality should be further confirmed in other similar prospective studies. Moreover, we did not perform a complete nutritional assessment since this study lacks data about body composition. Further studies are warranted to correlate the CONUT score with body composition parameters such as body cell mass or muscle mass. Finally, the impact of statin therapy (as regards total cholesterol) and the presence of hematological or infective diseases (as regards lymphocyte count) have not been investigated. Notwithstanding the above-mentioned limits, the study reflects the importance of using appropriate tools to stratify the nutritional risk at admission to the hospital, in order to prompt necessary nutritional interventions that could be effective in reducing mortality. Current guidelines propose other nutritional tools such as NRS-2002, MUST, and GLIM Criteria, which are more standardized and focused on nutritional status. These tools investigate the amount and the speed of weight loss, the BMI, the reduced dietary intake, the severity of disease and, in the case of the GLIM criteria, also the loss of muscle mass. We also recognized the value of such an approach in clinical practice . However, such a nutritional approach is still not widely spread until now in medical departments . We thus decided to test another simple score as an objective and rapid method to predict prognosis. The CONUT score was demonstrated to be a simple, objective, and predictive method for this purpose, at least for hospital LOS and probably also for hospital mortality.
The CONUT score is a simple and reliable nutrition-related tool for stratifying the risk of higher LOS and predict mortality at admission. Given the relevance and ease of performing, health professionals should be incentivized to use the CONUT score in clinical practice to prompt personalized nutritional support. Indeed, we observed that early nutritional intervention (within 48 h of admission) could reduce in-hospital mortality. The predictive role of different CONUT score cut-off values needs to be validated in populations with different diseases. Further studies are needed to confirm our preliminary results in large and multicentric medical cohorts.
|
Molecular Pathology, Oxidative Stress, and Biomarkers in Obstructive Sleep Apnea | 4cf34854-d682-441c-b0df-a87fbd84899f | 10058074 | Pathology[mh] | Obstructive sleep apnea syndrome (OSAS) affects from 9% to 39% of the adult population, with a higher incidence in males and the elderly, and is the most common form of respiratory sleep disorder . It is characterized by recurrent, complete, or partial upper airway obstruction due to their collapse, with consequent hypopnea or apnea, leading to hypoventilation and chronic intermittent hypoxemia (IH) and increasing blood carbon dioxide partial pressure . The OSAS consequences do not only concern excessive daytime sleepiness; they also independently favor the development of cardiovascular pathologies as an independent risk factor for hypercholesterolemia and hypertension, obesity, diabetes, and neuropsychological diseases such as depression . Therefore, OSAS patients have higher cardiovascular-related morbidity compared with non-OSAS ones . The diagnosis of OSAS follows the diagnostic criteria codified by the American Academy of Sleep Medicine (AASM) after the exclusion of other pathologies that may be the cause of the apnea/hypopnea events . Furthermore, it is possible to make a diagnosis of OSAS when one is faced with anamnestic data, collected by interviewing the patient or those who sleep with him, with reported episodes of falling asleep when awake, excessive daytime sleepiness, non-refreshing sleep, tiredness, or insomnia, breathing, snoring, wheezing, choking, loud snoring, and at least 5 episodes of apnea, hypopnea, or breathing-related awakenings per hour in polysomnography. Alternatively, even in the absence of anamnestic data, it is possible to diagnose OSAS when polysomnography shows 15 or more apneas, hypopneas, or awakenings related to respiratory events per hour with evidence of respiratory effort in all or part of them in polysomnography . To measure the degree of OSAS, the polysomnographic apnea-hypopnea index (AHI) is used . Depending on the severity of the disease, treatment options include surgical interventions, lifestyle modifications, continuous positive airway pressure (CPAP), oral appliances such as mandibular advancement, and hypoglossal nerve stimulation . However, the OSAS diagnosis and treatment currently do not take into consideration what happens at the molecular and cellular level, which instead causes systemic complications related to OSAS. This review aims to recognize molecular mechanisms, markers, and potential treatments useful for counteracting the cellular damage resulting from OSAS. This could help to better understand which patients are candidates, not only for the use of CPAP or MAD (mandibular advancement device), but also for medical therapy to counteract or reverse cellular damage. 2.1. Oxidative Stress (OS) OS is an imbalance between reactive oxygen species (ROS) production and antioxidant capacity. The hypoxia and reoxygenation cycles cause a change in the oxidative balance, leading to the formation of ROS and a decrease in endogenous antioxidant molecules . ROS react with organic molecules, impairing their functions, altering cellular metabolism, and causing cell damage. They are considered one of the main mechanisms responsible for cardiovascular complications in patients with OSAS, and their production correlates with the severity of the disease . The presence of oxidative imbalance has also been observed in uvular tissues removed after uvulopalatoplasty in patients with OSAS and correlates with the severity of the disease . The human organism tries to adapt to the condition of lack of oxygen with the production of specific molecules useful for cell survival in hypoxic conditions such as Hypoxia Induced Factor-1α (HIF-1α) and Vascular endothelial growth factor (VEGF) . HIF-1 α, instead, is one of the main actors in oxygen homeostasis. Its levels, together with those of NF-kB, correlate with the severity of the disease measured by the AHI and desaturation number, and the levels of surfactant protein D (SPD) are reduced in an inverse manner . There is no significant variation in HIF-1α levels during the day . The alteration of circulating HIF-1α levels is chronic; it has been observed that a single night with CPAP in severe OSAS patients does not significantly modify its expression . The expression of HIF-1α and NF-kB is reduced after two months of continuous nasal CPAP use, and the SPD levels increase . VEGF is increased as the nocturnal oxygen saturation decreases . OS also underlies the increase in myeloperoxidase (MPO), intracellular adhesion molecule 1 (ICAM-1), vascular cell adhesion protein (VCAM-1), L-selectin, and E-selectin . The differences in molecular expression in OSAS patients also concern the reduction of the morning levels of the Rho-associated protein kinase (ROCK) 1 and 2 molecules and the circulating nitric oxide, as well as the expression of endothelial eNOS . These observations were confirmed in vivo, with an observed down-regulation of eNOS and an increase in nitrotyrosine . VE-cadherin cleavage is one of the mechanisms of endothelial dysfunction and in OSAS patients, the circulating plasma values of the soluble form of VE-cadherin (sVE) were increased, suggesting an augmented endothelial permeability. This mechanism appears to be associated with ROS production, and activation of HIF-1, VEGF, and tyrosine kinase pathways . Another mechanism that contributes to the endothelial dysfunction in IH is the production of extracellular vesicles by red blood cells, and their pathogenetic mechanism involves decreased eNOS, increased Endothelin-1 (ET-1) production via the Erk1/2 pathway, and phosphorylation via the PI3K/AKT pathway ( . OSAS-induced endothelial dysfunction has also been related to some microRNAs, such as miR-630 in infantile OSAS and miR-30a, miR-34a-5p, and miR-193 in mouse models . CPAP is widely used in the treatment of OSAS, and its systemic benefits have also been observed at the molecular level regarding endothelial function. Myocardial damage is also linked to microRNA expression alterations such as miR-146a-5p . 2.1.1. Mitochondrial Involvement ROS are mainly produced at the mitochondrial level during the reoxygenation phase . Mitochondrial damage is confirmed by the reduction of circulating mitochondrial DNA in conjunction with the worsening of OSAS . Studies from mouse models showed that mitochondrial OS is also the basis of the damage to auditory hair cells, with aberrant mitochondrial morphology and overexpression of PGC-1α and Tfam mRNA . The role of mitochondria in recurrent hypoxia damage is not limited to neuronal cells, and studies on genioglossus and palatine muscles have also shown damage at this level . 2.1.2. Inflammatory Signaling IH leads to the formation of proinflammatory factors such as tumor necrosis factor (TNF), C-reactive protein (CRP), and interleukins (IL)-6 and -8 , and elevated values of NF-kB and TNF-α are linked to OSAS and daytime sleepiness. Cytokines related to NF-kB are also consequently increased, such as IL-8 , whereas IL-17 levels are also correlated with OSAS and its severity, as well as an inverse correlation with Vitamin D levels in enrolled patients . CRP and TNF-α levels decrease after OSAS surgery, but they still remain higher than those of healthy control groups . Further molecules whose expression is altered are Osteoprotegerine, Chitinase 3-like protein 1 (YKL-40), and Cardiotrophin-1 (CT-1), and their value correlates with AHI . 2.1.3. Peripheral Blood Cells Various changes in peripheral blood cells have been observed. Overexpression of Toll-Like Receptor (TLR) was observed in circulating monocytes of OSAS patients , and the TLR 6 gene is upregulated in peripheral blood cells through DNA methylation. In particular, the cytosine-phosphate-guanine (CPG) site number 1 is hypermethylated in patients with severe OSAS, and its methylation is reduced after at least 6 months of CPAP therapy . Epigenetic studies also identified hypermethylation of interleukin 1 receptor 2 (IL1R2) and androgen receptors with increased expression of both . 2.1.4. Cardiovascular Implication of OSAS Inflammation The role of inflammation has also been observed in the aortas of OSAS mouse models, evidencing an accumulation and proliferation of pro-inflammatory metabolic M1-like macrophages highly expressing CD36 and an increase in the transcription of atherogenic pathways, inflammation, and OS . Increased circulating fibrinogen values are associated with an elevated risk of cardiovascular events as well as OSAS, correlating high serum levels with the severity of the disease . The inflammatory molecules increased in OSAS patients are numerous and also include heat shock protein 70, tissue fat, monocyte chemotactic protein-1, and highly sensitive C-reactive protein . The prothrombotic state due to OSAS is also documented by the increase in tissue plasminogen activators 1 and 2 and the decrease in TGF-β and urokinase-type plasminogen activator . Epigenetic dysregulation of DNA in OSAS was shown by aberrant methylation of the formyl peptide receptor (FPR) 1, 2, 3 genes, and FPR1 overexpression and the deficiency of FPR2 and FPR3 were associated with OSAS and its severity as well as with the development of diabetes mellitus and cardiovascular diseases. Literature evidence did not show a correlation between OS and cardiovascular risk in OSAS patients . Neurotrophins are proteins that regulate the nervous system; alterations of some of them, such as brain-derived neurotrophic factor (BDNF) and nerve growth factor (NGF), can also lead to cardiovascular complications. BDNF is related to cardiomyocyte contractility and its alterations with atherosclerosis and hypertension, and NGF plays a role in atrial autonomic OSAS-induced atrial fibrillation . 2.2. Circulating Metabolites Previous research has shown that OSAS induces an increase in the blood levels of endocannabinoids such as anandamide and ethanolamide, which are associated with an increase in blood pressure and cardiovascular risk . Likewise, an increase in the blood levels of adenosine, epinephrine, norepinephrine, and aldosterone was also observed , as was an increase in retinoids, carotenoids, and tocopherol, which increase the susceptibility to vascular pathologies . Levels of saturated fatty acids and n-3 fatty acids also correlate with sleep quality, duration, and rapid eye movements. An increase in lactic acid and some fatty acids such as arabinose, arabitol, cellulose, glyceraldehyde, and threitol has been observed in these patients , and metabolomic studies have also shown an increase in glutamic acid, deoxy sugar, arachidonic acid, phosphatidylethanolamine, sphingomyelin, and lysophosphatidylcholine. The metabolic alterations mentioned above seem to be attributable to the hypoxia induced by OSAS . In addition to the metabolites, the regulatory metabolism molecules are also impaired in OSAS patients. OSAS is independently correlated with insulin resistance and fatty liver disease, and several genes involved in cholesterol metabolism were impaired, such as malic enzyme and acetyl coenzyme A (CoA) synthetase, or acetyl-CoA carboxylase, stearoyl-CoA desaturases 1 and mitochondrial glycerol- 3-phosphate acyltransferase . OSAS also induces adipose tissue inflammation and dysfunction . The circulating omentin expression increased in OSAS . Moreover, OSAS patients have significantly increased levels of leptin, affecting sleep, ventilation, and upper airway defenses . 2.3. Urine Molecules Urine analysis from OSAS patients revealed high concentrations of adrenaline, noradrenaline, and homocysteine, which are associated with increased cardiovascular risk, as well as leukotriene E4. Increased levels of homovanillic acid, a metabolite of dopamine, and 3-4-dihydroxyphenylacetic acid were also measured . Urinary isoprostane 8, which is associated with daytime sleepiness and OSAS, as well as being an element of endothelial damage , and an increase in urinary leukotriene-4 has also been observed, which is related to atherosclerosis . The significance of these altered urine metabolites in OSAS is unknown. 2.4. Neurotransmitters Several neurotransmitters could play a role in the pathogenesis of the disease. Histamine is responsible for the state of arousal in the central nervous system. In studies on mouse models, it also seems to have a role in the neuromuscular transmission that occurs from the hypoglossal nerve to the genioglossus, both of which are altered in OSAS patients . Starting from the observation that apneas increase during REM sleep, due to motor inhibition of the cervical-cephalic muscles, several authors investigated the potential mechanism of muscular inactivation. REM sleep-related upper airway collapse is due to a change in norepinephrine and serotonin secretion by cranial nerve XII. The secretory patterns and increased distribution of α1-adrenoceptors are impaired in mouse models of IH during sleep . Loss of muscle tone leading to recurrent airway collapse appears to be related to reduced activity of pharyngeal motoneurons, also due to decreased stimulation of cholinergic acetylcholine receptors. This molecule, therefore, not only acts at the level of the central nervous system with different concentrations depending on whether we are in a state of sleep or wakefulness, but its action also varies at a peripheral level and could be one of the mechanisms of muscular hypotonia pharyngeal in OSAS patients . Grace et al., observed that the G-protein-coupled inwardly rectifying potassium (GIRK) channels are also involved in muscular inactivation during REM sleep. Therefore, the authors hypothesized that the potential target to counteract the loss of muscle tone is a potassium channel, and to avoid having an undesirable systemic effect, they suggested the inwardly rectifying potassium channel Kir2.4, which is expressed almost exclusively in the motor nuclei of the cranial nerves. To date, there are no clinical studies with this hypothesis. The brain neurotrophic factor (BDNF) is a neurotrophin responsible for neuron growth, development, and plasticity. Therefore, it has a role in memory and learning mechanisms. It can cross the brain-blood barrier, and its blood level dysregulation is related to depression and cognitive decline. Despite there being no significant difference in BDNF levels between OSAS and control patients, the morning molecule expression is related to age and oxygen saturation during sleep. The proBDNF, a derived product of BDNF, relates to age and HIF-1α morning quantity . BDNF and proBDNF evening concentrations are higher in patients with alteration measured with the Athens Insomnia Scale (AIS), Pittsburg Sleep Quality Index (PSQUI) positive, and lower in those with Beck Depression Inventory (BDI) positive . There is also a correlation between plasmatic BDNF levels and the oxygen desaturation index, and a negative one with the AHI. It has been hypothesized that BDNF could be involved in apoptosis-related neural injury, contributing to OSAS-induced cognitive degeneration and psychiatric pathologies . OSAS patients with depression have lower levels of BDNF and pro-BDNF compared with OSAS patients with no mood disturbances . IH could also be responsible for peripheric neural damage. In mice models, IH-induced OS decreases BDNF and pro-BDNF expression, which increases retinal cell apoptosis . BDNF expression is altered in cognitive diseases (e.g., Alzheimer’s disease, Parkinson’s related dementia, etc.) BDNF is also involved in nociception, and some authors have hypothesized his involvement in the greater pain susceptibility of OSAS patients compared with the general population . Further neurotrophins are known and are related to OSAS, such as the nerve growth factor (NGF), neurotrophins 3 and 4, and one neurotrophic factor, the glial cell-line-derived neurotrophic factor (GDNF) . NGF is responsible for sympathetic neurons’ maturation, differentiation, and survival. His precursor protein, the proNGF, has the opposite action . Its neurological involvement in OSAS has not been observed, but it has been related to cardiac autonomic nervous system disturbances and pediatric neurogenic tonsillar inflammation and hypertrophy . GDNF is essential for dopaminergic system development and respiratory pattern generation. It has been observed that its levels are lower in patients with OSAS . OSAS exacerbates Parkinson’s disease; sleep disturbance severity and motor dysfunction have been related to IL-6 levels. In the same study, Kaminska et al. observed a correlation between BDNF and increased sleepiness . OSAS, Neurocognition and Neurofilament Oxidative stress has repercussions in many parts of the body, including the central nervous system. Therefore, OS in OSAS is also responsible for neurocognitive dysfunction. The imbalance between oxidization and antioxidants, as previously cited, could cause neuron injury in brain regions most susceptible to hypoxia and oxidative stress, such as the hippocampus and cerebral cortex regions . Different domains could be altered in OSAS patients, such as attention/vigilance, memory, global cognitive, and executive function. However, CPAP treatment seems to improve some but not all executive functions in different degrees of cognitive dysfunction. Compromised cognition could be partially reversed after CPAP. As mentioned above, several studies highlighted the association between OS and nervous system diseases such as Parkinson’s disease, Alzheimer’s disease, and epilepsy . In this direction, identification and quantification of neuronal biomarkers of axonal damage could improve the diagnostic accuracy and the prognostic assessment in the management of neurological disease. Neurofilament light chain (NFL) is a neuronal protein, exclusively located in the neuronal cytoplasm, whose levels increase in serum and cerebrospinal fluid (CSF) proportionally to the degree of neuronal axonal damage, and it is a valuable biomarker in several neurological disorders. In OSAS patients, it could be interesting to monitor NFL variations after CPAP treatment aimed to evaluate neuronal recovery. The neuronal damage has also been confirmed by the first studies, which show an increase in the expression of neurofilament (NFL) and its correlation with the severity of the disease . Further studies are needed to measure the impact of CPAP on this parameter. 2.5. Potential Therapies 2.5.1. Antioxidants There are many potentially beneficial molecules for patients with OSAS, but most of them have not been tested on humans. Manganese superoxide dismutase is protective against cortical neuron oxidative damage by IH in mouse models . Adiponectin also proved to be useful in counteracting mitochondrial damage in the genioglossus muscle of OSAS mice . In addition, ROS scavenger administration with Endavarone in mouse models of IH has been tested, showing a significant reduction in cognitive impairment associated with increased brain expression of phosphorylated-cAMP response element-binding (p-CREB) . The molecules tested in humans have not been studied in courts large enough to give indications for their use. Vitamin C and N-acetylcysteine (NAC) have shown interesting results in the reduction of OS in OSAS , and NAC reduces OS in OSAS through the reduction of peroxidized lipids and the increase in glutathione. Surprisingly, patients who received it continuously also had improvements in sleep parameters . Vitamin C, on the other hand, proved to be effective in improving the endothelial function of OSAS patients in a study that took as its reference the diameter of the brachial artery, an indirect indicator of endothelial function . Lastly, Leptin is both a drug capable of reducing free radicals, OS, and atherosclerosis in patients with OSAS . To date, the only confirmed antioxidant therapy is CPAP itself, which has shown to be able to reverse many of the molecular alterations observed in vivo, such as eNOS, nitro-tyrosine, and NF-kB in the endothelium and circulating TNF-α . 2.5.2. Non-Antioxidant-Based Therapy Estrogen-related receptor-α (ERR-α) is downregulated in OSAS patients and its ligand-binding induces the expression of fast-type muscle fibers in palatopharyngeal muscles. The interaction between estrogens and ERR-α could be a therapeutic target to reverse the muscle remodeling typical of these patients . They inhibit the overexpression of HIF-1α induced by chronic IH and improve the endurance and regeneration of the genioglossus muscle in OSAS animal models . Estrogens, in particular 17β-estradiol (E2) and a resveratrol dimer (RD), have a protective action against OSAS by limitation of HIF-1α action. A pilot study in OSAS patients evidenced that Desipramine reduced airway collapse. At the same time, its anti-inflammatory properties could be beneficial in counteracting the systemic effects of OSAS, but further clinical studies are needed on a larger scale to evaluate its application . The use of sedatives in the treatment of OSAS appears to be counterintuitive. However, it has been hypothesized that trazodone may reduce the respiratory arousal threshold and upper airway obstruction. The first phenomenon occurred in an experimental group, while the second was not significant, and the magnitude of the threshold change was not sufficient to counteract the changes due to mechanical obstruction . The Phase II Pharmacotherapy of Apnea by Cannabimimetic Enhancement (PACE) has shown encouraging preliminary results for Dronabinol. A reduction in the AHI, a reduction in the feeling of sleepiness, and good satisfaction in the treated patients have been observed . Sildenafil involves the inhibition of cyclin guanosine monophosphate phosphodiesterase 5, resulting in an increase in cyclic guanosine monophosphate and NO. Its experimentation in a randomized controlled trial in which it was compared with a placebo, however, showed a worsening of the disease . In a randomized trial, the combination of Atomoxetine, a norepinephrine reuptake inhibitor, and antimuscarinic Oxybutynin, taken before going to sleep, was shown to be able to reduce the severity of OSAS, and further studies on larger sample sizes are needed . It is important to observe that many molecules have only been tested in mouse models, such as Astragaloside IV, which showed an improvement of hypoxia-induced endothelial function ; Tauroursodeoxycholic acid, against hepatic damage induced by HI ; Pitavastatin, showing a reversal of IH-induced myocardial hypertrophy, cardiac function, perivascular fibrosis and inflammatory indices ; Allopurinol also showed beneficial effects in mouse models of OSAS with a reduction of lipid peroxidation and an improvement in cardiac function ; and for all molecules, clinical trials are needed. 2.6. Diagnostic Biomarkers Some authors have attempted to identify OSAS biomarkers. From the studies of Fleming et al. it has emerged that altered values of glycated hemoglobin, c-reactive protein, and erythropoietin may be useful in OSAS diagnosis . Variants of the 5-hydroxytryptamine receptor 2A have also been studied as markers of risk and severity of OSAS. It has been observed that some of them can be protective, while others predispose to greater severity of the disease . The study of the overexpression of the genes disintegrin and metalloproteinase domain 29 (ADAM29), solute carrier family 18 (vesicular acetylcholine) member 3 (SLC18A3), and fibronectin-like domain-containing leucine-rich transmembrane protein 2 (FLRT2) showed how these latter are overexpressed in Asian subjects with OSAS and could be used to screen patients, at risk for the severe form of the disease . Numerous genetic polymorphisms have been associated with the development of OSAS, but none of them have been validated in clinical trials as useful for screening or diagnosis . MicroRNAs could be helpful for diagnosis, but we still don’t have clinical validations in OSAS , such as the downregulation of miR-664a-3p , and dysregulation of miR-126-3p, miR-26a-5p, and miR-107, which associate with arterial hypertension in OSAS patients . The most accurate OS markers in OSAS are thioredoxin, malondialdehyde, superoxide dismutase, and iron reduction . Thioredoxin (TRX) concentration is a marker of disease severity as it is proportional to that of AHI and is inversely related to oxygen saturation . Although several molecules related to OS were not significantly increased in OSAS, glutathione, 8-isoprostane, substances reactive to barbituric acid (TBARS), catalase activity, copper-zinc superoxide dismutase (SOD), and products of lipid peroxidation . However, there is no agreement between SOD and malondialdehyde (a TBARS), because, in some studies, their levels seem to be correlated with the severity of OSAS . The markers of the loss of antioxidant capacity in OSAS patients have been observed in several studies as the lowering of the antioxidant power of reduced iron (FRAP), the concentration of reduced iron, and the total serum antioxidant status (TAS) . OS is an imbalance between reactive oxygen species (ROS) production and antioxidant capacity. The hypoxia and reoxygenation cycles cause a change in the oxidative balance, leading to the formation of ROS and a decrease in endogenous antioxidant molecules . ROS react with organic molecules, impairing their functions, altering cellular metabolism, and causing cell damage. They are considered one of the main mechanisms responsible for cardiovascular complications in patients with OSAS, and their production correlates with the severity of the disease . The presence of oxidative imbalance has also been observed in uvular tissues removed after uvulopalatoplasty in patients with OSAS and correlates with the severity of the disease . The human organism tries to adapt to the condition of lack of oxygen with the production of specific molecules useful for cell survival in hypoxic conditions such as Hypoxia Induced Factor-1α (HIF-1α) and Vascular endothelial growth factor (VEGF) . HIF-1 α, instead, is one of the main actors in oxygen homeostasis. Its levels, together with those of NF-kB, correlate with the severity of the disease measured by the AHI and desaturation number, and the levels of surfactant protein D (SPD) are reduced in an inverse manner . There is no significant variation in HIF-1α levels during the day . The alteration of circulating HIF-1α levels is chronic; it has been observed that a single night with CPAP in severe OSAS patients does not significantly modify its expression . The expression of HIF-1α and NF-kB is reduced after two months of continuous nasal CPAP use, and the SPD levels increase . VEGF is increased as the nocturnal oxygen saturation decreases . OS also underlies the increase in myeloperoxidase (MPO), intracellular adhesion molecule 1 (ICAM-1), vascular cell adhesion protein (VCAM-1), L-selectin, and E-selectin . The differences in molecular expression in OSAS patients also concern the reduction of the morning levels of the Rho-associated protein kinase (ROCK) 1 and 2 molecules and the circulating nitric oxide, as well as the expression of endothelial eNOS . These observations were confirmed in vivo, with an observed down-regulation of eNOS and an increase in nitrotyrosine . VE-cadherin cleavage is one of the mechanisms of endothelial dysfunction and in OSAS patients, the circulating plasma values of the soluble form of VE-cadherin (sVE) were increased, suggesting an augmented endothelial permeability. This mechanism appears to be associated with ROS production, and activation of HIF-1, VEGF, and tyrosine kinase pathways . Another mechanism that contributes to the endothelial dysfunction in IH is the production of extracellular vesicles by red blood cells, and their pathogenetic mechanism involves decreased eNOS, increased Endothelin-1 (ET-1) production via the Erk1/2 pathway, and phosphorylation via the PI3K/AKT pathway ( . OSAS-induced endothelial dysfunction has also been related to some microRNAs, such as miR-630 in infantile OSAS and miR-30a, miR-34a-5p, and miR-193 in mouse models . CPAP is widely used in the treatment of OSAS, and its systemic benefits have also been observed at the molecular level regarding endothelial function. Myocardial damage is also linked to microRNA expression alterations such as miR-146a-5p . 2.1.1. Mitochondrial Involvement ROS are mainly produced at the mitochondrial level during the reoxygenation phase . Mitochondrial damage is confirmed by the reduction of circulating mitochondrial DNA in conjunction with the worsening of OSAS . Studies from mouse models showed that mitochondrial OS is also the basis of the damage to auditory hair cells, with aberrant mitochondrial morphology and overexpression of PGC-1α and Tfam mRNA . The role of mitochondria in recurrent hypoxia damage is not limited to neuronal cells, and studies on genioglossus and palatine muscles have also shown damage at this level . 2.1.2. Inflammatory Signaling IH leads to the formation of proinflammatory factors such as tumor necrosis factor (TNF), C-reactive protein (CRP), and interleukins (IL)-6 and -8 , and elevated values of NF-kB and TNF-α are linked to OSAS and daytime sleepiness. Cytokines related to NF-kB are also consequently increased, such as IL-8 , whereas IL-17 levels are also correlated with OSAS and its severity, as well as an inverse correlation with Vitamin D levels in enrolled patients . CRP and TNF-α levels decrease after OSAS surgery, but they still remain higher than those of healthy control groups . Further molecules whose expression is altered are Osteoprotegerine, Chitinase 3-like protein 1 (YKL-40), and Cardiotrophin-1 (CT-1), and their value correlates with AHI . 2.1.3. Peripheral Blood Cells Various changes in peripheral blood cells have been observed. Overexpression of Toll-Like Receptor (TLR) was observed in circulating monocytes of OSAS patients , and the TLR 6 gene is upregulated in peripheral blood cells through DNA methylation. In particular, the cytosine-phosphate-guanine (CPG) site number 1 is hypermethylated in patients with severe OSAS, and its methylation is reduced after at least 6 months of CPAP therapy . Epigenetic studies also identified hypermethylation of interleukin 1 receptor 2 (IL1R2) and androgen receptors with increased expression of both . 2.1.4. Cardiovascular Implication of OSAS Inflammation The role of inflammation has also been observed in the aortas of OSAS mouse models, evidencing an accumulation and proliferation of pro-inflammatory metabolic M1-like macrophages highly expressing CD36 and an increase in the transcription of atherogenic pathways, inflammation, and OS . Increased circulating fibrinogen values are associated with an elevated risk of cardiovascular events as well as OSAS, correlating high serum levels with the severity of the disease . The inflammatory molecules increased in OSAS patients are numerous and also include heat shock protein 70, tissue fat, monocyte chemotactic protein-1, and highly sensitive C-reactive protein . The prothrombotic state due to OSAS is also documented by the increase in tissue plasminogen activators 1 and 2 and the decrease in TGF-β and urokinase-type plasminogen activator . Epigenetic dysregulation of DNA in OSAS was shown by aberrant methylation of the formyl peptide receptor (FPR) 1, 2, 3 genes, and FPR1 overexpression and the deficiency of FPR2 and FPR3 were associated with OSAS and its severity as well as with the development of diabetes mellitus and cardiovascular diseases. Literature evidence did not show a correlation between OS and cardiovascular risk in OSAS patients . Neurotrophins are proteins that regulate the nervous system; alterations of some of them, such as brain-derived neurotrophic factor (BDNF) and nerve growth factor (NGF), can also lead to cardiovascular complications. BDNF is related to cardiomyocyte contractility and its alterations with atherosclerosis and hypertension, and NGF plays a role in atrial autonomic OSAS-induced atrial fibrillation . ROS are mainly produced at the mitochondrial level during the reoxygenation phase . Mitochondrial damage is confirmed by the reduction of circulating mitochondrial DNA in conjunction with the worsening of OSAS . Studies from mouse models showed that mitochondrial OS is also the basis of the damage to auditory hair cells, with aberrant mitochondrial morphology and overexpression of PGC-1α and Tfam mRNA . The role of mitochondria in recurrent hypoxia damage is not limited to neuronal cells, and studies on genioglossus and palatine muscles have also shown damage at this level . IH leads to the formation of proinflammatory factors such as tumor necrosis factor (TNF), C-reactive protein (CRP), and interleukins (IL)-6 and -8 , and elevated values of NF-kB and TNF-α are linked to OSAS and daytime sleepiness. Cytokines related to NF-kB are also consequently increased, such as IL-8 , whereas IL-17 levels are also correlated with OSAS and its severity, as well as an inverse correlation with Vitamin D levels in enrolled patients . CRP and TNF-α levels decrease after OSAS surgery, but they still remain higher than those of healthy control groups . Further molecules whose expression is altered are Osteoprotegerine, Chitinase 3-like protein 1 (YKL-40), and Cardiotrophin-1 (CT-1), and their value correlates with AHI . Various changes in peripheral blood cells have been observed. Overexpression of Toll-Like Receptor (TLR) was observed in circulating monocytes of OSAS patients , and the TLR 6 gene is upregulated in peripheral blood cells through DNA methylation. In particular, the cytosine-phosphate-guanine (CPG) site number 1 is hypermethylated in patients with severe OSAS, and its methylation is reduced after at least 6 months of CPAP therapy . Epigenetic studies also identified hypermethylation of interleukin 1 receptor 2 (IL1R2) and androgen receptors with increased expression of both . The role of inflammation has also been observed in the aortas of OSAS mouse models, evidencing an accumulation and proliferation of pro-inflammatory metabolic M1-like macrophages highly expressing CD36 and an increase in the transcription of atherogenic pathways, inflammation, and OS . Increased circulating fibrinogen values are associated with an elevated risk of cardiovascular events as well as OSAS, correlating high serum levels with the severity of the disease . The inflammatory molecules increased in OSAS patients are numerous and also include heat shock protein 70, tissue fat, monocyte chemotactic protein-1, and highly sensitive C-reactive protein . The prothrombotic state due to OSAS is also documented by the increase in tissue plasminogen activators 1 and 2 and the decrease in TGF-β and urokinase-type plasminogen activator . Epigenetic dysregulation of DNA in OSAS was shown by aberrant methylation of the formyl peptide receptor (FPR) 1, 2, 3 genes, and FPR1 overexpression and the deficiency of FPR2 and FPR3 were associated with OSAS and its severity as well as with the development of diabetes mellitus and cardiovascular diseases. Literature evidence did not show a correlation between OS and cardiovascular risk in OSAS patients . Neurotrophins are proteins that regulate the nervous system; alterations of some of them, such as brain-derived neurotrophic factor (BDNF) and nerve growth factor (NGF), can also lead to cardiovascular complications. BDNF is related to cardiomyocyte contractility and its alterations with atherosclerosis and hypertension, and NGF plays a role in atrial autonomic OSAS-induced atrial fibrillation . Previous research has shown that OSAS induces an increase in the blood levels of endocannabinoids such as anandamide and ethanolamide, which are associated with an increase in blood pressure and cardiovascular risk . Likewise, an increase in the blood levels of adenosine, epinephrine, norepinephrine, and aldosterone was also observed , as was an increase in retinoids, carotenoids, and tocopherol, which increase the susceptibility to vascular pathologies . Levels of saturated fatty acids and n-3 fatty acids also correlate with sleep quality, duration, and rapid eye movements. An increase in lactic acid and some fatty acids such as arabinose, arabitol, cellulose, glyceraldehyde, and threitol has been observed in these patients , and metabolomic studies have also shown an increase in glutamic acid, deoxy sugar, arachidonic acid, phosphatidylethanolamine, sphingomyelin, and lysophosphatidylcholine. The metabolic alterations mentioned above seem to be attributable to the hypoxia induced by OSAS . In addition to the metabolites, the regulatory metabolism molecules are also impaired in OSAS patients. OSAS is independently correlated with insulin resistance and fatty liver disease, and several genes involved in cholesterol metabolism were impaired, such as malic enzyme and acetyl coenzyme A (CoA) synthetase, or acetyl-CoA carboxylase, stearoyl-CoA desaturases 1 and mitochondrial glycerol- 3-phosphate acyltransferase . OSAS also induces adipose tissue inflammation and dysfunction . The circulating omentin expression increased in OSAS . Moreover, OSAS patients have significantly increased levels of leptin, affecting sleep, ventilation, and upper airway defenses . Urine analysis from OSAS patients revealed high concentrations of adrenaline, noradrenaline, and homocysteine, which are associated with increased cardiovascular risk, as well as leukotriene E4. Increased levels of homovanillic acid, a metabolite of dopamine, and 3-4-dihydroxyphenylacetic acid were also measured . Urinary isoprostane 8, which is associated with daytime sleepiness and OSAS, as well as being an element of endothelial damage , and an increase in urinary leukotriene-4 has also been observed, which is related to atherosclerosis . The significance of these altered urine metabolites in OSAS is unknown. Several neurotransmitters could play a role in the pathogenesis of the disease. Histamine is responsible for the state of arousal in the central nervous system. In studies on mouse models, it also seems to have a role in the neuromuscular transmission that occurs from the hypoglossal nerve to the genioglossus, both of which are altered in OSAS patients . Starting from the observation that apneas increase during REM sleep, due to motor inhibition of the cervical-cephalic muscles, several authors investigated the potential mechanism of muscular inactivation. REM sleep-related upper airway collapse is due to a change in norepinephrine and serotonin secretion by cranial nerve XII. The secretory patterns and increased distribution of α1-adrenoceptors are impaired in mouse models of IH during sleep . Loss of muscle tone leading to recurrent airway collapse appears to be related to reduced activity of pharyngeal motoneurons, also due to decreased stimulation of cholinergic acetylcholine receptors. This molecule, therefore, not only acts at the level of the central nervous system with different concentrations depending on whether we are in a state of sleep or wakefulness, but its action also varies at a peripheral level and could be one of the mechanisms of muscular hypotonia pharyngeal in OSAS patients . Grace et al., observed that the G-protein-coupled inwardly rectifying potassium (GIRK) channels are also involved in muscular inactivation during REM sleep. Therefore, the authors hypothesized that the potential target to counteract the loss of muscle tone is a potassium channel, and to avoid having an undesirable systemic effect, they suggested the inwardly rectifying potassium channel Kir2.4, which is expressed almost exclusively in the motor nuclei of the cranial nerves. To date, there are no clinical studies with this hypothesis. The brain neurotrophic factor (BDNF) is a neurotrophin responsible for neuron growth, development, and plasticity. Therefore, it has a role in memory and learning mechanisms. It can cross the brain-blood barrier, and its blood level dysregulation is related to depression and cognitive decline. Despite there being no significant difference in BDNF levels between OSAS and control patients, the morning molecule expression is related to age and oxygen saturation during sleep. The proBDNF, a derived product of BDNF, relates to age and HIF-1α morning quantity . BDNF and proBDNF evening concentrations are higher in patients with alteration measured with the Athens Insomnia Scale (AIS), Pittsburg Sleep Quality Index (PSQUI) positive, and lower in those with Beck Depression Inventory (BDI) positive . There is also a correlation between plasmatic BDNF levels and the oxygen desaturation index, and a negative one with the AHI. It has been hypothesized that BDNF could be involved in apoptosis-related neural injury, contributing to OSAS-induced cognitive degeneration and psychiatric pathologies . OSAS patients with depression have lower levels of BDNF and pro-BDNF compared with OSAS patients with no mood disturbances . IH could also be responsible for peripheric neural damage. In mice models, IH-induced OS decreases BDNF and pro-BDNF expression, which increases retinal cell apoptosis . BDNF expression is altered in cognitive diseases (e.g., Alzheimer’s disease, Parkinson’s related dementia, etc.) BDNF is also involved in nociception, and some authors have hypothesized his involvement in the greater pain susceptibility of OSAS patients compared with the general population . Further neurotrophins are known and are related to OSAS, such as the nerve growth factor (NGF), neurotrophins 3 and 4, and one neurotrophic factor, the glial cell-line-derived neurotrophic factor (GDNF) . NGF is responsible for sympathetic neurons’ maturation, differentiation, and survival. His precursor protein, the proNGF, has the opposite action . Its neurological involvement in OSAS has not been observed, but it has been related to cardiac autonomic nervous system disturbances and pediatric neurogenic tonsillar inflammation and hypertrophy . GDNF is essential for dopaminergic system development and respiratory pattern generation. It has been observed that its levels are lower in patients with OSAS . OSAS exacerbates Parkinson’s disease; sleep disturbance severity and motor dysfunction have been related to IL-6 levels. In the same study, Kaminska et al. observed a correlation between BDNF and increased sleepiness . OSAS, Neurocognition and Neurofilament Oxidative stress has repercussions in many parts of the body, including the central nervous system. Therefore, OS in OSAS is also responsible for neurocognitive dysfunction. The imbalance between oxidization and antioxidants, as previously cited, could cause neuron injury in brain regions most susceptible to hypoxia and oxidative stress, such as the hippocampus and cerebral cortex regions . Different domains could be altered in OSAS patients, such as attention/vigilance, memory, global cognitive, and executive function. However, CPAP treatment seems to improve some but not all executive functions in different degrees of cognitive dysfunction. Compromised cognition could be partially reversed after CPAP. As mentioned above, several studies highlighted the association between OS and nervous system diseases such as Parkinson’s disease, Alzheimer’s disease, and epilepsy . In this direction, identification and quantification of neuronal biomarkers of axonal damage could improve the diagnostic accuracy and the prognostic assessment in the management of neurological disease. Neurofilament light chain (NFL) is a neuronal protein, exclusively located in the neuronal cytoplasm, whose levels increase in serum and cerebrospinal fluid (CSF) proportionally to the degree of neuronal axonal damage, and it is a valuable biomarker in several neurological disorders. In OSAS patients, it could be interesting to monitor NFL variations after CPAP treatment aimed to evaluate neuronal recovery. The neuronal damage has also been confirmed by the first studies, which show an increase in the expression of neurofilament (NFL) and its correlation with the severity of the disease . Further studies are needed to measure the impact of CPAP on this parameter. Oxidative stress has repercussions in many parts of the body, including the central nervous system. Therefore, OS in OSAS is also responsible for neurocognitive dysfunction. The imbalance between oxidization and antioxidants, as previously cited, could cause neuron injury in brain regions most susceptible to hypoxia and oxidative stress, such as the hippocampus and cerebral cortex regions . Different domains could be altered in OSAS patients, such as attention/vigilance, memory, global cognitive, and executive function. However, CPAP treatment seems to improve some but not all executive functions in different degrees of cognitive dysfunction. Compromised cognition could be partially reversed after CPAP. As mentioned above, several studies highlighted the association between OS and nervous system diseases such as Parkinson’s disease, Alzheimer’s disease, and epilepsy . In this direction, identification and quantification of neuronal biomarkers of axonal damage could improve the diagnostic accuracy and the prognostic assessment in the management of neurological disease. Neurofilament light chain (NFL) is a neuronal protein, exclusively located in the neuronal cytoplasm, whose levels increase in serum and cerebrospinal fluid (CSF) proportionally to the degree of neuronal axonal damage, and it is a valuable biomarker in several neurological disorders. In OSAS patients, it could be interesting to monitor NFL variations after CPAP treatment aimed to evaluate neuronal recovery. The neuronal damage has also been confirmed by the first studies, which show an increase in the expression of neurofilament (NFL) and its correlation with the severity of the disease . Further studies are needed to measure the impact of CPAP on this parameter. 2.5.1. Antioxidants There are many potentially beneficial molecules for patients with OSAS, but most of them have not been tested on humans. Manganese superoxide dismutase is protective against cortical neuron oxidative damage by IH in mouse models . Adiponectin also proved to be useful in counteracting mitochondrial damage in the genioglossus muscle of OSAS mice . In addition, ROS scavenger administration with Endavarone in mouse models of IH has been tested, showing a significant reduction in cognitive impairment associated with increased brain expression of phosphorylated-cAMP response element-binding (p-CREB) . The molecules tested in humans have not been studied in courts large enough to give indications for their use. Vitamin C and N-acetylcysteine (NAC) have shown interesting results in the reduction of OS in OSAS , and NAC reduces OS in OSAS through the reduction of peroxidized lipids and the increase in glutathione. Surprisingly, patients who received it continuously also had improvements in sleep parameters . Vitamin C, on the other hand, proved to be effective in improving the endothelial function of OSAS patients in a study that took as its reference the diameter of the brachial artery, an indirect indicator of endothelial function . Lastly, Leptin is both a drug capable of reducing free radicals, OS, and atherosclerosis in patients with OSAS . To date, the only confirmed antioxidant therapy is CPAP itself, which has shown to be able to reverse many of the molecular alterations observed in vivo, such as eNOS, nitro-tyrosine, and NF-kB in the endothelium and circulating TNF-α . 2.5.2. Non-Antioxidant-Based Therapy Estrogen-related receptor-α (ERR-α) is downregulated in OSAS patients and its ligand-binding induces the expression of fast-type muscle fibers in palatopharyngeal muscles. The interaction between estrogens and ERR-α could be a therapeutic target to reverse the muscle remodeling typical of these patients . They inhibit the overexpression of HIF-1α induced by chronic IH and improve the endurance and regeneration of the genioglossus muscle in OSAS animal models . Estrogens, in particular 17β-estradiol (E2) and a resveratrol dimer (RD), have a protective action against OSAS by limitation of HIF-1α action. A pilot study in OSAS patients evidenced that Desipramine reduced airway collapse. At the same time, its anti-inflammatory properties could be beneficial in counteracting the systemic effects of OSAS, but further clinical studies are needed on a larger scale to evaluate its application . The use of sedatives in the treatment of OSAS appears to be counterintuitive. However, it has been hypothesized that trazodone may reduce the respiratory arousal threshold and upper airway obstruction. The first phenomenon occurred in an experimental group, while the second was not significant, and the magnitude of the threshold change was not sufficient to counteract the changes due to mechanical obstruction . The Phase II Pharmacotherapy of Apnea by Cannabimimetic Enhancement (PACE) has shown encouraging preliminary results for Dronabinol. A reduction in the AHI, a reduction in the feeling of sleepiness, and good satisfaction in the treated patients have been observed . Sildenafil involves the inhibition of cyclin guanosine monophosphate phosphodiesterase 5, resulting in an increase in cyclic guanosine monophosphate and NO. Its experimentation in a randomized controlled trial in which it was compared with a placebo, however, showed a worsening of the disease . In a randomized trial, the combination of Atomoxetine, a norepinephrine reuptake inhibitor, and antimuscarinic Oxybutynin, taken before going to sleep, was shown to be able to reduce the severity of OSAS, and further studies on larger sample sizes are needed . It is important to observe that many molecules have only been tested in mouse models, such as Astragaloside IV, which showed an improvement of hypoxia-induced endothelial function ; Tauroursodeoxycholic acid, against hepatic damage induced by HI ; Pitavastatin, showing a reversal of IH-induced myocardial hypertrophy, cardiac function, perivascular fibrosis and inflammatory indices ; Allopurinol also showed beneficial effects in mouse models of OSAS with a reduction of lipid peroxidation and an improvement in cardiac function ; and for all molecules, clinical trials are needed. There are many potentially beneficial molecules for patients with OSAS, but most of them have not been tested on humans. Manganese superoxide dismutase is protective against cortical neuron oxidative damage by IH in mouse models . Adiponectin also proved to be useful in counteracting mitochondrial damage in the genioglossus muscle of OSAS mice . In addition, ROS scavenger administration with Endavarone in mouse models of IH has been tested, showing a significant reduction in cognitive impairment associated with increased brain expression of phosphorylated-cAMP response element-binding (p-CREB) . The molecules tested in humans have not been studied in courts large enough to give indications for their use. Vitamin C and N-acetylcysteine (NAC) have shown interesting results in the reduction of OS in OSAS , and NAC reduces OS in OSAS through the reduction of peroxidized lipids and the increase in glutathione. Surprisingly, patients who received it continuously also had improvements in sleep parameters . Vitamin C, on the other hand, proved to be effective in improving the endothelial function of OSAS patients in a study that took as its reference the diameter of the brachial artery, an indirect indicator of endothelial function . Lastly, Leptin is both a drug capable of reducing free radicals, OS, and atherosclerosis in patients with OSAS . To date, the only confirmed antioxidant therapy is CPAP itself, which has shown to be able to reverse many of the molecular alterations observed in vivo, such as eNOS, nitro-tyrosine, and NF-kB in the endothelium and circulating TNF-α . Estrogen-related receptor-α (ERR-α) is downregulated in OSAS patients and its ligand-binding induces the expression of fast-type muscle fibers in palatopharyngeal muscles. The interaction between estrogens and ERR-α could be a therapeutic target to reverse the muscle remodeling typical of these patients . They inhibit the overexpression of HIF-1α induced by chronic IH and improve the endurance and regeneration of the genioglossus muscle in OSAS animal models . Estrogens, in particular 17β-estradiol (E2) and a resveratrol dimer (RD), have a protective action against OSAS by limitation of HIF-1α action. A pilot study in OSAS patients evidenced that Desipramine reduced airway collapse. At the same time, its anti-inflammatory properties could be beneficial in counteracting the systemic effects of OSAS, but further clinical studies are needed on a larger scale to evaluate its application . The use of sedatives in the treatment of OSAS appears to be counterintuitive. However, it has been hypothesized that trazodone may reduce the respiratory arousal threshold and upper airway obstruction. The first phenomenon occurred in an experimental group, while the second was not significant, and the magnitude of the threshold change was not sufficient to counteract the changes due to mechanical obstruction . The Phase II Pharmacotherapy of Apnea by Cannabimimetic Enhancement (PACE) has shown encouraging preliminary results for Dronabinol. A reduction in the AHI, a reduction in the feeling of sleepiness, and good satisfaction in the treated patients have been observed . Sildenafil involves the inhibition of cyclin guanosine monophosphate phosphodiesterase 5, resulting in an increase in cyclic guanosine monophosphate and NO. Its experimentation in a randomized controlled trial in which it was compared with a placebo, however, showed a worsening of the disease . In a randomized trial, the combination of Atomoxetine, a norepinephrine reuptake inhibitor, and antimuscarinic Oxybutynin, taken before going to sleep, was shown to be able to reduce the severity of OSAS, and further studies on larger sample sizes are needed . It is important to observe that many molecules have only been tested in mouse models, such as Astragaloside IV, which showed an improvement of hypoxia-induced endothelial function ; Tauroursodeoxycholic acid, against hepatic damage induced by HI ; Pitavastatin, showing a reversal of IH-induced myocardial hypertrophy, cardiac function, perivascular fibrosis and inflammatory indices ; Allopurinol also showed beneficial effects in mouse models of OSAS with a reduction of lipid peroxidation and an improvement in cardiac function ; and for all molecules, clinical trials are needed. Some authors have attempted to identify OSAS biomarkers. From the studies of Fleming et al. it has emerged that altered values of glycated hemoglobin, c-reactive protein, and erythropoietin may be useful in OSAS diagnosis . Variants of the 5-hydroxytryptamine receptor 2A have also been studied as markers of risk and severity of OSAS. It has been observed that some of them can be protective, while others predispose to greater severity of the disease . The study of the overexpression of the genes disintegrin and metalloproteinase domain 29 (ADAM29), solute carrier family 18 (vesicular acetylcholine) member 3 (SLC18A3), and fibronectin-like domain-containing leucine-rich transmembrane protein 2 (FLRT2) showed how these latter are overexpressed in Asian subjects with OSAS and could be used to screen patients, at risk for the severe form of the disease . Numerous genetic polymorphisms have been associated with the development of OSAS, but none of them have been validated in clinical trials as useful for screening or diagnosis . MicroRNAs could be helpful for diagnosis, but we still don’t have clinical validations in OSAS , such as the downregulation of miR-664a-3p , and dysregulation of miR-126-3p, miR-26a-5p, and miR-107, which associate with arterial hypertension in OSAS patients . The most accurate OS markers in OSAS are thioredoxin, malondialdehyde, superoxide dismutase, and iron reduction . Thioredoxin (TRX) concentration is a marker of disease severity as it is proportional to that of AHI and is inversely related to oxygen saturation . Although several molecules related to OS were not significantly increased in OSAS, glutathione, 8-isoprostane, substances reactive to barbituric acid (TBARS), catalase activity, copper-zinc superoxide dismutase (SOD), and products of lipid peroxidation . However, there is no agreement between SOD and malondialdehyde (a TBARS), because, in some studies, their levels seem to be correlated with the severity of OSAS . The markers of the loss of antioxidant capacity in OSAS patients have been observed in several studies as the lowering of the antioxidant power of reduced iron (FRAP), the concentration of reduced iron, and the total serum antioxidant status (TAS) . Patients with OSAS often have an altered lipidic profile. The molecular imbalances above mentioned could determine alterations detectable before OSAS leads to the typical increase in total cholesterol, triglycerides, low-density lipoprotein, high-density lipoprotein, and low-density lipoprotein cholesterol . The metabolic imbalance observed in the OSAS leading to an increase in glycolysis products seems to be also attributable to the action of HIF-1α which also induces glycolytic enzymes . However, the list of impaired molecules in OSAS is wide, such as cardiolipin, phosphatidylcholine, phosphatidylethanolamine, bile acids, and oxylipids . Since a correlation between OSAS and Vitamin D levels has been observed, its concentration measurement and possible supplementation could be useful, not to improve apneas but for the systemic damage connected to the deficiency . IH causes increased production of ROS and a reduction of endogenous antioxidant molecules . OS is a crucial component of dysfunctional pathologies associated with OSAS, such as obesity, hypertension, dyslipidemia, sympathetic activation, and diabetes. It has even been hypothesized that the OS produced by IH favors the development of obesity, thus favoring the development of OSAS . It has been observed that the reduction of antioxidant enzymes occurs through the methylation of DNA, which is reversible with the normalization of breathing, inverting ROS production. Along with it, the chemosensory reflex of the carotid body and hypertension, which are impaired following IH, also stabilize . Aldosterone levels increase with the increasing severity of OSAS. Aldosterone is a molecule related to resistant hypertension, just as OSAS is a disease related to the same disorder . The interaction between the increase in all the proinflammatory molecules and the endothelium could be one of the causes of the increased cardiovascular risk in OSAS, such as TNF, interleukins, NF-kB, TLR receptors, myeloid-related protein-8/14, the accumulation of pro-inflammatory metabolic M1-lie macrophages highly expressing CD36, fibrinogen, shock protein 70, monocyte chemotactic protein 1, highly sensitive C-reactive protein, P-selectin, soluble CD40, ICAM-1, VCAM-1, L-selectin, E-selectin and MPO . Some authors hypothesized the role of microRNAs (miR-126-3p, miR-26a-5p, and miR-107) in the diagnosis of arterial hypertension in OSAS patients . In our opinion, they could be useful for better understanding the etiopathogenetic mechanisms, but they are certainly less relevant from a hypertension diagnostic point of view. In the brain, chronic IH is associated with hippocampal cortical damage. The OS produced by the cycles of ischemia and reoxygenation with ROS production is thought to mimic that of stroke. Confirmation of this phenomenon was investigated in mouse models, and an increase in ROS and OS response markers was observed. Recurrent ischemia induced for prolonged times induces an increase in the molecules produced by the action of ROS, such as oxidized proteins, peroxidized lipids, and oxidized nucleic acids, with activation of caspase 3 and neuronal cell apoptosis. Further, confirming the role of OS in neuronal degeneration following OSAS, it was observed that mice overexpressing ROS scavenger molecules were less susceptible to neuronal damage following recurrent ischemia . Oxidative damage from IH also affects areas of the sleep-wake rhythm, which could aggravate the persistent feeling of sleepiness . Regarding the potential neurocognitive dysfunction induced by OSAS, it should also be noted that markers of Alzheimer’s disease (Aβ40, t-tau, p-tau) are increased in affected patients . In the study of OSAS-related cognitive impairment, it was observed that it is associated with the expression of miR-26b and miR-207 . The chronic IH that occurs in OSAS patients also leads to the formation of ROS at the cortical level, and their production increases above all during the re-oxygenation phases. ROS are considered responsible for cortical oxidative damage and the reduction of neurocognitive functions. The maximum production of ROS appears to occur at the mitochondrial level . Mitochondrial OS is also responsible for damage to auditory sensory neuronal hair cells. The study of mouse models of hypoxia has highlighted alterations in mitochondrial morphology, and alterations in mRNA expression and leaves us with the prospect of experimenting with drugs active at this level to prevent hearing loss in OSAS patients . It should be evaluated whether a mechanic counteracting airway collapse, associated with substances that reduce or reverse mitochondrial oxidative damage to cortical neurons, can alleviate neuronal damage. Several molecules are potentially useful in counteracting OS, cardiovascular, and neuronal damage due to IH, such as NAC, Vitamin C, Leptin, Dronabinol, and a combination of Atomoxetine and Oxybutinin . From a clinical point of view, however, it is also useful to know which drugs are not suggested in patients suffering from a specific pathology. Indeed, even if the randomized trial on Sildenafil showed a negative impact on OSAS, the study is useful from a clinical point of view. Patients with OSAS suffer more frequently from erectile dysfunction; therefore, it is necessary to remember in their treatment not to administer this drug as it is pejorative for their disease . The only confirmed antioxidant therapy is CPAP itself, which is able to reverse many of the molecular alterations . Therefore, the utility of molecular antioxidant therapies must be viewed with caution. The mechanisms underlying cell damage could be disrupted by the restoration of normal nocturnal oxygenation through CPAP. This phenomenon has been observed, for example, with the return to normal values of eNOS, nitrotyrosine, and NF-kB in the endothelium of OSAS subjects after the use of CPAP and, as noted by Ryan et al., with TNF-α . Similarly, CPAP also reduces silent brain infarctions, which are usually increased in subjects with OSAS. Therefore, also in this case, the mechanical action of the positive pressure and the rebalancing of nocturnal oxygenation are sufficient to interrupt the phenomenon . CPAP rebalances the expression of numerous pro-inflammatory and pro-coagulation molecules, but it is not always able to restore the levels of healthy subjects. This is the case with inhibitors of plasminogen-1 activation and TGF-β . Importantly, not all damages appear to be reversible with CPAP. Aortic injury mediated by pro-inflammatory metabolic M1-like macrophages highly expressing CD36, upregulation of atherogenic pathway transcription, inflammation, and OS in mouse models, once triggered, is not reversed by a return to normal oxygenation values . Despite the identification of numerous molecules whose levels are significantly increased in patients with OSAS, studies that have tried to identify diagnostic or severity biomarkers have not led to conclusive results . Plasma thioredoxin (TRX) is a marker of OS, and CPAP is able to reduce its concentration after 1 month of treatment . In this direction, there are several molecules that can potentially support the clinician in selecting patients for specific antioxidant therapy, such as MMP-2, -9, highly sensitive C-reactive protein, soluble receptors for advanced glycation end-products (sRAGE), and copper (Cu). We know that cellular hypoxia induces acidosis, but an interesting observation has also been made on the role that IH has on the ability of cells to respond to it. As an effect of IH, the overproduced ROS induces the release of Ca 2+ and the entry of Na + through the activation of the Na + /Ca 2+ exchanger, resulting in an increase in Na + ions that inhibit the activity of the Na + /H + exchanger, leading to an accumulation of H + ions and acidosis . 4.1. Literature Research We performed a narrative literature review with articles from PubMed, Embase, and the Cochrane Central Register of Controlled Trials. We considered articles concerning molecular and metabolic alterations, OS, biomarkers, and antioxidant therapy in OSAS patients. We also analyzed the bibliography of the selected manuscripts for further relevant articles. 4.2. Inclusion and Exclusion Criteria We have considered articles in the English language without time limits. We preferred in vivo clinical analyses, but manuscripts describing animal models were not excluded if useful for understanding and completing the mechanisms analyzed. Unpublished studies were not considered for the present review. 4.3. Data Selection As this is a narrative review, the decision regarding the inclusion of each article was addressed jointly by the authors. After a careful selection of sources, the collected evidence was discussed by the authors and summarized in this manuscript. We performed a narrative literature review with articles from PubMed, Embase, and the Cochrane Central Register of Controlled Trials. We considered articles concerning molecular and metabolic alterations, OS, biomarkers, and antioxidant therapy in OSAS patients. We also analyzed the bibliography of the selected manuscripts for further relevant articles. We have considered articles in the English language without time limits. We preferred in vivo clinical analyses, but manuscripts describing animal models were not excluded if useful for understanding and completing the mechanisms analyzed. Unpublished studies were not considered for the present review. As this is a narrative review, the decision regarding the inclusion of each article was addressed jointly by the authors. After a careful selection of sources, the collected evidence was discussed by the authors and summarized in this manuscript. OSAS is not only a pathology related to sleep dysfunction but also has a significant systemic impact. OS and IH lead to impaired endothelial function, osteoporosis, metabolic alterations, systemic inflammation, cardiovascular complications, central and peripheral neuronal degeneration, and pulmonary remodeling. To counteract systemic effects, therapies based on NAC, Vitamin C, Leptin, or a combination of Dronabinol and Atomoxetine appear to have promising results. Currently, the only approved therapy is CPAP, which is also capable of reversing many of the observed molecular alterations. Therefore, drug therapy can be useful in the treatment of all those dysfunctions that remain even after the restoration of normal nocturnal oxygenation. OSAS patients are more frequently subject to erectile dysfunction; however, Sildenafil should not be prescribed because it worsens the underlying disease. OSAS patients are also more prone to vitamin D deficiency; therefore, this should be sought in newly diagnosed patients and corrected. In conclusion, it is important to discover new biomarkers through innovative diagnostic tools , opening a new translational phase aimed at tuning oxidative profiles in OSAS patients. |
Biofilm Formation and Control of Foodborne Pathogenic Bacteria | f6909921-6845-42d3-853e-34827a88429b | 10058477 | Microbiology[mh] | Biofilms are defined as communities of microorganisms that are attached to living or abiotic surfaces, and they are common to the growth patterns of microorganisms in nature. Biofilms offer resistance to extreme environments and can protect microorganisms from ultraviolet (UV) radiation, extreme pH, extreme temperature, high salinity, high pressure, malnutrition, antibiotics, etc., thus acting as “protective clothing” for microorganisms . The resistance of biofilms to environmental extremes allows for the creation of suitable habitats for microbial populations and facilitates material and information exchange between microorganisms; thus, biofilms are self-protective mechanisms in microbial growth . The morphological structure, sensitivity to environmental factors, and biological characteristics of microorganisms in biofilms are quite different from those of plankton, and the three-dimensional structure of biofilms also appears to provide a natural barrier and protective layer for microorganisms . In addition, studies have now established that biofilms are the primary source of contamination during food contamination and that the persistence of biofilms on food contact surfaces and equipment is a key factor that serves as an enduring source of contamination. In the food industry, some microorganisms that are inherent in food products are harmless to consumers and in some cases may provide some benefits (for example, microorganisms introduced in the form of ferments in fermented foods, probiotics in yogurt, Saccharomyces cerevisiae in rice wine, etc.). Therefore, unless there is excessive growth or visible food spoilage, no effort is usually made to remove such microorganisms from the processing environment. However, biofilms formed by pathogenic microorganisms and decaying microorganisms are unfavourable microbial reservoirs. Such microbial cells are likely to contaminate raw materials and food during processing, leading to food spoilage and economic losses to producers . Pathogenic and putrefying bacteria are also major obstacles in the food industry and healthcare sector, as their ability to form biofilms shields them from ordinary cleaning procedures and allows them to persist in the environment. This persistence results in an increased microbial burden on the food processing environment and in the final food product, which further leads to spoilage and reduced shelf life, as well as increased risks from outbreaks of infectious diseases originating from food sources. Biofilms are a substantial problem in many food processing sectors, including dairy processing, seafood processing, meat processing, food brewing, and fresh produce . In the dairy processing industry, the most common bacteria associated with food contact surfaces include Lactobacillus, Listeria , Enterobacter , Micrococcus, Bacillus , Staphylococcus , Streptococcus, and Pseudomonas . In the meat processing industry, the main pathogens that need to be controlled include Staphylococcus aureus , Campylobacter , Escherichia coli O157:H7, Salmonella, and Listeria monocytogenes . In the fish product processing industry, the main pathogenic bacteria that need to be controlled include Escherichia coli , Vibrio , Listeria monocytogenes , Clostridium , Salmonella enteritidis, and Staphylococcus aureus . In fact, these foodborne pathogens often form biofilms on living and abiotic surfaces during infections, resulting in cross-contamination and food safety issues. To gain a better understanding of microbial biofilms and foodborne pathogens and determine elimination strategies to avoid food contamination, researchers have carried out a large number of studies. This review describes bacterial biofilm formation, elaborates on the problem associated with biofilms in the food industry, enumerates several kinds of common foodborne pathogens in biofilms, summarizes the current strategies used to eliminate or control harmful bacterial biofilms, introduces current and emerging control strategies, and emphasizes the future development prospects with respect to bacterial biofilms.
Biofilms are bacterial aggregation membranes formed by microorganisms, and these adhere to the surfaces of living or nonliving solids. Bacteria wrap themselves in an extracellular matrix by secreting extracellular polymers. In short, these biological membranes are attached to the surfaces of complex microbial communities; the microbes produce a polymer matrix consisting of extracellular polymers that are mainly composed of proteins, lipids, polysaccharides, and nucleic acids (RNA and DNA outside the cell (eDNA)), and this matrix forms a highly hydrated mixture that helps support the biofilms and their three-dimensional structures . Biofilms can consist of individual microbial species or different combinations of protozoan, bacterial, archaeal, filamentous fungal, yeast, and algal species that form a complex three-dimensional microbiome or form extracellular polymer structures such as flocs or granules [ , , ]. Extracellular polymers provide protection for biofilm residents by concentrating nutrients, preventing the entry of biocides, isolating metals and toxins, and preventing desiccation. The ability of microorganisms to form biofilms has been proven to be an adaptive property of microorganisms . Currently, scientists generally accept that biofilms are the primary way by which bacteria survive and grow in natural environments. Scientists at the U.S. National Institutes of Health have shown that approximately 80% of persistent bacterial infections are connected to biofilms . The formation of biofilms affords a new survival mechanism, offering bacteria more suitable habitats than those of planktonic microbes. It facilitates stronger growth capacity, easier access to nutritional resources, higher survival rates when exposed to biocides, stronger capacity to evade a body’s immune system, higher biological productivity and interactions, and higher environmental stability in nutrient-poor environments [ , , ]. Thus, biofilms provide protection for bacteria and shields them from adverse environmental pressures and antimicrobial agents under certain conditions to achieve a more favourable external environment. However, if foodborne pathogens form biofilms, the risks from foodborne disease infections can be strongly exacerbated, which can cause major public health risks and lead to adverse economic consequences. Therefore, the study of biofilms and their elimination strategies in the food industry is one of the most important research areas today.
The formation and maturation of biofilms is a continuous, dynamic, and complex process that depends on the matrix, culture medium, intrinsic characteristics of cells, signalling molecules, cell metabolism, and genetic control . Biofilm formation consists of five successive steps: (1) reversible attachment; (2) irreversible adhesion; (3) early development of biofilm structure (formation of small colonies); (4) biofilm maturation; and (5) cell separation and diffusion ( ). The formation of bacterial biofilms begins with the uptake of organic molecules (such as proteins, lipids, polysaccharides, fatty acids, etc.) or inorganic molecules (such as inorganic salt, water, etc.) to form an appropriate surface layer, which is then embedded in a heterogeneous structure of extracellular polymers (EPS) in single or mixed communities . Once bacteria are attached to a living or nonliving surface, they communicate with each other through an extracellular signalling system based on quorum sensing (QS) . QS can regulate the whole stage of biofilm formation, activating certain genes in bacteria to secrete extracellular matrices, such as EPS and proteins, and gradually form a complete and mature biofilm structure. QS intercellular communication is also a controlling factor in biofilm maturation. QS is a process by which chemical communication between bacterial cells mediates the production, release, and accumulation of extracellular signal molecules. Chemical signal molecules are called autoinducers. These autoinducers are continuously produced by bacterial cells, so the level of autoinducers increases as the cell number increases ( ). When autoinducers reach a minimum threshold range, these autoinducers are able to induce triggered signal transduction cascades that lead to multicellular responses in microbial populations. In other multicellular reactions, this mechanism can be involved in regulating biofilm formation, especially during the production of extracellular polysaccharides and the formation of channels or columnar structures. The formation of these structures ensures the transport of nutrients to cells in a biofilm community . In addition, bacteria usually integrate the information encoded in some QS automatic induction factors into the control of gene expression to achieve mutual communication between microorganisms . The normal operation of the QS system requires the participation of signal molecules, and different types of bacteria secrete different signal molecules, such as acylated homoserine lactones (AHLs) secreted by Gram-negative bacteria, autoinducing peptides (AIPs) secreted by Gram-positive bacteria, and autoinducer-2 (AI-2) secreted by both Gram-negative and Gram-positive bacteria. This shows that the regulatory mechanisms of Gram-positive and Gram-negative bacteria involved in biofilm formation may be different. Biofilm formation is a process by which microorganisms alter their phenotype to adapt to environmental stresses or immune responses. Throughout the formation of multispecies biofilms, biofilm regulatory genes are activated and function accordingly. Furthermore, interactions between multiple species increase the possibility for biofilms to regulate genetic changes. Therefore, the formation mechanism of multispecies biofilms is intimately connected to QS, EPS, biofilm regulation genes, and additional elements. Evidence now exists that interactions between different species can significantly boost the resistance of multispecies biofilms to biocides . Multispecies biofilms are characterized by greater resistance to disinfectants than single-species biofilms. Therefore, it is crucial to fully understand the mechanisms and environmental conditions that control biofilm formation to reduce the microbial risks associated with biofilm formation.
Foodborne pathogens and their biofilms are the main causes of foodborne diseases, which strongly threaten food industry development and human health. On Earth, approximately 40% to 80% of microorganisms are capable of forming biofilms . Biofilms can form rapidly in food industry environments, and different microorganisms can grow on food substrates and food industry infrastructure and may lead to the formation of biofilms. In the food processing industry, microbial biofilms can appear on surfaces that come into contact with food or on surfaces that do not. In addition, a particularly important point in the food processing industry is that some biofilm-forming microorganisms that are present in food plant environments are human pathogens that can form biofilm structures on different artificial substrates, such as stainless steel, polyethylene, wood, glass, polypropylene, and rubber . At the same time, many studies have emphasized that human immunodeficiency virus and foodborne diseases are largely caused by biofilms that form on the surfaces of equipment in the food processing and medical fields , and these biofilms serve as potential hosts for pathogens and are a constant source of infection and cross-contamination. Biofilms are the cause of approximately 60% of the world’s foodborne outbreaks, so the formation and presence of microbial biofilms in food processing environments is a major concern and poses risks to food safety . In food processing environments, pollutants mainly originate from the surrounding air, equipment, or food surfaces, and if cleaning is not adequate, they are very likely to lead to the formation and accumulation of microbial biofilms, which can lead to the spoilage of food, resulting in severe public health risks and adverse economic consequences for consumers . At the same time, biofilms also create substantial technological challenges in the food industry because biofilms may prevent heat from flowing through the surfaces of equipment, increase the frictional resistance of fluids on surfaces, lead to mechanical clogging of fluid handling systems, and increase the rate of metal surface corrosion, resulting in losses in productivity . In summary, biofilms generate risks for direct pathogen contamination in the food industry, as well as the contamination of instruments and processing equipment. Biofilms are a major challenge in the food industry, as they allow bacteria to bind to a range of surfaces, such as wood, polypropylene, glass, plastic, rubber, stainless steel, and even food, in just a few minutes; mature biofilms can then form in a matter of days (or even hours) . Biofilm formation is harmful in most cases . In the food industry, foodborne pathogens can form biofilms, which can lead to food spoilage, harming the health of consumers . In hospital settings, biofilms can persist on surfaces of medical devices and patient tissues, resulting in persistent infection . In the dairy industry, some thermophilic and cryophilic bacteria are often present during the processing, pasteurisation, and preservation stages of dairy products. For example, Geobacillus spp., which can grow at temperatures up to 65 °C, and its heat-resistant spores have been shown to have significant adverse effects on the production of milk powder . The ability of cryophilic bacteria to thrive at refrigerated temperatures complicates the storage of dairy products, resulting in shorter shelf lives for dairy products. Pseudomonas was the most common psychrophilic bacterium responsible for spoilage. Without heat treatment, it can achieve high populations and form biofilms under the low-temperature conditions of milk cooling tanks and pipe walls; it can usually secrete enzymes and reduce the thermal stability of fats via protease secretion, which is a major cause of milk spoilage . In addition, some examples have emerged of biofilm-related diseases in food safety. For example, the ability of the lungs of patients with cystic fibrosis to form Pseudomonas aeruginosa biofilms is a classic instance of biofilm involvement in chronic infections. Because Pseudomonas aeruginosa accumulates in biofilms, this chronic infection is usually incurable and ultimately leads to death in cystic fibrosis patients . Staphylococcus aureus can cause food poisoning by producing enterotoxins. Biofilms can cause diseases associated with human infections, such as otitis media, bacterial vaginitis, gingivitis, dental plaque, urinary tract infection, middle ear infection, catheter, and prosthetic joint infection, contact lens infection, or cystic fibrosis . Some of these infections are associated with antibiotic resistance and can be fatal, such as cystic fibrosis infection, heart valve infection, and endocarditis . In addition, studies have reported that infections connected with microbial biofilms can be incredibly challenging to treat and cure because the pathogenic bacteria present in biofilm communities often exhibit strong virulence and a high degree of antimicrobial tolerance or resistance, allowing them to survive even under fairly aggressive antimicrobial therapy regimens. Therefore, it is very important to find new and effective strategies to eliminate or control the formation of biofilms that harbour harmful pathogens. Food security is an issue of global concern. As risks from foodborne pathogen infections increase, managers of food manufacturing and processing plants must carry out more thorough and frequent assessments of pathogen growth. In addition, in light of the negative effects that pathogenic bacterial biofilms have on several areas of human health, the prevention, control, and elimination of harmful biofilms have become key issues in this field. Biofilm control methods and their uses in the food industry are briefly described in .
Globally, emphasis has been placed on biofilm formation by bacterial pathogens, particularly in the medical and food industries, because of the potential health risks and public health problems associated with biofilms. For example, biofilms not only have antimicrobial resistance and mechanical persistence but also produce virulence factors, among other substances [ , , ], all of which can lead to severe human health problems. The growth of biofilms, which may contain food-spoiling bacteria and foodborne pathogens, in food processing environments results in an increased potential for microbial contamination of processed products. Biofilms protect the microorganisms from disinfectants, increasing their survival rate and the likelihood of the subsequent contamination of food, leading to shorter shelf lives and risks associated with disease transmission. As a result, it is necessary and important to understand and control biofilm formation and to find strategies for biofilm elimination to reduce the microbial risks associated with biofilm formation. Overall, current strategies for controlling harmful biofilms fall into three main categories : (i) modifying abiotic surface features to prevent biofilm formation; (ii) regulating signalling pathways to inhibit biofilm formation and stimulate biofilm diffusion; and (iii) applying external forces to eradicate biofilms. In the food industry, the best strategy to eradicate bacterial biofilms is to prevent biofilm formation. This can be conducted by avoiding the formation of bacterial biofilms in key locations via methods such as aseptic processing, regular disinfection cleaning, and the sterilisation of equipment after use. However, in most cases, especially during food processing, it is neither possible nor cost-effective to sterilize all environments in production areas. Therefore, other measures must be taken to decrease the population of harmful bacteria and biofilms in manufacturing areas. It is necessary to prevent the formation of biofilms by carrying out regular cleaning and disinfection so that cells do not attach firmly (reversibly) to contact surfaces. The disinfectants most widely used in sanitary disinfection programmes in the food industry are quaternary ammonium compounds (QACs), hypochlorites, peroxides (peracetic acid and hydrogen peroxide), chloramines, iodine, ozone, aldehydes (formaldehyde, glutaraldehyde, paraformaldehyde), and phenols. Today, alkyl amines, chlorine dioxide, and quaternary ammonium mixtures are also included in disinfection programmes . They react with various components of bacterial cells and thus exhibit harmful effects on bacterial cells. Representative compounds for the most common disinfectants used in sanitary disinfection programmes in the food industry are briefly described in . In the food industry, heat treatment is also a measure employed to decrease the number of harmful bacteria and biofilm populations in production areas. Steam is a promising heat treatment technology for biofilm inactivation . Compared with other conventional heat treatments, steam heat treatment technology has the following advantages . First, steam heat treatment technology can operate in an oxygen-free environment, and steam has a high heat transfer capacity during condensation. In addition, steam can easily access surface fissures or fractures in cells, thus effectively eliminating foodborne pathogens. Some studies have shown that steam pasteurisation is an efficient approach for the rapid inactivation of foodborne pathogens due to the high heat capacity . Kim et al. demonstrated that the inactivation of biofilms on diverse nonbiological surfaces can be accomplished extremely well using steam heat treatment technologies. The capacity of superheated steam heat treatment to destroy foodborne pathogens is large . In recent years, more effective and environmentally friendly control strategies have also been discovered to eliminate or control the formation of harmful biofilms. For example, subinhibitory concentrations of ibuprofen have been demonstrated to reduce biofilm formation by E. coli , Staphylococcus aureus , Streptococcus pneumoniae , and Candida albicans on abiotic surfaces . Bacteriophages and phage lysosomes have also been shown to be useful as antibiofilm agents to achieve better control of biofilm formation . In contrast to phage lysosomes, phages can not only directly kill bacteria but also induce host bacterial expression of EPS degradation enzymes, thus accelerating the removal of mature biofilms . Furthermore, combined techniques can also be used in which multiple bacteriophages or phage lysates are employed to achieve a broad spectrum of antibacterial effects. In addition, the development or research into other physical surface decontamination technologies for the eradication of bacteria from biofilms has become increasingly popular in recent years, including photodynamic inactivation using pulsed ultraviolet light, electron beam irradiation, steam heating, irradiation at 405 nm, or surface treatment with ozone, ultrasound, or gaseous chlorine dioxide . Several novel biofilm eliminations and control methods used in the food industry are briefly summarized in . All of these cutting-edge methods hold out hope for the future in preventing biofilm formation in the food industry. These techniques are now the subject of ongoing studies, but we think that such products will undoubtedly become available soon. We also hope that researchers can uncover the complete mechanisms of biofilm antibiotic resistance in the near future to lay a solid foundation for the design and development of new biofilm antibiotics.
Food contamination by pathogenic microorganisms has developed into a significant public health issue and has resulted in significant economic losses worldwide. Foodborne pathogenic bacteria can adhere to food by forming biofilms and survive for long periods of time on surfaces that come in contact with food, resulting in postprocessing contamination, a reduction in product quality and shelf life, and potential disease transmission, which are significant food safety problems in the food industry. A summary of common foodborne pathogens and their implications for food safety are briefly described in . In the following sections, we discuss the food safety and clinical aspects connected with the five most significant foodborne bacterial pathogens ( Listeria monocytogenes , Salmonella enteritidis , Pseudomonas aeruginosa , Staphylococcus aureus , and Escherichia coli ), as well as their ability to form biofilms on various surfaces. 6.1. Listeria monocytogenes Foodborne pathogens can adhere to food items by forming biofilms and survive for long periods of time on surfaces that come in contact with food, and this introduces significant issues related to food safety in the food industry. Listeria monocytogenes is the main pathogen related to foodborne diseases around the world and has a high fatality rate and hospitalisation rate . Listeria monocytogenes is a gram-positive, aerobic, nonsporoforming, rod-shaped bacterium that belongs to the genus Listeria firmicide. Out of the 17 listeria species described, it is the only pathogenic species and is the pathogenic factor in listeriosis . It can infect a large number of host organs, such as the liver, spleen, cerebrospinal fluid, and blood, among which the liver is the main site of infection . Meningitis, sepsis, and other central nervous system infections are common in Listeria patients. In healthy adults, diarrhea and fever are the primary symptoms . In pregnant women, listeriosis may cause fever, diarrhoea, spontaneous abortion, or stillbirth . In neonates, listeriosis can cause septicemia, pneumonia, and meningitis . Listeria monocytogenes may potentially cause a noninvasive illness commonly referred to as febrile gastroenteritis or noninvasive gastroenteritis, which has been associated with contaminated deli meats, chocolate milk, cheese, smoked fish and corn . Pregnant women, foetuses or newborns, the elderly, and people with compromised immune systems are at high risk for developing diseases such as sepsis, meningitis, or gastroenteritis. In general, aminopenicillin or benzylpenicillin alone or in combination with aminoglycosides are the antibiotics typically recommended for Listeria monocytogenes infection [ , , ]. The pathogenic factors in listeriosis are ubiquitous in nature and can invade the food processing environment. According to a two-year survey conducted by Wu et al. , Listeria monocytogenes had the highest contamination rate in China’s food industry, accounting for approximately 20%. Additionally, 99% of listeriosis cases were brought on by the consumption of food tainted with Listeria monocytogenes , and only a small number of cases were brought on by pathogens found in the environment . At the same time, food processing contamination is the primary transmission route for Listeria monocytogenes . Therefore, the first line of defence in the control and prevention of listeriosis is routine cleaning, disinfection processes, and the application of appropriate food hygiene standards during food preparation. Quaternary ammonium compounds and chlorine-based biocides are the two biocides most often utilized for Listeria monocytogenes , when used at recommended dosages, are highly effective against Listeria monocytogenes . 6.2. Salmonella enterica Salmonella enterica is among the most prevalent foodborne pathogens worldwide and has been linked to high-profile outbreaks in many foods. It has two species, S. Enterica and S. Bongori , and more than 2500 known serum variants, and it is a gram-negative, facultatively anaerobic, flagellated enterobacter. Salmonella is a human and animal pathogen that causes salmonellosis, which is the most typical (85%) foodborne illness . Salmonellosis is an infectious disease, and the pathogen responsible causes human illnesses that begins with gastroenteritis and end with systemic infections. Approximately 99.5% of all isolates of pathogenic Salmonella in humans and other warm-blooded animals were of the species Salmonella enterica . Therefore, Salmonella enterica is among the most important intestinal bacterial foodborne pathogens . Salmonellosis usually manifests as gastroenteritis, accompanied by fever, diarrhoea, and abdominal cramps. The symptoms of salmonellosis are usually mild and can be cured without therapy within 1–4 days, but in severe cases, salmonellosis can lead to acute gastroenteritis, food poisoning, sepsis, etc. . The severity of disease manifestation depends on a patient’s susceptibility to the pathogen and the virulence of the particular serum variant. Due to the widespread incidence and severity of this disease, the prevalence of Salmonella enterica in a country’s food supply has been considered an important benchmark for public health . The major source of Salmonella enterica infections in humans is food, such as eggs, egg products, and poultry meat . Contaminated food is the primary mode of transmission of Salmonella enteritidis , which has a high survival rate and can thrive on undercooked or improperly stored meat and animal products. In some instances, Salmonella infections can sometimes persist for several years without showing any overt clinical symptoms in both people or animals. Therefore, salmonellosis is a significant issue for human health because large numbers of animal hosts exist, transmission is easy, and carriers can be asymptomatic. In addition, the persistence of Salmonella in the food industry is a primary food health issue, as it can form biofilms in food processing environments and become a potential host for food contamination. Therefore, Salmonella produces biofilms, which are key components of its pathogenicity. Salmonella , similar to other bacterial pathogens, can exist in a wide range of cell surface structures (particularly those with protein-like and carbohydrate-like properties), which may enable effective aggregation of the bacterial cells with those of other species and thus promote the formation of single or multiple biofilm cell communities. Salmonella biofilms can exist not only on biological surfaces but also on abiotic surfaces such as concrete, stainless steel, ceramic tile, glass, granite, quartz stone, rubber, and synthetic plastics . By encouraging the creation of virulence factors, and due to the mechanical resistance and antimicrobial resistance components of biofilms, they improve the odds of microbial survival . 6.3. Staphylococcus aureus Staphylococcus aureus is one of the most common foodborne pathogens related to food safety problems . Staphylococcus aureus is an enterotoxin-producing gram-positive bacterium that is often parasitic on the skin, throat, intestines, stomach, nasal cavity, carbuncle, and sores of humans and animals. It is a zoonotic pathogen that can lead to cardiovascular infection, surgical site infection, lower respiratory tract site infection, cystic fibrosis pulmonary infection, endocarditis, and pneumonia in humans and animals . Additionally, Staphylococcus aureus is a highly adaptive microbe that can live in a variety of environments (such as air, sewage, and soil) by forming biofilms. Moreover, Staphylococcus aureus is highly capable of forming biofilms on the surfaces of food, on food processing equipment, and in water, which are sources of cross-contamination of food . The control of Staphylococcus aureus in environments where food is processed is also complicated by its propensity to adhere to food-contacting surfaces and form biofilms. The ability of this species to form biofilms and achieve cell adhesion is connected to the production of polysaccharide intercellular adhesion (PIA). PIA is encoded by the ICA operon, which contains the icaADBC gene cluster. These four genes encode proteins that mediate PIA synthesis and elongation . The World Health Organization has classified Staphylococcus aureus as a high-priority species on its list of antibiotic-resistant bacteria that are dangerous to human health . Staphylococcus aureus forms biofilms as one of its most effective survival strategies; thus, it is difficult to treat even with antibiotics and causes a severe burden in medical settings. The primary issue related to Staphylococcus aureus biofilms is their resistance to antibiotics and their host defence mechanisms, and the biofilms’ properties confer increased resistance to Staphylococcus aureus pathogenic strains to antibiotics and host defence factors . This resistance is mainly achieved through the diffusion barrier formed by the polysaccharide matrix . Therefore, the emergence of MRSA strains is an issue in public health, and due to the formation of Staphylococcus aureus biofilms, bacterial sensitivity to antibiotics and even to vancomycin has decreased, making the removal of Staphylococcus aureus difficult . Staphylococcus aureus is a pathogenic bacterium that colonizes 30% to 50% of healthy people, and it can adhere to surfaces such as glass, metal, plastic, and host tissues. Infections associated with Staphylococcus aureus biofilms include osteomyelitis, endocarditis, chronic wound infections, eye infections, multimicrobial biofilm infections, and kidney abscesses . In general, to prevent Staphylococcus aureus from adhering to biological or abiotic surfaces, anti-adhesive agents such as calcium chelators, silver nanoparticles, aryl rhodamine, and chitosan can be applied to surfaces . In addition, nucleases, proteases, dispersin B, lysococcin, and hyaluronic acid lysase can disrupt and inhibit biofilms through different biofilm dispersal mechanisms. 6.4. Pseudomonas aeruginosa One of the most virulent pathogens, Pseudomonas aeruginosa , is a major contributor to a number of acute infections . In 2017, the World Health Organization designated it as a pathogen that requires high-priority research and the development of new medications . Pseudomonas aeruginosa , which belongs to the genus Pseudomonas, is an aerobic gram-negative opportunistic pathogen. It is also an important water source and conditional foodborne pathogenic bacterium that is widely distributed and resistant to adverse environments. It mainly exists in soil, dust, and water and small numbers in human intestines. It is mainly parasitic in the genital parts, anus, external auditory canal, and armpit, and can also temporarily parasitize skin surfaces. Pseudomonas aeruginosa has many pathogenic factors and is a completely pathogenic bacterium that leads to human acute intestinal diseases and skin inflammation. In addition, in individuals with severe conditions as well as those who have burns, surgical wounds, foot ulcers, and diabetes, Pseudomonas aeruginosa can lead to severe acute and chronic infections; for example, Pseudomonas aeruginosa is a significant contributor to cystic fibrosis . Pseudomonas aeruginosa infection can also occur in healthy individuals, causing external auditory canal inflammation, otitis media, keratitis, and folliculitis . If not properly treated in the acute state, Pseudomonas aeruginosa can form biofilms, establishing a chronic biofilm infection that is difficult to eradicate. Biofilm formation is certainly one of the most significant factors affecting virulence in the pathogenesis of Pseudomonas aeruginosa infections . Biofilms allow these pathogens to attach to different surfaces, providing protection from severe environmental factors and the immunological systems of hosts. In addition to these basic protections, biofilms provide microbes with a safe haven for antibiotic resistance in vivo, leading to the emergence of the MDR phenotype. Therefore, biofilm formation is a key reason that Pseudomonas aeruginosa has become a hospital pathogen and is an important indicator of the persistence of chronic bacteria. In addition, Pseudomonas aeruginosa has a distinct advantage: it can move and easily travel from one niche to another without difficulty . Three movement types have been observed, namely, smattering motion, swimming motion and convulsive motion, which enable Pseudomonas aeruginosa to exist in various environments . Currently, controlling Pseudomonas aeruginosa infections is enormously challenging because of the emergence of antibiotic-resistant strains. Pseudomonas aeruginosa is the most frequent pathogen detected in hospital-associated infections (HAI) and is the second-most frequent cause of ventilator-associated pneumonia in the United States . On the one hand, Pseudomonas aeruginosa is capable of producing a variety of virulence factors, including elastase, flagella, alkaline protease, type IV pili, lipopolysaccharide, exotoxin A, phospholipase, pyocyanin, pyochelin, pyoverdine, and Pseudomonas quinolone signal (PQS) . On the other hand, Pseudomonas aeruginosa , has a genome that is relatively large compared with that of other prokaryotes and has abnormal chromosome regulation of genes, which helps the species adapt to various environmental conditions, thus increasing the incidence of disease and mortality, and it is intimately connected to the rise of antibiotic resistance . Another cause of Pseudomonas aeruginosa resistance is the formation of biofilms. Because biofilms are not degraded by antimicrobial agents (such as disinfectants), heat, or drying and remain on living or abiotic surfaces, especially in hospitals, they can lead to contamination and the spread of infectious illnesses. Therefore, biofilms are a key contributor to infectious illnesses because they act as a barrier between the immune system and antibiotic drugs . Pseudomonas aeruginosa plays a major role in hospital-acquired infections, particularly in burn patients, so knowledge of these strains is of particular epidemiological importance for the prevention and control of Pseudomonas aeruginosa infections. 6.5. Escherichia coli Escherichia coli is common in both humans and animals as part of the regular flora and is generally harmless to humans . However, the development of virulence factors causes some strains of E. coli to become pathogenic, and as a result, they rank among one of the most prevalent foodborne pathogens associated with food safety issues. E. coli is a gram-negative, nonsporoforming, metabolically active, rod-shaped bacterium. A few special serotypes of E. coli exhibit pathogenicity, and according to their different pathogenicity, they can be roughly divided into enteroaggregative E. coli (EAEC), enterohemorrhagic E. coli (EHEC), enteroinvasive E. coli (EIEC), enteropathogenic E. coli (EPEC), enterotoxigenic E. coli (ETEC), and diffuse-adhering E. coli (DAEC), which commonly manifest in human infections as diarrhoeal illness. Among them, EHEC can produce Shiga toxin, which can cause diarrhoea, haemorrhagic enteritis, haemolytic uraemic syndrome (HUS), thrombotic thrombocytopenic purpura (TTP), and other diseases through foodborne infections . Children, the elderly, and immunocompromised people may even develop systemic infections or acute renal failure. Meningitis, sepsis, and urinary tract infections are frequently ascribed to extraintestinal E. coli pathotype infections, including neonatal meningitis-associated E. coli (NMEC), sepsis-associated E. coli (SEPEC), and uropathogenic E. coli (UPEC), respectively . Because the presence of E. coli implies unsanitary conditions in the food industry, it serves as a hygiene indicator. Regulation 2073/2005 of the European Commission states that the amount of E. coli that can be found in certain meat products (such as minced beef) cannot be more than 500 CFU/g . The main pathogenic strain of E. coli is EHEC- O157:H7, which can cause infectious diarrhoea and haemorrhagic enteritis. It mainly causes human infection through contaminated food, including fresh meat, fruits, vegetables, raw milk, and dairy products. E. coli O157:H7 has a strong pathogenic capacity and is resistance to gastric acid, and it is destructive to cells. E. coli , similar to most foodborne microbes, can form biofilms by adhering to a range of food-contact surfaces. Biofilms are more resistant to environmental stresses, such as UV light exposure, sanitising agents, nutritional and oxidative stresses, and desiccation. Consequently, biofilms are important for both public health and the economy because they cause chronic illnesses that are challenging to cure and are resistant to cleaning and sanitation .
Foodborne pathogens can adhere to food items by forming biofilms and survive for long periods of time on surfaces that come in contact with food, and this introduces significant issues related to food safety in the food industry. Listeria monocytogenes is the main pathogen related to foodborne diseases around the world and has a high fatality rate and hospitalisation rate . Listeria monocytogenes is a gram-positive, aerobic, nonsporoforming, rod-shaped bacterium that belongs to the genus Listeria firmicide. Out of the 17 listeria species described, it is the only pathogenic species and is the pathogenic factor in listeriosis . It can infect a large number of host organs, such as the liver, spleen, cerebrospinal fluid, and blood, among which the liver is the main site of infection . Meningitis, sepsis, and other central nervous system infections are common in Listeria patients. In healthy adults, diarrhea and fever are the primary symptoms . In pregnant women, listeriosis may cause fever, diarrhoea, spontaneous abortion, or stillbirth . In neonates, listeriosis can cause septicemia, pneumonia, and meningitis . Listeria monocytogenes may potentially cause a noninvasive illness commonly referred to as febrile gastroenteritis or noninvasive gastroenteritis, which has been associated with contaminated deli meats, chocolate milk, cheese, smoked fish and corn . Pregnant women, foetuses or newborns, the elderly, and people with compromised immune systems are at high risk for developing diseases such as sepsis, meningitis, or gastroenteritis. In general, aminopenicillin or benzylpenicillin alone or in combination with aminoglycosides are the antibiotics typically recommended for Listeria monocytogenes infection [ , , ]. The pathogenic factors in listeriosis are ubiquitous in nature and can invade the food processing environment. According to a two-year survey conducted by Wu et al. , Listeria monocytogenes had the highest contamination rate in China’s food industry, accounting for approximately 20%. Additionally, 99% of listeriosis cases were brought on by the consumption of food tainted with Listeria monocytogenes , and only a small number of cases were brought on by pathogens found in the environment . At the same time, food processing contamination is the primary transmission route for Listeria monocytogenes . Therefore, the first line of defence in the control and prevention of listeriosis is routine cleaning, disinfection processes, and the application of appropriate food hygiene standards during food preparation. Quaternary ammonium compounds and chlorine-based biocides are the two biocides most often utilized for Listeria monocytogenes , when used at recommended dosages, are highly effective against Listeria monocytogenes .
Salmonella enterica is among the most prevalent foodborne pathogens worldwide and has been linked to high-profile outbreaks in many foods. It has two species, S. Enterica and S. Bongori , and more than 2500 known serum variants, and it is a gram-negative, facultatively anaerobic, flagellated enterobacter. Salmonella is a human and animal pathogen that causes salmonellosis, which is the most typical (85%) foodborne illness . Salmonellosis is an infectious disease, and the pathogen responsible causes human illnesses that begins with gastroenteritis and end with systemic infections. Approximately 99.5% of all isolates of pathogenic Salmonella in humans and other warm-blooded animals were of the species Salmonella enterica . Therefore, Salmonella enterica is among the most important intestinal bacterial foodborne pathogens . Salmonellosis usually manifests as gastroenteritis, accompanied by fever, diarrhoea, and abdominal cramps. The symptoms of salmonellosis are usually mild and can be cured without therapy within 1–4 days, but in severe cases, salmonellosis can lead to acute gastroenteritis, food poisoning, sepsis, etc. . The severity of disease manifestation depends on a patient’s susceptibility to the pathogen and the virulence of the particular serum variant. Due to the widespread incidence and severity of this disease, the prevalence of Salmonella enterica in a country’s food supply has been considered an important benchmark for public health . The major source of Salmonella enterica infections in humans is food, such as eggs, egg products, and poultry meat . Contaminated food is the primary mode of transmission of Salmonella enteritidis , which has a high survival rate and can thrive on undercooked or improperly stored meat and animal products. In some instances, Salmonella infections can sometimes persist for several years without showing any overt clinical symptoms in both people or animals. Therefore, salmonellosis is a significant issue for human health because large numbers of animal hosts exist, transmission is easy, and carriers can be asymptomatic. In addition, the persistence of Salmonella in the food industry is a primary food health issue, as it can form biofilms in food processing environments and become a potential host for food contamination. Therefore, Salmonella produces biofilms, which are key components of its pathogenicity. Salmonella , similar to other bacterial pathogens, can exist in a wide range of cell surface structures (particularly those with protein-like and carbohydrate-like properties), which may enable effective aggregation of the bacterial cells with those of other species and thus promote the formation of single or multiple biofilm cell communities. Salmonella biofilms can exist not only on biological surfaces but also on abiotic surfaces such as concrete, stainless steel, ceramic tile, glass, granite, quartz stone, rubber, and synthetic plastics . By encouraging the creation of virulence factors, and due to the mechanical resistance and antimicrobial resistance components of biofilms, they improve the odds of microbial survival .
Staphylococcus aureus is one of the most common foodborne pathogens related to food safety problems . Staphylococcus aureus is an enterotoxin-producing gram-positive bacterium that is often parasitic on the skin, throat, intestines, stomach, nasal cavity, carbuncle, and sores of humans and animals. It is a zoonotic pathogen that can lead to cardiovascular infection, surgical site infection, lower respiratory tract site infection, cystic fibrosis pulmonary infection, endocarditis, and pneumonia in humans and animals . Additionally, Staphylococcus aureus is a highly adaptive microbe that can live in a variety of environments (such as air, sewage, and soil) by forming biofilms. Moreover, Staphylococcus aureus is highly capable of forming biofilms on the surfaces of food, on food processing equipment, and in water, which are sources of cross-contamination of food . The control of Staphylococcus aureus in environments where food is processed is also complicated by its propensity to adhere to food-contacting surfaces and form biofilms. The ability of this species to form biofilms and achieve cell adhesion is connected to the production of polysaccharide intercellular adhesion (PIA). PIA is encoded by the ICA operon, which contains the icaADBC gene cluster. These four genes encode proteins that mediate PIA synthesis and elongation . The World Health Organization has classified Staphylococcus aureus as a high-priority species on its list of antibiotic-resistant bacteria that are dangerous to human health . Staphylococcus aureus forms biofilms as one of its most effective survival strategies; thus, it is difficult to treat even with antibiotics and causes a severe burden in medical settings. The primary issue related to Staphylococcus aureus biofilms is their resistance to antibiotics and their host defence mechanisms, and the biofilms’ properties confer increased resistance to Staphylococcus aureus pathogenic strains to antibiotics and host defence factors . This resistance is mainly achieved through the diffusion barrier formed by the polysaccharide matrix . Therefore, the emergence of MRSA strains is an issue in public health, and due to the formation of Staphylococcus aureus biofilms, bacterial sensitivity to antibiotics and even to vancomycin has decreased, making the removal of Staphylococcus aureus difficult . Staphylococcus aureus is a pathogenic bacterium that colonizes 30% to 50% of healthy people, and it can adhere to surfaces such as glass, metal, plastic, and host tissues. Infections associated with Staphylococcus aureus biofilms include osteomyelitis, endocarditis, chronic wound infections, eye infections, multimicrobial biofilm infections, and kidney abscesses . In general, to prevent Staphylococcus aureus from adhering to biological or abiotic surfaces, anti-adhesive agents such as calcium chelators, silver nanoparticles, aryl rhodamine, and chitosan can be applied to surfaces . In addition, nucleases, proteases, dispersin B, lysococcin, and hyaluronic acid lysase can disrupt and inhibit biofilms through different biofilm dispersal mechanisms.
One of the most virulent pathogens, Pseudomonas aeruginosa , is a major contributor to a number of acute infections . In 2017, the World Health Organization designated it as a pathogen that requires high-priority research and the development of new medications . Pseudomonas aeruginosa , which belongs to the genus Pseudomonas, is an aerobic gram-negative opportunistic pathogen. It is also an important water source and conditional foodborne pathogenic bacterium that is widely distributed and resistant to adverse environments. It mainly exists in soil, dust, and water and small numbers in human intestines. It is mainly parasitic in the genital parts, anus, external auditory canal, and armpit, and can also temporarily parasitize skin surfaces. Pseudomonas aeruginosa has many pathogenic factors and is a completely pathogenic bacterium that leads to human acute intestinal diseases and skin inflammation. In addition, in individuals with severe conditions as well as those who have burns, surgical wounds, foot ulcers, and diabetes, Pseudomonas aeruginosa can lead to severe acute and chronic infections; for example, Pseudomonas aeruginosa is a significant contributor to cystic fibrosis . Pseudomonas aeruginosa infection can also occur in healthy individuals, causing external auditory canal inflammation, otitis media, keratitis, and folliculitis . If not properly treated in the acute state, Pseudomonas aeruginosa can form biofilms, establishing a chronic biofilm infection that is difficult to eradicate. Biofilm formation is certainly one of the most significant factors affecting virulence in the pathogenesis of Pseudomonas aeruginosa infections . Biofilms allow these pathogens to attach to different surfaces, providing protection from severe environmental factors and the immunological systems of hosts. In addition to these basic protections, biofilms provide microbes with a safe haven for antibiotic resistance in vivo, leading to the emergence of the MDR phenotype. Therefore, biofilm formation is a key reason that Pseudomonas aeruginosa has become a hospital pathogen and is an important indicator of the persistence of chronic bacteria. In addition, Pseudomonas aeruginosa has a distinct advantage: it can move and easily travel from one niche to another without difficulty . Three movement types have been observed, namely, smattering motion, swimming motion and convulsive motion, which enable Pseudomonas aeruginosa to exist in various environments . Currently, controlling Pseudomonas aeruginosa infections is enormously challenging because of the emergence of antibiotic-resistant strains. Pseudomonas aeruginosa is the most frequent pathogen detected in hospital-associated infections (HAI) and is the second-most frequent cause of ventilator-associated pneumonia in the United States . On the one hand, Pseudomonas aeruginosa is capable of producing a variety of virulence factors, including elastase, flagella, alkaline protease, type IV pili, lipopolysaccharide, exotoxin A, phospholipase, pyocyanin, pyochelin, pyoverdine, and Pseudomonas quinolone signal (PQS) . On the other hand, Pseudomonas aeruginosa , has a genome that is relatively large compared with that of other prokaryotes and has abnormal chromosome regulation of genes, which helps the species adapt to various environmental conditions, thus increasing the incidence of disease and mortality, and it is intimately connected to the rise of antibiotic resistance . Another cause of Pseudomonas aeruginosa resistance is the formation of biofilms. Because biofilms are not degraded by antimicrobial agents (such as disinfectants), heat, or drying and remain on living or abiotic surfaces, especially in hospitals, they can lead to contamination and the spread of infectious illnesses. Therefore, biofilms are a key contributor to infectious illnesses because they act as a barrier between the immune system and antibiotic drugs . Pseudomonas aeruginosa plays a major role in hospital-acquired infections, particularly in burn patients, so knowledge of these strains is of particular epidemiological importance for the prevention and control of Pseudomonas aeruginosa infections.
Escherichia coli is common in both humans and animals as part of the regular flora and is generally harmless to humans . However, the development of virulence factors causes some strains of E. coli to become pathogenic, and as a result, they rank among one of the most prevalent foodborne pathogens associated with food safety issues. E. coli is a gram-negative, nonsporoforming, metabolically active, rod-shaped bacterium. A few special serotypes of E. coli exhibit pathogenicity, and according to their different pathogenicity, they can be roughly divided into enteroaggregative E. coli (EAEC), enterohemorrhagic E. coli (EHEC), enteroinvasive E. coli (EIEC), enteropathogenic E. coli (EPEC), enterotoxigenic E. coli (ETEC), and diffuse-adhering E. coli (DAEC), which commonly manifest in human infections as diarrhoeal illness. Among them, EHEC can produce Shiga toxin, which can cause diarrhoea, haemorrhagic enteritis, haemolytic uraemic syndrome (HUS), thrombotic thrombocytopenic purpura (TTP), and other diseases through foodborne infections . Children, the elderly, and immunocompromised people may even develop systemic infections or acute renal failure. Meningitis, sepsis, and urinary tract infections are frequently ascribed to extraintestinal E. coli pathotype infections, including neonatal meningitis-associated E. coli (NMEC), sepsis-associated E. coli (SEPEC), and uropathogenic E. coli (UPEC), respectively . Because the presence of E. coli implies unsanitary conditions in the food industry, it serves as a hygiene indicator. Regulation 2073/2005 of the European Commission states that the amount of E. coli that can be found in certain meat products (such as minced beef) cannot be more than 500 CFU/g . The main pathogenic strain of E. coli is EHEC- O157:H7, which can cause infectious diarrhoea and haemorrhagic enteritis. It mainly causes human infection through contaminated food, including fresh meat, fruits, vegetables, raw milk, and dairy products. E. coli O157:H7 has a strong pathogenic capacity and is resistance to gastric acid, and it is destructive to cells. E. coli , similar to most foodborne microbes, can form biofilms by adhering to a range of food-contact surfaces. Biofilms are more resistant to environmental stresses, such as UV light exposure, sanitising agents, nutritional and oxidative stresses, and desiccation. Consequently, biofilms are important for both public health and the economy because they cause chronic illnesses that are challenging to cure and are resistant to cleaning and sanitation .
The majority of bacteria in nature are found in biofilms. Biofilm properties provide protection against environmental pressures and enhance resistance to antimicrobial agents, contributing to microbial persistence and toxicity. Antimicrobial resistance is significantly increased when bacteria form biofilms, so the formation of bacterial biofilms is a significant cause of many persistent and chronic infectious diseases. In addition, infections associated with bacterial biofilms are challenging to treat and are highly resistant to both host immune systems and antibiotics, which poses a substantial challenge for the treatment of biofilm-associated infections. Food poisoning is a general term for illnesses caused by the consumption of foods typically contaminated with bacteria, viruses, toxins, or parasites. Food is rich in nutrition and is thus suitable for the growth and reproduction of pathogens; therefore, microbial biofilms easily form on food. Microbial biofilms may contain a significant number of decaying and pathogenic microorganisms, so the presence of biofilms on surfaces that come into contact with food is generally considered harmful to human health. The economic losses incurred during epidemics of foodborne pathogens mean that the formation and presence of bacterial biofilms can have a significant influence on businesses that process food, so impeding their capacity to survive under these circumstances is a particularly appealing goal for both food sector workers and researchers. Although research on bacterial biofilms is advancing at present, the bacterial biofilm formation mechanisms still need further study. Understanding the specifics of biofilm formation and how signalling pathways are regulated by biofilm formation can help us identify novel targets for the development of highly effective small peptide or protein inhibitors that have strong antibiofilm properties. Additionally, as we continue to explore the mechanisms underlying the biofilm life cycle in the future, more powerful antibiofilm drugs may be discovered, and synthetic derivatives with structural modifications may be designed to create more powerful inhibitors or alter the way we apply them to achieve more effective and rapid suppression or elimination of harmful biological membranes. Today, traditional control strategies, including mechanical and manual cleaning, chemical cleaning, aseptic processing, periodic disinfection cleaning, final sterilisation of equipment, and heat treatment, are still in use and are being further developed. To satisfy the requirements for food safety set out by the food processing industry, more effective and ecologically friendly control strategies must be developed owing to the rising resistance of biofilms to traditional disinfection procedures. In recent years, it has become increasingly popular to develop or investigate emerging strategies for controlling or eliminating biofilms, which include the use of enzyme treatments, phage treatments, pulsed ultraviolet light treatments, steam heat treatment technologies, cold plasma technologies, electron beam irradiation technologies, irradiation at 405 nm, or surface treatment with ozone, ultrasound, or gaseous chlorine dioxide. Different treatments for biofilms at different periods will make the removal of biofilms more efficient, so these emerging biofilm control strategies can provide a new, diverse, and targeted solution for food safety. In addition, biofilm inhibition and QS by natural biological agents will also help to address biofilm issues. Studies have already demonstrated that the extracellular polymers produced by bacteria can sustain the high osmotic pressure inside biofilms, improving the biofilm’s capacity to absorb nutrients from the environment and fuel biofilm proliferation. Therefore, future studies could focus on methods to regulate the osmotic pressure of biofilms to eliminate or control the formation of hazardous biofilms while simultaneously promoting the formation of beneficial biofilms. In addition, combined technologies, which integrate two or more different control technologies, are promising new approaches to eliminate or control the formation of harmful biofilms. For instance, a combination of chemical agents and UV irradiation can effectively remove Pseudomonas aeruginosa biofilms. Combined technologies not only provide synergistic effects but also reduce material and energy consumption. Therefore, the development of combined technologies to eliminate or control the formation of harmful biofilms in the future may also become an attractive research focus. Future research into biofilm control will require a multidisciplinary approach, and although there may be many difficulties, these will be overcome as the research progresses. We hope that this summary will serve as a reference and provide effective strategies for the prevention, suppression, and even eradication of biofilms.
|
Analogues of Anticancer Natural Products: Chiral Aspects | 57469c9d-fee6-4c91-a076-08ebcfac6e2e | 10058835 | Pharmacology[mh] | Cancer is one of the leading causes of death globally, second only behind ischemic heart disease. Lung, prostate, colorectal, stomach, and liver cancer are the most common types of cancer in men, while breast, colorectal, lung, cervical, and thyroid cancer are the most common among women . The primary treatment modalities encompass surgery, chemotherapy, radiation, immunotherapy, etc. However, the mainstay treatment is based on chemotherapy which employs various compounds of natural and synthetic origin that can kill cancer cells or stop their unwanted proliferation . The compounds used in the chemotherapy of cancer disease are quite varied in structure and mechanism of action, comprising alkylating agents; antimetabolite analogues of folic acid, pyrimidine, and purine; natural products; hormones and hormone antagonists; and a variety of agents directed at specific molecular targets. The majority of anticancer agents interact with DNA or its precursors, inhibiting the synthesis of new genetic material and causing damage to DNA in both normal and malignant cells . The rapidly expanding knowledge of cancer biology has brought about the discovery of entirely new and more cancer specific targets (e.g., growth factor receptors, intracellular signalling pathways, epigenetic processes, tumour vascularity, DNA repair effects, and cell death pathways . Throughout history, natural products have been the basis of therapy for a variety of diseases. More than half of the currently used drugs are based on natural compounds . Besides molecules isolated directly from natural sources and applied as such in therapy, this category encompasses also compounds derived from them using chemical methodology, as well as fully synthetic compounds which employ the natural compounds as structural models for the preparation of more efficient analogues. These derivatisation and modification strategies provide drugs with improved pharmacological activities and facilitate the overcoming of inherent drawbacks associated with many drug-like compounds isolated from natural sources, such as poor aqueous solubility and marked adverse reactions . Thus, pharmacomodulation employs a natural compound with known biological activity as the origin, lead, prototype, or series head. Combinatorial chemistry in combination with high-throughput synthesis provides large libraries of bioactive compounds, making it possible to identify new lead molecules . Manipulation of biosynthetic pathways constitutes another powerful tool for the preparation of derivatives of bioactive compounds . Biosynthesis often affords structures not accessible through chemical synthesis due to their high structural and stereochemical complexity. Semisynthetic derivatives of natural products also play an important role in the development of prodrugs . Today, the fraction of anticancer drugs related in one way or another to natural sources amounts to over 60% . Although in the 1990s they temporarily fell out of favour with commercial pharmaceutical research due to the emergence of targeted therapies, recently we have encountered a revived interest in this category of bioactive compounds. According to a study of new and approved drugs for cancer by the United States Food and Drug Administration (FDA), from 1940s–2010, of the 175 small molecules, 74.8% were other than synthetic . Historically, (terrestrial) plants constitute the first major source of natural products. The main categories of anticancer natural compounds of herbal origin (including their semisynthetic derivatives) comprise taxanes, vinca alkaloids, camptothecins, and podophyllotoxins. Representative members of these drug classes are shown in . Taxanes are among the most important chemotherapeutic agents in clinical use . They belong to the group of microtubulin-stabilising agents. The parent compound of the taxane class is paclitaxel, isolated from the bark of the Pacific yew tree ( Taxus brevifolia ). Docetaxel is a semisynthetic analogue with improved anticancer activity along with better pharmacokinetic properties. A number of structural analogues have been developed with a view to overcome the limitations of paclitaxel and docetaxel . Podophyllotoxin is a lignin isolated from the roots and rhizomes of Podophyllum species. It has a long history of use in traditional medicine for various indications. Podophyllotoxin inhibits the polymerisation of tubulin, destabilising microtubules and preventing cell division. Its use in antineoplastic therapy is impaired especially by low bioavailability and high toxicity. Thus, several semisynthetic analogues have been developed, the most successful of them being etoposide, teniposide, and etopophos . They are irreversible inhibitors of topoisomerase II, inducing DNA cleavage. Vinca alkaloids are microtubulin-disrupting agents, originally isolated from the periwinkle plant Catharanthus roseus . Most important among them are vinblastine and vincristine, efficacious anticancer drugs in clinical practice . Numerous semisynthetic analogues have been developed, including vinorelbine, vindesine, vincamine, and vinflunine . These particular compounds show marked differences in their spectrum of activity as well as toxicity profiles . Consequently, they have differing clinical application areas in antineoplastic therapy. Camptothecin is an alkaloid found in the bark of the Chinese tree Camptotheca acuminata . Its anticancer activity relates to the inhibition of topoisomerase I via the formation of a ternary complex between the enzyme, DNA, and camptothecin, preventing DNA relegation . Despite its marked antineoplastic effect, its clinical utility is limited due to severe adverse reactions, as well as unsatisfactory solubility and bioavailability . In order to improve its pharmacological profile and to reduce the side effects, many semisynthetic analogues have been prepared and evaluated . Thus, irinotecan and topotecan have found their way into clinical practice; other examples include belotecan, silatecan, cositecan, exatecan, lurtotecan, and rubitecan. Microbial-based compounds are among the oldest and most important chemotherapeutic agents in use, including bleomycin, actinomycin, ansamycin, anthracyclines, epothilones, and enediynes, among others . Examples of their structures are given in . A substantial majority of antitumour antibiotics originate from various Streptomyces species. A wide variety of structural analogues with improved pharmacological profiles can be obtained by a combination of genetic engineering techniques and methods of organic synthesis . A modern and useful technique for the identification of new microbial secondary metabolites is genome mining . Recently there has been increasing interest in the discovery of new cytotoxic compounds from unconventional sources, such as plant-associated microorganisms or marine habitats. Despite the enormous biodiversity of marine organisms, only a small fraction of marine habitats has been pharmacologically explored. Recent advancements in isolation and purification techniques, structure elucidation, synthetic modification, and biological assays rendered possible the isolation and pharmacological evaluation of numerous unique anticancer compounds from ocean habitats . In this regard, a major source of anticancer compounds are various marine sponges. In addition, diverse organisms such as molluscs, tunicates, algae, marine microbes, and various chordates can be sources of bioactive agents of astounding structural diversity . shows the structures of selected marine-derived molecules with antineoplastic activity. Cytarabine is a synthetic drug modelled after the natural compound found in the Caribbean sponge Tectitethya crypta . Trabectedin is an antitumour chemotherapeutic drug discovered in the extract from the sea squirt Ecteinascidia turbinata . Eribulin is a synthetic analogue of the marine natural product halichondrin B (found in the sponge Halichondria okadai ), both compounds being potent mitotic inhibitors. An interesting anticancer agent is brentuximab vedotin, a semisynthetic bioconjugate prepared from the chimeric monoclonal antibody brentuximab and monomethyl auristatin E, a synthetic analogue derived from dolastatins, natural peptides occurring in the marine mollusc Dolabella auricularia . Stereoisomers are compounds which differ only in the three-dimensional arrangement of their constituent atoms in space. Such isomers may be divided into two groups—enantiomers and diastereoisomers. Enantiomers are pairs of compounds which are non-superimposable mirror images of each other and, in terms of physicochemical properties, differ only in their ability to rotate plane polarised light. Such isomers are called chiral and are referred to as optical isomers. Diastereoisomers are stereoisomers which do not appear as mirror images of each other. They can be chiral or geometrical ( cis / trans ) isomers. A mixture of equal quantities of two enantiomers is called racemate or racemic mixture. Chirality is a property inherent to all biological systems. Biomacromolecules composed of simpler chiral subunits (amino acids, sugars, lipids) fold into complex three-dimensional architectures, exhibiting supramolecular chirality. Chiral macromolecular scaffolds contain asymmetric binding sites and catalytic centres capable of recognising and discriminating between individual stereoisomers of other chiral molecules . Most often, chirality arises from the presence of asymmetric centres in organic molecules; generally, these are tetracoordinate centres to which four different atoms or group are connected, such as in the chiral chemotherapy agent melphalan ( A). Much less frequently, chirality is caused by atropoisomerism, i.e., the hindered rotation about a single bond, e.g., ortho-substituted biphenyl derivatives are chiral due to restricted rotation; an interesting example is gossypol, a bioactive yellow pigment of natural origin ( B). Many biochemical processes during drug action require interaction with chiral biomolecules, hence it is not surprising that enzyme and receptor systems frequently exhibit a stereochemical preference towards one of a pair of enantiomers. Enantiomers may differ both quantitatively and qualitatively in their biological activities. The Easson Stedman hypothesis is generally used to explain the difference in the biological activity of enantiomers. It asserts that the difference in activity is caused by differential binding of the pair of enantiomers to the common binding site . The more active enantiomers must be involved in a minimum of three intermolecular interactions with the receptor surface; the less potent enantiomer only interacts with two sites . In some cases, especially in the interaction of drugs with enzymes, it is asserted that a fourth location, either a direction requirement or an additional binding site is essential to discriminate between the enantiomers . The differential pharmacodynamic and toxicological properties of the enantiomers of chiral drugs have been known for a number of years, affecting essentially all categories of drugs . At one extreme, one enantiomer may be devoid of any biological activity; at the other extreme, both enantiomers may exhibit qualitatively different biological activities. Furthermore, the required activity may reside in both enantiomers, but the adverse effects can be predominantly associated with only one enantiomer, or the enantiomers may have opposite effects on the same biological target . These stereoselective differences may arise not only from drug interactions at the pharmacological receptors but also from pharmacokinetic events . Differences between enantiomers may occur during their absorption, distribution, metabolism, and excretion. Thus, following the administration of a racemic drug, the individual enantiomers do not reach their site of action in equal concentrations . As a result of advances in chemical techniques, especially in the methodology of stereoselective syntheses and stereospecific analyses, together with regulatory requirements, the number of chiral drugs submitted for approval to regulatory authorities as single enantiomers rather than racemates has increased considerably . Compared to the end of the last century, when about 55% of clinically used drugs were chiral and half of them were used as racemates, the current trend in the development of new drugs is mainly towards substances containing one single enantiomeric form . In addition to new chemical entities, a number of established racemic drugs have been re-evaluated as potential single enantiomer products with the possibility of an improved therapeutic profile or application in other therapeutic indications. As an example of this “chiral switching“ concept, the originally racemic nonsteroidal anti-inflammatory drugs (e.g., ibuprofen, ketoprofen) were marketed as single ( S )-enantiomers, since this enantiomer is mainly responsible for the antiphlogistic effect . Several studies demonstrated the association of these anti-inflammatory drugs with decreased cancer incidence and recurrence . Interestingly, the ( R )-enantiomer of flurbiprofen has been found effective against colon and prostate cancer as well as against formation of glioblastomas in in vitro and in vivo models . Investigating the influence of varying configuration at the chiral centre on biological effects is rapidly becoming a significant part in the discovery of novel chemotherapeutic agents . Chirality is one of the important factors that determine the structure of a particular anticancer drug and its interaction with cancer molecular targets. This is true not only for purely organic drugs but also for metal-based drugs developed for anticancer applications . The chiral metal-based anticancer drugs have been comprehensively reviewed in the literature and the influence of chirality on the antineoplastic effect of synthetic organic compounds was surveyed by Valentova et al. . The antitumour activity of natural and synthetic chiral flavonoids is the subject of a literature review by Pinto et al. . This review deals with recent studies on diverse chiral antineoplastic agents exhibiting different mechanisms in their anticancer effects. We focus on selected examples of chiral natural anticancer compounds and their analogues to provide the reader with an overview of the most important developments. Only studies which report the anticancer activities of both enantiomers, or a comparison between a single enantiomer and the racemate are considered, thus allowing the assessment of the influence of stereochemical arrangement on antitumour activity. 4.1. Chiral Xanthones Many naturally occurring xanthones, those isolated from plants as well as marine sources, are chiral and exhibit interesting biological activities . Chemically, xanthones are compounds with an oxygen-containing dibenzo-pyrone heterocyclic scaffold—9H-xanthen-9-ones ( 1 ) . Within this class of compounds and their synthetic derivatives, the main biological activities reported have been antitumour and antimicrobial activities . The renowned Garcinia plants widespread in tropical zones of Asia, Australia, and America are a major source of natural polyprenylated xanthones and benzophenones with significant antitumour activity . Different mechanisms can play a role in the cytotoxic effects of xanthones, including the induction of apoptosis, cell proliferation arrest and autophagy, and inhibition of telomerase. They also demonstrated antimetastasis, anti-angiogenesis and anti-inflammatory activities . Several Garcinia species are important in local medicine, and some are cultivated for their fruit or as ornamentals . Recently, three pairs of newly discovered polyisoprenylated xanthone enantiomers, (±) paucinervins L, M, N ( 2 – 7 ) and two new xanthones, (−) paucinervin O and paucinervin P, along with thirteen known xanthones were isolated from the stem of Garcinia paucinervi . All isolated xanthones were evaluated for anticancer activity against the myeloid–promyelocytic cell line HL-60, the human prostate cancer cell line PC-3, and the colon adenocarcinoma cell line Caco-2 . Enantiomeric pairs of paucinervins L-N exhibited the strongest antiproliferative effects against the HL-60 cell line (IC 50 in the range 0.8–8 μM). Interestingly, the xanthones with dextro-rotation showed more a potent effect that those with laevo-rotation (−). In the case of paucinervin M, the (−)-enantiomer was ten times more cytotoxic than the -enantiomer. Differences in antitumour activity of the chiral synthetic derivatives of xanthones were reviewed by Fernandez et al. . The synthetic analogues of xanthone-4-acetic acid, one of the most studied xanthones with regard to its pharmacological activities, are worth mentioning . The dimethyl analogue of xanthone-4-acetic acid is a tumour vascular-disrupting agent leading to vascular collapse and tumour necrosis by immunomodulation and the action of cytokines. . Its chiral analogues ( 8 ) exhibited enantioselectivity in their antitumour activity—the ability to cause early haemorrhagic necrosis of colon tumours in mice. The ( S )- enantiomer of 5-methyl-α-xanthone-4-acetic was much more dose-potent than the ( R )-(−) enantiomer in both in vitro and in vivo tumour assays. This suggests that the enantiomers have different intrinsic activities rather than differing in their in vivo metabolism . 4.2. Chiral Baicalin The different anti-neoplastic activities in different cell lines were also demonstrated with the chiral derivatives of baicalin, a flavonoid extracted from Scullaria baicalensis Georg, and used as a potential antitumour active ingredient in Chinese traditional medicine . Chiral derivatives of baicalin were prepared by combining baicalin with either D- or L-phenylalanine methyl ester . Antitumour activities of chiral derivatives of baicalin—BAL ( 9 ) (derived from L-phenylalanine methyl ester) and BAD ( 10 ) (derived from D-phenylalanine methyl ester) were investigated against lung (A549, H460, Calu-1) and breast cancer cell lines (MBA-M-435, MCF-7, T47D) in in vitro and in vivo studies. The prepared derivatives had a stronger inhibitory effect on lung cancer cell lines, especially on the A549 cell line, compared with pure baicalin. The antiproliferative activity of BAL was more remarkable than that of BAD. The inhibition rates of 50 mg/mL BA, BAD, and BAL on A549 cells at 48 h were 31.1%, 88.9%, and 94.1%, respectively. In breast cancer lines, both BAL and BAD exhibited stronger inhibitory activity in T47D cells compared with baicalin. In contrast, BAL and BAD did not inhibit the proliferation of MDA-M-435 cells and exhibited inhibition in MCF-7 cells only at high concentrations. BAL and BAD had a good inhibitory effect on subcutaneous tumour growth in nude mice in in vivo experiments, and the effect was shown to follow the order BAL > BAD > baicalin, which was consistent with the results in vitro. The higher antitumour activity of BAL compared with BAD and baicalin was related to the promotion of apoptosis of tumour cells via the phosphatidylinositol 3-kinase signalling pathway . 4.3. Chiral Derivatives of Ricinoleic Acid ( R )-(Z)-ricinoleic acid (RA) ( 11 – 12 ) is a natural fatty acid and is the main component of castor oil from Ricinus communis L., seeds. Many synthetic derivatives of RA with interesting biological activities have been obtained . In particular, amides, esters, and glycosides exhibited potent antiproliferative and cytotoxic activities . The modification of the parent compounds by amines resulted in increased cytotoxicity of the obtained products against HT29, HCT116, MCF-7, and AGS cancer cells (human colorectal adenocarcinoma cell line, human colorectal carcinoma cell line, human breast adenocarcinoma cell line, and human gastric adenocarcinoma cell line, respectively). The antitumour effect was observed for both enantiomeric forms. The most promising cytotoxic effects in terms of anticancer potential were obtained for ethanolamine-derived amides ( 13 – 14 ) . Blaszczyk and co-workers reported the synthesis and cytotoxic activity of both ( R )- and ( S )-enantiomers of ricinoleic acid amides and their acetates. The ricinoleic acid amides as well as acetate derivatives of ethanolamine amides were studied ( 15 – 22 ) to demonstrate the influence of the stereogenic centre on their potential anticancer activity. The cytotoxic effect of the prepared compounds was evaluated against several cancer cell lines (HT29, HTC116, AGS, MFC7). Subsequently, the mechanism of cytotoxicity by the prepared enantiomers of RA-amide derivatives was evaluated using HT29 cancer cells. The ability to induce oxidative stress, DNA damage, and apoptosis was tested. Prepared compounds caused DNA damage and induced apoptotic and necrotic cell death. In most cases, only slight differences between the activities of the two enantiomers were observed. In the case of ( R )- and ( S )-enantiomers of one of the tested acetates ( 21 , 22 ), a significant difference in the ability to induce DNA damage was observed, which showed the impact of the stereogenic centre on the activities of these compounds . 4.4. Chiral Anthramycin Derivatives Derivatives of anthramycin belong to antibiotics produced by various actinomycetes. Their selective cytotoxic activity towards tumour cells makes them a possible source of anticancer agents . Mieczkowski et al. reported the synthesis of novel chiral anthramycin analogues possessing a fused piperazine ring instead of a pyrrole and evaluated their cytotoxic activity in several cancer cell lines . Some of them were prepared as enantiomerically pure ( S )- and ( R )- isomers and were tested as single enantiomers for their antiproliferative potential on human biphenotypic B myelomonocytic leukaemia (MV-4-11) and human urinary bladder (TCC-UM-IC-3) cell lines. Cisplatin was used as a positive control (IC 50 was 0,4 and 4,8, respectively). Most of the tested compounds showed similar cytotoxic effects in both cell lines (IC 50 in the range of 10–44 μM). A significant difference between enantiomers was observed only in the case of ( S ) and ( R ) isomers of the derivative with a biphenyl substituent ( 23 – 25 ). The ( S )-configuration at the chiral centre and the presence of a hydrophobic 4-biphenyl substituent were determined as key structural features responsible for the cytotoxic effect. Cell cycle arrest at the G1/S checkpoint and apoptosis associated with production of reactive oxygen species were also encountered in the most effective compounds . 4.5. Derivatives of Tetrahydroquinolin-8-Amines Substituted tetrahydroquinolins are important structures present in a wide variety of natural alkaloids and synthetic analogues with high biological activity as potent antitumour agents . Amino-quinoline derivatives have been reported to have antiproliferative activity due to their ability to induce mitochondrial dysfunction by increasing ROS levels in the sensitive cervical epithelioid carcinoma cell line HeLaS3 and in the multi-drug resistant human cervical cancer KB-vin cell line . Based on these findings, Facchety et al. investigated a new series of chiral derivatives of 2-methyl-5,6,7,8-tetrahydroquinolin-8-amine for their cytotoxic activity against a panel of human cancer cell lines: human T-lymphocyte (CEM), cervix carcinoma (HeLa), and dermal microvascular endothelial (HMEC-1) cells, as well as colorectal adenocarcinoma (HT-29), ovarian carcinoma (A2780), and biphasic mesothelioma (MSTO-211H) cells. The influence of spatial arrangement of compounds on their biological effect is sometimes difficult to predict. In the case of chiral tetrahydroquinolin derivatives, the cytotoxic effect of enantiomers manifested differently depending on the type of cancer cells. In order to evaluate the different interaction of each enantiomer with their biological targets, the active compounds in the series were synthesised in an enantiomerically pure form—metylphenol derivative ( 26 ); pyridine derivative ( 27 ), and imidazole derivative ( 28 ) . Both enantiomers of prepared compounds were evaluated for their in vitro antiproliferative activity in three human tumour cell lines (HT-29, A2780, and MSTO-211H). All enantiomers showed a marked antiproliferative activity in A2780 cells (IC 50 5.4–17.2 μM). Remarkable differences between biological activities of the enantiomers were found in imidazole derivatives. The most effective was ( R )- 28 and the least active was ( S )- 28 . Conversely, the two chiral forms of metylphenol and pyridine derivatives did not show any difference in terms of IC 50 , suggesting a similar cytotoxic effect. This behaviour was also confirmed in MSTO-211H cells and indeed, a comparable cytotoxic effect was observed in cells incubated with ( R )- 27 and (S)- 27 , while both ( S )- 26 and ( R )- 26 were inactive in this cell line. On the other hand, the ( R )- 28 enantiomer, unlike ( S )- 28 , which appears ineffective, induced an appreciable inhibition of cell growth. Regarding colorectal adenocarcinoma cells (HT-29), they appeared resistant towards all synthesized compounds (IC 50 > 20 μM). For the most active pyridine derivative, ( R )- 27 , the mechanism of the cytotoxic effect was investigated. The compound was able to affect cell cycle phases and to induce mitochondrial membrane depolarisation and cellular ROS production in A2780 cells . Many naturally occurring xanthones, those isolated from plants as well as marine sources, are chiral and exhibit interesting biological activities . Chemically, xanthones are compounds with an oxygen-containing dibenzo-pyrone heterocyclic scaffold—9H-xanthen-9-ones ( 1 ) . Within this class of compounds and their synthetic derivatives, the main biological activities reported have been antitumour and antimicrobial activities . The renowned Garcinia plants widespread in tropical zones of Asia, Australia, and America are a major source of natural polyprenylated xanthones and benzophenones with significant antitumour activity . Different mechanisms can play a role in the cytotoxic effects of xanthones, including the induction of apoptosis, cell proliferation arrest and autophagy, and inhibition of telomerase. They also demonstrated antimetastasis, anti-angiogenesis and anti-inflammatory activities . Several Garcinia species are important in local medicine, and some are cultivated for their fruit or as ornamentals . Recently, three pairs of newly discovered polyisoprenylated xanthone enantiomers, (±) paucinervins L, M, N ( 2 – 7 ) and two new xanthones, (−) paucinervin O and paucinervin P, along with thirteen known xanthones were isolated from the stem of Garcinia paucinervi . All isolated xanthones were evaluated for anticancer activity against the myeloid–promyelocytic cell line HL-60, the human prostate cancer cell line PC-3, and the colon adenocarcinoma cell line Caco-2 . Enantiomeric pairs of paucinervins L-N exhibited the strongest antiproliferative effects against the HL-60 cell line (IC 50 in the range 0.8–8 μM). Interestingly, the xanthones with dextro-rotation showed more a potent effect that those with laevo-rotation (−). In the case of paucinervin M, the (−)-enantiomer was ten times more cytotoxic than the -enantiomer. Differences in antitumour activity of the chiral synthetic derivatives of xanthones were reviewed by Fernandez et al. . The synthetic analogues of xanthone-4-acetic acid, one of the most studied xanthones with regard to its pharmacological activities, are worth mentioning . The dimethyl analogue of xanthone-4-acetic acid is a tumour vascular-disrupting agent leading to vascular collapse and tumour necrosis by immunomodulation and the action of cytokines. . Its chiral analogues ( 8 ) exhibited enantioselectivity in their antitumour activity—the ability to cause early haemorrhagic necrosis of colon tumours in mice. The ( S )- enantiomer of 5-methyl-α-xanthone-4-acetic was much more dose-potent than the ( R )-(−) enantiomer in both in vitro and in vivo tumour assays. This suggests that the enantiomers have different intrinsic activities rather than differing in their in vivo metabolism . The different anti-neoplastic activities in different cell lines were also demonstrated with the chiral derivatives of baicalin, a flavonoid extracted from Scullaria baicalensis Georg, and used as a potential antitumour active ingredient in Chinese traditional medicine . Chiral derivatives of baicalin were prepared by combining baicalin with either D- or L-phenylalanine methyl ester . Antitumour activities of chiral derivatives of baicalin—BAL ( 9 ) (derived from L-phenylalanine methyl ester) and BAD ( 10 ) (derived from D-phenylalanine methyl ester) were investigated against lung (A549, H460, Calu-1) and breast cancer cell lines (MBA-M-435, MCF-7, T47D) in in vitro and in vivo studies. The prepared derivatives had a stronger inhibitory effect on lung cancer cell lines, especially on the A549 cell line, compared with pure baicalin. The antiproliferative activity of BAL was more remarkable than that of BAD. The inhibition rates of 50 mg/mL BA, BAD, and BAL on A549 cells at 48 h were 31.1%, 88.9%, and 94.1%, respectively. In breast cancer lines, both BAL and BAD exhibited stronger inhibitory activity in T47D cells compared with baicalin. In contrast, BAL and BAD did not inhibit the proliferation of MDA-M-435 cells and exhibited inhibition in MCF-7 cells only at high concentrations. BAL and BAD had a good inhibitory effect on subcutaneous tumour growth in nude mice in in vivo experiments, and the effect was shown to follow the order BAL > BAD > baicalin, which was consistent with the results in vitro. The higher antitumour activity of BAL compared with BAD and baicalin was related to the promotion of apoptosis of tumour cells via the phosphatidylinositol 3-kinase signalling pathway . ( R )-(Z)-ricinoleic acid (RA) ( 11 – 12 ) is a natural fatty acid and is the main component of castor oil from Ricinus communis L., seeds. Many synthetic derivatives of RA with interesting biological activities have been obtained . In particular, amides, esters, and glycosides exhibited potent antiproliferative and cytotoxic activities . The modification of the parent compounds by amines resulted in increased cytotoxicity of the obtained products against HT29, HCT116, MCF-7, and AGS cancer cells (human colorectal adenocarcinoma cell line, human colorectal carcinoma cell line, human breast adenocarcinoma cell line, and human gastric adenocarcinoma cell line, respectively). The antitumour effect was observed for both enantiomeric forms. The most promising cytotoxic effects in terms of anticancer potential were obtained for ethanolamine-derived amides ( 13 – 14 ) . Blaszczyk and co-workers reported the synthesis and cytotoxic activity of both ( R )- and ( S )-enantiomers of ricinoleic acid amides and their acetates. The ricinoleic acid amides as well as acetate derivatives of ethanolamine amides were studied ( 15 – 22 ) to demonstrate the influence of the stereogenic centre on their potential anticancer activity. The cytotoxic effect of the prepared compounds was evaluated against several cancer cell lines (HT29, HTC116, AGS, MFC7). Subsequently, the mechanism of cytotoxicity by the prepared enantiomers of RA-amide derivatives was evaluated using HT29 cancer cells. The ability to induce oxidative stress, DNA damage, and apoptosis was tested. Prepared compounds caused DNA damage and induced apoptotic and necrotic cell death. In most cases, only slight differences between the activities of the two enantiomers were observed. In the case of ( R )- and ( S )-enantiomers of one of the tested acetates ( 21 , 22 ), a significant difference in the ability to induce DNA damage was observed, which showed the impact of the stereogenic centre on the activities of these compounds . Derivatives of anthramycin belong to antibiotics produced by various actinomycetes. Their selective cytotoxic activity towards tumour cells makes them a possible source of anticancer agents . Mieczkowski et al. reported the synthesis of novel chiral anthramycin analogues possessing a fused piperazine ring instead of a pyrrole and evaluated their cytotoxic activity in several cancer cell lines . Some of them were prepared as enantiomerically pure ( S )- and ( R )- isomers and were tested as single enantiomers for their antiproliferative potential on human biphenotypic B myelomonocytic leukaemia (MV-4-11) and human urinary bladder (TCC-UM-IC-3) cell lines. Cisplatin was used as a positive control (IC 50 was 0,4 and 4,8, respectively). Most of the tested compounds showed similar cytotoxic effects in both cell lines (IC 50 in the range of 10–44 μM). A significant difference between enantiomers was observed only in the case of ( S ) and ( R ) isomers of the derivative with a biphenyl substituent ( 23 – 25 ). The ( S )-configuration at the chiral centre and the presence of a hydrophobic 4-biphenyl substituent were determined as key structural features responsible for the cytotoxic effect. Cell cycle arrest at the G1/S checkpoint and apoptosis associated with production of reactive oxygen species were also encountered in the most effective compounds . Substituted tetrahydroquinolins are important structures present in a wide variety of natural alkaloids and synthetic analogues with high biological activity as potent antitumour agents . Amino-quinoline derivatives have been reported to have antiproliferative activity due to their ability to induce mitochondrial dysfunction by increasing ROS levels in the sensitive cervical epithelioid carcinoma cell line HeLaS3 and in the multi-drug resistant human cervical cancer KB-vin cell line . Based on these findings, Facchety et al. investigated a new series of chiral derivatives of 2-methyl-5,6,7,8-tetrahydroquinolin-8-amine for their cytotoxic activity against a panel of human cancer cell lines: human T-lymphocyte (CEM), cervix carcinoma (HeLa), and dermal microvascular endothelial (HMEC-1) cells, as well as colorectal adenocarcinoma (HT-29), ovarian carcinoma (A2780), and biphasic mesothelioma (MSTO-211H) cells. The influence of spatial arrangement of compounds on their biological effect is sometimes difficult to predict. In the case of chiral tetrahydroquinolin derivatives, the cytotoxic effect of enantiomers manifested differently depending on the type of cancer cells. In order to evaluate the different interaction of each enantiomer with their biological targets, the active compounds in the series were synthesised in an enantiomerically pure form—metylphenol derivative ( 26 ); pyridine derivative ( 27 ), and imidazole derivative ( 28 ) . Both enantiomers of prepared compounds were evaluated for their in vitro antiproliferative activity in three human tumour cell lines (HT-29, A2780, and MSTO-211H). All enantiomers showed a marked antiproliferative activity in A2780 cells (IC 50 5.4–17.2 μM). Remarkable differences between biological activities of the enantiomers were found in imidazole derivatives. The most effective was ( R )- 28 and the least active was ( S )- 28 . Conversely, the two chiral forms of metylphenol and pyridine derivatives did not show any difference in terms of IC 50 , suggesting a similar cytotoxic effect. This behaviour was also confirmed in MSTO-211H cells and indeed, a comparable cytotoxic effect was observed in cells incubated with ( R )- 27 and (S)- 27 , while both ( S )- 26 and ( R )- 26 were inactive in this cell line. On the other hand, the ( R )- 28 enantiomer, unlike ( S )- 28 , which appears ineffective, induced an appreciable inhibition of cell growth. Regarding colorectal adenocarcinoma cells (HT-29), they appeared resistant towards all synthesized compounds (IC 50 > 20 μM). For the most active pyridine derivative, ( R )- 27 , the mechanism of the cytotoxic effect was investigated. The compound was able to affect cell cycle phases and to induce mitochondrial membrane depolarisation and cellular ROS production in A2780 cells . 5.1. Combretastatin A-4 Analogues Microtubules are important components of the cytoskeleton formed by polymerisation of the α- and β-subunit. Microtubules play a role in separating the daughter chromosomes to opposite poles during mitosis. The disruption of microtubules will result in the interruption of mitosis and leads to apoptosis of the cells . Natural compounds such as colchicine (isolated from Colchicum autumnale ) and combretastatin CA-4 ( 29 ) (isolated from the bark of the African tree Combretum caffum ) with strong tubulin inhibitory activity served as templates for preparing more potent synthetic derivatives ( 30 – 33 ) . Zhou and co-workers presented the synthesis and biological evaluation of diverse chiral β-lactam-bridged combretastatin A-4 analogues. In the cytotoxicity studies, the majority of the prepared target compounds displayed moderate to potent anti-proliferative activities against four human cancer cell lines (A2780, Hela, SKOV-3, and MDA-MB-231). The studies of structure–activity relationships revealed that the absolute configurations of the chiral C-4 atoms were critically important for the activity; more specifically, the ( S )-configuration for 3-methylene-substituted series and the same orientation for other analogues. On this basis, trans -configuration of substituents at the 3,4-positions of the β-lactam scaffold benefits the antiproliferative activity. Among all the synthesised compounds, derivatives ( 32 ) and ( 33 ) turned out to be the most potent and were selected for further pharmacological studies. The co-crystal structures of tubulin in complex as determined by X-ray crystallography showed that derivatives ( 32 ) and ( 33 ) bind to the same site as colchicine with a similar binding mode. 5.2. Analogues of 4-Arylisochromenes Weak inhibitory activity against tubulin polymerisation was also found in the 4-arylisochromenes derivatives isolated from the peel of Musa sapine tum. L (banana) . Li et al. prepared more effective chiral 4-arylisochromenes ( 34 ) which are structural analogues of the natural inhibitor (±)-7,8-dihydroxy-3-methylisochroman-4-one. Antiproliferative activity of prepared compounds was manifested against a panel of cancer cells: epithelial carcinoma (KB), ileocecal adenocarcinoma (HCT-8), breast cancer (MDA-MB-231), chronic myelogenous leukaemia (K562), and hepatocellular carcinoma (H22) cells, with IC 50 values ranging from 10 to 25 nM. The racemic form and the ( R )-enantiomer ( 34 ) were the most active against the K562 cell line with an IC 50 value of 10 nM, which was more potent than the combretastatin (CA-4) (IC 50 = 15 nM) used as a positive control, whereas the ( S )-enantiomer ( 35 ) displayed a significant decrease of activity (IC 50 = 460 nM). Chiral isomers of 4-arylisochromenes ( 34 ) also showed potent inhibitory activity against tubulin polymerisation. The ( R )-enantiomer was slightly more potent than racemate, whereas the ( S )-enantiomer displayed a significantly lower activity. The difference in activity for single enantiomers was illustrated by molecular modelling studies with tubulin crystal structures (PDB, 5lyj). The ( R )--isomer exhibited very similar positioning with that of CA-4. The phenolic hydroxyl and 4-methyl groups of the (R)-enantiomer and CA-4 formed hydrogen bonds with Thr179 and Cys241 residues, respectively. The oxygen atom in the isochromene ring interacted with the Asn258 residue by a weak hydrogen bond. On the other side, the binding pose of the ( S )-enantiomer was flipped over 180° compared with that of CA-4, which may explain why both the antitubulin and antiproliferative activity of the ( S )-enantiomer decreased dramatically . 5.3. Taxol Isomers Nowadays, molecular docking is important in the investigation of the interaction between ligands and proteins and is among the most basic strategies for drug discovery. Molecular docking studies of interactions between active ligands and β-tubulin proteins have been utilised in the search for the most active chiral paclitaxel isomers. Paclitaxel (trade name Taxol ® ) is active in breast, ovarian, lung, bladder, prostate, melanoma, oesophageal, and other types of solid tumour cancers . The molecule exerts its anticancer activity by inhibiting mitosis through enhancement of the polymerisation of tubulin and consequent stabilisation of microtubules . Paclitaxel is produced by extraction from the bark of yew trees ( Taxus brevifolia ), which grow very slowly. Therefore, obtaining paclitaxel from natural sources is not sufficient, and this has prompted extensive searches for alternative sources, including semisynthesis, cellular culture production and chemical synthesis. Taxol contains 11 chiral centres which makes it a very difficult target for total synthesis . Ghadari et al. investigated the effect of variations of chiral centres of Taxol ( 35 ) on the binding to β-tubulin by molecular modelling methods. They studied the hypothetical Taxol ligands which have been obtained by changing the configuration of atoms on one of the chiral centres. The binding activities of 12 different diastereoisomers were compared to the activity of the original Taxol structure. In docking studies, the structures with better binding towards the protein were selected for further investigation using molecular dynamic simulation methods. The results showed that the structures with reversed configuration on the 5 and 8 chiral centres ( 36,37 ) have better affinity towards β-tubulin in comparison with Taxol and are thus good candidate compounds for further experimental studies. Derivatives with reversed configurations 1, 3, and 9 have similar affinity towards β-tubulin in comparison with Taxol. This work provides new opportunities for simplifying future preparation of synthetic analogues of Taxol by omitting the chiral centres which are not essential for the anticancer activity. 5.4. Maytansinoids Significant differences in antitumour activity of enantiomeric forms based on interaction with microtubules were seen in chiral maytansinoids, the synthetic derivatives of maytansine. This compound was originally isolated from the African shrub Maytenus ovatus and belongs to the most potent microtubule inhibitors . Some maytansinoid structures have been prepared in order to be linked to monoclonal tumour-specific antibodies . It has been reported that maytansinoids with an L-configuration of the methyl group at the C3 position exhibited 100–400-fold higher antitumour activity than those with a D-configuration . Based on these results, Li et al. determined the high-resolution crystal structure of the tubulin complex with maytansinol and two stereoisomers of C3-ester side-chain derivatives D-DM1-SMe and L-DM1SMe. ( 38 – 40 ) . The study of crystal structures revealed differences at the C3 side chain in D-DM1-SMe and L-DM1-SMe. The carbonyl oxygen atom of the ester moiety and the tail thiomethyl group at the C3 chain of L-DM1-SMe create strong intramolecular interactions with the hydroxyl at position 9 and the benzene ring, respectively, fixing the bioactive conformation and enhancing the binding activity. The C3 side chain of D-DM1-SMe is swung to the opposite direction, thereby losing the ability to create intramolecular interactions. The conformational differences may provide an explanation for how the chirality of the methyl group at the C3 position affects the anticancer activity . Microtubules are important components of the cytoskeleton formed by polymerisation of the α- and β-subunit. Microtubules play a role in separating the daughter chromosomes to opposite poles during mitosis. The disruption of microtubules will result in the interruption of mitosis and leads to apoptosis of the cells . Natural compounds such as colchicine (isolated from Colchicum autumnale ) and combretastatin CA-4 ( 29 ) (isolated from the bark of the African tree Combretum caffum ) with strong tubulin inhibitory activity served as templates for preparing more potent synthetic derivatives ( 30 – 33 ) . Zhou and co-workers presented the synthesis and biological evaluation of diverse chiral β-lactam-bridged combretastatin A-4 analogues. In the cytotoxicity studies, the majority of the prepared target compounds displayed moderate to potent anti-proliferative activities against four human cancer cell lines (A2780, Hela, SKOV-3, and MDA-MB-231). The studies of structure–activity relationships revealed that the absolute configurations of the chiral C-4 atoms were critically important for the activity; more specifically, the ( S )-configuration for 3-methylene-substituted series and the same orientation for other analogues. On this basis, trans -configuration of substituents at the 3,4-positions of the β-lactam scaffold benefits the antiproliferative activity. Among all the synthesised compounds, derivatives ( 32 ) and ( 33 ) turned out to be the most potent and were selected for further pharmacological studies. The co-crystal structures of tubulin in complex as determined by X-ray crystallography showed that derivatives ( 32 ) and ( 33 ) bind to the same site as colchicine with a similar binding mode. Weak inhibitory activity against tubulin polymerisation was also found in the 4-arylisochromenes derivatives isolated from the peel of Musa sapine tum. L (banana) . Li et al. prepared more effective chiral 4-arylisochromenes ( 34 ) which are structural analogues of the natural inhibitor (±)-7,8-dihydroxy-3-methylisochroman-4-one. Antiproliferative activity of prepared compounds was manifested against a panel of cancer cells: epithelial carcinoma (KB), ileocecal adenocarcinoma (HCT-8), breast cancer (MDA-MB-231), chronic myelogenous leukaemia (K562), and hepatocellular carcinoma (H22) cells, with IC 50 values ranging from 10 to 25 nM. The racemic form and the ( R )-enantiomer ( 34 ) were the most active against the K562 cell line with an IC 50 value of 10 nM, which was more potent than the combretastatin (CA-4) (IC 50 = 15 nM) used as a positive control, whereas the ( S )-enantiomer ( 35 ) displayed a significant decrease of activity (IC 50 = 460 nM). Chiral isomers of 4-arylisochromenes ( 34 ) also showed potent inhibitory activity against tubulin polymerisation. The ( R )-enantiomer was slightly more potent than racemate, whereas the ( S )-enantiomer displayed a significantly lower activity. The difference in activity for single enantiomers was illustrated by molecular modelling studies with tubulin crystal structures (PDB, 5lyj). The ( R )--isomer exhibited very similar positioning with that of CA-4. The phenolic hydroxyl and 4-methyl groups of the (R)-enantiomer and CA-4 formed hydrogen bonds with Thr179 and Cys241 residues, respectively. The oxygen atom in the isochromene ring interacted with the Asn258 residue by a weak hydrogen bond. On the other side, the binding pose of the ( S )-enantiomer was flipped over 180° compared with that of CA-4, which may explain why both the antitubulin and antiproliferative activity of the ( S )-enantiomer decreased dramatically . Nowadays, molecular docking is important in the investigation of the interaction between ligands and proteins and is among the most basic strategies for drug discovery. Molecular docking studies of interactions between active ligands and β-tubulin proteins have been utilised in the search for the most active chiral paclitaxel isomers. Paclitaxel (trade name Taxol ® ) is active in breast, ovarian, lung, bladder, prostate, melanoma, oesophageal, and other types of solid tumour cancers . The molecule exerts its anticancer activity by inhibiting mitosis through enhancement of the polymerisation of tubulin and consequent stabilisation of microtubules . Paclitaxel is produced by extraction from the bark of yew trees ( Taxus brevifolia ), which grow very slowly. Therefore, obtaining paclitaxel from natural sources is not sufficient, and this has prompted extensive searches for alternative sources, including semisynthesis, cellular culture production and chemical synthesis. Taxol contains 11 chiral centres which makes it a very difficult target for total synthesis . Ghadari et al. investigated the effect of variations of chiral centres of Taxol ( 35 ) on the binding to β-tubulin by molecular modelling methods. They studied the hypothetical Taxol ligands which have been obtained by changing the configuration of atoms on one of the chiral centres. The binding activities of 12 different diastereoisomers were compared to the activity of the original Taxol structure. In docking studies, the structures with better binding towards the protein were selected for further investigation using molecular dynamic simulation methods. The results showed that the structures with reversed configuration on the 5 and 8 chiral centres ( 36,37 ) have better affinity towards β-tubulin in comparison with Taxol and are thus good candidate compounds for further experimental studies. Derivatives with reversed configurations 1, 3, and 9 have similar affinity towards β-tubulin in comparison with Taxol. This work provides new opportunities for simplifying future preparation of synthetic analogues of Taxol by omitting the chiral centres which are not essential for the anticancer activity. Significant differences in antitumour activity of enantiomeric forms based on interaction with microtubules were seen in chiral maytansinoids, the synthetic derivatives of maytansine. This compound was originally isolated from the African shrub Maytenus ovatus and belongs to the most potent microtubule inhibitors . Some maytansinoid structures have been prepared in order to be linked to monoclonal tumour-specific antibodies . It has been reported that maytansinoids with an L-configuration of the methyl group at the C3 position exhibited 100–400-fold higher antitumour activity than those with a D-configuration . Based on these results, Li et al. determined the high-resolution crystal structure of the tubulin complex with maytansinol and two stereoisomers of C3-ester side-chain derivatives D-DM1-SMe and L-DM1SMe. ( 38 – 40 ) . The study of crystal structures revealed differences at the C3 side chain in D-DM1-SMe and L-DM1-SMe. The carbonyl oxygen atom of the ester moiety and the tail thiomethyl group at the C3 chain of L-DM1-SMe create strong intramolecular interactions with the hydroxyl at position 9 and the benzene ring, respectively, fixing the bioactive conformation and enhancing the binding activity. The C3 side chain of D-DM1-SMe is swung to the opposite direction, thereby losing the ability to create intramolecular interactions. The conformational differences may provide an explanation for how the chirality of the methyl group at the C3 position affects the anticancer activity . The ubiquitin–proteasome pathway is the most important intracellular protein degradation system, and it is involved in processes such as apoptosis, cell survival, cell-cycle progression, DNA repair, and antigen presentation, among others. The proteosomal system includes different kinds of enzymes, which are modified by binding of several regulatory complexes to the core particle (the 20S proteasome) . Inhibitors of the 20S proteasome (targeting the 20S catalytic particle) are an important class of drugs for the treatment of liquid tumours, such as multiple myeloma and mantle cell lymphoma, and they are being investigated for other diseases as well . Bortezomib was the first proteasomal inhibitor to be approved by the US Food and Drug Administration. Carfilzomib and Ixazomib have recently been approved, and more drugs are in development . However, these protease inhibitors have not demonstrated sufficient activity against solid tumours, and peripheral neuropathy is a dose-limiting toxic side effect for their clinical use . Non-peptide inhibitors targeting different components of the proteasome system appear to be a promising alternative for the treatment of solid tumours. Anchoori et al. presented the development of novel derivatives targeting the 19S regulatory particle unit which contains the ubiquitin receptor RPN13, RA 183, and RA375. The preparation of new derivatives was rationalised. To improve their specificity and potency, several libraries of molecules were generated to probe the pharmacophore of the benzylidenepiperidone core unit and to identify the active compounds. Based on these findings and molecular modelling data, they introduced a methyl group at the ring carbon atom next to the nitrogen, and thus prepared chiral derivatives of perspective RPN13 inhibitors ( 41 – 42 ) . The docking studies suggested the potential for differing RPN13 binding and cytotoxicity potencies of the racemic form R414 and the ( S )-isomer R413S, and for weaker toxicity for R413R. Consequently, each form of active compound was synthesised and tested against several ovarian cancer cell lines. The cytotoxicity studies confirmed the theoretical predictions. RA413S was 5-fold more cytotoxic for HeLa cells than RA413R (23 nM vs. 172 nM). Similar phenomena appeared also in additional cell lines derived from ovarian cancer (e.g., SKOV3, TOV21G) and cervical cancer (HeLa, CaSki, SiHa). The cytotoxicity of RA413S for normal human cells was much weaker (IC 50 > 100 nM). The mechanism of antitumour activity for the active ( S )-isomer RA413S and the racemate was further evaluated. The cancer cell toxicity was associated with improved binding to RPN13 lysates, ATP depletion, mitochondrial damage, oxidative stress, and glutathione and NF-κB inhibition . Peroxisome proliferator-activated receptors (PPARs) belong to the group of nuclear receptors. They exist in three different isoforms: PPARα, PPARβ, and PPARγ, and are mainly produced in brown adipose tissue, gut, immune cells, liver, kidney, heart, and other tissues. PPARs play major regulatory roles in energy homeostasis and metabolic function by activation of fatty acid metabolism and stimulation of glucogenesis . The modulatory function of PPARs-α and -γ is evident in immunity inflammation, vascular functions, cellular proliferation, differentiation, development, and apoptosis . PPARs have become interesting therapeutic targets for the treatment of various diseases—dyslipidaemia, type 2 diabetes, cardiovascular diseases, obesity, cancer, and metabolic diseases . PPAR modulators, including agonists and antagonists, could represent a novel strategy for preventing and treating multiple types of cancer . The antitumour effect of PPARα and PPARγ in various types of cancers, both in laboratory and in clinical settings, have been recently reviewed . The synthesised PPARγ modulators thiazolidinediones (TZDs), also known as glitazones, are involved in clinical phase trials for the treatment of prostate cancer, liver cancer, melanoma, and lung cancer. In order to diminish the side effects of TZD treatment, novel PPAR ligands with different molecular scaffolds are being developed . Sabatino et al. prepared and biologically evaluated new chiral derivatives of phenoxyacetic acid, acting as PPARγ partial agonists ( 43 – 49 ) . Their antiproliferative activity was evaluated in colorectal carcinoma cell lines (HT-29 and CRC). All compounds exhibited an antiproliferative effect in the range of 31–82% of residual vitality; with respect to the 60 % produced by the full i PPARγ agonist rosiglitazone. The compounds 45 ( RS )-, ( S )-isomers, and 49 ( RS ) forms were subjected to further evaluation since they combine the best antiproliferative activity (31–47% of residual vitality) and a limited trans activation (efficacy ranging between 55% and 65%) in comparison with the effects of all other compounds. Piemontese et al. designed a new class of dual PPARα/γ agonists based on the 2-oxy-propanoid acid moiety linked to diphenylmethane ( 50 ) . This structural skeleton is an active pharmacophore for the activation of PPARα/γ subtypes. Prepared diphenylmethane derivatives were tested for their agonist activity towards the human PPARα, PPARβ, and PPARγ subtypes. The highest activity was obtained in R1 compounds (47% activation compared to reference compounds). Single enantiomers were prepared to evaluate the influence of configuration on receptor activation. Unexpectedly, both enantiomers of R1 displayed similar activity towards all PPAR subtypes. To rationalise this effect, docking experiments were performed. The docking experiment predicted that both ( S )-1 and ( R )-enantiomers favourably bind to the PPARγ ligand-binding domain, adopting a similar U-shaped configuration that wraps around H3. In case of PPARα, both enantiomers fit the PPARα pocket well. The carboxylate head groups form the well-recognised H-bonding network with residues Y464, Y314, and S280, which is supposed to be critical for PPARα ligands’ activity. The antiproliferative activities of racemate and both enantiomers were evaluated against HT-29 cells, with the ( S )-enantiomer eliciting a more robust activity than the ( R )-enantiomer. The alternative antiproliferative pathways were tested. The ability of the compounds to inhibit cell proliferation in colon cancer lines seems to be due to downregulation of Wnt/β-catenin signalling which is overexpressed in the majority of colorectal cancers. The ( S )- and ( R )-enantiomers strongly influenced mitochondrial function, as they activated the carnitine shuttle system through upregulation of the carnitine/acylcarnitine carrier and carnitine palmitoyl-transferase genes . Chirality can be considered one of the major topics in the design, discovery, development, and marketing of new drugs. Chirality plays an important role for biological activities, so when a chiral centre is present in a drug, both enantiomers must be studied for the evaluation of their pharmacological properties. One enantiomer of a chiral drug may be a medicine for a particular disease, whereas another enantiomer of the same molecule may not only be inactive but can even be toxic. This review outlines a variety of some recent examples of structurally diverse natural anticancer chiral compounds and their analogues exhibiting different mechanisms in their anticancer effect. The present survey represents up-to-date studies of the difference in biological activities between single enantiomers of anticancer agents and their racemic mixtures. The influence of stereoselectivity on anticancer activity is difficult to generalise, as it is manifested specifically for each individual chiral compound as well as in dependence on the type of cellular targets. The stereospecificity of new anticancer agents manifests itself in the cytotoxicity effect at the cellular level or in the interaction with subcellular structures. The awareness of the stereochemistry of anticancer compounds can help to understand some critical processes underlying their toxicity towards cancer cells and can provide a rational basis for the design of new antitumour drugs. |
Pharmacogenetics of the Primary and Metastatic Osteosarcoma: Gene Expression Profile Associated with Outcome | 70f46fe4-fa05-450c-829a-0212f7271d50 | 10059037 | Pharmacology[mh] | Osteosarcoma (OS) is the most common primary malignant bone tumor in children and adolescents . The overall survival probabilities have not improved during the last 30 years. Since then, the treatment has consisted of complete tumor resection after neoadjuvant chemotherapy, followed by adjuvant chemotherapy . According to the EURAMOS-1 study results, the MAP regimen (Methotrexate, Adriamycin—doxorubicin, Platinol—cisplatin) must be considered the standard chemotherapy treatment for high-grade osteosarcoma . Equivalent chemotherapeutic drug doses may lead to wide interpatient variability in treatment response, and it may be due to pharmacokinetic (Absorption, Distribution, Metabolism and Elimination—ADME) and pharmacodynamic (receptors and targets) differences in drugs . The biological mechanisms involved in genetic variability are the differences in gene expression, epigenetics and genetic polymorphism . Pharmacogenetics investigations have been explored in OS to understand the variability in treatment outcomes among patients . Many pharmacogenomic studies have been conducted in OS and are beginning to yield insights into how to modify and improve chemotherapeutic approaches . However, these studies have been focused merely on single nucleotide polymorphisms (SNPs) . Moreover, a major priority in OS management is pulmonary metastasis, as this is the primary cause of death . Therefore, the aim of the present study was to investigate the gene expression profile in a pharmacogenetic context, to the best of our knowledge, for the first time in paired OS specimens. The gene panel was designed based on MAP pharmacokinetic and pharmacodynamic modeling, as well as cell death and DNA damage repair processes that could be related to MAP response and OS tumorigenesis. The present study has investigated 32 genes involved in many processes, such as apoptosis—B-cell lymphoma 2 like 1 ( BCL2L1 ), caspase 3 ( CASP3 ) and Fas ligand ( FASLG ); cell cycle—cyclin-dependent kinase 1 ( CDK1 ); damage recognition—high mobility group box 1 ( HMGB1 ); DNA repair—excision repair cross-complementing 1 ( ERCC1 ), excision repair cross-complementing 2 ( ERCC2 ) and mutS homolog 2 ( MSH2 ); detoxification—glutathione S-transferase ( GSTM1, GSTM3, GSTP1 and GSTT1 ) and superoxide dismutase 1 ( SOD1 ); doxorubicin pathway—DNA topoisomerase II alpha ( TOP2A ); folate pathway—dihydrofolate reductase ( DHFR ) gamma-glutamyl hydrolase ( GGH ) and methylenetetrahydrofolate reductase ( MTHFR ); influx transport—solute carrier family ( SLC19A1 , SLC22A1 and SLC31A1 ); and efflux transport—ATP binding cassette ( ABCB1 , ABCC1 , ABCC2 , ABCC3 , ABCC4 , ABCC5 , ABCC6 , ABCC10 , ABCC11 and ABCG2 ), ATPase copper transporting beta ( ATP7B ) and ralA binding protein 1 ( RALBP1 ). 2.1. Gene Expression Profile in Primary OS, Metastatic OS and Normal Bone The expression of 32 target genes was investigated in OS and normal bone specimens. In some cases, this expression was different when comparing pre-chemotherapy with post-chemotherapy specimens, as well as in the primary and metastatic OS. The present investigation did not detect the expression of the HMGB1 gene, neither in the OS nor in the normal bone specimens. The relative quantification (RQ) of the target genes with statistically significant results in all analyzed specimens: pre-chemotherapy (B), post-chemotherapy (S), metastasis (M) and normal bone (NB) are presented in . The RQ of the target genes with non-statistically significant results is presented in the . Regarding the comparisons between primary OS and normal bone, evaluated by the Mann-Whitney test, higher expression of GSTM3, GGH, ABCC10 and SLC22A1 genes in OS was observed. ( p = 0.037; p = 0.042; p = 0.013; and p = 0.015, respectively). The comparisons between pre- and post-chemotherapy specimens were analyzed using the Wilcoxon test since all these samples were paired. The post-chemotherapy specimens presented higher expression of the BCL2L1 , FASLG , ABCB1 , ABCC2 and ABCG2 genes than the pre-chemotherapy specimens ( p = 0.012; p = 0.037; p = 0.004; p = 0.039; and p = 0.042, respectively). The post-chemotherapy specimens presented lower expression of the CASP3 , CDK1 , MSH2 , GSTM3 , SOD1 , TOP2A , DHFR , GGH , ABCC10 and SLC19A1 genes than the pre-chemotherapy specimens ( p = 0.037; p < 0.001; p = 0.001; p < 0.001; p = 0.004; p < 0.0001; p = 0.025; p < 0.0001; p = 0.009; and p < 0.001, respectively). The gene expression in metastasis specimens was compared with the pre-chemotherapy specimens using the Wilcoxon and Mann-Whitney test (paired and non-paired samples, respectively) because only 14 patients had developed metastasis. The metastasis specimens presented higher expression of ABCC1 , ABCC3 and ABCC4 genes ( p = 0.049; p = 0.057 (trend); and p = 0.039, respectively) and lower expression of ERCC2, MSH2, SOD1, TOP2A, ABCC10 and SLC22A1 genes ( p = 0.043; p = 0.043; p = 0.048; p = 0.005; p = 0.049; and p = 0.017, respectively). 2.2. Gene Expression Profile Associated with Clinical Parameters The clinical parameters were associated with gene expression, evaluated by the Mann-Whitney test . Tumors from patients who were metastatic at diagnosis presented higher expression of ABCB1 , ABCC6 , ABCC10 , BCL2L1 and SLC19A1 genes ( p = 0.039; p = 0.048; p = 0.048; p = 0.026; and p = 0.010, respectively) than tumors from patients who were non-metastatic at diagnosis. Poor responders’ tumors presented higher expression of ERCC1 and TOP2A genes ( p = 0.021 and p = 0.036, respectively) and lower expression of ABCC3 , FASLG and SLC22A1 genes ( p = 0.031; p = 0.017; and p = 0.014, respectively) than good responders’ tumors. The sizes of tumors resected in surgery were also associated with the expression of the investigated genes. Large tumors of 12 cm or more presented lower expression of ABCG2 , CASP3 and MSH2 genes ( p = 0.027; p = 0.033; and p = 0.045, respectively) than small tumors. The local control was conducted with surgery that could be either conservative or an amputation. Tumors from patients who underwent amputation presented higher expression of ABCC11 , DHFR , ERCC1 , GSTM3 , SLC19A1 and TOP2A genes ( p = 0.002; p = 0.007; p = 0.0042; p = 0.022; p = 0.002. and p = 0.010, respectively) and lower expression of FASLG , MTHFR and SLC22A1 genes ( p < 0.0001; p = 0.003; and p = 0.024, respectively) than tumors from patients who underwent conservative surgery. In the relapse analyses, one patient was excluded because he had disease progression during the first treatment and died before he reached remission. Tumors from patients who relapsed presented higher expression of TOP2A ( p = 0.038) and lower expression of ABCC3 , ABCC5 and FASLG genes ( p = 0.026; e p = 0.051; p = 0.050, respectively) than patients who not relapsed. Regarding the ABCC5 result, this was only a trend of statistical significance. 2.3. Gene Expression Profile Associated with OAS and EFS As shown in , patients with high expression of the ABCC5 and BCL2L1 genes in the pre-chemotherapy biopsy had a trend towards worse OAS ( p = 0.051; HR = 3.42) and EFS ( p = 0.058; HR = 3.27), respectively, compared with patients with low expression of the ABCC5 and BCL2L1 genes. Moreover, patients with high expression of the ABCC3 gene in the pre-chemotherapy biopsy had worse EFS compared with patients with low expression of the ABCC3 gene ( p = 0.048; HR = 3.41). Patients with high expression of the TOP2A gene in the post-chemotherapy specimens had worse OAS ( p = 0.015; HR = 5.37) and EFS ( p = 0.005; HR = 6.36) compared with patients with low expression of the TOP2A gene. Furthermore, patients with low expression of the RALBP1A gene in the post-chemotherapy specimens had a trend towards worse OAS compared with patients with high expression of the RALBP1 gene ( p = 0.051; HR = 3.40). Patients with low expression of the BCL2L1 and MTHFR genes in the metastasis had worse OAS ( p = 0.018, HR = 3.53; and p = 0.027, HR = 3.27, respectively) and worse EFS ( p = 0.019, HR = 3.29; and p = 0.024, HR = 3.16, respectively) compared with patients with high expression of the BCL2L1 and MTHFR genes. Moreover, patients with low expression of the ABCC2, RALBP1 and SOD1 genes in the metastasis specimens had worse EFS ( p = 0.048, HR = 3.16; p = 0.022, HR = 3.26; and p = 0.027, HR = 3.14, respectively) compared with patients with high expression of the ABCC2, RALBP1 and SOD1 genes. The expression of 32 target genes was investigated in OS and normal bone specimens. In some cases, this expression was different when comparing pre-chemotherapy with post-chemotherapy specimens, as well as in the primary and metastatic OS. The present investigation did not detect the expression of the HMGB1 gene, neither in the OS nor in the normal bone specimens. The relative quantification (RQ) of the target genes with statistically significant results in all analyzed specimens: pre-chemotherapy (B), post-chemotherapy (S), metastasis (M) and normal bone (NB) are presented in . The RQ of the target genes with non-statistically significant results is presented in the . Regarding the comparisons between primary OS and normal bone, evaluated by the Mann-Whitney test, higher expression of GSTM3, GGH, ABCC10 and SLC22A1 genes in OS was observed. ( p = 0.037; p = 0.042; p = 0.013; and p = 0.015, respectively). The comparisons between pre- and post-chemotherapy specimens were analyzed using the Wilcoxon test since all these samples were paired. The post-chemotherapy specimens presented higher expression of the BCL2L1 , FASLG , ABCB1 , ABCC2 and ABCG2 genes than the pre-chemotherapy specimens ( p = 0.012; p = 0.037; p = 0.004; p = 0.039; and p = 0.042, respectively). The post-chemotherapy specimens presented lower expression of the CASP3 , CDK1 , MSH2 , GSTM3 , SOD1 , TOP2A , DHFR , GGH , ABCC10 and SLC19A1 genes than the pre-chemotherapy specimens ( p = 0.037; p < 0.001; p = 0.001; p < 0.001; p = 0.004; p < 0.0001; p = 0.025; p < 0.0001; p = 0.009; and p < 0.001, respectively). The gene expression in metastasis specimens was compared with the pre-chemotherapy specimens using the Wilcoxon and Mann-Whitney test (paired and non-paired samples, respectively) because only 14 patients had developed metastasis. The metastasis specimens presented higher expression of ABCC1 , ABCC3 and ABCC4 genes ( p = 0.049; p = 0.057 (trend); and p = 0.039, respectively) and lower expression of ERCC2, MSH2, SOD1, TOP2A, ABCC10 and SLC22A1 genes ( p = 0.043; p = 0.043; p = 0.048; p = 0.005; p = 0.049; and p = 0.017, respectively). The clinical parameters were associated with gene expression, evaluated by the Mann-Whitney test . Tumors from patients who were metastatic at diagnosis presented higher expression of ABCB1 , ABCC6 , ABCC10 , BCL2L1 and SLC19A1 genes ( p = 0.039; p = 0.048; p = 0.048; p = 0.026; and p = 0.010, respectively) than tumors from patients who were non-metastatic at diagnosis. Poor responders’ tumors presented higher expression of ERCC1 and TOP2A genes ( p = 0.021 and p = 0.036, respectively) and lower expression of ABCC3 , FASLG and SLC22A1 genes ( p = 0.031; p = 0.017; and p = 0.014, respectively) than good responders’ tumors. The sizes of tumors resected in surgery were also associated with the expression of the investigated genes. Large tumors of 12 cm or more presented lower expression of ABCG2 , CASP3 and MSH2 genes ( p = 0.027; p = 0.033; and p = 0.045, respectively) than small tumors. The local control was conducted with surgery that could be either conservative or an amputation. Tumors from patients who underwent amputation presented higher expression of ABCC11 , DHFR , ERCC1 , GSTM3 , SLC19A1 and TOP2A genes ( p = 0.002; p = 0.007; p = 0.0042; p = 0.022; p = 0.002. and p = 0.010, respectively) and lower expression of FASLG , MTHFR and SLC22A1 genes ( p < 0.0001; p = 0.003; and p = 0.024, respectively) than tumors from patients who underwent conservative surgery. In the relapse analyses, one patient was excluded because he had disease progression during the first treatment and died before he reached remission. Tumors from patients who relapsed presented higher expression of TOP2A ( p = 0.038) and lower expression of ABCC3 , ABCC5 and FASLG genes ( p = 0.026; e p = 0.051; p = 0.050, respectively) than patients who not relapsed. Regarding the ABCC5 result, this was only a trend of statistical significance. As shown in , patients with high expression of the ABCC5 and BCL2L1 genes in the pre-chemotherapy biopsy had a trend towards worse OAS ( p = 0.051; HR = 3.42) and EFS ( p = 0.058; HR = 3.27), respectively, compared with patients with low expression of the ABCC5 and BCL2L1 genes. Moreover, patients with high expression of the ABCC3 gene in the pre-chemotherapy biopsy had worse EFS compared with patients with low expression of the ABCC3 gene ( p = 0.048; HR = 3.41). Patients with high expression of the TOP2A gene in the post-chemotherapy specimens had worse OAS ( p = 0.015; HR = 5.37) and EFS ( p = 0.005; HR = 6.36) compared with patients with low expression of the TOP2A gene. Furthermore, patients with low expression of the RALBP1A gene in the post-chemotherapy specimens had a trend towards worse OAS compared with patients with high expression of the RALBP1 gene ( p = 0.051; HR = 3.40). Patients with low expression of the BCL2L1 and MTHFR genes in the metastasis had worse OAS ( p = 0.018, HR = 3.53; and p = 0.027, HR = 3.27, respectively) and worse EFS ( p = 0.019, HR = 3.29; and p = 0.024, HR = 3.16, respectively) compared with patients with high expression of the BCL2L1 and MTHFR genes. Moreover, patients with low expression of the ABCC2, RALBP1 and SOD1 genes in the metastasis specimens had worse EFS ( p = 0.048, HR = 3.16; p = 0.022, HR = 3.26; and p = 0.027, HR = 3.14, respectively) compared with patients with high expression of the ABCC2, RALBP1 and SOD1 genes. The expression of genes analyzed in the present study was investigated for the first time in paired OS specimens. When comparing OS samples obtained pre-and post-treatment, as well as from metastases, we detected different levels of expression of the selected genes. Moreover, this study showed that the genes related to a treatment response could be associated with OS tumorigenesis. TOP2A is a target for several anticancer agents, such as doxorubicin, and a variety of mutations in this gene have been associated with the development of drug resistance. This nuclear enzyme is involved in processes such as chromosome condensation, chromatid separation and the relief of torsional stress that occurs during DNA transcription and replication . A meta-analysis showed that high TOP2A expression is associated with a worse prognosis in many types of cancer . In OS, the presence of TOP2A amplification tends to relate to a worse overall survival rate . The present study showed an association between high expression of TOP2A and poor response, amputation and relapse. Moreover, high expression was also associated with worse OAS and EFS. DHFR , GGH and MTHFR are genes involved in the methotrexate pathway and response . Methotrexate resistance in human OS cells is associated with an amplification and/or overexpression of its target, the DHFR . We observed that patients who underwent amputation presented metastasis with higher DHFR expression than patients who underwent conservative surgery. Increased levels of GGH led to a decreased accumulation of polyglutamated MTX and MTX resistance . In OS, the ratio between the patients and the controls for the polymorphisms GGH_452T/C, GGH_401T/C and GGH_16T/C was greater than 1.5. The GGH_401C/T variant enhanced promoter activity, increasing protein expression . In the present study, it was observed that OS presented higher GGH expression than in normal bone. The rs1801133 polymorphism of the MTHFR has been the most frequently studied in OS and leads to a C to T substitution, resulting in decreased enzymatic activity. In OS, the TT genotype was significantly associated with toxicity . However, we observed that low expression in OS metastasis was associated with worse OAS and EFS. Moreover, the patients who underwent amputation had lower MTHFR expression in the primary tumor compared with the patients who underwent conservative surgery. Regarding apoptosis, BCL2L1 , CASP3 and FASLG were investigated in the present study. The longer isoform of BCL2L1 acts as an apoptotic inhibitor and the shorter isoform acts as an apoptotic activator . We found that high BCL2L1 expression in the primary tumor was associated with metastasis at diagnosis and a worse EFS. Nevertheless, in the metastatic tumor, low expression was associated with worse OAS and EFS. This discrepancy in our results could be explained by the theory that the primary tumor expresses the longer isoform (anti-apoptosis) and a metastatic tumor expresses the shorter isoform (pro-apoptosis), since chemotherapy drugs stimulate the production of the shorter isoform . The G allele of the variant rs2720376, linked with lower CASP3 expression, was associated with a lower EFS in OS . In this study, low CASP3 expression was associated with large tumors. The variant rs763110, linked to a lower FasL expression, was associated with a lower EFS in OS . We found that low FASLG expression was associated with a poor response, amputation and relapse. Regarding genes related to detoxification of the chemotherapeutic drugs, GSTM1 and GSTP1 presented no association with the outcome in OS. The polymorphism in GSTM3 (AA versus BB) has been associated with OS risk . The present study observed an association between high GSTM3 expression and amputation. Moreover, OS presented higher GSTM3 expression than normal bone. The GSTM3 polymorphism could confer different efficiencies in the metabolism of carcinogens and has been shown to modulate various cancers’ risk . The null GSTT1 genotype was associated with OS risk . Resistant cell lines of OS showed lower SOD1 expression than their parental cells . We found that metastasis presented lower SOD1 expression than the primary tumor, and patients with low expression of the SOD1 gene in the metastasis had worse EFS. Genes involved in the repair of DNA adducts induced by cisplatin, which thereby influence cisplatin efficacy, have been investigated by the largest number of studies on OS . ERCC1 positivity has presented an association with poor EFS and OAS in OS . The present study showed an association between high expression and poor response and amputation. The ERCC2 rs1799793 polymorphism was related to the high risk of OS development . This study showed that metastasis presented lower ERCC2 expression than primary tumors. We found that low MSH2 expression was associated with large primary tumors. Metastasis specimens presented lower expression than primary tumors. In OS, the variant rs4638843 in MSH2 was associated with a worse EFS . Moreover, a wide investigation of childhood cancers found germline mutations of MSH2 in OS . Taken together, our results showed that metastasis in OS presents low expression of the ERCC2 and MSH2 genes compared with pre-chemotherapy biopsy, which could be related to decreased ability to repair DNA damage in metastasis, possibly resulting in genetic alterations accumulation and more aggressive cancer . High efflux transporter gene expression and low influx transporter gene expression are the main resistance mechanisms related to cisplatin, doxorubicin and methotrexate, in vitro. Moreover, many polymorphisms in these genes have been related to treatment response in OS . The present study, for the first time, investigated the expression of transporter genes in paired specimens. The results showed that the tumor biopsy presented high ABCC6 and ABCC10 expression, and metastasis presented high ABCB1 expression when metastasis was present at diagnosis. The patients with high ABCC3 and ABCC5 expression in biopsy presented worse EFS and OAS, respectively, and patients with low ABCC2 expression in metastasis presented worse EFS.The patients with low RALBP1 expression in surgery and metastasis presented worse OAS and EFS, respectively. OS presented higher ABCC10 and SLC22A1 expression than normal bone. However, low SLC22A1 expression was associated with a poor response and amputation, probably due to its influx function. This is the first investigation regarding SLC22A1 and OS. SLC22A1 could be activated by miR-21, which is overexpressed in OS and was associated with tumorigenesis . Moreover, metastasis presented higher ABCC1 , ABCC3 and ABCC4 expression and lower SLC22A1 and ABCC10 expression than the primary tumor. This pattern could contribute to the lower intracellular concentration of the chemotherapeutic drugs. Consequently, it could contribute to the mechanism of resistance in metastasis, which is the main cause of death in OS patients. Therefore, with the knowledge of the metastasis profile, it is possible to develop new strategies for these patients. CBT-1 ® is an adjunct to chemotherapy in all cancer types with multi-drug resistance. Eight clinical trials are evaluating CBT-1 ® in patients with many cancer types, such as acute myelogenous leukemia, breast, non-Hodgkin’s lymphoma, Hodgkin’s disease, lung, and sarcoma . Moreover, CBT-1® was able to revert the ABCB1/ABCC1-mediated resistance against doxorubicin in OS cell lines . In the future, it could be interesting to evaluate CBT-1 ® in metastatic OS patients In conclusion, the present study identified associations between OS outcome and expression of the genes TOP2A , DHFR , MTHFR , BCL2L1 , CASP3 , FASLG , GSTM3 , SOD1 , ABCB1 , ABCC2 , ABCC3 , ABCC5 , ABCC6 , ABCC10 , ABCC11 , ABCG2 , RALBP1 , SLC19A1 , SLC22A1 , ERCC1 and MSH2 . In addition, the pre-chemotherapy biopsy from OS patients had higher gene expression of ABCC10 , GGH , GSTM3 and SLC22A1 compared with bone specimens obtained from healthy subjects, and the metastasis specimens showed a high expression profile of ABCC1 , ABCC3 and ABCC4 and low expression of SLC22A1 and ABCC10 , which is possibly an important factor for resistance in OS metastasis. In summary, we found that the expression of genes related MAP pharmacokinetic and pharmacodynamic modeling, as well as cell death and DNA damage repair processes are associated with OS tumorigenesis and MAP response in OS patients. Therefore, in the future, our findings may contribute to clinical management as prognostic markers and also as possible therapeutic targets. 4.1. Patients and Specimens We investigated 80 paired specimens obtained from 33 patients with diagnoses of OS. These patients were admitted to the Pediatric Oncology Institute (IOP/GRAACC/UNIFESP) between 2006 and 2016. The average age at diagnosis was 13 years old. Of 33 OS patients, 14 patients presented pulmonary metastasis. Thus, we investigated 33 biopsy specimens (pre-chemotherapy), 33 surgery specimens (post-chemotherapy) and 14 pulmonary metastasis specimens. Five normal bone tissues were used as a control; they were obtained from orthopedic surgeries of five healthy individuals that underwent trauma and did not present either genetic disorders or bone diseases. This study had the Research Ethics Committee approval from the Federal University of Sao Paulo (N° 0189/2016), and all patients agreed to participate by informed consent. All patients were treated following the GLATO (Grupo Latino Americano de Tratamento de Osteossarcoma—Latin American Group of Osteosarcoma Treatment) protocol of 2006, which is based on high doses of cisplatin, doxorubicin and methotrexate. All clinical data are summarized in . 4.2. Gene Expression (qRT-PCR) The expression of 32 genes involved with pharmacogenetics was measured by a quantitative reverse transcription PCR (qRT-PCR). All frozen tissues were submitted to an RNA extraction using TRIzol ® Reagent (Thermo Fisher Scientific, Waltham, MA, USA). The cDNA was synthesized using SuperScript ® Vilo™ Master Mix (Invitrogen, Waltham, MA, USA). The qRT-PCR was performed in triplicate using TaqMan ® Gene Expression Assays (Thermo Fisher Scientific, Waltham, MA, USA) . The ACTB and GAPDH genes were used as endogenous controls. Normal bone was used as a calibrator. 4.3. Statistical Analyses Data analyses were performed using GraphPad Prism version 6.0 for Windows (GraphPad Software, San Diego, CA, USA). The gene expression measured by relative quantification was compared using nonparametric tests: the Wilcoxon and Mann-Whitney tests. The overall survival (OAS) and event-free survival (EFS) were calculated using the Kaplan-Meier method and the survival curves were compared using the log-rank test. The time of relapse was considered the time from the OS diagnosis until the relapse event. For OAS and EFS analyses, the median value of each gene and specimen type (biopsy, surgery or metastasis) was the cut-off that defined high or low expression. Statistical significance was considered when p < 0.05. We investigated 80 paired specimens obtained from 33 patients with diagnoses of OS. These patients were admitted to the Pediatric Oncology Institute (IOP/GRAACC/UNIFESP) between 2006 and 2016. The average age at diagnosis was 13 years old. Of 33 OS patients, 14 patients presented pulmonary metastasis. Thus, we investigated 33 biopsy specimens (pre-chemotherapy), 33 surgery specimens (post-chemotherapy) and 14 pulmonary metastasis specimens. Five normal bone tissues were used as a control; they were obtained from orthopedic surgeries of five healthy individuals that underwent trauma and did not present either genetic disorders or bone diseases. This study had the Research Ethics Committee approval from the Federal University of Sao Paulo (N° 0189/2016), and all patients agreed to participate by informed consent. All patients were treated following the GLATO (Grupo Latino Americano de Tratamento de Osteossarcoma—Latin American Group of Osteosarcoma Treatment) protocol of 2006, which is based on high doses of cisplatin, doxorubicin and methotrexate. All clinical data are summarized in . The expression of 32 genes involved with pharmacogenetics was measured by a quantitative reverse transcription PCR (qRT-PCR). All frozen tissues were submitted to an RNA extraction using TRIzol ® Reagent (Thermo Fisher Scientific, Waltham, MA, USA). The cDNA was synthesized using SuperScript ® Vilo™ Master Mix (Invitrogen, Waltham, MA, USA). The qRT-PCR was performed in triplicate using TaqMan ® Gene Expression Assays (Thermo Fisher Scientific, Waltham, MA, USA) . The ACTB and GAPDH genes were used as endogenous controls. Normal bone was used as a calibrator. Data analyses were performed using GraphPad Prism version 6.0 for Windows (GraphPad Software, San Diego, CA, USA). The gene expression measured by relative quantification was compared using nonparametric tests: the Wilcoxon and Mann-Whitney tests. The overall survival (OAS) and event-free survival (EFS) were calculated using the Kaplan-Meier method and the survival curves were compared using the log-rank test. The time of relapse was considered the time from the OS diagnosis until the relapse event. For OAS and EFS analyses, the median value of each gene and specimen type (biopsy, surgery or metastasis) was the cut-off that defined high or low expression. Statistical significance was considered when p < 0.05. |
Induced Coma, Death, and Organ Transplantation: A Physiologic, Genetic, and Theological Perspective | 5996d22c-9ee8-473f-b8d3-a32a520f3544 | 10059721 | Forensic Medicine[mh] | The word “death” is derived from the Greek word “thanatos” [θάνατος] and refers to both a physical punishment or spiritual punishment [Rom. 6:23]. Clinical death may have different meanings. Classically, cessation of vital functions, including cessation of heartbeat and respiration are the two criteria that are necessary to declare a person dead. In the clinic, the death certificate is issued if brain electrical activity is no longer detectable. The concept of brain death has been challenged by the Church. However, in 2015 the German Bishops Conference stated that brain death is the best and safest criterion to declare a person dead . After death, some of the organs may still be alive and can be used for transplantation. Quality criteria for the organs that are used for transplantation include organ pathophysiology and metabolic viability. Although gene expression is more sensitive than pathophysiological changes, the organs that are used for transplantation are not tested for gene expression or epigenetic modifications. However, gene activity does not cease after a person is declared dead, questioning the very definition of death. Indeed, the discovery that many genes are still working up to 48 h after death [ , , , ] has implications for our very definition of death, and in particular for forensics and organ transplantation.
2.1. The Heart Heart viability after death is short, 4–6 h. Currently, in the clinic, only hearts from donors meeting brain death criteria are accepted for heart transplantation. Indeed, modern resuscitation methods and availability of modern mechanical life-supporting systems have required the introduction of the concept of brain death. Thus, death by neurologic criteria means permanent cessation of the brainstem and cerebral functions, including no respiratory drive and lack of response to stimulation of the cranial nerve or to pain stimuli. For an adult patient who received a heart transplant, the median survival time following heart transplantation now approaches 12.2 years. More recently, because of increased demand, heart organs that are obtained from donors following circulatory death, i.e., the irreversible cessation of all circulatory and respiratory function, are also being used [ , , , ]. Gene set enrichment analysis and gene ontology analysis were used to study the transcriptional profile of the human left ventricle (LV) and right ventricle (RV) after 0, 4, and 8 h of cold ischemic storage in a preservation solution. Quite surprisingly, the LV and RV showed distinct, opposing genetic responses to cold storage including changes related to inflammation, including NFκB activation, oxidative phosphorylation, and fatty acid metabolism pathways after 8 h of storage . 2.2. The Lung Lung viability after death is similar to that of the heart, i.e., 4–6 h. Unfortunately, brain death leads to hemodynamic neuroendocrine abnormalities and metabolic decline, causing neurogenic pulmonary edema that limits the viability of the lungs. Since the percentage of lung retrieval rate is lower when compared to other organs that are used for transplantation, lungs after cardiocirculatory death (DCD) and brain death are increasingly used in the clinic [ , , ]. Gene expression in ex vivo normothermic perfused lungs has revealed increases in endothelial markers of inflammation and decreases in circulating leukocyte transcripts coding for CCL2, CCL3, CCR1, and CCR2, suggesting that lung recovery after normothermic perfusion follows specific stages, i.e., cellular death, cellular preservation, cellular reorganization, and cellular invasion . 2.3. The Liver The liver can survive a longer time outside of the living body (8–12 h) and both hypothermic and normothermic perfusion machines are being used to extend the liver viability up to 24 h. Normothermic machine perfusion (NMP) used for liver preservation provides a near-physiological environment to the liver and maintains the liver at 37 °C in a physiological state through the delivery of oxygen and nutrients. NMP allows organ therapy and quantitation of many metabolic and dynamic parameters to assess graft quality and viability. Hypothermic machine perfusion used for preservation in a cold environment relies on the reduction of the metabolic rate and, therefore, does not allow assessment of graft viability that is usually done by measuring lactate production, biliary biomarker, or transaminases . A common stressor to all organs after stress is the lack of oxygen. Thus, global gene expression profiles in liver biopsies after cold storage and at 90 min postreperfusion using the Affymetrix GeneChip Human Gene 1.0 ST array revealed significant alteration in transcripts coding for antioxidant, immunological, lipid biosynthesis, cell development, and growth pathways . 2.4. The Kidney Kidney viability after death is considerably long, 24–36 h. Similarly, both hypothermic and normothermic perfusion machines are being used to preserve and extend kidney viability. However, the major limiting factor for using kidneys after death is circulatory death causing kidney ischemia, organ damage, and viability impairment. One approach to preserve kidney viability is to keep the kidney, after being washed of blood, in a cold solution and attached to a hypothermic working machine at 4 °C. At this temperature, the rate of metabolism is 10% of that at normal physiological temperature. Gene expression analysis of biopsies that were taken from kidneys of deceased persons and kept in the cold solution has shown that indeed inflammation that was caused by cold ischemia was the major factor for transplant rejection at 3 months. Other activated genetic pathways were related to cell cycle/growth (e.g., IGFBP5, CSNK2A2), signal transduction (e.g., RASGRP3), immune response (e.g., CD83, BCL3, MX1), and metabolism (e.g., ENPP4, GBA3) . 2.5. The Brain Brains cannot be transplanted. Therefore, information about postmortem transcriptomics was obtained in a different way. Thus, to mimic the postmortem interval, the transcription patterns and histological features of postmortem brains were compared to slices of healthy cortical tissue surrounding the epileptic focus that were obtained immediately at surgery of epileptic patients. Within a few hours, a selective reduction in the transcriptional activity of neuronal genes with relative preservation of housekeeping gene expression was recorded, suggesting that cessation of brain electrical activity is a valid criterion for brain death. However, at the same time, there was a reciprocal increase in astroglial and microglial gene expression, resulting in the conversion to a reactive phenotype that increased gradually for at least 24 h after tissue removal. Histologically, neurons were rapidly degenerating and caused strong glial cell reactivity by activating their processes . 2.6. Death Genes in Animal Models More detailed technical studies of the death genes are available in animal models such as mice and zebrafish. A complex genetic analysis has shown that in postmortem tissues from animal models, 99% of gene transcripts decreased in abundance within 30 min after death while 1% had increased in abundance for up to 96 h postmortem. Quite interestingly, by function, the most abundant transcripts were involved in transport, stress, apoptosis, immunity, inflammation, development, epigenetic regulation, and cancer . Remarkably, this pattern of gene expression in animal models was tissue-specific. Thus, in the mouse, brain transcripts increased in the first hour and then gradually decreased, while mRNA gradually decreased with increased postmortem time (PMT) in the mouse liver suggesting that mRNAs are more stable in the brain than in the liver . Similar findings have been reported for humans, i.e., genes that are involved in wound healing and contracting heart muscle were active for at least 12 h after death in people who had died from multiple trauma, heart attack, or suffocation . 2.7. Gene Expression in Deep Coma We asked if there are similarities between gene expression in the brain after death and gene expression in induced coma. By searching the available literature, we found that indeed, seven genes were expressed both in the postmortem brains and induced coma: Cdc42 , Csnk2a1 , Gadd45a , Cdc42 , Tnfrsf14 , Prdx2 , Tpr , Rasa1 , and four genes had a similar function both in the postmortem brains and induced coma, Klkb1 , Bcl2 , Ier3 , and Zfand2a . By functionality, we grouped genes according to their function: regulator of energy metabolism ( Gadd45a ), apoptosis ( Bcl2 , Ier3 ), catabolic processes ( Zfand2a ), inflammation ( Klkb1 , Tnfrsf14 ), cell cycle control, and cancer ( Csnk2a1 , Cdc42 , Rasa1 , Prdx2 , Tpr ) ( ). Gadd45a encodes the growth arrest and DNA damage-inducible protein 45 alpha, a key regulator of energy metabolism and longevity by regulating epigenetic DNA methylation . Gadd45a transcript levels are also increased under stressful conditions . Gadd45 family members regulate this function through multiple cellular processes (e.g., DNA demethylation, gene expression, RNA stability, MAPK signaling) . During development, the Gadd45a gene is required for neural cell proliferation and is used as a pan-neural and neural crest marker . It has been hypothesized that GADD45 proteins are essential for brain function and their dysfunction might underlie pathophysiological conditions such as neuropsychiatric disorders and misexpression of Gadd45 family members were associated with psychiatric diseases . Thus, unpredictable chronic mild stress (UCMS) reduces the expression of Gadd45 family members, suggesting that Gadd45 family members are new putative targets for UCMS treatment . Bcl2 gene encodes an integral outer mitochondrial membrane protein that blocks the apoptotic death of some cells, including lymphocytes. Constitutive expression of BCL2 is thought to be the cause of follicular lymphoma, mature T-cell lymphomas , prostate cancer , and renal cell carcinomas . Most importantly for the brain death, stress activates cell survival and cell death signaling pathways that converge on mitochondria, a process that is controlled by the activities of BCL-2 proteins. Indeed, BCL-2 proteins play a significant role in initiating or inhibiting apoptosis during neuronal development and injury. Emerging evidence suggests that BCL-2 is involved in maintaining both mitochondrial bioenergetics and neuronal membrane potential by modulation of Ca 2+ signaling during acute neuronal injury and thus plays a role in neuroprotection . Ier3 is a stress inducible gene that encodes for an immediate early response 3 (IER3) protein that is involved in the protection of cells from Fas - or tumor necrosis factor type alpha-induced apoptosis. Increasing evidence suggests that IER3 functions either as an oncogene or as a tumor suppressor in various human cancers . Thus, it was recently reported that IER3 induces the apoptosis of cervical cancer cells and that its expression is downregulated in patients with cervical cancer via its ubiquitination, followed by proteasomal degradation . The Zfand2a mRNA is normally strongly expressed in the eye, brain, and heart of mice. The ZFAND2A/AIRAP protein encoded by Zfand2a mRNA is part of the proteasome complex and is involved in proteasome-mediated ubiquitin-dependent accelerated protein catabolic processes in the skeletal muscles and protein targeting to ER, resulting in a loss of muscle mass by increasing ATP hydrolysis [ , , ]. Therefore, pharmacological antagonists of ZFAND5 may serve as a treatment of the debilitating loss of muscle protein in a variety of cachectic disorders, as well as in these muscular dystrophies . A recent study also identified ZFAND2A/AIRAP as a novel stress-regulated survival factor implicated in the stabilization of the antiapoptotic protein cIAP2 that is highly expressed in several cancers, including melanoma. An increased expression of cIAP2 blocks the pro-apoptotic pathways and increases resistance to the drug bortezomib in melanoma cells. Since ZFAND2A/AIRAP-siRNA-mediated knockdown affects cIAP2 protein stability in melanoma cells, ZFAND2A/AIRAP downregulation markedly enhances bortezomib anticancer activity in melanoma . Klkb1 transcript encodes a glycoprotein that is part of the plasma contact activation system and plays a role in the defense response to infections by participating in the surface-dependent activation of blood coagulation, fibrinolysis, and kinin generation. Indeed, it was recently shown that kallikrein (KK) activates the complement system by cleaving C3 and factor B, with subsequent formation of an active complement C3 convertase to yield active components C3a and C3b and trigger complement activation . Recent evidence suggests that kallikrein-kinin and the coagulation system contribute to autoimmune CNS diseases such as multiple sclerosis (MS) by mediating transendothelial trafficking of inflammatory cells across the blood brain barrier (BBB). Since pharmacological inhibition of plasma prekallikrein (PKK) or KK, respectively, protects mice from experimental autoimmune encephalomyelitis (EAE) by reducing transendothelial leukocyte trafficking, PK inhibition may offer a strategy for the treatment of MS . Tnfrsf14 transcript encodes a member of the tumor necrosis factor (TNF) receptor superfamily that has a pivotal role in T-cell-mediated adaptive immunity and immune diseases. The encoded protein functions in signal transduction pathways that activate inflammatory and inhibitory T-cell immune responses . TNFR superfamily molecules including TNFRSF14, are constitutive or inducible expressed on T cells and play important roles in protective immunity, inflammatory, autoimmune diseases, and tumor immunotherapy . In the brain, Tnfrsf14 transcripts showed a positive correlation with a WHO grade of glioma and TNFRSF14 was significantly increased in a mesenchymal glioma subtype. Indeed, a higher TNFRSF14 expression was significantly associated with a shorter survival for glioma patients and Cox regression models revealed that TNFRSF14 expression was an independent variable for predicting survival . One gene that is shared by both conditions is Csnk2a1 . Csnk2a mRNA encodes casein kinase II, a constitutively active serine/threonine protein kinase that phosphorylates hundreds of acidic proteins, including casein. It is involved in various signaling pathways and cellular processes, including cell cycle control, apoptosis, and circadian rhythmicity. In the brain, casein kinase 2 (CK2), is highly expressed in infections, Alzheimer’ disease (AD) patients, and contributes to the AD pathology by regulating NR2B-mediated neurotransmission . CK2 has enhanced expression/activity in a plethora of human diseases and numerous cancers, including glioblastoma (GBM). Therefore, inhibitors of CK2 may be good candidates for the inhibition of tumor growth as has been shown in GBM xenograft mouse models . Cdc42 transcript encodes a small GTPase of the Rho-subfamily, which regulates signaling pathways that control diverse cellular functions including cell morphology, migration, endocytosis, cell cycle progression, and epigenetic regulation. Indeed, CDC42 can downregulate the inhibitor of DNA binding 4 (ID4) which has been shown to be overexpressed in colorectal adenocarcinomas through hypermethylation of its promoter . In the brain, the small GTPase, CDC42, plays an essential role in neurogenesis and brain development. Thus, CDC42 stimulates mTORC1 activity and thereby upregulates tissue-specific transcription factors that are essential for neural progenitor formation. Further, by promoting EGFR degradation, it was found that CDC42b and ACK stimulate autophagy and trigger neural progenitor cells to differentiate into neurons . The protein that is encoded by Rasa1 mRNA is part of the GAP1 family of GTPase-activating proteins which regulate multiple cellular signaling pathways including those that control cell growth, differentiation, and survival. RASA1 is involved in numerous physiological processes such as angiogenesis, cell proliferation, and apoptosis. Mutations leading to changes in the binding sites of RASA1 are associated with basal cell carcinomas and multiple tumor types of the lung, intestines, liver, and breast . Prdx2 gene encodes a member of the peroxiredoxin family of antioxidant enzymes, that reduce hydrogen peroxide and alkyl hydroperoxides and thus plays an antioxidant protective role in cells. Thus, PRDX2 reduces the production of reactive oxygen species by catalyzing the reduction of hydrogen peroxide to water, thereby protecting neurons against oxidative stress. However, overexpression of PRDX2 accelerates brain damage after stroke by activating an inflammatory response . Therefore, restoration of redox balance may play an important role in minimizing the detrimental effects of oxidative damage in neurodegenerative disorders. PRDX2 is also an antioxidant and molecular chaperone that can be secreted by tumor cells. Increasing evidence suggests that the role of PRDXs may go beyond the antioxidant properties and could be related to the regulation of cell signaling. Thus, induction of oxidative stress via H 2 O 2 treatment leads to the overexpression of PRXD-2 and inhibition of cancer cell proliferation . Recent results also indicate that PRDX2 promotes cell survival and inflammation-associated myocardial hypertrophy . Tpr gene encodes a large coiled-coil protein that forms intranuclear filaments that are attached to the inner surface of nuclear pore complexes (NPCs). During oncogene-induced senescence (OIS), heterochromatin forms internal senescence-associated heterochromatin foci (SAHFs). Recently it was shown that the nucleoporin TPR is necessary for both the formation and maintenance of SAHFs . Of note, TPR expression is significantly increased in lung cancer tissues and correlated with poor prognosis .
Heart viability after death is short, 4–6 h. Currently, in the clinic, only hearts from donors meeting brain death criteria are accepted for heart transplantation. Indeed, modern resuscitation methods and availability of modern mechanical life-supporting systems have required the introduction of the concept of brain death. Thus, death by neurologic criteria means permanent cessation of the brainstem and cerebral functions, including no respiratory drive and lack of response to stimulation of the cranial nerve or to pain stimuli. For an adult patient who received a heart transplant, the median survival time following heart transplantation now approaches 12.2 years. More recently, because of increased demand, heart organs that are obtained from donors following circulatory death, i.e., the irreversible cessation of all circulatory and respiratory function, are also being used [ , , , ]. Gene set enrichment analysis and gene ontology analysis were used to study the transcriptional profile of the human left ventricle (LV) and right ventricle (RV) after 0, 4, and 8 h of cold ischemic storage in a preservation solution. Quite surprisingly, the LV and RV showed distinct, opposing genetic responses to cold storage including changes related to inflammation, including NFκB activation, oxidative phosphorylation, and fatty acid metabolism pathways after 8 h of storage .
Lung viability after death is similar to that of the heart, i.e., 4–6 h. Unfortunately, brain death leads to hemodynamic neuroendocrine abnormalities and metabolic decline, causing neurogenic pulmonary edema that limits the viability of the lungs. Since the percentage of lung retrieval rate is lower when compared to other organs that are used for transplantation, lungs after cardiocirculatory death (DCD) and brain death are increasingly used in the clinic [ , , ]. Gene expression in ex vivo normothermic perfused lungs has revealed increases in endothelial markers of inflammation and decreases in circulating leukocyte transcripts coding for CCL2, CCL3, CCR1, and CCR2, suggesting that lung recovery after normothermic perfusion follows specific stages, i.e., cellular death, cellular preservation, cellular reorganization, and cellular invasion .
The liver can survive a longer time outside of the living body (8–12 h) and both hypothermic and normothermic perfusion machines are being used to extend the liver viability up to 24 h. Normothermic machine perfusion (NMP) used for liver preservation provides a near-physiological environment to the liver and maintains the liver at 37 °C in a physiological state through the delivery of oxygen and nutrients. NMP allows organ therapy and quantitation of many metabolic and dynamic parameters to assess graft quality and viability. Hypothermic machine perfusion used for preservation in a cold environment relies on the reduction of the metabolic rate and, therefore, does not allow assessment of graft viability that is usually done by measuring lactate production, biliary biomarker, or transaminases . A common stressor to all organs after stress is the lack of oxygen. Thus, global gene expression profiles in liver biopsies after cold storage and at 90 min postreperfusion using the Affymetrix GeneChip Human Gene 1.0 ST array revealed significant alteration in transcripts coding for antioxidant, immunological, lipid biosynthesis, cell development, and growth pathways .
Kidney viability after death is considerably long, 24–36 h. Similarly, both hypothermic and normothermic perfusion machines are being used to preserve and extend kidney viability. However, the major limiting factor for using kidneys after death is circulatory death causing kidney ischemia, organ damage, and viability impairment. One approach to preserve kidney viability is to keep the kidney, after being washed of blood, in a cold solution and attached to a hypothermic working machine at 4 °C. At this temperature, the rate of metabolism is 10% of that at normal physiological temperature. Gene expression analysis of biopsies that were taken from kidneys of deceased persons and kept in the cold solution has shown that indeed inflammation that was caused by cold ischemia was the major factor for transplant rejection at 3 months. Other activated genetic pathways were related to cell cycle/growth (e.g., IGFBP5, CSNK2A2), signal transduction (e.g., RASGRP3), immune response (e.g., CD83, BCL3, MX1), and metabolism (e.g., ENPP4, GBA3) .
Brains cannot be transplanted. Therefore, information about postmortem transcriptomics was obtained in a different way. Thus, to mimic the postmortem interval, the transcription patterns and histological features of postmortem brains were compared to slices of healthy cortical tissue surrounding the epileptic focus that were obtained immediately at surgery of epileptic patients. Within a few hours, a selective reduction in the transcriptional activity of neuronal genes with relative preservation of housekeeping gene expression was recorded, suggesting that cessation of brain electrical activity is a valid criterion for brain death. However, at the same time, there was a reciprocal increase in astroglial and microglial gene expression, resulting in the conversion to a reactive phenotype that increased gradually for at least 24 h after tissue removal. Histologically, neurons were rapidly degenerating and caused strong glial cell reactivity by activating their processes .
More detailed technical studies of the death genes are available in animal models such as mice and zebrafish. A complex genetic analysis has shown that in postmortem tissues from animal models, 99% of gene transcripts decreased in abundance within 30 min after death while 1% had increased in abundance for up to 96 h postmortem. Quite interestingly, by function, the most abundant transcripts were involved in transport, stress, apoptosis, immunity, inflammation, development, epigenetic regulation, and cancer . Remarkably, this pattern of gene expression in animal models was tissue-specific. Thus, in the mouse, brain transcripts increased in the first hour and then gradually decreased, while mRNA gradually decreased with increased postmortem time (PMT) in the mouse liver suggesting that mRNAs are more stable in the brain than in the liver . Similar findings have been reported for humans, i.e., genes that are involved in wound healing and contracting heart muscle were active for at least 12 h after death in people who had died from multiple trauma, heart attack, or suffocation .
We asked if there are similarities between gene expression in the brain after death and gene expression in induced coma. By searching the available literature, we found that indeed, seven genes were expressed both in the postmortem brains and induced coma: Cdc42 , Csnk2a1 , Gadd45a , Cdc42 , Tnfrsf14 , Prdx2 , Tpr , Rasa1 , and four genes had a similar function both in the postmortem brains and induced coma, Klkb1 , Bcl2 , Ier3 , and Zfand2a . By functionality, we grouped genes according to their function: regulator of energy metabolism ( Gadd45a ), apoptosis ( Bcl2 , Ier3 ), catabolic processes ( Zfand2a ), inflammation ( Klkb1 , Tnfrsf14 ), cell cycle control, and cancer ( Csnk2a1 , Cdc42 , Rasa1 , Prdx2 , Tpr ) ( ). Gadd45a encodes the growth arrest and DNA damage-inducible protein 45 alpha, a key regulator of energy metabolism and longevity by regulating epigenetic DNA methylation . Gadd45a transcript levels are also increased under stressful conditions . Gadd45 family members regulate this function through multiple cellular processes (e.g., DNA demethylation, gene expression, RNA stability, MAPK signaling) . During development, the Gadd45a gene is required for neural cell proliferation and is used as a pan-neural and neural crest marker . It has been hypothesized that GADD45 proteins are essential for brain function and their dysfunction might underlie pathophysiological conditions such as neuropsychiatric disorders and misexpression of Gadd45 family members were associated with psychiatric diseases . Thus, unpredictable chronic mild stress (UCMS) reduces the expression of Gadd45 family members, suggesting that Gadd45 family members are new putative targets for UCMS treatment . Bcl2 gene encodes an integral outer mitochondrial membrane protein that blocks the apoptotic death of some cells, including lymphocytes. Constitutive expression of BCL2 is thought to be the cause of follicular lymphoma, mature T-cell lymphomas , prostate cancer , and renal cell carcinomas . Most importantly for the brain death, stress activates cell survival and cell death signaling pathways that converge on mitochondria, a process that is controlled by the activities of BCL-2 proteins. Indeed, BCL-2 proteins play a significant role in initiating or inhibiting apoptosis during neuronal development and injury. Emerging evidence suggests that BCL-2 is involved in maintaining both mitochondrial bioenergetics and neuronal membrane potential by modulation of Ca 2+ signaling during acute neuronal injury and thus plays a role in neuroprotection . Ier3 is a stress inducible gene that encodes for an immediate early response 3 (IER3) protein that is involved in the protection of cells from Fas - or tumor necrosis factor type alpha-induced apoptosis. Increasing evidence suggests that IER3 functions either as an oncogene or as a tumor suppressor in various human cancers . Thus, it was recently reported that IER3 induces the apoptosis of cervical cancer cells and that its expression is downregulated in patients with cervical cancer via its ubiquitination, followed by proteasomal degradation . The Zfand2a mRNA is normally strongly expressed in the eye, brain, and heart of mice. The ZFAND2A/AIRAP protein encoded by Zfand2a mRNA is part of the proteasome complex and is involved in proteasome-mediated ubiquitin-dependent accelerated protein catabolic processes in the skeletal muscles and protein targeting to ER, resulting in a loss of muscle mass by increasing ATP hydrolysis [ , , ]. Therefore, pharmacological antagonists of ZFAND5 may serve as a treatment of the debilitating loss of muscle protein in a variety of cachectic disorders, as well as in these muscular dystrophies . A recent study also identified ZFAND2A/AIRAP as a novel stress-regulated survival factor implicated in the stabilization of the antiapoptotic protein cIAP2 that is highly expressed in several cancers, including melanoma. An increased expression of cIAP2 blocks the pro-apoptotic pathways and increases resistance to the drug bortezomib in melanoma cells. Since ZFAND2A/AIRAP-siRNA-mediated knockdown affects cIAP2 protein stability in melanoma cells, ZFAND2A/AIRAP downregulation markedly enhances bortezomib anticancer activity in melanoma . Klkb1 transcript encodes a glycoprotein that is part of the plasma contact activation system and plays a role in the defense response to infections by participating in the surface-dependent activation of blood coagulation, fibrinolysis, and kinin generation. Indeed, it was recently shown that kallikrein (KK) activates the complement system by cleaving C3 and factor B, with subsequent formation of an active complement C3 convertase to yield active components C3a and C3b and trigger complement activation . Recent evidence suggests that kallikrein-kinin and the coagulation system contribute to autoimmune CNS diseases such as multiple sclerosis (MS) by mediating transendothelial trafficking of inflammatory cells across the blood brain barrier (BBB). Since pharmacological inhibition of plasma prekallikrein (PKK) or KK, respectively, protects mice from experimental autoimmune encephalomyelitis (EAE) by reducing transendothelial leukocyte trafficking, PK inhibition may offer a strategy for the treatment of MS . Tnfrsf14 transcript encodes a member of the tumor necrosis factor (TNF) receptor superfamily that has a pivotal role in T-cell-mediated adaptive immunity and immune diseases. The encoded protein functions in signal transduction pathways that activate inflammatory and inhibitory T-cell immune responses . TNFR superfamily molecules including TNFRSF14, are constitutive or inducible expressed on T cells and play important roles in protective immunity, inflammatory, autoimmune diseases, and tumor immunotherapy . In the brain, Tnfrsf14 transcripts showed a positive correlation with a WHO grade of glioma and TNFRSF14 was significantly increased in a mesenchymal glioma subtype. Indeed, a higher TNFRSF14 expression was significantly associated with a shorter survival for glioma patients and Cox regression models revealed that TNFRSF14 expression was an independent variable for predicting survival . One gene that is shared by both conditions is Csnk2a1 . Csnk2a mRNA encodes casein kinase II, a constitutively active serine/threonine protein kinase that phosphorylates hundreds of acidic proteins, including casein. It is involved in various signaling pathways and cellular processes, including cell cycle control, apoptosis, and circadian rhythmicity. In the brain, casein kinase 2 (CK2), is highly expressed in infections, Alzheimer’ disease (AD) patients, and contributes to the AD pathology by regulating NR2B-mediated neurotransmission . CK2 has enhanced expression/activity in a plethora of human diseases and numerous cancers, including glioblastoma (GBM). Therefore, inhibitors of CK2 may be good candidates for the inhibition of tumor growth as has been shown in GBM xenograft mouse models . Cdc42 transcript encodes a small GTPase of the Rho-subfamily, which regulates signaling pathways that control diverse cellular functions including cell morphology, migration, endocytosis, cell cycle progression, and epigenetic regulation. Indeed, CDC42 can downregulate the inhibitor of DNA binding 4 (ID4) which has been shown to be overexpressed in colorectal adenocarcinomas through hypermethylation of its promoter . In the brain, the small GTPase, CDC42, plays an essential role in neurogenesis and brain development. Thus, CDC42 stimulates mTORC1 activity and thereby upregulates tissue-specific transcription factors that are essential for neural progenitor formation. Further, by promoting EGFR degradation, it was found that CDC42b and ACK stimulate autophagy and trigger neural progenitor cells to differentiate into neurons . The protein that is encoded by Rasa1 mRNA is part of the GAP1 family of GTPase-activating proteins which regulate multiple cellular signaling pathways including those that control cell growth, differentiation, and survival. RASA1 is involved in numerous physiological processes such as angiogenesis, cell proliferation, and apoptosis. Mutations leading to changes in the binding sites of RASA1 are associated with basal cell carcinomas and multiple tumor types of the lung, intestines, liver, and breast . Prdx2 gene encodes a member of the peroxiredoxin family of antioxidant enzymes, that reduce hydrogen peroxide and alkyl hydroperoxides and thus plays an antioxidant protective role in cells. Thus, PRDX2 reduces the production of reactive oxygen species by catalyzing the reduction of hydrogen peroxide to water, thereby protecting neurons against oxidative stress. However, overexpression of PRDX2 accelerates brain damage after stroke by activating an inflammatory response . Therefore, restoration of redox balance may play an important role in minimizing the detrimental effects of oxidative damage in neurodegenerative disorders. PRDX2 is also an antioxidant and molecular chaperone that can be secreted by tumor cells. Increasing evidence suggests that the role of PRDXs may go beyond the antioxidant properties and could be related to the regulation of cell signaling. Thus, induction of oxidative stress via H 2 O 2 treatment leads to the overexpression of PRXD-2 and inhibition of cancer cell proliferation . Recent results also indicate that PRDX2 promotes cell survival and inflammation-associated myocardial hypertrophy . Tpr gene encodes a large coiled-coil protein that forms intranuclear filaments that are attached to the inner surface of nuclear pore complexes (NPCs). During oncogene-induced senescence (OIS), heterochromatin forms internal senescence-associated heterochromatin foci (SAHFs). Recently it was shown that the nucleoporin TPR is necessary for both the formation and maintenance of SAHFs . Of note, TPR expression is significantly increased in lung cancer tissues and correlated with poor prognosis .
Clinical trials have shown that in human patients, prolonged induced coma in intensive care units is associated with cognitive impairment in survivors . Yet, at the synapse-level, a morphological correlation between prolonged anesthesia and cognition could be found only in mice that were subjected to prolonged general anesthesia. Thus, the authors used two-photon imaging of fluorescently labeled dendrites and synapses to assess synaptic turnover and found that synapse turnover was more than doubled in experimental anesthesia. An increased synapse turnover may result in impaired cognition . There is a common belief that most organelles, including mitochondria, deteriorate after death. Neuronal function in the brain requires a lot of energy and a reduction in ATP production in neurons can negatively affect neuronal function. Indeed, numerous studies have reported altered or decreased mitochondrial energy production in neurodegenerative diseases, neurometabolic disorders such as Leigh syndrome, and in psychiatric disorders such as schizophrenia and bipolar disorder [ , , , , , ]. However, it has been reported that there are plenty of structurally intact and functional mitochondria in postmortem mouse and human brains shortly after death. Moreover, mitochondria can be frozen for future functional assessments . Unfortunately, histological data that are obtained in studies on animals are often extrapolated to humans without considering the postmortem time. Autopsy tissue is rarely obtained within 2 h postmortem time (PMT). For example, in the mouse brain that was fixed at different PMT, significant metabolomic changes already occurred at 2 h PMT, whereas neuroanatomical changes occurred mostly at 5 h PMT. Thus, the levels of pyroglutamic acid, anandamide, and urea increased until 2 h PMT whereas GABA, creatinine, N-acetyl-aspartic acid, putrescine, and cadaverine reached higher levels at 5 h PMT. Other compounds such as cholesterol, cholesterol esters, and arachidonoylglycerol were stable during the study period . Another study investigated the time course of postmortem proteolytic degradation of selected tissue antigens and found that the preservation of antigens for Ki67, Vimentin, Pancytokeratin, and CD20 was tissue-dependent, i.e., the autolysis process was not appreciable in the first 5 days PMT. However, the liver and the spleen underwent rapid autolysis, while the kidney displayed an autolysis of the tubules after one day PMT and parcellular autolysis of the glomeruli at 5 days PMT .
In the clinic, the death certificate is issued if brain electrical activity is no longer detectable. The concept of brain death has been challenged by the Church. However, in 2015 the German Bishops Conference stated it is, nevertheless, the best and safest criterion to declare a person dead. In 1989 the Pontifical Academy of Science of the Roman Catholic Church also stated: “a person is dead when there has been total and irreversible loss of all capacity for integrating and coordinating physical and mental functions of the body as a unit” ( https://sites.sju.edu/icb/position-catholic-church-organ-donation/ , accessed on 14 March 2023). The reliability of the encephalogram (EEG) to confirm brain death remains, nevertheless, controversial. Indeed, several studies have shown that brain electrical activity persists for up to 72 h after the person is declared dead . By function, the most abundant transcripts were involved in transport, stress, apoptosis, immunity, inflammation, development, epigenetic regulation, and cancer . Within a few hours, there is a selective reduction in the transcriptional activity of neuronal genes with relative preservation of housekeeping genes, commonly used as a reference for RNA normalization, suggesting that indeed, cessation of brain electrical activity could be taken as a valid death criterium ( ). So, why do so many genes become activated after death? One possible explanation is that these genes are involved in cellular proliferation during development. In every living body there is continuous competition for resources, even if we are not aware of it. Thus, a rapid decay of genes whose expression normally suppress other genes, such as those involved in embryological development, might allow the usually quiet genes to become active for a short period of time. Since these genes are involved in cellular proliferation, it could be an attempt of cells to escape mortality and raises the question of organ genetics and viability used for transplantation after death. Secondly, once the inhibitory gene network disintegrates, it is natural for a cell to try escape mortality by proliferation. This immediately raises the next question about their long-term effects in the host. Since many such genes are shared both by the human brains after death and the brains in induced coma, it is legitimate to ask if long-term deep coma may have severe side effects that may result in cancer development. Another explanation is that many of the genes become active as part of physiological processes that aid healing or recovery after severe injury. For example, histological degeneration of neuronal structure during aging or brain injuries, is paralleled by activation of astroglia and microglia, a phenomenon that is seen both in the aging brain , Alzheimer disease , and brain injuries that are caused by stroke or TBI . Similarly, there were increases in abundance after death of genes that are required for innate ( Laao , Tox2 ) and adaptive immunity ( Ms4a17 , Usp18 ) and increased levels of the apoptotic genes ( Fosb , Bcl2l11 ) that are normally activated by infection and injury . Third, death can be seen as a condition of extreme stress . Hence, the increased abundance of specific transcripts shortly after death could be attributed to genes encoding proteins involved in resilient pathways (e.g., heat shock, Hsp , hypoxia-related, Hif1ab and oxidative stress, Gadd45a , March4 ) to compensate for extreme stress and increase survival chances . A closely related phenomenon is the near-death experience [NDE] which is an altered state of consciousness due to the decoupling of the psychosoma from the physical body both in clinical conditions such as multiple organ failure, late-stage cancer confusion and life-threatening events including traffic accidents, traumatic brain injury (TBI), physical assaults, and drug abuse. NDEs are not a rare phenomenon, occurring in approximately 10–23% of cardiac arrest survivors and 3% of TBI survivors [ , , , ]. Since NDE has been known worldwide for a long time and numerous cultural backgrounds, it suggests an underlying common neurobiological mechanism. Thus, it has been hypothesized that thanatosis, aka death-feigning, or tonic immobility, is a highly conserved defense and survival mechanism that could have an evolutionary origin . For example, when attacked by a predator, animals can feign death as a last resort to improve their chances of survival . Indeed, in the animal kingdom, thanatosis is an anti-predator strategy that is part of an innate defense cascade which is activated when mechanisms underlying the fight or flight response triggered by cortisol, are no longer possible [ , , , , ]. Behaviorally, it involves sudden onset of immobility, with or without loss of tonic muscular activity, and unresponsiveness to external stimuli while awareness is preserved . In humans, peritraumatic tonic immobility has been described as a cerebral defense mechanism in post-traumatic stress disorder that can happen during sexual assault [ , , ]. Proposed mechanisms underlying NDEs include cortical spreading depolarizations (CSDs), migraine aura, a predictor of near-death experiences . Indeed, a short-lasting variant of CSDs is considered the pathophysiological correlate of migraine aura while terminal CSDs occur during the dying process of the brain in humans , rats , and insects . At the molecular level, mechanisms underlying NDEs include the activity of the potent serotonergic psychedelic drug N,N-Dimethyltryptamine (DMT) or hypofunction of the N-methyl-D-aspartate receptor (NMDAR) [ , , ]. Quite interestingly, MDAR hypofunction is also induced by ketamine when used for recreational purposes , or serotonergic psychedelics (also referred to as serotonergic hallucinogens, psilocybin, LSD, and mescaline), including the endogenous serotonin 2A receptor agonist DMT, suggesting that endogenous NMDA antagonists with neuroprotective properties may be released in the proximity of death . In a study that was published in the journal Nature, it was shown that in mice, ketamine produced 1–3 Hz oscillations in neurons of layer 5 of the retrosplenial cortex, a region that is required for visuospatial navigation and episodic memory, that blocked the communication with other brain parts. In humans, a 3 Hz oscillation in the deep posteromedial cortex (analog to mouse retrosplenial cortex) can induce dissociation in patients who experienced a seizure aura or have been stimulated externally . We discovered a very interesting parallel between genes that were upregulated in the brain after death and genes that were upregulated in the brains subjected to induced coma, including genes involved in neurotransmission, proteosomal degradation, apoptosis, inflammation, and most interestingly, cancer. While in humans it is not possible to study gene activity related to these processes, it opens a new avenue of research on the side effects of induced coma in animal models. Organs that are used for transplantation are mostly obtained from donors that are declared dead after irreversible loss of brain functions . However, brain death causes a massive inflammatory response that triggers substantial circulatory and metabolic changes (e.g., hypoxia, pH) in the donor’s body that compromise organs vitality . Indeed, biochemical investigations of biopsies that were obtained from organs used for transplantation have revealed substantial increases in the expression of genes that are involved in oxidative stress, apoptosis, adhesion, and inflammation biomarkers. Therefore, postmortem time could have a significant effect on gene expression because of the physiological changes that take place in the body after organismal death . Yet to date, such studies have not been officially used to assess organ quality and/or transplantation success. Therefore, Pozhitkov and Noble have suggested a closer examination of the relationship between gene expression and postmortem time which might yield valuable information about organ quality. Thus, activation of many developmental genes, including Mdga2 , Ripply3 , or tumor suppressor Tnfrsf9 , and the oncogenic ( Tpr , Csnk2a1 ) transcripts in the first 24 h after death may increase the risk of carcinogenesis in the transplanted organs. Indeed, it is well established that recipients of transplanted solid organs have a higher risk of developing cancer than the general population . Worrisomely, cancer accounts for approximately 10–30% of deaths in recipients who received organs . Of concern is that 11 genes are upregulated both in postmortem brains and induced coma: Cdc42 , Csnk2a1 , Gadd45a , Prdx2 , Rasa1 , Tnfrsf14 , Tpr , Klkb1 , Bcl2 , Ier3 , and Zfand2a [ , , , , , , , , , , ]. Further, near-death experiences and induced coma that is required in numerous conditions associated with cardiopulmonary arrest with multiple organ failure, are always accompanied by a loss of consciousness [ , , , , , ]. For neurologists, these so-called near-death experiences may have a neural basis. Thus, recent data from the brain scan of a dying human brain obtained in a non-experimental, real-life acute care clinical setting provided the first evidence that the dying brain goes through a sudden flash of memories seconds before and after the heart stopped beating and advocate that the human brain may possess the capability to generate coordinated activity during the near-death period . Finally, one factor limiting the organ availability for transplantation is the religious belief. Although no religion obliges one to donate or refuse organs from deceased donors, its practice may be discouraged by Roma Gypsies, Shintoists, Native Americans, Confucians, and some Orthodox rabbis. Some South Asia Muslim ulemas (scholars) and muftis (jurists) also oppose donation of organs from deceased persons. However, they do not oppose xenotransplantation research . More recently, organ donation for the benefit of humans in need has been seen as “posthumous giving of organs and tissues can be a manifestation of love spreading also to the other side of death” . This position is assumed by the great religions of the world, especially by the Protestant, Catholic, and Orthodox Churches. Indeed, both Protestants and Catholics allow the removal of organs to save lives provided that organ removal does not violate the dignity or the eternal peace of death .
|
Long COVID-19 renal disease: A present medical need for nephrology | 30729bf6-59b8-44b8-b75b-e9075c617dd9 | 10060193 | Internal Medicine[mh] | • Secondary to organ damage due to serious illness (AKI) Preclinical studies show that after ischaemic renal injury and despite the return of serum creatinine concentrations to normal values, there is persistence of inflammation, renal fibrosis, abnormal gene expression profiles of the kidney, and functional impairment. Therefore, it is not clear if all cases of kidney involvement post-COVID-19 can be included as part of long COVID, because in most cases they can be explained by organ damage secondary to ischaemia and/or infection. A clear example is in a series of patients hospitalised for COVID-19 who had AKI and of whom 47% still had kidney damage at the time of hospital discharge. This kidney damage and subclinical inflammation and injury may persist for many months, resulting in a progressive decline in kidney function leading to CKD. A study from the USA using Veterans Health Administration (VHA) electronic health records to perform a comprehensive assessment of prolonged COVID-19 found that COVID-19 increased the risk of CKD and that this risk was higher among those with severe illness. • Secondary to the presence of the virus (long COVID) However, it seems that it is not only a previous AKI the explanation of post-COVID-19 kidney disease. In one study published in Cell Stem Cell, researchers studied kidney tissue from COVID-19 patients admitted to the ICU. They found scar tissue compared to intensive care patients with a non-COVID-19-related lung infection and a control group. Furthermore, in autopsy samples, they observed that SARS-COV-2 directly infects renal cells and is associated with increased renal tubulointerstitial fibrosis. So, based on that, they investigated the direct effects of the virus on the kidney, independent of the systemic effects. To do this, they infected with SARS-COV-2 kidney organoids (microkidneys) derived from human pluripotent stem cells and containing many different renal cells, except immune cells. They studied the virus’s direct effect on renal cells, regardless of possible side effects caused by immune cells or other systemic effects. Single-cell RNA sequencing indicated injury and dedifferentiation of the infected cells with activation of profibrotic signalling pathways. SARS-COV-2 infection also led to increased collagen 1 protein expression in organoids. The SARS-COV-2 protease inhibitor was able to improve the infection by SARS-COV-2 of renal cells. These results suggest that SARS-CoV-2 can directly infect renal cells and induce injury with subsequent fibrosis. That is, the virus causes direct cell damage, independent of the immune system. These data could explain both acute kidney injury in patients with COVID-19 and the development of chronic kidney disease in long COVID. We do not yet understand completely the complexity of the kidney damage, and there may be other mechanisms that may explain kidney injury in long COVID in addition to the direct effects of the virus, such as an abnormal immune response or autoimmunity, persistent inflammation, alterations in endothelial function and in the coagulation system, or alterations in the autonomic nervous system. All this requires from us to consolidate cases and research projects to make further progress. Also, changes in social (such as reduced social contact and loneliness), financial (such as job loss), and behavioural (such as changes in diet and exercise) conditions that people with COVID-19 may experience could explain some of the general complications of long COVID, including renal complications due to non-compliance of the treatment, lack of blood pressure (BP) control, etc.
There are still not many follow-up studies on kidney involvement in long COVID, but there may be loss of kidney function months after the illness. This could be a silent disease, which also occurs in patients who did not require hospital admission. In the already mentioned work of Bowe et al. a cohort of 1,726,683 US veterans diagnosed with COVID-19 of varying severity and 1,637,467 uninfected controls were studied, adjusting for prior comorbidities and other covariates to ensure the separation of the pure effect of COVID-19 on outcomes. They demonstrated that after 30 days of COVID-19 the hazard ratio (HR) of AKI is 1.94;decreased glomerular filtration (GFR) ≥30% HR:1.25 ≥ 40%, HR: 1.44, ≥50% HR:1.62 and major adverse renal events (MARE) (decrease in GFR >50%, end-stage kidney disease, or all-cause mortality) HR:1.66. These data correlate with the severity of the infection. However those who were not hospitalised also had loss of kidney function. Compared with non-infected controls, 30-day COVID-19 survivors exhibited a decreased GFR of −3.26, −5.20, and −7.69 ml/min/1.73 m 2 per year, respectively in non-hospitalised and hospitalised patients and patients admitted to intensive care during the acute phase of infection by COVID-19. What it is important is that although having AKI in the acute phase influenced the results of the loss of renal function post-COVID, the results of this study suggest that the deterioration of renal function was also observed in those who did not suffer AKI during the acute phase. Other studies also insist on the need for post-COVID renal surveillance, finding that in patients who had AKI in the hospital, AKI associated with COVID-19 was associated with a greater rate of decline in estimated GFR (eGFR) after discharge compared to AKI in patients without COVID-19, regardless of underlying comorbidities or AKI severity. Therefore, considering that there have been close to 500 million recorded cases of COVID-19 globally, a figure that is undoubtedly three or four times higher, and that non-hospitalised patients who did not have AKI in the acute phase also observed deterioration of renal function, it is clear the necessity for the presence of Nephrology in monitoring and prevention of kidney function deterioration.
The data we have discussed justify the critical importance of paying attention to kidney function and disease in the care of patients who have had COVID-19. However, neither in the clinical references of post-COVID syndrome nor in the consensus between societies is recognition granted to the importance of renal impairment post-COVID. One example is the Clinical Guide for LC patient care, in which 47 Spanish Societies and scientific entities participate and the Spanish Society of Nephrology is absent. It does not even include elemental urinalysis with sediment and detection of albuminuria. Therefore, it is time to develop an institutional, scientific and social rationale that enables progress to be made with the following objectives: 1. Scientific knowledge of renal abnormalities in patients who suffered from COVID-19. 2. Registry of kidney disease in patients who suffered from COVID-19 with follow-up over time. We recommend semi-annual or annual follow-up, based on initial data, renal function and elemental urine with urinary sediment. 3. Prevention of kidney damage associated with treatments derived from the numerous pathologies linked to post-COVID. 4. Social information and visibility of the importance of monitoring kidney disease, even in patients who, not having had symptoms, are unaware that their kidney function may have decreased due to the infection. 5. Development of lines of research with studies of risk factors, clinical and laboratory expression, renal functional reserve and biomarkers for kidney injury, among others, that improve knowledge of the mechanisms of kidney injury. 6. Development by a working group of the Spanish Society of Nephrology (S.E.N.) of uniform healthcare strategy processes in terms of diagnosis and follow-up of affected patients with a single registry. 7. Considering the aforementioned risks of kidney damage, the immediate incorporation of Nephrology in the multidisciplinary care of post-acute COVID-19 is necessary.
The authors declare that they did not receive any funding.
The authors declare that they have no conflicts of interest
|
Characterizing brain dynamics during ketamine-induced dissociation and subsequent interactions with propofol using human intracranial neurophysiology | 0145d0a7-efc5-42bc-a1d3-20481c40a8fa | 10060225 | Physiology[mh] | Ketamine is a dissociative anesthetic that has both anesthetic and psychoactive properties , . Intravenous induction doses (1–2 mg/kg) of ketamine result in a rapid loss of consciousness appropriate for general anesthesia , . At subanesthetic doses (0.5 mg/kg), ketamine produces a dissociative state, which includes gaps in memory, out-of-body experiences, and altered sensory perception – . In addition, intravenous administration of a subanesthetic dose of ketamine induces significant and rapid antidepressant-like response in depressed patients . Although ketamine was approved by the Food and Drug Administration (FDA) for adult patients with treatment-resistant depression , the neuropsychiatric side effects have limited its extensive use in clinical practice , . Defining the neural circuits engaged in ketamine’s rapid antidepressant and dissociative effects is an important priority that could facilitate the development of improved therapies with fewer side effects and greater safety. Ketamine is known to induce profound changes in brain oscillatory dynamics that appear to be correlated with its antidepressant and sensory dissociative activity , – . The electrophysiologic profile of subanesthetic ketamine in humans generally includes an increase of gamma oscillation power and a decrease of delta, alpha, and beta oscillation power , – . Oscillatory power changes have also been reported in patients with depression and have been used to differentiate depressive from healthy subjects . However, the relationships between these changes in oscillatory power and the neural circuit mechanisms of depression and dissociation are not well-understood. Previous studies suggest that at subanesthetic doses, ketamine preferentially blocks the NMDA receptors on GABAergic inhibitory interneurons, resulting in the disinhibition of downstream excitatory pyramidal neurons that is thought to facilitate increased gamma-band activity – . When GABA A agonists, such as benzodiazepines, are administered alongside ketamine, they mitigate dissociations, possibly by restoring inhibitory activity in the affected brain regions , . In addition, ketamine inhibits the hyperpolarization-activated cyclic nucleotide-gated potassium channel 1 (HCN1), a molecular target that is thought to play an important role in generating rhythmic EEG activity and is considered a novel therapeutic target for depressive disorders – . Studies have been conducted to investigate which cortical or subcortical structures play a major role in mediating this process. Previous work has showed that ketamine’s antidepressant effects are largely dependent upon its actions within the prefrontal cortex and the hippocampus . On the other hand, the reduction of alpha oscillations in the precuneus and temporal-parietal junction and the 3 Hz rhythm in the deep posteromedial cortex (PMC), as studied in rodents and a human patient, have been proposed as mechanisms for ketamine-induced dissociation , , . Functional connectivity analysis with fMRI and EEG suggest that ketamine disrupts the frontoparietal default mode network connectivity , . Although ketamine’s antidepressive and dissociative effects are known to co-occur whenever the drug is administered, these effects may in fact be mediated by distinct mechanisms within distinct neural circuits. If that were true, it might be possible to design novel therapeutics with greater specificity and fewer side effects. In this study, we measured intracranial EEG (iEEG) in human patients implanted with intracranial electrodes who were administered a subanesthetic dose of ketamine prior to induction of general anesthesia with propofol for electrode removal surgery. Our goal was to characterize the brain regions involved in different ketamine-induced rhythms in order to better understand their potential role in mediating ketamine’s dissociative and antidepressant properties. In addition to characterizing changes in canonical frequency bands associated with subanesthetic ketamine, we also looked for evidence of a 3 Hz rhythm recently implicated in ketamine-induced dissociation . To characterize the potential role of NMDA and HCN1 receptors in producing ketamine-induced oscillations, we analyzed the interactions between subanesthetic ketamine and propofol. Propofol is a positive GABA allosteric modulator and HCN1 blocker , . Propofol’s GABAergic activity would be expected to antagonize any ketamine-induced oscillations stemming from NMDA-mediated disinhibition. At the same time, propofol would be expected to further potentiate any ketamine-induced oscillations originating from HCN1 inhibition. We collected data from 10 epilepsy patients implanted with intracranial depth electrodes to identify sites of epileptogenic origin (Table , Supplementary Fig. and Supplementary ). The responses on the abbreviated Clinician-Administered Dissociative States Scale (CADSS) – questionnaire (Supplementary Fig. ) are summarized in Supplementary Table . The responses on the questionnaire are consistent with a dissociative state induced by subanesthetic ketamine. Ketamine and propofol-induced location- and frequency-dependent iEEG dynamics We observed distinct dynamic patterns in the iEEG after ketamine infusion, which changed after the administration of propofol. Figure shows the spectrogram and power spectra for 3 channels in the inferior frontal, middle temporal, and occipital cortices from an example subject. The spectrograms for other subjects are shown in Supplementary Fig. . Under ketamine, we observed increased gamma power (25–55 Hz) in the inferior frontal channel and decreased alpha power (8–15 Hz) in the middle temporal and occipital channels. After propofol was added, there was a large increase of power in the inferior frontal and middle temporal channels for nearly all frequencies, except for upper gamma band (40–55 Hz). In contrast, the reduction of alpha oscillations in the occipital channels was further enhanced with the addition of propofol. These results suggest that the iEEG dynamics induced by ketamine and propofol are location- and frequency-dependent. To understand how these brain dynamics mapped to different brain structures, we analyzed the changes in power for different cortical and subcortical structures, first after ketamine infusion and then after the addition of propofol. Ketamine induced an increase in gamma oscillation power and a reduction of low-frequency oscillation power We analyzed the changes in iEEG dynamics for different brain structures after ketamine infusion (Fig. , Supplementary Table and Supplementary Fig. ). For gamma frequencies (25-55 Hz), a greater than 100 dB increase in mean power after ketamine infusion compared with baseline was detected in frontal structures, which include the anterior and posterior cingulate (159.04 dB), superior frontal (153.03 dB), middle frontal (153.59 dB), orbitofrontal (133.68 dB), and inferior frontal (149.20 dB) areas. The mean power increase in precentral, postcentral, isthmus cingulate, temporal structures, lingual, pericalcarine, hippocampus, amygdala, striatum, and insula, was between 19.07 and 96.37 dB. A decrease in mean gamma power was detected in occipital channels (−42.96 dB). For beta frequencies (15-25 Hz), while an increase in power was detected in hippocampus and amygdala (4.43 dB), a decrease in power was detected for middle frontal (−14.26 dB), precentral (−18.21 dB), postcentral (−36.00 dB), isthmus cingulate (−10.53 dB), parietal (−15.90 dB) and temporal structures (−8.40 dB), as well as the lingual and pericalcarine (−19.58 dB), and the occipital cortices (−43.81 dB). No other structural labels showed changes in power after ketamine infusion (i.e., confidence intervals overlapped zero). For alpha frequencies (8-15 Hz), the decrease of mean alpha power was observed for nearly all structure labels with the largest reduction in postcentral (−33.55 dB) and occipital cortices (−32.07 dB). For theta rhythms (4-8 Hz), we identified an increase of power in insula (3.88 dB) cortex and decrease of power in superior frontal (−5.65 dB), precentral (−9.96 dB), postcentral (−4.50 dB), parietal (−7.75 dB) and temporal structures (−5.83 dB), lingual and pericalcarine (−11.68 dB), as well as the occipital cortices (−20.52 dB) and striatum (−1.85 dB). For slow (0.1-1 Hz) and delta frequencies (1-4 Hz), the decrease in power was observed in most of the structural labels (slow: −1.51 to −3.51 dB, delta: −1.40 to −12.91 dB), except for orbitofrontal, isthmus cingulate, striatum, and insula cortex, which did not showed changes in power after ketamine infusion. Propofol reversed the gamma band iEEG dynamics induced by ketamine in frontal regions and caused a further reduction of occipital alpha oscillation power Adding the propofol (Fig. , Supplementary Table and Supplementary Fig. ) reversed the gamma power (40–55 Hz) increase in anterior and posterior cingulate (−61.26 dB), superior frontal (−68.32 dB), middle frontal (−134.51 dB), orbitofrontal (−52.27 dB), and inferior frontal (−61.86 dB) regions of the brain, as well as the gamma power decrease in the occipital cortex (18.85 dB). In addition, propofol further intensified the gamma power increase at precentral (49.39 dB), postcentral (67.34 dB), isthmus cingulate (9.78 dB), hippocampus and amygdala (20.56 dB). The presence of propofol reversed the alpha power (8-15 Hz) decrease induced by ketamine for most of the structural labels (33.33 to 158.32 dB) except for occipital cortices (−35.20 dB), which showed a further reduction in power after propofol administration. In addition, propofol increased the beta power (15-25 Hz) for nearly all structural labels (22.00 to 214.90 dB). For theta rhythms (4-8 Hz), propofol also increased theta power in most of the structural labels (17.35 to 68.88 dB). The addition of propofol reversed the power reduction induced by ketamine at slow (0.1-1 Hz, 7.31 to 38.66 dB) and delta (1-4 Hz, 10.29 to 92.64 dB) oscillations for all the structural labels. Subanesthetic doses of ketamine induced an increase of 3 Hz oscillation in posteromedial cortex (PMC) We studied the spatial distribution of 3 Hz rhythms after the administration of ketamine and propofol (Fig. , Supplementary Table and Supplementary Fig. ). We identified a dramatic increase of 3-4 Hz oscillatory power after ketamine infusion in posterior (2.05 dB) and isthmus (1.00 dB) cingulate cortex, which are part of the PMC, as well as the pars opercularis (2.90 dB) located within the inferior frontal cortex (Fig. ). We then analyzed the spectrum of the oscillatory activity within PMC by plotting the power differences after ketamine relative to baseline for posterior and isthmus cingulate cortex as a function of the frequency (Fig. ). We found that the increase of iEEG power after ketamine peaked between 3 to 6 Hz. The addition of propofol greatly increased the 3-4 Hz power in most brain regions (6.34 to 24.57 dB), including the posterior and isthmus cingulate cortex, suggesting that the effects of ketamine and propofol on this 3-4 Hz rhythm may be additive rather than antagonistic (Fig. ). We observed distinct dynamic patterns in the iEEG after ketamine infusion, which changed after the administration of propofol. Figure shows the spectrogram and power spectra for 3 channels in the inferior frontal, middle temporal, and occipital cortices from an example subject. The spectrograms for other subjects are shown in Supplementary Fig. . Under ketamine, we observed increased gamma power (25–55 Hz) in the inferior frontal channel and decreased alpha power (8–15 Hz) in the middle temporal and occipital channels. After propofol was added, there was a large increase of power in the inferior frontal and middle temporal channels for nearly all frequencies, except for upper gamma band (40–55 Hz). In contrast, the reduction of alpha oscillations in the occipital channels was further enhanced with the addition of propofol. These results suggest that the iEEG dynamics induced by ketamine and propofol are location- and frequency-dependent. To understand how these brain dynamics mapped to different brain structures, we analyzed the changes in power for different cortical and subcortical structures, first after ketamine infusion and then after the addition of propofol. We analyzed the changes in iEEG dynamics for different brain structures after ketamine infusion (Fig. , Supplementary Table and Supplementary Fig. ). For gamma frequencies (25-55 Hz), a greater than 100 dB increase in mean power after ketamine infusion compared with baseline was detected in frontal structures, which include the anterior and posterior cingulate (159.04 dB), superior frontal (153.03 dB), middle frontal (153.59 dB), orbitofrontal (133.68 dB), and inferior frontal (149.20 dB) areas. The mean power increase in precentral, postcentral, isthmus cingulate, temporal structures, lingual, pericalcarine, hippocampus, amygdala, striatum, and insula, was between 19.07 and 96.37 dB. A decrease in mean gamma power was detected in occipital channels (−42.96 dB). For beta frequencies (15-25 Hz), while an increase in power was detected in hippocampus and amygdala (4.43 dB), a decrease in power was detected for middle frontal (−14.26 dB), precentral (−18.21 dB), postcentral (−36.00 dB), isthmus cingulate (−10.53 dB), parietal (−15.90 dB) and temporal structures (−8.40 dB), as well as the lingual and pericalcarine (−19.58 dB), and the occipital cortices (−43.81 dB). No other structural labels showed changes in power after ketamine infusion (i.e., confidence intervals overlapped zero). For alpha frequencies (8-15 Hz), the decrease of mean alpha power was observed for nearly all structure labels with the largest reduction in postcentral (−33.55 dB) and occipital cortices (−32.07 dB). For theta rhythms (4-8 Hz), we identified an increase of power in insula (3.88 dB) cortex and decrease of power in superior frontal (−5.65 dB), precentral (−9.96 dB), postcentral (−4.50 dB), parietal (−7.75 dB) and temporal structures (−5.83 dB), lingual and pericalcarine (−11.68 dB), as well as the occipital cortices (−20.52 dB) and striatum (−1.85 dB). For slow (0.1-1 Hz) and delta frequencies (1-4 Hz), the decrease in power was observed in most of the structural labels (slow: −1.51 to −3.51 dB, delta: −1.40 to −12.91 dB), except for orbitofrontal, isthmus cingulate, striatum, and insula cortex, which did not showed changes in power after ketamine infusion. Adding the propofol (Fig. , Supplementary Table and Supplementary Fig. ) reversed the gamma power (40–55 Hz) increase in anterior and posterior cingulate (−61.26 dB), superior frontal (−68.32 dB), middle frontal (−134.51 dB), orbitofrontal (−52.27 dB), and inferior frontal (−61.86 dB) regions of the brain, as well as the gamma power decrease in the occipital cortex (18.85 dB). In addition, propofol further intensified the gamma power increase at precentral (49.39 dB), postcentral (67.34 dB), isthmus cingulate (9.78 dB), hippocampus and amygdala (20.56 dB). The presence of propofol reversed the alpha power (8-15 Hz) decrease induced by ketamine for most of the structural labels (33.33 to 158.32 dB) except for occipital cortices (−35.20 dB), which showed a further reduction in power after propofol administration. In addition, propofol increased the beta power (15-25 Hz) for nearly all structural labels (22.00 to 214.90 dB). For theta rhythms (4-8 Hz), propofol also increased theta power in most of the structural labels (17.35 to 68.88 dB). The addition of propofol reversed the power reduction induced by ketamine at slow (0.1-1 Hz, 7.31 to 38.66 dB) and delta (1-4 Hz, 10.29 to 92.64 dB) oscillations for all the structural labels. We studied the spatial distribution of 3 Hz rhythms after the administration of ketamine and propofol (Fig. , Supplementary Table and Supplementary Fig. ). We identified a dramatic increase of 3-4 Hz oscillatory power after ketamine infusion in posterior (2.05 dB) and isthmus (1.00 dB) cingulate cortex, which are part of the PMC, as well as the pars opercularis (2.90 dB) located within the inferior frontal cortex (Fig. ). We then analyzed the spectrum of the oscillatory activity within PMC by plotting the power differences after ketamine relative to baseline for posterior and isthmus cingulate cortex as a function of the frequency (Fig. ). We found that the increase of iEEG power after ketamine peaked between 3 to 6 Hz. The addition of propofol greatly increased the 3-4 Hz power in most brain regions (6.34 to 24.57 dB), including the posterior and isthmus cingulate cortex, suggesting that the effects of ketamine and propofol on this 3-4 Hz rhythm may be additive rather than antagonistic (Fig. ). In this study, we show, in humans, a detailed description of the principal oscillatory changes in cortical and subcortical structures after the administration of a subanesthetic dose of ketamine. Using intraoperative recordings from intracranial electrodes in 10 patients with epilepsy, we found that ketamine increased gamma oscillations within prefrontal cortical areas and the hippocampus—structures previously implicated in ketamine’s antidepressant effects . Furthermore, our studies provide direct evidence of a ketamine-induced 3 Hz oscillation in posteromedial cortex that has been proposed as a mechanism for its dissociative effects . By analyzing changes in neural oscillations after the addition of propofol in 7 out of 10 subjects, we were also able to identify putative NMDA-mediated brain dynamics that could be antagonized by propofol’s GABAergic activity, as well as possible HCN1-mediated effects where both drugs showed an additive effect. Overall, our results suggest that ketamine engages different neural circuits in distinct frequency-dependent patterns of activity to produce its antidepressant and dissociative sensory effects. These insights may help guide the development of brain dynamic biomarkers and novel therapeutics for depression. For gamma frequencies (25–55 Hz), we observed a remarkable increase in power in frontal and limbic structures that are consistent with previous reports employing non-invasive EEG in humans under both subanesthetic and anesthetic doses of ketamine , – , . We found that the gamma band activity was reversed after the subsequent addition of propofol in prefrontal cortical structures. We propose that the ketamine-induced gamma power increase and its subsequent reversal by propofol could be explained by an antagonist mechanism (Fig. , top panel ). Ketamine preferentially blocks the NMDA receptors on GABAergic inhibitory interneurons, resulting in disinhibition of the downstream excitatory pyramidal neurons, which mediates the increased gamma-band activity – . When propofol, a GABA agonist, is administered alongside ketamine, it antagonizes the gamma power increase by restoring some of the inhibitory activity in the prefrontal cortex. The increase in gamma spectral power anteriorly following subanesthetic ketamine infusion may reflect a shift of brain activity from a globally balanced state to a disorganized and autonomous state . The changes in gamma band activity in sensory cortices may contribute to the discoordination of higher-order functional networks and perceptual distortions produced by subanesthetic doses of ketamine , , . In contrast, for alpha frequencies (8–15 Hz), we detected a large reduction in iEEG power after ketamine infusion for all brain regions studied, with the largest reductions occurring in posterior sensory cortices. When propofol was subsequently administered, the reduction in alpha power was reversed in most brain regions, suggesting a similar NMDA-dependent mechanism as described above for gamma activity. However, in posterior sensory structures (lingual, pericalcarine and occipital cortices), the addition of propofol further attenuated alpha power. We attribute this additive behavior to ketamine and propofol’s shared inhibition of HCN1 channels (Fig. , middle panel ). HCN1 channels have been identified as an important molecular target for ketamine’s action . Knockout of HCN1 channels abolishes the ketamine-induced loss-of-right reflex, a behavioral correlate of unconsciousness in rodents . Propofol also inhibits HCN1 channels and the HCN1 knock-out mice are known to be less sensitive to unconsciousness due to propofol . Modeling studies suggest that reductions in hyperpolarization-activated cationic current ( Ih ) mediated by HCN1 can abolish occipital alpha rhythms by silencing thalamocortical cells . The reduction of alpha power in occipital regions is also observed during anesthetic doses of ketamine , , propofol-induced unconsciousness , as well as sleep , , suggesting the loss of occipital alpha rhythms may be a hallmark for disrupted sensory processing in different states of altered arousal . We found that subanesthetic doses of ketamine induced a 3 Hz oscillation in PMC in humans, consistent with previous studies in mice after administration of ketamine and in an epileptic patient during a pre-seizure aura as well as in response to electrical stimulation of epileptic foci . Vesuna, et al., 2020, showed that there are NMDA receptors and HCN1 channels in the homologous deep retrosplenial (RSP) cortex in mice, both of which are required for generating the observed 3 Hz rhythmic activity . Knockout of HCN1 channels abolished ketamine-induced rhythms in RSP and the dissociation-related behavior in mice, whereas optogenetic inhibition of long-range inputs to the RSP enhanced ketamine-induced oscillations . Vesuna et al., proposed that ketamine blockade of NMDA receptors could hyperpolarize membrane potentials in PMC, activating intrinsic HCN1 channels and permitting rhythmic dynamics. We propose that the same effect could occur with propofol by way of a GABA-mediated hyperpolarization (Fig. , bottom panel ). Although both ketamine and propofol-induced 3 Hz rhythms in PMC, dissociation was only detected after ketamine. This may be because propofol suppresses arousal and induces unconsciousness, which would supersede any perceived dissociative effects. Besides its dissociative effects, subanesthetic ketamine has been shown to have a powerful antidepressant effect. The oscillatory circuit dynamics produced by ketamine may be related to this antidepressant effect. Subjects with a history of depression have been observed to have higher amplitude delta and theta oscillations compared to controls during a working memory task . Consistent with this observation, we found that ketamine reduces delta and theta oscillation power. Patients with depression have also been reported to have increased activity in alpha, beta, and theta bands at the occipital and parietal regions of the brain . Accordingly, we identified a global reduction of power at theta, alpha and beta frequencies, with the largest reduction in occipital and parietal regions after ketamine infusion. Gamma oscillations have also been discussed as a potential biomarker for depression. Changes in gamma rhythms can vary according to behavioral states and task conditions, but there are a few studies suggesting that reduced gamma power is associated with depression. One EEG study found that subjects with high depression scores had reduced resting gamma power in the anterior cingulate cortex . Another MEG study showed that depressed subjects with lower baseline gamma and higher ketamine-induced gamma had a better response to ketamine than those with higher baseline gamma . It has also known that the prefrontal cortex and hippocampus are implicated in ketamine’s antidepressant response . The dramatic increase in gamma rhythms we identified in those brain regions with subanesthetic doses of ketamine are consistent with previous studies. In this study, although we did not directly measure clinical depression nor antidepressant effects, we inferred that our results could be related to ketamine’s antidepressant effects, based on the neuroanatomy of the brain oscillations we identified and prior literature that showed associations among depression, brain dynamics, and functional neuroanatomy. Future studies investigating brain dynamics after ketamine infusion in depressed patients are needed. In this study, we focused primarily on the role of NMDA receptors, which appear to play a central role in mediating ketamine’s effects on brain dynamics as well as its antidepressant effects . The role of other receptors such as AMPA , that have been suggested to play an important role in ketamine’s antidepressant effects should also be investigated in the future. In follow-up studies it would also be interesting to explore the relationship between EEG oscillatory dynamics and the intensity level of dissociation, which could not be addressed in the current study due to our limited sample size and the limited resolution of dissociation assessment. Cross-frequency coupling analysis could be an additional topic of interest for characterizing the interactions between oscillations at different frequency bands. Our results also show how the combination of ketamine and propofol could contribute to unconsciousness through a shared mechanism, providing an explanation for why propofol and ketamine appear to work synergistically to maintain unconsciousness when administered during general anesthesia . Overall, we find that ketamine has distinct dynamic effects on neural systems known to mediate cognition, depression, and sensory processing by way of multiple dissociable neuropharmacological mechanisms. The neural circuit mechanisms underlying ketamine-induced oscillatory dynamics, and their potential links to antidepressive and dissociative effects as proposed in this study, may have important implications for the development of novel therapies with fewer side effects and greater safety. Subject recruitment Patients with medication-refractory epilepsy implanted with intracranial depth electrodes to locate their seizure onset zone were recruited from Massachusetts General Hospital and Brigham and Women’s Hospital. Electrode placement was determined by the clinical team independent of this study. Ten patients (five male and five female) aged 22 to 59 years old were recruited. Subjects’ demographic and electrode information are summarized in Table . This study was approved by the Institutional Review Board (IRB) covering the two hospitals (Mass General Brigham Human Research Committee). Informed consent was obtained from all subjects prior to the study. Experimental procedure All experiments were conducted during stereotactic neurosurgery for removal of the intracranial depth electrodes in the operating room at the Massachusetts General Hospital or the Brigham and Women’s Hospital. Participants were implanted with multi-lead depth electrodes (a.k.a. stereotactic EEG, sEEG) to confirm the hypothesized seizure focus, and located epileptogenic tissue in relation to essential cortex, thus directing surgical treatment. Depth electrodes (Ad-tech Medical, Racine WI, USA, or PMT, Chanhassen, MN, USA) with diameters of 0.8–1.0 mm and consisting of 8–16 platinum/iridium-contacts 1–2.4 mm long were stereotactically placed in locations deemed necessary for seizure localization by a multidisciplinary clinical team. The first period was a baseline recording of 5 min (Fig. ). The second period consisted of 14 min with continuous infusion of subanesthetic level of ketamine (total dose of 0.5 mg/kg over 14 min, Supplementary Fig. shows pharmacokinetic effects of different ketamine delivery schemes). At the end of ketamine infusion, a clinical research staff member administered the abbreviated CADSS questionnaire (Supplementary Fig. ) to the patients – . Because of limited time in the operating room, patients only answered yes or no to the questions. Immediately after the questionnaire, propofol bolus was given to the patients to induce general anesthesia. During the whole process, subjects were instructed to close their eyes to avoid eye-blink artifacts in the signal. Supplementary Fig. shows oxygen saturation (SpO 2 ), mean arterial pressure (MAP), pulse, and end-tidal CO 2 for the study period. iEEG signals were recorded using a Blackrock Cerebus system (Blackrock Microsystems) sampled at 2,000 Hz. Before each study, structural MRI scans were acquired for each subject (Siemens Trio 3 Tesla, T1-weighted magnetization-prepared rapid gradient echo, 1.3-mm slice thickness, 1.3 × 1 mm in-plane resolution, TR/TE = 2530/3.3 ms, 7° flip angle). iEEG preprocessing, power spectral analysis and statistical analysis Data analysis was performed using custom analysis code in MATLAB (R2021a). Raw iEEG data were notch filtered at 60 Hz and its harmonics, downsampled to 500 Hz, and detrended across the entire recording. The signals were then visually inspected, and channels with noise or artifacts were removed. Data were re-referenced with a bipolar montage. A total of 824 bipolar channels were generated for 10 subjects received ketamine, and 606 bipolar channels were generated for 7 subjects received propofol (Supplementary Fig. and Supplementary ). Spectral analysis was performed using the multitaper method, with window lengths of T = 2 sec with 0.5 sec overlap, time-bandwidth product TW = 3, number of tapers K = 5, and spectral resolution of 3 Hz , . The mean power spectral density for baseline, ketamine and propofol conditions were calculated by taking the average across each period. The power spectral density was converted to decibels (dB) to facilitate easier comparisons. The differences of power after ketamine infusion relative to baseline, and propofol relative to ketamine periods were calculated by subtracting the mean power spectral density in dB between each of the two conditions at different frequencies (slow: 0.1–1 Hz, delta: 1–4 Hz, theta: 4–8 Hz, alpha: 8–15 Hz, beta: 15–25 Hz, gamma: 25–55 Hz, low gamma: 25–40 Hz, upper gamma: 40–55 Hz). Our primary objective was to describe changes in iEEG power by reporting effect sizes and confidence intervals for changes in iEEG power in the indicated brain regions of interest (ROIs) after drug administration. We did not report p-values and thus did not correct for multiple comparisons. The bootstrap method was used to generate the 95% confidence interval around the mean differences in power for each structural label at each frequency using data from all subjects who had electrodes located within each structural label. The upper and lower bars represent the bootstrapped 95% confidence interval bounds. Structural parcellation of the brain The electrode positions in each subject’s brain were obtained by aligning the preoperative T1-weighted MRI with a postoperative CT/MRI using the Freesurfer (7.2) image analysis tool , . To identify the structural label and functional network for each of the electrodes, an electrode labeling algorithm (ELA) was employed . This algorithm estimated the probability of overlap of an expanding area around each electrode with brain structural labels that had been identified in the Desikan-Killiany-Tourville (DKT) 40 atlas using purely anatomical approaches – . Then the ELA used gradient descent to find the closest voxel in the template’s brain that gives similar regions and probabilities to transform the patients’ electrode coordinates to the template brain – . Based on DKT 40 atlas, we assigned the 824 electrodes from 10 subjects received ketamine to 49 structural labels, which were then further classified into 15 labels according to the anatomical locations and the mean differences of power after ketamine relative to the baseline condition. Likewise, we assigned the 606 electrodes collected from 7 subjects received propofol to 14 structural labels. We plotted all electrodes on Colin 27 template brain with colors per parcellated brain region indicating the differences in power for the ketamine infusion period relative to baseline, as well as for propofol bolus relative to the ketamine infusion period for each of the frequencies. Reporting summary Further information on research design is available in the linked to this article. Patients with medication-refractory epilepsy implanted with intracranial depth electrodes to locate their seizure onset zone were recruited from Massachusetts General Hospital and Brigham and Women’s Hospital. Electrode placement was determined by the clinical team independent of this study. Ten patients (five male and five female) aged 22 to 59 years old were recruited. Subjects’ demographic and electrode information are summarized in Table . This study was approved by the Institutional Review Board (IRB) covering the two hospitals (Mass General Brigham Human Research Committee). Informed consent was obtained from all subjects prior to the study. All experiments were conducted during stereotactic neurosurgery for removal of the intracranial depth electrodes in the operating room at the Massachusetts General Hospital or the Brigham and Women’s Hospital. Participants were implanted with multi-lead depth electrodes (a.k.a. stereotactic EEG, sEEG) to confirm the hypothesized seizure focus, and located epileptogenic tissue in relation to essential cortex, thus directing surgical treatment. Depth electrodes (Ad-tech Medical, Racine WI, USA, or PMT, Chanhassen, MN, USA) with diameters of 0.8–1.0 mm and consisting of 8–16 platinum/iridium-contacts 1–2.4 mm long were stereotactically placed in locations deemed necessary for seizure localization by a multidisciplinary clinical team. The first period was a baseline recording of 5 min (Fig. ). The second period consisted of 14 min with continuous infusion of subanesthetic level of ketamine (total dose of 0.5 mg/kg over 14 min, Supplementary Fig. shows pharmacokinetic effects of different ketamine delivery schemes). At the end of ketamine infusion, a clinical research staff member administered the abbreviated CADSS questionnaire (Supplementary Fig. ) to the patients – . Because of limited time in the operating room, patients only answered yes or no to the questions. Immediately after the questionnaire, propofol bolus was given to the patients to induce general anesthesia. During the whole process, subjects were instructed to close their eyes to avoid eye-blink artifacts in the signal. Supplementary Fig. shows oxygen saturation (SpO 2 ), mean arterial pressure (MAP), pulse, and end-tidal CO 2 for the study period. iEEG signals were recorded using a Blackrock Cerebus system (Blackrock Microsystems) sampled at 2,000 Hz. Before each study, structural MRI scans were acquired for each subject (Siemens Trio 3 Tesla, T1-weighted magnetization-prepared rapid gradient echo, 1.3-mm slice thickness, 1.3 × 1 mm in-plane resolution, TR/TE = 2530/3.3 ms, 7° flip angle). Data analysis was performed using custom analysis code in MATLAB (R2021a). Raw iEEG data were notch filtered at 60 Hz and its harmonics, downsampled to 500 Hz, and detrended across the entire recording. The signals were then visually inspected, and channels with noise or artifacts were removed. Data were re-referenced with a bipolar montage. A total of 824 bipolar channels were generated for 10 subjects received ketamine, and 606 bipolar channels were generated for 7 subjects received propofol (Supplementary Fig. and Supplementary ). Spectral analysis was performed using the multitaper method, with window lengths of T = 2 sec with 0.5 sec overlap, time-bandwidth product TW = 3, number of tapers K = 5, and spectral resolution of 3 Hz , . The mean power spectral density for baseline, ketamine and propofol conditions were calculated by taking the average across each period. The power spectral density was converted to decibels (dB) to facilitate easier comparisons. The differences of power after ketamine infusion relative to baseline, and propofol relative to ketamine periods were calculated by subtracting the mean power spectral density in dB between each of the two conditions at different frequencies (slow: 0.1–1 Hz, delta: 1–4 Hz, theta: 4–8 Hz, alpha: 8–15 Hz, beta: 15–25 Hz, gamma: 25–55 Hz, low gamma: 25–40 Hz, upper gamma: 40–55 Hz). Our primary objective was to describe changes in iEEG power by reporting effect sizes and confidence intervals for changes in iEEG power in the indicated brain regions of interest (ROIs) after drug administration. We did not report p-values and thus did not correct for multiple comparisons. The bootstrap method was used to generate the 95% confidence interval around the mean differences in power for each structural label at each frequency using data from all subjects who had electrodes located within each structural label. The upper and lower bars represent the bootstrapped 95% confidence interval bounds. The electrode positions in each subject’s brain were obtained by aligning the preoperative T1-weighted MRI with a postoperative CT/MRI using the Freesurfer (7.2) image analysis tool , . To identify the structural label and functional network for each of the electrodes, an electrode labeling algorithm (ELA) was employed . This algorithm estimated the probability of overlap of an expanding area around each electrode with brain structural labels that had been identified in the Desikan-Killiany-Tourville (DKT) 40 atlas using purely anatomical approaches – . Then the ELA used gradient descent to find the closest voxel in the template’s brain that gives similar regions and probabilities to transform the patients’ electrode coordinates to the template brain – . Based on DKT 40 atlas, we assigned the 824 electrodes from 10 subjects received ketamine to 49 structural labels, which were then further classified into 15 labels according to the anatomical locations and the mean differences of power after ketamine relative to the baseline condition. Likewise, we assigned the 606 electrodes collected from 7 subjects received propofol to 14 structural labels. We plotted all electrodes on Colin 27 template brain with colors per parcellated brain region indicating the differences in power for the ketamine infusion period relative to baseline, as well as for propofol bolus relative to the ketamine infusion period for each of the frequencies. Further information on research design is available in the linked to this article. Supplementary Information Peer Review File Description of Additional Supplementary Files Supplementary Movie 1 Reporting Summary |
Stress levels of a group of dentists while providing dental care under clinical, deep sedation, and general anesthesia | de1b680f-e6ec-4eb1-809d-281d110efe48 | 10060907 | Dental[mh] | Stress is commonly defined as occurring when an individual’s high expectations force their capacity to adapt . Occupational stress, defined as when the resources of an individual are not sufficient to cope with the needs of a situation, is a leading modern health and safety challenge . Health sector professionals have higher stress than other professionals, and it is usually caused by workload and patient‒doctor relationships [ – ]. Dentistry professions constantly require a high level of skill and attention. Especially in pediatric dentistry, behavior management problems in children, parental expectations, and parental behaviors can be more stressful and exhausting in practice [ – ]. General dental practitioners report that they are very stressed due to anxious children and their behaviors that prevent clinical procedures . Additionally, parents’ inability to understand the difficulties of treating anxious children and their high expectations are among the reasons that increase the stress of dental practitioners even more . A child’s level of cooperation and general behavior is critical for a dentist to choose the most suitable behavioral management approaches, such as tell-show-do, sedation, and general anesthesia . There are some treatment complications with both deep sedation and general anesthesia. Ensuring and maintaining airway patency is vital in the application of sedation. In the application of anesthesia, deep sedation is challenging in dental procedures due to the anatomical proximity of the surgical area to the airway and the risk of microaspiration of water, blood, saliva, and small particles of filling material when working in the open mouth . Dental treatment with general anesthesia is seen as a stressful situation for practitioners since some complications, such as neurological damage, cardiac and respiratory arrest, and even death, may occur . In response to any stress factors in humans, two biological systems are activated: the sympathetic nervous system in the period immediately after exposure to the factor and then the hypothalamic‒pituitary‒adrenal system . In healthy individuals, stimulation of the sympathetic nervous system at the beginning of the stress response begins with the secretion of epinephrine and norepinephrine from the adrenal medulla. These catecholamines cause the characteristic features of sympathetic nervous system activity, such as an increase in heart rate, mydriasis in the pupils, and acceleration of breathing. For this reason, sympathetic activity is measured by various evaluation methods, such as heart rate, blood pressure, and O 2 saturation . On the other hand, salivary cortisol has been accepted as a reliable biomarker of the hypothalamus–pituitary–adrenal system as a delayed stress response . A literature review showed that very few studies have investigated occupational stress related to pediatric dentistry . Furthermore, no study was found in which the stress situation in the three treatment protocols (clinical sedation, deep sedation, general anesthesia) was evaluated and compared. In the present study, we aimed to evaluate the stress experienced by dentists while treating children in all three treatment protocols by using objective and subjective (Dentists’ Stress Questionnaire) findings . The null hypothesis (H0) of this study was that there is a statistically significant difference between the stress levels of dentists while treating children with the different treatment protocols.
Study design This study was carried out in the University Pediatric Dentistry Department, Oral and Maxillofacial Surgery Operating Room, and Medicine Faculty, Biochemistry Department. According to the power analysis, the estimated number of samples was determined to be 9 patients in each group ( α = 0.05 and 1- β = 0.80). Each dentist treated 27 patients with 3 different treatment approaches, and the study resulted in a total of 108 patients. Inclusion and exclusion criteria Dentist standardization - Four dentists, who started their specialization training simultaneously at the Erciyes University Faculty of Dentistry, Department of Pediatric Dentistry, with equal clinical experience and training. - Women - Between the ages of 30 and 33 years - Weight of 52–55 kg - No systemic disease - Dentists who had actively treated patients for at least 2 years in a clinic with deep sedation and general anesthesia were included. Patient standardization Children with positive or definitely positive (Frankl 3, 4) behavior according to the Frankl Scale , with the data observed in the first session, were included in the clinical treatment group. According to the clinical examination of these patients, 36 healthy children aged 5–6 years who did not require pulpal treatment and whose caries level was 1–4 according to the ICDAS (International Caries Detection and Assessment System) were selected. The type of treatment was determined as compomer fillings applied to 2 primary molars after local anesthesia, and the duration of the treatment was limited to 30–60 min. Children aged 48–72 months and children with negative or absolutely negative behavior (Frankl 1, 2) according to the Frankl Scale were provided treatment under deep sedation. To provide standardization among patients suitable for sedation, 36 children whose dmft (decayed, missing, filled teeth index) score was less than their age and for whom the duration of the procedure was limited to between 30 and 40 min were included in the study. For general anesthesia, 36 healthy children whose dmft score was equal to or higher than their age were included. In addition, patients whose treatment time was limited to 30–60 min were included. The patients were randomly assigned to the dentists, and their treatment was carried out. The study did not include children with general health problems or children whose parents refused treatment with general anesthesia/sedation. For the general anesthesia and deep sedation group, after a minimum of 6 h of fasting, all patients were given midazolam for premedication before they were taken to the operating room. In the operating room, noninvasive standard monitoring was performed for all patients, including heart rate, noninvasive MAP, electrocardiogram, and SpO2. For the general anesthesia group, 2.5 mg/kg propofol, 0.6 mg/kg rocuronium, and 1 μg/kg fentanyl were used to induce the anesthesia. The most appropriate cuffed endotracheal tube was used for the intubation procedure. Sevoflurane (1 MAC) and a 50% oxygen-air mixture were applied to maintain general anesthesia. In the deep sedation group, anesthesia was initiated with propofol at a dose of 2 mg/kg. A nasal mask was applied to all patients in this group. The pressure mod of ventilatör was used for noninvasive ventilation during the deep sedation procedure. Additional intermittent propofol was used to achieve the appropriate sedation depth at which the dental treatment and ventilation could be performed comfortably. Noninvasive mean arterial blood pressure (MAP) and peripheral oxygen saturation (SpO2) were measured by Dräger Fabius Plus (Dräger Medical GmbH, Lübeck, Germany) and recorded 10 min before the dental treatment, at the 25th min of the treatment, and 30 min after the treatment for all three treatment approaches. Also, saliva samples were taken from 4 dentists in the same time intervals for the study. The dentists’ measurement scores compared to each other. The Saliva Swab Sample Collection (SpeciMAX™, Cat. No. A50696, Thermo Fisher Scientific Inc.) kit was used to collect saliva samples. The swab was placed under the tongue for 2 min, and it was ensured that the swab absorbed the saliva. Afterward, the swab was centrifuged and placed in saliva storage tubes with a perforated chamber for separating saliva and the remaining dry swab. Samples were centrifuged at 3000 rpm for 15 min; the saliva was cleared of debris and then poured into the bottom of the storage tube. Then, the saliva in the plastic saliva storage tube was stored at − 80 °C in an upright position until measurements were made . After the saliva samples were thawed at room temperature on the day of the measurements, they were taken to the University Medical Faculty Central Biochemistry Laboratory for analysis. Cortisol measurement protocol Salivary cortisol was measured by the electrochemiluminescence (ECLIA) method using the Cobas Cortisol ll kit (Roche Diagnostics GmbH; Mannheim, Germany). At the end of each patient’s treatment, after the samples were taken and the measurements were made, the dentists were asked to fill out the “Dentist Job Stress Questionnaire,” consisting of 6 questions reflecting their current stress . In the questionnaire, questions were asked to measure the physician’s current degree of work stress, and they were asked to choose the most suitable option for them. Statistics In the study, the analyses were made with the SPSS 25.0 program. Histograms, q–q plots, and Shapiro‒Wilk’s test were applied to assess data normality. Descriptive statistics are presented as the mean and standard deviation. The Kruskal‒Wallis test was performed to examine the differences in measurements according to the dentists and treatment approaches. The Mann‒Whitney U test was used to reveal the evaluation causing the difference. The Friedman test was used to examine the differences among the measurements in the treatment approaches before, during, and after the procedures. The Wilcoxon method was used to determine the time causing the difference. To test the reliability levels of the question groups in the study, the Co. Alpha analysis was applied. Spearman correlation analysis was applied to determine the relationship between stress levels and biological parameters. P values less than 0.05 were considered statistically significant in the study.
This study was carried out in the University Pediatric Dentistry Department, Oral and Maxillofacial Surgery Operating Room, and Medicine Faculty, Biochemistry Department. According to the power analysis, the estimated number of samples was determined to be 9 patients in each group ( α = 0.05 and 1- β = 0.80). Each dentist treated 27 patients with 3 different treatment approaches, and the study resulted in a total of 108 patients.
Dentist standardization - Four dentists, who started their specialization training simultaneously at the Erciyes University Faculty of Dentistry, Department of Pediatric Dentistry, with equal clinical experience and training. - Women - Between the ages of 30 and 33 years - Weight of 52–55 kg - No systemic disease - Dentists who had actively treated patients for at least 2 years in a clinic with deep sedation and general anesthesia were included. Patient standardization Children with positive or definitely positive (Frankl 3, 4) behavior according to the Frankl Scale , with the data observed in the first session, were included in the clinical treatment group. According to the clinical examination of these patients, 36 healthy children aged 5–6 years who did not require pulpal treatment and whose caries level was 1–4 according to the ICDAS (International Caries Detection and Assessment System) were selected. The type of treatment was determined as compomer fillings applied to 2 primary molars after local anesthesia, and the duration of the treatment was limited to 30–60 min. Children aged 48–72 months and children with negative or absolutely negative behavior (Frankl 1, 2) according to the Frankl Scale were provided treatment under deep sedation. To provide standardization among patients suitable for sedation, 36 children whose dmft (decayed, missing, filled teeth index) score was less than their age and for whom the duration of the procedure was limited to between 30 and 40 min were included in the study. For general anesthesia, 36 healthy children whose dmft score was equal to or higher than their age were included. In addition, patients whose treatment time was limited to 30–60 min were included. The patients were randomly assigned to the dentists, and their treatment was carried out. The study did not include children with general health problems or children whose parents refused treatment with general anesthesia/sedation. For the general anesthesia and deep sedation group, after a minimum of 6 h of fasting, all patients were given midazolam for premedication before they were taken to the operating room. In the operating room, noninvasive standard monitoring was performed for all patients, including heart rate, noninvasive MAP, electrocardiogram, and SpO2. For the general anesthesia group, 2.5 mg/kg propofol, 0.6 mg/kg rocuronium, and 1 μg/kg fentanyl were used to induce the anesthesia. The most appropriate cuffed endotracheal tube was used for the intubation procedure. Sevoflurane (1 MAC) and a 50% oxygen-air mixture were applied to maintain general anesthesia. In the deep sedation group, anesthesia was initiated with propofol at a dose of 2 mg/kg. A nasal mask was applied to all patients in this group. The pressure mod of ventilatör was used for noninvasive ventilation during the deep sedation procedure. Additional intermittent propofol was used to achieve the appropriate sedation depth at which the dental treatment and ventilation could be performed comfortably. Noninvasive mean arterial blood pressure (MAP) and peripheral oxygen saturation (SpO2) were measured by Dräger Fabius Plus (Dräger Medical GmbH, Lübeck, Germany) and recorded 10 min before the dental treatment, at the 25th min of the treatment, and 30 min after the treatment for all three treatment approaches. Also, saliva samples were taken from 4 dentists in the same time intervals for the study. The dentists’ measurement scores compared to each other. The Saliva Swab Sample Collection (SpeciMAX™, Cat. No. A50696, Thermo Fisher Scientific Inc.) kit was used to collect saliva samples. The swab was placed under the tongue for 2 min, and it was ensured that the swab absorbed the saliva. Afterward, the swab was centrifuged and placed in saliva storage tubes with a perforated chamber for separating saliva and the remaining dry swab. Samples were centrifuged at 3000 rpm for 15 min; the saliva was cleared of debris and then poured into the bottom of the storage tube. Then, the saliva in the plastic saliva storage tube was stored at − 80 °C in an upright position until measurements were made . After the saliva samples were thawed at room temperature on the day of the measurements, they were taken to the University Medical Faculty Central Biochemistry Laboratory for analysis.
- Four dentists, who started their specialization training simultaneously at the Erciyes University Faculty of Dentistry, Department of Pediatric Dentistry, with equal clinical experience and training. - Women - Between the ages of 30 and 33 years - Weight of 52–55 kg - No systemic disease - Dentists who had actively treated patients for at least 2 years in a clinic with deep sedation and general anesthesia were included.
Children with positive or definitely positive (Frankl 3, 4) behavior according to the Frankl Scale , with the data observed in the first session, were included in the clinical treatment group. According to the clinical examination of these patients, 36 healthy children aged 5–6 years who did not require pulpal treatment and whose caries level was 1–4 according to the ICDAS (International Caries Detection and Assessment System) were selected. The type of treatment was determined as compomer fillings applied to 2 primary molars after local anesthesia, and the duration of the treatment was limited to 30–60 min. Children aged 48–72 months and children with negative or absolutely negative behavior (Frankl 1, 2) according to the Frankl Scale were provided treatment under deep sedation. To provide standardization among patients suitable for sedation, 36 children whose dmft (decayed, missing, filled teeth index) score was less than their age and for whom the duration of the procedure was limited to between 30 and 40 min were included in the study. For general anesthesia, 36 healthy children whose dmft score was equal to or higher than their age were included. In addition, patients whose treatment time was limited to 30–60 min were included. The patients were randomly assigned to the dentists, and their treatment was carried out. The study did not include children with general health problems or children whose parents refused treatment with general anesthesia/sedation. For the general anesthesia and deep sedation group, after a minimum of 6 h of fasting, all patients were given midazolam for premedication before they were taken to the operating room. In the operating room, noninvasive standard monitoring was performed for all patients, including heart rate, noninvasive MAP, electrocardiogram, and SpO2. For the general anesthesia group, 2.5 mg/kg propofol, 0.6 mg/kg rocuronium, and 1 μg/kg fentanyl were used to induce the anesthesia. The most appropriate cuffed endotracheal tube was used for the intubation procedure. Sevoflurane (1 MAC) and a 50% oxygen-air mixture were applied to maintain general anesthesia. In the deep sedation group, anesthesia was initiated with propofol at a dose of 2 mg/kg. A nasal mask was applied to all patients in this group. The pressure mod of ventilatör was used for noninvasive ventilation during the deep sedation procedure. Additional intermittent propofol was used to achieve the appropriate sedation depth at which the dental treatment and ventilation could be performed comfortably. Noninvasive mean arterial blood pressure (MAP) and peripheral oxygen saturation (SpO2) were measured by Dräger Fabius Plus (Dräger Medical GmbH, Lübeck, Germany) and recorded 10 min before the dental treatment, at the 25th min of the treatment, and 30 min after the treatment for all three treatment approaches. Also, saliva samples were taken from 4 dentists in the same time intervals for the study. The dentists’ measurement scores compared to each other. The Saliva Swab Sample Collection (SpeciMAX™, Cat. No. A50696, Thermo Fisher Scientific Inc.) kit was used to collect saliva samples. The swab was placed under the tongue for 2 min, and it was ensured that the swab absorbed the saliva. Afterward, the swab was centrifuged and placed in saliva storage tubes with a perforated chamber for separating saliva and the remaining dry swab. Samples were centrifuged at 3000 rpm for 15 min; the saliva was cleared of debris and then poured into the bottom of the storage tube. Then, the saliva in the plastic saliva storage tube was stored at − 80 °C in an upright position until measurements were made . After the saliva samples were thawed at room temperature on the day of the measurements, they were taken to the University Medical Faculty Central Biochemistry Laboratory for analysis.
Salivary cortisol was measured by the electrochemiluminescence (ECLIA) method using the Cobas Cortisol ll kit (Roche Diagnostics GmbH; Mannheim, Germany). At the end of each patient’s treatment, after the samples were taken and the measurements were made, the dentists were asked to fill out the “Dentist Job Stress Questionnaire,” consisting of 6 questions reflecting their current stress . In the questionnaire, questions were asked to measure the physician’s current degree of work stress, and they were asked to choose the most suitable option for them.
In the study, the analyses were made with the SPSS 25.0 program. Histograms, q–q plots, and Shapiro‒Wilk’s test were applied to assess data normality. Descriptive statistics are presented as the mean and standard deviation. The Kruskal‒Wallis test was performed to examine the differences in measurements according to the dentists and treatment approaches. The Mann‒Whitney U test was used to reveal the evaluation causing the difference. The Friedman test was used to examine the differences among the measurements in the treatment approaches before, during, and after the procedures. The Wilcoxon method was used to determine the time causing the difference. To test the reliability levels of the question groups in the study, the Co. Alpha analysis was applied. Spearman correlation analysis was applied to determine the relationship between stress levels and biological parameters. P values less than 0.05 were considered statistically significant in the study.
When the findings were evaluated according to the measurement times, systolic and diastolic blood pressure measurements for all dentists before the procedure were similar in the clinical, general anesthesia, and deep sedation groups ( P > 0.05). Both systolic and diastolic blood pressure measurements were found to be higher in the deep sedation group ( P < 0.05). It was determined that the systolic blood pressure measurements after treatment did not differ among the clinical, general anesthesia, or deep sedation groups ( P > 0.05), but the diastolic blood pressure measurements did. The measurements of the deep sedation group were higher than those of the clinical sedation and general anesthesia groups ( P < 0.05). Heart rate and oxygen saturation measurements before, during, and after the treatment did not differ among the clinical sedation, general anesthesia, and deep sedation groups ( P > 0.05). It was observed that cortisol measurements before and during treatment were not at different levels among the clinical sedation, general anesthesia, and deep sedation groups ( P > 0.05). After treatment, the cortisol measurements in the deep sedation group were higher than those in the clinical and general anesthesia groups ( P < 0.05) (Table ). When the systolic and diastolic blood pressure measurements were examined according to the procedure times, it was observed that the systolic and diastolic blood pressure measurements were similar before, during, and after treatment in the clinical and general anesthesia groups ( P > 0.05). In the deep sedation group, systolic and diastolic blood pressures measured during the procedure were shown to be high. The heart rate measurements were at similar levels before, during, and after treatment in the clinical and general anesthesia groups ( P > 0.05). In the deep sedation group, it was determined that the heart rate measurements during the procedure were higher than those before and after the procedure ( P < 0.05). When the authors evaluated the oxygen saturation measurements, the oxygen saturation value during the procedure was lower than that before and after the procedure ( P < 0.05). Cortisol measurements were found to be at different levels according to the processing times shown in Table ( P < 0.05). It was determined that the stress levels obtained by using the applied questionnaire differed among the clinical sedation, general anesthesia, and deep sedation groups. It was observed that the stress level of the dentists in the deep sedation group was higher than that of dentists in the clinical sedation and general anesthesia groups ( P < 0.05). In this study, there was a correlation between the stress level and systolic blood pressure values of the dentists in the clinical sedation, general anesthesia, and deep sedation groups before and after the procedure ( P < 0.05). According to the survey results, the preprocedural systolic blood pressures of the dentists reporting high stress levels were higher. It was determined that the stress level of dentists during the protocols and the systolic blood pressure and heart rate measurements were moderately strong and positively correlated ( P < 0.05) (Table ). The relationship between the stress level of the dentists and oxygen saturation and cortisol values are shown in Table . The study showed that systolic and diastolic blood pressure values and heart rate measurements differed among dentists before, during, and after treatment. It was observed that the difference was because dentist-1 had higher systolic and diastolic blood pressure values and heart rate measurements than all the other dentists ( P < 0.05). Oxygen saturation and salivary cortisol measurements before, during, and after treatment were found to be at similar levels in all dentists ( P > 0.05) (Table ).
Several studies related to occupational stress have been conducted among dentists . In healthy individuals, some physiological changes occur due to exposure to stress, with the secretion of adrenaline at the beginning of stress, and various values such as heart rate and blood pressure increase . On the other hand, salivary cortisol is accepted as a reliable biomarker of the hypothalamus–pituitary–adrenal system as a delayed stress response . In addition, cortisol measurements have advantages such as noninvasiveness of the measurement, easy sampling during an individual’s daily activities, and no long preliminary preparation phase for the samples . Anabuki et al. used cortisol measurements in stress evaluation in addition to questionnaire evaluations in their study of providing treatment with moderate sedation . Compared to the previous study, in which the stress levels that pediatric dentists were exposed to while providing treatment with 3 different treatment approaches were measured, in the present study, we aimed to make additional measurements that determined stress occurring during treatment via objective data. In the literature, although some studies have measured the occupational stress and burnout of dentists , no studies have investigated the effects of different treatment approaches on stress. Therefore, comparing the present study with related research is not easy. For example, in the study by Ronneberg et al. , which evaluated the stress levels of dentists during the application of various procedures and treatments for patients, they stated that dentists were more stressed when treating pediatric patients. Therefore, dentists specializing in the treatment of pediatric patients were included in the study. When the results were examined, it was observed that the O 2 saturation values of the dentists in all 3 treatment approach groups decreased during the procedure. In the literature, it is reported to be difficult for dentists to work with reinforced protective equipment during the pandemic . Therefore, in this study, we assumed that the decrease in O 2 saturation values may be because the study was carried out during the COVID-19 pandemic since dentists worked with extra protective equipment such as N95 masks, surgical masks, glasses, and shields. In the current study, different results were obtained regarding the cortisol values of the dentists in the 3 treatment approach groups. It was found that the postprocedure salivary cortisol values of the dentists were lower than the values obtained before and during treatment while caring for patients in the clinic. Gomes et al. showed that even if pediatric patients have no dental experience, they become stressed, and their cortisol levels increase even with noninvasive treatments such as professional dental prophylaxis. Professional dental prophylaxis increases salivary cortisol in children with dental behavioral management problems: a longitudinal study . In addition, the difficulties experienced while treating a pediatric patient may not always be related to dental anxiety. It has been shown that children, even under midazolam sedation, may not cooperate with a dental procedure without experiencing dental anxiety. Oral midazolam reduces cortisol levels during local anesthesia in children: a randomized controlled trial . Ronnenberg et al. reported that the negative behaviors of pediatric patients during treatment caused stress for dentists . Additionally, in studies, dentists and students stated that they were stressed while applying local anesthesia and restorative treatment in a clinic . In this study, we think that the dentists were stressed before and during the procedure because they did not know what kind of patient they would encounter before providing treatment in the clinic. They were worried that the treatment would be unsuccessful due to the unexpected reactions of the pediatric patients. The compatibility of the pediatric patients can be seen as a reason for the decrease in posttreatment cortisol values. For this reason, stress at the end of the treatment may have decreased since no adverse events occurred during the procedures. When the patients were treated under general anesthesia, it was observed that the salivary cortisol values before treatment were higher than the values obtained during and after treatment. It is thought that the decrease in the salivary cortisol value during treatment with general anesthesia is because this treatment approach is safer for the dentist treating the pediatric patient: the dentist can work without the requirement of giving commands, since the patient is completely under control. Anabuki et al. concluded in their study that the salivary cortisol values of pediatric dentists using moderate sedation decreased during treatment. The reason for this decrease was explained by the fact that although the dentists included in the study stated that they were stressed in the questionnaire, they did not give physiological responses as a result of getting used to their work routines. In this study, when the salivary cortisol values were examined for moderate sedation, it was determined that this value increased during the procedure. This result demonstrates that dentists are more stressed when applying treatment with moderate sedation. Rasmussen et al. , Davidovich et al. , and Ronnenberg et al. reported that experience is an important factor in reducing stress. Dentists specializing in pediatric dentistry were included in this study, while specialist dentists were evaluated in the study of Anabuki et al. The difference in the experience of the pediatric dentists may have led to these different results . Song et al. revealed that general occupational stress in dentists stems from interpersonal relationships with patients and that a specialist’s responsibilities, rather than their workload, have a significant relationship with stress . This study determined that dentists were more nervous during sedation treatment according to the survey results, consistent with the objective results. Since compatible patients were selected for clinical sedation in the study, we think that there was no stress arising from the relationship with the patient. However, as Song et al. mentioned in their study, stress may occur when using general anesthesia and especially deep sedation due to the burden of responsibility for the patient. Occupational stress and factors affecting burnout are related to personal factors as well as external factors such as workload and workplace . Considering the results obtained in this study, the authors observed that the stress symptoms of dentist-1, measured before, during, and after treatment with all treatment protocols, were higher than those of the other included dentists. Although standardization was attempted in the study, the reason why the values of one of the dentists were higher than those of the other dentists can be explained by the fact that the stress factor is personal, although it also depends on the treatment approach and additional factors. This study should be evaluated in light of some limitations. The first is that including more dentists with different training and experience in the study may have led to different results. Second, measuring the daily stress values of the dentists (e.g., trait anxiety, chronic stress) may contribute to the reliability of the results. Within the scope of the study, compatible children were included in the clinical treatment group in terms of standardization of the treatment duration, but it is a limitation that a noncompatible group was not included in the study. In conclusion, it has been determined that dentists who care for pediatric patients are more stressed when providing treatment under deep sedation. The study provides information on the stress caused by routine treatment approaches in the practice of pediatric dentistry. Therefore, the stress of pediatric dentists is clinically important. It is known that occupational stress also affects functional abilities. To reduce stress, it is important to first determine in which situation and treatment approach it occurs more. In this way, measures can be taken to increase the health and treatment quality of dentists, who spend most of the day providing dental treatment for children.
|
SDHB immunohistochemistry for prognosis of pheochromocytoma and paraganglioma: A retrospective and prospective analysis | 1a0277eb-6ac9-454e-9e6d-19e312c9f0dd | 10061060 | Anatomy[mh] | Pheochromocytomas and paragangliomas (PCC/PGL) are rare tumors originating from chromaffin cells and arising from the adrenal medulla and extra-adrenal sites. PGL are subdivided into sympathetic and parasympathetic PGL, depending on their location and catecholamine production. Previously, PCC/PGL were classified as benign or malignant based on the presence of metastatic lesions at nonchromaffin sites . However, since the 4th edition of WHO Endocrine Tumor Classification, the term ‘malignant’ was not used and was replaced by ‘metastatic’ in PCC/PGL, and all PCC/PGL have been considered as tumors with metastatic potential due to the lack of a histological system endorsed for defining the biological aggressiveness . This approach was maintained in the current WHO classification . Given the fact that 15~40% of PCC/PGL patients will have recurrence , and 5%~35% of PCC/PGL patients can develop metastasis , whose five-year survival ranges from 50% to 70% , all PCC/PGL patients should receive at least ten years of follow-up and annual reexaminations , and high-risk patients should be offered lifelong annual follow-up. Current evaluation of the tumor progression potential of PCC/PGL is based on multifactorial risk assessments, including the presence of SDHB mutation, larger tumor size, extra-adrenal location, dopamine hypersecretion, and several histopathologic scoring systems . The first scoring system was Pheochromocytoma of the Adrenal Gland Scaled Score (PASS), proposed by Thompson in 2002 , and then several groups formulated different scoring systems, including GAPP, ASES, and COPPs . Approximately 40% of PCC/PGL carry a germline mutation in one of at least 20 genes . Among these germline mutations, SDHB mutation often displays an increased risk of metastasis and recurrence and is the strongest indicator of metastasis in PCC/PGL , rather than other subunits of SDH . The SDHB gene encodes for SDHB protein assembled in the mitochondria to form succinate dehydrogenase (SDH), a key respiratory enzyme, that has an essential role in cell energy production. With absent or abnormal SDHB, the mitochondrial complex II fails to assemble and loses its enzymatic activity, causing the accumulation of succinate, inhibiting α-KG-dependent dioxygenases, and resulting in a pseudo hypoxic state, hypermethylation, and subsequent invasive behavior . As previously mentioned, SDHB mutation in PCC/PGL patients is associated with an increased risk of tumor progression, and several studies have shown that SDHB gene mutations can be detected by the loss of SDHB staining by immunohistochemistry (IHC) , we hypothesized that loss of SDHB expression in PCC/PGL tumors can function as a risk factor for progressive tumors in the current study. Remarkably, SDHB IHC becomes negative most commonly in the setting where is bi-allelic inactivity in any of the four SDH subunit genes (SDHA, SDHB, SDHC, and SDHD) , and loss of function of both alleles almost always occurs in the presence of a germline mutation in PCC/PGL, but it can also occur in somatic mutation isolated to neoplastic cells . In addition, hypermethylation of the SDHC promoter, resulting in the silencing of the SDHC gene and subsequent inactivity of the SDH complex, can also cause completely negative cytoplasmic staining for SDHB . Recent studies have shown that SDHB protein loss can be used as an IHC biomarker to independently predict the prognosis of PCC/PGL patients , and a modified GAPP system (M-GAPP) and COPPs scoring system also incorporates loss of SDHB staining for analyzing PCC/PGL risk . Despite these above studies have included SDHB IHC in the investigation of the prognosis of PCC/PGL, no reliable prospective study or long-term follow-up for PCC/PGL focusing on SDHB IHC is available at present. In the present study, we analyzed the value of SDHB IHC for distinguishing between high-risk or low-risk of progressive tumors in large retrospective and prospective series in our center.
Study populations We performed a retrospective and prospective study of PCC/PGL patients diagnosed in Ruijin Hospital, Shanghai Jiao Tong University School of Medicine from 2002 to 2020 and collected their tumors after surgery in our center. The retrospective study was composed of a series of 274 tumors from 274 patients who underwent an operation in our center from 2002 to 2014, including 195 PCC and 79 PGL, and this series was based on the population of our previous study . Patients in the retrospective study were selected according to the following inclusion criteria: histologically confirmed diagnosis of PCC/PGL and there was no perioperative mortality. At first, we included 414 patients in our retrospective study. The exclusion criteria were as follows: 1) Patients whose surgery was not performed in our hospital or whose tumor specimen was not available (n=109); 2) Incomplete pathological information (n=12); 4) Patients who were lost to follow-up (n=19). Thus 274 patients were finally included in the retrospective analysis . The prospective study was performed on 217 tumors from 217 patients, which included 153 PCC and 64 PGL, and they received surgery in our center from 2015 to 2020. Patients in the prospective study were selected according to the same inclusion criteria as the retrospective study, and initially, 833 patients were included in the prospective analysis. The exclusion criteria were as follows: 1) Patients whose surgery was not performed in our hospital (n=241); 2) Patients who refused to perform additional SDHB IHC (n=360); 3) Without survival details (n=15) . Thus 217 patients were finally included in the prospective analysis. Their baseline clinical characteristics, including age at diagnosis, sex, primary tumor location, primary tumor size, and catecholamine type were obtained by preoperational examinations. We have categorized the catecholamine-producing phenotype of PCC/PGL as adrenergic phenotype, in case the plasma metanephrine (MN), urinary adrenalin, or urinary dopamine levels exceeds the upper limits; noradrenergic phenotype, in case the normetanephrine (NMN) or urinary noradrenaline exceeds the upper limits without elevated MN/adrenaline/dopamine; or nonfunctioning phenotype, in case catecholamines were within the normal range, according to MNs, urinary adrenalin and dopamine levels. The postoperative pathological evaluation included TNM classification (categorized as I, II, III, or IV), and the TNM staging classification was based on the system set by the American Joint Committee on Cancer (AJCC) . IHC of SDHB Formalin-fixed, paraffin-embedded tissue specimens were obtained and handled by standard surgical oncology procedures. Serial 5-μm sections were prepared and deparaffinized with xylene and gradient concentrations of alcohol (100%, 95%, 80%, and 75%) and then rehydrated. Immunoreactions of these slides were incubated overnight in a solution of SDHB antibody (rabbit polyclonal HPA002868; Sigma-Aldrich Corp, St Louis, MO, USA;1:500). Human adrenal gland tissue was used as the positive control and showed strong granular staining in the cytoplasm. And in the tumor, endothelial cells, sustentacular cells, and lymphocytes served as the internal positive control for each sample. For the retrospective series, SDHB IHC was carried out on all tumors in the Endocrine and Metabolic Diseases department of Ruijin Hospital. For the prospective study, SDHB IHC was consecutively performed in the Pathology department of Ruijin Hospital, allowing simultaneous detection of SDHB protein after surgery. All stained slides were analyzed blindly, by two independent observers (QW and JX), without any knowledge of the clinical data. The interpretation criteria used in this study followed the reported one . In order to determine whether SDHB IHC is negative, we rely on the consensus of the two observers. And only complete loss of SDHB staining was defined as SDHB immunonegative , otherwise, we judged it as immunopositive . Detection of SDHB germline mutation In the retrospective population, all 274 patients underwent SDHB mutation analysis. DNA was collected from peripheral blood and isolated using standard procedure. SDHB mutation analysis was performed by commercial Sanger sequencing of PCR products. Follow-up procedure Postoperative patients were followed up every six months for the first year after surgery, then annually. Laboratory examinations and abdominal computed tomography or magnetic resonance imaging were performed at every visit. All patients were followed up until 2022. The study was approved by the Ethics Committee of the Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, and informed consent, which also permitted the collection of specimens and clinical data, was obtained from patients. Disease-free survival (DFS) time was defined as the date of surgery to the date of tumor progression, including recurrence or metastasis, confirmed by radiological and/or clinicopathological findings, and death. Overall survival (OS) time was defined as the interval between the date of surgery of and death or the latest follow-up. Statistical analysis Descriptive statistics were presented as frequencies, percentages, median, and range. Comparisons of baseline characteristics between the two series were performed by t-test and Chi-square tests. The odds ratio (OR) was calculated by Logistic regression. Survival rates were compared by the Kaplan-Meier curves and differences in survival time were analyzed by log-rank tests. Hazard ratios were performed by univariate and multivariate Cox regression models. The risk ratio (RR) was calculated by Poisson regression with robust error variance. In multivariate analyses, we adjusted the baseline characteristics of PCC/PGL. All confidence intervals (CIs) were stated at the 95% confidence level. P<0.05 was considered statistically significant. The statistical data visualization was performed in the R 4.0.3 environment, using RStudio software, and all statistical analyses were performed by RStudio V 1.3.1093 (Boston, Massachusetts, USA) and SPSS Statistics version 28 (IBM, Armonk, NY, USA).
We performed a retrospective and prospective study of PCC/PGL patients diagnosed in Ruijin Hospital, Shanghai Jiao Tong University School of Medicine from 2002 to 2020 and collected their tumors after surgery in our center. The retrospective study was composed of a series of 274 tumors from 274 patients who underwent an operation in our center from 2002 to 2014, including 195 PCC and 79 PGL, and this series was based on the population of our previous study . Patients in the retrospective study were selected according to the following inclusion criteria: histologically confirmed diagnosis of PCC/PGL and there was no perioperative mortality. At first, we included 414 patients in our retrospective study. The exclusion criteria were as follows: 1) Patients whose surgery was not performed in our hospital or whose tumor specimen was not available (n=109); 2) Incomplete pathological information (n=12); 4) Patients who were lost to follow-up (n=19). Thus 274 patients were finally included in the retrospective analysis . The prospective study was performed on 217 tumors from 217 patients, which included 153 PCC and 64 PGL, and they received surgery in our center from 2015 to 2020. Patients in the prospective study were selected according to the same inclusion criteria as the retrospective study, and initially, 833 patients were included in the prospective analysis. The exclusion criteria were as follows: 1) Patients whose surgery was not performed in our hospital (n=241); 2) Patients who refused to perform additional SDHB IHC (n=360); 3) Without survival details (n=15) . Thus 217 patients were finally included in the prospective analysis. Their baseline clinical characteristics, including age at diagnosis, sex, primary tumor location, primary tumor size, and catecholamine type were obtained by preoperational examinations. We have categorized the catecholamine-producing phenotype of PCC/PGL as adrenergic phenotype, in case the plasma metanephrine (MN), urinary adrenalin, or urinary dopamine levels exceeds the upper limits; noradrenergic phenotype, in case the normetanephrine (NMN) or urinary noradrenaline exceeds the upper limits without elevated MN/adrenaline/dopamine; or nonfunctioning phenotype, in case catecholamines were within the normal range, according to MNs, urinary adrenalin and dopamine levels. The postoperative pathological evaluation included TNM classification (categorized as I, II, III, or IV), and the TNM staging classification was based on the system set by the American Joint Committee on Cancer (AJCC) .
Formalin-fixed, paraffin-embedded tissue specimens were obtained and handled by standard surgical oncology procedures. Serial 5-μm sections were prepared and deparaffinized with xylene and gradient concentrations of alcohol (100%, 95%, 80%, and 75%) and then rehydrated. Immunoreactions of these slides were incubated overnight in a solution of SDHB antibody (rabbit polyclonal HPA002868; Sigma-Aldrich Corp, St Louis, MO, USA;1:500). Human adrenal gland tissue was used as the positive control and showed strong granular staining in the cytoplasm. And in the tumor, endothelial cells, sustentacular cells, and lymphocytes served as the internal positive control for each sample. For the retrospective series, SDHB IHC was carried out on all tumors in the Endocrine and Metabolic Diseases department of Ruijin Hospital. For the prospective study, SDHB IHC was consecutively performed in the Pathology department of Ruijin Hospital, allowing simultaneous detection of SDHB protein after surgery. All stained slides were analyzed blindly, by two independent observers (QW and JX), without any knowledge of the clinical data. The interpretation criteria used in this study followed the reported one . In order to determine whether SDHB IHC is negative, we rely on the consensus of the two observers. And only complete loss of SDHB staining was defined as SDHB immunonegative , otherwise, we judged it as immunopositive .
In the retrospective population, all 274 patients underwent SDHB mutation analysis. DNA was collected from peripheral blood and isolated using standard procedure. SDHB mutation analysis was performed by commercial Sanger sequencing of PCR products.
Postoperative patients were followed up every six months for the first year after surgery, then annually. Laboratory examinations and abdominal computed tomography or magnetic resonance imaging were performed at every visit. All patients were followed up until 2022. The study was approved by the Ethics Committee of the Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, and informed consent, which also permitted the collection of specimens and clinical data, was obtained from patients. Disease-free survival (DFS) time was defined as the date of surgery to the date of tumor progression, including recurrence or metastasis, confirmed by radiological and/or clinicopathological findings, and death. Overall survival (OS) time was defined as the interval between the date of surgery of and death or the latest follow-up.
Descriptive statistics were presented as frequencies, percentages, median, and range. Comparisons of baseline characteristics between the two series were performed by t-test and Chi-square tests. The odds ratio (OR) was calculated by Logistic regression. Survival rates were compared by the Kaplan-Meier curves and differences in survival time were analyzed by log-rank tests. Hazard ratios were performed by univariate and multivariate Cox regression models. The risk ratio (RR) was calculated by Poisson regression with robust error variance. In multivariate analyses, we adjusted the baseline characteristics of PCC/PGL. All confidence intervals (CIs) were stated at the 95% confidence level. P<0.05 was considered statistically significant. The statistical data visualization was performed in the R 4.0.3 environment, using RStudio software, and all statistical analyses were performed by RStudio V 1.3.1093 (Boston, Massachusetts, USA) and SPSS Statistics version 28 (IBM, Armonk, NY, USA).
Baseline characteristics of patients In the retrospective series, the median follow-up duration was 167 months, and the median follow-up duration of the prospective series was 28 months. During follow-up, 38 of 264 (14.4%) patients developed metastasis or recurrence, and 22 patients died in the retrospective series, while in the prospective series, 10 of 213 (4.7%) patients developed metastasis or recurrences, and 1 patient died. Overall, 23 patients died during our follow-up. Compared with the retrospective series, the prospective population showed similar baseline characteristics, including gender, primary tumor location, and catecholamine type, but differed in the age of diagnosis, primary tumor size, and TNM classification . In the retrospective series, the median age at diagnosis of PCC/PGL was 44 years old, ranging from 11 to 79, and 56.9% (156/274) of individuals were female. There were 58.4% of patients had a larger (≥ 5cm) primary tumor size and 28.8% of primary tumors were located in extra-adrenal sites (75 in the abdomen/pelvis, two in the thorax, and two located in the head/neck region). Preoperative catecholamine levels were available in 89.4% (245/274) patients, including 60.8% (149/245) adrenergic subtype, 32.2% (79/245) noradrenergic subtype, and 6.9% (17/245) nonfunctioning subtype. TNM classifications were obtained in all patients, composed of 3.6% stage 4 tumors, 3.6% stage 3 tumors, 62.8% stage 2 tumors, and 29.9% stage 1 tumors. The SDHB staining was negative in 12 (4.4%) tumors. In the prospective series, the median age at diagnosis was 48 years old, with a range of 16 to 77 years old, and 53.0% (115/217) of the patients were female. 38.7% of patients had primary tumors that were bigger (≥ 5 cm) in size, and 29.5% of primary tumors were of extra-adrenal origin, including 62 located in the abdomen/pelvis, one in the thorax, and one in the head/neck region. Preoperative catecholamine levels were available for all patients, and 57.6% were adrenergic subtype, 38.2% were noradrenergic subtype, and 4.1% were nonfunctioning subtypes. The percentages of stage 4 tumors, stage 3 tumors, stage 2 tumors, and stage 1 tumors by TNM classification, were 1.8%, 10.1%, 45.2%, and 42.9%, respectively. The SDHB staining was negative in 17 (7.8%) tumors. Retrospective study Among the 274 patients, ten patients had metastatic lesions when they received an operation in our hospital, thus only 264 individuals were included in further analysis, except for the OS analysis. During follow-up, there were 46 patients that developed progressive tumors or died, six with SDHB staining and 40 with SDHB staining. 66.7% (6/9) of participants in the SDHB group had progressive tumors or died compared with 15.7% (40/255) in the SDHB group (OR: 10.75, 95% CI: 2.72-52.60, P=0.001). After adjusted by other baseline parameters, including primary tumor size, primary tumor location, age at diagnosis, sex, catecholamine type, and TNM classification, SDHB was still associated with poor outcomes (OR: 11.68, 95% CI: 2.58-64.45, P=0.002). The Kaplan-Meier curve showed SDHB patients were more likely to develop progressive tumors or died (P<0.001) . The multivariate Cox regression analysis showed that SDHB IHC, even adjusted by other clinicopathological factors, was still a statistically significant independent predictor for DFS (HR: 6.89, 95% CI: 2.41-19.70, P<0.001) . The median DFS in the retrospective series was 163 months (95% CI: 3-14 months), and the tumor-free rate of the 264 patients was 94.7% at five years and 90.9% at ten years. For the OS analysis, the Kaplan-Meier curve revealed that SDHB patients had a higher possibility to die (P<0.001) . In addition, we compared the performance of SDHB IHC with SDHB germline mutation in predicting progressive tumors in our retrospective series, and the result is shown in . There were eight patients with SDHB germline mutation, and six of them were included in this comparison because two patients already had metastatic sites when they underwent surgery. Based on our result, we found that among the predictive index scores, the specificity of SDHB germline mutation was higher (99.08%) compared with SDHB IHC (98.62%). However, the sensitivity for predicting progressive behavior was higher in SDHB IHC (13.04%) compared with SDHB germline mutation (8.7%), and so was the negative predictive value. Prospective study There were four patients who already had metastatic lesions when they were enrolled in the prospective study, thus only 213 individuals were included in the further analysis, except for the OS analysis. Ten patients had poor medical outcomes including recurrences, metastasis, and death during follow-up. 18.8% (3/16) of participants in the SDHB group had progressive tumors or died compared with 3.6% (7/197) in the SDHB group (RR: 5.28, 95% CI: 1.51-18.47, P=0.009). After regression analysis of mixed effects, including primary tumor size, primary tumor location, age at diagnosis, sex, catecholamine type, and TNM classification, SDHB IHC was still an independent predictor associated with prognosis (RR: 3.35, 95% CI: 1.20-9.38, P=0.021) . However, the Kaplan-Meier curve showed no difference in DFS or OS between the two groups .
In the retrospective series, the median follow-up duration was 167 months, and the median follow-up duration of the prospective series was 28 months. During follow-up, 38 of 264 (14.4%) patients developed metastasis or recurrence, and 22 patients died in the retrospective series, while in the prospective series, 10 of 213 (4.7%) patients developed metastasis or recurrences, and 1 patient died. Overall, 23 patients died during our follow-up. Compared with the retrospective series, the prospective population showed similar baseline characteristics, including gender, primary tumor location, and catecholamine type, but differed in the age of diagnosis, primary tumor size, and TNM classification . In the retrospective series, the median age at diagnosis of PCC/PGL was 44 years old, ranging from 11 to 79, and 56.9% (156/274) of individuals were female. There were 58.4% of patients had a larger (≥ 5cm) primary tumor size and 28.8% of primary tumors were located in extra-adrenal sites (75 in the abdomen/pelvis, two in the thorax, and two located in the head/neck region). Preoperative catecholamine levels were available in 89.4% (245/274) patients, including 60.8% (149/245) adrenergic subtype, 32.2% (79/245) noradrenergic subtype, and 6.9% (17/245) nonfunctioning subtype. TNM classifications were obtained in all patients, composed of 3.6% stage 4 tumors, 3.6% stage 3 tumors, 62.8% stage 2 tumors, and 29.9% stage 1 tumors. The SDHB staining was negative in 12 (4.4%) tumors. In the prospective series, the median age at diagnosis was 48 years old, with a range of 16 to 77 years old, and 53.0% (115/217) of the patients were female. 38.7% of patients had primary tumors that were bigger (≥ 5 cm) in size, and 29.5% of primary tumors were of extra-adrenal origin, including 62 located in the abdomen/pelvis, one in the thorax, and one in the head/neck region. Preoperative catecholamine levels were available for all patients, and 57.6% were adrenergic subtype, 38.2% were noradrenergic subtype, and 4.1% were nonfunctioning subtypes. The percentages of stage 4 tumors, stage 3 tumors, stage 2 tumors, and stage 1 tumors by TNM classification, were 1.8%, 10.1%, 45.2%, and 42.9%, respectively. The SDHB staining was negative in 17 (7.8%) tumors.
Among the 274 patients, ten patients had metastatic lesions when they received an operation in our hospital, thus only 264 individuals were included in further analysis, except for the OS analysis. During follow-up, there were 46 patients that developed progressive tumors or died, six with SDHB staining and 40 with SDHB staining. 66.7% (6/9) of participants in the SDHB group had progressive tumors or died compared with 15.7% (40/255) in the SDHB group (OR: 10.75, 95% CI: 2.72-52.60, P=0.001). After adjusted by other baseline parameters, including primary tumor size, primary tumor location, age at diagnosis, sex, catecholamine type, and TNM classification, SDHB was still associated with poor outcomes (OR: 11.68, 95% CI: 2.58-64.45, P=0.002). The Kaplan-Meier curve showed SDHB patients were more likely to develop progressive tumors or died (P<0.001) . The multivariate Cox regression analysis showed that SDHB IHC, even adjusted by other clinicopathological factors, was still a statistically significant independent predictor for DFS (HR: 6.89, 95% CI: 2.41-19.70, P<0.001) . The median DFS in the retrospective series was 163 months (95% CI: 3-14 months), and the tumor-free rate of the 264 patients was 94.7% at five years and 90.9% at ten years. For the OS analysis, the Kaplan-Meier curve revealed that SDHB patients had a higher possibility to die (P<0.001) . In addition, we compared the performance of SDHB IHC with SDHB germline mutation in predicting progressive tumors in our retrospective series, and the result is shown in . There were eight patients with SDHB germline mutation, and six of them were included in this comparison because two patients already had metastatic sites when they underwent surgery. Based on our result, we found that among the predictive index scores, the specificity of SDHB germline mutation was higher (99.08%) compared with SDHB IHC (98.62%). However, the sensitivity for predicting progressive behavior was higher in SDHB IHC (13.04%) compared with SDHB germline mutation (8.7%), and so was the negative predictive value.
There were four patients who already had metastatic lesions when they were enrolled in the prospective study, thus only 213 individuals were included in the further analysis, except for the OS analysis. Ten patients had poor medical outcomes including recurrences, metastasis, and death during follow-up. 18.8% (3/16) of participants in the SDHB group had progressive tumors or died compared with 3.6% (7/197) in the SDHB group (RR: 5.28, 95% CI: 1.51-18.47, P=0.009). After regression analysis of mixed effects, including primary tumor size, primary tumor location, age at diagnosis, sex, catecholamine type, and TNM classification, SDHB IHC was still an independent predictor associated with prognosis (RR: 3.35, 95% CI: 1.20-9.38, P=0.021) . However, the Kaplan-Meier curve showed no difference in DFS or OS between the two groups .
In this study, we included two series of PCC/PGL patients, composed of 274 tumors for retrospective study and 217 tumors for prospective study, to investigate the clinical significance of SDHB IHC. And this is a relatively large-scale study, using SDHB IHC of PCC/PGL tumors as a predictive factor to evaluate prognosis. We discovered that SDHB patients had a much worse prognosis than SDHB patients and that the lack of SDHB protein expression was an independent risk factor linked to progressive tumors. Both the retrospective analysis and the prospective investigation found that patients with SDHB staining were more likely to experience adverse outcomes when considering the proportion of developing progressing tumors. Additionally, SDHB patients had significantly shorter DFS and OS time than SDHB individuals in survival analyses. However, we only found this tendency in the retrospective analysis, not the prospective study. Given the fact that between the retrospective series and prospective series, the median age at diagnosis, primary tumor size, and TNM classification were statistically different, which showed that in the retrospective series, patients were diagnosed with PCC/PGL at an earlier age, had a larger proportion of bigger tumors and late TNM stages. Thus, we thought that no statistical difference found in the prospective study was caused by the above clinicopathological differences and the mean follow-up duration of the prospective series was not as long as that of the retrospective series. Previous studies which performed SDHB IHC also showed that the loss of SDHB protein is associated with poor prognosis , and they discovered that patients with SDHB tumors had considerably lower OS and DFS than those with SDHB tumors. Compared to the research mentioned above, our study had a larger sample size and a longer follow-up time. However, there was a study that contradicted our conclusions, which found SDHB was associated with shorter DFS in PCC patients , and we supposed it may be a result of different inclusion criteria for patients. The SDHB protein is a catalytic subunit of succinate-ubiquinone oxidoreductase, also known as succinate dehydrogenase, or mitochondrial complex II, which links the electron transport chain and the Krebs cycle, and catalyzes succinate into fumarate. Other SDH-related genes (SDHA, SDHC, and SDHD) are autosomal inheritance and their corresponding proteins are assembled at the inner membrane of mitochondrial to form the SDH enzyme, except SDHAF2 encodes a protein that helps the insertion of FAD into SDHA . When SDH enzymatic activity decreased in SDHx-mutated tumors, SDHB expression levels by Western blotting and IHC were reduced as well . Comparing transcriptional and translational efficiency for indicating SDHB germline mutation, Chunzhang Yang and colleagues found that SDHB-related PCC/PGL tumors demonstrated unreduced mRNA expression but significantly declined protein expression than non-SDHB PCC/PGL , and most SDH-mutated tumors were negative for SDHB IHC, while non-SDH related tumors nearly all showed expression of SDHB by IHC . These results indicated that SDHB IHC is a favorable predictor for SDH-related tumors, better than transcriptional mRNA expression. In addition to the loss of SDHB expression, we also found that larger tumor size, extra-adrenal location, noradrenergic phenotype, and earlier age at diagnosis were significantly associated with shorter DFS in PCC/PGL, in agreement with previous findings . Despite the fact that numerous studies have attempted to evaluate biomarkers for predicting PCC/PGL prognosis, there was no definite standard that can predict the clinical behavior of PCC/PGL, and the 5th series of the WHO Classification of Endocrine and Neuroendocrine Tumours does not support any of scoring systems but does not forbid their use in individual practices either . As our team analyzed before, SDHB mutation showed a higher odds ratio of metastasis , and previous studies had indicated that SDHB IHC can detect the presence of an SDH mutation . Based on our findings, the SDHB germline mutation demonstrated high specificity in the retrospective study compared with IHC; however, SDHB IHC was more sensitive and had a higher negative predictive value, which meant that more high-risk individuals and more low-risk individuals were divided into the poor prognosis group and the better prognosis group, respectively, and high-risk individuals would receive more stringent reexaminations and follow-up. And considering that many centers could not satisfy the need to undergo genetic tests for every PCC/PGL patient, this study can benefit PCC/PGL patients by using SDHB IHC to predict prognosis, as IHC is an easy procedure that could be performed in nearly all centers. However, our study still has several limitations, such as the subjective definition of the IHC results which may decrease the predictive value of SDHB protein expression and this is an observational study, without any intervention, with a limited number of cases that had the main outcome. One of the reasons is that PCC/PGL are generally slow-growing tumors and metastasis commonly happens several years or even decades after the initial surgery, rendering investigation more difficult, thus a multicenter cohort study with a longer follow-up time is needed. The second reason may be the insufficient utilization of detection methods for metastasis or recurrence sites, such as 18F-FDOPA PET/CT and 68Ga-DOTATATE PET/CT. And at last, our proposed procedure can only be done after patients have undergone surgery and tumor tissue is available for pathological study. In conclusion, we suggest that loss of SDHB protein can be regarded as a biomarker of adverse outcomes in PCC/PGL. Therefore, we recommend that patients with SDHB tumors into consideration with more rigorous follow-up protocols, as they are more likely to develop progressive tumors.
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.
The studies involving human participants were reviewed and approved by the ethics committee of the Ruijin Hospital, Shanghai Jiao Tong University School of Medicine. Written informed consent to participate in this study was provided by the participants’ legal guardian/next of kin. Written informed consent was obtained from the individual(s), and minor(s)’ legal guardian/next of kin, for the publication of any potentially identifiable images or data included in this article.
GN and WW conceptualized the study. LJ, JX, YJ, CZ, WZ, and LY designed the study. XZ and LW collected the data. YY and TS did the analysis and drafted the article. All authors contributed to critical revisions and final approval of the article.
|
Editorial: Community series in SIRT family in endocrinology | 38139299-bc52-46ef-8a0c-9e814a1e8356 | 10061062 | Physiology[mh] | All authors listed have made a substantial, direct and intellectual contribution to the work, and approved it for publication. |
Exploring culturally-preferred communication approaches for increased uptake of voluntary medical male circumcision (VMMC) services in rural Malawi | f187bd24-cb76-4964-94e2-8495c120d2ee | 10061708 | Health Communication[mh] | In 2010, approximately 54, 000 new HIV infections were registered in Malawi. In response to this alarming figure, Malawi committed to reducing new infections by 75% by 2020 . By that year, Malawi had registered 19,000 new infections . In light of Malawi’s prevailing HIV burden, voluntary medical male circumcision (VMMC) became a key HIV prevention strategy as per joint World Health Organization (WHO) and The Joint United Nations Programme on HIV/AIDS (UNAIDS) recommendation of 2007 . Supported by evidence from three randomized studies conducted in South Africa, Uganda and Kenya which revealed that circumcision was effective in reducing HIV transmission among heterosexual men by 60–70%, in 2007, both UNAIDS and WHO recommended VMMC as a new HIV prevention measure . The recommendation specifically targeted 14 countries with high HIV burden in the sub-Saharan Africa region, including Malawi . Globally, by 2019 nearly 26.8 million cumulative male circumcisions for HIV prevention were performed between 2008 and 2019 in the 15 priority countries of East and Southern Africa . This number of circumcisions had averted about 340,000 new cases of HIV by 2019. As a result of the registered success, it was projected that by 2030 some 1.8 million new infections would be prevented . In the same period, Malawi performed cumulatively 887,205 medical circumcisions representing approximately 31% of the target for 2020 . As in the case of Uganda, Malawi’s VMMC implementation faced some hesitancy due to among others, lack of local evidence on benefits of the intervention . Malawi developed its first national policy on VMMC in 2012, which was incorporated into a mix of HIV prevention strategies in the country . However, performance towards the set targets has generally been slacking . With VMMC prevalence currently at 31% of men aged between 15 and 49, Malawi trailed the 60% target by almost 50% . Communication has been integral to VMMC implementation and specifically in VMMC demand creation in Malawi . The first VMMC communication strategy in Malawi was implemented from 2012 to 2016 . The 2010 Situation Analysis Report recommended a VMMC communication strategy to create demand for services . At regional level, findings of a study in Kenya adds weight behind communication as a tool for increasing VMMC uptake . Accordingly, a communication strategy was embedded into the implementation package for VMMC for purposes of demand creation . Malawi Government acknowledges that VMMC demand creation through communication interventions has proved challenging . Though there is a high awareness of VMMC’s health benefits among the target audience, there is low uptake among both circumcising and non-circumcising communities of Malawi . Several studies have shown that most tribes and religious groups regard circumcision as ‘amoral’ and ‘intrusive’ to their social-cultural values, beliefs, and traditions; thus, also viewed as ‘threatening’ their identity of tribes and religious groups . In addition, fear of pain, the long recovery period, perceived fertility loss, and medical complications are some other factors fueling the resistance to medical circumcision among adults . The impact of these factors on uptake among adult men in Malawi suggests the need for efforts to increase uptake . Communication is integral to countering these factors by boosting knowledge, understanding and positive attitudes toward the service and in turn stimulating demand and uptake among Malawian men, boys, parents/guardians and partners . Hence, the Malawi government developed the first VMMC Communication Strategy in 2012 to guide design and implementation of strategic communications within the framework of the national VMMC policy towards achieving the 80% uptake of VMMC among eligible males aged 10–49 by 2025 . By 2007, the Sub-Sahara African Region had the highest HIV prevalence globally and had very low male circumcision prevalence allegedly due to the widespread presence of Christianity which did not encourage male circumcision and the historical legacy of British colonialism . Following three randomized clinical trials in Uganda, Kenya and South Africa in 2005, in 2007 WHO and UNAIDS recommended to 14 East and Southern Africa countries, to take VMMC as an extra arsenal in the fight against HIV transmission . These countries included Botswana, Ethiopia, Kenya, Lesotho, Malawi, Mozambique, Namibia, Rwanda, South Africa, Swaziland, Uganda, United Republic of Tanzania, Zambia and Zimbabwe . By 2017 nearly 15,269,720 million boys and men had been circumcised in a decade with support from PEPFAR alone and figures have continued to rise . In 2013, UNAIDS estimated that the world had 35 Million people living with HIV; in the same year, 2.1 million more people got infected and 1.5 people died of AIDS-related illnesses . In Malawi, HIV prevalence among persons aged between 15 and 49 has been reducing from 16.4% in 1999 to 8.5 in 2020 . Overall, new infections among the 15–49 brackets were reducing. While in 2011 the country registered 55,000 new infections, it only registered 19, 000 in 2020 . This, however, remains an unacceptably very high number compared to the population size of approximately 19 million people and calls for concerted effort in a drive to cut the number of new infections to 0 by 2030 . Although national average adult HIV prevalence hovered around 8.5% the country had witnessed about 10, 000 HIV-related deaths at the end of 2020 . In Malawi, Male Circumcision is most common in Southern Region with prevalence rate at 47% among men aged 15–49 seconded by Central Region at 15% and lastly at 6% in the Northern Region of the country . By 2018, overall VMMC performance in the country hovered around 30% of the total target population of males in the age bracket of 15 to 49 . According to available data, low performance of VMMC in Malawi could be attributed to social-cultural factors and perceived adverse effects following the procedure at social-ecological level . Specifically, VMMC was perceived by both traditionally circumcising and non-circumcising ethnic groups as a threat to cultural identity, religious beliefs and values . At personal level, males avoided VMMC for fear of being tested for HIV at the health facility; concerns of meeting travel costs to the health facility; concerns over losing income during the recuperation period; fear of infidelity on the other partner among married couples either during the recuperation period (on the part of males) and after recuperation (on the part of females); unwillingness to abstain from sexual activity six weeks after undergoing the VMMC procedure and perception that VMMC diminishes sexual pleasure . Related to VMMC communication, Mhagama et al had concluded that although VMMC was purportedly voluntary, most up-takers did not do so voluntarily . According to the study that was conducted in Lilongwe, men were influenced to uptake VMMC mostly by peer pressure and the need for conformity; partner/girlfriend demand and considerations; and advice from health personnel . That men opt for VMMC under duress impedes on VMMC uptake . It is therefore the task of communication to bring to the fore the salience of voluntariness in VMMC in order to increase uptake among men. The word “communication” finds its roots in the Latin word, “communis” which translates “common”. The key purpose of communication interventions is to create a common understanding between the communicator and the audience . Linking “communication” to behavior change, Carl Hovland defines communication as: “…the process by which an individual (the communicator) transmits stimuli (usually verbal symbols) to modify the behavior of the other individuals (communicates)” . In order to effectively identify the preferred communication approaches for the promotion of VMMC services among the Yao of Mangochi and the Chewa of Dowa in Southern and Central regions respectively, these researchers used Laswell’s Transmission Theory which is anchored by the dictum “Who says What, in Which Channel, to Whom and for What Effect?” (as indicated in Fig. : A typical conception of Laswell’s construct as a graphic model of communication ) . This theory was named after Harold Dwight Laswell an American political scientist and a communication theorist who in 1948, while he was a professor at Yale Law School, wrote in his article entitled: “The Structure and Function of Communication in Society” , : …the most convenient way to describe an act of communication is to answer the following questions: a. Who b. Says What c. In Which Channel d. To Whom e. With What Effect? We call that the “5Ws” model . This framework effectively divides up the field of communication into five key areas of study and research. These researchers framed questions from these five fields to unearth the preferred communication approaches to be used by health communicators in both circumcising and non-circumcising communities. Informants answered these five most important questions: (a) What is your preferred source of information for VMMC information? (b) What is your preferred message for VMMC communication interventions? (c) What is your preferred audience for VMMC messaging? (d) What is your preferred channel for VMMC communication interventions? (e) What are your expected effects (or impacts) of VMMC communication interventions? When data were collected and analyzed these researchers noted, just like other scholars have, that feedback is the key missing element in Laswell’s 5Ws Model since communication is a two-way and not a one-way road . They noted through this study too that current communication interventions meant to promote VMMC services in Malawi did not pay adequate attention to these five elements. These researchers also discovered that, in Malawi, the most popular and preferred communication approaches were community engagement and interpersonal communication since they allowed the audience to give real-time feedback to communicators on VMMC messages. This study was informed by the works and commentaries on the Social-Ecological Model, Laswell’s Transmission Theory, and numerous communication and behaviour change theories. Secondly, investigators collected data among the predominantly circumcising Yao of Mangochi and the predominantly non-circumcising Chewa of Dowa. Researchers used FGDs, KIIs and IDIs to gain a comprehensive and deep understanding of the values, beliefs and traditions that underlie the resistance to VMMC. Data were thematically analyzed to decipher the people’s communication needs and preferred communication approaches by VMMC campaigners. Sample population Total number of participants in this study was 276 (160 males and 116 females. We conducted 24 focus group discussions (FGDs); 13 in Mangochi involving 118 participants (66 males and 52 females), 11 in Dowa involving 97 participants (54 males and 43 females). This makes a total of 215 participants for both Mangochi and Dowa. There were 9 key informant interviews involving 14 participants 4 interviews of Mangochi (7 males 0 females) and 5 interviews in Dowa involving 7 participants (4 males and 3 females). There were 16 in-depth interviews, involving 19 participants (14 males and 5 females). In Dowa there were 7 in-depth interviews involving 10 people (6 males and 4 females). In Mangochi there were 9 in-depth interviews involving 10 participants (9 males and 1 female). There were 5 life histories involving 5 participants (2 in Dowa and 3 in Mangochi). In Dowa 1 was male and 2 were females. In Mangochi there was 1 male and 1 female. There were two PRAs involving 23 people. In Dowa there were 11 participants (6 males and 5 females). In Mangochi there were 12 participants (6 males and 6 females) The interview guides during this study were based on the five key questions as indicated in Fig. below. By collecting data from 160 males and 116 females including youths and minors, we explored the social and cultural values that influence people’s decisions to seek health care. Specifically, the study explored the reasons why uptake of VMMC services was low. We also explored the preferred communication approaches that VMMC campaigners could use in order to motivate men and boys to access services. Figure below shows the details of data collection channels disintegrated according to districts, research tools and participant categories. Total number of participants in this study was 276 (160 males and 116 females. We conducted 24 focus group discussions (FGDs); 13 in Mangochi involving 118 participants (66 males and 52 females), 11 in Dowa involving 97 participants (54 males and 43 females). This makes a total of 215 participants for both Mangochi and Dowa. There were 9 key informant interviews involving 14 participants 4 interviews of Mangochi (7 males 0 females) and 5 interviews in Dowa involving 7 participants (4 males and 3 females). There were 16 in-depth interviews, involving 19 participants (14 males and 5 females). In Dowa there were 7 in-depth interviews involving 10 people (6 males and 4 females). In Mangochi there were 9 in-depth interviews involving 10 participants (9 males and 1 female). There were 5 life histories involving 5 participants (2 in Dowa and 3 in Mangochi). In Dowa 1 was male and 2 were females. In Mangochi there was 1 male and 1 female. There were two PRAs involving 23 people. In Dowa there were 11 participants (6 males and 5 females). In Mangochi there were 12 participants (6 males and 6 females) The interview guides during this study were based on the five key questions as indicated in Fig. below. By collecting data from 160 males and 116 females including youths and minors, we explored the social and cultural values that influence people’s decisions to seek health care. Specifically, the study explored the reasons why uptake of VMMC services was low. We also explored the preferred communication approaches that VMMC campaigners could use in order to motivate men and boys to access services. Figure below shows the details of data collection channels disintegrated according to districts, research tools and participant categories. Findings of this study revolved around the five critical questions of transmission inquiry as suggested by Harold Laswell namely ‘who says what, to whom, through what channel, and with what effect? ’ Five questions framed the presentation of results namely: (a) What is the preferred source of information for VMMC information? (b) What is the preferred message for VMMC communication interventions? (c) What is the preferred channel for VMMC communication interventions? (d) What is the preferred audience for VMMC messaging? (e) What are the expected effects (or impacts) of VMMC communication interventions? Communication ‘noise’ behind the low attention to information on VMMC services in the two study sites We first report on the cultural and religious factors that impeded VMMC communication in the two data collection sites. Overall, this study found that royalty to culture and religion prevented people from the two study sites from acting on messages. These messages were being disseminated by various VMMC implementing agencies, predominantly NGOs, with support from various development partners, under the leadership of the Ministry of Health. However, the traditionally circumcising Yao people avoided VMMC services on the fear that it undermined jando, a cultural initiation ritual where boys were circumcised to symbolize transition from childhood to adulthood. VMMC messages and the VMMC program as a whole, were perceived to be in conflict with their cultural identities and, therefore, a threat to the survival of their initiation practices which, for the circumcising Yaos, went beyond the mere cutting of the foreskin but also inculcated character in the initiates: “Ours was the circumcision of the brain not of the penis as they are portraying it now… As Yaos we were circumcised for cleanliness, moral discipline and transition to manhood,” [KII, Yao Culture Expert ]. Jando also had a political significance since boys were prepared for fiduciary duties towards their local chief and enabled them to practice a ritual that was sanctioned by Islamic Scriptures. “…This time around you are telling the chiefs that circumcision should be done at the hospital. The chief cannot promote this because he is benefiting nothing ….,” [ KII, Yao Cultural Expert & Sheikh]. On the other hand, being a predominantly traditionally non-circumcising and Christian ethnic group, the Chewa people did not pay much attention to VMMC messages. Their understanding was twofold: first, male circumcision was not part of their culture. Secondly, male circumcision was not part of their religion but an attempt by government to popularize Islam in their area. Informants, therefore, equated getting circumcised with becoming a Yao or a Muslim. One man said: “ I am a Chewa man, why should I go for circumcision? Do you want me to become a Yao?” Another one was more conscious of his faith: “ As Christians, we do not make [put] emphasis on circumcision,” [IDI, Religious leader - Dowa]. These researchers posit, therefore, that cultural and religious perception, accounted for low uptake of VMMC services in the two data collection communities. Culturally-preferred communication approaches based on Laswell’s theory Based on Harold Laswell’s fivefold question we now present results of the study as follows: What is the preferred source of information for VMMC information? Both cultures under study perceived the chief as a primary source of information and a mouth-piece of the community. Hence any information to be delivered to the community needed to pass through and vetted by him or her. Despite that among the Yao allegiance was determined by one’s economic standing, among the Chewa it was more of the cultural royalty. From that end, the study informants first noted that promoters of VMMC had overlooked this important community entry protocol thereby creating a communication barrier. “They must tell the chief, [and] the chief will announce to his people that the hospital personnel have something to say, so everyone will go to the meeting…The chief will open the floor and hand it over to the hospital staff to address the people.” [FGD, Unmarried Youth – Dowa]. From the foregoing, the chief was traditionally, the primary source of information. However, for lack of technical knowledge and skills on the subject of VMMC services study participants in both communities identified health workers as a preferred source of information. Study participants held that health workers were well trained hence knowledgeable, experienced and holding reliable information. “…medical personnel should be assigned to spread the message of circumcision because they are well trained. If they were to convey the message it will be better understood,” [IDI, Traditional Leader – Dowa]. What is the preferred message for VMMC communication interventions? The most preferred message that communities want to hear from those implementing VMMC interventions is benefits that VMMC offers to communities, households, and men and boys in particular. Regarding pre-adolescents, parents, clan leaders, community and religious leaders from Mangochi district indicated that they needed to be schooled on how VMMC provided long-term health benefits to males. For example, they needed to learn about such long-term benefits as the reduction of the risk of contracting HIV during traditional circumcision rituals during ndagala . Parents and guardians also wanted assurance that VMMC offered their wards protection against HIV infection during heterosexual relationships before and after they began to raise their future families. An unmarried youth in Dowa narrated as follows: “ I think the most vital information is that people should know the benefits… of circumcision and to my understanding circumcision primarily protects you from cervical cancer so if they are told the women will encourage the men to go,” [ FP1, Dowa Unmarried Youth]. A female respondent in Dowa was of the view that men were not provided with sufficient details regarding VMMC procedure. From her perspective, VMMC messages needed to clearly explain what exactly made a circumcised person different from a non-circumcised man. She observed that VMMC messages had scanty details on what the circumcision procedure involved and what really happened to the man. She argued that men, who were the target of opportunity for VMMC interventions, were blank on details of the procedure: “…We just said they circumcise each other but we did not really know what they really cut. People just say the ‘foreskin’ of the penis others just say the ‘[head of the] penis’ so when you think about it, eeh! It’s scaring.” [FGD, Gate Keepers - Dowa]. The study established that the role of VMMC in the prevention of cervical cancer was more attractive than the role that it played in the reduction of HIV infection since everybody in the area knew at least one woman who had died of cervical cancer. They, therefore, wanted messages to focus on how VMMC would protect females from cervical cancer. “The protection from cervical cancer in women is most important in this area…we don’t have the desire for promiscuity and our women don’t have the desire of seeking sexual satisfaction in bed from circumcised men… Let us put HIV aside, currently the lives of people are at risk. You can manage HIV when infected, but there is no cure for cancer,” [FGD, Traditional Leaders - Dowa]. Female participants wanted VMMC messages to focus more on how their sons would be circumcised by trained medical personnel, which would result in less pain, loss of blood and reduce the risk of complications. “…in the past we had STIs like chindoko mabomu but they were treatable but these days there is AIDS which is very dangerous. There is [cervical] cancer and chisonono too. VMMC is helpful since we hear that it protects our youth and their [future] families from these. These are messages that we need hear about,” [FGD, Adult Women – Mangochi]. VMMC messages also needed to fight against stigma and must assure both Christians and Muslims that VMMC did not contradict the Bible and the Qur’an respectively. One Sheikh, for example, was pleased with VMMC, as a religious authority: “…when I look at the current situation as an Islamic leader, I am very pleased because what is happening now [in clinics] is real jando as prescribed by the Islamic faith.… VMMC fully agrees with Islam.” [KII, Religious Leaders - Mangochi] What is the preferred channel for VMMC communication interventions? Regarding the channel of communication for VMMC intervention, many informants reported that the media, particularly radio was prominently used in both study communities. “…to avoid the huge cost [that would come with village meetings] then I can only agree with the media approaches that are being used currently… just think about how MBC Radio One is boasting that 87% of the Malawi population are able to hear them… Now if 87% are listening to MBC [alone] how about those listening to Zodiak?” [IDI, Religious Leader – Mangochi]. However, not many informants were excited about the use of such open channels to promote male circumcision both in Dowa and Mangochi district. In the two districts the use of open media channels was frowned upon for infringing on the secrecy that surrounds rites of passage rituals. First, male circumcision among the Yao was not a subject for public discussion. Second, although both the Quran and the Bible legitimized it, the praxis in the Muslim Yao community was to veil it in secrecy since those not circumcised and females, in particular, were not supposed to know what happened out in the bush: “In the past, circumcision was a total secret such that even the boy’s mother did not know why her child was going to ndagala . The majority of the mothers celebrated jando blindly. And when we were there they strictly told us not to tell anybody what had happened to us. They told us that when we went back home we must bath with pants on lest the mother notices the difference in the appearance of the member….” [KII, Religious Leaders - Mangochi]. Another key concern with the use of radio, television and other forms of mass media was the sexual explicitness that such messages were associated with. This concern was particularly common among adult men and women who feared that such messages would have negative consequences since they promoted sexual immorality among the youth. This, they argued, was contrary to the purpose of male circumcision in the Yao culture: “The problem comes in because … [through the media] you are telling people that once you have been circumcised, you are free from HIV. These are wrong messages…You will [now] hear a man telling a woman that ‘I was circumcised you can come for sex I can’t contract diseases,’” [KII, Yao Culture Expert - Mangochi]. Informants’ most preferred communication channel both among the Yaos of Mangochi and the Chewas of Dowa district was face-to-face engagement meetings attended by health promotion experts on one hand and community leaders, cultural gatekeepers and community members on the other. Among these methods were: meetings where traditional leaders would bring their subjects to one place and engage them in a dialogue; cultural performances such as gule wankulu; religious gatherings such as regular weekly services of worship; interactive drama and door-to-door visits just to mention some. The key reasons for preference were twofold: first, unlike the mass media, they provided room to control the type of audience to hear the message or not. Secondly, they provided an opportunity for the audience to ask questions and seek clarifications as messages are delivered. Engagement meetings organized by traditional leaders were important since chiefs and their councils of advisors acted as a bridge between citizens and government and its stakeholders: “… gulewankulu characters can dance then the drama can come next to teach things like these or you can introduce the topic at first so that the people can have an idea so when the drama on circumcision comes it will entertain them,” [FGD, Gatekeepers – Dowa]. Although informants from all Protestant Christian churches including the Catholic leadership in Dowa sounded very negative about using the church to promote VMMC messages, in Mangochi the Catholic Church was already disseminating VMMC messages. I was also carrying out surgeries. One gulupa said: “ … [Children] are also assisted to get circumcised. When they get to that stage, they are kept at the parish, which is like a simba ….” In Chowe area, Mangochi, FGD participants proposed that VMMC messages should be disseminated through their mosques and churches. This is how one traditional leader put it: “In this community, Churches and Mosques are more ideal. Religious gatherings are easy places to disseminate VMMC information,” [FGD, Traditional Leaders, Mangochi]. In Mangochi, a representative of an NGO added that they used males from within the community to deliver VMMC messages to fellow men. The health worker posited that men tended to listen better to health messages when delivered by their peers. “ We have community mobilizers who are recruited from within communities. They reach out to men in the communities,” [KII Participant, NGO Worker – Mangochi]. Another NGO worker said: “Our main approach with demand creation is interpersonal communication.” [KII Participant, NGO Worker - Mangochi] . What is the preferred audience for VMMC messaging? The study established that VMMC messages should target all community members including men, women and boys to avoid misunderstandings that may arise from misinterpretation of the messages particularly among married couples. This is what one IDI participant said: “These days I think that everyone [should be targeted] because, in a family, the man may hear the message alone but when he brings the message into his household the wife may misinterpret it… If the message was received by all - the man, the woman and the boy … people would easily embrace it,” [IDI, Traditional Leader – Dowa]. This view was supported by another community leader who argued that the message must go to both males and females to avoid disagreements between married couples: “The message has to go to both because the women are the ones who stay with the children. They can get them circumcised,” [FGD, Gatekeepers – Dowa]. The study also established that another key entry point into VMMC among the Yao men of Mangochi was the pre-adolescence age bracket of 5 to 10 years. They were minors such that they needed consent from their parents or guardians or those in authority of a village of institution. Two KII participants in Mangochi suggested that VMMC messaging must also target teachers and schools: “…I think that teachers also are very important in their own right because they interact with the community [and learners] in particular. I have a view that teachers can help us clear a lot of misconceptions …,” [KII, Service Providers – Mangochi]. This view point agrees with the observation that one participant in an FGD with Service Providers made on the same matter in Dowa. He lamented that although schools had a lot of HIV/AIDS-related clubs authorities had failed to utilize them to popularized VMMC services among adolescents and youth. “At school we have HIV&AIDS youth clubs but because they were not oriented on [VMMC] services … it is hard for you [as a teacher] to explain it in clear details [to learners] particularly the benefits of VMMC… The main topic in youth clubs is HIV&AIDS so I think if HSAs and teachers were oriented we can pass on the knowledge when we are with the youth clubs ,” [FGD, Service Providers – Dowa]. What are the expected effects (or impact) of messages on VMMC services on the audience and policy? Regarding the expected effects (or impact) of VMMC communication interventions and the VMMC program in general, the study established that the anticipated effects were at four different levels: at personal level, at household level, at community level and at policy level. At personal level, interventions were expected to create awareness of VMMC and how it contributes to HIV prevention in heterosexual men, cervical cancer (among women) and other sexually transmitted infections. There was also an expectation of increased preventive post-circumcision behaviour among VMMC clients: “…one man did not observe the six weeks -sexual-abstinence window like he was instructed by the service provider. They resumed sex earlier and he developed complications,” [KII, Service Providers, Chowe – Mangochi]. In Dowa, the overall anticipated impact of VMMC communication interventions was the reduction of cervical cancer cases that were considered to be on the rise. One sex workers’ expectation had this to say: “The main issue is reduction of diseases; it is the same when the hospital is giving counseling on condoms, they say ‘every man should abstain from sex. However, if they fail then using a condom is Plan B’. It is the same with circumcision. It is Plan B,” [FGD, Commercial Sex Worker - Dowa]. In Dowa, there was also fear that VMMC communication interventions could also lead to rising culture of promiscuity that could lead to increased HIV prevalence: “It [VMMC] is encouraging promiscuity among the youth here because if one boy is told that he is very sweet by a girl he wants to sleep with every girl…,” [FGD, Service Provider – Dowa]. Community members and health personnel working in local health facilities wanted to see household leaders, particularly mothers and clan leaders, to mobilize adult men and boys under their influence for services since they were primary producers of health. Sexual partners, mothers, aunties and grandmothers were considered to have the duty of motivating male members of their households to access services. They were also expected to support circumcised males to adopt post-circumcision prevention measures: “…females are the ones who go to the hospital… when the mother and other females are taught and understand [advantages of VMMC] it will be easy for the child to get circumcised,” [FGD, Community Leaders, Dowa]. Some traditional circumcisers were still reported using one razor blade or knife to circumcise more than one initiate. But, change had already started happening at community level since VMMC communication intervention rolled out in Mangochi: “Two weeks ago I was on a tour … Angaliba themselves were asking us: ‘Please, government should send us circumcision knives for use in initiation camps ,” [IDI, Senior Chief – Mangochi]. In Mangochi, informants felt that interventions enabled Yao Muslims practice jando as prescribed in the Holy Quran. One Yao Islamic cleric posited that VMMC already contributed to the demystification of male circumcision which, in the Yao culture, had been shrouded in secrecy from women, children and uncircumcised males for centuries: “…I am very pleased because what is happening now [in clinics] is real jando as prescribed by the Islamic faith.… VMMC fully agrees with Islam,” [KII, Religious Leaders - Mangochi] . In Dowa district, gatekeepers also expected non-formal institutions such as the gulewankulu supporting the VMMC campaign through information dissemination. They also expected it to lead in the reduction of cancer-related illness and deaths and to harness VMMC clients as advocates for services in their respective communities. In Dowa interventions were also expected to lobby for a more stable availability of funds to support community outreach programs: “In the past we used to go on outreach programs and we would discuss (VMMC) with community members… We have no money for outreach programs,” [FGD, Service Providers– Mponela, Dowa]. Although they completely rejected the proposals to adopt male circumcision as a standard cultural practice in gulewankulu bases called dambwe , the greatest change was that the Chewas were, nonetheless, ready to use gule wankulu characters as crowd puller to community meetings to ensure that VMMC messages quickly diffused into communities. “…They can spread the message via songs since gule wankulu characters have a special talent when it comes to composing songs,” [FGD, Community Leaders – Dowa]. Informants both in Mangochi and Dowa also wanted to see secondary facility (i.e. district-level-hospital) circumcisers and community-level-hospital (i.e. local health facilities) personnel working together when carrying out circumcisions. To achieve this, informants were of the view that the district health system needed to ensure that there was adequate coordination and collaboration by all key stakeholders in the VMMC service delivery value chain in each district. “If, on average, they had at least enrolled one nurse or clinician from each rural facility in the district …These could have been assigned to promote VMMC services at community level,” [KII, Service Provider, Mangochi]. Another service provider in Dowa : “The job is done by the top dogs so… we can’t follow-up because we don’t know how they counsel [clients]. Maybe there isn’t any follow-up [in their plans],” [FGD, Service Providers -Dowa]. Informants also complained that VMMC circumcisions that were offered at health facilities lacked privacy for clients. They expected this to change if the program would be successful. In Mangochi, due to lack of a proper transport arrangements, older men were sometimes being ferried from rural areas to the secondary facility on the same open lorry with boys as young as 5 years. At one facility in Dowa, informants reported that circumcisions were taking place in a room whose entry was inside the maternity ward such that adult men shunned services. Many parents who were willing to get their sons circumcised shunned services. “ … they had no choice due to lack of proper space … [VMMC clients] had to go past a group of pregnant women [in the maternity wing]. Boys were shy. The room should have [had] two doors so that after getting circumcised they should use the other door… ,” [FGD, Service Providers – Dowa]. At two facilities in Mangochi, informants told these researchers that VMMC had not been integrated into other services offered by facility and no special day or time had been allocated to these services since services were just being provided when circumcisers came all the way from Mangochi District Hospital. The other challenge both in Mangochi and Dowa was the long distances to service points. In the FGD with VMMC clients in Dowa said: “Hospitals are very far from our area so it costs a lot of money to go there and sometimes we don’t have the money,” [FGD, VMMC Clients – Dowa] . We first report on the cultural and religious factors that impeded VMMC communication in the two data collection sites. Overall, this study found that royalty to culture and religion prevented people from the two study sites from acting on messages. These messages were being disseminated by various VMMC implementing agencies, predominantly NGOs, with support from various development partners, under the leadership of the Ministry of Health. However, the traditionally circumcising Yao people avoided VMMC services on the fear that it undermined jando, a cultural initiation ritual where boys were circumcised to symbolize transition from childhood to adulthood. VMMC messages and the VMMC program as a whole, were perceived to be in conflict with their cultural identities and, therefore, a threat to the survival of their initiation practices which, for the circumcising Yaos, went beyond the mere cutting of the foreskin but also inculcated character in the initiates: “Ours was the circumcision of the brain not of the penis as they are portraying it now… As Yaos we were circumcised for cleanliness, moral discipline and transition to manhood,” [KII, Yao Culture Expert ]. Jando also had a political significance since boys were prepared for fiduciary duties towards their local chief and enabled them to practice a ritual that was sanctioned by Islamic Scriptures. “…This time around you are telling the chiefs that circumcision should be done at the hospital. The chief cannot promote this because he is benefiting nothing ….,” [ KII, Yao Cultural Expert & Sheikh]. On the other hand, being a predominantly traditionally non-circumcising and Christian ethnic group, the Chewa people did not pay much attention to VMMC messages. Their understanding was twofold: first, male circumcision was not part of their culture. Secondly, male circumcision was not part of their religion but an attempt by government to popularize Islam in their area. Informants, therefore, equated getting circumcised with becoming a Yao or a Muslim. One man said: “ I am a Chewa man, why should I go for circumcision? Do you want me to become a Yao?” Another one was more conscious of his faith: “ As Christians, we do not make [put] emphasis on circumcision,” [IDI, Religious leader - Dowa]. These researchers posit, therefore, that cultural and religious perception, accounted for low uptake of VMMC services in the two data collection communities. Based on Harold Laswell’s fivefold question we now present results of the study as follows: What is the preferred source of information for VMMC information? Both cultures under study perceived the chief as a primary source of information and a mouth-piece of the community. Hence any information to be delivered to the community needed to pass through and vetted by him or her. Despite that among the Yao allegiance was determined by one’s economic standing, among the Chewa it was more of the cultural royalty. From that end, the study informants first noted that promoters of VMMC had overlooked this important community entry protocol thereby creating a communication barrier. “They must tell the chief, [and] the chief will announce to his people that the hospital personnel have something to say, so everyone will go to the meeting…The chief will open the floor and hand it over to the hospital staff to address the people.” [FGD, Unmarried Youth – Dowa]. From the foregoing, the chief was traditionally, the primary source of information. However, for lack of technical knowledge and skills on the subject of VMMC services study participants in both communities identified health workers as a preferred source of information. Study participants held that health workers were well trained hence knowledgeable, experienced and holding reliable information. “…medical personnel should be assigned to spread the message of circumcision because they are well trained. If they were to convey the message it will be better understood,” [IDI, Traditional Leader – Dowa]. What is the preferred message for VMMC communication interventions? The most preferred message that communities want to hear from those implementing VMMC interventions is benefits that VMMC offers to communities, households, and men and boys in particular. Regarding pre-adolescents, parents, clan leaders, community and religious leaders from Mangochi district indicated that they needed to be schooled on how VMMC provided long-term health benefits to males. For example, they needed to learn about such long-term benefits as the reduction of the risk of contracting HIV during traditional circumcision rituals during ndagala . Parents and guardians also wanted assurance that VMMC offered their wards protection against HIV infection during heterosexual relationships before and after they began to raise their future families. An unmarried youth in Dowa narrated as follows: “ I think the most vital information is that people should know the benefits… of circumcision and to my understanding circumcision primarily protects you from cervical cancer so if they are told the women will encourage the men to go,” [ FP1, Dowa Unmarried Youth]. A female respondent in Dowa was of the view that men were not provided with sufficient details regarding VMMC procedure. From her perspective, VMMC messages needed to clearly explain what exactly made a circumcised person different from a non-circumcised man. She observed that VMMC messages had scanty details on what the circumcision procedure involved and what really happened to the man. She argued that men, who were the target of opportunity for VMMC interventions, were blank on details of the procedure: “…We just said they circumcise each other but we did not really know what they really cut. People just say the ‘foreskin’ of the penis others just say the ‘[head of the] penis’ so when you think about it, eeh! It’s scaring.” [FGD, Gate Keepers - Dowa]. The study established that the role of VMMC in the prevention of cervical cancer was more attractive than the role that it played in the reduction of HIV infection since everybody in the area knew at least one woman who had died of cervical cancer. They, therefore, wanted messages to focus on how VMMC would protect females from cervical cancer. “The protection from cervical cancer in women is most important in this area…we don’t have the desire for promiscuity and our women don’t have the desire of seeking sexual satisfaction in bed from circumcised men… Let us put HIV aside, currently the lives of people are at risk. You can manage HIV when infected, but there is no cure for cancer,” [FGD, Traditional Leaders - Dowa]. Female participants wanted VMMC messages to focus more on how their sons would be circumcised by trained medical personnel, which would result in less pain, loss of blood and reduce the risk of complications. “…in the past we had STIs like chindoko mabomu but they were treatable but these days there is AIDS which is very dangerous. There is [cervical] cancer and chisonono too. VMMC is helpful since we hear that it protects our youth and their [future] families from these. These are messages that we need hear about,” [FGD, Adult Women – Mangochi]. VMMC messages also needed to fight against stigma and must assure both Christians and Muslims that VMMC did not contradict the Bible and the Qur’an respectively. One Sheikh, for example, was pleased with VMMC, as a religious authority: “…when I look at the current situation as an Islamic leader, I am very pleased because what is happening now [in clinics] is real jando as prescribed by the Islamic faith.… VMMC fully agrees with Islam.” [KII, Religious Leaders - Mangochi] What is the preferred channel for VMMC communication interventions? Regarding the channel of communication for VMMC intervention, many informants reported that the media, particularly radio was prominently used in both study communities. “…to avoid the huge cost [that would come with village meetings] then I can only agree with the media approaches that are being used currently… just think about how MBC Radio One is boasting that 87% of the Malawi population are able to hear them… Now if 87% are listening to MBC [alone] how about those listening to Zodiak?” [IDI, Religious Leader – Mangochi]. However, not many informants were excited about the use of such open channels to promote male circumcision both in Dowa and Mangochi district. In the two districts the use of open media channels was frowned upon for infringing on the secrecy that surrounds rites of passage rituals. First, male circumcision among the Yao was not a subject for public discussion. Second, although both the Quran and the Bible legitimized it, the praxis in the Muslim Yao community was to veil it in secrecy since those not circumcised and females, in particular, were not supposed to know what happened out in the bush: “In the past, circumcision was a total secret such that even the boy’s mother did not know why her child was going to ndagala . The majority of the mothers celebrated jando blindly. And when we were there they strictly told us not to tell anybody what had happened to us. They told us that when we went back home we must bath with pants on lest the mother notices the difference in the appearance of the member….” [KII, Religious Leaders - Mangochi]. Another key concern with the use of radio, television and other forms of mass media was the sexual explicitness that such messages were associated with. This concern was particularly common among adult men and women who feared that such messages would have negative consequences since they promoted sexual immorality among the youth. This, they argued, was contrary to the purpose of male circumcision in the Yao culture: “The problem comes in because … [through the media] you are telling people that once you have been circumcised, you are free from HIV. These are wrong messages…You will [now] hear a man telling a woman that ‘I was circumcised you can come for sex I can’t contract diseases,’” [KII, Yao Culture Expert - Mangochi]. Informants’ most preferred communication channel both among the Yaos of Mangochi and the Chewas of Dowa district was face-to-face engagement meetings attended by health promotion experts on one hand and community leaders, cultural gatekeepers and community members on the other. Among these methods were: meetings where traditional leaders would bring their subjects to one place and engage them in a dialogue; cultural performances such as gule wankulu; religious gatherings such as regular weekly services of worship; interactive drama and door-to-door visits just to mention some. The key reasons for preference were twofold: first, unlike the mass media, they provided room to control the type of audience to hear the message or not. Secondly, they provided an opportunity for the audience to ask questions and seek clarifications as messages are delivered. Engagement meetings organized by traditional leaders were important since chiefs and their councils of advisors acted as a bridge between citizens and government and its stakeholders: “… gulewankulu characters can dance then the drama can come next to teach things like these or you can introduce the topic at first so that the people can have an idea so when the drama on circumcision comes it will entertain them,” [FGD, Gatekeepers – Dowa]. Although informants from all Protestant Christian churches including the Catholic leadership in Dowa sounded very negative about using the church to promote VMMC messages, in Mangochi the Catholic Church was already disseminating VMMC messages. I was also carrying out surgeries. One gulupa said: “ … [Children] are also assisted to get circumcised. When they get to that stage, they are kept at the parish, which is like a simba ….” In Chowe area, Mangochi, FGD participants proposed that VMMC messages should be disseminated through their mosques and churches. This is how one traditional leader put it: “In this community, Churches and Mosques are more ideal. Religious gatherings are easy places to disseminate VMMC information,” [FGD, Traditional Leaders, Mangochi]. In Mangochi, a representative of an NGO added that they used males from within the community to deliver VMMC messages to fellow men. The health worker posited that men tended to listen better to health messages when delivered by their peers. “ We have community mobilizers who are recruited from within communities. They reach out to men in the communities,” [KII Participant, NGO Worker – Mangochi]. Another NGO worker said: “Our main approach with demand creation is interpersonal communication.” [KII Participant, NGO Worker - Mangochi] . What is the preferred audience for VMMC messaging? The study established that VMMC messages should target all community members including men, women and boys to avoid misunderstandings that may arise from misinterpretation of the messages particularly among married couples. This is what one IDI participant said: “These days I think that everyone [should be targeted] because, in a family, the man may hear the message alone but when he brings the message into his household the wife may misinterpret it… If the message was received by all - the man, the woman and the boy … people would easily embrace it,” [IDI, Traditional Leader – Dowa]. This view was supported by another community leader who argued that the message must go to both males and females to avoid disagreements between married couples: “The message has to go to both because the women are the ones who stay with the children. They can get them circumcised,” [FGD, Gatekeepers – Dowa]. The study also established that another key entry point into VMMC among the Yao men of Mangochi was the pre-adolescence age bracket of 5 to 10 years. They were minors such that they needed consent from their parents or guardians or those in authority of a village of institution. Two KII participants in Mangochi suggested that VMMC messaging must also target teachers and schools: “…I think that teachers also are very important in their own right because they interact with the community [and learners] in particular. I have a view that teachers can help us clear a lot of misconceptions …,” [KII, Service Providers – Mangochi]. This view point agrees with the observation that one participant in an FGD with Service Providers made on the same matter in Dowa. He lamented that although schools had a lot of HIV/AIDS-related clubs authorities had failed to utilize them to popularized VMMC services among adolescents and youth. “At school we have HIV&AIDS youth clubs but because they were not oriented on [VMMC] services … it is hard for you [as a teacher] to explain it in clear details [to learners] particularly the benefits of VMMC… The main topic in youth clubs is HIV&AIDS so I think if HSAs and teachers were oriented we can pass on the knowledge when we are with the youth clubs ,” [FGD, Service Providers – Dowa]. What are the expected effects (or impact) of messages on VMMC services on the audience and policy? Regarding the expected effects (or impact) of VMMC communication interventions and the VMMC program in general, the study established that the anticipated effects were at four different levels: at personal level, at household level, at community level and at policy level. At personal level, interventions were expected to create awareness of VMMC and how it contributes to HIV prevention in heterosexual men, cervical cancer (among women) and other sexually transmitted infections. There was also an expectation of increased preventive post-circumcision behaviour among VMMC clients: “…one man did not observe the six weeks -sexual-abstinence window like he was instructed by the service provider. They resumed sex earlier and he developed complications,” [KII, Service Providers, Chowe – Mangochi]. In Dowa, the overall anticipated impact of VMMC communication interventions was the reduction of cervical cancer cases that were considered to be on the rise. One sex workers’ expectation had this to say: “The main issue is reduction of diseases; it is the same when the hospital is giving counseling on condoms, they say ‘every man should abstain from sex. However, if they fail then using a condom is Plan B’. It is the same with circumcision. It is Plan B,” [FGD, Commercial Sex Worker - Dowa]. In Dowa, there was also fear that VMMC communication interventions could also lead to rising culture of promiscuity that could lead to increased HIV prevalence: “It [VMMC] is encouraging promiscuity among the youth here because if one boy is told that he is very sweet by a girl he wants to sleep with every girl…,” [FGD, Service Provider – Dowa]. Community members and health personnel working in local health facilities wanted to see household leaders, particularly mothers and clan leaders, to mobilize adult men and boys under their influence for services since they were primary producers of health. Sexual partners, mothers, aunties and grandmothers were considered to have the duty of motivating male members of their households to access services. They were also expected to support circumcised males to adopt post-circumcision prevention measures: “…females are the ones who go to the hospital… when the mother and other females are taught and understand [advantages of VMMC] it will be easy for the child to get circumcised,” [FGD, Community Leaders, Dowa]. Some traditional circumcisers were still reported using one razor blade or knife to circumcise more than one initiate. But, change had already started happening at community level since VMMC communication intervention rolled out in Mangochi: “Two weeks ago I was on a tour … Angaliba themselves were asking us: ‘Please, government should send us circumcision knives for use in initiation camps ,” [IDI, Senior Chief – Mangochi]. In Mangochi, informants felt that interventions enabled Yao Muslims practice jando as prescribed in the Holy Quran. One Yao Islamic cleric posited that VMMC already contributed to the demystification of male circumcision which, in the Yao culture, had been shrouded in secrecy from women, children and uncircumcised males for centuries: “…I am very pleased because what is happening now [in clinics] is real jando as prescribed by the Islamic faith.… VMMC fully agrees with Islam,” [KII, Religious Leaders - Mangochi] . In Dowa district, gatekeepers also expected non-formal institutions such as the gulewankulu supporting the VMMC campaign through information dissemination. They also expected it to lead in the reduction of cancer-related illness and deaths and to harness VMMC clients as advocates for services in their respective communities. In Dowa interventions were also expected to lobby for a more stable availability of funds to support community outreach programs: “In the past we used to go on outreach programs and we would discuss (VMMC) with community members… We have no money for outreach programs,” [FGD, Service Providers– Mponela, Dowa]. Although they completely rejected the proposals to adopt male circumcision as a standard cultural practice in gulewankulu bases called dambwe , the greatest change was that the Chewas were, nonetheless, ready to use gule wankulu characters as crowd puller to community meetings to ensure that VMMC messages quickly diffused into communities. “…They can spread the message via songs since gule wankulu characters have a special talent when it comes to composing songs,” [FGD, Community Leaders – Dowa]. Informants both in Mangochi and Dowa also wanted to see secondary facility (i.e. district-level-hospital) circumcisers and community-level-hospital (i.e. local health facilities) personnel working together when carrying out circumcisions. To achieve this, informants were of the view that the district health system needed to ensure that there was adequate coordination and collaboration by all key stakeholders in the VMMC service delivery value chain in each district. “If, on average, they had at least enrolled one nurse or clinician from each rural facility in the district …These could have been assigned to promote VMMC services at community level,” [KII, Service Provider, Mangochi]. Another service provider in Dowa : “The job is done by the top dogs so… we can’t follow-up because we don’t know how they counsel [clients]. Maybe there isn’t any follow-up [in their plans],” [FGD, Service Providers -Dowa]. Informants also complained that VMMC circumcisions that were offered at health facilities lacked privacy for clients. They expected this to change if the program would be successful. In Mangochi, due to lack of a proper transport arrangements, older men were sometimes being ferried from rural areas to the secondary facility on the same open lorry with boys as young as 5 years. At one facility in Dowa, informants reported that circumcisions were taking place in a room whose entry was inside the maternity ward such that adult men shunned services. Many parents who were willing to get their sons circumcised shunned services. “ … they had no choice due to lack of proper space … [VMMC clients] had to go past a group of pregnant women [in the maternity wing]. Boys were shy. The room should have [had] two doors so that after getting circumcised they should use the other door… ,” [FGD, Service Providers – Dowa]. At two facilities in Mangochi, informants told these researchers that VMMC had not been integrated into other services offered by facility and no special day or time had been allocated to these services since services were just being provided when circumcisers came all the way from Mangochi District Hospital. The other challenge both in Mangochi and Dowa was the long distances to service points. In the FGD with VMMC clients in Dowa said: “Hospitals are very far from our area so it costs a lot of money to go there and sometimes we don’t have the money,” [FGD, VMMC Clients – Dowa] . Both cultures under study perceived the chief as a primary source of information and a mouth-piece of the community. Hence any information to be delivered to the community needed to pass through and vetted by him or her. Despite that among the Yao allegiance was determined by one’s economic standing, among the Chewa it was more of the cultural royalty. From that end, the study informants first noted that promoters of VMMC had overlooked this important community entry protocol thereby creating a communication barrier. “They must tell the chief, [and] the chief will announce to his people that the hospital personnel have something to say, so everyone will go to the meeting…The chief will open the floor and hand it over to the hospital staff to address the people.” [FGD, Unmarried Youth – Dowa]. From the foregoing, the chief was traditionally, the primary source of information. However, for lack of technical knowledge and skills on the subject of VMMC services study participants in both communities identified health workers as a preferred source of information. Study participants held that health workers were well trained hence knowledgeable, experienced and holding reliable information. “…medical personnel should be assigned to spread the message of circumcision because they are well trained. If they were to convey the message it will be better understood,” [IDI, Traditional Leader – Dowa]. The most preferred message that communities want to hear from those implementing VMMC interventions is benefits that VMMC offers to communities, households, and men and boys in particular. Regarding pre-adolescents, parents, clan leaders, community and religious leaders from Mangochi district indicated that they needed to be schooled on how VMMC provided long-term health benefits to males. For example, they needed to learn about such long-term benefits as the reduction of the risk of contracting HIV during traditional circumcision rituals during ndagala . Parents and guardians also wanted assurance that VMMC offered their wards protection against HIV infection during heterosexual relationships before and after they began to raise their future families. An unmarried youth in Dowa narrated as follows: “ I think the most vital information is that people should know the benefits… of circumcision and to my understanding circumcision primarily protects you from cervical cancer so if they are told the women will encourage the men to go,” [ FP1, Dowa Unmarried Youth]. A female respondent in Dowa was of the view that men were not provided with sufficient details regarding VMMC procedure. From her perspective, VMMC messages needed to clearly explain what exactly made a circumcised person different from a non-circumcised man. She observed that VMMC messages had scanty details on what the circumcision procedure involved and what really happened to the man. She argued that men, who were the target of opportunity for VMMC interventions, were blank on details of the procedure: “…We just said they circumcise each other but we did not really know what they really cut. People just say the ‘foreskin’ of the penis others just say the ‘[head of the] penis’ so when you think about it, eeh! It’s scaring.” [FGD, Gate Keepers - Dowa]. The study established that the role of VMMC in the prevention of cervical cancer was more attractive than the role that it played in the reduction of HIV infection since everybody in the area knew at least one woman who had died of cervical cancer. They, therefore, wanted messages to focus on how VMMC would protect females from cervical cancer. “The protection from cervical cancer in women is most important in this area…we don’t have the desire for promiscuity and our women don’t have the desire of seeking sexual satisfaction in bed from circumcised men… Let us put HIV aside, currently the lives of people are at risk. You can manage HIV when infected, but there is no cure for cancer,” [FGD, Traditional Leaders - Dowa]. Female participants wanted VMMC messages to focus more on how their sons would be circumcised by trained medical personnel, which would result in less pain, loss of blood and reduce the risk of complications. “…in the past we had STIs like chindoko mabomu but they were treatable but these days there is AIDS which is very dangerous. There is [cervical] cancer and chisonono too. VMMC is helpful since we hear that it protects our youth and their [future] families from these. These are messages that we need hear about,” [FGD, Adult Women – Mangochi]. VMMC messages also needed to fight against stigma and must assure both Christians and Muslims that VMMC did not contradict the Bible and the Qur’an respectively. One Sheikh, for example, was pleased with VMMC, as a religious authority: “…when I look at the current situation as an Islamic leader, I am very pleased because what is happening now [in clinics] is real jando as prescribed by the Islamic faith.… VMMC fully agrees with Islam.” [KII, Religious Leaders - Mangochi] Regarding the channel of communication for VMMC intervention, many informants reported that the media, particularly radio was prominently used in both study communities. “…to avoid the huge cost [that would come with village meetings] then I can only agree with the media approaches that are being used currently… just think about how MBC Radio One is boasting that 87% of the Malawi population are able to hear them… Now if 87% are listening to MBC [alone] how about those listening to Zodiak?” [IDI, Religious Leader – Mangochi]. However, not many informants were excited about the use of such open channels to promote male circumcision both in Dowa and Mangochi district. In the two districts the use of open media channels was frowned upon for infringing on the secrecy that surrounds rites of passage rituals. First, male circumcision among the Yao was not a subject for public discussion. Second, although both the Quran and the Bible legitimized it, the praxis in the Muslim Yao community was to veil it in secrecy since those not circumcised and females, in particular, were not supposed to know what happened out in the bush: “In the past, circumcision was a total secret such that even the boy’s mother did not know why her child was going to ndagala . The majority of the mothers celebrated jando blindly. And when we were there they strictly told us not to tell anybody what had happened to us. They told us that when we went back home we must bath with pants on lest the mother notices the difference in the appearance of the member….” [KII, Religious Leaders - Mangochi]. Another key concern with the use of radio, television and other forms of mass media was the sexual explicitness that such messages were associated with. This concern was particularly common among adult men and women who feared that such messages would have negative consequences since they promoted sexual immorality among the youth. This, they argued, was contrary to the purpose of male circumcision in the Yao culture: “The problem comes in because … [through the media] you are telling people that once you have been circumcised, you are free from HIV. These are wrong messages…You will [now] hear a man telling a woman that ‘I was circumcised you can come for sex I can’t contract diseases,’” [KII, Yao Culture Expert - Mangochi]. Informants’ most preferred communication channel both among the Yaos of Mangochi and the Chewas of Dowa district was face-to-face engagement meetings attended by health promotion experts on one hand and community leaders, cultural gatekeepers and community members on the other. Among these methods were: meetings where traditional leaders would bring their subjects to one place and engage them in a dialogue; cultural performances such as gule wankulu; religious gatherings such as regular weekly services of worship; interactive drama and door-to-door visits just to mention some. The key reasons for preference were twofold: first, unlike the mass media, they provided room to control the type of audience to hear the message or not. Secondly, they provided an opportunity for the audience to ask questions and seek clarifications as messages are delivered. Engagement meetings organized by traditional leaders were important since chiefs and their councils of advisors acted as a bridge between citizens and government and its stakeholders: “… gulewankulu characters can dance then the drama can come next to teach things like these or you can introduce the topic at first so that the people can have an idea so when the drama on circumcision comes it will entertain them,” [FGD, Gatekeepers – Dowa]. Although informants from all Protestant Christian churches including the Catholic leadership in Dowa sounded very negative about using the church to promote VMMC messages, in Mangochi the Catholic Church was already disseminating VMMC messages. I was also carrying out surgeries. One gulupa said: “ … [Children] are also assisted to get circumcised. When they get to that stage, they are kept at the parish, which is like a simba ….” In Chowe area, Mangochi, FGD participants proposed that VMMC messages should be disseminated through their mosques and churches. This is how one traditional leader put it: “In this community, Churches and Mosques are more ideal. Religious gatherings are easy places to disseminate VMMC information,” [FGD, Traditional Leaders, Mangochi]. In Mangochi, a representative of an NGO added that they used males from within the community to deliver VMMC messages to fellow men. The health worker posited that men tended to listen better to health messages when delivered by their peers. “ We have community mobilizers who are recruited from within communities. They reach out to men in the communities,” [KII Participant, NGO Worker – Mangochi]. Another NGO worker said: “Our main approach with demand creation is interpersonal communication.” [KII Participant, NGO Worker - Mangochi] . The study established that VMMC messages should target all community members including men, women and boys to avoid misunderstandings that may arise from misinterpretation of the messages particularly among married couples. This is what one IDI participant said: “These days I think that everyone [should be targeted] because, in a family, the man may hear the message alone but when he brings the message into his household the wife may misinterpret it… If the message was received by all - the man, the woman and the boy … people would easily embrace it,” [IDI, Traditional Leader – Dowa]. This view was supported by another community leader who argued that the message must go to both males and females to avoid disagreements between married couples: “The message has to go to both because the women are the ones who stay with the children. They can get them circumcised,” [FGD, Gatekeepers – Dowa]. The study also established that another key entry point into VMMC among the Yao men of Mangochi was the pre-adolescence age bracket of 5 to 10 years. They were minors such that they needed consent from their parents or guardians or those in authority of a village of institution. Two KII participants in Mangochi suggested that VMMC messaging must also target teachers and schools: “…I think that teachers also are very important in their own right because they interact with the community [and learners] in particular. I have a view that teachers can help us clear a lot of misconceptions …,” [KII, Service Providers – Mangochi]. This view point agrees with the observation that one participant in an FGD with Service Providers made on the same matter in Dowa. He lamented that although schools had a lot of HIV/AIDS-related clubs authorities had failed to utilize them to popularized VMMC services among adolescents and youth. “At school we have HIV&AIDS youth clubs but because they were not oriented on [VMMC] services … it is hard for you [as a teacher] to explain it in clear details [to learners] particularly the benefits of VMMC… The main topic in youth clubs is HIV&AIDS so I think if HSAs and teachers were oriented we can pass on the knowledge when we are with the youth clubs ,” [FGD, Service Providers – Dowa]. Regarding the expected effects (or impact) of VMMC communication interventions and the VMMC program in general, the study established that the anticipated effects were at four different levels: at personal level, at household level, at community level and at policy level. At personal level, interventions were expected to create awareness of VMMC and how it contributes to HIV prevention in heterosexual men, cervical cancer (among women) and other sexually transmitted infections. There was also an expectation of increased preventive post-circumcision behaviour among VMMC clients: “…one man did not observe the six weeks -sexual-abstinence window like he was instructed by the service provider. They resumed sex earlier and he developed complications,” [KII, Service Providers, Chowe – Mangochi]. In Dowa, the overall anticipated impact of VMMC communication interventions was the reduction of cervical cancer cases that were considered to be on the rise. One sex workers’ expectation had this to say: “The main issue is reduction of diseases; it is the same when the hospital is giving counseling on condoms, they say ‘every man should abstain from sex. However, if they fail then using a condom is Plan B’. It is the same with circumcision. It is Plan B,” [FGD, Commercial Sex Worker - Dowa]. In Dowa, there was also fear that VMMC communication interventions could also lead to rising culture of promiscuity that could lead to increased HIV prevalence: “It [VMMC] is encouraging promiscuity among the youth here because if one boy is told that he is very sweet by a girl he wants to sleep with every girl…,” [FGD, Service Provider – Dowa]. Community members and health personnel working in local health facilities wanted to see household leaders, particularly mothers and clan leaders, to mobilize adult men and boys under their influence for services since they were primary producers of health. Sexual partners, mothers, aunties and grandmothers were considered to have the duty of motivating male members of their households to access services. They were also expected to support circumcised males to adopt post-circumcision prevention measures: “…females are the ones who go to the hospital… when the mother and other females are taught and understand [advantages of VMMC] it will be easy for the child to get circumcised,” [FGD, Community Leaders, Dowa]. Some traditional circumcisers were still reported using one razor blade or knife to circumcise more than one initiate. But, change had already started happening at community level since VMMC communication intervention rolled out in Mangochi: “Two weeks ago I was on a tour … Angaliba themselves were asking us: ‘Please, government should send us circumcision knives for use in initiation camps ,” [IDI, Senior Chief – Mangochi]. In Mangochi, informants felt that interventions enabled Yao Muslims practice jando as prescribed in the Holy Quran. One Yao Islamic cleric posited that VMMC already contributed to the demystification of male circumcision which, in the Yao culture, had been shrouded in secrecy from women, children and uncircumcised males for centuries: “…I am very pleased because what is happening now [in clinics] is real jando as prescribed by the Islamic faith.… VMMC fully agrees with Islam,” [KII, Religious Leaders - Mangochi] . In Dowa district, gatekeepers also expected non-formal institutions such as the gulewankulu supporting the VMMC campaign through information dissemination. They also expected it to lead in the reduction of cancer-related illness and deaths and to harness VMMC clients as advocates for services in their respective communities. In Dowa interventions were also expected to lobby for a more stable availability of funds to support community outreach programs: “In the past we used to go on outreach programs and we would discuss (VMMC) with community members… We have no money for outreach programs,” [FGD, Service Providers– Mponela, Dowa]. Although they completely rejected the proposals to adopt male circumcision as a standard cultural practice in gulewankulu bases called dambwe , the greatest change was that the Chewas were, nonetheless, ready to use gule wankulu characters as crowd puller to community meetings to ensure that VMMC messages quickly diffused into communities. “…They can spread the message via songs since gule wankulu characters have a special talent when it comes to composing songs,” [FGD, Community Leaders – Dowa]. Informants both in Mangochi and Dowa also wanted to see secondary facility (i.e. district-level-hospital) circumcisers and community-level-hospital (i.e. local health facilities) personnel working together when carrying out circumcisions. To achieve this, informants were of the view that the district health system needed to ensure that there was adequate coordination and collaboration by all key stakeholders in the VMMC service delivery value chain in each district. “If, on average, they had at least enrolled one nurse or clinician from each rural facility in the district …These could have been assigned to promote VMMC services at community level,” [KII, Service Provider, Mangochi]. Another service provider in Dowa : “The job is done by the top dogs so… we can’t follow-up because we don’t know how they counsel [clients]. Maybe there isn’t any follow-up [in their plans],” [FGD, Service Providers -Dowa]. Informants also complained that VMMC circumcisions that were offered at health facilities lacked privacy for clients. They expected this to change if the program would be successful. In Mangochi, due to lack of a proper transport arrangements, older men were sometimes being ferried from rural areas to the secondary facility on the same open lorry with boys as young as 5 years. At one facility in Dowa, informants reported that circumcisions were taking place in a room whose entry was inside the maternity ward such that adult men shunned services. Many parents who were willing to get their sons circumcised shunned services. “ … they had no choice due to lack of proper space … [VMMC clients] had to go past a group of pregnant women [in the maternity wing]. Boys were shy. The room should have [had] two doors so that after getting circumcised they should use the other door… ,” [FGD, Service Providers – Dowa]. At two facilities in Mangochi, informants told these researchers that VMMC had not been integrated into other services offered by facility and no special day or time had been allocated to these services since services were just being provided when circumcisers came all the way from Mangochi District Hospital. The other challenge both in Mangochi and Dowa was the long distances to service points. In the FGD with VMMC clients in Dowa said: “Hospitals are very far from our area so it costs a lot of money to go there and sometimes we don’t have the money,” [FGD, VMMC Clients – Dowa] . Both the Chewa and Yao people performed rites of passage with the same purpose of shaping the character of boys and transitioning them from childhood to adulthood. Nonetheless, s they differed in the rituals they performed for both boys’ initiation rites. While the Chewa initiated boys throughgulewamkulu the Yao people initiated them through jando where male circumcision was administered both for cultural and religious beliefs. The negative attitude to VMMC observable among the two ethnic groups is the major reason why there is low uptake of VMMC services both among the Chewa people of Dowa and the Yao people of Mangochi. They feared that the procedure was in conflict with their cultural identities, traditions and value systems. In particular, both data collection communities perceived VMMC as lacking in character formation which was the main drive behind rites of passage rituals in their areas. However, the two cultures also differed on motivators for VMMC services. A fraction of the Yao people accepted VMMC because they did not have to pay anything to access services, except meeting transport costs, reduced pain, quicker healing of wounds, reduced risk of contracting HIV due to the use of sterilized razor blades and receiving services in a hygienic environment. Some Chewa people, on the other hand, were motivated to access VMMC services mainly because of its perceived ability to protect women from cervical cancer. On culturally-preferred communication approaches, generally, both study communities opted for interpersonal face-to-face communication approaches since they gave them the opportunity to interface with health promoters and provide real-time feedback on the messages that they disseminated. The study also established that since the two ethnic groups were different from each other culturally and religiously such that, although they had similar communication needs in many respects, these two groups had different communication needs. This phenomenon suggested that communication programs should not be generic but must be designed to address the specific needs of each treatment group. Regarding the source of VMMC messages , informants in both study groups posited that trained health workers were preferred. Health workers were preferred to disseminate such information because they were considered trained and well experience in disseminating health information hence expected to possess the required competence to carry out the task of reaching out to various audiences with VMMC messages. This finding is also consistent with other studies that have been conducted in Malawi and elsewhere on VMMC communication and have shown that health personnel are among the most trusted and preferred sources of information among males . The Ministry of Health in Kenya, for example, prioritized capacity building of health personnel including those involved in health communication . It was intriguing, however, to learn that much as informants preferred health workers as sources of VMMC messages, study participants both in Mangochi and Dowa were of the view that community entry must be done in consultation with leaders of their communities. Thus, iformation delivery must take place in the presence of chiefs or anyone delegated to stand in. The study demonstrated that health workers needed to work with chiefs in order to create a platform on which the message can be disseminated to the maximum number of people in the area. The contact between the health experts and traditional leaders provides the first opportunity for establishing a shared vision between health experts and the target audience. This suggest that the starting point for designing a good health communication campaign is the establishment of a shared vision between the health promoters and the target audience . This finding agrees with Lozare et al. who positulated that since the primary producers of health are the people themselves there is need to engage them in the most meaningful way . In the Malawi setting, when the chief and the rest of the cultural structures reject a message, the whole community follow suit. Generally, informants approved the current practice where health workers were at the center of message dissemination. However, they expressed worry that, in the current set up VMMC workers from the district hospital or the NGO partners, invaded their villages and communities with loud hailing before adequately engaging traditional leaders, cultural gatekeepers and without building partnerships with cultural structures that carried out similar work in the local community. They further deplored the failure by health experts to hold community meetings through which community members would have a chance to ask questions and seek clarification on aspects of VMMC that were not clear. Regarding messaging , informants wanted messages that were contextual and unambiguous and promoted HIV prevention. They expressed worry that although VMMC had the potential of positively contributing to the reduction of HIV prevalence sexually explicit messages, which were frequently heard in the media, were perceived to have the potential of increasing promiscuity among the youth, a phenomenon that could potentially worsen the HIV situation in their areas. This notion reflects the general Malawian worldview that the function of sex surpasses the procreation and sensual function as it connotes survival of the extended family, clan and tribe . In his writings Augustine Musopole argues that, in Africa sexuality is one of those vital forces that make life secure, meaningful and worthwhile . He, therefore, advises that it is largely a taboo to openly discuss sex and sexuality across many African cultures where adolescents were separated from the community and taken to the bush to be taught sexual rituals, moral values, and cultural norms . In this study informants generally agreed that messages that put emphasis on the sexual pleasure as a benefit of VMMC services were not welcome to the majority of Malawians. Both in Dowa and Mangochi, older informants expressed concern that sexual pleasure was dominating the VMMC messaging instead of focusing on HIV prevention outcomes. These findings add value to the conclusions of the study in Malawi by Patrick Makono et al. who observed that lack of a proper research prior to the design of Malawi’s Ndife Otsogola VMMC brand, resulted in messages that failed to effectively promote the uptake of VMMC services among men and boys in Malawi . Both study sites preferred messages that focused on health benefits of VMMC such as the prevention of HIV and cervical cancer. The Chewa people, on one hand, were almost complacent about their vulnerability to HIV infection because they did not practice polygamy and their instruction, both in the cultural and religious initiation rituals, was perceived to be strong on its emphasis on sexual abstinence. They, therefore, felt that if VMMC messages were to appeal to the majority of them, they needed to create an awareness of how male circumcision contributed to the reduction of deaths due to cervical cancer among women. Yao informants, on the other hand, posited that both HIV and cervical cancer prevention was a motivating factor. They, therefore, preferred messages that created the awareness, that facility circumcision service reduced pain, ensured early healing of the wound and the risk of post-op complications. They also preferred messages that portrayed VMMC services as being cheaper and more hygienic than traditional circumcision. Findings on the need for benefit-centered messages and their indispensability in promoting the uptake of VMMC services are similar to outcomes of other studies conducted in Malawi. For example, Mhagama et al. and Makono, et al. established that men and boys will opt for medical circumcision after obtaining benefit-centered information from medical personnel and personalized information from their peers . In a study conducted in KwaZulu-Natal in South Africa by George Gavin et al. established that knowledge-rich personalized-information was a vital facilitator of service uptake in South Africa . Another study carried out in Rakai, Uganda, by Ssekubugu, et al., demonstrated that males acted on VMMC messages that emphasize on health benefits . Regarding channels of communication , radio came out prominently as a common channel that government agencies and NGOs were using to disseminate messages on VMMC services. However, as Tilson et al. have noted, one key obstacle to effective communication in the past pertained to conceptualizing communication as a simple one-way transmission from the source to a receiver with the intention of producing some effect . Many study participants pointed out that radio and loud hailing was not interactive enough because it did not provide them with an opportunity to ask questions and seek clarification on the messages that it delivered to listeners. They, instead, preferred community-based interpersonal communication to any other form of communication. For example, the Chewa people of Mwancheka area, in Dowa district, preferred approaches that brought families and communities together. Messages could therefore be delivered during funeral receptions, religious functions, drama performances and gule wankulu functions, just to mention a few. One community leader said: “Even when there is a funeral the chief can stand [up] and deliver information on VMMC.” Study participants in Dowa were completely opposed to mainstreaming male circumcision into the male-dominated gulewankulu. However, they were open to the use of gulewamkulu to accelerate the dissemination of messages on VMMC services in their communities. It is for this reason that communication channels that enabled families and communities to collectively have direct access to VMMC messages in an interactive manner were preferred. This view point was consistent with Benjamin Lozare’s assertion that since the households and communities were the primary producers of health, communication approaches that sought to empower such units needed to be prioritized . About the audience , the key problem with VMMC campaigns in Malawi, according to informants from both study ethnic groups, was the focus on the individual male rather than every ‘individual citizen’ of the community in the VMMC messaging. It was recommended by study communities that VMMC messages must target the man, the boy, the woman, and all social structures in the social-ecology. They argued that the message must be directed to close family members, peers of eligible men and boys, sexual partners and caregivers who happen to be females generally. Decisions either to access VMMC or not are not entirely personal for men or boys. Like it was discovered in Uganda, this study, both in Dowa and Mangochi, found that for married men, female sexual partners had a strong say on the decision a man makes . This finding also agrees with studies conducted in Papua New Guinea, Zimbabwe and Botswana. In these studies, it was observed that women played a significant role to improve uptake of voluntary medical male circumcision (VMMC) and in the post-op behaviors of the client . These studies established that women could motivate men, particularly their sexual partners, to access VMMC services. Women were willing to engage their husbands on the benefits of VMMC such as improved penile hygiene, reduced risk of HIV transmission and other sexually transmitted infections, as well as increased sexual satisfaction . In the matrilineal marriage system which is generally practiced in Malawi and in the study areas i.e. Mangochi and Dowa, extended family members particularly uncles, aunts, and grandmothers, as well as clan heads (eni mbumba) , have a great deal of influence in decision making processes at household level . Among the Chewa people of Dowa for example, the woman generally moves to live in the man’s village under a practice called chitengwa while among the Yao of Mangochi, the man is expected to move and live in the woman’s village under a practice called chikamwini . This means that apart from aiming messages at eligible men and boys, the campaign must also aim at reaching the entire community, particularly women, if access to services is to improve. This study has unearthed mixed preferred effects of VMMC messages. These preferred effects , according to the Social-Ecological Model that was used in identifying informants to this study, manifested at four different levels of the social ecology namely: at individual level of a circumcised male; at household level; at community or society level and, finally, at policy level. The preferred effect of VMMC messages on the primary audience, adult men and boys, is sixfold. First, is an awareness and an understanding of VMMC with a view to demystify the concept of voluntary medical male circumcision - a phenomenon that would lead to increased uptake of services, reduced transmission of HIV, cervical cancer among women and girls and sexually transmitted infections. Secondly, an assurance both among Christians and Muslims that VMMC had not been introduced neither to contradict the Bible or the Qur’an respectively nor for political reasons since its purpose is purely biomedical. Thirdly, informants mentioned increased post-circumcision preventive behaviour among clients. Fourthly, they expected increased sexual pleasure and a reduced risk of complications due to early sexual resumption and a reduced fear of pain particularly among adults. At household level informants expected two key impacts. First, mothers and female household members, as primary producers of health , were expected to use their ‘authority of matrilineality’ to encourage men to access services and get their sons, particularly infants and pre-adolescents and adolescents, circumcised at the facility. Secondly, clan leaders were expected to play their rightful role as mwini mtundu or mwini mbumba , by encouraging men, pre-adolescents and adolescents to get circumcised at the facility and supporting them adequately after accessing services. At community level in Mangochi, informants expected to see the reduction of three risky practices: adult sexual instruction that initiates as young as 5 or 6 years were exposed to in rituals called kusasa fumbi for girls and kutaya mafuta for boys; sexually explicit language and songs and the use one razor blade or knife to circumcise more than one initiate by traditional circumcisers. In Dowa, community leaders and gatekeepers expected to see the gulewankulu community support the cause of the VMMC campaign by disseminating messages in their own communities; the increased adoption of VMMC services by adult men who were said to be reticent on access to health services generally. Lastly, the key effect that informants expected from VMMC communication programming was community development as a result of good health at household level. These authors have noted, however, that feedback is the key missing link in Laswell’s Transmission Theory since the theory discusses communication purely as a linear process on the assumption that all there is in a communicative process are sources of information, messages, the audience, medium or channel through which sources speak, and effects. This observation was also made by Peng Wenxiu . This was not complete because, for real effect to occur, like Lozare puts it, the communicator (source of the message) and the audience (the consumer of the message) need to have a common vision or understanding for true behaviour change to occur . For this to happen there is need for those involved in VMMC communication to appreciate that communication is not a one-way lane but a two-way road with interaction between the communicator and the audience. It is for this reason that this study concluded that community engagement and interpersonal communication which provide room for real-time feedback in any communicative event are the preferred communication interventions for VMMC services among Yaos and Chewas of Southern and Central Malawi respectively. Participants both in Mangochi and Dowa raised a serious concern that health practitioners involved in the promotion of VMMC services in their districts did not adequately engage them in their villages. This, they observed, robbed them of the chance to ask questions and to seek clarification on matters that concerned them or were not clear regarding facility circumcision. Study strength The study has clarified five key elements in the communication continuum of VMMC interventions including source, audience, message, channel, effect and feedback and has demonstrated the place of communication in health promotion. The study has demonstrated the indispensability of interpersonal communication health communication with a special emphasis on the need for real-time feedback. It has also provided a platform for further inquiry into health promotion programming particularly in the use of Laswell’s Transmission Theory in health communication in the African context. Limitations Being a by-product of a purely qualitative study, this report lacks statistical basis upon which conclusions could effectively be grounded. Although the principal investigator sought the services of two Yao insiders (a male and a female) and one Chewa insider (a female in-training health worker) to collect and support data analysis, he is aware that his own personal biases may have been introduced in the data analysis and presentation of findings. To counter the bias the data analysis approach was guided by the thematic framework approach whereby rigorous reading and re-reading of each transcript was undertaken with the view to getting familiar with the content of the transcripts. The repeated review of transcripts informed the identification of codes which were then classified into sub-themes which were further synthesized into bigger themes. This effectively minimized the possibility of introducing biases into the content. The codes were developed from the emerging issues from the responses as was required by the study questions which fully underwent a review and approach of the National Committee on Research Ethics in the Social Sciences and Humanities (NCRSH). In this regard, the approach to data collection and analysis was guided by the stipulated ethical obligations by a competent review IRB board implying that the researchers were guided and made sure to adhere to the stipulated ethical obligations and principles of thematic framework approach which is the analysis methodology followed. Further to this, this is a product of team work involving the corresponding author and a team of co-authors who participated in the drafting of manuscript and review of the same suffice to say that co-authors were from different cultural backgrounds. The study has clarified five key elements in the communication continuum of VMMC interventions including source, audience, message, channel, effect and feedback and has demonstrated the place of communication in health promotion. The study has demonstrated the indispensability of interpersonal communication health communication with a special emphasis on the need for real-time feedback. It has also provided a platform for further inquiry into health promotion programming particularly in the use of Laswell’s Transmission Theory in health communication in the African context. Being a by-product of a purely qualitative study, this report lacks statistical basis upon which conclusions could effectively be grounded. Although the principal investigator sought the services of two Yao insiders (a male and a female) and one Chewa insider (a female in-training health worker) to collect and support data analysis, he is aware that his own personal biases may have been introduced in the data analysis and presentation of findings. To counter the bias the data analysis approach was guided by the thematic framework approach whereby rigorous reading and re-reading of each transcript was undertaken with the view to getting familiar with the content of the transcripts. The repeated review of transcripts informed the identification of codes which were then classified into sub-themes which were further synthesized into bigger themes. This effectively minimized the possibility of introducing biases into the content. The codes were developed from the emerging issues from the responses as was required by the study questions which fully underwent a review and approach of the National Committee on Research Ethics in the Social Sciences and Humanities (NCRSH). In this regard, the approach to data collection and analysis was guided by the stipulated ethical obligations by a competent review IRB board implying that the researchers were guided and made sure to adhere to the stipulated ethical obligations and principles of thematic framework approach which is the analysis methodology followed. Further to this, this is a product of team work involving the corresponding author and a team of co-authors who participated in the drafting of manuscript and review of the same suffice to say that co-authors were from different cultural backgrounds. The key finding of the study is that community stakeholder engagement, complemented by various interpersonal communication and small-group approaches such as village town-hall meetings, drama, religious and cultural gatherings (including gulewankulu festivities for the Chewa people), is the preferred communication approach both in Yao and Chewa communities. The main reason is that apart from its strong emphasis on exchanging knowledge between the source and the audience in a human-to-human fashion, community stakeholder engagement and interpersonal communication allow the communicator and the audience to foster a common vision which, according to Ben Lozare, is a prerequisite for behavioral change. This study also established that one critical element that must be created in every health communication situation is feedback . It is this element that makes the development of a common vision between the health communicator and the target audience. It is in this context that true adoption of a desired behavior can be achieved. |
Increased synovial immunohistochemistry reactivity of TGF-β1 in erosive peripheral psoriatic arthritis | 920667f4-afd1-49ce-a36b-237b6989dba1 | 10061727 | Anatomy[mh] | Psoriatic arthritis (PsA) belongs to the spondylarthritis (SpA) group, which encompasses a group of diseases sharing genetic, pathophysiological, clinical, and radiological features, where bone remodeling alterations appear when there is an imbalance between bone resorption and bone formation. Osteolysis in patients with rheumatoid arthritis (RA) results from an imbalance in which bone resorption by osteoclasts is favored over bone formation by osteoblasts,with the inflammatory cells infiltrating the synovial tissues, resulting in synovial hyperplasia, angiogenesis, cartilage destruction, and bone erosion in the diarthrodial joints, where local and systemic factors interrupt the physiological bone remodeling process . Depending on the local microenvironment, inflammation has very different effects on bone, inducing bone loss in the joints and periarticular bone or bone formation in the enthesis areas in patients with axial and peripheral SpA . In the last years, a great deal has been learned about the differentiation and function of osteoclasts , and now it is appreciated that osteoblast-mediated bone formation is also inhibited in the rheumatoid joint, limiting erosion repair , while osteoblast function is increased for producing new bone in SpA . The physio-pathological differences between RA and SpA are rooted in the interaction network of proinflammatory cytokines and are probably related to the different expression of IL-17 A and TNFα, two cytokines that strongly promote osteoclastogenesis and the development of focal bone erosions . TNFα is the main proinflammatory cytokine in RA and promotes bone erosion by triggering osteoclast on through the RANK-RANKL system and by suppressing osteoblast bone formation through the overexpression of Dickkopf-1 (Dkk1), a potent inhibitor of the anabolic Wnt bone signaling pathway . In RA, where the overexpression of TNFα is higher than IL-17 A, bone resorption prevails over bone formation. Unlike TNFα, IL-17 A also promotes osteogenesis, particularly at inflamed sites that experience mechanical stress, as is the case with entheses in animal models of SpA, where fibroblast-like synoviocytes exposed to IL-17 A differentiate into osteoblasts. Bone morphogenetic protein (BMP) signaling could have an anti-inflammatory role in the control and maintenance of low levels of pro-inflammatory factors in healthy joints or in the early stages of RA. It is a critical pathway for osteoblast differentiation and function and this pathway plays a role in bone formation in SpA, with higher BMP2 and 4 serum levels in the serum of ankylosing spondylitis (AS) patients and having a significant correlation with spinal radiograph scores by developing spinal fusion . PsA is an immune-mediated chronic inflammatory arthritis, where skin-resident cells such as keratinocytes, γδT cells, and innate lymphoid cells express IL-17 A, which also inhibits osteoblasts and osteocytes function through Wnt signaling and IL-17 A is a bone-destroying cytokine that is involved in immune-mediated bone diseases and can exert a negative effect on bone by promoting osteoclastogenesis and initiates an immunologic cascade that is associated with synovial inflammation, bone destruction, and juxta-articular new bone formation . Furthermore, the clinical trials of IL-17 A inhibitors in PsA have shown improvement in the signs and symptoms of active PsA , and they can inhibit the progression of bone erosion and maintain bone stability . Therefore, we hypothesize the synovial expression of TGF-β1 is a differentiating factor for the development of erosive peripheral disease in psoriatic arthritis (TGF-β1 can prime IL-17 A), and synovial expression of Dkk1 may be related to peripheral joint destruction as in rheumatoid arthritis. The present study aimed to analyze proteins of inflammation and bone destruction and regeneration in serum, gene expression, and immunohistochemistry (IHC) reactivity in the synovial membrane in patients with PsA, compared with rheumatoid arthritis, osteoarthritis, and AS.
Patients and samples This is a descriptive cross-sectional study in patients fulfilling the CASPAR criteria for psoriatic arthritis (n = 15) to determine the expression of the biomarkers that may be associated with joint inflammation (TGF-β1, IL-23, IL-6, IL-17 A, and IL-22) and with destruction or new bone formation (DKK1, Sclerostin, BMP2, BMP4, Wnt1, and Wnt5a) using immunohistochemistry and qRT-PCR in synovial tissue and by measuring the synovial fluid and serum levels. The expression of these proteins in patients with PsA was compared to their expression in synovial tissue samples from the knee joint and serum from patients with RA (n = 8), OA (n = 18), and AS (n = 4). The patients underwent knee arthroscopy between 2009 and 2013 because they clinically presented with knee joint swelling and tenderness that did not improve with the medical treatment indicated for their diagnosed disease. In December 2018, it was verified that the patient’s clinical diagnosis had not changed. Those patients whose macroscopic or microscopic characteristics of the synovial membrane samples presented characteristics different from the underlying pathology (e.g., deposits of urate or PPCD crystals, pigmented villonodular synovitis or vasculitis) that could interfere with the determinations made were excluded, as well as poorly labeled or unlocatable samples. Patients were evaluated every 3 to 4 months, and the treatment was changed if the disease remained active (≥ 2 swollen and tender joints). Therapy with methotrexate was initiated (up to 20 mg/week if tolerated), and if no response occurred or adverse events were noted, patients were switched to anti-TNF-a or to combined therapy, according to their rheumatologist’s judgment. Clinical and biological data (tender/swollen joint counts 66/68, C-reactive protein (CRP) and erythrocyte sedimentation rate (ESR), disease-modifying antirheumatic drugs (DMARDs), and biologic therapy administered) were collected during the study inclusion and the last clinical control. None of the patients was being treated with biologics, and eight of the patients with PsA were on treatment with methotrexate (MTX). The Kellgren-Lawrence (KL) scale was used to establish the degree of osteoarthritis in the knee . The evaluation of axial radiological damage was also performed using sacroiliac X-ray and through the classification of sacroiliitis according to the modified New York criteria as well as to the presence/absence of erosive peripheral joint damage. This study was carried out at the A Coruña Biomedical Research Institute (INIBIC). The study was approved by the Clinical Research Ethics Committee (CEIC), under number 2011/301, and performed according to the Declaration of Helsinki. All study patients provided written informed consent. Arthroscopies At first, peripheral blood was extracted from a forearm vein. Synovial tissue samples were obtained surgically from the knee joint by means of arthroscopy under local anesthesia. Arthroscopy was performed under diagnostic and/or therapeutic (lavage) conditions with a 2.7-mm arthroscope (Storz, Tullingen, Germany). Eight synovial tissue samples were obtained from the suprapatellar pouch and the medial and lateral gutter in each patient. Four samples were fixed in 4% formaldehyde and embedded in paraffin wax for immunohistochemistry, and the remaining four were collected on RLT lysis buffer (Qiagen, Crawley, West Sussex, UK) for RNA extraction . Arthroscopic joint lavage is a formal joint lavage that takes place in addition to a visual inspection of the structures of the knee joint at the same time . The amount of irrigation fluid (saline serum 0.9% at 5ºC) used was 5000 mL, and the procedure took place over 30 min. The biopsies from each patient were collected, carved, and fixed in the operating room and, together with the blood samples, transferred to the INIBIC facilities in the shortest possible time (less than 1 h). Once in the laboratory, the samples were registered in the data bank (Biobank) using NorayBanks software to guarantee the confidentiality of the procedure. After encoding, they were processed according to the techniques described below. Histopathological analysis and immunohistochemistry: quantification of protein expression in IHC staining The synovial biopsies from each patient were fixed in the operating room, some were immersed in 4% formaldehyde for up to 24 h, and others were immersed in OCT (cryoprotective medium). They were transported using post-physical fixation in dry ice for storage following an established order in a -80ºC chest in the Basic Research Laboratory at the INIBIC. Once the inclusion was finished, the sections were stained with hematoxylin-eosin and Masson’s trichrome (H-E, MM-classical histological stains) for the first morphological study. The histopathological analysis was carried out by a pathologist other than the immunohistochemistry, who made a detailed description of it, without images attached to the report. Indirect immunohistochemical techniques (with paraffin peroxidase) were used in all the biopsies, in which the mouse monoclonal anti-BMP2 ab6285 antibody (clone 65529.111) of Abcam® and rabbit monoclonal anti-Dkk1 antibody were used as the primary antibodies. ab109416 (clone EPR4759) from Abcam®, rabbit anti-BMP2 monoclonal antibody ab 124,715 (clone EPR6211) from Abcam®, mouse monoclonal anti-Wnt5a ab86720 (clone 3D10) from Abcam®, mouse monoclonal antibody anti-TGF-β1 ab64715 (clone 2Ar2) from Abcam®, Abcam® ab79056 rabbit anti-IL-17 A polyclonal antibody, and Dako® K-5007 antibody (EnVision ™ Detection Systems Peroxidase / DAB), were used as the secondary antibodies. The samples were pretreated with tris-EDTA at pH 9 in Retriever (0.1 M sodium citrate pH 6.1 for Wnt5a), and the positive controls were BMP2 for human small bowel tissue (1: 5000), BMP4 for human colon tissue (1: 1000), Dkk1 for human placenta tissue (1: 1000), Wnt5a for human thyroid tissue (1: 1000), TGF-β1 for human articular cartilage tissue and osteoarthritis (1:50), and IL-17 A for human lymphatic node (1: 1000), and the negatives of the technique were conducted without the use of a primary antibody. The chromophore that was used includes the chromogen diaminobenzidine (DAB, brown color) and the peroxidase substrate (H2O2). The samples were counterstained with Gill III’s hematoxylin-eosin, dehydrated, and rinsed with xylene. DPX (acrylic resin) was used as a coverslip mounting medium. Image capture was performed with the Olympus BX61 microscope, and the analysis and quantification of the samples were conducted with the Nikon Eclipse microscope, using NIS Elements imaging software. Determination of serum levels of proteins: enzyme-linked ImmunoSorbent Assay (ELISA) In this study, the double sandwich ELISA (DAS) was performed in triplicate, in 96-well microplates. The expression of BMP2, BMP4, Dkk1, Wnt1, Wnt5a, sclerostin, TGF- β1, IL-6, IL-17 A, and IL-22 in the serum was quantified according to the manufacturer’s specifications of each ELISA KIT. The absorbance was measured in a microplate reader at 450 nm, with the results being extrapolated to the standard curve in each case. Determination of gene expression in synovial tissue: quantitative real-time PCR (qRT-PCR) From the biopsies obtained during the arthroscopies, we performed RNA extraction with TRIzol®Reagent (Life Technologies) following the manufacturer’s instructions. The genetic material obtained was quantified by spectrophotometry with NanoDrop ™ (Thermo Scientific). Using the Superscript® VILO ™ cDNA synthesis kit (Invitrogen), it was performed the reverse transcription of the RNA. The cDNA that was obtained was amplified by qRT-PCR with LightCycler 480 II equipment (Roche) using the Taqman probes (Light-Cycler® 480 Probes Master (Roche). The 60 S ribosomal protein L13a (RPL13a) was used to normalize the results that were obtained. The qRT-PCR results were analyzed with the qBase plus software (Biogazelle). Statistical analysis The results were expressed as a percentage or as the median and interquartile range (IQR) and categoric variables as frequencies and percentages. Comparisons of qualitative variables were made using the chi-square test and Fisher’s exact test if applicable. The Wilcoxon rank sum test or the Kruskal-Wallis’s test were used to compare the distribution of the numeric variables between the groups. The correlation between the numeric variables was expressed by the Spearman correlation coefficient, and the null hypothesis was tested (coefficient = zero). The correlation between two categoric variables or between one numeric and one categoric variable was assessed by using Fisher´s exact test and the Wilcoxon rank sum test or the Kruskal-Wallis’s test. Univariate and multivariate logistic regression models were performed to evaluate the association of the proteins related to gene expression with demographics, clinical, radiological, and therapeutic features. The data were statistically analyzed with the SPSS version 21 program (IBM SPSS Statistics). Values of p < 0.05 were considered statistically significant.
This is a descriptive cross-sectional study in patients fulfilling the CASPAR criteria for psoriatic arthritis (n = 15) to determine the expression of the biomarkers that may be associated with joint inflammation (TGF-β1, IL-23, IL-6, IL-17 A, and IL-22) and with destruction or new bone formation (DKK1, Sclerostin, BMP2, BMP4, Wnt1, and Wnt5a) using immunohistochemistry and qRT-PCR in synovial tissue and by measuring the synovial fluid and serum levels. The expression of these proteins in patients with PsA was compared to their expression in synovial tissue samples from the knee joint and serum from patients with RA (n = 8), OA (n = 18), and AS (n = 4). The patients underwent knee arthroscopy between 2009 and 2013 because they clinically presented with knee joint swelling and tenderness that did not improve with the medical treatment indicated for their diagnosed disease. In December 2018, it was verified that the patient’s clinical diagnosis had not changed. Those patients whose macroscopic or microscopic characteristics of the synovial membrane samples presented characteristics different from the underlying pathology (e.g., deposits of urate or PPCD crystals, pigmented villonodular synovitis or vasculitis) that could interfere with the determinations made were excluded, as well as poorly labeled or unlocatable samples. Patients were evaluated every 3 to 4 months, and the treatment was changed if the disease remained active (≥ 2 swollen and tender joints). Therapy with methotrexate was initiated (up to 20 mg/week if tolerated), and if no response occurred or adverse events were noted, patients were switched to anti-TNF-a or to combined therapy, according to their rheumatologist’s judgment. Clinical and biological data (tender/swollen joint counts 66/68, C-reactive protein (CRP) and erythrocyte sedimentation rate (ESR), disease-modifying antirheumatic drugs (DMARDs), and biologic therapy administered) were collected during the study inclusion and the last clinical control. None of the patients was being treated with biologics, and eight of the patients with PsA were on treatment with methotrexate (MTX). The Kellgren-Lawrence (KL) scale was used to establish the degree of osteoarthritis in the knee . The evaluation of axial radiological damage was also performed using sacroiliac X-ray and through the classification of sacroiliitis according to the modified New York criteria as well as to the presence/absence of erosive peripheral joint damage. This study was carried out at the A Coruña Biomedical Research Institute (INIBIC). The study was approved by the Clinical Research Ethics Committee (CEIC), under number 2011/301, and performed according to the Declaration of Helsinki. All study patients provided written informed consent.
At first, peripheral blood was extracted from a forearm vein. Synovial tissue samples were obtained surgically from the knee joint by means of arthroscopy under local anesthesia. Arthroscopy was performed under diagnostic and/or therapeutic (lavage) conditions with a 2.7-mm arthroscope (Storz, Tullingen, Germany). Eight synovial tissue samples were obtained from the suprapatellar pouch and the medial and lateral gutter in each patient. Four samples were fixed in 4% formaldehyde and embedded in paraffin wax for immunohistochemistry, and the remaining four were collected on RLT lysis buffer (Qiagen, Crawley, West Sussex, UK) for RNA extraction . Arthroscopic joint lavage is a formal joint lavage that takes place in addition to a visual inspection of the structures of the knee joint at the same time . The amount of irrigation fluid (saline serum 0.9% at 5ºC) used was 5000 mL, and the procedure took place over 30 min. The biopsies from each patient were collected, carved, and fixed in the operating room and, together with the blood samples, transferred to the INIBIC facilities in the shortest possible time (less than 1 h). Once in the laboratory, the samples were registered in the data bank (Biobank) using NorayBanks software to guarantee the confidentiality of the procedure. After encoding, they were processed according to the techniques described below.
The synovial biopsies from each patient were fixed in the operating room, some were immersed in 4% formaldehyde for up to 24 h, and others were immersed in OCT (cryoprotective medium). They were transported using post-physical fixation in dry ice for storage following an established order in a -80ºC chest in the Basic Research Laboratory at the INIBIC. Once the inclusion was finished, the sections were stained with hematoxylin-eosin and Masson’s trichrome (H-E, MM-classical histological stains) for the first morphological study. The histopathological analysis was carried out by a pathologist other than the immunohistochemistry, who made a detailed description of it, without images attached to the report. Indirect immunohistochemical techniques (with paraffin peroxidase) were used in all the biopsies, in which the mouse monoclonal anti-BMP2 ab6285 antibody (clone 65529.111) of Abcam® and rabbit monoclonal anti-Dkk1 antibody were used as the primary antibodies. ab109416 (clone EPR4759) from Abcam®, rabbit anti-BMP2 monoclonal antibody ab 124,715 (clone EPR6211) from Abcam®, mouse monoclonal anti-Wnt5a ab86720 (clone 3D10) from Abcam®, mouse monoclonal antibody anti-TGF-β1 ab64715 (clone 2Ar2) from Abcam®, Abcam® ab79056 rabbit anti-IL-17 A polyclonal antibody, and Dako® K-5007 antibody (EnVision ™ Detection Systems Peroxidase / DAB), were used as the secondary antibodies. The samples were pretreated with tris-EDTA at pH 9 in Retriever (0.1 M sodium citrate pH 6.1 for Wnt5a), and the positive controls were BMP2 for human small bowel tissue (1: 5000), BMP4 for human colon tissue (1: 1000), Dkk1 for human placenta tissue (1: 1000), Wnt5a for human thyroid tissue (1: 1000), TGF-β1 for human articular cartilage tissue and osteoarthritis (1:50), and IL-17 A for human lymphatic node (1: 1000), and the negatives of the technique were conducted without the use of a primary antibody. The chromophore that was used includes the chromogen diaminobenzidine (DAB, brown color) and the peroxidase substrate (H2O2). The samples were counterstained with Gill III’s hematoxylin-eosin, dehydrated, and rinsed with xylene. DPX (acrylic resin) was used as a coverslip mounting medium. Image capture was performed with the Olympus BX61 microscope, and the analysis and quantification of the samples were conducted with the Nikon Eclipse microscope, using NIS Elements imaging software.
In this study, the double sandwich ELISA (DAS) was performed in triplicate, in 96-well microplates. The expression of BMP2, BMP4, Dkk1, Wnt1, Wnt5a, sclerostin, TGF- β1, IL-6, IL-17 A, and IL-22 in the serum was quantified according to the manufacturer’s specifications of each ELISA KIT. The absorbance was measured in a microplate reader at 450 nm, with the results being extrapolated to the standard curve in each case.
From the biopsies obtained during the arthroscopies, we performed RNA extraction with TRIzol®Reagent (Life Technologies) following the manufacturer’s instructions. The genetic material obtained was quantified by spectrophotometry with NanoDrop ™ (Thermo Scientific). Using the Superscript® VILO ™ cDNA synthesis kit (Invitrogen), it was performed the reverse transcription of the RNA. The cDNA that was obtained was amplified by qRT-PCR with LightCycler 480 II equipment (Roche) using the Taqman probes (Light-Cycler® 480 Probes Master (Roche). The 60 S ribosomal protein L13a (RPL13a) was used to normalize the results that were obtained. The qRT-PCR results were analyzed with the qBase plus software (Biogazelle).
The results were expressed as a percentage or as the median and interquartile range (IQR) and categoric variables as frequencies and percentages. Comparisons of qualitative variables were made using the chi-square test and Fisher’s exact test if applicable. The Wilcoxon rank sum test or the Kruskal-Wallis’s test were used to compare the distribution of the numeric variables between the groups. The correlation between the numeric variables was expressed by the Spearman correlation coefficient, and the null hypothesis was tested (coefficient = zero). The correlation between two categoric variables or between one numeric and one categoric variable was assessed by using Fisher´s exact test and the Wilcoxon rank sum test or the Kruskal-Wallis’s test. Univariate and multivariate logistic regression models were performed to evaluate the association of the proteins related to gene expression with demographics, clinical, radiological, and therapeutic features. The data were statistically analyzed with the SPSS version 21 program (IBM SPSS Statistics). Values of p < 0.05 were considered statistically significant.
Women were more prevalent in the OA group (p = 0.036), and these patients were older (< 0.0001). Disease duration was higher in the OA group and in the AS group and was associated with a longer period of disease evolution while for the PsA and RA groups was about 3 years (p 0.002). The peripheral joint disease for all groups was mono-articular (knee joint target), except for PsA which had an oligo-articular joint count (0.033). CRP was higher in the AS group, and this was the only group of patients with radiologic sacroiliac damage and positive HLA-B27. Out of 15 PsA patients, only four had erosions on X-ray films, and eight patients were treated with MTX (Table ). From 45 patients, histopathological analysis of the synovial membrane of PsA patients showed an increase in vessel density in PsA compared with RA, OA, or AS (p 0.035). Other features had not shown differences among patients (Table ). No statistically significant differences were observed regarding hyperplasia of the synovial lining, villous papillae, lymphoplasmacytic accumulations, lymphocytic or plasmacytoid infiltrates, and fibrosis. For the histopathological study, 45 valid samples of the synovial membrane were available, but only from 42 patients for IHC, extraction, and purification of RNA for synovial mRNA expression analysis. Blood was obtained to measure the protein levels in the serum, and only valid samples were available from 38 patients. The IL-17 A gene expression in the synovial membrane (Table ) was higher in PsA patients (p = 0.018) and was positively correlated with IL-23 (p = 0.025), and Dkk1 (p = 0.022), and negatively with BMP2 (p = 0.033), and BMP4 (p = 0.01) (Fig. ). Patients with diagnoses of peripheral SpA (PsA and AS with knee arthritis) had a higher mRNA expression ratio of IL17A than patients with RA or OA (p = 0.001), and of IL-23, but this last was not statistically significant (Fig. ). Considering spondyloarthritis patients as a group with shared pathophysiological characteristics, both had higher IL17A gene expression. Separately, patients with AS had higher synovial IL17A gene expression than in OA or RA (p 0.01). However, these data were even higher for PsA patients (p 0.009), as can be seen in Fig. . There were no differences between PsA and AS regarding IL-23. When was considered the presence of treatment with MTX, lower levels of IL-17 A gene expression were not observed in patients being treated with MTX (p = 0.703). The IHC expression for TGF-β1 was higher in PsA patients than in the other diseases (Table ). The IHC reactivity for TGF-β1 in the synovial tissue (Fig. ) was higher in patients with psoriatic arthritis (p 0.010) and was positively correlated with IL-17 A (r = 0.389, p = 0.012) and Dkk1 (r = 0.388, p = 0.012), (Fig. ). Nevertheless, IL-17 A and Dkk1 IHC% were not statistically different among group diseases (p = 0.448 and p = 0.323, respectively), although the results were slightly higher for IL-17 A for PsA and AS (Table ). When PsA patients’ group was divided between patients with (n = 6) or without (n = 9) radiologic erosive damage (Fig. ), the TGF-β1 serum levels were significantly increased in PsA with erosions (p = 0.024). Following treatment with MTX, patients had lower IL-17 A gene expression in the synovial membrane than the other patients did (p < 0.0001), and in the regression logistic analysis (adjusted by age, gender, disease duration, and KL scale in the knee target), it had a correlation with erosive disease (p = 0.027). The RA patients obtained higher serum concentrations of BMP2 than the other groups (Table ). Additionally, Dkk1 in PsA, like in AS, obtained higher serum levels than in RA and OA; BMP4 and sclerostin were higher in OA; Wnt1, Wnt5b, TGF-β1, and IL-22 were higher in AS; the serum concentration of IL-6 was lower in OA, and higher levels of IL-17 A were seen in OA with respect to the other pathologies.
As the main source of IL-17 A, differentiated Th17 cells require TGF-β1 and IL-23 , and TGF-β1 is a pleiotropic cytokine that is involved in both suppressive and inflammatory immune responses . TGF-β1 controls the proliferation, survival, activation, and differentiation of B cells as well as the development and functions of innate cells, including natural killer (NK) cells, macrophages, dendritic cells, and granulocytes. TGF-β1 is normally generated by macrophages during apoptotic cell clearance and contributes to buffering the inflammatory sequelae that are associated with phagocytosis, playing an important role in maintaining the tolerance by controlling survival, proliferation, and differentiation of the Th17 lymphocytes , and Th17 cells can be both immunoregulatory and pathogenic . Our data showed that the immunohistochemical expression of TGF-β1 in the synovial tissue was higher in patients with PsA compared to those with RA, OA, and AS. Interestingly, patients with PsA who had developed bone erosions had the highest levels of TGF-β1 expression in IHC and these data could reflect a tissue healing effect. In addition, in our patients, IL-17 A expression was also higher in PsA than it was in RA, and it was especially higher than it was in OA, and it seems to be like what is observed in AS, which shares pathogenic mechanisms with PsA. The generation of regulatory Th17 cells is promoted by the combination of TGF-β1 and IL-6 . Pathogenic Th17 cells, however, require further stimulation with IL-23 , although pathogenic Th17 cells can also be induced in cell culture without TGF-β1 and in the presence of IL-6, IL-1β, and IL-23. At low concentrations, TGF-β1 synergizes with IL-6 and IL-21 to promote IL-23 receptor expression and Th17-cell differentiation, whereas high TGF-β1 concentrations repress IL-23 receptor expression and promote Treg-cell differentiation . Our findings show that patients with higher IL-23 gene expression were correlated with high expression levels of IL-17 A. Since Dkk1 is an inhibitor of the Wnt pathway, which normally induces new bone formation, one might expect Dkk1 concentrations to be progressively lower along a spectrum of diseases that increase bone formation. However, consistent with most studies, the Dkk1 concentrations were seen as no different in patients with peripheral PsA compared to healthy controls and were higher in patients with AS compared to those with PsA . In our report, Dkk1 gene expression was higher in AS patients and correlated with IL17A, and serum levels were higher in PsA and AS but did not show statistical significance. Jadon et al. found lower Dkk1 concentrations in patients with PsA with vertebral erosions and Li et al. showed the combined action of TNFα and IL-17 A on hMSCs, which increased osteogenesis through the inhibition of Dkk1 and RANKL gene expression. BMPs are considered cytokines that stimulate the formation of bone and cartilage, and they have been shown to play important roles in migration, proliferation, apoptosis, and differentiation of several cell types . BMP2 induces bone and cartilage formation , and Osta et al. saw that IL-17 decreased TNFα-induced BMP2 inhibition in vitro, with IL17A potentiating the effect of TNFα, and this fact may explain the ligaments ossification mechanisms as observed in AS. Our data show lower BMP2 and 4 gene expression in patients with PsA and higher IL-17 A synovial mRNA expression correlating with Dkk1. This fact is in accordance with the pathways that are implicated in new bone formation in SpA and include the IL-23/IL-17 axis and the BMP and Wnt signaling pathways . Diarra et al. were able, by inhibiting Dkk-1, to reverse the bone destructive pattern of a mouse model of rheumatoid arthritis to the bone-forming pattern of osteoarthritis. Decrease expression levels of the Wnt antagonists, Dkk1 and sclerostin, in SpA lead to increase Wnt signaling and osteoblast function, resulting in new bone formation. Mechanical stress also promotes inflammation and new bone formation in the entheses . Moreover, Zhang et al. showed that IL17 significantly inhibited osteoblasts differentiation induced by BMP2 in a murine model. We found a higher serum level of IL-17 A in OA compared with PsA. In general, IL-17 A is prominently produced by the Th17 subtype of Th cells. These cells can act as either pathogenic or protective, depending on the cytokine milieu that stimulates them. Moreover, γδ T cells (another important source of IL-17) do not seem altered in the synovial tissue in patients with OA. Several studies suggested that serum IL-17 A is increased in patients with more severe OA (KL grade 2–4) with no correlation regarding levels in SF or synovium expression , and we have observed these high serum levels in older patients with severe knee OA. Patients who were following treatment with MTX when the biopsy was performed had lower IL-17 A gene expression in the synovial membrane than the other patients, and it was correlated with erosive disease. It is known MTX dose-dependently suppresses the production of IL-17 A at the mRNA level by PBMCs from healthy donors and RA patients . Our study is limited due to its retrospective design and single-center approach and future longitudinal studies are needed to evaluate the real role that TGF-β1 plays in PsA with erosive damage in peripheral joints and its interaction with other cytokines involved in inflammation such as IL17A and Dkk-1. Moreover, the size of the patient’s samples is a bit small, but these studies provide estimates for biomarkers and effect sizes with respect to clinical outcomes, which are necessary to calculate the sample size and statistical power for future studies. Very small samples undermine the internal and external validity of a study. That is why these results should be considered with caution and this study should be replicated with larger sample size. The findings in the synovial membrane will be more enriching in PsA together with the study of peripheral entheses and techniques for obtaining enthesis tissue samples for study are currently being perfected . In summary, IL-17 A gene expression in the synovial membrane of patients with psoriatic arthritis is positively correlated with traditional osteo-destructive proteins and negatively correlated with the bone-forming proteins in peripheral arthritis. TGF-β1 (which is necessary for the activation of Th17 cells, but also involved in regeneration processes) immunoreactivity in synovial tissue, was higher in patients with erosive psoriatic arthritis in relation to the increased levels of IL-17 A and Dkk1 in the IHC.
|
Uptake of pharmacist recommendations by patients after discharge: Implementation study of a patient-centered medicines review service | fa216cc7-2f93-42d4-ab4b-c6bef5698918 | 10061906 | Patient-Centered Care[mh] | Polypharmacy, defined here as the taking of five or more medicines concurrently, is associated with a high prevalence of potentially inappropriate medicine (PIM – defined in supplementary Table ) use and occurs frequently in those aged 65 years or over [ – ]. PIM use results in poor outcomes including falls, emergency department visits, increased costs, adverse events, and functional decline . Deprescribing - the patient-centred, supervised process of dose reduction or cessation of PIMs - has been identified as part of good prescribing but as limited and reactive rather than proactive, generally occurring because of an adverse event . Deprescribing does not appear to be part of current hospital inpatient practice . Yet the simple count of prescribed medicines at discharge has been shown to outperform complex indicators of therapy quality, such as Beers’ list 2019 and STOPP criteria Version 2 when identifying people at risk and predicting poor outcomes . In Australia, up to 30% of hospital admissions for patients over 75 years of age have been found to be medicine-related, with up to three-quarters potentially preventable, the single most important predictor being the number of medicines taken . The risk of harm and of poor adherence rises with the addition of each new medicine , with harm described to be at epidemic proportions . Transitions from hospital to primary care further increase the risk for reasons that include increased medicine sensitivity due to deconditioning and ongoing recovery from acute illness, inaccuracies in medicine reconciliation, insufficient patient education, poor communication with primary care and unexplained medicine changes [ – ]. As many as 44% of patients do not follow medicine changes initiated in hospital, continuing to take discontinued medicines, failing to implement dosage changes or to take newly prescribed medicines , which may themselves be potentially inappropriate . While the best strategies to combat PIM use in primary care remain unclear [ , , ], effective transitional pharmacist-led strategies have been described [ – ]. They have included medicine reconciliation and review in the context of multidisciplinary care, patient counselling, communication with primary care providers and post-discharge follow-up. Although patient engagement in understanding and managing their medicines is strongly encouraged, it is uncommon [ , – ]. Transitional patient-centred care has been described as poorly understood and a missed opportunity for pharmacists , such care recognised as improving patient satisfaction and decision making and reducing adverse events and readmissions [ – ]. A paradigm shift in such care is needed . Australian hospital safety and quality standards state that patients and their caregivers should be actively involved in their care, and that they should receive verbal and written information in ways that are meaningful to them . Patient-directed education or coaching has been shown to be the most influential component of multicomponent interventions for successful transitions . However, there is limited research on the impact of pharmacy health coaching , or how well patient-centred care is applied to medicine management in Australian hospitals . Patients have been reported to arrive at hospital taking PIMs, have PIMs commenced and be discharged on PIMs . To address this problem, an implementation program for a discharge medicine review service was begun in 2006 with the development of prescribing appropriateness criteria for older Australians . This criteria set was applied in a scoping study , which found a high incidence of PIM use at our hospital. A randomised controlled trial subsequently applied the criteria during medicine review at discharge in intervention patients, sent to patients’ general practitioners (GPs) for actioning. No significant difference in criteria-based recommendations between intervention and control groups were found at follow-up. GPs implemented a relatively low number (42%) of recommendations . This led to a new intervention strategy; the patient and/or caregiver were made the driver of change in reducing their use of PIMs. A patient-centred discharge medicines review service was commenced in 2016. This study aims to identify the processes, barriers and facilitators that influenced the implementation and intervention effectiveness of this service. For example, limited organisational resources and low leadership engagement have been identified as barriers to implementation of transitional care innovations, whereas adaptability of innovations and high perceived benefit by users identified as facilitators . Implementing research into healthcare practice can be complex and unpredictable, with failure common [ – ]. A post-implementation (post hoc) study of these factors was conducted, such studies being commonly used to analyse and explain the implementation process . A prospective audit was conducted to determine the effectiveness of the resulting patient-centred intervention. Aims of the study To describe an implementation program in the development of a patient-centred medicine review service; to assess service impact on older patients and their caregivers actioning recommendations after discharge from hospital. Ethics approvals Ethics approval was obtained from the Human Research Ethics Committee of The University of Sydney for each phase of the intervention process, begun in 2006 (project numbers 2011-2015/10043, 2019/209). Approval was also obtained from the Hospitals Medical Executive Committee. Written informed consent was obtained from all individual patients or their caregivers.
To describe an implementation program in the development of a patient-centred medicine review service; to assess service impact on older patients and their caregivers actioning recommendations after discharge from hospital.
Ethics approval was obtained from the Human Research Ethics Committee of The University of Sydney for each phase of the intervention process, begun in 2006 (project numbers 2011-2015/10043, 2019/209). Approval was also obtained from the Hospitals Medical Executive Committee. Written informed consent was obtained from all individual patients or their caregivers.
Implementation process Many different implementation frameworks have been developed to plan, guide, and evaluate implementation efforts [ – ]. Implementation (or process evaluation) dimensions (defined in supplementary Table ) recommended by the Cochrane Qualitative and Implementation Methods Group were identified by the authors post-intervention that determined the resulting intervention. To gain a broad understanding of determinants of practice (that is, barriers or facilitators), a checklist resulting from a synthesis of frameworks was chosen to identify determinants responsible for achieving the desired outcome. Combining different frameworks may enable a more comprehensive study . Reporting was guided by the “Standards for Reporting Implementation Studies” checklist . Intervention Setting The intervention, a prospective post-hospital audit of recommendations made to patients and/or caregivers at discharge, was carried out at a private, not-for-profit 55 bed hospital in Sydney Australia. Patients were admitted for exacerbations of chronic medical conditions such as heart failure, Parkinson’s disease, chronic obstructive pulmonary disease/asthma, degenerative spinal disease, and inflammatory bowel disease; for rehabilitation after heart, spinal, joint, gastrointestinal, breast or gynaecologic surgery, or trauma from motor vehicle accidents or falls; for palliative care due to metastatic disease; and for management of infections such as cellulitis, pneumonia or urosepsis. Chronic medical conditions and medicines were representative of older Australian community patients . Patients were admitted under the care of one of three geriatricians, rehabilitation specialists or one of two palliative care physicians, supported by two staff doctors. Multidisciplinary care was provided by nursing staff, physiotherapists, occupational therapists, dieticians, social workers, and a discharge planner. The clinical pharmacist (BJB) was an experienced medicines review pharmacist. Eligibility criteria All patients 65 years or older were eligible. There were no other exclusion criteria. Specifically, patients were not excluded if taking less than five medicines, cognitively impaired, whose second language was English, were being discharged to residential or supportive care, lived distant from the hospital, had a terminal illness, or had vision or hearing impairment. Intervention Between July 2019 and March 2020, a convenience sample of 100 patients were recruited for follow-up after discharge. Between one to four patients were discharged daily, the first alternating with the last on a non-alphabetized list being recruited daily. Where cognitive impairment was present, as determined by a Montreal Cognitive Assessment (MoCA) test score of less than 26/30, or where there was language, hearing or vision difficulties, a caregiver was recruited. Two to three days before discharge, the pharmacist explained to the patient and/or caregiver that sometimes, the benefit of taking certain medicines may be unclear, or the dose may need adjustment. A safer or cheaper medicine or even no medicine at all may be more appropriate. Permission to review their medicines, make recommendations and follow them up was sought, an information sheet provided, and a consent form signed. A medicine list would be provided that detailed the best times to take their medicines, brand names, purpose, cost considerations, relevant side effects and easy-to-understand recommendations to assist with management. Medicines were then reconciled, and reviewed utilizing validated prescribing appropriateness criteria, shown in this setting to detect approximately three quarters of all causes of medicine-related problems (MRPs) . A comprehensive medicine review was conducted according to the protocol of the Pharmaceutical Society of Australia , including opportunities for non-pharmacologic care. Patient-directed education was provided during a discharge interview, timing facilitated by allied health staff. Patients/caregivers were encouraged to discuss with their GPs those recommendations important to them for prescription medicines, and to consider for themselves their use of non-prescription medicines. The pharmacist acted as the patient/caregivers’ advocate in proactively addressing PIM use, catering to patient/caregiver health literacy. The discharge medicine list with recommendations and pharmacist contact details was sent separately to GPs, and where appropriate to aged care facilities, community nurses and pharmacies. Where patients had no GP, support was given finding one. Because it was necessary for all patients to have their medicines reconciled and reviewed and to receive discharge counselling, a control group was not possible. The time taken for each activity was recorded to determine the cost of the service. This included finding medical notes and walking corridors. Patients were invited to fill in a general hospital feedback form at discharge as part of standard practice. Ten to fourteen days after discharge, each patient or caregiver was contacted, either by phone or in person. Enquiry was made about the actioning of each recommendation, and the results including GP response recorded. Patients’ reports of changes to medicine use were accepted as truthful. Where there had been no visit to a GP or specialist doctor, support and reassurance was provided, and a repeat contact time made. The patient journey consisted of six stages (Figure ), fitted into episodes of physiotherapy/hydrotherapy attendance, sleep, and mealtimes. Reporting followed the STROBE checklist for observational studies . Data analysis Data were entered into Microsoft Excel (version 2203), checked for normality, and analyzed using descriptive statistics.
Many different implementation frameworks have been developed to plan, guide, and evaluate implementation efforts [ – ]. Implementation (or process evaluation) dimensions (defined in supplementary Table ) recommended by the Cochrane Qualitative and Implementation Methods Group were identified by the authors post-intervention that determined the resulting intervention. To gain a broad understanding of determinants of practice (that is, barriers or facilitators), a checklist resulting from a synthesis of frameworks was chosen to identify determinants responsible for achieving the desired outcome. Combining different frameworks may enable a more comprehensive study . Reporting was guided by the “Standards for Reporting Implementation Studies” checklist .
The intervention, a prospective post-hospital audit of recommendations made to patients and/or caregivers at discharge, was carried out at a private, not-for-profit 55 bed hospital in Sydney Australia. Patients were admitted for exacerbations of chronic medical conditions such as heart failure, Parkinson’s disease, chronic obstructive pulmonary disease/asthma, degenerative spinal disease, and inflammatory bowel disease; for rehabilitation after heart, spinal, joint, gastrointestinal, breast or gynaecologic surgery, or trauma from motor vehicle accidents or falls; for palliative care due to metastatic disease; and for management of infections such as cellulitis, pneumonia or urosepsis. Chronic medical conditions and medicines were representative of older Australian community patients . Patients were admitted under the care of one of three geriatricians, rehabilitation specialists or one of two palliative care physicians, supported by two staff doctors. Multidisciplinary care was provided by nursing staff, physiotherapists, occupational therapists, dieticians, social workers, and a discharge planner. The clinical pharmacist (BJB) was an experienced medicines review pharmacist.
All patients 65 years or older were eligible. There were no other exclusion criteria. Specifically, patients were not excluded if taking less than five medicines, cognitively impaired, whose second language was English, were being discharged to residential or supportive care, lived distant from the hospital, had a terminal illness, or had vision or hearing impairment.
Between July 2019 and March 2020, a convenience sample of 100 patients were recruited for follow-up after discharge. Between one to four patients were discharged daily, the first alternating with the last on a non-alphabetized list being recruited daily. Where cognitive impairment was present, as determined by a Montreal Cognitive Assessment (MoCA) test score of less than 26/30, or where there was language, hearing or vision difficulties, a caregiver was recruited. Two to three days before discharge, the pharmacist explained to the patient and/or caregiver that sometimes, the benefit of taking certain medicines may be unclear, or the dose may need adjustment. A safer or cheaper medicine or even no medicine at all may be more appropriate. Permission to review their medicines, make recommendations and follow them up was sought, an information sheet provided, and a consent form signed. A medicine list would be provided that detailed the best times to take their medicines, brand names, purpose, cost considerations, relevant side effects and easy-to-understand recommendations to assist with management. Medicines were then reconciled, and reviewed utilizing validated prescribing appropriateness criteria, shown in this setting to detect approximately three quarters of all causes of medicine-related problems (MRPs) . A comprehensive medicine review was conducted according to the protocol of the Pharmaceutical Society of Australia , including opportunities for non-pharmacologic care. Patient-directed education was provided during a discharge interview, timing facilitated by allied health staff. Patients/caregivers were encouraged to discuss with their GPs those recommendations important to them for prescription medicines, and to consider for themselves their use of non-prescription medicines. The pharmacist acted as the patient/caregivers’ advocate in proactively addressing PIM use, catering to patient/caregiver health literacy. The discharge medicine list with recommendations and pharmacist contact details was sent separately to GPs, and where appropriate to aged care facilities, community nurses and pharmacies. Where patients had no GP, support was given finding one. Because it was necessary for all patients to have their medicines reconciled and reviewed and to receive discharge counselling, a control group was not possible. The time taken for each activity was recorded to determine the cost of the service. This included finding medical notes and walking corridors. Patients were invited to fill in a general hospital feedback form at discharge as part of standard practice. Ten to fourteen days after discharge, each patient or caregiver was contacted, either by phone or in person. Enquiry was made about the actioning of each recommendation, and the results including GP response recorded. Patients’ reports of changes to medicine use were accepted as truthful. Where there had been no visit to a GP or specialist doctor, support and reassurance was provided, and a repeat contact time made. The patient journey consisted of six stages (Figure ), fitted into episodes of physiotherapy/hydrotherapy attendance, sleep, and mealtimes. Reporting followed the STROBE checklist for observational studies .
Data were entered into Microsoft Excel (version 2203), checked for normality, and analyzed using descriptive statistics.
Implementation Processes and determinants identifying actions taken in the implementation of a discharge medicines review service appear in Table . Processes of context, fidelity, implementer engagement, intervention quality and reach (definitions supplementary Table ) appeared in each phase, as did the following determinants: feasibility; mandate, authority, and accountability; quality assurance and patient safety systems; source of the recommendation. The most commonly occurring determinants were capacity to plan change; implementer engagement; and patient needs, beliefs, knowledge, and motivation. Intervention The implemented service was audited between July 2019 and March 2020. Of the 166 patients recruited, 66 were excluded; 11 were transferred to other hospitals due to the occurrence of an acute medical condition such as bleeding or chest pain, or for a procedure unavailable onsite; six left before interview; no recommendations requiring follow-up were made for 33 patients; nine patients were uncontactable after discharge; three had not seen a doctor within four weeks of discharge; three were admitted to another hospital within two weeks of discharge, and one patients family refused follow-up, leaving 100 patients. All patients/caregivers received a discharge medicine list and review form, and all agreed to participate in a medicines discharge interview and to consider discussing those recommendations important to them with their GP. All patients were followed-up. The pharmacist did not communicate directly with GPs, nor did any GP contact the pharmacist. Mean participant age was 83.1 years, mean total number of medicines 10.4, with a mean number of 8.9 medical conditions per patient. Of 100 patients, five took less than 5 regular medicines, 48 took five to nine regular medicines, and 47 took 10 regular medicines or more - classed as hyper polypharmacy . Fifty six percent of patients were counselled in the presence of a caregiver. Of 368 recommendations made to 100 patients/caregivers, 351 (95%) were actioned, with 284 (77% of those actioned) reported to be implemented and 206 (21%) regularly taken medicines deprescribed – 141 ceased and 65 medicines reduced in dose (Table ). There were 340 causes of a medicine-related problems (MRPs - 3.4 per patient), classified according to a validated system . The top 10 categories represented 92% (312/340) of all causes of MRPs, the most common being: Medicine not effective for the indication treated; medicine was not the most safe/effective; and indication does not warrant medicine treatment (Table ) Medicines for acid-related disorders, multivitamins, complementary and alternative medicines, and mineral supplements were the most common medicines ceased. Gabapentinoids, opiates, proton pump inhibitors and statins were the most common medicines reduced in dose. The time taken to reconcile, review and interview patients/caregivers averaged 63.6 minutes/patient. Recommendations not actioned (17 or 4.6% of the total number) occurred if patients/caregivers decided they were unimportant. Recommendations not implemented occurred because medicines were continued despite evidence provided of poor or absent effectiveness, or GPs considering recommendations unnecessary. Examples included non-discontinuation of glucosamine and prescription of proton pump inhibitors despite apparent lack of indication. Oral feedback about the service from attending doctors and nursing staff, and written feedback from patients presented at patient care committee meetings, was consistently positive with respect to the quality and usefulness of the service. Examples of medicine management recommendations made to patients appear in supplementary Table , according to the cause of their medicine related problem.
Processes and determinants identifying actions taken in the implementation of a discharge medicines review service appear in Table . Processes of context, fidelity, implementer engagement, intervention quality and reach (definitions supplementary Table ) appeared in each phase, as did the following determinants: feasibility; mandate, authority, and accountability; quality assurance and patient safety systems; source of the recommendation. The most commonly occurring determinants were capacity to plan change; implementer engagement; and patient needs, beliefs, knowledge, and motivation.
The implemented service was audited between July 2019 and March 2020. Of the 166 patients recruited, 66 were excluded; 11 were transferred to other hospitals due to the occurrence of an acute medical condition such as bleeding or chest pain, or for a procedure unavailable onsite; six left before interview; no recommendations requiring follow-up were made for 33 patients; nine patients were uncontactable after discharge; three had not seen a doctor within four weeks of discharge; three were admitted to another hospital within two weeks of discharge, and one patients family refused follow-up, leaving 100 patients. All patients/caregivers received a discharge medicine list and review form, and all agreed to participate in a medicines discharge interview and to consider discussing those recommendations important to them with their GP. All patients were followed-up. The pharmacist did not communicate directly with GPs, nor did any GP contact the pharmacist. Mean participant age was 83.1 years, mean total number of medicines 10.4, with a mean number of 8.9 medical conditions per patient. Of 100 patients, five took less than 5 regular medicines, 48 took five to nine regular medicines, and 47 took 10 regular medicines or more - classed as hyper polypharmacy . Fifty six percent of patients were counselled in the presence of a caregiver. Of 368 recommendations made to 100 patients/caregivers, 351 (95%) were actioned, with 284 (77% of those actioned) reported to be implemented and 206 (21%) regularly taken medicines deprescribed – 141 ceased and 65 medicines reduced in dose (Table ). There were 340 causes of a medicine-related problems (MRPs - 3.4 per patient), classified according to a validated system . The top 10 categories represented 92% (312/340) of all causes of MRPs, the most common being: Medicine not effective for the indication treated; medicine was not the most safe/effective; and indication does not warrant medicine treatment (Table ) Medicines for acid-related disorders, multivitamins, complementary and alternative medicines, and mineral supplements were the most common medicines ceased. Gabapentinoids, opiates, proton pump inhibitors and statins were the most common medicines reduced in dose. The time taken to reconcile, review and interview patients/caregivers averaged 63.6 minutes/patient. Recommendations not actioned (17 or 4.6% of the total number) occurred if patients/caregivers decided they were unimportant. Recommendations not implemented occurred because medicines were continued despite evidence provided of poor or absent effectiveness, or GPs considering recommendations unnecessary. Examples included non-discontinuation of glucosamine and prescription of proton pump inhibitors despite apparent lack of indication. Oral feedback about the service from attending doctors and nursing staff, and written feedback from patients presented at patient care committee meetings, was consistently positive with respect to the quality and usefulness of the service. Examples of medicine management recommendations made to patients appear in supplementary Table , according to the cause of their medicine related problem.
Continuing positive feedback and the results of this study resulted in our non-government, not-for-profit (private) hospital commencing and continuing to pay for a non-dispensing or cognitive pharmacy service. Facilitators influencing the implementation of transitional care innovations have been identified and include the benefits and usefulness of the innovation to healthcare providers; patient satisfaction resulting in high buy-in from healthcare providers and management; quality of information transfer; clear roles and responsibilities of key team members; support from allied health and administrative staff; and regular communication and feedback about the innovation . These facilitators appear in this study. Gaining the approval of the Hospital’s executive officers, board of management and medical committee was considered critical in legitimizing the clinical role of the pharmacist. The Hospital supported implementation from inception, providing organizational and policy support. Allied healthcare team support was also essential to facilitate implementation, contributing to the design and evaluation of the service at each stage. This has been found to make interventions more likely to be effective at ward level and represented a participatory action research approach . Such an approach has been used to improve care of delirium in older inpatients and to address inappropriate psychotropic medicine use in residential care . Staff understood that the pharmacist taking time to talk to patients/caregivers about medicines was fundamental to patient care. Patient-centered care appeared to be of low priority in Australian hospitals and internationally [ , , ], featuring poor delivery of information [ , – ]. Transition interventions involving caregivers also appeared uncommon [ , , ] and often with poor pharmacist involvement . Caregivers need to be recognized as partners in management to reduce communication failures and share information received by patients [ , , ]. Care delivered in this study motivated patients/caregivers to become effective facilitators of medicine management change after discharge. Educating patients/caregivers facilitated crossing the primary-secondary interface, where the pharmacist was made the person for accurately determining and explaining the appropriateness of patients’ medicines and providing it in plainly written form . Such a model of pharmacist care did not appear to be standard practice . In a realist synthesis of pharmacist-conducted medicine reviews in discharged patients , factors likely to lead to beneficial outcomes were discussed. Corresponding to these factors, this study engaged healthcare professionals, patients, and caregivers; recruited patients in a trusted environment supportive of the integral role and skill of the pharmacist; established hospital organizational support; provided a pharmacist who understood the critical role of medicine review and integration with staff; and had access to comprehensive information about patients . Handover at transitions of care involved transfer of responsibility to GPs. However, in this study, PIM use was identified and discussed with the patient/caregiver, who were requested to take it up with their GP if it concerned them. This differed from standard practice of pharmacists making recommendations directly to GPs. . GPs then had their attention directed to PIM use by a concerned patient. This proved effective in influencing GPs decision-making behavior (the “nudge” strategy ) through overcoming personal cognitive biases, habits, fear of upsetting the patient, therapeutic inertia (failure to alter therapy when indicated ) or psychological reactance – a motivational state that affirms a person’s freedom of choice, even if opposite to a recommendation . The presence of MRPs after discharge was not unusual, as hospital doctors may not review long-term medicines unrelated to the current admission, viewing it as the GPs role . After discharge, the GP may assume that medicines have been evaluated and were appropriate to continue. Lack of hospital review represented a lost opportunity, as most older Australians were willing to stop one or more of their regular medicines if their GP said they could . Strengths and limitations The behavioural nudge featured in this study requires confirmation . Cost of the service appeared dependent upon pharmacist time per patient. Follow-up was short, although persistence of discharge medicine changes following medicine review have been demonstrated . Patients/caregivers reports of medicine changes were accepted as truthful, with no further form of validation. This study was performed in a small hospital by a single pharmacist, limiting generalisability. No clinical outcomes were reported. However, the implementation process delivered a funded service judged effective by management. There were no patient exclusion criteria other than age, adding to real-world impact.
The behavioural nudge featured in this study requires confirmation . Cost of the service appeared dependent upon pharmacist time per patient. Follow-up was short, although persistence of discharge medicine changes following medicine review have been demonstrated . Patients/caregivers reports of medicine changes were accepted as truthful, with no further form of validation. This study was performed in a small hospital by a single pharmacist, limiting generalisability. No clinical outcomes were reported. However, the implementation process delivered a funded service judged effective by management. There were no patient exclusion criteria other than age, adding to real-world impact.
An implementation program resulted in the commencement of a paid patient-centred discharge medicine review service with an implementation rate of recommendations exceeding that of a previous effort. Failure of patient centred care appeared common in hospitals. This, combined with low rates of medicine review in those recently discharged from hospital , meant that the epidemic of medicine-related harm may remain undiminished.
Additional file 1. Additional file 2.
|
Experience of a modified chest tube suture-fixation technique in uniportal thoracoscopic pulmonary resection | bb055c0d-91f1-41ae-8504-f0d1be385971 | 10061984 | Suturing[mh] | In recent years, video-assisted thoracic surgery (VATS) has evolved as a minimally invasive surgery for diseases in the field of thoracic surgery . The technique has many advantages, including causing less trauma to the chest wall, early remission of postoperative pain, less bleeding, improved cardiopulmonary function, lower complication rates, and rapid recovery. Moreover, patients who start early adjuvant chemotherapy show better immunological responses and stress hormone responses , reduced length of hospital stay, and chest tube indwelling time. VATS allows faster recovery of patients to their normal life and work. Notably, concerns regarding less pain and improved cosmesis fueled the evolution of uniportal access. The uniportal video-assisted thoracic surgery (U-VATS) technique was discovered following the modification and development of the two-port and three-port technique . The technique causes less damage to the integrity of chest wall, eliminate compression of the intercostal nerves with the use of a poking thoracoscope, and reduces local pain associated with the postoperative incision . In 2022, an estimated 1,918,030 new cancer cases and 609,360 cancer-related fatalities were reported in the United States, including approximately 350 deaths per day from lung cancer, making it the leading cause of cancer-related death. With the development of diagnosis and treatment of pulmonary nodules, lung cancer presents a younger trend and the rapid rehabilitation surgery concept has been ingrained into the routines of people. Therefore, most thoracic surgeons may focus on various aspects, including hilar and lung segmental anatomy, surgical techniques, endoscopic instruments, and thoracoscopy surgery, with the aim of improving the capability of surgery. Ignoring surgical incision healing is also important in the outcome and the patient aesthetic requirements. However, considering that there is only one incision, the chest tube is inserted into the thoracic cavity after which an incision suture is formed. Patient activity, abrasion of the drainage tube, fat liquefaction, diabetes, steroids user, chronic kidney/liver disease and incision leakage may increase the possibility of delayed healing after removal of the tube. Generally, delayed healing of the incision, secondary debridement and suture, and scar hyperplasia may occur following chest tube removal. Currently, there is no standard surgical method for effective placement of the chest tube. The conventional methods of intermittent suture and chest tube fixation often increase the formation of hypertrophic scar in postoperative incisions, which does not promote rapid recovery. In the demix suture procedure, appropriate tension, skin temperature, tissue swelling, and adequate healing of the incision skin site are critical to incision healing. Herein, we improved the drainage tube suture-fixation approach which can be potentially applied in many clinical settings. Methods This retrospective study analyzed the clinical characteristics data of 116 patients who underwent U-VATS to treat lung bullae or pulmonary nodules between October 2019 and October 2021 in Zhengzhou People’s Hospital. All the surgical procedures were successfully performed by the same surgical group. Patients were divided into two groups based on different suture techniques i.e., 72 patients who used improved suture methods (active group) and 44 patients who used traditional suture methods (control group). Signed informed consent form was obtained from all participants before the study and ethical approval was provided by the ethics committee of the Zhengzhou People’s Hospital (20,220,124). Inclusion criteria: (1) no clear surgical contraindications in the preoperative examination; (2) successfully completed with U-VATS; (3) a thoracic drainage tube (20 F) was placed at the dorsal part of the chest through the incision, and a 10 F drainage tube was placed percutaneously through the 7th intercostal space on the posterior axillary line; (4) patients with elective surgery. Exclusion criteria: (1) patients in the poor physical condition and unable to tolerate the operation; (2) conversion to multiport VATS or thoracotomy surgery intraoperatively; (3) the drainage tube repositioned without going through the surgical incision; (4) second operation through a similar incision or emergency operation. Surgical approach All patients were given general anesthesia with double-lumen endotracheal intubation. The incision was located between the 4th or 5th rib of the midaxillary line and the front armpit of the surgical incision, with a length of about 3 cm. An incision was made into the skin and subcutaneous tissue, followed by the placement of the protective sleeve of incision (Changzhou Haida Medical Instruments Co., Ltd., disposable incision retractor, HRB-70/70 − 35/25). The procedure was completed using a 10 mm 30° lens with the surgeon located on the ventral side and the camera-holder assistant located at the back of the patient. After the operation, a 20 F drainage tube (Guangdong Xianlai Medical Instrument Co., Ltd., disposable multifunctional drainage tube, S10B) was placed on the dorsal part of the incision, and a single-chamber thoracic drainage bottle (Suzhou New District Ben Q Polymer Medical Instrument Co., Ltd., disposable thoracic drainage device, water-sealed, single-chamber type) was attached. Subsequently, a 10 F drainage tube was percutaneously placed between the mid-axillary line and the 7th intercostal space of the posterior axillary line below the incision (Shenzhen Cooper Technology Development Co., Ltd., disposable drainage catheter and accessories, DC-1025), after which the anti-reflux drainage bag was connected (Coloplast Medical Products Co., Ltd., disposable drainage bag, 1030). Finally, the repair was completed, two chest drainage tubes were positioned, and the chest wall was closed, thereby ending the procedure. Tube suture-fixation method In the active group, the muscle layer was interrupted by suturing with Vicryl Plus (2/0 VCP345H). First, a needle was sutured at the ventral subcutaneous tissue close to the drainage tube as the fixation line of the 20 F drainage tube. The tube was then fixed out of the skin after subcutaneous knotting was passed through the entire layer of the skin. The muscle layer was sutured, and the subcutaneous tissue layer was sutured intermittently with Vicryl Plus (3/0 VCP311H), carefully to ensure that there was no dead space and suitable tension under the skin. Thereafter, the wound was closed starting at the dorsal side 1 cm, away from the edge of the incision with Vicryl Plus (3/0 VCP311H) and left long enough to re-tighten after the removing of the chest tube in the dorsal side. Similar to the subcutaneous tissue suture technique, the needle was placed horizontally through the subcutaneous tissue by passing through the opposite sides of the wound. Notably, the suture continued around the chest tube until the needle reached the other end of the tube. Eventually, the ventral side of the drainage tube was knotted to separate the drainage tube and the remaining incision (Fig. ); the remaining wounds were closed using knotless sutures(Video 1). In the control group, the same suture method was also used to close the muscle, subcutaneous tissue, and skin, except the drainage tube side. The “U-shaped” or “8-shaped” suture line [Vicryl Plus (2/0 VCP345H)] was reserved to knot and close the hole after removing the drainage tube, or the skin was sutured again at the time of removing the drainage tube. Notably, another line externally fixed the tube (Fig. ). Secondary fixation of drainage tube and post-treatment of chest tube removal Once the patient was returned to the ward, a 3 M tape was used to perform secondary fixation of the thoracic drainage tube to prevent unplanned secondary chest tube removal (Fig. ). No unplanned chest tube removal occurred among the 116 patients during this period. On the seconding morning after operation, chest radiographs were taken beside the ward bed, and patients intensity and mode of activity, including getting out of bed and inflating a balloon to promote recovery, were determined based on inspection results. The tube was only removed when the patient was able to deeply inhale and hold, the drainage number was less than 150 mL per 24 h with a clear color, the chest radiograph displayed good ipsilateral lung re-expansion without air leakage. The other drainage tube was kept smooth to relieve the associated pain and enable the patient to resume normal activities. After removing the larger-sized chest tube, the secured thread was pulled forward to tighten the suture and covered in the sterile dressing. The wound was then sealed with a zipper. The leftover thread was cut off after one day and nothing was left over at the scar of the chest tube site. In the control group, the drainage tube was closed by directly knotting the reserved line or suturing it. In both groups, Vaseline and sterile dressings were applied. Follow-up and evaluation indexes The pain score was observed one day after the operation. The indwelling time of the larger-sized chest tube, removal time, wound healing grade at the discharged day, and the wound scar satisfaction were observed one month after the operation (Figs. , and ). A patient-controlled analgesic pump was routinely applied on the day of surgery. The pain was assessed once a day after the operation using of a visual analog scale (VAS, vertical numerical scale ranging from 0 to 10 marked off in units of 1 point, with 0 score indicating no pain and 10 points indicating the worst possible pain. 1~3 points indicated mild pain; 4~6 points indicated moderate pain; and 7~10 points indicated severe pain). There was interference due to postoperative chest tube irritation, coughing, and post-exercise pain; therefore, the patient was asked to indicate the pain index in the steady state. The development of subcutaneous emphysema was rare during the first postoperative day. The total time from disinfection to completion of dressing when the drainage tube was removed was recorded as the chest tube removal time. The incision healing included incision leakage, secondary sutures, and wound infection. The healing was classified into grades A, B, or C based on the incision healing at the time of discharge (Fig. ). Grade A healing represents excellent healing without any adverse reactions; grade B healing refers to poor healing with inflammatory reactions, including redness, induration, hematoma, and effusion, but without purulent; grade C healing indicates that the incision was purulent and needed debridement and drainage. Generally, sutures were removed 10~14 days after removing the thoracic drainage tube based on incision healing. The control group sutures were removed normally, whereas the sutures of active group were disinfected and cut off the skin; the remaining sutures subcutaneously were absorbed after a few days. The Patient Scar Assessment Questionnaire (PASQ) was used to assess scar satisfaction by patients one month after surgery. According to the symptom, vascularization, pigmentation, thickness, relief, and scar pliability, incision scar satisfaction was divided into three responses i.e., 1 = very satisfied; 2 = satisfied; and 3 = dissatisfied. “Very satisfied” means that the incision had healed well with a slightly scar; “satisfied” refers to the incision having a small amount of exudation, delayed healing within three days, with a moderate scar; and “dissatisfied” indicates that the incision was purulent, the incision had not healed after changing the dressing for more than 2 weeks, or apparent scar hyperplasia and pigmentation. Incision scar satisfaction = (the number of patients with very satisfied + number of patients with satisfied)/the total number of patients ×100%. Statistical analyses All statistical analyses were performed using SPSS version 23.0 software (SPSS, Inc., Chicago, IL, USA). Data were expressed as the mean ± standard deviation. Independent sample t-tests were used to compare the differences of parameter variables. The count data was expressed as rate, and the comparison between groups was performed by χ 2 test or Fisher’s exact test. A P -value less than 0.05 was considered statistically significant.
This retrospective study analyzed the clinical characteristics data of 116 patients who underwent U-VATS to treat lung bullae or pulmonary nodules between October 2019 and October 2021 in Zhengzhou People’s Hospital. All the surgical procedures were successfully performed by the same surgical group. Patients were divided into two groups based on different suture techniques i.e., 72 patients who used improved suture methods (active group) and 44 patients who used traditional suture methods (control group). Signed informed consent form was obtained from all participants before the study and ethical approval was provided by the ethics committee of the Zhengzhou People’s Hospital (20,220,124). Inclusion criteria: (1) no clear surgical contraindications in the preoperative examination; (2) successfully completed with U-VATS; (3) a thoracic drainage tube (20 F) was placed at the dorsal part of the chest through the incision, and a 10 F drainage tube was placed percutaneously through the 7th intercostal space on the posterior axillary line; (4) patients with elective surgery. Exclusion criteria: (1) patients in the poor physical condition and unable to tolerate the operation; (2) conversion to multiport VATS or thoracotomy surgery intraoperatively; (3) the drainage tube repositioned without going through the surgical incision; (4) second operation through a similar incision or emergency operation.
All patients were given general anesthesia with double-lumen endotracheal intubation. The incision was located between the 4th or 5th rib of the midaxillary line and the front armpit of the surgical incision, with a length of about 3 cm. An incision was made into the skin and subcutaneous tissue, followed by the placement of the protective sleeve of incision (Changzhou Haida Medical Instruments Co., Ltd., disposable incision retractor, HRB-70/70 − 35/25). The procedure was completed using a 10 mm 30° lens with the surgeon located on the ventral side and the camera-holder assistant located at the back of the patient. After the operation, a 20 F drainage tube (Guangdong Xianlai Medical Instrument Co., Ltd., disposable multifunctional drainage tube, S10B) was placed on the dorsal part of the incision, and a single-chamber thoracic drainage bottle (Suzhou New District Ben Q Polymer Medical Instrument Co., Ltd., disposable thoracic drainage device, water-sealed, single-chamber type) was attached. Subsequently, a 10 F drainage tube was percutaneously placed between the mid-axillary line and the 7th intercostal space of the posterior axillary line below the incision (Shenzhen Cooper Technology Development Co., Ltd., disposable drainage catheter and accessories, DC-1025), after which the anti-reflux drainage bag was connected (Coloplast Medical Products Co., Ltd., disposable drainage bag, 1030). Finally, the repair was completed, two chest drainage tubes were positioned, and the chest wall was closed, thereby ending the procedure.
In the active group, the muscle layer was interrupted by suturing with Vicryl Plus (2/0 VCP345H). First, a needle was sutured at the ventral subcutaneous tissue close to the drainage tube as the fixation line of the 20 F drainage tube. The tube was then fixed out of the skin after subcutaneous knotting was passed through the entire layer of the skin. The muscle layer was sutured, and the subcutaneous tissue layer was sutured intermittently with Vicryl Plus (3/0 VCP311H), carefully to ensure that there was no dead space and suitable tension under the skin. Thereafter, the wound was closed starting at the dorsal side 1 cm, away from the edge of the incision with Vicryl Plus (3/0 VCP311H) and left long enough to re-tighten after the removing of the chest tube in the dorsal side. Similar to the subcutaneous tissue suture technique, the needle was placed horizontally through the subcutaneous tissue by passing through the opposite sides of the wound. Notably, the suture continued around the chest tube until the needle reached the other end of the tube. Eventually, the ventral side of the drainage tube was knotted to separate the drainage tube and the remaining incision (Fig. ); the remaining wounds were closed using knotless sutures(Video 1). In the control group, the same suture method was also used to close the muscle, subcutaneous tissue, and skin, except the drainage tube side. The “U-shaped” or “8-shaped” suture line [Vicryl Plus (2/0 VCP345H)] was reserved to knot and close the hole after removing the drainage tube, or the skin was sutured again at the time of removing the drainage tube. Notably, another line externally fixed the tube (Fig. ).
Once the patient was returned to the ward, a 3 M tape was used to perform secondary fixation of the thoracic drainage tube to prevent unplanned secondary chest tube removal (Fig. ). No unplanned chest tube removal occurred among the 116 patients during this period. On the seconding morning after operation, chest radiographs were taken beside the ward bed, and patients intensity and mode of activity, including getting out of bed and inflating a balloon to promote recovery, were determined based on inspection results. The tube was only removed when the patient was able to deeply inhale and hold, the drainage number was less than 150 mL per 24 h with a clear color, the chest radiograph displayed good ipsilateral lung re-expansion without air leakage. The other drainage tube was kept smooth to relieve the associated pain and enable the patient to resume normal activities. After removing the larger-sized chest tube, the secured thread was pulled forward to tighten the suture and covered in the sterile dressing. The wound was then sealed with a zipper. The leftover thread was cut off after one day and nothing was left over at the scar of the chest tube site. In the control group, the drainage tube was closed by directly knotting the reserved line or suturing it. In both groups, Vaseline and sterile dressings were applied.
The pain score was observed one day after the operation. The indwelling time of the larger-sized chest tube, removal time, wound healing grade at the discharged day, and the wound scar satisfaction were observed one month after the operation (Figs. , and ). A patient-controlled analgesic pump was routinely applied on the day of surgery. The pain was assessed once a day after the operation using of a visual analog scale (VAS, vertical numerical scale ranging from 0 to 10 marked off in units of 1 point, with 0 score indicating no pain and 10 points indicating the worst possible pain. 1~3 points indicated mild pain; 4~6 points indicated moderate pain; and 7~10 points indicated severe pain). There was interference due to postoperative chest tube irritation, coughing, and post-exercise pain; therefore, the patient was asked to indicate the pain index in the steady state. The development of subcutaneous emphysema was rare during the first postoperative day. The total time from disinfection to completion of dressing when the drainage tube was removed was recorded as the chest tube removal time. The incision healing included incision leakage, secondary sutures, and wound infection. The healing was classified into grades A, B, or C based on the incision healing at the time of discharge (Fig. ). Grade A healing represents excellent healing without any adverse reactions; grade B healing refers to poor healing with inflammatory reactions, including redness, induration, hematoma, and effusion, but without purulent; grade C healing indicates that the incision was purulent and needed debridement and drainage. Generally, sutures were removed 10~14 days after removing the thoracic drainage tube based on incision healing. The control group sutures were removed normally, whereas the sutures of active group were disinfected and cut off the skin; the remaining sutures subcutaneously were absorbed after a few days. The Patient Scar Assessment Questionnaire (PASQ) was used to assess scar satisfaction by patients one month after surgery. According to the symptom, vascularization, pigmentation, thickness, relief, and scar pliability, incision scar satisfaction was divided into three responses i.e., 1 = very satisfied; 2 = satisfied; and 3 = dissatisfied. “Very satisfied” means that the incision had healed well with a slightly scar; “satisfied” refers to the incision having a small amount of exudation, delayed healing within three days, with a moderate scar; and “dissatisfied” indicates that the incision was purulent, the incision had not healed after changing the dressing for more than 2 weeks, or apparent scar hyperplasia and pigmentation. Incision scar satisfaction = (the number of patients with very satisfied + number of patients with satisfied)/the total number of patients ×100%.
All statistical analyses were performed using SPSS version 23.0 software (SPSS, Inc., Chicago, IL, USA). Data were expressed as the mean ± standard deviation. Independent sample t-tests were used to compare the differences of parameter variables. The count data was expressed as rate, and the comparison between groups was performed by χ 2 test or Fisher’s exact test. A P -value less than 0.05 was considered statistically significant.
U-VATS was successfully performed in all 116 patients, including 41 men and 75 women, with an average age of 55.78 ± 12.17 years. The participants included 73 patients who underwent limited lung resection surgical methods [wedge (included lung bullae and pulmonary nodules resection), segmentectomy], 25 patients who underwent pulmonary lobectomy, and 18 patients who underwent greater surgery (included lobe + limited lung resection and multi-site limited resection). The median indwelling time of the larger-sized chest drainage tube was 4 days, whereas the length of hospital stay was 7.34 ± 1.71 days. The intensity of pain one day after operation was mostly moderate. The incision sutured again or knotting with the reserved sutures after removing the drainage tube was avoided in the active group, so the chest tube removal time of active group was a great shorter than that in the control group, and a reduction in the chest tube was removed period would avoid patient inconvenience and improve satisfaction. The surgical incision had healed without any complications in the vast majority of cases. However, the incision healing grade and incision scar satisfaction was more lifted appearance than control group. Table comprehensively shows the clinical characteristics of the patients. There were no significant differences in clinical features, surgical methods, chest tube indwelling time, and postoperative pain scores between the two groups ( P > 0.05). However, a significant difference was identified in the chest tube removal time, incision healing grade, and incision scar satisfaction ( P < 0.05) were better in the active group was superior to the control group.
Initially described in 2004 by Rocco for a wedge resection, the U-VATS technique is appealing to most surgeons because it involves potentially less postoperative pain and is responsible for higher patient satisfaction regarding the number of incisions. Gonzalez first used the technique for thoracoscopic lobectomy; as a result, its application has been broadened given the increased experience. Over the years, the U-VATS technique has been applied to various types of operation, including lobectomy, segmentectomy, sleeve resection of lung surgery [ – ], and mediastinal tumor surgery, including thymectomy and esophagectomy . After a thoracic operation, the chest drainage is routinely placed with an underwater seal. However, the position at which the tube is placed differs between the center and posterior edge of the incision. We recently placed on the dorsal part of the incision. Furthermore, the suture method, nutritional status, drainage tube diameter, and chest tube indwelling time have varying effects on wound healing. At present, limited studies have investigated the effect of the abdominal drainage tube e.g., the indwelling drainage tube effect on postoperative recovery after laparoscopic appendicitis . Limited studies, however, have explored the effect of the thoracic drainage tube. This study aims to explore a modified suture-fixation technique, geared towards improving the clinical application. A suture is important in maintaining moderate tension, thereby ensuring adequate healing of the incision site. The ideal surgical suture should meet the following requirements: moderate tension, precise hemostasis, no dead space, and no permanent or only a few suture marks. Despite the many types of sutures, it is important to select suture materials matching the tension and healing speed of the tissue. After removing the chest drainage tube, the wound usually takes approximately 10~14 days to heal. The Vicryl Plus is a type of absorbable glycan lactic acid suture, with a tensile strength of about 75% when sutured for 14 days and approximately 25% when sutured for 28 days. The suture is completely absorbed after 56–70 days. Several studies indicate that the suture yields satisfactory results in clinical practice [ – ]. The clinical efficacy and suitability of knotless barbed absorbable sutures have long been reported with remarkable benefits [ – ]. However, the Vicryl Plus material could yield better clinical outcomes due to the hard texture and the cost. With the improvement of the popularization of absorbable sutures and conditions for beauty, there is a rapid rise in surgeries including “small thyroid low suprasternal incisions and circular areola incisions for implantation of prostheses” in the Department of thyroid gland and breast as well as plastic surgery, “concealed incisions in the cavity” in the Department of Otolaryngology, and “single port laparoscopy, anastomosis, and specimen collection” in the Department of Gastrointestinal Surgery and Obstetrics and Gynecology. Therefore, surgeons have focused on minimally invasive and cosmetic requirements. Thoracic surgery is mostly categorized in grade three or four operations, where the trauma is relatively larger. Noteworthy, the healing of the surgical incision intuitively indicates the mental stress of the patient and promotes rapid recovery. However, cases of delayed healing or even poor healing have been reported after removing the tube, substantially increasing the mental burden and extending the recovery time. This is unlike the concept of rapid recovery. In this regard, the present study evaluated a modified suture method that can properly fix the chest tube, accelerate incision healing, and increase the beauty of the incision simultaneously. After thoracic surgery, proper indwelling and adequate drainage of the tube contribute to early postoperative getting out of bed and reducing the rate of postoperative complications, hence an important part of perioperative treatment and rapid recovery. The currently available drainage tube fixation methods and post-chest tube removal treatment approaches involve ligation of extracutaneous sutures, followed by ligation and fixation again with sutures approximately 1 cm above the drainage tube out of the skin. After removing the drainage tube, the suture can be immediately ligated when the “U-shaped” or “8-shaped” suture is reserved during the operation. A re-suture under local anesthesia is necessary without a reserved thread. After closing the wound, it is covered with Vaseline gauze and multi-layer dressing to promote healing. Although research on the fixation of the drainage tube has been documented, most of these tubes are thin. At present, most of the used drainage tubes are larger-sized or both larger-sized, and ultrafine tubes are used simultaneously to promote activities of thoracic hemorrhage and rapid changes in the condition after thoracic surgery. This study investigated a larger-sized tube suture-fixation method. After a certain period of exploration and accumulation, this method, combined with a 3 M tape for secondary fixation, yielded the desired effect on the fixation of the chest drainage tube. Results revealed that none of the patients in this group had unplanned chest tube removal. Besides, the chest tube removal time wound healing grade, and incision scar satisfaction was significantly different unlike that of ordinary suture methods in the same period, suggesting that it merits clinical promotion. Our experience is as follows: Given that U-VATS incisions are usually like “trapezoidal”, skin incisions were small, whereas subcutaneous and muscle layer incisions were large. Therefore, since the focus should be channeled towards the suture of the subcutaneous and muscle layers on both sides of the wound to prevent leakage near the drainage tube, we recommended an interrupted sutures method. After suturing the muscular layers on both sides of the drainage tube, the drainage tube was unable to slide up and down. The muscle layer was sutured with Vicryl Plus (2/0 VCP345H). The suture was left on the ventral side of the drainage tube and used as the drainage tube fixation line. After several ligations, the drainage tube was ligated approximately 1 cm outside the skin. The subcutaneous tissue was sutured using Vicryl Plus (3/0 VCP311H), and the skin was sutured from the dorsal side to the ventral side by an intradermal suture method. The needle was inserted approximately 1 cm outside the incision at the side of the drainage tube (the dorsal side), and approximately 5 cm was reserved outside the skin as a reserved line for removing the tube. The needle was placed horizontally through the subcutaneous tissue by passing through the opposite sides of the wound similar to the continuous subcutaneous suture technique. A knot was tied on the inside of the drainage tube and the incision to prevent leakage or poor healing near the drainage tube opening from spreading to the entire incision. The suture continues around the chest tube until the needle reached the other end of the incision. When cutting off the bundling drainage line, the end should be left into the subcutaneous tissue to facilitate suture absorption and avoid cutting the intradermal suture. When the tube is removed, the Vaseline gauze is compressed, the reserved suture was simultaneously tightened, and then the wound was covered with multiple dressings. Eventually, the wound was sealed with a zipper, and the suture was temporarily covered in the dressings. In the short term after chest tube removal, the patient was instructed to press the chest tube removal site when coughing, and avoid excessive lifting of the ipsilateral arm to increase the tension of the incision simultaneously. After one day, the sutures were pulled and knotted outside the skin, before cutting the excess sutures. After 10~14 days, sutures were removed based on the healing of the incision. Our team often uses an ultrafine chest tube (10 F) combined with a traditional larger-sized tube (20 F) after pulmonary U-VATS. Nonetheless, a larger-sized tube was often placed in the incision, increasing the risk of poor wound healing. With the deepening concept of rapid rehabilitation and the development of thoracoscopy technology as well as Subxiphoid uniportal VATS , the use of two 10 F pigtail tubes instead of the traditional larger-sized tube , as well as exploration of tubeless thoracoscopic surgery without endotracheal intubation and no chest drainage tube after surgery, has proven to be safe and feasible in a specific selected patient population [ – ]. However, this approach has limitations. First, it is still too early to be an alternative to routine surgical procedures and cannot be applied to wider populations . This study was inspired by several explorations of suture technology and drainage tube placement [ , – ]. Given that this is a preliminary retrospective study, it has compelling limitations, including the retrospective nature and subjective bias, differences caused by different group members, no record of duration for each suture, and inadequate follow-up period. Additionally, we excluded patient factors in the analysis, including obesity, diabetes, steroid user, chronic kidney/liver disease, etc. Although the maximum indwelling time of the larger-sized chest drainage tube was 13 days in the modified method, it can significantly increase with more experience. Therefore, additional scar assessment is necessary.
In conclusion, the thoracic drainage tube modified suture technique and chest tube fixation method described in this work are safe and effective with substantial cosmetic outcome. Moreover, the technique is a key component for the enhanced recovery surgery (ERAS) of thoracic surgery patients. We believe that this modified chest tube suture-fixation technique will be used in more patients.
Below is the link to the electronic supplementary material. Video 1: Experience of a modified chest tube suture-fixation technique in uniportal thoracoscopic pulmonary resection.
|
360° Virtual reality to improve patient education and reduce anxiety towards atrial fibrillation ablation | c217b033-c059-4c52-b1fd-a0951d2f795b | 10062331 | Patient Education as Topic[mh] | Catheter ablation for atrial fibrillation (AF) has emerged as an important rhythm control strategy and is used more often and earlier in the disease process. With the increasing prevalence of arrhythmia, catheter ablation for AF by means of pulmonary vein isolation is by far the most commonly performed cardiac ablation procedure worldwide. As recent data demonstrated that a shorter time between first AF documentation and catheter ablation is associated with lower arrhythmic recurrences and that catheter ablation is related to symptom reduction and quality of life improvement, , the number of AF ablation procedures is expected to keep growing. AF catheter ablation is associated with typical procedure-related complications and may lead to pre-/periprocedural anxiety and stress in patients. Therefore, adequate preparation and information provision for patients and their relatives is essential. The current European Society of Cardiology (ESC) guidelines recommend the organization of AF management in an integrated care model. One fundamental component of the integrated care approach in AF is patient involvement. Patients should be provided with appropriate education and instructions, be empowered to self-manage their condition and should be actively involved in treatment decision-making. Especially patient education may influence patient-reported outcomes such as post-procedural complications, quality of life, therapy adherence, anxiety, as well as disease-related knowledge and satisfaction. However, effective patient instruction takes time and requires adequate resources. Virtual reality (VR) may provide new dimensions for education as patients can virtually experience the hospital ward and operating room, and receive a step-by-step explanation of the course of their hospitalization and the procedure itself. Being able to review and discuss these experiences with relatives later may increase the patient's involvement in the procedure. Recently, VR has been used for providing periprocedural ‘distraction’ (during sedation and catheter ablation) to reduce patients’ experience of pain, anxiety, consumption of pain medication, and procedure length. However, little is known about the use of VR as preprocedural education tool for patients planning for AF ablation. In this study, we assessed the effects of a lifelike 360° VR preprocedural patient education video on information provision, procedure-related knowledge, satisfaction, and the level of worries in patients planned for AF ablation. At the same time, we evaluated the feasibility of a disposable cardboard VR viewer for home use in this setting.
Study design This study is embedded in the prospective Intensive Molecular and Electropathological Characterization of patientS undergOing atriaL fibrillATion ablatION (ISOLATION) cohort study (ClinicalTrials.gov identifier: NCT04342312) performed at the Maastricht University Medical Centre + (MUMC+), Maastricht, The Netherlands. The ISOLATION study comprises an extensive standardized diagnostic work up pre-AF ablation via an ambulant pre-ablation AF care pathway. The study aims to identify predictors of successful AF ablation in the following domains: (i) clinical factors, (ii) AF patterns, (iii) anatomical characteristics, (iv) electrophysiological characteristics, (v) circulating biomarkers, and (vi) genetic background, and has been described in detail elsewhere. The study was performed in compliance with the Declaration of Helsinki and approved by the Institutional Review Board at the MUMC+/Maastricht University (NL 70787.068.19; METC 19–052). All patients provided written informed consent. Study population We prospectively enrolled consecutive patients (≥18 years of age) with symptomatic paroxysmal or persistent AF scheduled for AF ablation, who were included in the ISOLATION study and who visited the pre-ablation AF care pathway of the MUMC + between January 2020 and August 2021. Patients were excluded from the ISOLATION study if they were deemed unfit to participate due to a serious medical condition before ablation, at the discretion of their treating physician, and if they underwent an emergency ablation procedure. All the patients qualified as candidates for either the intervention group (VR group) or the control group. All patients were alternatively assigned in a 1:1 ratio to the control or VR group, such that one patient would be included in the control group and the next will be included in the VR group. Patients were scheduled for this pathway on a random basis and for that reason, re-randomization has been waived. The control group received standard preprocedural information through oral counselling and information leaflets. The VR group received the standard information, as well as a short dedicated 360° VR video via in-hospital wireless rechargeable VR headset [Oculus © Go standalone VR headset (Meta Platforms Inc., California, United States)] and a disposable cardboard VR viewer for home use in combination with patients’ own smartphone and a free available app. 360° virtual reality video The VR video used in this study consisted of a short (4:18 min) educational recording in the Dutch language. It provides a short oral summary of preparation advices, an overview of the ablation procedure and catheterization environment, as well as post-procedural (self-)care and self-monitoring knowledge, while the video images show a nearly lifelike insight into the patient's tour across the hospital ward and operating room in a 360° environment ( Figure ). The VR video was developed at the MUMC + by a multidisciplinary team that consisted of EP nurses, a cardiologist, an anaesthesiologist, a marketing and communication advisor and the hospital’s patient advisory board. The patient advisory board consisted of a representative group of volunteers who were all (ex)patients or patients’ relatives. Prior to writing the script and the video recordings, they were asked about their wishes and expectations for such an informative video. This input was included in the realization of the film. At a later stage, they re-evaluated the raw cut of the film and provided comments that were included in the final cut of the film. Prior to the implementation of the VR film at the outpatient clinic, the board tested and approved the VR film. The patient advisory board was not involved in the further design and implementation of the present study. The video was realized in collaboration with Infor-Med, a health technological platform focussed on virtual medical information, and could be downloaded and seen via the free available Infor-Med app (Visyon 360° Immersive Experiences). Questionnaires Patients were asked to complete the Amsterdam Preoperative Anxiety and Information Scale (APAIS) together with some additional questions concerning procedural experience, ease of use of the disposable cardboard VR viewer and in-hospital VR headset, and satisfaction with both pre- and post-ablation via the SurveyMonkey online survey tool (SurveyMonkey Inc., San Mateo, California, United States, www.surveymonkey.com ). A link to each questionnaire was sent via email to the patients. The pre-ablation questionnaire was sent within the first week after visiting the AF outpatient clinic pre-ablation work up pathway. The post-ablation questionnaire was sent within the first week after AF ablation, and only to patients that completed the first questionnaire. An email reminder was sent one week after the initial contact to encourage those who did not complete the questionnaires. Translations of the complete questionnaires are provided in , . Data collection Baseline patient characteristics (demographics and medical history) were retrieved from patients’ electronic case report forms. Data on information provision, procedure-related knowledge, satisfaction, and the level of worries pre- and post-ablation were obtained from the questionnaires. The results of the questionnaires were collected in the secured SurveyMonkey cloud, only accessible to authorized investigators. All data points were exported in comma-separated values format. Data analysis The results of both questionnaires were pointed on a 5-point Likert scale from 1 = strongly disagree to 5 = strongly agree. Ratings of strongly disagree, disagree, and neutral were grouped together and ratings of agree and strongly agree were grouped together. The number of patients with a Likert scale score of 4 or 5 (agree or strongly agree) was assessed per statement. Both questionnaires comprised statements regarding information provision, procedure-related knowledge, satisfaction, and level of worries. Information provision was represented by the following statement: ‘I have been clearly informed’. Four statements represented procedure-related knowledge: ‘I know what to expect, before during and after the procedure’, ‘I know in which environment I will be treated, this reassures me’, ‘I know what will happen during the procedure’ and ‘I know how to prepare for the procedure’. Satisfaction was presented by the statements: ‘The patient information made me feel more positive about the procedure’, ‘The patient information made me feel more positive about the anaesthesia’ and ‘The patient information has answered all my questions regarding the procedure’. Level of worries was presented by the following statements: ‘I am worried about the procedure’ and ‘I am worried about the anaesthesia. Furthermore, both questionnaires comprised the six statements of the APAIS. Four statements represented anxiety of AF ablation or anaesthesia: ‘I am worried about the procedure’; ‘The procedure is on my mind continually’, ‘I am worried about the anaesthesia’ and ‘The anaesthesia is on my mind continually’. Two statements represented the need for information: ‘I would like to know as much as possible about the AF ablation procedure’ and ‘I would like to know as much as possible about the anaesthesia’. The answers to those statements were evaluated in two scales: the anxiety score and the information score. The anxiety score was the sum of the four statements representing anxiety with a scoring range from 4 to 20. The information score was the sum of the two statements representing the need for information, with a scoring range from 2 to 10. Patients with an anxiety score ≥11 were considered as anxiety cases. A value of ≥11 on the anxiety scale is the validated cut-off for anxious patients in the clinical setting. Patients with an information score below 5 were classified as having no or little information requirement, patients with an information score between 5 and 7 had an average information requirement and those with an information score of 8–10 were classified as having a high information requirement, as described elsewhere. Statistical analysis All continuous variables were tested for normality with the Shapiro–Wilk test. Variables with normal distribution were expressed as mean ± standard deviation (SD). Non-parametric variables were presented as median [interquartile range (IQR)] and categorical variables as numbers (n) with percentages (%). Differences in continuous parameters were compared using independent-samples t-tests in case of parametric variables and Mann–Whitney U tests in case of non-parametric variables. For the comparison of categorical data, the McNemar tests (paired comparison) or Pearson’s chi-squared tests (unpaired comparison) were used. A two-sided P -value of 0.050 was considered statistically significant. For database management and statistical analysis, we used IBM SPSS Version 28 (IBM Corporation, Somers, New York, USA).
This study is embedded in the prospective Intensive Molecular and Electropathological Characterization of patientS undergOing atriaL fibrillATion ablatION (ISOLATION) cohort study (ClinicalTrials.gov identifier: NCT04342312) performed at the Maastricht University Medical Centre + (MUMC+), Maastricht, The Netherlands. The ISOLATION study comprises an extensive standardized diagnostic work up pre-AF ablation via an ambulant pre-ablation AF care pathway. The study aims to identify predictors of successful AF ablation in the following domains: (i) clinical factors, (ii) AF patterns, (iii) anatomical characteristics, (iv) electrophysiological characteristics, (v) circulating biomarkers, and (vi) genetic background, and has been described in detail elsewhere. The study was performed in compliance with the Declaration of Helsinki and approved by the Institutional Review Board at the MUMC+/Maastricht University (NL 70787.068.19; METC 19–052). All patients provided written informed consent.
We prospectively enrolled consecutive patients (≥18 years of age) with symptomatic paroxysmal or persistent AF scheduled for AF ablation, who were included in the ISOLATION study and who visited the pre-ablation AF care pathway of the MUMC + between January 2020 and August 2021. Patients were excluded from the ISOLATION study if they were deemed unfit to participate due to a serious medical condition before ablation, at the discretion of their treating physician, and if they underwent an emergency ablation procedure. All the patients qualified as candidates for either the intervention group (VR group) or the control group. All patients were alternatively assigned in a 1:1 ratio to the control or VR group, such that one patient would be included in the control group and the next will be included in the VR group. Patients were scheduled for this pathway on a random basis and for that reason, re-randomization has been waived. The control group received standard preprocedural information through oral counselling and information leaflets. The VR group received the standard information, as well as a short dedicated 360° VR video via in-hospital wireless rechargeable VR headset [Oculus © Go standalone VR headset (Meta Platforms Inc., California, United States)] and a disposable cardboard VR viewer for home use in combination with patients’ own smartphone and a free available app.
The VR video used in this study consisted of a short (4:18 min) educational recording in the Dutch language. It provides a short oral summary of preparation advices, an overview of the ablation procedure and catheterization environment, as well as post-procedural (self-)care and self-monitoring knowledge, while the video images show a nearly lifelike insight into the patient's tour across the hospital ward and operating room in a 360° environment ( Figure ). The VR video was developed at the MUMC + by a multidisciplinary team that consisted of EP nurses, a cardiologist, an anaesthesiologist, a marketing and communication advisor and the hospital’s patient advisory board. The patient advisory board consisted of a representative group of volunteers who were all (ex)patients or patients’ relatives. Prior to writing the script and the video recordings, they were asked about their wishes and expectations for such an informative video. This input was included in the realization of the film. At a later stage, they re-evaluated the raw cut of the film and provided comments that were included in the final cut of the film. Prior to the implementation of the VR film at the outpatient clinic, the board tested and approved the VR film. The patient advisory board was not involved in the further design and implementation of the present study. The video was realized in collaboration with Infor-Med, a health technological platform focussed on virtual medical information, and could be downloaded and seen via the free available Infor-Med app (Visyon 360° Immersive Experiences).
Patients were asked to complete the Amsterdam Preoperative Anxiety and Information Scale (APAIS) together with some additional questions concerning procedural experience, ease of use of the disposable cardboard VR viewer and in-hospital VR headset, and satisfaction with both pre- and post-ablation via the SurveyMonkey online survey tool (SurveyMonkey Inc., San Mateo, California, United States, www.surveymonkey.com ). A link to each questionnaire was sent via email to the patients. The pre-ablation questionnaire was sent within the first week after visiting the AF outpatient clinic pre-ablation work up pathway. The post-ablation questionnaire was sent within the first week after AF ablation, and only to patients that completed the first questionnaire. An email reminder was sent one week after the initial contact to encourage those who did not complete the questionnaires. Translations of the complete questionnaires are provided in , .
Baseline patient characteristics (demographics and medical history) were retrieved from patients’ electronic case report forms. Data on information provision, procedure-related knowledge, satisfaction, and the level of worries pre- and post-ablation were obtained from the questionnaires. The results of the questionnaires were collected in the secured SurveyMonkey cloud, only accessible to authorized investigators. All data points were exported in comma-separated values format.
The results of both questionnaires were pointed on a 5-point Likert scale from 1 = strongly disagree to 5 = strongly agree. Ratings of strongly disagree, disagree, and neutral were grouped together and ratings of agree and strongly agree were grouped together. The number of patients with a Likert scale score of 4 or 5 (agree or strongly agree) was assessed per statement. Both questionnaires comprised statements regarding information provision, procedure-related knowledge, satisfaction, and level of worries. Information provision was represented by the following statement: ‘I have been clearly informed’. Four statements represented procedure-related knowledge: ‘I know what to expect, before during and after the procedure’, ‘I know in which environment I will be treated, this reassures me’, ‘I know what will happen during the procedure’ and ‘I know how to prepare for the procedure’. Satisfaction was presented by the statements: ‘The patient information made me feel more positive about the procedure’, ‘The patient information made me feel more positive about the anaesthesia’ and ‘The patient information has answered all my questions regarding the procedure’. Level of worries was presented by the following statements: ‘I am worried about the procedure’ and ‘I am worried about the anaesthesia. Furthermore, both questionnaires comprised the six statements of the APAIS. Four statements represented anxiety of AF ablation or anaesthesia: ‘I am worried about the procedure’; ‘The procedure is on my mind continually’, ‘I am worried about the anaesthesia’ and ‘The anaesthesia is on my mind continually’. Two statements represented the need for information: ‘I would like to know as much as possible about the AF ablation procedure’ and ‘I would like to know as much as possible about the anaesthesia’. The answers to those statements were evaluated in two scales: the anxiety score and the information score. The anxiety score was the sum of the four statements representing anxiety with a scoring range from 4 to 20. The information score was the sum of the two statements representing the need for information, with a scoring range from 2 to 10. Patients with an anxiety score ≥11 were considered as anxiety cases. A value of ≥11 on the anxiety scale is the validated cut-off for anxious patients in the clinical setting. Patients with an information score below 5 were classified as having no or little information requirement, patients with an information score between 5 and 7 had an average information requirement and those with an information score of 8–10 were classified as having a high information requirement, as described elsewhere.
All continuous variables were tested for normality with the Shapiro–Wilk test. Variables with normal distribution were expressed as mean ± standard deviation (SD). Non-parametric variables were presented as median [interquartile range (IQR)] and categorical variables as numbers (n) with percentages (%). Differences in continuous parameters were compared using independent-samples t-tests in case of parametric variables and Mann–Whitney U tests in case of non-parametric variables. For the comparison of categorical data, the McNemar tests (paired comparison) or Pearson’s chi-squared tests (unpaired comparison) were used. A two-sided P -value of 0.050 was considered statistically significant. For database management and statistical analysis, we used IBM SPSS Version 28 (IBM Corporation, Somers, New York, USA).
During the inclusion period, of 157 patients planned for AF ablation in the MUMC+, 134 patients (85.4%) completed the pre-ablation questionnaire and were included in this analysis. The median age was 66 (58–72) years, and 51 patients (38.1%) were female. Of the included patients, 66 (49.2%) were assigned to the control group and 68 (50.7%) to the VR group. Clinical characteristics were comparable between both groups and presented in Table . Preprocedural ablation questionnaire The number of patients that were worried about the ablation procedure was lower in the VR group compared to the control group [13 (19.1%) vs. 27 (40.9%), P = 0.006] ( Figure and , ). With regards to sex subgroups, especially the number of males that agreed or strongly agreed with the statement that they had worries about the ablation procedure was lower in the VR group than in the control group [5 (11.1%) vs. 16 (42.1%), P = 0.001]. In the VR subgroup, females (as compared with males) were more worried about the ablation procedure [8 (34.8%) vs. 5 (11.1%), P = 0.026] and anaesthesia [5 (21.7%) vs. 2 (4.4%), P = 0.039] ( Figure and , ). The number of patients <65 years that worried about the ablation procedure was lower in the VR group as compared to the controls [7 (21.2%) vs. 13 (50.0%), P = 0.020]. In patients aged ≥65 years, the VR group (vs. control group) was reported to be clearly better informed about the catheterization laboratory environment [29 (82.9%) vs. 22 (55.0%), P = 0.010] ( Figure and , ). Amsterdam preoperative anxiety and information scale Information and anxiety scores were comparable between VR and the control group (see , ). With regards to sex subgroups, females (compared with males) in the VR group had higher anxiety scores [11 (9–14) vs. 9 (6–12), P = 0.049] (see , ). Additionally, there were no differences in information and anxiety scores when comparing younger females with older females as well as with younger males (see , ). In-hospital virtual reality headset vs. disposable cardboard virtual reality viewer Of 66 patients, 74.2% indicated the in-hospital VR headset be realistic. The in-hospital VR headset was reported to be valuable by 65.2%, easy to understand by 72.7% and easy to use by 62.1%. There were no differences in experiences with in-hospital VR headset between younger and older patients, and males and females (see , ). In total, 32 patients in the VR group (48.5%) used the disposable cardboard VR viewer at home. Of these patients, 19 (59.4%) agreed or strongly agreed that the disposable cardboard was easy to use and 25 (78.1%) still had the cardboard VR viewer after the ablation procedure. In 22 patients (68.8%), home use of the video resulted in discussions with relatives. Fourteen patients (43.7%) thought the in-hospital VR headset was of no added value above the disposable cardboard VR viewer ( Figure ). Post-procedural questionnaire and pre- and post-comparison Of the 134 included patients, 101 patients (75.4%) completed the post-ablation questionnaire. Answers to the post-ablation statements were comparable between VR and the control group (see , ). The number of VR patients that were satisfied with the preprocedural information provision was higher post-ablation than pre-ablation [40 (83.3%) vs. 29 (60.4%), P = 0.007] ( Figure and , ). This finding holds true when we performed subanalyses in females (see , ). Controls aged <65 years were less satisfied based on preprocedural information provision post-ablation as compared with pre-ablation [15 (65.2%) vs. 22 (95.7%), P = 0.039] (see , ).
The number of patients that were worried about the ablation procedure was lower in the VR group compared to the control group [13 (19.1%) vs. 27 (40.9%), P = 0.006] ( Figure and , ). With regards to sex subgroups, especially the number of males that agreed or strongly agreed with the statement that they had worries about the ablation procedure was lower in the VR group than in the control group [5 (11.1%) vs. 16 (42.1%), P = 0.001]. In the VR subgroup, females (as compared with males) were more worried about the ablation procedure [8 (34.8%) vs. 5 (11.1%), P = 0.026] and anaesthesia [5 (21.7%) vs. 2 (4.4%), P = 0.039] ( Figure and , ). The number of patients <65 years that worried about the ablation procedure was lower in the VR group as compared to the controls [7 (21.2%) vs. 13 (50.0%), P = 0.020]. In patients aged ≥65 years, the VR group (vs. control group) was reported to be clearly better informed about the catheterization laboratory environment [29 (82.9%) vs. 22 (55.0%), P = 0.010] ( Figure and , ).
Information and anxiety scores were comparable between VR and the control group (see , ). With regards to sex subgroups, females (compared with males) in the VR group had higher anxiety scores [11 (9–14) vs. 9 (6–12), P = 0.049] (see , ). Additionally, there were no differences in information and anxiety scores when comparing younger females with older females as well as with younger males (see , ).
Of 66 patients, 74.2% indicated the in-hospital VR headset be realistic. The in-hospital VR headset was reported to be valuable by 65.2%, easy to understand by 72.7% and easy to use by 62.1%. There were no differences in experiences with in-hospital VR headset between younger and older patients, and males and females (see , ). In total, 32 patients in the VR group (48.5%) used the disposable cardboard VR viewer at home. Of these patients, 19 (59.4%) agreed or strongly agreed that the disposable cardboard was easy to use and 25 (78.1%) still had the cardboard VR viewer after the ablation procedure. In 22 patients (68.8%), home use of the video resulted in discussions with relatives. Fourteen patients (43.7%) thought the in-hospital VR headset was of no added value above the disposable cardboard VR viewer ( Figure ).
Of the 134 included patients, 101 patients (75.4%) completed the post-ablation questionnaire. Answers to the post-ablation statements were comparable between VR and the control group (see , ). The number of VR patients that were satisfied with the preprocedural information provision was higher post-ablation than pre-ablation [40 (83.3%) vs. 29 (60.4%), P = 0.007] ( Figure and , ). This finding holds true when we performed subanalyses in females (see , ). Controls aged <65 years were less satisfied based on preprocedural information provision post-ablation as compared with pre-ablation [15 (65.2%) vs. 22 (95.7%), P = 0.039] (see , ).
This study presents the development process, feasibility, and added value of a lifelike 360° VR patient education video about AF ablation. The VR video provides a short oral summary of preparation advices, an overview of the ablation procedure and catheterization environment as well as post-procedural (self-)care and self-monitoring knowledge, while the video images show a nearly lifelike insight into the patient's tour across the hospital ward and operating room in a 360° environment. We showed that this 360° VR patient education video led to better information provision and procedural knowledge as compared to standard oral and written information. Besides, it reduced preprocedural worries and anxiety in the overall population. Moreover, the video increased patient satisfaction after going through the procedure as patients indicated to be felt better prepared in retrospect. Impact of the virtual reality intervention on patient-related outcomes The use of VR is increasingly embraced in patient education and may encourage the active involvement of patients in their treatment, and provide new dimensions for education provision. VR has the potential to be a positive intervention with regards to patient knowledge, satisfaction, and anxiety. VR lets patients virtually experience the hospital environment and offers the possibility to review and discuss this with family or relatives at home repeatedly, as also shown in the present study. VR glasses allow the creation of a real-life experience also for patients with reduced mobility, and previous studies showed good applicability also in elderly patients. This is in line with findings from our current study, where older patients (≥65 years) in the VR group reported higher patient satisfaction scores and felt less worried about the ablation procedure than those in the control group. Nevertheless, combined information and anxiety scores in the APAIS were comparable between the VR and the control group. As a comparison, previous studies demonstrated that video-based information provision did not increase patients’ anxiety towards a procedure significantly and that oral-, written- and video-based information can be considered equivalent to patient anxiety scores. Interestingly, within our cohort, it turned out that the subgroup of (younger) females was more worried about the ablation procedure after watching the VR film as compared with males, where this difference was not observed in the control group. This would line up with the idea that comprehensive and detailed patient information, especially regarding the risks of an invasive procedure, but also the confrontation and enlivening of an upcoming procedure can cause adverse effects on patients’ anxiety. These findings show that when using mobile health (mHealth) products, their possible side effects such as reinforcement of anxiety must also be considered. Further studies would be necessary to confirm these findings and, in particular, to investigate the possible influence of gender on side effects in more detail (within-patient differences). However, previous studies indicated that patients prefer detailed procedure-based information. The use of mHealth tools should be supported by a structured, integrated and patient-centred management pathway, as also widely described in current literature and guidelines. , This might allow close support of patients in case of important side effects. In addition, the subgroup of patients older than 65 years indicated less worried about the ablation procedure and were more likely to have a ≥ 80% satisfaction score based on procedure-based information provision after watching the VR film as compared to controls. Age should therefore certainly not be an exclusion criterion for using mHealth tools. Disposable/reusable cardboard vs. commercially available virtual reality headset Within our study group, patients evaluated both in-hospital VR as well as disposable cardboard VR positively. Patients who used the disposable cardboard VR viewer showed high satisfaction scores and a majority of patients even kept using the disposable VR glasses at home after the procedure and showed it to friends or family at home. Home use of the disposable cardboard VR viewer initiated discussions in the private environment of patients. These discussions with relatives could have had an (positive) effect on attitudes and anxiety preceding AF ablation. However, further qualitative data would be required to state this. Another advantage of the disposable cardboard VR compared to the in-hospital standalone headset is its relatively low price. Additionally, patients can use it more flexibly and repetitively, or even for different VR programmes within the cardiology department. However, the implementation of VR in clinical AF care, as well as training for the use of in-hospital VR glasses might be time-consuming for healthcare providers. Time and personnel required for maintenance as well as costs for hygienic cleaning and wear and tear of the VR device must also be considered. Implementation of virtual reality in current daily cardiology practice There was no difference between our study groups regarding the use of (telephone) contact moments between the patient and the hospital. However, no further data reported by caregivers with regard to expenses in time or personnel, as well as possible impact on the patient–provider relationship was collected within the scope of this study. A higher workload, especially when implementing new technology, might still be a point of discussion. Compared with oral face-to-face information provision, video-based information provision, however, was previously considered equivalent or requires even fewer physicians’ time. This might be important in an expected lack of medical and caregiver staff in the coming decades, where modern mHealth technologies might help in overcoming obstacles. The current ESC guidelines recommend providing AF care through a multidisciplinary AF team approach rather than by one single healthcare professional alone. Such an approach would allow important roles for nurses and allied health professionals in educating patients and coordinating AF care, while closely collaborating with a medical specialist. Therefore, the integration of a multidisciplinary team for the management of AF ablation patients might improve patient education provision and reduce professionals’ workload. Implementation of VR-based patient education could improve patient understanding of therapy, improve patient compliance and ameliorate shared decision-making. Therefore, the use of VR could enhance the patient–provider relationship in a real-world setting. This is an important aspect to consider as VR should support but not replace the real-world patient–provider relationship. Currently, mHealth tools and VR are finding their way into modern disease management, but the participation of healthcare providers in the early stages of mHealth and VR developments will be still necessary to ensure and guide patient-centred development. Furthermore, Goldberger et al. pointed out that video-based information provision may provide a more standardized and complete story than oral discussion. Another study discussed the advantage of video material in patient education for patients with low literacy skills and explained this by a less stressful environment for the patient when compared with in-hospital discussion with a doctor or a nurse. In addition, the TeleCheck-AF project showed that age is no barrier to remote patient information and education. Interactivity is also an important part of holistic patient education, which we aimed to provide in our intervention group by creating a 360° lifelike VR environment. This might make procedure-based information provision even more explanatory and make patients feel more engaged, as discussed in prior work. One obstacle in the broad implementation of VR-based patient information, however, might be funding. Further studies need to be conducted regarding cost-effectiveness. VR will become widely used in the teaching and training of the physicians, but our study also shows its possibilities in the telemedical management of complex diseases. In addition, VR may also play a role in remote digital patient care pathways such as the VIRTUAL-SAFARI project. However, there is still a lack of medical guidelines in this area that needs to be overcome. Current and future perspectives of virtual reality in cardiovascular care VR has the potential for applications in direct patient treatment. Primary applications in cardiology are centred on education. VR can assist patients and health care providers in cardiovascular care, planning and performing complex cardiovascular interventions and complementing traditional learning methods. However, some critical considerations should be addressed. When using VR, cybersickness, a new and yet incompletely defined condition related to motion sickness, and accommodation problems have to be considered. Even though we did not include feedback mechanisms on such side effects within our study, none of the participants in the VR mentioned such side effects towards the treating physician by themselves or upon asking. Within the recently published state-of-the-art review paper, a difference in acceptance with regard to age and gender was reported to be present in clinical settings incorporating VR. The authors suggest the development of age and gender-neutral VR applications. Interestingly, within our cohort, the subgroup of younger females was more worried about the ablation procedure after watching the VR film as compared with males, where this difference was not observed in the control group. Next to the idea that comprehensive and detailed patient information, especially regarding risks of an invasive procedure, could cause adverse effects on patients’ anxiety, possible gender-specific effects of the overall VR setup could be considered. When using mHealth products, reinforcement of anxiety and gender or age-specific influences should therefore be considered. Limitations The current study was conducted within a small population. While the findings could have been more distinctly different, we believe this does not affect the main target of the study. We especially want to introduce a new way of providing information and show that it is useful and results in higher patient satisfaction and fewer worries. Further, the lack of standardization for oral patient information provision (as compared with written or VR) is well known and should be considered in a study like this, since an open conversation between doctor and patient has individual differences. Another limitation within our study is the missing questionnaires or feedback possibilities on possible side effects of VR glasses, such as motion sickness or eye strain, which were reported in prior work. However, none of the participants in the VR did report such side effects towards the treating physician explicitly. Finally, findings should be cautiously interpreted as this was a non-randomized study and may have limited generalizability.
The use of VR is increasingly embraced in patient education and may encourage the active involvement of patients in their treatment, and provide new dimensions for education provision. VR has the potential to be a positive intervention with regards to patient knowledge, satisfaction, and anxiety. VR lets patients virtually experience the hospital environment and offers the possibility to review and discuss this with family or relatives at home repeatedly, as also shown in the present study. VR glasses allow the creation of a real-life experience also for patients with reduced mobility, and previous studies showed good applicability also in elderly patients. This is in line with findings from our current study, where older patients (≥65 years) in the VR group reported higher patient satisfaction scores and felt less worried about the ablation procedure than those in the control group. Nevertheless, combined information and anxiety scores in the APAIS were comparable between the VR and the control group. As a comparison, previous studies demonstrated that video-based information provision did not increase patients’ anxiety towards a procedure significantly and that oral-, written- and video-based information can be considered equivalent to patient anxiety scores. Interestingly, within our cohort, it turned out that the subgroup of (younger) females was more worried about the ablation procedure after watching the VR film as compared with males, where this difference was not observed in the control group. This would line up with the idea that comprehensive and detailed patient information, especially regarding the risks of an invasive procedure, but also the confrontation and enlivening of an upcoming procedure can cause adverse effects on patients’ anxiety. These findings show that when using mobile health (mHealth) products, their possible side effects such as reinforcement of anxiety must also be considered. Further studies would be necessary to confirm these findings and, in particular, to investigate the possible influence of gender on side effects in more detail (within-patient differences). However, previous studies indicated that patients prefer detailed procedure-based information. The use of mHealth tools should be supported by a structured, integrated and patient-centred management pathway, as also widely described in current literature and guidelines. , This might allow close support of patients in case of important side effects. In addition, the subgroup of patients older than 65 years indicated less worried about the ablation procedure and were more likely to have a ≥ 80% satisfaction score based on procedure-based information provision after watching the VR film as compared to controls. Age should therefore certainly not be an exclusion criterion for using mHealth tools.
Within our study group, patients evaluated both in-hospital VR as well as disposable cardboard VR positively. Patients who used the disposable cardboard VR viewer showed high satisfaction scores and a majority of patients even kept using the disposable VR glasses at home after the procedure and showed it to friends or family at home. Home use of the disposable cardboard VR viewer initiated discussions in the private environment of patients. These discussions with relatives could have had an (positive) effect on attitudes and anxiety preceding AF ablation. However, further qualitative data would be required to state this. Another advantage of the disposable cardboard VR compared to the in-hospital standalone headset is its relatively low price. Additionally, patients can use it more flexibly and repetitively, or even for different VR programmes within the cardiology department. However, the implementation of VR in clinical AF care, as well as training for the use of in-hospital VR glasses might be time-consuming for healthcare providers. Time and personnel required for maintenance as well as costs for hygienic cleaning and wear and tear of the VR device must also be considered.
There was no difference between our study groups regarding the use of (telephone) contact moments between the patient and the hospital. However, no further data reported by caregivers with regard to expenses in time or personnel, as well as possible impact on the patient–provider relationship was collected within the scope of this study. A higher workload, especially when implementing new technology, might still be a point of discussion. Compared with oral face-to-face information provision, video-based information provision, however, was previously considered equivalent or requires even fewer physicians’ time. This might be important in an expected lack of medical and caregiver staff in the coming decades, where modern mHealth technologies might help in overcoming obstacles. The current ESC guidelines recommend providing AF care through a multidisciplinary AF team approach rather than by one single healthcare professional alone. Such an approach would allow important roles for nurses and allied health professionals in educating patients and coordinating AF care, while closely collaborating with a medical specialist. Therefore, the integration of a multidisciplinary team for the management of AF ablation patients might improve patient education provision and reduce professionals’ workload. Implementation of VR-based patient education could improve patient understanding of therapy, improve patient compliance and ameliorate shared decision-making. Therefore, the use of VR could enhance the patient–provider relationship in a real-world setting. This is an important aspect to consider as VR should support but not replace the real-world patient–provider relationship. Currently, mHealth tools and VR are finding their way into modern disease management, but the participation of healthcare providers in the early stages of mHealth and VR developments will be still necessary to ensure and guide patient-centred development. Furthermore, Goldberger et al. pointed out that video-based information provision may provide a more standardized and complete story than oral discussion. Another study discussed the advantage of video material in patient education for patients with low literacy skills and explained this by a less stressful environment for the patient when compared with in-hospital discussion with a doctor or a nurse. In addition, the TeleCheck-AF project showed that age is no barrier to remote patient information and education. Interactivity is also an important part of holistic patient education, which we aimed to provide in our intervention group by creating a 360° lifelike VR environment. This might make procedure-based information provision even more explanatory and make patients feel more engaged, as discussed in prior work. One obstacle in the broad implementation of VR-based patient information, however, might be funding. Further studies need to be conducted regarding cost-effectiveness. VR will become widely used in the teaching and training of the physicians, but our study also shows its possibilities in the telemedical management of complex diseases. In addition, VR may also play a role in remote digital patient care pathways such as the VIRTUAL-SAFARI project. However, there is still a lack of medical guidelines in this area that needs to be overcome.
VR has the potential for applications in direct patient treatment. Primary applications in cardiology are centred on education. VR can assist patients and health care providers in cardiovascular care, planning and performing complex cardiovascular interventions and complementing traditional learning methods. However, some critical considerations should be addressed. When using VR, cybersickness, a new and yet incompletely defined condition related to motion sickness, and accommodation problems have to be considered. Even though we did not include feedback mechanisms on such side effects within our study, none of the participants in the VR mentioned such side effects towards the treating physician by themselves or upon asking. Within the recently published state-of-the-art review paper, a difference in acceptance with regard to age and gender was reported to be present in clinical settings incorporating VR. The authors suggest the development of age and gender-neutral VR applications. Interestingly, within our cohort, the subgroup of younger females was more worried about the ablation procedure after watching the VR film as compared with males, where this difference was not observed in the control group. Next to the idea that comprehensive and detailed patient information, especially regarding risks of an invasive procedure, could cause adverse effects on patients’ anxiety, possible gender-specific effects of the overall VR setup could be considered. When using mHealth products, reinforcement of anxiety and gender or age-specific influences should therefore be considered.
The current study was conducted within a small population. While the findings could have been more distinctly different, we believe this does not affect the main target of the study. We especially want to introduce a new way of providing information and show that it is useful and results in higher patient satisfaction and fewer worries. Further, the lack of standardization for oral patient information provision (as compared with written or VR) is well known and should be considered in a study like this, since an open conversation between doctor and patient has individual differences. Another limitation within our study is the missing questionnaires or feedback possibilities on possible side effects of VR glasses, such as motion sickness or eye strain, which were reported in prior work. However, none of the participants in the VR did report such side effects towards the treating physician explicitly. Finally, findings should be cautiously interpreted as this was a non-randomized study and may have limited generalizability.
In patients scheduled for AF ablation, a lifelike 360° VR preprocedural educational video in addition to standard preprocedural information through oral counselling and information leaflets led to higher satisfaction, better information provision, procedure-related knowledge, and fewer worries regarding the procedure as compared with written and oral patient information. In addition, a disposable cardboard VR viewer is feasible as it is easy to use and amplifies conversations with relatives about the upcoming ablation procedure.
euac246_Supplementary_Data Click here for additional data file.
|
Rotor mechanism and its mapping in atrial fibrillation | a2fde0e9-5e16-406a-ba78-e80fa7defe3b | 10062333 | Physiology[mh] | Re-entry is central to the maintenance of atrial fibrillation (AF). Advanced mapping methods allow identifying a unique spiral-wave generator re-entry mechanism known as the rotor, which is a critical driver of AF in both experimental and clinical models. However, the properties of rotors and the effect of rotor ablation need to be defined further. We review the history and current status of basic and clinical science approaches to rotor identification and ablation. In 1913, Mines et al. first proposed a re-entrant excitation mechanism based on anatomical disorders. Then, Lewis et al. integrated such concepts into the hypothesis of ‘cyclic excitation caused by re-entry’ ( Figure ). The hypothesis states that the activity of the re-entrant circuit around the anatomical barrier may produce atrial flutter or cause fibrillation-like excitation depending on the different sizes of the re-entrant circuit and tissue refractory periods. In 1949, Scherf et al. found that a small dose of aconitine administered to the epicardium of the right atrial appendage in dogs induced atrial tachycardia or AF, with focal excitation as its underlying mechanism ( Figure ). In 1959, Moe et al. proposed the multi-wavelet theory and published the classical computer-model study , ( Figure ), in which an estimated 15–30 electrical wavelets were required for AF maintenance. This theory was validated in animal models of AF more than 20 years later, thanks to the advent of high-density mapping. Unlike Moe’s results, Allessie et al. reported that in an in vivo canine model, AF was maintained by only 4–6 wavelets, and Wang et al. showed that class IC anti-arrhythmic drugs suppressed AF by reducing the number of wavelets. After that, the multi-wavelet mechanism was widely accepted. However, with only 4–6 wavelets in an AF model, both atria are likely to recover from the previous excitation, resulting in wavelet fusion and AF termination. Therefore, what are the structural and electrophysiological bases for maintaining these wavelets? Is it possible that a high-frequency ‘mother rotor’ forms multiple wavelets after collision and fragmentation? In 1973, Allessie et al. (1973) documented functional re-entry independent of anatomical barriers in rabbit atria, which have the smallest-sized circuit, minimum re-excitation gap, and refractory centre. They developed the ‘leading circle hypothesis’ ( Figure ) and posited that the shorter the wavelength and the larger the atria, the more simultaneous re-entry circuits can be accommodated and that their stable existence would underlie AF maintenance. However, AF suppression was independent of wavelength reduction in experimental models and patients with AF. Subsequently, studies on functional re-entrant rhythm led to the emergence of the concept of spiral waves ( Figure ). The spiral wave was first discovered in 1990 in ex vivo sheep hearts, and spiral waves that moved rapidly in the ventricle were then shown to cause ventricular fibrillation. , However, stable rotation-like excitation was not detected using isochronous excitation mapping in the ex vivo sheep AF model. The introduction of more sophisticated signal analysis methods, including optical phase mapping, allowed to define and description spiral waves in greater detail, and in 1998, using optical phase mapping, Gray et al. recorded the rotor-like activity of spiral waves, which were triggered by heterogeneity in tissue conduction and refractory periods. Morphological characteristics of the rotor The rotor, as a ‘spiral wave generator,’ is a curved ‘vortex’ formed by spin motion in the two-dimensional (2D) plane. Morphologically, the spiral wave is connected by a curved wavefront (solid red line in Figure ) and wavetail (red dashed line in Figure ). The wavefront represents the depolarized region, which continuously conducts excitations outward; the wavetail represents the cardiomyocytes that have completed depolarization and are recovering to a resting state, and the area between the wavefront and the tail represents cardiomyocytes in the absolute refractory period. The connection point between the wavefront and the tail is the tip of the spiral wave (red point in Figure ), at which all excitation states converge, shifting the cardiomyocytes to a non-excitable state; therefore, this connection point is also called a phase singularity (PS). During the spin motion of the spiral wave around the PS, the PS can meander to form the corresponding trajectory, and the region enclosed by this trajectory is the core of the spiral wave (blue circle in Figure ). , At the PS, the shorter action potential duration and slower conduction velocity enable re-entry near the core, where the wavefront/tail meets and the excitable gap diminishes. With the development of optical mapping technology, some studies hypothesized that the atrial re-entries recorded optically from the surfaces corresponded to three-dimensional (3D) scroll waves spanning the thickness of the wall ( Figure ). , Such a scroll wave’s behaviour is embodied in its organizing axis or filament (green line in Figure ), a largely quiescent tube about which the scroll rotates. The filament is not always ‘I-shaped’ and extends between the two surfaces; its tension and stability are determined by heterogeneous myocardial thickness, stretch, and remodelling (i.e. ionic and anatomic remodelling). The mathematical approach is to calculate the filament shape as a ‘minimal path,’ i.e. the scroll wave follows the ‘least resistance’ rule that yields the prediction for stationary filaments on a purely geometrical basis ( Figure ). In vitro AF model, scroll waves with a bent L-shaped or U-shaped filament were observed by combining endo-epicardial optical mapping (Figure and ). Recently, Allessie and colleagues first proposed the presence of endocardial–epicardial dissociation (EED). After that, direct evidence of EED by simultaneous endo-epicardial mapping showed that EED was characterized by significant temporal heterogeneity, transitioning of preferential activation between the myocardial layers, and transmural conduction. With non-identical surface activation patterns on the endocardial and epicardial surfaces, the activity of scroll waves in the 3D myocardial wall thickness was hypothesized to conform to either ectopic discharge or scroll waves. Also, dyssynchronous activations on the endo- and epicardial surfaces further increased AF complexity, which may make endocardial mapping and ablation insufficient to address the AF mechanism. Electrophysiological characteristics of the rotor In high-density optical mapping and computer mathematical models, , the activity of rotors is characterized by wavebreak, meandering, and variable wavelength. (i) Wavebreak develops when the rotor encounters an anatomical or functional barrier and splits into two or more daughter rotors, which rotate in opposing spirals around two new PSs. Most of these daughter rotors are unstable and will dissipate by colliding with each other or refractory tissue; in contrast, relatively stable daughter rotors can continue to generate new daughter rotors. Atrial fibrillation is maintained when the rate of wavebreak is greater than or equal to the rate of rotor extinction. This phenomenon also partly explains the difficulty in the self-termination of persistent AF. (ii) Meandering is when the PS of a rotor is mobile with a usually uncertain movement range and direction. Heterogeneity of cardiac ion channels influences the meandering trajectory. Calvo et al. found that the rotor meanders towards areas with lower rectifying potassium current (IK1) distribution, lower regional myocardial excitability, and more extended refractory periods. In some optical mapping experiments, the rotor was usually anchored in a region with significant heterogeneity in myocardial thickness and myocardial fibre alignment. , Thus, rotor meandering is closely related to the electrical and anatomical heterogeneity of the myocardium (3). Rotor wavelength is variable, defined as the distance from the wavefront to the tail. Under different ion channel conditions, the rotor’s wavelength, core area, and excitable gap will change according to ion channel conditions. Nattel et al. summarized the effects of ion channels on the electrophysiological characteristics of the rotor: a reduction in sodium current leads to a reduction in rotor propagation and rotation speed, a decrease in wavefront curvature (WC), an increase in the core area, and an expansion of the meandering range. An increase in IK1 leads to a reduction in rotor wavelength, core area, and meandering range, while a decrease in outward I K1 leads to an increase in rotor wavelength, core area, and meandering range. Therefore, different ion channel states have different effects on the rotor wavelength and electrophysiological characteristics, which underlie the considerable complexity and uncertainty of rotor activity, rendering clinical mapping and identification of rotors more challenging. Generation mechanism of the rotor In 1978, Krinsky first hypothesized the process of spiral wave formation in a heterogeneous moiety as follows ( Figure ): in the myocardium with a highly heterogeneous refractory period, a pair of activation waves randomly propagates into the 2D myocardial tissue, with the coupling interval of these two waves being infinitely close to the refractory period of the region with higher refractoriness. Under these conditions, refractoriness heterogeneities result in wavebreaks, and the edges of wavebreaks evolve into a spiral wave. However, this mathematical model has specific requirements for the heterogeneity of both tissue refractory period and coupling interval of the two pulses, limiting its reproduction in animal models of AF. In 1996, Cabo et al. proposed the ‘vortex shedding’ theory of spiral wave initiation ( Figure ): in cardiac tissue, partial blockade of the membrane sodium channels or high-frequency (HF) excitation may result in an unexcitable obstacle with sharp edges, which may destabilize the propagation of electrical excitation waves, causing the formation of self-sustained vortices and turbulent cardiac electrical activity. The formation of such vortices, which visually resembles vortex shedding in hydrodynamic turbulent flows, is a potential mechanism leading to the spontaneous initiation of uncontrolled high-frequency excitation of the heart. This theory provides the experimental and theoretical bases for forming spiral waves in pathological conditions (e.g. post-myocardial infarction ventricular myocardium and atrial myocardium with persistent AF), which combine severe fibrosis and reduced Na + current density and/or excitability. Notably, in studies by Krinsky and Cabo et al., the WC would increase when the wavefront reached the edge of the refractory period region/obstacle. The latter phenomenon has been explained using the ‘source-sink relationship’ theory ( Figure ), in which, during excitation conduction, the wavefront represents the depolarized region that is defined as the ‘source.’ In contrast, the cardiomyocytes in the resting state in front of the wavefront are defined as the ‘sink’. When the excitation conducts in a narrow region, the ‘source’ and the ‘sink’ activated in front of the wavefront are in a high matching state, and its WC is relatively small. When the wavefront propagates to the adjacent larger region, the ‘sink’ in front of the wavefront suddenly increases. At the same time, the ‘source’ does not change significantly. At this time, the matching level of the ‘source’ and ‘sink’ decreases, and the WC increases in order to facilitate excitation forward propagation. Administration of Na + channel blockers may decrease myocardial excitability and reduce the ‘source,’ which increases the source-sink-mismatch, further facilitating the formation of ‘vortex shedding.’ In addition, the morphology of spiral waves suggests that the ‘source-sink’ away from the core has a higher matching level and a small WC (wavefront at ‘a’ in Figure ); in contrast, the wavefront closer to the core has a lower ‘source-sink’ matching relationship and a larger WC (wavefront at ‘b’ in Figure ), i.e. a ‘source-sink’ mismatch. When this ‘source-sink’ mismatch in the wavefront reaches an extreme, the ‘source’ is insufficient to activate the ‘sink’ in front of it, eventually forming the PS. Therefore, a ‘source-sink’ mismatch has also been considered a critical mechanism for spiral wave formation. The rotor, as a ‘spiral wave generator,’ is a curved ‘vortex’ formed by spin motion in the two-dimensional (2D) plane. Morphologically, the spiral wave is connected by a curved wavefront (solid red line in Figure ) and wavetail (red dashed line in Figure ). The wavefront represents the depolarized region, which continuously conducts excitations outward; the wavetail represents the cardiomyocytes that have completed depolarization and are recovering to a resting state, and the area between the wavefront and the tail represents cardiomyocytes in the absolute refractory period. The connection point between the wavefront and the tail is the tip of the spiral wave (red point in Figure ), at which all excitation states converge, shifting the cardiomyocytes to a non-excitable state; therefore, this connection point is also called a phase singularity (PS). During the spin motion of the spiral wave around the PS, the PS can meander to form the corresponding trajectory, and the region enclosed by this trajectory is the core of the spiral wave (blue circle in Figure ). , At the PS, the shorter action potential duration and slower conduction velocity enable re-entry near the core, where the wavefront/tail meets and the excitable gap diminishes. With the development of optical mapping technology, some studies hypothesized that the atrial re-entries recorded optically from the surfaces corresponded to three-dimensional (3D) scroll waves spanning the thickness of the wall ( Figure ). , Such a scroll wave’s behaviour is embodied in its organizing axis or filament (green line in Figure ), a largely quiescent tube about which the scroll rotates. The filament is not always ‘I-shaped’ and extends between the two surfaces; its tension and stability are determined by heterogeneous myocardial thickness, stretch, and remodelling (i.e. ionic and anatomic remodelling). The mathematical approach is to calculate the filament shape as a ‘minimal path,’ i.e. the scroll wave follows the ‘least resistance’ rule that yields the prediction for stationary filaments on a purely geometrical basis ( Figure ). In vitro AF model, scroll waves with a bent L-shaped or U-shaped filament were observed by combining endo-epicardial optical mapping (Figure and ). Recently, Allessie and colleagues first proposed the presence of endocardial–epicardial dissociation (EED). After that, direct evidence of EED by simultaneous endo-epicardial mapping showed that EED was characterized by significant temporal heterogeneity, transitioning of preferential activation between the myocardial layers, and transmural conduction. With non-identical surface activation patterns on the endocardial and epicardial surfaces, the activity of scroll waves in the 3D myocardial wall thickness was hypothesized to conform to either ectopic discharge or scroll waves. Also, dyssynchronous activations on the endo- and epicardial surfaces further increased AF complexity, which may make endocardial mapping and ablation insufficient to address the AF mechanism. In high-density optical mapping and computer mathematical models, , the activity of rotors is characterized by wavebreak, meandering, and variable wavelength. (i) Wavebreak develops when the rotor encounters an anatomical or functional barrier and splits into two or more daughter rotors, which rotate in opposing spirals around two new PSs. Most of these daughter rotors are unstable and will dissipate by colliding with each other or refractory tissue; in contrast, relatively stable daughter rotors can continue to generate new daughter rotors. Atrial fibrillation is maintained when the rate of wavebreak is greater than or equal to the rate of rotor extinction. This phenomenon also partly explains the difficulty in the self-termination of persistent AF. (ii) Meandering is when the PS of a rotor is mobile with a usually uncertain movement range and direction. Heterogeneity of cardiac ion channels influences the meandering trajectory. Calvo et al. found that the rotor meanders towards areas with lower rectifying potassium current (IK1) distribution, lower regional myocardial excitability, and more extended refractory periods. In some optical mapping experiments, the rotor was usually anchored in a region with significant heterogeneity in myocardial thickness and myocardial fibre alignment. , Thus, rotor meandering is closely related to the electrical and anatomical heterogeneity of the myocardium (3). Rotor wavelength is variable, defined as the distance from the wavefront to the tail. Under different ion channel conditions, the rotor’s wavelength, core area, and excitable gap will change according to ion channel conditions. Nattel et al. summarized the effects of ion channels on the electrophysiological characteristics of the rotor: a reduction in sodium current leads to a reduction in rotor propagation and rotation speed, a decrease in wavefront curvature (WC), an increase in the core area, and an expansion of the meandering range. An increase in IK1 leads to a reduction in rotor wavelength, core area, and meandering range, while a decrease in outward I K1 leads to an increase in rotor wavelength, core area, and meandering range. Therefore, different ion channel states have different effects on the rotor wavelength and electrophysiological characteristics, which underlie the considerable complexity and uncertainty of rotor activity, rendering clinical mapping and identification of rotors more challenging. In 1978, Krinsky first hypothesized the process of spiral wave formation in a heterogeneous moiety as follows ( Figure ): in the myocardium with a highly heterogeneous refractory period, a pair of activation waves randomly propagates into the 2D myocardial tissue, with the coupling interval of these two waves being infinitely close to the refractory period of the region with higher refractoriness. Under these conditions, refractoriness heterogeneities result in wavebreaks, and the edges of wavebreaks evolve into a spiral wave. However, this mathematical model has specific requirements for the heterogeneity of both tissue refractory period and coupling interval of the two pulses, limiting its reproduction in animal models of AF. In 1996, Cabo et al. proposed the ‘vortex shedding’ theory of spiral wave initiation ( Figure ): in cardiac tissue, partial blockade of the membrane sodium channels or high-frequency (HF) excitation may result in an unexcitable obstacle with sharp edges, which may destabilize the propagation of electrical excitation waves, causing the formation of self-sustained vortices and turbulent cardiac electrical activity. The formation of such vortices, which visually resembles vortex shedding in hydrodynamic turbulent flows, is a potential mechanism leading to the spontaneous initiation of uncontrolled high-frequency excitation of the heart. This theory provides the experimental and theoretical bases for forming spiral waves in pathological conditions (e.g. post-myocardial infarction ventricular myocardium and atrial myocardium with persistent AF), which combine severe fibrosis and reduced Na + current density and/or excitability. Notably, in studies by Krinsky and Cabo et al., the WC would increase when the wavefront reached the edge of the refractory period region/obstacle. The latter phenomenon has been explained using the ‘source-sink relationship’ theory ( Figure ), in which, during excitation conduction, the wavefront represents the depolarized region that is defined as the ‘source.’ In contrast, the cardiomyocytes in the resting state in front of the wavefront are defined as the ‘sink’. When the excitation conducts in a narrow region, the ‘source’ and the ‘sink’ activated in front of the wavefront are in a high matching state, and its WC is relatively small. When the wavefront propagates to the adjacent larger region, the ‘sink’ in front of the wavefront suddenly increases. At the same time, the ‘source’ does not change significantly. At this time, the matching level of the ‘source’ and ‘sink’ decreases, and the WC increases in order to facilitate excitation forward propagation. Administration of Na + channel blockers may decrease myocardial excitability and reduce the ‘source,’ which increases the source-sink-mismatch, further facilitating the formation of ‘vortex shedding.’ In addition, the morphology of spiral waves suggests that the ‘source-sink’ away from the core has a higher matching level and a small WC (wavefront at ‘a’ in Figure ); in contrast, the wavefront closer to the core has a lower ‘source-sink’ matching relationship and a larger WC (wavefront at ‘b’ in Figure ), i.e. a ‘source-sink’ mismatch. When this ‘source-sink’ mismatch in the wavefront reaches an extreme, the ‘source’ is insufficient to activate the ‘sink’ in front of it, eventually forming the PS. Therefore, a ‘source-sink’ mismatch has also been considered a critical mechanism for spiral wave formation. Principle of optical mapping technology The optical mapping technology was first used to record action potentials in nerve cells in 1973. In 1976, Salama et al. first recorded the electrical activity of cardiomyocytes using optical mapping technology. Later in 1990, Davidenko et al. detected the spiral wave and demonstrated that it could induce ventricular tachycardia using optical mapping in animal hearts. The principle of optical mapping technology is based on wavelength-dependent light–tissue interactions, including photon scattering, absorption, reflection, and fluorescence effects. Optical mapping technology mainly uses a voltage-sensitive fluorescent dye as a marker, which allows the transmembrane potential change of cells to be manifested as a change in fluorescent substance brightness that an optical detection device can record. Voltage-sensitive fluorescent dyes are usually calcium chelators. After cardiomyocytes are marked with fluorescent dyes, intra-cellular calcium ions are combined with fluorescent dyes to allow immediate detection of intracellular calcium ion concentration changes by optical detection equipment, thus reflecting the process of cardiomyocyte depolarization and repolarization. Computer processing will enable us to understand the depolarization and repolarization of different regions of myocardial tissue and more intuitively perceive the electrical activity process of myocardial tissue over time, thereby rendering optical mapping techniques suitable for dynamic observation of the rotor. An example of optical mapping in a canine AF model from our previous research is presented in Figure . Application of optical mapping in atrial fibrillation models In 1995, Gray et al. first recorded the activity of rotors on a ventricular fibrillation model using optical mapping. After that, this research group found rotors in an isolated sheep heart model for AF and indicated that rotors were formed based on the heterogeneity and refractory periods of tissue conduction, with temporal and spatial periodicity. Several years later, other researchers using a combination of endo-epicardial optical mapping reported that the patterns of rotor activation on the endocardial and epicardial myocardium surface could be identical or otherwise in the normal sheep heart. , Various non-identical endo-epicardial activation patterns, such as multiple centrifugal breakthrough activations, wavebreaks, and short-lived re-entries, suggest the existence of transmural propagation of rotational activity throughout the 3D thickness myocardium in association with increased stability of AF. In 2015, Hansen and colleagues used optical mapping on both the endocardium and the epicardium of human hearts and demonstrated a delay in action potential during transmural conduction, and that the rotors mapped in the endo- or epicardium with multiple expression forms, such as a spiral-wave like a re-entry, stable foci or breakthrough, or spatially unstable breakthrough. The appearance of these activation forms depends on a variety of factors: (i) the spiral shape of the 3D scroll wave: I-shaped, L-shaped, or U-shaped, and the complexity of its shape largely depends on the thickness of the local myocardium; (ii) the conduction characteristics of the local myocardium, which determines the number of breakthroughs on the surface of the myocardium. Because the rotor has a complex spatiotemporal structure, attention is called to the different manifestations of rotors when mapping the endocardium. The optical mapping technology was first used to record action potentials in nerve cells in 1973. In 1976, Salama et al. first recorded the electrical activity of cardiomyocytes using optical mapping technology. Later in 1990, Davidenko et al. detected the spiral wave and demonstrated that it could induce ventricular tachycardia using optical mapping in animal hearts. The principle of optical mapping technology is based on wavelength-dependent light–tissue interactions, including photon scattering, absorption, reflection, and fluorescence effects. Optical mapping technology mainly uses a voltage-sensitive fluorescent dye as a marker, which allows the transmembrane potential change of cells to be manifested as a change in fluorescent substance brightness that an optical detection device can record. Voltage-sensitive fluorescent dyes are usually calcium chelators. After cardiomyocytes are marked with fluorescent dyes, intra-cellular calcium ions are combined with fluorescent dyes to allow immediate detection of intracellular calcium ion concentration changes by optical detection equipment, thus reflecting the process of cardiomyocyte depolarization and repolarization. Computer processing will enable us to understand the depolarization and repolarization of different regions of myocardial tissue and more intuitively perceive the electrical activity process of myocardial tissue over time, thereby rendering optical mapping techniques suitable for dynamic observation of the rotor. An example of optical mapping in a canine AF model from our previous research is presented in Figure . In 1995, Gray et al. first recorded the activity of rotors on a ventricular fibrillation model using optical mapping. After that, this research group found rotors in an isolated sheep heart model for AF and indicated that rotors were formed based on the heterogeneity and refractory periods of tissue conduction, with temporal and spatial periodicity. Several years later, other researchers using a combination of endo-epicardial optical mapping reported that the patterns of rotor activation on the endocardial and epicardial myocardium surface could be identical or otherwise in the normal sheep heart. , Various non-identical endo-epicardial activation patterns, such as multiple centrifugal breakthrough activations, wavebreaks, and short-lived re-entries, suggest the existence of transmural propagation of rotational activity throughout the 3D thickness myocardium in association with increased stability of AF. In 2015, Hansen and colleagues used optical mapping on both the endocardium and the epicardium of human hearts and demonstrated a delay in action potential during transmural conduction, and that the rotors mapped in the endo- or epicardium with multiple expression forms, such as a spiral-wave like a re-entry, stable foci or breakthrough, or spatially unstable breakthrough. The appearance of these activation forms depends on a variety of factors: (i) the spiral shape of the 3D scroll wave: I-shaped, L-shaped, or U-shaped, and the complexity of its shape largely depends on the thickness of the local myocardium; (ii) the conduction characteristics of the local myocardium, which determines the number of breakthroughs on the surface of the myocardium. Because the rotor has a complex spatiotemporal structure, attention is called to the different manifestations of rotors when mapping the endocardium. Panoramic mapping Intracardiac panoramic mapping The focal impulse and rotor modulation (FIRM) methodology was first used clinically in the CONFIRM study, published in 2012 by Narayan et al. The basket-like 64 electrodes would simultaneously contact the endocardium of the atria and record potentials, and the proprietary computer software was used to analyse the electrical signals collected by the basket electrodes to guide the ablation. In the study, rotors were mapped in 97% of the study population, with an average of 2.1 ± 1.0 rotors per patient, and rotor ablation terminated AF or prolonged AF cycle length in 86% of the patients, and 82.4% of the patients were in sinus rhythm during a mean follow-up of 273 days. These favourable outcomes, however, were not reproduced at many other centres . In a meta-analysis, circumferential pulmonary vein isolation, combined with a rotor ablation strategy guided by a basket catheter, was not found superior to pulmonary vein isolation alone. The flaws seen with inter-spline bunching resulted in a loss of coverage and contact, and only 63.1% of the inter-electrode distances were less than the most stringent spatial resolution required for the identification of rotors in human AF. Moreover, a computer simulation study found that several high-density mapping catheters (AFocusII and PentaRay) had inter-electrode spacings below minimum resolution (11.9 mm), suggesting that these catheters have a higher resolution to locate PSs if placed over the rotor core accurately. Although these catheters have satisfactory mapping resolution, they are not panoramic mapping devices that could show overall atrial electrical activity. Therefore, both panoramic mapping catheters and local high-density mapping catheters have advantages and disadvantages, and there are no comparative clinical studies to determine which would provide superior performance. Recently, electrographic flow (EGF) mapping (Ablamap® software; Ablacon, Wheat Ridge, CO) emerged as a novel method to identify AF driver by estimating atrial cardiac action potential flow. An algorithm that combined Green’s minimal bending energy algorithm and Horn–Schunck Flow algorithm was used to analysed the time-domain information from the unipolar electrograms collected by a basket-like catheter and converted it into the space domain of flow vectors for the identification of singularities where the flow vector angles around a point covered 360°. One strength of this method is the potential to distinguish active and passive rotors. Theoretically, it can make up for insufficient local resolution of the basket-like catheter. The ongoing FLOW-AF study (NCT 04473963) will verify the efficacy of driver ablation–guided EGF mapping. In 2020, the real-time electrogram analysis for the driver (RADAR) system was introduced to guide driver ablation in a prospective, multi-centre study. The RADAR system was used in conjunction with a standard mapping system. High-density contact mapping was performed using a coronary sinus catheter as a reference to record electrograms in various anatomic locations. These electrograms were then sorted and stitched together to create a panoramic 3D conduction vector map for each coronary sinus phase. Sites of rotational activity and focal impulses were identified in the individual conduction vector maps. The initial outcome of using the RADAR system was promising, with 55% of patients experiencing AF termination after driver ablation and 82% remaining AF-free during follow-up. Non-invasive electrocardiographic imaging Recently, the non-invasive electrocardiographic imaging (ECGi) technique was used to map rotors . Using an external cardiac 3D mapping system (ECVUE system, Medtronic), patients with AF wore a vest with 252 electrodes and underwent CT scanning to obtain the relationship between the 3D geometry of both atria and the positions of 252 electrodes on the vest. Atrial activation during AF was obtained by phase mapping of the recorded ECG. Active driving and passive propagation zones were identified by analysing all accumulated images for each patient. A driver is defined as a focal breakthrough, when the image presents a centrifugal activation from a point or zone, or a spiral wave, when the wave is rotated around a centre on phase progression. In this study, the rate of AF termination during rotor ablation was 70%, and 85% of patients were free from AF recurrence at a 1-year follow-up. The latter AFACART study and the TARGET-AF1 trial also observed favourable outcomes of non-invasive ECGi-guided driver ablation for PerAF. However, the system has some limitations. First, epicardial far-field unipolar signal mapping results in poor stability of atrial signal quality and makes it challenging to distinguish micro-re-entrant from focal breakthrough; second, the activation data are a composite of endocardial, epicardial, and intramural patterns and interactions, and therefore, cannot differentiate the signals from overlapping cardiac anatomy sites, such as coronary sinus and inter-atrial septum. Third, phase mapping tends to introduce false rotors during complex activation patterns. , The conduction of two waves in opposite directions on both sides of a conduction block line may be misinterpreted as a rotor in phase analysis. Hence, although this system is a panoramic high-resolution mapping technology, it has a non-negligible false-positive rate in AF rotor mapping due to the inherent deficiency of external signals. Non-contact charge density mapping The charge density (CD) mapping is performed by a combined imaging and multi-electrode mapping system (AcQMap, Acutus Medical, Carlsbad, CA), which has 48 ultrasound transducers to reconstruct atrial chamber anatomy and 48 low-impedance, high-fidelity electrodes for recording biopotential signals. In brief, an inverse algorithm, based on the principles of electrostatic field theory, obtains the global distribution of CD sources across the endocardial surface generated by ultrasonic imaging and then creates global maps of cardiac activation displayed as a spatiotemporal window. The UNCOVER AF Trial reported a 72.5% success rate of single ablation for PerAF guided by the AcQMap system. The key advantages of CD mapping include (i) global and continuous mapping of AF with a reduction in far-field interference; (ii) enabling the identification of both re-entrant and focal activation; (iii) rapid (<5 min) and non-contact reconstruction of atrial anatomy leading to avoidance of some problems encountered by the impedance-based approach. The major limitations of this mapping method are the inability to study epicardial or transmural mechanisms and the inability to properly and successfully identify the so-called rotor activity. Local high-density mapping Dominant frequency mapping In 2005, Haïssaguerre’s group performed dominant frequency (DF) mapping by using spectral analysis to construct hierarchical gradients of activation frequencies at different atrial regions. The DF is calculated from the cycle length of local electrograms, which correlate with the local AF frequency, i.e. a faster local AF waveform frequency is associated with a higher DF. In theory, localized sites with the highest frequency activity are markers of critical AF drivers or rotors. Ablation of these high DF sites achieved AF termination and prolongation of the AF cycle length. However, the latter radiofrequency ablation of drivers of atrial fibrillation study failed to demonstrate the superiority of localized high-frequency source ablation over PVI in either patient with paroxysmal AF or persistent AF. The technical limitations of DF mapping may account for these results. The local potentials become more complex as AF progresses; for example, with local fragmented potentials or double potentials, the value obtained from this DF calculation may be faster than that of real drivers or rotors, precluding accurate DF measurement. Basic research has also confirmed that the DF of fragmented potentials generated by the collision of wavefronts around rotors is consistent with that of the rotor core region. Therefore, these limitations of DF mapping cannot be ignored, and this method can be used only as a reference for rotor potential analysis. In 2017, the CARTO-Finder, a novel technique module of the CARTO mapping system (Biosense Webster), enabled a dynamic evaluation of rotational and focal activation during AF and was used to identify areas of rotor domains. , Ablation of rotor domains effectively eliminated frequency gradients, and 70% of patients with long-standing persistent AF were free of AF recurrence at 12 months. However, only 13 patients were included in this study, and the efficacy of using CARTO-Finder to target AF drivers will be further confirmed by the ongoing larger multi-centre study (ClinicalTrials.gov Identifier: NCT03064451). Phase similarity mapping In 2013, Lin and colleagues demonstrated that regions of AF drivers were characterized by regular, organized, and rapid repetitive activities with a high degree of similarity in bipolar electrogram configuration. After that, in their single-centre, randomized study, a novel similarity index (SI) was calculated by using phase analysis of the bipolar signal acquired by the AFocus II catheter to identify potential AF drivers for ablation. The obtained electrogram signal underwent phase calculation simulation through a computer system to filter out interference signals and far-field contamination, theoretically improving the accuracy of rotor identification. An SI vector field was then constructed between a pair of the nearest electrodes to analyse the electrical wave propagation around AF drivers based on curvature and divergence forces, and therefore, can distinguish a rotor from a focal source or random fractionated. In this study, an average of 2.6 ± 0.89 high SI regions were mapped per patient, and the rate of AF termination was superior (68% vs. 27%) to that of the control group with a fractionated potential ablation strategy, with a high success rate of 83% in long-term follow-up. Dispersion mapping In 2017, Seitz et al. used the PentaRay catheter to identify the region with rotor activity in AF. This mapping is based on the theoretical premise that when a PentaRay catheter is placed at the centre of a rotor, its electrode branches will record the sequential excitation phases of rotor wavefronts, i.e. the asynchrony of the bipolar potential or ‘dispersion electrograms.’ However, the study methodology did not exclude fragmented potentials in the dispersion analysis and the area with both fragmented and non-fragmented potentials as an ablation target. Despite the excellent termination rate of AF, extensive ablation is performed on a wide area of atria (15.8–29% of LA surface area), and the unintentional creation of an unnecessary pro-arrhythmic scar cannot be ignored. Recently, the results from our study , showed that HF and dispersion electrogram-guided rotor ablation abruptly terminated AF, substantially improving long-term AF elimination. Notably, it showed a decreased ablation area (4.3–13.3% of LA surface area) and suggested that most fragmented potentials are the by-products of re-entrant rotors that break down at their boundaries. Therefore, local high-density mapping provides more detailed information to distinguish the rotor from complicated AF potentials despite not providing a panoramic imaging of the whole atrium. Stochastic trajectory analysis of ranked signals mapping Stochastic trajectory analysis of ranked signals (STAR) is a novel mapping method that compares activation times across electrode pairs to identify atrial regions with earlier activation than neighbouring areas. The author defined region as an AF driver if it leads for more than 75% of wavefronts during a recording period. In 2020, Honarbakhsh et al. confirmed the efficacy of STAR-guided ablation in a single-centre, prospective study, and the success rate was 81.5%. The STAR mapping method has some potential advantages: it allows both global and sequential mapping and is compatible with different commercially available multi-polar catheters (CARTO, Rhythmia, and EnSite). However, this method cannot distinguish the driver mechanism between rotational and focal. Furthermore, the ablation of early activation areas depended on the operator’s interpretation. The limited coverage and contact of basket catheters and the position and orientation of the catheter may also influence the mapping accuracy. Focal source and trigger computational algorithm In 2020, Chauhan et al. reported their proprietary focal source and trigger (FaST) computational algorithm, which allowed automatically identifying sustained periodic bipolar and unipolar QS electrogram morphology during AF. Both patients with paroxysmal and persistent AF were enrolled in this validation study. PVI plus FaST site ablation decreased AF recurrence compared with PVI alone, but the result did not reach statistical significance ( P = 0.064). One advantage of the FaST mapping is time efficiency: the mean ablation time of FaST sites outside the PV was only 8.5 ± 5.1 min. Notwithstanding, the FaST mapping was not performed comprehensively in this study and may miss some potential focal sources in the right atrium. Furthermore, FaST signals may arise from mechanisms other than focal sources, such as endo-epicardial breakthroughs. Repetitive-regular activities mapping Recently, Pappone and colleagues introduced a novel real-time integrated mapping technique to identify potential driver regions characterized by repetitive-regular activities (RRas). This mapping method consists of three components: a regular cycle length map, a fragmentation map, and a peak-to-peak voltage measurement. The software integrated the three mapping results and allowed an automated evaluation of multiple mechanisms during AF by identifying regions exhibiting RRas, fragmentation, the conduction velocity of consistent wavefronts, and electrically silent areas. Therefore, it provided a patient-tailored strategy. In this study, a modified PVI, plus RRa site ablation, achieved a higher rate of AF termination (61% vs. 30%) and less AF recurrence (73% vs. 50%) compared with a modified PVI alone. One drawback of this approach is that it cannot distinguish the specific electrophysiological characteristic of the repetitive-regular electrograms, such as rotors, focal sources, or micro re-entries. Driver ablation vs. conventional approach: pooled analysis from clinical trials At present, the overall AF termination rate of rotor ablation is 44.5 ± 15.5% ( Figure ), the rate of freedom from AF/AT is 57.7 ± 7.7%, and the rate of freedom from AF is 69.5 ± 4.6% ( Figure ). , presents clinical studies of different currently available methods for rotor mapping. Although the termination and success rates were different among various mapping methods, the overall mean values were similar. Considerable discrepancy in the rate of successful ablation and AF termination was observed among studies of FIRM-guided approach, which is also the method with the largest number of published studies. A study reported that the AF termination rate and long-term success rate were only 5% and 21% by FIRM-guided ablation. The reasons for the difference in results are the defects of the mapping technology itself and the limitations of the study, such as the type of AF included, the follow-up time, and the sample size. Thus, outcomes of single-arm studies were significantly limited by high heterogeneity. However, the number of case-control clinical studies of traditional ablation strategies is relatively small, and we conducted a meta-analysis of 10 clinical trials with published case-control studies ( Figure ). The significant pooled OR for freedom from AF/AT in these 10 studies using the fixed-effects model was 0.53 [CI, 0.40–0.69 ( P = 0.037); I 2 = 42% ( P < 0.0001)]. Among the ten studies, only Tilz’s findings did not support driver ablation, and the study included paroxysmal AF and performed FIRM-guided ablation alone without PVI. Two other studies also compared AF driver-only ablation to PVI. , Although Seitz’s conclusions favour driver ablation, the incidence of atrial tachycardia is as high as about 30%. If these three studies are excluded, freedom from AF/AT produced an OR of 0.45 (CI, 0.32–0.62; P < 0.0001), with minimal heterogeneity between studies (I 2 = 0%). It supports the possible benefit of driver ablation as an additional strategy in improving freedom from all arrhythmias compared with conventional ablation alone. The controversary of driver studies Although the overall results of the meta-analysis favoured driver ablation, we cannot ignore the inconsistency of clinical findings. Because there are several factors affecting the outcome of AF ablation, only in terms of mapping methods and ablation strategies, first of all, there are differences in the false-positive rate of different mapping devices, and the success rate is naturally different. Furthermore, the area of ablation lesions and different methods of ablation (linear or patch lesion) can also affect the incidence of iatrogenic arrhythmias. However, for the acute outcome of ablation, driver ablation achieved a higher AF termination rate than conventional ablation, , , , , suggesting an intervention in the maintenance mechanism of AF. Previous studies have also confirmed a favourable long-term outcome in patients with procedural AF termination. , Secondly, the results of studies may also be impacted by sample size and different control groups. Currently, there is still a lack of large-sample randomized controlled studies to directly compared driver ablation and conventional ablation. Previous studies set different ablation strategies as control groups, including PVI, , PVI + linear ablation, CAFE ablation, or ‘stepwise’ strategy. , It was therefore difficult to compare studies with one another. Moreover, there is also a lack of evidence comparing driver ablation with new strategies that have demonstrated satisfactory clinical outcomes, such as posterior BOX isolation or vein of Marshall ethanol infusion. In addition, most clinical studies have been conducted in ablation-naive patients, , , , and a part of these patients may be benefited by PVI alone. In the study of Seitz et al., PVI was not performed, and regions of AF termination were frequently located in the PV antrum. Therefore, enrichment of the study population for patients having extra-PV drivers would likely increase the success rate of adjuvant ablation approaches. Intracardiac panoramic mapping The focal impulse and rotor modulation (FIRM) methodology was first used clinically in the CONFIRM study, published in 2012 by Narayan et al. The basket-like 64 electrodes would simultaneously contact the endocardium of the atria and record potentials, and the proprietary computer software was used to analyse the electrical signals collected by the basket electrodes to guide the ablation. In the study, rotors were mapped in 97% of the study population, with an average of 2.1 ± 1.0 rotors per patient, and rotor ablation terminated AF or prolonged AF cycle length in 86% of the patients, and 82.4% of the patients were in sinus rhythm during a mean follow-up of 273 days. These favourable outcomes, however, were not reproduced at many other centres . In a meta-analysis, circumferential pulmonary vein isolation, combined with a rotor ablation strategy guided by a basket catheter, was not found superior to pulmonary vein isolation alone. The flaws seen with inter-spline bunching resulted in a loss of coverage and contact, and only 63.1% of the inter-electrode distances were less than the most stringent spatial resolution required for the identification of rotors in human AF. Moreover, a computer simulation study found that several high-density mapping catheters (AFocusII and PentaRay) had inter-electrode spacings below minimum resolution (11.9 mm), suggesting that these catheters have a higher resolution to locate PSs if placed over the rotor core accurately. Although these catheters have satisfactory mapping resolution, they are not panoramic mapping devices that could show overall atrial electrical activity. Therefore, both panoramic mapping catheters and local high-density mapping catheters have advantages and disadvantages, and there are no comparative clinical studies to determine which would provide superior performance. Recently, electrographic flow (EGF) mapping (Ablamap® software; Ablacon, Wheat Ridge, CO) emerged as a novel method to identify AF driver by estimating atrial cardiac action potential flow. An algorithm that combined Green’s minimal bending energy algorithm and Horn–Schunck Flow algorithm was used to analysed the time-domain information from the unipolar electrograms collected by a basket-like catheter and converted it into the space domain of flow vectors for the identification of singularities where the flow vector angles around a point covered 360°. One strength of this method is the potential to distinguish active and passive rotors. Theoretically, it can make up for insufficient local resolution of the basket-like catheter. The ongoing FLOW-AF study (NCT 04473963) will verify the efficacy of driver ablation–guided EGF mapping. In 2020, the real-time electrogram analysis for the driver (RADAR) system was introduced to guide driver ablation in a prospective, multi-centre study. The RADAR system was used in conjunction with a standard mapping system. High-density contact mapping was performed using a coronary sinus catheter as a reference to record electrograms in various anatomic locations. These electrograms were then sorted and stitched together to create a panoramic 3D conduction vector map for each coronary sinus phase. Sites of rotational activity and focal impulses were identified in the individual conduction vector maps. The initial outcome of using the RADAR system was promising, with 55% of patients experiencing AF termination after driver ablation and 82% remaining AF-free during follow-up. Non-invasive electrocardiographic imaging Recently, the non-invasive electrocardiographic imaging (ECGi) technique was used to map rotors . Using an external cardiac 3D mapping system (ECVUE system, Medtronic), patients with AF wore a vest with 252 electrodes and underwent CT scanning to obtain the relationship between the 3D geometry of both atria and the positions of 252 electrodes on the vest. Atrial activation during AF was obtained by phase mapping of the recorded ECG. Active driving and passive propagation zones were identified by analysing all accumulated images for each patient. A driver is defined as a focal breakthrough, when the image presents a centrifugal activation from a point or zone, or a spiral wave, when the wave is rotated around a centre on phase progression. In this study, the rate of AF termination during rotor ablation was 70%, and 85% of patients were free from AF recurrence at a 1-year follow-up. The latter AFACART study and the TARGET-AF1 trial also observed favourable outcomes of non-invasive ECGi-guided driver ablation for PerAF. However, the system has some limitations. First, epicardial far-field unipolar signal mapping results in poor stability of atrial signal quality and makes it challenging to distinguish micro-re-entrant from focal breakthrough; second, the activation data are a composite of endocardial, epicardial, and intramural patterns and interactions, and therefore, cannot differentiate the signals from overlapping cardiac anatomy sites, such as coronary sinus and inter-atrial septum. Third, phase mapping tends to introduce false rotors during complex activation patterns. , The conduction of two waves in opposite directions on both sides of a conduction block line may be misinterpreted as a rotor in phase analysis. Hence, although this system is a panoramic high-resolution mapping technology, it has a non-negligible false-positive rate in AF rotor mapping due to the inherent deficiency of external signals. Non-contact charge density mapping The charge density (CD) mapping is performed by a combined imaging and multi-electrode mapping system (AcQMap, Acutus Medical, Carlsbad, CA), which has 48 ultrasound transducers to reconstruct atrial chamber anatomy and 48 low-impedance, high-fidelity electrodes for recording biopotential signals. In brief, an inverse algorithm, based on the principles of electrostatic field theory, obtains the global distribution of CD sources across the endocardial surface generated by ultrasonic imaging and then creates global maps of cardiac activation displayed as a spatiotemporal window. The UNCOVER AF Trial reported a 72.5% success rate of single ablation for PerAF guided by the AcQMap system. The key advantages of CD mapping include (i) global and continuous mapping of AF with a reduction in far-field interference; (ii) enabling the identification of both re-entrant and focal activation; (iii) rapid (<5 min) and non-contact reconstruction of atrial anatomy leading to avoidance of some problems encountered by the impedance-based approach. The major limitations of this mapping method are the inability to study epicardial or transmural mechanisms and the inability to properly and successfully identify the so-called rotor activity. The focal impulse and rotor modulation (FIRM) methodology was first used clinically in the CONFIRM study, published in 2012 by Narayan et al. The basket-like 64 electrodes would simultaneously contact the endocardium of the atria and record potentials, and the proprietary computer software was used to analyse the electrical signals collected by the basket electrodes to guide the ablation. In the study, rotors were mapped in 97% of the study population, with an average of 2.1 ± 1.0 rotors per patient, and rotor ablation terminated AF or prolonged AF cycle length in 86% of the patients, and 82.4% of the patients were in sinus rhythm during a mean follow-up of 273 days. These favourable outcomes, however, were not reproduced at many other centres . In a meta-analysis, circumferential pulmonary vein isolation, combined with a rotor ablation strategy guided by a basket catheter, was not found superior to pulmonary vein isolation alone. The flaws seen with inter-spline bunching resulted in a loss of coverage and contact, and only 63.1% of the inter-electrode distances were less than the most stringent spatial resolution required for the identification of rotors in human AF. Moreover, a computer simulation study found that several high-density mapping catheters (AFocusII and PentaRay) had inter-electrode spacings below minimum resolution (11.9 mm), suggesting that these catheters have a higher resolution to locate PSs if placed over the rotor core accurately. Although these catheters have satisfactory mapping resolution, they are not panoramic mapping devices that could show overall atrial electrical activity. Therefore, both panoramic mapping catheters and local high-density mapping catheters have advantages and disadvantages, and there are no comparative clinical studies to determine which would provide superior performance. Recently, electrographic flow (EGF) mapping (Ablamap® software; Ablacon, Wheat Ridge, CO) emerged as a novel method to identify AF driver by estimating atrial cardiac action potential flow. An algorithm that combined Green’s minimal bending energy algorithm and Horn–Schunck Flow algorithm was used to analysed the time-domain information from the unipolar electrograms collected by a basket-like catheter and converted it into the space domain of flow vectors for the identification of singularities where the flow vector angles around a point covered 360°. One strength of this method is the potential to distinguish active and passive rotors. Theoretically, it can make up for insufficient local resolution of the basket-like catheter. The ongoing FLOW-AF study (NCT 04473963) will verify the efficacy of driver ablation–guided EGF mapping. In 2020, the real-time electrogram analysis for the driver (RADAR) system was introduced to guide driver ablation in a prospective, multi-centre study. The RADAR system was used in conjunction with a standard mapping system. High-density contact mapping was performed using a coronary sinus catheter as a reference to record electrograms in various anatomic locations. These electrograms were then sorted and stitched together to create a panoramic 3D conduction vector map for each coronary sinus phase. Sites of rotational activity and focal impulses were identified in the individual conduction vector maps. The initial outcome of using the RADAR system was promising, with 55% of patients experiencing AF termination after driver ablation and 82% remaining AF-free during follow-up. Recently, the non-invasive electrocardiographic imaging (ECGi) technique was used to map rotors . Using an external cardiac 3D mapping system (ECVUE system, Medtronic), patients with AF wore a vest with 252 electrodes and underwent CT scanning to obtain the relationship between the 3D geometry of both atria and the positions of 252 electrodes on the vest. Atrial activation during AF was obtained by phase mapping of the recorded ECG. Active driving and passive propagation zones were identified by analysing all accumulated images for each patient. A driver is defined as a focal breakthrough, when the image presents a centrifugal activation from a point or zone, or a spiral wave, when the wave is rotated around a centre on phase progression. In this study, the rate of AF termination during rotor ablation was 70%, and 85% of patients were free from AF recurrence at a 1-year follow-up. The latter AFACART study and the TARGET-AF1 trial also observed favourable outcomes of non-invasive ECGi-guided driver ablation for PerAF. However, the system has some limitations. First, epicardial far-field unipolar signal mapping results in poor stability of atrial signal quality and makes it challenging to distinguish micro-re-entrant from focal breakthrough; second, the activation data are a composite of endocardial, epicardial, and intramural patterns and interactions, and therefore, cannot differentiate the signals from overlapping cardiac anatomy sites, such as coronary sinus and inter-atrial septum. Third, phase mapping tends to introduce false rotors during complex activation patterns. , The conduction of two waves in opposite directions on both sides of a conduction block line may be misinterpreted as a rotor in phase analysis. Hence, although this system is a panoramic high-resolution mapping technology, it has a non-negligible false-positive rate in AF rotor mapping due to the inherent deficiency of external signals. The charge density (CD) mapping is performed by a combined imaging and multi-electrode mapping system (AcQMap, Acutus Medical, Carlsbad, CA), which has 48 ultrasound transducers to reconstruct atrial chamber anatomy and 48 low-impedance, high-fidelity electrodes for recording biopotential signals. In brief, an inverse algorithm, based on the principles of electrostatic field theory, obtains the global distribution of CD sources across the endocardial surface generated by ultrasonic imaging and then creates global maps of cardiac activation displayed as a spatiotemporal window. The UNCOVER AF Trial reported a 72.5% success rate of single ablation for PerAF guided by the AcQMap system. The key advantages of CD mapping include (i) global and continuous mapping of AF with a reduction in far-field interference; (ii) enabling the identification of both re-entrant and focal activation; (iii) rapid (<5 min) and non-contact reconstruction of atrial anatomy leading to avoidance of some problems encountered by the impedance-based approach. The major limitations of this mapping method are the inability to study epicardial or transmural mechanisms and the inability to properly and successfully identify the so-called rotor activity. Dominant frequency mapping In 2005, Haïssaguerre’s group performed dominant frequency (DF) mapping by using spectral analysis to construct hierarchical gradients of activation frequencies at different atrial regions. The DF is calculated from the cycle length of local electrograms, which correlate with the local AF frequency, i.e. a faster local AF waveform frequency is associated with a higher DF. In theory, localized sites with the highest frequency activity are markers of critical AF drivers or rotors. Ablation of these high DF sites achieved AF termination and prolongation of the AF cycle length. However, the latter radiofrequency ablation of drivers of atrial fibrillation study failed to demonstrate the superiority of localized high-frequency source ablation over PVI in either patient with paroxysmal AF or persistent AF. The technical limitations of DF mapping may account for these results. The local potentials become more complex as AF progresses; for example, with local fragmented potentials or double potentials, the value obtained from this DF calculation may be faster than that of real drivers or rotors, precluding accurate DF measurement. Basic research has also confirmed that the DF of fragmented potentials generated by the collision of wavefronts around rotors is consistent with that of the rotor core region. Therefore, these limitations of DF mapping cannot be ignored, and this method can be used only as a reference for rotor potential analysis. In 2017, the CARTO-Finder, a novel technique module of the CARTO mapping system (Biosense Webster), enabled a dynamic evaluation of rotational and focal activation during AF and was used to identify areas of rotor domains. , Ablation of rotor domains effectively eliminated frequency gradients, and 70% of patients with long-standing persistent AF were free of AF recurrence at 12 months. However, only 13 patients were included in this study, and the efficacy of using CARTO-Finder to target AF drivers will be further confirmed by the ongoing larger multi-centre study (ClinicalTrials.gov Identifier: NCT03064451). Phase similarity mapping In 2013, Lin and colleagues demonstrated that regions of AF drivers were characterized by regular, organized, and rapid repetitive activities with a high degree of similarity in bipolar electrogram configuration. After that, in their single-centre, randomized study, a novel similarity index (SI) was calculated by using phase analysis of the bipolar signal acquired by the AFocus II catheter to identify potential AF drivers for ablation. The obtained electrogram signal underwent phase calculation simulation through a computer system to filter out interference signals and far-field contamination, theoretically improving the accuracy of rotor identification. An SI vector field was then constructed between a pair of the nearest electrodes to analyse the electrical wave propagation around AF drivers based on curvature and divergence forces, and therefore, can distinguish a rotor from a focal source or random fractionated. In this study, an average of 2.6 ± 0.89 high SI regions were mapped per patient, and the rate of AF termination was superior (68% vs. 27%) to that of the control group with a fractionated potential ablation strategy, with a high success rate of 83% in long-term follow-up. Dispersion mapping In 2017, Seitz et al. used the PentaRay catheter to identify the region with rotor activity in AF. This mapping is based on the theoretical premise that when a PentaRay catheter is placed at the centre of a rotor, its electrode branches will record the sequential excitation phases of rotor wavefronts, i.e. the asynchrony of the bipolar potential or ‘dispersion electrograms.’ However, the study methodology did not exclude fragmented potentials in the dispersion analysis and the area with both fragmented and non-fragmented potentials as an ablation target. Despite the excellent termination rate of AF, extensive ablation is performed on a wide area of atria (15.8–29% of LA surface area), and the unintentional creation of an unnecessary pro-arrhythmic scar cannot be ignored. Recently, the results from our study , showed that HF and dispersion electrogram-guided rotor ablation abruptly terminated AF, substantially improving long-term AF elimination. Notably, it showed a decreased ablation area (4.3–13.3% of LA surface area) and suggested that most fragmented potentials are the by-products of re-entrant rotors that break down at their boundaries. Therefore, local high-density mapping provides more detailed information to distinguish the rotor from complicated AF potentials despite not providing a panoramic imaging of the whole atrium. Stochastic trajectory analysis of ranked signals mapping Stochastic trajectory analysis of ranked signals (STAR) is a novel mapping method that compares activation times across electrode pairs to identify atrial regions with earlier activation than neighbouring areas. The author defined region as an AF driver if it leads for more than 75% of wavefronts during a recording period. In 2020, Honarbakhsh et al. confirmed the efficacy of STAR-guided ablation in a single-centre, prospective study, and the success rate was 81.5%. The STAR mapping method has some potential advantages: it allows both global and sequential mapping and is compatible with different commercially available multi-polar catheters (CARTO, Rhythmia, and EnSite). However, this method cannot distinguish the driver mechanism between rotational and focal. Furthermore, the ablation of early activation areas depended on the operator’s interpretation. The limited coverage and contact of basket catheters and the position and orientation of the catheter may also influence the mapping accuracy. Focal source and trigger computational algorithm In 2020, Chauhan et al. reported their proprietary focal source and trigger (FaST) computational algorithm, which allowed automatically identifying sustained periodic bipolar and unipolar QS electrogram morphology during AF. Both patients with paroxysmal and persistent AF were enrolled in this validation study. PVI plus FaST site ablation decreased AF recurrence compared with PVI alone, but the result did not reach statistical significance ( P = 0.064). One advantage of the FaST mapping is time efficiency: the mean ablation time of FaST sites outside the PV was only 8.5 ± 5.1 min. Notwithstanding, the FaST mapping was not performed comprehensively in this study and may miss some potential focal sources in the right atrium. Furthermore, FaST signals may arise from mechanisms other than focal sources, such as endo-epicardial breakthroughs. Repetitive-regular activities mapping Recently, Pappone and colleagues introduced a novel real-time integrated mapping technique to identify potential driver regions characterized by repetitive-regular activities (RRas). This mapping method consists of three components: a regular cycle length map, a fragmentation map, and a peak-to-peak voltage measurement. The software integrated the three mapping results and allowed an automated evaluation of multiple mechanisms during AF by identifying regions exhibiting RRas, fragmentation, the conduction velocity of consistent wavefronts, and electrically silent areas. Therefore, it provided a patient-tailored strategy. In this study, a modified PVI, plus RRa site ablation, achieved a higher rate of AF termination (61% vs. 30%) and less AF recurrence (73% vs. 50%) compared with a modified PVI alone. One drawback of this approach is that it cannot distinguish the specific electrophysiological characteristic of the repetitive-regular electrograms, such as rotors, focal sources, or micro re-entries. In 2005, Haïssaguerre’s group performed dominant frequency (DF) mapping by using spectral analysis to construct hierarchical gradients of activation frequencies at different atrial regions. The DF is calculated from the cycle length of local electrograms, which correlate with the local AF frequency, i.e. a faster local AF waveform frequency is associated with a higher DF. In theory, localized sites with the highest frequency activity are markers of critical AF drivers or rotors. Ablation of these high DF sites achieved AF termination and prolongation of the AF cycle length. However, the latter radiofrequency ablation of drivers of atrial fibrillation study failed to demonstrate the superiority of localized high-frequency source ablation over PVI in either patient with paroxysmal AF or persistent AF. The technical limitations of DF mapping may account for these results. The local potentials become more complex as AF progresses; for example, with local fragmented potentials or double potentials, the value obtained from this DF calculation may be faster than that of real drivers or rotors, precluding accurate DF measurement. Basic research has also confirmed that the DF of fragmented potentials generated by the collision of wavefronts around rotors is consistent with that of the rotor core region. Therefore, these limitations of DF mapping cannot be ignored, and this method can be used only as a reference for rotor potential analysis. In 2017, the CARTO-Finder, a novel technique module of the CARTO mapping system (Biosense Webster), enabled a dynamic evaluation of rotational and focal activation during AF and was used to identify areas of rotor domains. , Ablation of rotor domains effectively eliminated frequency gradients, and 70% of patients with long-standing persistent AF were free of AF recurrence at 12 months. However, only 13 patients were included in this study, and the efficacy of using CARTO-Finder to target AF drivers will be further confirmed by the ongoing larger multi-centre study (ClinicalTrials.gov Identifier: NCT03064451). In 2013, Lin and colleagues demonstrated that regions of AF drivers were characterized by regular, organized, and rapid repetitive activities with a high degree of similarity in bipolar electrogram configuration. After that, in their single-centre, randomized study, a novel similarity index (SI) was calculated by using phase analysis of the bipolar signal acquired by the AFocus II catheter to identify potential AF drivers for ablation. The obtained electrogram signal underwent phase calculation simulation through a computer system to filter out interference signals and far-field contamination, theoretically improving the accuracy of rotor identification. An SI vector field was then constructed between a pair of the nearest electrodes to analyse the electrical wave propagation around AF drivers based on curvature and divergence forces, and therefore, can distinguish a rotor from a focal source or random fractionated. In this study, an average of 2.6 ± 0.89 high SI regions were mapped per patient, and the rate of AF termination was superior (68% vs. 27%) to that of the control group with a fractionated potential ablation strategy, with a high success rate of 83% in long-term follow-up. In 2017, Seitz et al. used the PentaRay catheter to identify the region with rotor activity in AF. This mapping is based on the theoretical premise that when a PentaRay catheter is placed at the centre of a rotor, its electrode branches will record the sequential excitation phases of rotor wavefronts, i.e. the asynchrony of the bipolar potential or ‘dispersion electrograms.’ However, the study methodology did not exclude fragmented potentials in the dispersion analysis and the area with both fragmented and non-fragmented potentials as an ablation target. Despite the excellent termination rate of AF, extensive ablation is performed on a wide area of atria (15.8–29% of LA surface area), and the unintentional creation of an unnecessary pro-arrhythmic scar cannot be ignored. Recently, the results from our study , showed that HF and dispersion electrogram-guided rotor ablation abruptly terminated AF, substantially improving long-term AF elimination. Notably, it showed a decreased ablation area (4.3–13.3% of LA surface area) and suggested that most fragmented potentials are the by-products of re-entrant rotors that break down at their boundaries. Therefore, local high-density mapping provides more detailed information to distinguish the rotor from complicated AF potentials despite not providing a panoramic imaging of the whole atrium. Stochastic trajectory analysis of ranked signals (STAR) is a novel mapping method that compares activation times across electrode pairs to identify atrial regions with earlier activation than neighbouring areas. The author defined region as an AF driver if it leads for more than 75% of wavefronts during a recording period. In 2020, Honarbakhsh et al. confirmed the efficacy of STAR-guided ablation in a single-centre, prospective study, and the success rate was 81.5%. The STAR mapping method has some potential advantages: it allows both global and sequential mapping and is compatible with different commercially available multi-polar catheters (CARTO, Rhythmia, and EnSite). However, this method cannot distinguish the driver mechanism between rotational and focal. Furthermore, the ablation of early activation areas depended on the operator’s interpretation. The limited coverage and contact of basket catheters and the position and orientation of the catheter may also influence the mapping accuracy. In 2020, Chauhan et al. reported their proprietary focal source and trigger (FaST) computational algorithm, which allowed automatically identifying sustained periodic bipolar and unipolar QS electrogram morphology during AF. Both patients with paroxysmal and persistent AF were enrolled in this validation study. PVI plus FaST site ablation decreased AF recurrence compared with PVI alone, but the result did not reach statistical significance ( P = 0.064). One advantage of the FaST mapping is time efficiency: the mean ablation time of FaST sites outside the PV was only 8.5 ± 5.1 min. Notwithstanding, the FaST mapping was not performed comprehensively in this study and may miss some potential focal sources in the right atrium. Furthermore, FaST signals may arise from mechanisms other than focal sources, such as endo-epicardial breakthroughs. Recently, Pappone and colleagues introduced a novel real-time integrated mapping technique to identify potential driver regions characterized by repetitive-regular activities (RRas). This mapping method consists of three components: a regular cycle length map, a fragmentation map, and a peak-to-peak voltage measurement. The software integrated the three mapping results and allowed an automated evaluation of multiple mechanisms during AF by identifying regions exhibiting RRas, fragmentation, the conduction velocity of consistent wavefronts, and electrically silent areas. Therefore, it provided a patient-tailored strategy. In this study, a modified PVI, plus RRa site ablation, achieved a higher rate of AF termination (61% vs. 30%) and less AF recurrence (73% vs. 50%) compared with a modified PVI alone. One drawback of this approach is that it cannot distinguish the specific electrophysiological characteristic of the repetitive-regular electrograms, such as rotors, focal sources, or micro re-entries. At present, the overall AF termination rate of rotor ablation is 44.5 ± 15.5% ( Figure ), the rate of freedom from AF/AT is 57.7 ± 7.7%, and the rate of freedom from AF is 69.5 ± 4.6% ( Figure ). , presents clinical studies of different currently available methods for rotor mapping. Although the termination and success rates were different among various mapping methods, the overall mean values were similar. Considerable discrepancy in the rate of successful ablation and AF termination was observed among studies of FIRM-guided approach, which is also the method with the largest number of published studies. A study reported that the AF termination rate and long-term success rate were only 5% and 21% by FIRM-guided ablation. The reasons for the difference in results are the defects of the mapping technology itself and the limitations of the study, such as the type of AF included, the follow-up time, and the sample size. Thus, outcomes of single-arm studies were significantly limited by high heterogeneity. However, the number of case-control clinical studies of traditional ablation strategies is relatively small, and we conducted a meta-analysis of 10 clinical trials with published case-control studies ( Figure ). The significant pooled OR for freedom from AF/AT in these 10 studies using the fixed-effects model was 0.53 [CI, 0.40–0.69 ( P = 0.037); I 2 = 42% ( P < 0.0001)]. Among the ten studies, only Tilz’s findings did not support driver ablation, and the study included paroxysmal AF and performed FIRM-guided ablation alone without PVI. Two other studies also compared AF driver-only ablation to PVI. , Although Seitz’s conclusions favour driver ablation, the incidence of atrial tachycardia is as high as about 30%. If these three studies are excluded, freedom from AF/AT produced an OR of 0.45 (CI, 0.32–0.62; P < 0.0001), with minimal heterogeneity between studies (I 2 = 0%). It supports the possible benefit of driver ablation as an additional strategy in improving freedom from all arrhythmias compared with conventional ablation alone. Although the overall results of the meta-analysis favoured driver ablation, we cannot ignore the inconsistency of clinical findings. Because there are several factors affecting the outcome of AF ablation, only in terms of mapping methods and ablation strategies, first of all, there are differences in the false-positive rate of different mapping devices, and the success rate is naturally different. Furthermore, the area of ablation lesions and different methods of ablation (linear or patch lesion) can also affect the incidence of iatrogenic arrhythmias. However, for the acute outcome of ablation, driver ablation achieved a higher AF termination rate than conventional ablation, , , , , suggesting an intervention in the maintenance mechanism of AF. Previous studies have also confirmed a favourable long-term outcome in patients with procedural AF termination. , Secondly, the results of studies may also be impacted by sample size and different control groups. Currently, there is still a lack of large-sample randomized controlled studies to directly compared driver ablation and conventional ablation. Previous studies set different ablation strategies as control groups, including PVI, , PVI + linear ablation, CAFE ablation, or ‘stepwise’ strategy. , It was therefore difficult to compare studies with one another. Moreover, there is also a lack of evidence comparing driver ablation with new strategies that have demonstrated satisfactory clinical outcomes, such as posterior BOX isolation or vein of Marshall ethanol infusion. In addition, most clinical studies have been conducted in ablation-naive patients, , , , and a part of these patients may be benefited by PVI alone. In the study of Seitz et al., PVI was not performed, and regions of AF termination were frequently located in the PV antrum. Therefore, enrichment of the study population for patients having extra-PV drivers would likely increase the success rate of adjuvant ablation approaches. Limitations of mapping technology AF is neither a purely focal nor a stable re-entry in nature, and sequential local high-density mapping may be limited by global resolution. Although panoramic mapping enables a real-time evaluation of propagation of the entire chamber, it also has many limitations, such as low spatial resolution and suboptimal electrode–tissue contact. The current mapping tools are only based on the analysis of unipolar or bipolar electrograms. The unipolar electrograms are vulnerable to far-field potentials, and the morphology of bipolar electrograms is affected by inter-electrode distances and wavefront direction. When the electrode resolution is insufficient, the interference is significant, and when the potential complexity is high, it is difficult to identify the false positives and false negatives of the rotor. Before optical mapping can be safely applied to the human body, computer and artificial intelligence technology should be the best way to identify rotors accurately. For example, in the research of RADAR, high-density mapping and computer signal processing technology are integrated to avoid the loss of mapping information and comprehensively analyse the overall electrical activation characteristics of AF, and the signal processing is rapid. Insufficient understanding of the rotor mechanism Although the theoretical research on rotors has been far ahead of understanding clinical practice, the clinical AF phenomenon is much more complicated than animal and in vitro models. In addition to the rotor phenomenon, micro re-entry, focal activity, or double-layer activation also participate in the maintenance of AF. However, these theories are not mutually exclusive, and different mechanisms of AF might exist in the same patient. This still needs to be revealed by clinical research and basic research. AF is neither a purely focal nor a stable re-entry in nature, and sequential local high-density mapping may be limited by global resolution. Although panoramic mapping enables a real-time evaluation of propagation of the entire chamber, it also has many limitations, such as low spatial resolution and suboptimal electrode–tissue contact. The current mapping tools are only based on the analysis of unipolar or bipolar electrograms. The unipolar electrograms are vulnerable to far-field potentials, and the morphology of bipolar electrograms is affected by inter-electrode distances and wavefront direction. When the electrode resolution is insufficient, the interference is significant, and when the potential complexity is high, it is difficult to identify the false positives and false negatives of the rotor. Before optical mapping can be safely applied to the human body, computer and artificial intelligence technology should be the best way to identify rotors accurately. For example, in the research of RADAR, high-density mapping and computer signal processing technology are integrated to avoid the loss of mapping information and comprehensively analyse the overall electrical activation characteristics of AF, and the signal processing is rapid. Although the theoretical research on rotors has been far ahead of understanding clinical practice, the clinical AF phenomenon is much more complicated than animal and in vitro models. In addition to the rotor phenomenon, micro re-entry, focal activity, or double-layer activation also participate in the maintenance of AF. However, these theories are not mutually exclusive, and different mechanisms of AF might exist in the same patient. This still needs to be revealed by clinical research and basic research. Studies on AF mechanisms have gradually revealed the complicated properties and diverse 3D forms of rotors that play an essential role in the occurrence and development of the AF substrate. However, limitations in current research methodology and the complexity of the rotor mechanism have rendered it difficult to achieve complementary results between basic science and clinical studies. Although initial studies of rotor ablation have yielded favourable outcomes, there is still a lack of large multi-centre studies to verify the efficacy of rotor ablation. Despite various rotor mapping methods, the current mapping approaches are limited by inadequate resolution. Further theoretical and technological developments are warranted to address the limitations discussed and allow the translation of basic and clinical science results to treat patients with AF. euad002_Supplementary_Data Click here for additional data file. |
Tonsil biopsy to detect chronic wasting disease in white-tailed deer ( | 70e2ab50-906b-4aa2-af33-e8ab06f75bbb | 10062608 | Anatomy[mh] | Since its initial identification several decades ago, the incidence of chronic wasting disease (CWD) in North American wild and farmed cervid populations has increased. Since animals can transmit infection months to years before developing clinical signs, strategies to limit transmission depend on detecting affected stock during early infection. Diagnosis of CWD was initially limited to postmortem examination of clinical animals, where histologic analysis of the central nervous system would reveal characteristic spongiform neurodegeneration in advanced cases . Antibody-based diagnostics were developed when the role of abnormally folded prion protein was recognized as central to prion diseases, accumulating in neural tissue before spongiform neurodegeneration and, in the case of cervids with CWD, even earlier in specific lymphoid tissues . Immunohistochemistry (IHC) provides a great deal of diagnostic certainty. A sample is identified as positive when immunostaining specific for disease forms of the prion protein (e.g., PrP CWD ) can be visualized in expected locations such as lymphoid follicles . The official regulatory test of the United States Department of Agriculture (USDA) used for postmortem diagnosis of CWD is IHC of the medial retropharyngeal lymph nodes (MRPLNs) and the obex . Though the MRPLNs are a site of early PrP CWD accumulation in white-tailed deer (WTD) ( Odocoileus virginianus ) , biopsy of the MRPLNs requires a surgical approach and is thus an impractical tissue source for routine antemortem diagnosis. Abnormal prion protein can also accumulate in the recto-anal mucosa-associated lymphoid tissue (RAMALT) in sheep, WTD, and elk, which is readily sampled through superficial mucosa biopsy in living animals . However, the diagnostic sensitivity of IHC using RAMALT samples can vary between 25% and 95%, depending on animal species and genetic variability within the prion protein gene ( PRNP ) . Furthermore, IHC detection of PrP CWD in the RAMALT of WTD can vary from 12 to 27 months after infection . Accumulation of PrP CWD in the palatine tonsils is a relatively early event in mule deer ( Odocoileus hemnionus ) and WTD , and some antemortem tonsil biopsy data have been published . From a retrospective study of WTD , PrP CWD was detected in tonsil biopsies by IHC as early as six months post-inoculation. In the present study, we report the IHC diagnostic sensitivity of a two-bite tonsil biopsy from 79 field cases in farmed WTD. All study deer were preclinical, and all samples, including tonsil biopsies, were collected postmortem. Also evaluated were the potential associations of infection stage, PRNP genotype at codon 96, and tonsil follicle metrics on detection of PrP CWD by tonsil biopsy IHC.
Sample collection The study was carried out using tissues collected postmortem by employees of, and under the authority of, USDA-APHIS and Texas state regulatory agencies. These WTD herds were depopulated as an official regulatory action due to the presence of CWD in the herds. No animals were euthanized for the purpose of this study. All study samples were collected opportunistically postmortem. All deer were considered preclinical and appeared healthy at the time of depopulation. Antemortem biopsy of the tonsil was performed in some of these animals as conducted by local regulatory agencies. Regulatory tissue samples (left and right MRPLNs, obex) were collected and submitted to the USDA National Veterinary Services Laboratories (NVSL) (Ames, IA) for official CWD IHC testing. After collecting the regulatory samples, a two-bite tonsil biopsy procedure was conducted as previously described , preserving the contralateral tonsil for unbiased metrics. In brief, the tongue was reflected and two biopsies were collected in situ from the left tonsil using a 6 mm ovarian biopsy instrument inserted into the left tonsillar crypt at a dorsolateral angle. When the tonsillar crypt was not large enough to insert the biopsy instrument, a bite of the overlying epithelium was first removed to expose the tonsil. The biopsies were placed into a tissue cassette with a sponge and put in 10% formalin. The biopsy technique mimicked the antemortem process as much as possible. To limit variation in the biopsy technique , all the samples were collected at diagnostic laboratories and a single operator (TAN) utilized the same procedure across depopulation groups. After tonsil biopsy, both whole tonsils were removed and placed in 10% formalin. The tonsil samples were held until the official CWD diagnostic reports were received from NVSL. Immunohistochemistry Immunohistochemistry (IHC) was conducted at NVSL using the standard operating procedures for detecting PrP CWD as previously described . Briefly, 5 μm tissue sections were mounted on positively charged glass slides (Fisher Scientific), oven dried, treated with formic acid, rinsed with Tris buffer (pH 7.5), and subjected to hydrated autoclaving using DIVA antigen retrieval solution (Biocare Medical) and a decloaking chamber (Biocare Medical). Immunostaining was carried out using an automated immunostainer and associated reagents (Ventana Medical Systems) as well as the Anti-Prion (99) Research Kit, RTU (Ventana Medical Systems). The main reagents of these kits included decloaker solution, antibody block, monoclonal antibody F99, alkaline phosphatase-conjugated anti-mouse IgG secondary antibody, fast red chromogen, and hematoxylin. Each automated run included tissue controls from CWD-infected and non-infected deer. Data collection and statistical analyses Age was either precisely known from records or only known at the birth year level. For quantitative purposes, ages were recoded into one-year age groups such that precise ages were rounded up if equal or greater than a one-half year. The genotypes of the prion protein gene ( PRNP ) at codon 96—coding for the amino acids glycine (G) and serine (S)—were determined by a commercial service (GeneCheck). Genotypes at other codons were not determined. The stage of preclinical infection was classified by IHC of both MRPLNs and the obex, where ‘early’ stage deer had PrP CWD accumulation in MRPLN follicles but not the obex, and ‘late’ stage deer had accumulation at both tissue locations. The PrP CWD -positive and total numbers of lymphoid follicles were counted in a thin section of the whole tonsil. Data were analyzed and graphed using the procedures available in SAS 9.4 (SAS Institute Inc.). Basic statistics and histogram plots were produced using the UNIVARIATE procedure. The FREQ procedure was used to calculate diagnostic sensitivities, exact 95% confidence limits (CLs), and measures of agreement (Cohen’s kappa coefficient, κ; McNemar’s Q test for 2x2 contingencies, Q M ; Cochran’s Q test for stratified contingencies, Q C ). Values of κ were categorized as one of six agreement levels : none = 0–0.20, minimum = 0.21–0.39, weak = 0.40–0.59, moderate = 0.60–0.79, strong = 0.80–0.90, almost perfect > 0.90. The LOGISTIC procedure was used to test the association of stage of infection with genotype, sex, and age group and included first and second-order effects. The GLIMMIX procedure was used to model the effects of genotype, sex, and age on the total follicle counts of the whole tonsil sample (distribution: negative binomial). The likelihood ratio ( Q LR ), 95% CLs, and fit plots were used to assess the significance of each regression model.
The study was carried out using tissues collected postmortem by employees of, and under the authority of, USDA-APHIS and Texas state regulatory agencies. These WTD herds were depopulated as an official regulatory action due to the presence of CWD in the herds. No animals were euthanized for the purpose of this study. All study samples were collected opportunistically postmortem. All deer were considered preclinical and appeared healthy at the time of depopulation. Antemortem biopsy of the tonsil was performed in some of these animals as conducted by local regulatory agencies. Regulatory tissue samples (left and right MRPLNs, obex) were collected and submitted to the USDA National Veterinary Services Laboratories (NVSL) (Ames, IA) for official CWD IHC testing. After collecting the regulatory samples, a two-bite tonsil biopsy procedure was conducted as previously described , preserving the contralateral tonsil for unbiased metrics. In brief, the tongue was reflected and two biopsies were collected in situ from the left tonsil using a 6 mm ovarian biopsy instrument inserted into the left tonsillar crypt at a dorsolateral angle. When the tonsillar crypt was not large enough to insert the biopsy instrument, a bite of the overlying epithelium was first removed to expose the tonsil. The biopsies were placed into a tissue cassette with a sponge and put in 10% formalin. The biopsy technique mimicked the antemortem process as much as possible. To limit variation in the biopsy technique , all the samples were collected at diagnostic laboratories and a single operator (TAN) utilized the same procedure across depopulation groups. After tonsil biopsy, both whole tonsils were removed and placed in 10% formalin. The tonsil samples were held until the official CWD diagnostic reports were received from NVSL.
Immunohistochemistry (IHC) was conducted at NVSL using the standard operating procedures for detecting PrP CWD as previously described . Briefly, 5 μm tissue sections were mounted on positively charged glass slides (Fisher Scientific), oven dried, treated with formic acid, rinsed with Tris buffer (pH 7.5), and subjected to hydrated autoclaving using DIVA antigen retrieval solution (Biocare Medical) and a decloaking chamber (Biocare Medical). Immunostaining was carried out using an automated immunostainer and associated reagents (Ventana Medical Systems) as well as the Anti-Prion (99) Research Kit, RTU (Ventana Medical Systems). The main reagents of these kits included decloaker solution, antibody block, monoclonal antibody F99, alkaline phosphatase-conjugated anti-mouse IgG secondary antibody, fast red chromogen, and hematoxylin. Each automated run included tissue controls from CWD-infected and non-infected deer.
Age was either precisely known from records or only known at the birth year level. For quantitative purposes, ages were recoded into one-year age groups such that precise ages were rounded up if equal or greater than a one-half year. The genotypes of the prion protein gene ( PRNP ) at codon 96—coding for the amino acids glycine (G) and serine (S)—were determined by a commercial service (GeneCheck). Genotypes at other codons were not determined. The stage of preclinical infection was classified by IHC of both MRPLNs and the obex, where ‘early’ stage deer had PrP CWD accumulation in MRPLN follicles but not the obex, and ‘late’ stage deer had accumulation at both tissue locations. The PrP CWD -positive and total numbers of lymphoid follicles were counted in a thin section of the whole tonsil. Data were analyzed and graphed using the procedures available in SAS 9.4 (SAS Institute Inc.). Basic statistics and histogram plots were produced using the UNIVARIATE procedure. The FREQ procedure was used to calculate diagnostic sensitivities, exact 95% confidence limits (CLs), and measures of agreement (Cohen’s kappa coefficient, κ; McNemar’s Q test for 2x2 contingencies, Q M ; Cochran’s Q test for stratified contingencies, Q C ). Values of κ were categorized as one of six agreement levels : none = 0–0.20, minimum = 0.21–0.39, weak = 0.40–0.59, moderate = 0.60–0.79, strong = 0.80–0.90, almost perfect > 0.90. The LOGISTIC procedure was used to test the association of stage of infection with genotype, sex, and age group and included first and second-order effects. The GLIMMIX procedure was used to model the effects of genotype, sex, and age on the total follicle counts of the whole tonsil sample (distribution: negative binomial). The likelihood ratio ( Q LR ), 95% CLs, and fit plots were used to assess the significance of each regression model.
Seventy-nine (31 female and 48 male) WTD deer from nine herds were identified as infected with CWD by official testing at NVSL. The data collected from these animals are provided in . The dates of birth were known for 47 deer; the ages of 29 deer were recorded in whole years. Ages ranged from two deer less than 1 year of age to two deer aged 9.29 years (mean of original age data, 4.01 years). The median of ages grouped by year was 3 years. PRNP codon 96 genotypes included 56 GG, 17 GS, and one SS deer; genotype was not available for two female and three male deer. The age and sex distributions for genotypes GG and GS are shown in ; the single SS deer was a 6.33-year-old female. Accumulation of PrP CWD in MRPLN follicles was observed in all 79 deer. The stage of infection was classified as early preclinical in 42 deer (25 male, 17 female) and late preclinical in 37 deer (23 male, 14 female). The age distributions of deer in early and late preclinical stages of infection are shown for PRNP codon 96 genotypes GG and GS in ; the SS deer was in early preclinical infection. The probability of a deer being in an early stage of preclinical infection was not dependent on age group, PRNP genotype (where codon 96 was either GG or GS), sex, or any interaction of these factors ( Q LR , P = 0.2037). Diagnostic sensitivity of tonsil IHC using sections of whole and biopsy sample types Upon official diagnosis, the paired postmortem samples of whole tonsil and two-bite tonsil biopsy were submitted to NVSL for evaluation by IHC. Postmortem tonsil biopsies from all deer had more than six lymphoid follicles present. Tonsil biopsies collected antemortem were considered inconclusive if accumulation of PrP CWD was not observed and fewer than six lymphoid follicles were present in thin sections. Accumulation of PrP CWD in antemortem tonsil biopsies was detected in 14 of 36 WTD in which antemortem sampling had been conducted . Of the 22 WTD in which PrP CWD was not detected antemortem, 12 were detected in a postmortem biopsy. Conversely, tonsil accumulation of PrP CWD in antemortem biopsies was not detected in any WTD in which accumulation was not detected in the postmortem biopsy. Hereafter, all results are for postmortem samples. Accumulation of PrP CWD was observed in a thin section of whole tonsil in 69 deer (diagnostic sensitivity = 87.3%) and in the tonsil biopsy samples of 57 deer (72.2%). Accumulation of PrP CWD in a tonsil biopsy was only observed when accumulation was also observed in the whole tonsil. These paired estimates of general diagnostic sensitivity (i.e., without consideration of other factors) were significantly different ( Q M = 12.0, P exact = 0.0005). Furthermore, the agreement of diagnoses between sample types was categorized as weak (κ = 0.5460, 95% CLs: 0.3352, 0.7568) but was better than by chance alone ( P exact < 0.0001). The following analyses compare diagnostic sensitivities and agreement as stratified by stage of preclinical infection and by genotype at PRNP codon 96. The agreement of results between sample types depended on the stage of preclinical infection ( Q c = 20.7170; P < 0.0001). From deer in late preclinical infection, there were no discordant pairs of results (that is, there was perfect agreement between sample types), yielding a joint tonsil IHC diagnostic sensitivity of 91.9% (exact 95% CLs: 78.1%, 98.3%). For deer in early preclinical infection, there was minimum agreement between tonsil sample types (κ = 0.3898, 95% CLs: 0.1587, 0.6210) but which was better than by chance alone ( P exact < 0.0019). The early preclinical diagnostic sensitivity of whole tonsil IHC was 83.3% (exact 95% CLs: 68.6%, 93.0%) and for tonsil biopsy IHC was 54.8% (exact 95% CLs: 38.7%, 70.2%); these estimates were significantly different ( Q M = 12.000, P exact = 0.0005). The agreement of results between sample types significantly depended on genotype (GG vs GS) stratified by stage of infection ( Q c = 21.3860; P < 0.0001). For deer in late preclinical infection, there were no discordant pairs of results for either genotype, yielding a joint estimate of tonsil IHC diagnostic sensitivity in GG deer of 92.6% (exact 95% CLs: 75.7%, 99.1%) and in GS deer of 85.7% (42.1%, 99.6%); a significant difference between these joint sensitivities was not detected ( Q C = 4.000, P = 0.1353). In contrast, during early preclinical infection there was minimum agreement of results between tonsil sample types when from GG deer (κ = 0.3596, 95% CLs: 0.0423, 0.6770) and when from GS deer (κ = 0.4444, 95% CLs: 0.0071, 0.8818). The agreement of results from early preclinical GG deer was significantly better than by chance alone ( P exact = 0.0328) but agreement from early preclinical GS deer was not ( P exact = 0.1667). For early preclinical GG deer, the tonsil IHC diagnostic sensitivity for whole samples was 89.7% (exact 95% CLs: 72.7%, 97.8%) but for tonsil biopsy samples was 65.5% (exact 95% CLs: 45.7%, 82.1%); these estimates were significantly different ( Q M = 7.000, P exact = 0.0156). For early preclinical GS deer, a statistical difference between tonsil IHC diagnostic sensitivity of whole samples (60.0%, 95% CLs: 26.2%, 87.8%) and tonsil biopsy (30.0%, 95% CLs: 6.7%, 65.3%) was not detected ( Q M = 3.000, P exact = 0.2500). Relationship of whole tonsil metrics with the probability of detecting PrP CWD in a tonsil biopsy The proportion of PrP CWD positive tonsil follicles was estimated by counting the total and positive numbers of follicles present in thin sections of the unbiopsied whole tonsil (N = 66 WTD; ). The total number of whole tonsil follicles counted was highly variable between deer (mean = 126.7, standard deviation = 47.5). The mean of whole tonsil follicle counts was marginally dependent on the animal’s age ( F = 4.83, P = 0.0316); the estimated reduction in mean total follicle count was 4.9 follicles per year. The probability of a false negative tonsil biopsy result in early preclinical deer was not significantly dependent on the whole tonsil total follicle count ( Q LR = 0.3717, P = 0.5421). In contrast, the probability of a false negative tonsil biopsy result in early preclinical deer was significantly dependent on the whole tonsil estimate of the proportion of positive follicles ( Q LR = 30.4393, P < 0.0001; ). The odds of a false negative result based on a two-bite tonsil biopsy from deer in early preclinical infection increased 1.617 (95% CLs: 1.226, 2.603) for each 0.1 unit decrease in positive proportion of whole tonsil follicles .
Upon official diagnosis, the paired postmortem samples of whole tonsil and two-bite tonsil biopsy were submitted to NVSL for evaluation by IHC. Postmortem tonsil biopsies from all deer had more than six lymphoid follicles present. Tonsil biopsies collected antemortem were considered inconclusive if accumulation of PrP CWD was not observed and fewer than six lymphoid follicles were present in thin sections. Accumulation of PrP CWD in antemortem tonsil biopsies was detected in 14 of 36 WTD in which antemortem sampling had been conducted . Of the 22 WTD in which PrP CWD was not detected antemortem, 12 were detected in a postmortem biopsy. Conversely, tonsil accumulation of PrP CWD in antemortem biopsies was not detected in any WTD in which accumulation was not detected in the postmortem biopsy. Hereafter, all results are for postmortem samples. Accumulation of PrP CWD was observed in a thin section of whole tonsil in 69 deer (diagnostic sensitivity = 87.3%) and in the tonsil biopsy samples of 57 deer (72.2%). Accumulation of PrP CWD in a tonsil biopsy was only observed when accumulation was also observed in the whole tonsil. These paired estimates of general diagnostic sensitivity (i.e., without consideration of other factors) were significantly different ( Q M = 12.0, P exact = 0.0005). Furthermore, the agreement of diagnoses between sample types was categorized as weak (κ = 0.5460, 95% CLs: 0.3352, 0.7568) but was better than by chance alone ( P exact < 0.0001). The following analyses compare diagnostic sensitivities and agreement as stratified by stage of preclinical infection and by genotype at PRNP codon 96. The agreement of results between sample types depended on the stage of preclinical infection ( Q c = 20.7170; P < 0.0001). From deer in late preclinical infection, there were no discordant pairs of results (that is, there was perfect agreement between sample types), yielding a joint tonsil IHC diagnostic sensitivity of 91.9% (exact 95% CLs: 78.1%, 98.3%). For deer in early preclinical infection, there was minimum agreement between tonsil sample types (κ = 0.3898, 95% CLs: 0.1587, 0.6210) but which was better than by chance alone ( P exact < 0.0019). The early preclinical diagnostic sensitivity of whole tonsil IHC was 83.3% (exact 95% CLs: 68.6%, 93.0%) and for tonsil biopsy IHC was 54.8% (exact 95% CLs: 38.7%, 70.2%); these estimates were significantly different ( Q M = 12.000, P exact = 0.0005). The agreement of results between sample types significantly depended on genotype (GG vs GS) stratified by stage of infection ( Q c = 21.3860; P < 0.0001). For deer in late preclinical infection, there were no discordant pairs of results for either genotype, yielding a joint estimate of tonsil IHC diagnostic sensitivity in GG deer of 92.6% (exact 95% CLs: 75.7%, 99.1%) and in GS deer of 85.7% (42.1%, 99.6%); a significant difference between these joint sensitivities was not detected ( Q C = 4.000, P = 0.1353). In contrast, during early preclinical infection there was minimum agreement of results between tonsil sample types when from GG deer (κ = 0.3596, 95% CLs: 0.0423, 0.6770) and when from GS deer (κ = 0.4444, 95% CLs: 0.0071, 0.8818). The agreement of results from early preclinical GG deer was significantly better than by chance alone ( P exact = 0.0328) but agreement from early preclinical GS deer was not ( P exact = 0.1667). For early preclinical GG deer, the tonsil IHC diagnostic sensitivity for whole samples was 89.7% (exact 95% CLs: 72.7%, 97.8%) but for tonsil biopsy samples was 65.5% (exact 95% CLs: 45.7%, 82.1%); these estimates were significantly different ( Q M = 7.000, P exact = 0.0156). For early preclinical GS deer, a statistical difference between tonsil IHC diagnostic sensitivity of whole samples (60.0%, 95% CLs: 26.2%, 87.8%) and tonsil biopsy (30.0%, 95% CLs: 6.7%, 65.3%) was not detected ( Q M = 3.000, P exact = 0.2500).
CWD in a tonsil biopsy The proportion of PrP CWD positive tonsil follicles was estimated by counting the total and positive numbers of follicles present in thin sections of the unbiopsied whole tonsil (N = 66 WTD; ). The total number of whole tonsil follicles counted was highly variable between deer (mean = 126.7, standard deviation = 47.5). The mean of whole tonsil follicle counts was marginally dependent on the animal’s age ( F = 4.83, P = 0.0316); the estimated reduction in mean total follicle count was 4.9 follicles per year. The probability of a false negative tonsil biopsy result in early preclinical deer was not significantly dependent on the whole tonsil total follicle count ( Q LR = 0.3717, P = 0.5421). In contrast, the probability of a false negative tonsil biopsy result in early preclinical deer was significantly dependent on the whole tonsil estimate of the proportion of positive follicles ( Q LR = 30.4393, P < 0.0001; ). The odds of a false negative result based on a two-bite tonsil biopsy from deer in early preclinical infection increased 1.617 (95% CLs: 1.226, 2.603) for each 0.1 unit decrease in positive proportion of whole tonsil follicles .
Early detection of CWD-infected cervids is key to mitigating the spread of disease. Of particular interest is the potential application of antemortem diagnostic testing to farmed WTD. This study determined the sensitivity of CWD IHC using a two-bite biopsy technique reported to produce optimal antemortem retrieval of tonsillar follicles from white-tailed deer under field conditions . In this study, diagnostically adequate numbers of follicles were obtained using this two-bite biopsy sampling technique in 79 preclinical, naturally infected farmed WTD from nine CWD-positive herds from across the United States. The study group included similar proportions of deer with early and late preclinical infections, and each infection stage was similarly represented by males and females, ages ranging from 6 months to 9 years, and the GG and GS genotypes of PRNP codon 96. The contralateral tonsil was collected intact to provide unbiased whole tonsil metrics to better understand factors that may affect the diagnostic sensitivity of tonsillar biopsy. The overall preclinical diagnostic sensitivity of CWD IHC using this unilateral two-bite tonsil biopsy technique was estimated to be 72% whereas the sensitivity from the paired whole tonsil was significantly higher at 87% . Animals with early preclinical infection—a stage defined in this study as official IHC detection of PrP CWD in MRPLN follicles but not the obex of WTD—are notoriously difficult to diagnose antemortem. Thus, even though the diagnostic sensitivity of tonsil biopsy for WTD in late preclinical infection was 92% and was the same as that achieved by examining the whole tonsil, the tonsil biopsy sensitivity for WTD with early preclinical infection was reduced to 55% despite an 83% sensitivity based on the whole tonsil. Furthermore, early preclinical detection was low at 30% in WTD bearing the GS genotype of PRNP codon 96 as compared to 66% detection in GG herd mates. The poor sensitivity of tonsil biopsy during early preclinical infection was strongly associated with the proportion of PrP CWD -positive tonsil follicles as estimated using follicle counts from the unbiopsied whole tonsil. As seen in , the detection of PrP CWD in at least 80% of tonsillar follicles (x-axis) was observed in 31 (or 94%, y-axis) of 33 late preclinical deer and in 14 (42%) of 33 early preclinical deer. Thus, it is not surprising that PrP CWD was detected by tonsil biopsy in all 45 of these deer. But false negative results from tonsil biopsies occurred when tonsil estimates fell below 80% positive follicles. In WTD with early preclinical infection and PrP CWD present in the whole tonsil sample (N = 33), PrP CWD was not detected by tonsil biopsy in two deer with respective estimates of 76% and 28% positive tonsil follicles, and in 9 of 12 (75%) deer in which the tonsil estimates were less than 20% positive follicles (range 4% to 19%). The odds of a false negative biopsy result during early preclinical infection increased by approximately 1.6 for every 10% decrease in the estimated proportion of positive tonsil follicles. As such, the chance of a false negative result from a unilateral two-bite tonsil biopsy was greater than 50% (probability 0.5) when the tonsils of WTD with early preclinical infection were estimated to have 20% or fewer positive follicles. Other sample types and novel detection methods have been studied in naturally infected WTD as potential antemortem tests for CWD and, in each case, diagnostic sensitivity was negatively impacted by the PRNP genotype at codon 96 and for deer during the early stage of infection (all using the same definition as in this study). Deer with GS and SS codon 96 polymorphisms are still susceptible to CWD. However, the amount of PrP CWD staining is significantly less in these animals as demonstrated in a controlled intranasal inoculation . From a meta-analysis of IHC-based diagnosis using RAMALT biopsy , the sensitivity was 68% overall but was only 42% in GS deer and only 36% for deer in early preclinical infection. Newer assay methods detect the misfolding activity associated with prions and have the potential to detect far lower amounts of PrP CWD than are routinely detected by immunoassays, including IHC. In one application of the real-time quaking-induced conversion (RT-QuIC) assay , the sensitivity of this misfolding assay applied to RAMALT biopsies was 69% overall but only 39% in GS deer and only 25% for deer in early preclinical infection. When the protein misfolding cyclic amplification (PMCA) assay was optimized for use with cervid blood samples , the sensitivity was 79% overall but only 57% in GS deer and 53% for deer in early preclinical infection.
While this study demonstrates some potential for using CWD IHC and tonsil biopsy as an antemortem diagnostic in naturally infected farmed WTD, detection was limited during early preclinical infection and in deer bearing the GS genotype at PRNP codon 96. This is not surprising given these same factors have been observed to negatively impact the measured diagnostic sensitivity of other antemortem sample types (e.g., RAMALT and blood), even when tested using protein misfolding assays. Thus, evaluations of CWD IHC applied to a two-bite biopsy sample of the palatine tonsil must also consider the potential impact of these limitations on its intended application.
S1 Table Whole tonsil and tonsil biopsy results for seventy-nine preclinical white-tail deer naturally infected with chronic wasting disease. (XLSX) Click here for additional data file.
|
How has medical student learning changed with the pivot to online delivery of ophthalmology in the pandemic? | 5dd0f4f1-e2f5-4259-937d-ccbe0e860671 | 10062639 | Ophthalmology[mh] | The global lockdown and government mandated social distancing measures enforced following the COVID-19 pandemic required that the clinical attachment Ophthalmology module be delivered online. This pivot to the online environment during the COVID-19 pandemic incorporated the flipped classrooms (FC) approach and offered a student-centred learning environment where the traditional face-to-face teaching pedagogy was transitioned into an online distance module using the Blackboard Collaborate (BBc) platform . A pre pandemic review into the effectiveness of a FC concluded that the approach is a promising teaching strategy that can be utilised to promote learner engagement and motivation , however, there is little evidence to correlate the implementation of the FC and learner outcomes. In this study, the FC approach was adopted not only as a means to deliver content but also because it afforded the opportunity to engage students in this online environment, enhance problem solving skills and to promote deep learning . Overall, in line with Blooms taxonomy and the FC pedagogy our revised module aimed to move from lower order to higher order cognition . Introduction of technology is not an entirely new concept in medical education or ophthalmology, where studies have previously examined the effectiveness of video webcasts , and how digital learning compared to traditional didactic teaching interventions . While there has been increasing interest in the use of a FC approach in medical education and ophthalmology there is controversy in the literature regarding this form of teaching particularly with regard to demonstrating positive changes in knowledge and skills. Several systematic reviews have investigated the efficacy of flipped classrooms , and while many studies reported the effectiveness of the flipped classroom in terms of learner outcomes categorised according to Kirkpatrick’s framework and Blooms taxonomy, the varying methods for implementation have produced varying results . Additionally, there is a paucity of research detailing changes in these aspects with respect to the use of FC to provide a means of providing continued education for ophthalmology students during the pandemic. A review of developments in medical education in response to the COVID-19 pandemic found that seven studies described utilising the FC educational intervention BEME Guide No 63 identified two studies describing the use of a combination of videoconferencing, FC (with question-and-answer time), video review of surgical procedures, and surgical simulators and BEME Guide No 64 identified further studies which described the incorporation of active learning strategies in response to the pandemic including FC . One study by Sud et al described their pivot of undergraduate ophthalmology and positive student feedback from this intervention in terms of ease of accessing material, asynchronous nature and ability to view multiple times. The student feedback also detailed some negative aspects of the intervention related to the diminished interactivity of this approach and the increased time for any doubts to be clarified. However, they do not detail any changes in knowledge gain or skill acquisitions. There is significant interest in, and positives attributed to the FC, however, Diel et al suggest that further work is required to assess the efficacy of the FC approach for ophthalmology in a completely virtual setting, as necessitated by COVID-19 physical distancing constraints . In response to the consequences of the COVID-19 pandemic, we pivoted a didactic and clinical ophthalmology module to an online flipped classroom approach encompassing the same learning objectives for 4th year medical students. We hypothesised that the flipped classroom delivery would result in more engagement with the material, better problem solving and so impact positively on their learning. Thus, as the literature regarding an entirely online flipped classroom approach for ophthalmology is limited our study is timely as it examines the perceptions of both students and faculty with the online FC pedagogy, representing Kirkpatrick–level 1- reaction . Additionally, the evaluation of the impact of introducing this pedagogical model had on knowledge gain was determined, representing Kirkpatrick Level 2- learning . Ethics This study was reviewed and approved by the Research and Ethics Committee (REC) of the RCSI, University of Medicine and Health Sciences and was conducted according to the principles expressed in the Declaration of Helsinki. Written informed consent was be obtained from all participants (REC 202006015). Module description Our objective was to consistently address the learning outcomes despite the change in delivery supporting constructive alignment whilst responding to the changes precipitated the by COVID-19 pandemic regarding social distancing etc (REF Irish Government Guidelines on closing in March 2020). Subjects and study design Study populations All students and faculty involved in the delivery of this module were invited to participate. Participants (243 students) in this study were 4 th year senior cycle medical students enrolled in RCSI on an Ophthalmology clinical attachment that takes place 20 times during the academic year. Traditional delivery (TD) of the 4 th year ophthalmology clinical attachment began on the 23 rd September 2019 in the Royal Victoria Eye and Ear Hospital (RVEEH) and finished 09 th March 2020. As a result of the global pandemic, an online distance module was devised. Students engaged in an online flipped classroom (OFC) which commenced on March 16 th 2020 and finished on the 22 nd May 2020. Faculty members (n = 5) were involved in both TD and OFC between the 23 rd September 2019 and the 22 nd May 2020 . As this is a natural experiment, two groups were spontaneously created; Group 1 = TD and Group 2 = OFC group. Traditional delivery (TD) group (pre COVID-19 usual ophthalmology module) Students (129 students) attended didactic lectures (Red eye, Sudden loss of vision, Gradual loss of vision, Ocular trauma) prior to commencing their clinical attachment. Students were assigned into groups by the SARA (Student, Academic & Regulatory Affairs) office RCSI (10–12 in each) before commencing their clinical attachment week. The traditional ophthalmology module consisted of a review of pre-recorded videos of ophthalmic examinations performed by ophthalmologists in the Royal Victoria Eye and Ear Hospital (RVEEH) on patients and simulated patients. Students attended clinical tutorials (Cataract, Glaucoma, Diabetic Retinopathy, Age-related Macular Degeneration (AMD)), which consisted of 60-min small group teaching sessions (face-to-face lectures with 15-minute question and answer session) led by an Ophthalmologist. Students then engaged in 60-min of patient led teaching which consisted of taking patient history, reading patient charts, examining patients etc. Students also engaged in practical skills sessions (Patient Based Teaching, Clinical Skills, Outpatients Department) and peer led teaching sessions, where students collaboratively presented in groups of 2 or 3 a joint presentation on a topic assigned at the beginning of the week. Knowledge was tested upon completion of the module via a multiple- choice question (MCQ) exam. Clinical Competency (skills) were assessed by practical examination of fundoscopy skills. Online Flipped Classroom (OFC) group (during COVID-19 online ophthalmology module) Students (114 students) were assigned into groups by the SARA office RCSI (12–18 in each) before attending a week long online ophthalmology course facilitated by ophthalmology department from the RVEEH. This week consisted of an online flipped classroom pedagogical approach where students reviewed pre-recorded lecture content and slide sets of the lecture material, which were made available in advance of online interactive classroom sessions. The small group didactic lectures were replaced by PowerPoint presentations with recorded audio. Clinical tutorials which were small group teaching sessions followed by patient encounters in the pre-COVID19 traditional delivery course were converted to interactive BBc tutorials on-line. Specifically students were asked to watch the pre-recorded video lectures (Cataract, Glaucoma, Diabetic Retinopathy, AMD) online in advance of the one hour interactive session led by an Ophthalmologist. This was supplemented by slide sets of the didactic lecture material without audio. After the pre-class lecture, students attended a synchronous, online, live interactive session on the same topic. These BBc sessions included problem solving, clinical vignettes and MCQs relating to the recorded lecture. Facilitators prepared clinical cases and related MCQs that addressed learning outcomes and promoted engagement for use during the interactive online session. The facilitator encouraged problem solving using the poll feature of BBc which encouraged both discussion and active learning. To facilitate a student-centred Peer Assisted Learning (PAL) pedagogical approach students were provided with details on how to prepare PowerPoint presentations with audio for online peer-led teaching sessions. Knowledge was tested upon completion of the module via a multiple- choice question (MCQ) exam. details the Red eye session format for TD versus OFC groups. A similar format was used for all topics delivered as part of the OFC approach. Instrument and data collection The majority of studies investigating perceptions of and satisfaction with FC utilise a questionnaire (either standardised or in house) to garner feedback from the students . The CEQ is utilised annually by universities in Australia and the UK, among undergraduate students, to determine student satisfaction and to identify areas for improvement . There is a significant body of evidence in the literature supporting the reliability and validity of this questionnaire in higher education . Observable differences in respondent scores in diverse fields of study or between institutions offering similar programmes in the same field demonstrates the discriminant validity of the CEQ . In relation to investigations specifically focussing on FC and ophthalmology students Tang, Lin and Zhu et al have utilised the CEQ , which Broomfield et al determined was an appropriate instrument for course evaluation in medical education . To date Paul Ramsden’s CEQ exists in three different formats: the CEQ23; CEQ30; and CEQ36. Each item of the questionnaire is answered using a standard 5-point Likert scale where the levels of agreement ranged from “strongly agree” (scoring a “1”) to “strongly disagree” (scoring a “5”). The CEQ36 measures six constructs established as important learning environment features within the context of higher education : Good teaching (GT) Clear goals and standards (CG) Appropriate assessment (AA) Appropriate workload (AW) Generic skills (GS) Emphasis on Independence (IN) To investigate key stakeholders’ perceptions of and satisfaction with the online flipped classroom, all students (n = 243) and faculty (n = 5) were invited to complete the CEQ36 online via Survey Monkey. Additionally, final anonymised MCQ exam scores were obtained for each student in the study. Statistical analysis Descriptive statistics were used to describe the characteristics of the two groups and Chi-square test/Fisher exact test, or independent samples t test used to explore differences between the groups. The scores of the MCQ final exam were compared using independent samples t test. The questionnaire data given to students were analysed using Mann-Whitney-U tests, to explore potential differences between the groups. Data was collected from each staff member in relation to both teaching methods and hence the staff questionnaire data were analysed using Wilcoxon signed rank tests. No adjustment was made for multiple comparisons. During analysis responses for the ‘agree’ and ‘strongly agree’ categories were combined, similarly responses for ‘disagree’ and ‘strongly disagree’ category were combined. All statistical analyses were performed in GraphPad Prism V5 or Stata v13. Digital training Blackboard Collaborate (BBc) has previously been shown to have utility as a platform to support nursing students placement learning. Several studies have highlighted the importance training to develop student’s digital literacy to facilitate student engagement with this form of technology . To support this guides to the use of BBc were prepared and provided to the students ahead of the online module. Digital training was provided to ophthalmology faculty along with support guides for the use of the BBc platform. This study was reviewed and approved by the Research and Ethics Committee (REC) of the RCSI, University of Medicine and Health Sciences and was conducted according to the principles expressed in the Declaration of Helsinki. Written informed consent was be obtained from all participants (REC 202006015). Our objective was to consistently address the learning outcomes despite the change in delivery supporting constructive alignment whilst responding to the changes precipitated the by COVID-19 pandemic regarding social distancing etc (REF Irish Government Guidelines on closing in March 2020). Study populations All students and faculty involved in the delivery of this module were invited to participate. Participants (243 students) in this study were 4 th year senior cycle medical students enrolled in RCSI on an Ophthalmology clinical attachment that takes place 20 times during the academic year. Traditional delivery (TD) of the 4 th year ophthalmology clinical attachment began on the 23 rd September 2019 in the Royal Victoria Eye and Ear Hospital (RVEEH) and finished 09 th March 2020. As a result of the global pandemic, an online distance module was devised. Students engaged in an online flipped classroom (OFC) which commenced on March 16 th 2020 and finished on the 22 nd May 2020. Faculty members (n = 5) were involved in both TD and OFC between the 23 rd September 2019 and the 22 nd May 2020 . As this is a natural experiment, two groups were spontaneously created; Group 1 = TD and Group 2 = OFC group. All students and faculty involved in the delivery of this module were invited to participate. Participants (243 students) in this study were 4 th year senior cycle medical students enrolled in RCSI on an Ophthalmology clinical attachment that takes place 20 times during the academic year. Traditional delivery (TD) of the 4 th year ophthalmology clinical attachment began on the 23 rd September 2019 in the Royal Victoria Eye and Ear Hospital (RVEEH) and finished 09 th March 2020. As a result of the global pandemic, an online distance module was devised. Students engaged in an online flipped classroom (OFC) which commenced on March 16 th 2020 and finished on the 22 nd May 2020. Faculty members (n = 5) were involved in both TD and OFC between the 23 rd September 2019 and the 22 nd May 2020 . As this is a natural experiment, two groups were spontaneously created; Group 1 = TD and Group 2 = OFC group. Students (129 students) attended didactic lectures (Red eye, Sudden loss of vision, Gradual loss of vision, Ocular trauma) prior to commencing their clinical attachment. Students were assigned into groups by the SARA (Student, Academic & Regulatory Affairs) office RCSI (10–12 in each) before commencing their clinical attachment week. The traditional ophthalmology module consisted of a review of pre-recorded videos of ophthalmic examinations performed by ophthalmologists in the Royal Victoria Eye and Ear Hospital (RVEEH) on patients and simulated patients. Students attended clinical tutorials (Cataract, Glaucoma, Diabetic Retinopathy, Age-related Macular Degeneration (AMD)), which consisted of 60-min small group teaching sessions (face-to-face lectures with 15-minute question and answer session) led by an Ophthalmologist. Students then engaged in 60-min of patient led teaching which consisted of taking patient history, reading patient charts, examining patients etc. Students also engaged in practical skills sessions (Patient Based Teaching, Clinical Skills, Outpatients Department) and peer led teaching sessions, where students collaboratively presented in groups of 2 or 3 a joint presentation on a topic assigned at the beginning of the week. Knowledge was tested upon completion of the module via a multiple- choice question (MCQ) exam. Clinical Competency (skills) were assessed by practical examination of fundoscopy skills. Students (114 students) were assigned into groups by the SARA office RCSI (12–18 in each) before attending a week long online ophthalmology course facilitated by ophthalmology department from the RVEEH. This week consisted of an online flipped classroom pedagogical approach where students reviewed pre-recorded lecture content and slide sets of the lecture material, which were made available in advance of online interactive classroom sessions. The small group didactic lectures were replaced by PowerPoint presentations with recorded audio. Clinical tutorials which were small group teaching sessions followed by patient encounters in the pre-COVID19 traditional delivery course were converted to interactive BBc tutorials on-line. Specifically students were asked to watch the pre-recorded video lectures (Cataract, Glaucoma, Diabetic Retinopathy, AMD) online in advance of the one hour interactive session led by an Ophthalmologist. This was supplemented by slide sets of the didactic lecture material without audio. After the pre-class lecture, students attended a synchronous, online, live interactive session on the same topic. These BBc sessions included problem solving, clinical vignettes and MCQs relating to the recorded lecture. Facilitators prepared clinical cases and related MCQs that addressed learning outcomes and promoted engagement for use during the interactive online session. The facilitator encouraged problem solving using the poll feature of BBc which encouraged both discussion and active learning. To facilitate a student-centred Peer Assisted Learning (PAL) pedagogical approach students were provided with details on how to prepare PowerPoint presentations with audio for online peer-led teaching sessions. Knowledge was tested upon completion of the module via a multiple- choice question (MCQ) exam. details the Red eye session format for TD versus OFC groups. A similar format was used for all topics delivered as part of the OFC approach. The majority of studies investigating perceptions of and satisfaction with FC utilise a questionnaire (either standardised or in house) to garner feedback from the students . The CEQ is utilised annually by universities in Australia and the UK, among undergraduate students, to determine student satisfaction and to identify areas for improvement . There is a significant body of evidence in the literature supporting the reliability and validity of this questionnaire in higher education . Observable differences in respondent scores in diverse fields of study or between institutions offering similar programmes in the same field demonstrates the discriminant validity of the CEQ . In relation to investigations specifically focussing on FC and ophthalmology students Tang, Lin and Zhu et al have utilised the CEQ , which Broomfield et al determined was an appropriate instrument for course evaluation in medical education . To date Paul Ramsden’s CEQ exists in three different formats: the CEQ23; CEQ30; and CEQ36. Each item of the questionnaire is answered using a standard 5-point Likert scale where the levels of agreement ranged from “strongly agree” (scoring a “1”) to “strongly disagree” (scoring a “5”). The CEQ36 measures six constructs established as important learning environment features within the context of higher education : Good teaching (GT) Clear goals and standards (CG) Appropriate assessment (AA) Appropriate workload (AW) Generic skills (GS) Emphasis on Independence (IN) To investigate key stakeholders’ perceptions of and satisfaction with the online flipped classroom, all students (n = 243) and faculty (n = 5) were invited to complete the CEQ36 online via Survey Monkey. Additionally, final anonymised MCQ exam scores were obtained for each student in the study. Descriptive statistics were used to describe the characteristics of the two groups and Chi-square test/Fisher exact test, or independent samples t test used to explore differences between the groups. The scores of the MCQ final exam were compared using independent samples t test. The questionnaire data given to students were analysed using Mann-Whitney-U tests, to explore potential differences between the groups. Data was collected from each staff member in relation to both teaching methods and hence the staff questionnaire data were analysed using Wilcoxon signed rank tests. No adjustment was made for multiple comparisons. During analysis responses for the ‘agree’ and ‘strongly agree’ categories were combined, similarly responses for ‘disagree’ and ‘strongly disagree’ category were combined. All statistical analyses were performed in GraphPad Prism V5 or Stata v13. Blackboard Collaborate (BBc) has previously been shown to have utility as a platform to support nursing students placement learning. Several studies have highlighted the importance training to develop student’s digital literacy to facilitate student engagement with this form of technology . To support this guides to the use of BBc were prepared and provided to the students ahead of the online module. Digital training was provided to ophthalmology faculty along with support guides for the use of the BBc platform. 243 undergraduate medical students who received TD (n = 129) or OFC (n = 114) delivery of the ophthalmology clinical attachment were invited to participate. All students in the TD and OFC attended in person or online, completed the module and MCQ examination. Twenty-three students in the TD group and 28 students in the OFC group agreed to take part in the study and completed an online CEQ. This represents an overall participation rate of 17.8% for the TD group and 24.6% for the OFC group. All faculty (n = 5) delivered both the TD and the OFC module and completed the questionnaire giving a response rate of 100%. The demographic distribution of the participants is presented in . No gender difference was observed between TD and FC groups when compared to the class as a whole (column 1–3). Overall analysis of the demographic data indicated that a greater proportion of younger females participated in the online survey overall and specifically in the TD cohort, while the OFC survey participants were more presentative of the class as a whole. Student perceptions summarizes the responses from the students regarding the six constructs established as important learning environment features within the context of higher education: Good Teaching (GT), Generic Skills (GS), Appropriate Assessment (AA), Appropriate Workload (AW), Clear Goals and Standards (CG), Emphasis on Independence (IN) . We observed significant differences between the responses of the TD and OFC groups regarding the learning experience, perceived value of the flipped classroom, teaching process, and the evaluation system. Good teaching scale There was no evidence of a difference in student perceptions about the amount of time staff spent commenting on their work (Q9) or the effort staff made to understand student difficulties (Q20). Both the TD and OFC group felt that lecturers were good at explaining (Q23) and worked hard to make their topics interesting (Q25). Furthermore, the TD and OFC students felt that faculty had an interest in what students had to say (Q31) and that they tried to get the best out of all the students (Q33). However, compared to the TD group who attended the on-site traditional clinical attachment, students felt that the OFC approach did not motivate students to do their best (Q4, P = 0.01) or provide adequate feedback on how they were doing (Q22, P = 0.041) on the GT scale. Clear goals and standards scale There was no evidence of a difference in student perceptions on the goals and standards (CG) scale specifically about what was expected from them (Q8 & 18), about the objectives of the course (Q24) and faculty expectations of students being made clear (Q35). The OFC group were significantly less satisfied with the CG. Specifically, students felt it was harder to determine what standard of work was expected (Q1, P = 0.046). Generic skills scale There was no evidence of a difference in student perceptions about the capacity for the TD or OFC course to sharpen their analytic skills (Q6), develop their ability to work as a team member (Q11), improve confidence about tackling unfamiliar problems (Q12), improve communication skills (Q13) or develop their ability to plan work (Q28). On the GS scale students felt that the OFC approach was significantly less beneficial in terms of helping develop their problem skills when compared to the TD group (Q2, P = 0.048). Appropriate assessment scale There was no evidence of a difference between the TD and OFC group in student perceptions of the impression faculty gave about learning from students (Q7), that a good memory is all that is required to do well on the course (Q10), and that staff are more interested in testing what students have memorised (Q17). There was also no evidence of a difference between student perceptions from the two groups in relation to asking too many questions about facts (Q26). However, there was difference between the TD and OFC group in their perceptions of the form that feedback was given (Q29) or that just by working hard around exam times they could get through the course (Q32). Appropriate workload scale There was no evidence of a difference in student perceptions in relation to the amount of topics covered in the syllabus (Q14), the amount of time given to learn (Q19), the pressure felt by students (Q27) or how the volume of work affects comprehension of topics (Q36). Interestingly, on the AW scale there was a significant difference in feeling about the workload being too heavy in TD group compared to the OFC group (Q5, P = 0.002). Emphasis on independence scale There was no evidence of a difference in student perceptions between the TD and OFC group on the IN scale regarding opportunities to choose the particular areas you want to study (Q3), that the course encouraged them to pursue their academic interests (Q15) or the opportunities to discuss faculty how they were going to learn in this course (Q30). Analysis of the IN scale indicates that students were dissatisfied with the level of choice afforded by the OFC. Students within the TD group found that they had little choice over how they would learn (Q16, P = 0.018) or over the work they had to do (Q21, P = 0.011). Additionally, students in the OFC group felt that they had little choice in the way they were assessed compared to the TD group (Q34, P > 0.001). Questions regarding the value of the flipped classroom Previous studies have highlighted questions within the CEQ survey, which provide insights into the perceived value of the flipped classroom , indicated by FC in . The FC scale questions overlap with the GT and GS scale; specifically questions 2, 4, 5, 11, 12, 13 and 28. Student survey responses indicated a lack of satisfaction with the online flipped classroom approach. As mentioned above students in the TD group felt that staff did less to motivate them (Q2, P = 0.048) and that there were less opportunities to improve their problem-solving skills (Q4, P = 0.01). While not reaching statistical significance within the FC scale a reduced percentage of students within the OFC group agreed that the online flipped classroom helped develop their ability to work as a team member compared to the TD group (Q11, 52.17% v 25% agree TD v OFC), tackle unfamiliar problems (Q12, 52.18% v 39. 28% agree TD v OFC) or develop their ability to plan their own work (Q28, 21.74% v 39.28% disagree TD v OFC). Overall, study participants showed no preference for traditional delivery of the ophthalmology over the online flipped classroom approach (Q 37). When asked to rate the statement “Overall, I am satisfied with the quality of this course” there was no evidence of a difference in the rating between the TD and the OFC group. Comparison of faculty perspectives between the traditional delivery and the online flipped classroom summarises the feedback from the faculty who taught the didactic and clinical–skills ophthalmology module on site and delivered the OFC ophthalmology module. There was no evidence of differences across the five scales. We compared trends in agreement from Face-to-Face teaching to online flipped classroom teaching for faculty, we defined changes as; agreement increased, stayed the same or decreased from face-to-face to the online approach . On the GS scale, this analysis revealed that more faculty agreed that they put a lot of time into commenting on students work (Q9) and providing feedback (Q22) for the online versus the face-to-face approach. CS scale showed faculty agreed that it was easy to determine the standard of work expected (Q1) while the aims and objectives of this course were not made very clear (Q 24) for the OFC versus the TD approach. Comparison of the GS scale indicated that faculty perceived that the OFC approach supported improvements in student problem solving skills (Q2), analytic skills (Q6) and ability to tackle unfamiliar problems (Q12). An increased number of faculty disagreed that lecturers asked questions specifically about facts (Q26) on the AA scale for OFC approach versus traditional face-to-face teaching. Comparison of trends for the IN scale indicated that faculty felt the online flipped classroom approach had increased opportunities for students to choose particular areas of study (Q3), that the course encouraged students to develop their own academic interests (Q15) and provided more options regarding how students were going to learn on the course (Q16). Comparison of overall student performance on final multiple-choice exam Next we compared students’ exam scores before and after the educational intervention for all students in the TD (n = 129) and OFC groups (n = 114) and students in the TD (n = 23) and OFC groups (n = 28) who responded to the survey. Students answered 20 ophthalmology multiple-choice questions (MCQ) as part of completing the course. Each question had the same weight, and the total score was converted into a 0–100 scale. Independent samples t test was used to compare the differences between the two groups. This analysis of the final exam MCQ score showed that there were no statistical differences between the TD and OFC group (P = 0.0651). Comparison of the final exam MCQ score for survey responders between the TD and OFC found no evidence of a statistical difference in the score achieved. Overall, this indicates that the OFC did not negatively influence knowledge gain. summarizes the responses from the students regarding the six constructs established as important learning environment features within the context of higher education: Good Teaching (GT), Generic Skills (GS), Appropriate Assessment (AA), Appropriate Workload (AW), Clear Goals and Standards (CG), Emphasis on Independence (IN) . We observed significant differences between the responses of the TD and OFC groups regarding the learning experience, perceived value of the flipped classroom, teaching process, and the evaluation system. There was no evidence of a difference in student perceptions about the amount of time staff spent commenting on their work (Q9) or the effort staff made to understand student difficulties (Q20). Both the TD and OFC group felt that lecturers were good at explaining (Q23) and worked hard to make their topics interesting (Q25). Furthermore, the TD and OFC students felt that faculty had an interest in what students had to say (Q31) and that they tried to get the best out of all the students (Q33). However, compared to the TD group who attended the on-site traditional clinical attachment, students felt that the OFC approach did not motivate students to do their best (Q4, P = 0.01) or provide adequate feedback on how they were doing (Q22, P = 0.041) on the GT scale. There was no evidence of a difference in student perceptions on the goals and standards (CG) scale specifically about what was expected from them (Q8 & 18), about the objectives of the course (Q24) and faculty expectations of students being made clear (Q35). The OFC group were significantly less satisfied with the CG. Specifically, students felt it was harder to determine what standard of work was expected (Q1, P = 0.046). Generic skills scale There was no evidence of a difference in student perceptions about the capacity for the TD or OFC course to sharpen their analytic skills (Q6), develop their ability to work as a team member (Q11), improve confidence about tackling unfamiliar problems (Q12), improve communication skills (Q13) or develop their ability to plan work (Q28). On the GS scale students felt that the OFC approach was significantly less beneficial in terms of helping develop their problem skills when compared to the TD group (Q2, P = 0.048). Appropriate assessment scale There was no evidence of a difference between the TD and OFC group in student perceptions of the impression faculty gave about learning from students (Q7), that a good memory is all that is required to do well on the course (Q10), and that staff are more interested in testing what students have memorised (Q17). There was also no evidence of a difference between student perceptions from the two groups in relation to asking too many questions about facts (Q26). However, there was difference between the TD and OFC group in their perceptions of the form that feedback was given (Q29) or that just by working hard around exam times they could get through the course (Q32). Appropriate workload scale There was no evidence of a difference in student perceptions in relation to the amount of topics covered in the syllabus (Q14), the amount of time given to learn (Q19), the pressure felt by students (Q27) or how the volume of work affects comprehension of topics (Q36). Interestingly, on the AW scale there was a significant difference in feeling about the workload being too heavy in TD group compared to the OFC group (Q5, P = 0.002). Emphasis on independence scale There was no evidence of a difference in student perceptions between the TD and OFC group on the IN scale regarding opportunities to choose the particular areas you want to study (Q3), that the course encouraged them to pursue their academic interests (Q15) or the opportunities to discuss faculty how they were going to learn in this course (Q30). Analysis of the IN scale indicates that students were dissatisfied with the level of choice afforded by the OFC. Students within the TD group found that they had little choice over how they would learn (Q16, P = 0.018) or over the work they had to do (Q21, P = 0.011). Additionally, students in the OFC group felt that they had little choice in the way they were assessed compared to the TD group (Q34, P > 0.001). Questions regarding the value of the flipped classroom Previous studies have highlighted questions within the CEQ survey, which provide insights into the perceived value of the flipped classroom , indicated by FC in . The FC scale questions overlap with the GT and GS scale; specifically questions 2, 4, 5, 11, 12, 13 and 28. Student survey responses indicated a lack of satisfaction with the online flipped classroom approach. As mentioned above students in the TD group felt that staff did less to motivate them (Q2, P = 0.048) and that there were less opportunities to improve their problem-solving skills (Q4, P = 0.01). While not reaching statistical significance within the FC scale a reduced percentage of students within the OFC group agreed that the online flipped classroom helped develop their ability to work as a team member compared to the TD group (Q11, 52.17% v 25% agree TD v OFC), tackle unfamiliar problems (Q12, 52.18% v 39. 28% agree TD v OFC) or develop their ability to plan their own work (Q28, 21.74% v 39.28% disagree TD v OFC). Overall, study participants showed no preference for traditional delivery of the ophthalmology over the online flipped classroom approach (Q 37). When asked to rate the statement “Overall, I am satisfied with the quality of this course” there was no evidence of a difference in the rating between the TD and the OFC group. Comparison of faculty perspectives between the traditional delivery and the online flipped classroom summarises the feedback from the faculty who taught the didactic and clinical–skills ophthalmology module on site and delivered the OFC ophthalmology module. There was no evidence of differences across the five scales. We compared trends in agreement from Face-to-Face teaching to online flipped classroom teaching for faculty, we defined changes as; agreement increased, stayed the same or decreased from face-to-face to the online approach . On the GS scale, this analysis revealed that more faculty agreed that they put a lot of time into commenting on students work (Q9) and providing feedback (Q22) for the online versus the face-to-face approach. CS scale showed faculty agreed that it was easy to determine the standard of work expected (Q1) while the aims and objectives of this course were not made very clear (Q 24) for the OFC versus the TD approach. Comparison of the GS scale indicated that faculty perceived that the OFC approach supported improvements in student problem solving skills (Q2), analytic skills (Q6) and ability to tackle unfamiliar problems (Q12). An increased number of faculty disagreed that lecturers asked questions specifically about facts (Q26) on the AA scale for OFC approach versus traditional face-to-face teaching. Comparison of trends for the IN scale indicated that faculty felt the online flipped classroom approach had increased opportunities for students to choose particular areas of study (Q3), that the course encouraged students to develop their own academic interests (Q15) and provided more options regarding how students were going to learn on the course (Q16). There was no evidence of a difference in student perceptions about the capacity for the TD or OFC course to sharpen their analytic skills (Q6), develop their ability to work as a team member (Q11), improve confidence about tackling unfamiliar problems (Q12), improve communication skills (Q13) or develop their ability to plan work (Q28). On the GS scale students felt that the OFC approach was significantly less beneficial in terms of helping develop their problem skills when compared to the TD group (Q2, P = 0.048). There was no evidence of a difference between the TD and OFC group in student perceptions of the impression faculty gave about learning from students (Q7), that a good memory is all that is required to do well on the course (Q10), and that staff are more interested in testing what students have memorised (Q17). There was also no evidence of a difference between student perceptions from the two groups in relation to asking too many questions about facts (Q26). However, there was difference between the TD and OFC group in their perceptions of the form that feedback was given (Q29) or that just by working hard around exam times they could get through the course (Q32). There was no evidence of a difference in student perceptions in relation to the amount of topics covered in the syllabus (Q14), the amount of time given to learn (Q19), the pressure felt by students (Q27) or how the volume of work affects comprehension of topics (Q36). Interestingly, on the AW scale there was a significant difference in feeling about the workload being too heavy in TD group compared to the OFC group (Q5, P = 0.002). There was no evidence of a difference in student perceptions between the TD and OFC group on the IN scale regarding opportunities to choose the particular areas you want to study (Q3), that the course encouraged them to pursue their academic interests (Q15) or the opportunities to discuss faculty how they were going to learn in this course (Q30). Analysis of the IN scale indicates that students were dissatisfied with the level of choice afforded by the OFC. Students within the TD group found that they had little choice over how they would learn (Q16, P = 0.018) or over the work they had to do (Q21, P = 0.011). Additionally, students in the OFC group felt that they had little choice in the way they were assessed compared to the TD group (Q34, P > 0.001). Previous studies have highlighted questions within the CEQ survey, which provide insights into the perceived value of the flipped classroom , indicated by FC in . The FC scale questions overlap with the GT and GS scale; specifically questions 2, 4, 5, 11, 12, 13 and 28. Student survey responses indicated a lack of satisfaction with the online flipped classroom approach. As mentioned above students in the TD group felt that staff did less to motivate them (Q2, P = 0.048) and that there were less opportunities to improve their problem-solving skills (Q4, P = 0.01). While not reaching statistical significance within the FC scale a reduced percentage of students within the OFC group agreed that the online flipped classroom helped develop their ability to work as a team member compared to the TD group (Q11, 52.17% v 25% agree TD v OFC), tackle unfamiliar problems (Q12, 52.18% v 39. 28% agree TD v OFC) or develop their ability to plan their own work (Q28, 21.74% v 39.28% disagree TD v OFC). Overall, study participants showed no preference for traditional delivery of the ophthalmology over the online flipped classroom approach (Q 37). When asked to rate the statement “Overall, I am satisfied with the quality of this course” there was no evidence of a difference in the rating between the TD and the OFC group. summarises the feedback from the faculty who taught the didactic and clinical–skills ophthalmology module on site and delivered the OFC ophthalmology module. There was no evidence of differences across the five scales. We compared trends in agreement from Face-to-Face teaching to online flipped classroom teaching for faculty, we defined changes as; agreement increased, stayed the same or decreased from face-to-face to the online approach . On the GS scale, this analysis revealed that more faculty agreed that they put a lot of time into commenting on students work (Q9) and providing feedback (Q22) for the online versus the face-to-face approach. CS scale showed faculty agreed that it was easy to determine the standard of work expected (Q1) while the aims and objectives of this course were not made very clear (Q 24) for the OFC versus the TD approach. Comparison of the GS scale indicated that faculty perceived that the OFC approach supported improvements in student problem solving skills (Q2), analytic skills (Q6) and ability to tackle unfamiliar problems (Q12). An increased number of faculty disagreed that lecturers asked questions specifically about facts (Q26) on the AA scale for OFC approach versus traditional face-to-face teaching. Comparison of trends for the IN scale indicated that faculty felt the online flipped classroom approach had increased opportunities for students to choose particular areas of study (Q3), that the course encouraged students to develop their own academic interests (Q15) and provided more options regarding how students were going to learn on the course (Q16). Next we compared students’ exam scores before and after the educational intervention for all students in the TD (n = 129) and OFC groups (n = 114) and students in the TD (n = 23) and OFC groups (n = 28) who responded to the survey. Students answered 20 ophthalmology multiple-choice questions (MCQ) as part of completing the course. Each question had the same weight, and the total score was converted into a 0–100 scale. Independent samples t test was used to compare the differences between the two groups. This analysis of the final exam MCQ score showed that there were no statistical differences between the TD and OFC group (P = 0.0651). Comparison of the final exam MCQ score for survey responders between the TD and OFC found no evidence of a statistical difference in the score achieved. Overall, this indicates that the OFC did not negatively influence knowledge gain. The primary aim of this study was to measure stakeholder satisfaction with our usual delivery format, which previously relied on a blend of didactic lectures and clinical skills sessions compared to a revised format, which relied on a FC format to facilitate continued delivery of the ophthalmology module. It was hypothesised that as an educational intervention in response to the COVID-19 pandemic, a FC approach would not only facilitate delivery of content but result in improved or equivalent levels of student satisfaction and knowledge gain as determined by a CEQ and MCQ examination. In pivoting from didactic face-to-face teaching to a solely online delivery of the ophthalmology module, we found that students were satisfied with the quality of this course in both the TD and the OFC groups. However, students in the OFC group were less satisfied with staff motivation of students and provision of feedback, compared to the TD group. OFC students also felt it was harder to determine what standard of work was expected and found the course less beneficial in terms of helping develop their problem-solving skills. Furthermore, students were dissatisfied with the level of choice afforded by the online FC, compared to TD group. Of note, while student responses indicated a lack of satisfaction with the OFC approach compared to traditional face-to-face teaching exam scores were consistent between both groups. Recent reviews have examined developments in medical education interventions in the wake of the pandemic . These reviews found that a significant proportion of interventions focused on pivoting traditional teaching online (53%) with a significant proportion of these studies using the same teaching approaches for face-to-face teaching and their online approach. These reviews noted there was a lack of interventional outcome data in the studies reviewed. More recently, Ferrara et al conducted an online survey to assess the changes ophthalmology residents and fellows have experienced in ophthalmology training related to the current COVID-19 pandemic and collected responses from 32 different countries. Based on the survey responses of ophthalmology trainees’ perspectives the authors suggest that incorporation of technological educational tools, can increase the capacity to cope successfully with the current situation and continued use may improve the effectiveness of training programs in the long term . However, the perspective of Irish ophthalmology students, trainees or fellows was not included in this study. Herein the current study describes an educational intervention and uniquely provides data on the perspectives of those involved: both student and faculty satisfaction and perceptions of an online flipped classroom approach for ophthalmology teaching. We devised a remote synchronous and asynchronous module that was instigated in response to the global COVID-19 pandemic. By adopting an online flipped classroom approach and incorporating elements of Blooms taxonomy we aimed to move away from recall and enhance critical thinking and application of knowledge . Overall, our results are in direct contrast to existing literature regarding the utility of a flipped classroom approach for delivering ophthalmology content to medical students. Previously it has been shown that students preferred the flipped classroom approach to the traditional lecture method as it helped them to develop problem solving, creative thinking and team working skills . However, we found significant dissatisfaction with the online flipped classroom approach among the OFC group. Significant levels of dissatisfaction were observed in problem solving, communication, staff motivation and provision of feedback. Previous groups examining a flipped classroom approach for ophthalmology students have reported higher levels of knowledge gain (increased exam scores) compared to the traditional teaching group . The current study found that exam outcome was consistent with both approaches: the remote flipped classroom approach did not improve knowledge gain for the OFC group compared to the prior didactic and clinical skills-based module. A recent review detailed the types of uncertainty faced by students, including those relating to the educational process and the global coronavirus pandemic, and explored the potential negative impact this can have on learning . Unger et al reported that students experienced a significant level of anxiety in relation to online learning and that this anxiety persisted beyond an initial 3-week practice period . Shahrvini et al similarly reported that students experienced anxiety with the transition of their curriculum online and perceived this mode of delivery had negatively impacted the quality of instruction . Students reported satisfaction with the flexibility of remote learning, however Shahrvini et al suggests that removal of face-to-face teaching and adoption of a solely online program, which leads to digital fatigue, coupled with loss of practical elements of the curriculum contributed to the students negative perceptions towards online learning . They suggest that video casted lectures uploaded in advance, electronic health record and telehealth training for students, in addition to training for teaching faculty to increase technological fluency should be considered to optimise remote learning curricula . A survey conducted by Mishra et al highlights that the COVID-19 lockdown has brought with it, uncertainty, anxiety and higher stress levels among ophthalmology trainees across India due to the disruption of training program schedules . Therefore, anxiety, uncertainty and digital fatigue may be factors contributing to student dissatisfaction with the online flipped classroom approach in our study. Faculty preparedness is another aspect that may have influenced student satisfaction with our educational intervention. Overall, despite literature describing a range of innovative ways to deliver teaching in response to the pandemic, there is relatively little existing in current literature focusing on faculty development or support. Of note the BEME Guide No 63 highlights three studies that describe measures used to support medical educators move to online including adapting existing programs for online delivery, curating resources, providing a webinar where best practices were shared and establishing a twitter community of practice . Furthermore, BEME Guide No 64 : described the faculty development intervention undertaken by Buckley who established a virtual faculty development session to bring together regionally dispersed teaching faculty groups together to learn with and from each other using social networking theory . This virtual faculty development initiative had a range of benefits including supporting faculty to embrace the new challenges of teaching online and importantly providing skills to allow faculty to attend to student well-being, however some issues with its implementation were highlighted. Specifically, that the online webinar did not permit informal side bar conversations, that organisers should consider inclusion of smaller break out rooms to promote discussion along with additional forms of follow up commination eg email . Faculty in RCSI, including those involved in ophthalmology teaching, attended an in-person training course before the closure of the campus, which aimed to increase faculty digital competence with the online delivery platform BBc as we pivoted delivery of our module online. Results from our study indicate that students felt on the ‘Good Teaching Scale’ that the faculty approach to delivery of the OFC did not motivate students to do their best or provide adequate feedback. This resulted in students feeling that the module was overall less beneficial in terms of helping them to develop their problem skills when compared to the TD group. Thus, in line with the main findings of the recent BEME reviews our results confirm that significant additional levels of support are required to support faculty to develop and implement online learning. Additional factors potentially contributing to the overall dissatisfaction with the course include the lack of clinical experiences in the redesigned ophthalmology curriculum, reduced digital efficacy and technical issues. Kostaki et al found that student engagement negatively correlated with technical difficulties and home distractions . Furthermore, they reported that computer self-efficacy positively contributed towards engagement as students could remedy technical difficulties . Brockman et al compared student perceptions of an online or in-person microbiology laboratory and found that while students have positive perceptions of digital online activities there was still a significant preference for a blend of online and in-person practical activities . There has been significant interest in the FC pedagogy a means to continue to deliver education during these extraordinary times . While many researchers have indicated that this approach should be maintained long term in an effort to modernise medical education , Fisher et al advised that the strategy for the flipped classroom is important, as the environment needs to facilitate learning rather than influence learning to be satisfactory for students . Change to curriculum Our investigations and comparison with contemporary literature indicate that students prefer face-to-face teaching with practical elements. While inclusion of online technology within the curriculum is essential to enhance student digital proficiency and future proof against future disruptions, elements of the traditional curriculum are indispensable. Therefore, future developments should include a blend of traditional classroom-based and remote learning approaches combined with in person practical elements including direct patient contact with mitigated risk. Additionally training of both faculty and students will help to increase digital proficiency and engagement as online elements are predicted to be a central feature on medical education developments. Study limitations The results from our investigations should be viewed in light of the limitations of this study. One such limitation is the lack of student participation in the study resulting in poor respondent rates. The ongoing global pandemic at the time of participant recruitment is a potential factor leading to a lack of study respondents. Additionally we have identified the delay from module completion to distribution of the post-clinical attachment survey as another factor influencing study participation. However, by using the widely validated CEQ we believe that this study offers a unique and important reflection of the impact that the pandemic has had on student learning during the pandemic. Another limitation is that that the OFC did not include clinical skills training which a staple in the TD approach is. It was not possible for our department to teach clinical skills online and consequently this aspect of the clinical attachment was not assessed for the OFC groups. Future studies seeking to determine changes in student and learning and student satisfaction with the OFC for delivery of Ophthalmology content should address these limitations by inclusion of our suggested changes to the curriculum. Our investigations and comparison with contemporary literature indicate that students prefer face-to-face teaching with practical elements. While inclusion of online technology within the curriculum is essential to enhance student digital proficiency and future proof against future disruptions, elements of the traditional curriculum are indispensable. Therefore, future developments should include a blend of traditional classroom-based and remote learning approaches combined with in person practical elements including direct patient contact with mitigated risk. Additionally training of both faculty and students will help to increase digital proficiency and engagement as online elements are predicted to be a central feature on medical education developments. The results from our investigations should be viewed in light of the limitations of this study. One such limitation is the lack of student participation in the study resulting in poor respondent rates. The ongoing global pandemic at the time of participant recruitment is a potential factor leading to a lack of study respondents. Additionally we have identified the delay from module completion to distribution of the post-clinical attachment survey as another factor influencing study participation. However, by using the widely validated CEQ we believe that this study offers a unique and important reflection of the impact that the pandemic has had on student learning during the pandemic. Another limitation is that that the OFC did not include clinical skills training which a staple in the TD approach is. It was not possible for our department to teach clinical skills online and consequently this aspect of the clinical attachment was not assessed for the OFC groups. Future studies seeking to determine changes in student and learning and student satisfaction with the OFC for delivery of Ophthalmology content should address these limitations by inclusion of our suggested changes to the curriculum. S1 Appendix Ophthalmology module learning outcomes. (DOCX) Click here for additional data file. |
Effect of a Senior Cardiology Nursing Role on Streamlining Assessment of Emergency Cardiology Presentations During COVID-19: An Observational Study | 797e72a2-e78c-47ad-865f-23c917452b65 | 10063155 | Internal Medicine[mh] | Cardiovascular disease (CVD) affects populations globally, with estimates by the World Health Organization that CVD was responsible for almost 18 million deaths in 2019, making it the number one cause of death worldwide . CVD refers to a number of conditions affecting the heart and blood vessels including coronary heart disease, heart failure, cardiomyopathy, conditions of rhythm and conduction, stroke and transient ischaemic attack, peripheral arterial disease, and aortic disease . CVD is the major cause of health-related burden in the world as measured by disability-adjusted life years, years of life lost, and years lived with disability . In addition, these CVD-related measures have been steadily increasing since 1990 . In 2017–2018, 1.2 million Australian adults suffered from some form of CVD, with CVD being the principle diagnosis for 5.2% of all hospitalisations . Coronary heart disease, heart failure and cardiomyopathy made up 40% of CVD-related hospitalisations and admissions whilst atrial fibrillation added another 12% . In 2018, a total of 343,290 presentations to Australian Emergency Departments (EDs) were coded as diseases of the circulatory system . The impact of cardiac-related problems on acute care settings places an ongoing strain on ED resources . Chest pain due to acute coronary syndrome (ACS) is a common presenting complaint in EDs and is expected to increase as the population ages . ACS is a time-critical event as the management aim is to minimise myocardial ischaemic time, and thus ACS needs to be promptly recognised and prioritised . One form of ACS is non-ST segment elevated myocardial infarction (NSTEMI), which may present with variable electrocardiography (ECG) changes and symptoms with patients requiring a number of diagnostic interventions in the ED . A dedicated ED clinician to coordinate these interventions, expedite the diagnostic process, and facilitate prompt and definitive treatment of patients with ACS has been shown to decrease patients’ myocardial ischaemic time . Studies also show that the involvement of a specialist cardiology nurse results in the efficient management of patients presenting with cardiac problems by ED staff . Specialist nurses are well utilised in health care as an integral part of the multidisciplinary team. A recent literature review identified many positive impacts on patients with heart failure, including a reduction in unplanned admissions to hospital and a reduction in hospital length of stay . In the United Kingdom, a nurse-led cardiac assessment team of 10 senior cardiac nurses located in the ED has been shown to save approximately six bed-days per day of operation, which translated to an annual saving of GBP£400,000 due to expedited cardiology admission and safe discharge . The Gold Coast Hospital and Health Service (GCHHS) in South-East Queensland, Australia, recognised a need to expedite the assessment of patients presenting to the ED with cardiac complaints. A new model of care was developed and formed part of the initial planning stages of the response to the COVID-19 pandemic . The overall aim was to optimise patient flow to either expedite discharge from the ED or facilitate admission to cardiology. A secondary aim was for this process to free up ED medical/nursing time and space for the anticipated increase in respiratory presentations. A new model of care for suspected cardiology presentations to ED was developed that incorporated a new senior nursing role, the Emergency Cardiac Coordinator (ECC) . The evaluation of the ECC role involved a review of the ECC interventions provided and the time from triage to cardiology consult (TTCC) as the primary measure of efficacy of the ECC role. To overcome the confounding effect of COVID-19, TTCC was compared directly between patients seen by the ECC and those not seen by the ECC on days that the ECC was working in the ED. The ECC role was in place from 14 April 2020 to 15 September 2020. The effect of COVID-19 on TTCC is also demonstrated.
This is an observational study assessing the effect of an intervention. The intervention was the introduction of the ECC role in the ED and the effect measured in the population of patients presenting with cardiac symptomology. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines were followed throughout the review (see ). Ethics approval was received by the GCHHS (Ref No. LNR2020QGC66709). Specific objectives of the study were to: 1. Identify the demographic and clinical profiles of patients referred to the ECC. 2. Describe the interventions made by the ECC in the ED. 3. Compare time from ED TTCC for patients who presented to the ED with a cardiac complaint between those with an ECC intervention and those without an ECC intervention. 4. Measure the effect of COVID-19 on TTCC. Setting The ECC role was evaluated at Gold Coast University Hospital (GCUH), a tertiary hospital within the GCHHS, South-East Queensland. GCUH has one of the busiest EDs in Australia with appproximately 350 presentations per day . ECC Role The ECC role was a new nurse-led model of care consisting of a single senior cardiology specialist nurse, based within the ED, that was trialled between 14 April 2020 and 15 September 2020. The aim of the role was to facilitate timely evaluation, treatment and referral for cardiology consultation of patients who presented to the ED with cardiac symptoms. Patients requiring an ECC review were identified by the ECC through monitoring the ED electronic medical record dashboard, or were referred by the ED medical officer, nursing team leader, or the cardiology on call medical officer. Common triggers for referral of suspected cardiac patients included: previously known to a cardiologist, involvement of multiple treating teams, previous cardiac history and complex social issues. Referrals to the ECC also occurred from clinicians in the ED when movement needed to occur due to bed pressures or if a patient had not been reviewed by the cardiology team within one hour. Not all suspected cardiology patients were referred and/or reviewed by the ECC. This was due to either an uncomplicated patient journey through the department or the limited ability of the ECC to see all possible cardiac patients. If multiple patients were referred for review, priority was given to complex patients and those who had been in the ED for the longest period of time. Discharged patients who required follow-up care were either referred to an outpatient cardiology clinic or referred back to their general practitioner. Referral to outpatient testing was followed up by the consulting cardiology team. The ECC role was in place from 07:30 to 15:30 during weekdays. There was no back-up cover for leave days, after-hours, or weekends. Data Sources Two data sources were used: 1. a spreadsheet completed by the ECC, and 2. the GCHHS integrated electronic Medical Record (ieMR) system. The data recorded by each are described below. Spreadsheet completed by the ECC A spreadsheet was used by the ECC to record the profile of each patient seen and the particular interventions provided by the ECC for that patient. Included the following: - Patient demographics: Patient unique identifier, age, gender, co-morbidities, presenting complaint, diagnosis. - Interventions per patient, classified as: • Comprehensive chart review including current presentation details, past medical history, current medications, previous investigations, and previous hospital admissions • Patient cardiac assessment involving review of physical signs and symptoms related to the cardiac presentation • Patient education about cardiac disease and risk factor modification • Clinical care of patient—Performing ECGs, pathology, observation measurement, clinical assessments, involvement in interventions e.g., cardioversion • Liaise between treating teams and admitting inpatient unit • Organise direct admission without face-to-face review by cardiac team to Cardiac Catheter Suite • Organise direct admission without face-to-face review by cardiac team to Cardiology inpatient unit • Transfer patient to Cardiology inpatient unit, Coronary Care Unit, Cardiac Catheter Suite • Identify need for and fast-track diagnostic testing • Medication review and reconciliation of discrepancies between medical plan and charted medications • Ensure appropriate medical plan in place • Escalation of care (to medical specialty, address barriers in movement, referral to other services) • Organise outpatient testing The GCHHS integrated electronic Medical Record (ieMR) system Cardiac-related presentations to ED were extracted and included all patients who presented during the six-month period the ECC role was in place, as well as eight months before and five months after. A cardiac presentation was defined as a subsequent diagnosis recorded as an International Classification of Diseases (ICD) Diagnosis of Diseases of Circulatory System, excluding cerebrovascular disease, or the patient was admitted to the Coronary Care Unit, Cardiology or Cardiothoracic Surgery departments, or was recorded as having had a cardiology consultation. Data included details of the ED presentation, including the presenting complaint, corresponding inpatient admission and, if the patient had surgery associated with that admission, details of the surgery. If the patient was discharged and re-admitted for surgery, this was not included as this was regarded as a new admission. Any episode of care change relating to a previous admission was not included. The following information was obtained: • Patient unique identifier, age, date of birth and gender • Reason for ED presentation • Time from triage to cardiology consultation (TTCC). These data were used to explore the following: a. TTCC was used as a measure of the efficacy of the ECC. To account for the contemporaneous effect of COVID-19 of TTCC, the effect of the ECC on TTCC was estimated by comparing the median TTCC of patients seen by the ECC with that of patients not seen by the ECC on days the ECC was in the ED. b. ED presentations in the eight months before the implementation of the ECC role and five months after were counted and cardiology-related presentations used to calculate TTCC. These data were used to determine the impact of COVID-19 on TTCC and the number of ED presentations per month, thus providing context for the effect of the ECC. Data Analysis The population of patients seen by the ECC was described by the median and interquartile range (IQR) for age after assessing normality by visual inspection and Shapiro-Wilk test. The distribution of ages was also represented by numbers and percentages per decade of life. Gender, co-morbidities, and presenting complaint were summarised as percentages. Time from ED triage to cardiology consultation (TTCC) was used as a measure of the effect of the ECC position. However, the ECC position began during the COVID pandemic which, it was anticipated, would influence TTCC due to increased presentations to ED. Thus, to describe the context into which the ECC was working we utilised an interrupted time series analysis (ITS) based on piecewise linear regression of the median weekly TTCC for all ED cardiology-referred patients from September 2019 to March 2021 with 26 March 2020 being marked as the beginning of the rise of presentations due to the COVID-19 pandemic. The ITS enables the immediate and ongoing effects of an event (COVID) to be estimated by modelling the before/after effect of COVID and the COVID×Time interaction effect. In addition, a graphical representation of these effects is produced. To directly estimate the effect of the ECC on TTCC we compared patients seen by the ECC to those not seen by the ECC on days the ECC worked. Comparisons were made by the two-sample Wilcoxon rank-sum (Mann–Whitney) test following an assessment of non-normality by the Shapiro-Wilk test. Analyses were performed overall and within each presentation category. All analyses were performed using Stata 17 (Stat Corp., College Station, TX, USA).
The ECC role was evaluated at Gold Coast University Hospital (GCUH), a tertiary hospital within the GCHHS, South-East Queensland. GCUH has one of the busiest EDs in Australia with appproximately 350 presentations per day .
The ECC role was a new nurse-led model of care consisting of a single senior cardiology specialist nurse, based within the ED, that was trialled between 14 April 2020 and 15 September 2020. The aim of the role was to facilitate timely evaluation, treatment and referral for cardiology consultation of patients who presented to the ED with cardiac symptoms. Patients requiring an ECC review were identified by the ECC through monitoring the ED electronic medical record dashboard, or were referred by the ED medical officer, nursing team leader, or the cardiology on call medical officer. Common triggers for referral of suspected cardiac patients included: previously known to a cardiologist, involvement of multiple treating teams, previous cardiac history and complex social issues. Referrals to the ECC also occurred from clinicians in the ED when movement needed to occur due to bed pressures or if a patient had not been reviewed by the cardiology team within one hour. Not all suspected cardiology patients were referred and/or reviewed by the ECC. This was due to either an uncomplicated patient journey through the department or the limited ability of the ECC to see all possible cardiac patients. If multiple patients were referred for review, priority was given to complex patients and those who had been in the ED for the longest period of time. Discharged patients who required follow-up care were either referred to an outpatient cardiology clinic or referred back to their general practitioner. Referral to outpatient testing was followed up by the consulting cardiology team. The ECC role was in place from 07:30 to 15:30 during weekdays. There was no back-up cover for leave days, after-hours, or weekends.
Two data sources were used: 1. a spreadsheet completed by the ECC, and 2. the GCHHS integrated electronic Medical Record (ieMR) system. The data recorded by each are described below. Spreadsheet completed by the ECC A spreadsheet was used by the ECC to record the profile of each patient seen and the particular interventions provided by the ECC for that patient. Included the following: - Patient demographics: Patient unique identifier, age, gender, co-morbidities, presenting complaint, diagnosis. - Interventions per patient, classified as: • Comprehensive chart review including current presentation details, past medical history, current medications, previous investigations, and previous hospital admissions • Patient cardiac assessment involving review of physical signs and symptoms related to the cardiac presentation • Patient education about cardiac disease and risk factor modification • Clinical care of patient—Performing ECGs, pathology, observation measurement, clinical assessments, involvement in interventions e.g., cardioversion • Liaise between treating teams and admitting inpatient unit • Organise direct admission without face-to-face review by cardiac team to Cardiac Catheter Suite • Organise direct admission without face-to-face review by cardiac team to Cardiology inpatient unit • Transfer patient to Cardiology inpatient unit, Coronary Care Unit, Cardiac Catheter Suite • Identify need for and fast-track diagnostic testing • Medication review and reconciliation of discrepancies between medical plan and charted medications • Ensure appropriate medical plan in place • Escalation of care (to medical specialty, address barriers in movement, referral to other services) • Organise outpatient testing The GCHHS integrated electronic Medical Record (ieMR) system Cardiac-related presentations to ED were extracted and included all patients who presented during the six-month period the ECC role was in place, as well as eight months before and five months after. A cardiac presentation was defined as a subsequent diagnosis recorded as an International Classification of Diseases (ICD) Diagnosis of Diseases of Circulatory System, excluding cerebrovascular disease, or the patient was admitted to the Coronary Care Unit, Cardiology or Cardiothoracic Surgery departments, or was recorded as having had a cardiology consultation. Data included details of the ED presentation, including the presenting complaint, corresponding inpatient admission and, if the patient had surgery associated with that admission, details of the surgery. If the patient was discharged and re-admitted for surgery, this was not included as this was regarded as a new admission. Any episode of care change relating to a previous admission was not included. The following information was obtained: • Patient unique identifier, age, date of birth and gender • Reason for ED presentation • Time from triage to cardiology consultation (TTCC). These data were used to explore the following: a. TTCC was used as a measure of the efficacy of the ECC. To account for the contemporaneous effect of COVID-19 of TTCC, the effect of the ECC on TTCC was estimated by comparing the median TTCC of patients seen by the ECC with that of patients not seen by the ECC on days the ECC was in the ED. b. ED presentations in the eight months before the implementation of the ECC role and five months after were counted and cardiology-related presentations used to calculate TTCC. These data were used to determine the impact of COVID-19 on TTCC and the number of ED presentations per month, thus providing context for the effect of the ECC.
A spreadsheet was used by the ECC to record the profile of each patient seen and the particular interventions provided by the ECC for that patient. Included the following: - Patient demographics: Patient unique identifier, age, gender, co-morbidities, presenting complaint, diagnosis. - Interventions per patient, classified as: • Comprehensive chart review including current presentation details, past medical history, current medications, previous investigations, and previous hospital admissions • Patient cardiac assessment involving review of physical signs and symptoms related to the cardiac presentation • Patient education about cardiac disease and risk factor modification • Clinical care of patient—Performing ECGs, pathology, observation measurement, clinical assessments, involvement in interventions e.g., cardioversion • Liaise between treating teams and admitting inpatient unit • Organise direct admission without face-to-face review by cardiac team to Cardiac Catheter Suite • Organise direct admission without face-to-face review by cardiac team to Cardiology inpatient unit • Transfer patient to Cardiology inpatient unit, Coronary Care Unit, Cardiac Catheter Suite • Identify need for and fast-track diagnostic testing • Medication review and reconciliation of discrepancies between medical plan and charted medications • Ensure appropriate medical plan in place • Escalation of care (to medical specialty, address barriers in movement, referral to other services) • Organise outpatient testing
Cardiac-related presentations to ED were extracted and included all patients who presented during the six-month period the ECC role was in place, as well as eight months before and five months after. A cardiac presentation was defined as a subsequent diagnosis recorded as an International Classification of Diseases (ICD) Diagnosis of Diseases of Circulatory System, excluding cerebrovascular disease, or the patient was admitted to the Coronary Care Unit, Cardiology or Cardiothoracic Surgery departments, or was recorded as having had a cardiology consultation. Data included details of the ED presentation, including the presenting complaint, corresponding inpatient admission and, if the patient had surgery associated with that admission, details of the surgery. If the patient was discharged and re-admitted for surgery, this was not included as this was regarded as a new admission. Any episode of care change relating to a previous admission was not included. The following information was obtained: • Patient unique identifier, age, date of birth and gender • Reason for ED presentation • Time from triage to cardiology consultation (TTCC). These data were used to explore the following: a. TTCC was used as a measure of the efficacy of the ECC. To account for the contemporaneous effect of COVID-19 of TTCC, the effect of the ECC on TTCC was estimated by comparing the median TTCC of patients seen by the ECC with that of patients not seen by the ECC on days the ECC was in the ED. b. ED presentations in the eight months before the implementation of the ECC role and five months after were counted and cardiology-related presentations used to calculate TTCC. These data were used to determine the impact of COVID-19 on TTCC and the number of ED presentations per month, thus providing context for the effect of the ECC.
The population of patients seen by the ECC was described by the median and interquartile range (IQR) for age after assessing normality by visual inspection and Shapiro-Wilk test. The distribution of ages was also represented by numbers and percentages per decade of life. Gender, co-morbidities, and presenting complaint were summarised as percentages. Time from ED triage to cardiology consultation (TTCC) was used as a measure of the effect of the ECC position. However, the ECC position began during the COVID pandemic which, it was anticipated, would influence TTCC due to increased presentations to ED. Thus, to describe the context into which the ECC was working we utilised an interrupted time series analysis (ITS) based on piecewise linear regression of the median weekly TTCC for all ED cardiology-referred patients from September 2019 to March 2021 with 26 March 2020 being marked as the beginning of the rise of presentations due to the COVID-19 pandemic. The ITS enables the immediate and ongoing effects of an event (COVID) to be estimated by modelling the before/after effect of COVID and the COVID×Time interaction effect. In addition, a graphical representation of these effects is produced. To directly estimate the effect of the ECC on TTCC we compared patients seen by the ECC to those not seen by the ECC on days the ECC worked. Comparisons were made by the two-sample Wilcoxon rank-sum (Mann–Whitney) test following an assessment of non-normality by the Shapiro-Wilk test. Analyses were performed overall and within each presentation category. All analyses were performed using Stata 17 (Stat Corp., College Station, TX, USA).
A total of 378 patients were reviewed by the ECC . The median patient age was 66 years (IQR 53, 76; youngest 17 yr; oldest 101 yr). Of these patients, 58.5% were male. Most patients reviewed presented with chest pain (269/378; 71.2%) which was either the only complaint or occurred in combination with other cardiac symptoms. The most common co-morbidities were pre-existing hypertension, 46.0% (174/378), a history of ischaemic heart disease, 31.2% (118/378), and high cholesterol, 27.8% (105/378). Out of the 378 patients reviewed, 112 patients went on to receive a cardiology consult. The remaining 266 patients were either discharged home or received other specialty reviews. ECC Interventions The ECC provided 1,727 interventions over the approximate five-month period . Almost all patients received a comprehensive chart review, 70.4% (266/378) received patient-centred cardiac disease education including risk factor optimisation and 68.8% (260/378) a targeted cardiac assessment. Over the 122 days the ECC worked the services provided equated to: • 14.2 interventions per day • 4.6 interventions per patient • 3.1 patients consulted per day. Time From Triage to Cardiology Consultation (TTCC) The period in which the ECC was in operation (see ) coincided with the initial stage of the COVID-19 pandemic in Queensland, Australia. The pandemic precipitated a massive increase in ED presentations from approximately 16,000 per month prior to the pandemic to a peak of almost 28,000 in August 2020 ( and ). Prior to COVID-19 the median weekly TTCC had been decreasing by approximately 0.015 hours per week (95% CI: 0.034, 0.003) but following COVID-19 this changed to a sustained increase of approximately 0.015 hours per week . The ITS model identified this as a change of 0.029 hours (1.74 minutes) per week (95%CI: 0.008, 0.051 hours; p=0.008) with an instantaneous decrease of 1.07 hours (95% CI: 1.73, 0.410 hours; p=0.002). As such it was difficult to assess the independent effect of the ECC on TTCC in this period, particularly as the ECC was only able to see approximately one third of the patients when present in the ED and only worked weekday day shifts. However, as there were more cardiology-related ED presentations than were possible for the ECC to see, we were able to directly compare TTCC in patients who the ECC saw to those who the ECC did not see on the days the ECC worked. This analysis, though it has limitations (see Discussion ), was considered least likely to be affected by bias favouring the ECC compared to other possible analyses, e.g., including patients from days the ECC did not work. (A) shows the distribution of presenting complaints between patients seen by the ECC and those not seen by the ECC on days the ECC worked (P=0.07). (B) provides a summary of the comparison of the TTCC between patients seen by the ECC and those not seen by the ECC on days the ECC worked for patients who presented to the ED and were referred for a cardiac consult. For the 325 patients who had a cardiac consult, the median TTCC was 2.07 hours for the 112 patients seen by the ECC compared to 2.58 hours for the 213 patients not seen by the ECC (p=0.007). Of the 176 patients who presented with chest pain, the median TTCC was 1.94 hours for the 72 patients seen by the ECC compared to 2.41 hours for those not seen by the ECC (p=0.06). The median TTCC for patients presenting with non-cardiac primary symptoms was most affected by ECC intervention; 1.77 hours compared to 3.05 hours (P=0.004). Presentations with palpitations/abnormal heart rate, respiratory distress, and altered level of consciousness had similar TTCCs.
The ECC provided 1,727 interventions over the approximate five-month period . Almost all patients received a comprehensive chart review, 70.4% (266/378) received patient-centred cardiac disease education including risk factor optimisation and 68.8% (260/378) a targeted cardiac assessment. Over the 122 days the ECC worked the services provided equated to: • 14.2 interventions per day • 4.6 interventions per patient • 3.1 patients consulted per day.
The period in which the ECC was in operation (see ) coincided with the initial stage of the COVID-19 pandemic in Queensland, Australia. The pandemic precipitated a massive increase in ED presentations from approximately 16,000 per month prior to the pandemic to a peak of almost 28,000 in August 2020 ( and ). Prior to COVID-19 the median weekly TTCC had been decreasing by approximately 0.015 hours per week (95% CI: 0.034, 0.003) but following COVID-19 this changed to a sustained increase of approximately 0.015 hours per week . The ITS model identified this as a change of 0.029 hours (1.74 minutes) per week (95%CI: 0.008, 0.051 hours; p=0.008) with an instantaneous decrease of 1.07 hours (95% CI: 1.73, 0.410 hours; p=0.002). As such it was difficult to assess the independent effect of the ECC on TTCC in this period, particularly as the ECC was only able to see approximately one third of the patients when present in the ED and only worked weekday day shifts. However, as there were more cardiology-related ED presentations than were possible for the ECC to see, we were able to directly compare TTCC in patients who the ECC saw to those who the ECC did not see on the days the ECC worked. This analysis, though it has limitations (see Discussion ), was considered least likely to be affected by bias favouring the ECC compared to other possible analyses, e.g., including patients from days the ECC did not work. (A) shows the distribution of presenting complaints between patients seen by the ECC and those not seen by the ECC on days the ECC worked (P=0.07). (B) provides a summary of the comparison of the TTCC between patients seen by the ECC and those not seen by the ECC on days the ECC worked for patients who presented to the ED and were referred for a cardiac consult. For the 325 patients who had a cardiac consult, the median TTCC was 2.07 hours for the 112 patients seen by the ECC compared to 2.58 hours for the 213 patients not seen by the ECC (p=0.007). Of the 176 patients who presented with chest pain, the median TTCC was 1.94 hours for the 72 patients seen by the ECC compared to 2.41 hours for those not seen by the ECC (p=0.06). The median TTCC for patients presenting with non-cardiac primary symptoms was most affected by ECC intervention; 1.77 hours compared to 3.05 hours (P=0.004). Presentations with palpitations/abnormal heart rate, respiratory distress, and altered level of consciousness had similar TTCCs.
A specialist nurse role in the ED, the ECC, was introduced to better facilitate and optimise the patient’s journey from presentation to discharge. The ECC reviewed 378 patients from 14 April to 15 September 2020. Most patients reviewed by the ECC presented with chest pain with the most common co-morbidities being hypertension followed by ischaemic heart disease and high cholesterol. The ECC provided 1,727 interventions, with almost all patients having a comprehensive chart review and most receiving education, cardiac assessments and facilitated liaison between treatment teams. On average, the ECC reviewed three patients and provided 14 interventions per day. The primary measure used to assess the effectiveness of the ECC was the time from triage to cardiology consult (TTCC) though this, unfortunately, did not capture the utility of the ECC in identifying individuals who, though presenting with possible cardiac signs, did not require a specialist cardiology consultation. The ECC role coincided with the COVID-19 pandemic which was associated with an overall, and sustained, increase in TTCC. However, for patients who required a cardiology consult, the ECC was seen to expedite the TTCC. This difference was particularly apparent for patients presenting with non-cardiac symptoms or chest pain. The results from this study is similar to those reported in a recent systematic review of 26 studies on nurse-initiated interventions which demonstrated decreased time to treatment and decreased hospital length of stay . That is, an experienced senior nurse with a focus on cardiology assessments, ordering of relevant tests, and liaison between ED and cardiology teams, allowed a more streamlined and efficient patient journey from ED to the cardiology team. The introduction of the ECC role resulted in an overall decreased TTCC and hence time spent within ED primarily due to earlier consultations by the cardiac team. There is evidence that decreased length of stay in ED leads to improved patient outcomes. An observational study for instance, showed a positive correlation between increased ED length of stay and greater total inpatient length of stay . Prolonged ED length of stay has also been linked to patient safety risks, including increased mortality . Of specific significance is the reduction in the time patients presenting with chest pain spent within the ED. It is possible that this was due to expedited diagnoses of NSTEMI following expert ECG assessment and prompt following up of required diagnostic tests in the ED. Patients who present with non-obvious cardiac signs and symptoms are also a challenge in the ED. We found that these patients, who ultimately required a cardiology consult, were also seen quicker by a cardiologist when the ECC was involved. The time to cardiac consultation did not change for patients who presented with palpitations, respiratory distress and altered level of consciousness. This could be due to ECC interventions occurring relatively late when patients present with palpitations as the cardiology team only becomes involved if symptoms are refractory to initial medications provided by the ED clinicians. This is similar for patients presenting with altered level of consciousness. Patients who presented with respiratory distress were likely tested for COVID-19. This involved the use of an isolation room and the requirement for personal protective equipment, hence there were delays in their initial referral to speciality areas that could not be avoided by ECC intervention. Most of the patients reviewed by the ECC presented with chest pain and, in terms of patient co-morbidities, hypertension, pre-existing ischaemic heart disease and high cholesterol were most prevalent. Evidence has shown that patients presenting to ED with chest pain and several cardiac risk factors have an increased likelihood of an ACS . The patient co-morbidity and risk factor data will be used to inform the future development of an ED cardiac prioritisation tool. Previous research that involved risk stratification of chest pain patients in the ED by a specialised nurse through a biomarker protocol showed it was effective and safe in deciding which patients could be safely discharged home . There are a number of limitations with this study. One limitation was that it was not possible to effectively measure the number of patients who were safely sent home or referred to a specialty other than cardiology due to ECC intervention. That is, the number of unnecessary cardiac consultations that were saved. Another limitation was that the ECC was introduced in March 2020 which coincided with the COVID-19 pandemic. The massive increase in ED presentations seen at GCUH was similar to increases reported in other Australian settings and was associated with increased TTCC. As TTCC was used as the primary measure of the effectiveness of the ECC role there was a risk that any ECC effect could be overwhelmed by the COVID-19 effect. To address this, as the ECC could only see approximately 35% of patients presenting with cardiac signs and symptoms on the days the ECC worked, we directly compared TTCC between the patients the ECC saw to those the ECC did not see on these days. This provided a biased measure of the effectiveness of the ECC role as it was likely that the most complex cases would have been prioritised to the ECC. This measure, also, did not estimate the overall effect of the presence of the ECC in the ED. However, as the ECC demonstrated significant decreases in TTCC for presentations such as chest pain and non-cardiac symptomology, and that only just over a third of potential patients were seen by the ECC, there is scope for an expanded role of ECC(s) and, possibly, a more targeted triaging to the ECC would positively impact ED-cardiac workflows. Another limitation included some inconsistences in definitions used in the classification of presenting complaints while the recording of co-morbidities in patients’ medical records varied in quality and standardisation, which may have contributed to low reporting of certain factors such as family history and obesity.
The introduction of a specialist cardiology nurse in the ED led to an overall improvement in the time patients were seen by the cardiology team. This impact was most noticeable for patients who presented with chest pain and non-cardiac symptoms. Data collected on patients’ presentations and cardiac risk factors will be used to inform the ongoing development of the ECC role and an ED prioritisation system to guide decision making and referral of patients for ECC review. The ECC provided interventions with the majority of patients receiving comprehensive chart reviews, targeted cardiac assessments, education on cardiac disease management and risk factor modification. The ECC role also facilitated liaison between clinical teams which resulted to decreased time spent in the ED through reduced TTCC. The introduction of the ECC role resulted in improved clinical handover between ED and cardiology teams, thereby streamlining patients’ hospital journey. Preliminary results from this study provided evidence for ongoing funding and implementation of this role with implications on future practice. Future research should explore the impact of advanced ED nursing roles on patients’ length of stay, health cost and patient outcomes such as morbidity and mortality.
The authors do not have any conflicts of interest.
|
Rare germline variants in pancreatic cancer and multiple primary cancers: an autopsy study | 90ee17ae-8170-4a2f-9b76-0791c8ac5f57 | 10063194 | Forensic Medicine[mh] | Most patients with pancreatic cancer are sporadic ( ) and the incidence of pancreatic cancer is estimated to increase further. The risk factors for pancreatic cancer include sex, smoking, alcohol consumption, diabetes and family history. Pancreatic cancer is refractory and has a low 5-year survival rate. The etiology of pancreatic cancer is not fully understood. The close agreement between the incidence and mortality rates indicates that only a small fraction of patients are detected in the early stages of efficient intervention. These features highlight the need for effective detection methods for high-risk populations ( ; ). However, little is known about germline variants that predispose to pancreatic cancer. Recent genetic analyses suggest that more than 5–10% of patients with pancreatic cancer carry rare germline variants such as ATM, BRCA1, BRCA2 and Lynch syndrome (LS)-related four DNA mismatch repair (MMR) genes ( MLH1, MSH2, MSH6 and PMS2 ) ( ; ; ; ; ). GWAS studies have been reported in patients with pancreatic cancer ( ; ; ; ; ). However, information regarding germline genes for pancreatic cancer in Japanese individuals without a family history is lacking. Here, we aimed to provide information on rare germline variants in the elderly with pancreatic cancer, multiple primary cancers and non-cancers, using 61 gene panel sequencing. Most cancers occur in the elderly, imposing great public health and socioeconomic burden ( ). This case-control study is based on a database of consecutive autopsies registered for a study of geriatric disease, a design that characterizes the risk of pancreatic cancer in elderly individuals with a negative family history. Furthermore, this study is unique in that it includes centenarians of the same age and sex in the case and control groups.
Patient selection and ethics committees Subjects for the case-control study were collected from consecutive autopsy samples of elderly individuals in Japan, referred to as JG-SNP: Japanese single nucleotide polymorphism for geriatric research ( ) for the study of geriatric diseases. Autopsy samples were obtained from patients with pancreatic cancer and multiple primary cancers and those without cancer who were enrolled at Tokyo Metropolitan Geriatric Hospital between 1995 and 2012 and had no relevant family history. The control group consisted of non-cancer patients whose sex, age, smoking and alcohol consumption matched the patients in the case group as closely as possible. Patient medical records were referenced to obtain clinical data. Written informed consent was obtained from a family member of each patient involved in this study for the collection and research. This study was approved by the ethics committees of Tokyo Medical and Dental University (approval No. 02016-011), Tokyo Metropolitan Geriatric Hospital (approval No. R16-55) and Juntendo University (approval No. M18-0240). DNA extraction, quantification and quality control DNA was extracted from the kidneys of autopsy samples using a standard phenol/chloroform extraction method. The purity and concentration of the DNA extracted from all samples were measured using a NanoDrop 1000 spectrophotometer (Thermo Scientific, Wilmington, Delaware, USA) and a Qubit fluorometer (Life Technologies, Carlsbad, California, USA). All the samples fulfilled the minimum quality and purity requirements of 1 mg in less than a 130-ml volume, with an optical density OD260/280 of 1.7–1.9 and an OD260/230 of greater than 2. DNA fragmentation was assessed by agarose gel (2%) electrophoresis. Panel sequencing The 61 selected genes have previously been reported to be associated or potentially associated with cancer and are used by healthcare providers at our medical institution to predict the risk of cancer. AmpliSeq Custom panel was designed with AmpliSeq 7.0.2 (Thermo Fisher Scientific, Waltham, Massachusetts, USA) that covers exons, exon-intron boundaries, 5′UTR and 3′UTR regions of AKT1 ( ; ) , APC ( ) , ATM ( ) , AXIN2 ( ) , BARD1 ( ) , BMPR1A ( ) , BRCA1, BRCA2 ( ) , BRIP1 ( ; ) , BUB1 ( ) , BUB1B ( ; ) , BUB3 ( ) , CDH1 ( ) , CDKN1B ( ; ) , CHEK2 ( ) , CNTN6 ( ) , ENG ( ; ) , EPCAM ( ; ) , EPHX1 ( ; ) , FANCC ( ) , FANCE ( ) , FAN1 ( ), GALNT12 ( ; ) , GREM1 ( ; ) , LRP6 ( ) , MBD4 ( ) , MCM9 ( ) , MYH11 ( ) , MLH1, MLH3, MSH2, MSH3, MSH6 ( ; ; ; ; ; ) , MUTYH ( ; ) , NFKBIZ ( ; ) , NTHL1 ( ) , PIK3CA ( ; ) , PMS1 ( ) , PMS2 ( ) , POLD1 ( ; ) , POLE ( ) , POLQ ( ; ) , PTEN ( ) , RAD52 ( ; ) , RBL1 ( ; ; ) , REV3L ( ) , RNF43 ( ) , RPS20 ( ) , SCG5 ( ) , SDHB ( ) , SDHD ( ) , SMAD4 ( ) , SMAD9 ( ) , SMARCA4 ( ; ; ) , STK11 ( ), TDRD3 ( ) , TGFBR2 ( ) , TP53 ( ) , UIMC1 ( ) , XAF1 ( ) and XRCC4 ( ) genes, together with upstream regions of GREM1, APC and MSH2 ( ). We used an Ion Chef instrument with Ion 520 and 530 kits (Thermo Fisher Scientific) for emulsion PCR, bead enrichment, and chip loading onto the Ion 530 chip (Thermo Fisher Scientific). The loaded chips were sequenced on an Ion GeneStudio S5 Plus sequencer (Thermo Fisher Scientific). The data obtained were analyzed using the Ion Reporter server 5.10 (Thermo Fisher Scientific). The variants were visually inspected using the Integrative Genomics Viewer [(IGV) Broad Institute].
Subjects for the case-control study were collected from consecutive autopsy samples of elderly individuals in Japan, referred to as JG-SNP: Japanese single nucleotide polymorphism for geriatric research ( ) for the study of geriatric diseases. Autopsy samples were obtained from patients with pancreatic cancer and multiple primary cancers and those without cancer who were enrolled at Tokyo Metropolitan Geriatric Hospital between 1995 and 2012 and had no relevant family history. The control group consisted of non-cancer patients whose sex, age, smoking and alcohol consumption matched the patients in the case group as closely as possible. Patient medical records were referenced to obtain clinical data. Written informed consent was obtained from a family member of each patient involved in this study for the collection and research. This study was approved by the ethics committees of Tokyo Medical and Dental University (approval No. 02016-011), Tokyo Metropolitan Geriatric Hospital (approval No. R16-55) and Juntendo University (approval No. M18-0240).
DNA was extracted from the kidneys of autopsy samples using a standard phenol/chloroform extraction method. The purity and concentration of the DNA extracted from all samples were measured using a NanoDrop 1000 spectrophotometer (Thermo Scientific, Wilmington, Delaware, USA) and a Qubit fluorometer (Life Technologies, Carlsbad, California, USA). All the samples fulfilled the minimum quality and purity requirements of 1 mg in less than a 130-ml volume, with an optical density OD260/280 of 1.7–1.9 and an OD260/230 of greater than 2. DNA fragmentation was assessed by agarose gel (2%) electrophoresis.
The 61 selected genes have previously been reported to be associated or potentially associated with cancer and are used by healthcare providers at our medical institution to predict the risk of cancer. AmpliSeq Custom panel was designed with AmpliSeq 7.0.2 (Thermo Fisher Scientific, Waltham, Massachusetts, USA) that covers exons, exon-intron boundaries, 5′UTR and 3′UTR regions of AKT1 ( ; ) , APC ( ) , ATM ( ) , AXIN2 ( ) , BARD1 ( ) , BMPR1A ( ) , BRCA1, BRCA2 ( ) , BRIP1 ( ; ) , BUB1 ( ) , BUB1B ( ; ) , BUB3 ( ) , CDH1 ( ) , CDKN1B ( ; ) , CHEK2 ( ) , CNTN6 ( ) , ENG ( ; ) , EPCAM ( ; ) , EPHX1 ( ; ) , FANCC ( ) , FANCE ( ) , FAN1 ( ), GALNT12 ( ; ) , GREM1 ( ; ) , LRP6 ( ) , MBD4 ( ) , MCM9 ( ) , MYH11 ( ) , MLH1, MLH3, MSH2, MSH3, MSH6 ( ; ; ; ; ; ) , MUTYH ( ; ) , NFKBIZ ( ; ) , NTHL1 ( ) , PIK3CA ( ; ) , PMS1 ( ) , PMS2 ( ) , POLD1 ( ; ) , POLE ( ) , POLQ ( ; ) , PTEN ( ) , RAD52 ( ; ) , RBL1 ( ; ; ) , REV3L ( ) , RNF43 ( ) , RPS20 ( ) , SCG5 ( ) , SDHB ( ) , SDHD ( ) , SMAD4 ( ) , SMAD9 ( ) , SMARCA4 ( ; ; ) , STK11 ( ), TDRD3 ( ) , TGFBR2 ( ) , TP53 ( ) , UIMC1 ( ) , XAF1 ( ) and XRCC4 ( ) genes, together with upstream regions of GREM1, APC and MSH2 ( ). We used an Ion Chef instrument with Ion 520 and 530 kits (Thermo Fisher Scientific) for emulsion PCR, bead enrichment, and chip loading onto the Ion 530 chip (Thermo Fisher Scientific). The loaded chips were sequenced on an Ion GeneStudio S5 Plus sequencer (Thermo Fisher Scientific). The data obtained were analyzed using the Ion Reporter server 5.10 (Thermo Fisher Scientific). The variants were visually inspected using the Integrative Genomics Viewer [(IGV) Broad Institute].
Population frequency of variant carriers We screened missense, nonsense and frameshift variants from panel sequences using a minor allele frequency (MAF) of less than 0.005 (0.5%) as a threshold in the jMorp 38KJPN dataset (version: 30 June 2022; accessed 30 December 2022) for approximately 38 000 Japanese individuals from the Tohoku Medical Megabank ( https://jmorp.megabank.tohoku.ac.jp/202206/variants ). Pathogenicity annotation In this study, we included rare germline variants of pathogenic (P), likely pathogenic (LP), and of uncertain significance (VUS). First, we classified rare variants using the ClinVar ( https://www.ncbi.nlm.nih.gov/clinvar ) database for primary interpretation of their clinical importance (accessed 30 December 2022). Next, all variants except benign and likely benign were classified according to American College of Medical Genetics and Genomics (ACMG) guidelines ( ). Supporting materials were obtained from National Center for Biotechnology Information ( https://www.ncbi.nlm.nih.gov ), gnomAD ( https://gnomad.broadinstitute.org ), COSMIC ( https://cancer.sanger.ac.uk/cosmic ) databases and InterVar ( https://wintervar.wglab.org ) bioinformatics software. Disease mutation in the Human Gene Mutation Database ( https://www.hgmd.cf.ac.uk/ac/index.php ) was noted for informative purposes (reference data: professional version 2022.2). Supplementary Figure 1, Supplemental digital content 1, http://links.lww.com/EJCP/A380 shows the variant classification process. Predicting functional effects of missense variants We classified the annotations predicting the effect of missense variants on protein function using the following three algorithms in silico (accessed 30 December 2022): PolyPhen-2 (Polymorphism Phenotyping v2.2.3; http://genetics.bwh.harvard.edu/pph2 ), SIFT ( https://sift.bii.a-star.edu.sg ) and LoFtool ( https://github.com/konradjk/loftee ). Polyphen-2 used the HumDiv model, an evaluation model for rare genes identified by genome-wide association studies, and the annotation by the respective algorithms of SIFT and LoFtool is mainly based on the Ensembl Variant Effect Predictor ( https://asia.ensembl.org/Homo_sapiens/Tools/VEP ). The predictive annotations for these three algorithms were classified as possibly damaging (PD), benign (B) for Polyphen-2, deleterious (D), deleterious low confidence (DLC), tolerated low confidence (TLC), tolerated (T) for SIFT, and PD, B for LoFtool. Statistical analysis We statistically compared the associations between patients with and without cancer. In addition, the patients were examined separately according to their sex. The association between rare germline variants and cancer risk was estimated using the odds ratio (OR), confidence interval (CI), and P value for Fisher’s exact test and calculated using the MedCalc ( https://www.medcalc.org ) software. Disease prevalence and gene counts were compared between groups using SPSS statistics software (version 20 IBM, USA) and confirmed using the parametric Student’s t -test. The significance threshold for all tests was set at P < 0.05.
We screened missense, nonsense and frameshift variants from panel sequences using a minor allele frequency (MAF) of less than 0.005 (0.5%) as a threshold in the jMorp 38KJPN dataset (version: 30 June 2022; accessed 30 December 2022) for approximately 38 000 Japanese individuals from the Tohoku Medical Megabank ( https://jmorp.megabank.tohoku.ac.jp/202206/variants ).
In this study, we included rare germline variants of pathogenic (P), likely pathogenic (LP), and of uncertain significance (VUS). First, we classified rare variants using the ClinVar ( https://www.ncbi.nlm.nih.gov/clinvar ) database for primary interpretation of their clinical importance (accessed 30 December 2022). Next, all variants except benign and likely benign were classified according to American College of Medical Genetics and Genomics (ACMG) guidelines ( ). Supporting materials were obtained from National Center for Biotechnology Information ( https://www.ncbi.nlm.nih.gov ), gnomAD ( https://gnomad.broadinstitute.org ), COSMIC ( https://cancer.sanger.ac.uk/cosmic ) databases and InterVar ( https://wintervar.wglab.org ) bioinformatics software. Disease mutation in the Human Gene Mutation Database ( https://www.hgmd.cf.ac.uk/ac/index.php ) was noted for informative purposes (reference data: professional version 2022.2). Supplementary Figure 1, Supplemental digital content 1, http://links.lww.com/EJCP/A380 shows the variant classification process.
We classified the annotations predicting the effect of missense variants on protein function using the following three algorithms in silico (accessed 30 December 2022): PolyPhen-2 (Polymorphism Phenotyping v2.2.3; http://genetics.bwh.harvard.edu/pph2 ), SIFT ( https://sift.bii.a-star.edu.sg ) and LoFtool ( https://github.com/konradjk/loftee ). Polyphen-2 used the HumDiv model, an evaluation model for rare genes identified by genome-wide association studies, and the annotation by the respective algorithms of SIFT and LoFtool is mainly based on the Ensembl Variant Effect Predictor ( https://asia.ensembl.org/Homo_sapiens/Tools/VEP ). The predictive annotations for these three algorithms were classified as possibly damaging (PD), benign (B) for Polyphen-2, deleterious (D), deleterious low confidence (DLC), tolerated low confidence (TLC), tolerated (T) for SIFT, and PD, B for LoFtool.
We statistically compared the associations between patients with and without cancer. In addition, the patients were examined separately according to their sex. The association between rare germline variants and cancer risk was estimated using the odds ratio (OR), confidence interval (CI), and P value for Fisher’s exact test and calculated using the MedCalc ( https://www.medcalc.org ) software. Disease prevalence and gene counts were compared between groups using SPSS statistics software (version 20 IBM, USA) and confirmed using the parametric Student’s t -test. The significance threshold for all tests was set at P < 0.05.
Patient characteristics are summarized in Table . The mean age of the patients was 82 years and ± 8.09 standard deviation (cancer patients: 82 ± 8.003; controls: 82 ± 8.21). Of the 90 patients with cancer, 72 (80%) had pancreatic cancer, including 49 with pancreatic cancer only (PaC) and 23 with pancreatic cancer and multiple primary cancers (PaMC). Eighteen (20%) patients had no pancreatic cancer in multiple primary cancers (NPaMC). The control group comprised 99 patients. We examined 61 cancer predisposition genes and identified 169 rare germline variants, including novel variants in 46 genes (Fig. a,b, Supplementary Table 1, Supplemental digital content 2, http://links.lww.com/EJCP/A381 and Supplementary Table 2, Supplemental digital content 3, http://links.lww.com/EJCP/A382 ). Pathogenic and likely pathogenic variants Table summarizes the rare germline P and LP variants. APC (P), BRCA2 (P), BUB1B (LP), ENG (LP) and MSH6 (P) were identified in patients with cancer. The frequency of P/LP variants was 6% in each cancer group: pancreatic cancer (PaC + PaMC; 4/72), NPaMC (1/18) and all-cancer (5/90) groups. We identified AXIN2 (LP), BMPR1A (P), MSH2 (LP) and SMARCA4 (LP; n = 2) in controls and the frequency of these variants (P/LP) was 5% (5/99). There was no statistically significant difference in the presence of pathogenic variants (P/LP) between cancer and non-cancer controls ( P = 1.00). Variants of uncertain significance We identified 159 rare germlines VUS. In cancer patients, 47 variants were identified in women and 36 variants were identified in men. In non-cancer patients, 43 variants were identified in women and 33 variants were identified in men. In the cancer group, 54% (49/90) of patients carried only VUS. The most frequent genes in patients with cancer were DNA polymerase theta ( POLQ; n = 6) and MSH6 (n = 4) in women and POLQ (n = 4) in men (Fig. a,b). We examined the association between VUS and the presence of cancer by using two statistical approaches in a case-control study. Table summarizes the association between VUS and the presence or absence of cancer, using Fisher’s exact test. MSH6 , four MMR genes ( MLH1, MSH2, MSH6 and PMS2 ), and POLQ were significantly associated with both pancreatic cancer and all-cancer groups. MSH6 was significantly associated with pancreatic cancer ( P = 0.029) and all-cancer groups ( P = 0.023; women: P = 0.044). The ORs for MMR genes ( MLH1, MSH2, MSH6 and PMS2 ) were significantly higher in both the pancreatic cancer (OR = 3.83; P = 0.025) and all-cancer groups (OR = 4; P = 0.019). POLQ was significantly associated with pancreatic cancer in men ( P = 0.027) and all-cancer groups (OR = 4; P = 0.041). Table summarizes the results of the Student’s t -tests. We examined the mean number of VUS for MMR genes ( MLH1, MSH2, MSH6 and PMS2 ). The mean number of VUS for MMR genes ( MLH1, MSH2, MSH6 and PMS2 ) was significantly higher in the pancreatic cancer group (cancer patient, 0.13 vs. control, 0.04; P = 0.033) than that in the control group. In the all-cancer group, the mean number of VUS for MMR genes ( MLH1, MSH2, MSH6 and PMS2 ) was significantly higher (0.14 vs. 0.04; P = 0.015) than that in the control group. The mean number of VUS was significantly higher in the all-cancer group (0.18 vs. 0.05; P = 0.044) compared to the control group in women. MMR genes ( MLH1, MSH2, MSH6 and PMS2 ) in multiple primary cancers One P variant and six VUS MMR genes ( MLH1, MSH2, MSH6 and PMS2 ) were identified in patients with multiple primary cancers (pancreatic, colorectal, gastric, bladder, cervical, esophageal, lung, skin and thyroid cancers). The details are presented in Supplementary Table 3, Supplemental digital content 4, http://links.lww.com/EJCP/A383 . Predicting functional effects of missense variants In silico analysis using Polyphen-2, SIFT and LoFtool was annotated in cancer patients and controls for predicted effects of missense variants on protein function (Supplementary Table 1, Supplemental digital content 2, http://links.lww.com/EJCP/A381 and Supplementary Table 2, Supplemental digital content 3, http://links.lww.com/EJCP/A382 ). All three algorithms listed VUS annotated as deleterious and possibly damaging in patients with pancreatic cancer (Fig. ). POLQ was abundant ( n = 7), and AXIN2 (n = 2), MSH3 (n = 2), CDH1, LRP6, MLH1, MSH2, MSH6, NTHL1 and LEV3L (n = 1 each) were identified, and the same variant of CDH1 , c.2638G>A (p.Glu880Lys), was also identified in one 67-year-old woman (NPaMC) with hepatocellular carcinoma and colon cancer. During data collection, some information was unclear, such as the rs ID numbers for variants, but was scrutinized as closely as possible. Further updating of information on the data sets used in this study is needed.
Table summarizes the rare germline P and LP variants. APC (P), BRCA2 (P), BUB1B (LP), ENG (LP) and MSH6 (P) were identified in patients with cancer. The frequency of P/LP variants was 6% in each cancer group: pancreatic cancer (PaC + PaMC; 4/72), NPaMC (1/18) and all-cancer (5/90) groups. We identified AXIN2 (LP), BMPR1A (P), MSH2 (LP) and SMARCA4 (LP; n = 2) in controls and the frequency of these variants (P/LP) was 5% (5/99). There was no statistically significant difference in the presence of pathogenic variants (P/LP) between cancer and non-cancer controls ( P = 1.00).
We identified 159 rare germlines VUS. In cancer patients, 47 variants were identified in women and 36 variants were identified in men. In non-cancer patients, 43 variants were identified in women and 33 variants were identified in men. In the cancer group, 54% (49/90) of patients carried only VUS. The most frequent genes in patients with cancer were DNA polymerase theta ( POLQ; n = 6) and MSH6 (n = 4) in women and POLQ (n = 4) in men (Fig. a,b). We examined the association between VUS and the presence of cancer by using two statistical approaches in a case-control study. Table summarizes the association between VUS and the presence or absence of cancer, using Fisher’s exact test. MSH6 , four MMR genes ( MLH1, MSH2, MSH6 and PMS2 ), and POLQ were significantly associated with both pancreatic cancer and all-cancer groups. MSH6 was significantly associated with pancreatic cancer ( P = 0.029) and all-cancer groups ( P = 0.023; women: P = 0.044). The ORs for MMR genes ( MLH1, MSH2, MSH6 and PMS2 ) were significantly higher in both the pancreatic cancer (OR = 3.83; P = 0.025) and all-cancer groups (OR = 4; P = 0.019). POLQ was significantly associated with pancreatic cancer in men ( P = 0.027) and all-cancer groups (OR = 4; P = 0.041). Table summarizes the results of the Student’s t -tests. We examined the mean number of VUS for MMR genes ( MLH1, MSH2, MSH6 and PMS2 ). The mean number of VUS for MMR genes ( MLH1, MSH2, MSH6 and PMS2 ) was significantly higher in the pancreatic cancer group (cancer patient, 0.13 vs. control, 0.04; P = 0.033) than that in the control group. In the all-cancer group, the mean number of VUS for MMR genes ( MLH1, MSH2, MSH6 and PMS2 ) was significantly higher (0.14 vs. 0.04; P = 0.015) than that in the control group. The mean number of VUS was significantly higher in the all-cancer group (0.18 vs. 0.05; P = 0.044) compared to the control group in women.
MLH1, MSH2, MSH6 and PMS2 ) in multiple primary cancers One P variant and six VUS MMR genes ( MLH1, MSH2, MSH6 and PMS2 ) were identified in patients with multiple primary cancers (pancreatic, colorectal, gastric, bladder, cervical, esophageal, lung, skin and thyroid cancers). The details are presented in Supplementary Table 3, Supplemental digital content 4, http://links.lww.com/EJCP/A383 .
In silico analysis using Polyphen-2, SIFT and LoFtool was annotated in cancer patients and controls for predicted effects of missense variants on protein function (Supplementary Table 1, Supplemental digital content 2, http://links.lww.com/EJCP/A381 and Supplementary Table 2, Supplemental digital content 3, http://links.lww.com/EJCP/A382 ). All three algorithms listed VUS annotated as deleterious and possibly damaging in patients with pancreatic cancer (Fig. ). POLQ was abundant ( n = 7), and AXIN2 (n = 2), MSH3 (n = 2), CDH1, LRP6, MLH1, MSH2, MSH6, NTHL1 and LEV3L (n = 1 each) were identified, and the same variant of CDH1 , c.2638G>A (p.Glu880Lys), was also identified in one 67-year-old woman (NPaMC) with hepatocellular carcinoma and colon cancer. During data collection, some information was unclear, such as the rs ID numbers for variants, but was scrutinized as closely as possible. Further updating of information on the data sets used in this study is needed.
We explored rare germline variants that encode proteins that predispose patients to sporadic pancreatic cancer. Panel sequencing of the 61 genes was performed using a case-control design in patients with a negative family history. Sequencing results revealed that 6% (4/72) of the patients, including one 104-year-old centenarian woman, with pancreatic cancer carried pathogenic cancer predisposition P/LP variants (Table ). VUS of LS-related germline MMR genes and POLQ were significantly associated with pancreatic cancer. Interestingly, these results suggest that trends in cancer risk may differ in women and men, respectively (Tables and ). Our report on the pathogenic frequency of rare germline variants provides evidence that reaffirms the importance of predicting pancreatic cancer risk using genetic screening in individuals without a family history. Previous studies have reported that the frequency of germline pathogenic variants in patients with familial pancreatic cancer ranges from approximately 5 to >10% ( ; ; ; ), supporting the frequency of 6% observed in the present study. Pancreatic cancer-predisposing genes were also present in multiple cancer phenotypes, and similar results were observed for the following two cancer subsets: all-cancer (5/90) and NPaMC (1/18) groups. Despite current cost concerns, we speculate that comprehensive genetic evaluations, such as multigene panels, may be beneficial for early detection and therapeutic measures even for individuals with a negative family history. We identified five genes associated with pancreatic cancer predisposition in patients with cancer. P/LP variants of APC, BRCA2, BUB1B and ENG were identified in patients with pancreatic cancer, and MSH6 was identified in an NPaMC patient with both colorectal and esophageal cancers. No pathogenic P or LP variants were identified in any of these five genes in the control group (Table ). In jMorp, a large Japanese database, BUB1B c.2441G>A (p.Arg814His) has been reported as a rare variant (MAF = 0.0039%); however, no records are available for APC c.4666dupA (p.Thr1556AsnfsX3), BRCA2 c.3853_3854insG (p.Glu1285fs), ENG c.728_729insCG (p.Pro244fs) and MSH6 c.1444C>T (p.Arg482Ter). The ENG c.728_729insCG variant is not listed in the ClinVar database (accessed 30 December 2022) and appears to be a novel variant. Meanwhile, the P/LP variants of AXIN2, BMPR1A, MSH2 and SMARCA4 were identified in 5% (5/99) of the subjects in the non-cancer control group, including one 104-year-old woman (Table ), which did not significantly differ from the rate in the cancer group, and therefore the significance of risk is uncertain. These results also provide evidence that cancer risk prediction should be interpreted with caution and emphasize the complexity of understanding cancer mechanisms. Germline variants can influence somatic variant patterns and genomic instability; ancestral populations with the same germline variants may exhibit different cancer phenotypes ( ). One of our interesting findings is that de novo APC c.4666dupA (p.Thr1556AsnfsX3), which has been reported to cause an aggressive Gardner syndrome in a 2-year-old European boy ( ), reproduced as a pancreatic, duodenal and colorectal cancer phenotype in a 104-year-old Japanese woman (Table ). The effect of predisposition, such as family history, other genes, or environment, on the phenotypic differences between the two patients, i.e. the boy and the centenarian, remains unclear. Whether the pathogenic variant APC c.4666dupA varies significantly by sex, age and race in terms of disease phenotype, age of onset and prevalence require clarification. Overall, the observed landscapes of rare germline variants were diverse and heterogeneous (Fig. a,b), suggesting that equivalent or complementary interactions between other susceptible genes and variants may influence cancer development. Of the 90 patients with cancer, 40% (36/90) carried no variants and 54% (49/90) carried only a potentially pathogenic variant, i.e. VUS. We then examined whether VUS carriers of specific cancer predisposition genes are more likely to develop cancer. Our study evaluated the potential effects of VUS on the risk of developing cancer in a case-control study and confirmed statistical associations among the following genes: MSH6 , four MMR genes ( MLH1, MSH2, MSH6 and PMS2 ) and POLQ . VUS for these genes was more abundant in patients with pancreatic cancer than in the non-cancer control group, and the same trend was observed in the all-cancer group (Tables and ). MSH6 tends to delay cancer development compared to MLH1 and MSH2 ( ; ; ). Patients with multiple primary cancers (PaMC and NPaMC) with variants in MLH1, MSH2, MSH6 and PMS2 showed a trend consistent with various LS-related cancers ( ; ; ; ; ; ), although whether they had LS was unknown (Supplementary Table 3, Supplemental digital content 4, http://links.lww.com/EJCP/A383 ). In the results of the three algorithms for protein function prediction (Fig. , Supplementary Table 1, Supplemental digital content 2, http://links.lww.com/EJCP/A381 and Supplementary Table 2, Supplemental digital content 3, http://links.lww.com/EJCP/A382 ), the predicted impact of VUS on gene expression in patients with pancreatic cancer included predicted possibly damaging and deleterious variants in MLH1, MSH2 and MSH6 . Interestingly, POLQ were more abundant than those of other genes and some VUS might be loss-of-function variants in DNA double-strand break repair-related pathways. Genetic POLQ dysregulation plays an important role in genomic double-strand break repair and genome maintenance and is associated with the risk of pancreatic, breast, ovarian and other cancers, and overexpression is associated with poor prognosis ( ; ; ). Here, our results also revealed that the VUS of MSH6 and POLQ suggested potentially different trends in association with cancer in women and men, respectively (Table ). Our results may indicate that VUS in cancer patients may affect the potential accumulation of cancer risk, although there is no evidence of a causal relationship between VUS and cancer risk at this stage. We could not examine microsatellite instability (MSI) as it relates to error repair in the germline; therefore, the association between MSI and the VUS of MMR genes is unknown. MMR gene-related cancers, including pancreatic cancer, are associated with MSI and MMR gene dysfunction, and the presence of MMR gene dysfunction or MSI-high is a known LS predictor and increases cancer risk ( ; ; ). MSI status is also interesting because it predicts immunotherapy susceptibility such as immune checkpoint inhibitors ( ; ; ). Previous studies that focused on sex differences in MSI in somatic MMR gene variants reported significant sex differences in esophageal and gastric cancers as well as an association among the differences in the incidence, prognosis, and treatment response of cancer ( ; ). Further research on germline variants including MSI may facilitate the development of personalized therapies that focus on sex-based differences. Information retrieved from the ClinVar database indicates no consensus on the involvement of VUS in cancer predisposition. The number of individual variant carriers of this study was limited and the small sample size was insufficient to establish the potential role of VUS in cancer. Nevertheless, the potential pathogenic risk of VUS is important for variant carriers. The clinical management of genetic diagnosis with risk prediction using VUS has several limitations. The current scrutiny of VUS pathogenicity has not kept pace with the number of VUS that have been detected. Indeed, many VUS are subsequently downgraded to benign variants whereas some are classified as pathogenic ( ; ). In addition, a study also reported that the coexistence of VUS in MMR genes with predicted nonpathogenicity with other VUS of MMR genes might subtly increase cancer risk ( ). Hence, identification of VUS, which is very common in patients with cancer, and the collection of information on at-risk VUS by functional and epidemiological analysis is warranted. Findings of previous studies, as well as those of the current study, have suggested the importance of cancer risk prediction in individuals carrying rare germline VUS that lack the P/LP variants. This study had several limitations. First, the rare germline variants identified in a population with a negative family history are likely de novo variants. However, no absolute evidence was found because no genetic information was available from the parents, families or relatives. Second, the effects of rare germline variants may be influenced by disease combination, smoking and drinking status, ethnicity, geographical region and database classification interpretation. Thus, the extent of variant reflection in the development of cancer remains unclear. Third, the presence of variants that were not included in the cohort and the uncertainty of the statistical power may stem from the small population size. Our results reflect only partial information on the individual patients in the study cohort. Prospective case-control studies including all age groups in various large populations of different ancestries are warranted to validate the current study findings. In conclusion, our results provide important evidence for P/LP variants in 6% of elderly patients with pancreatic cancer with a negative family history. As for VUS, MMR genes ( MLH1, MSH2, MSH6 and PMS2 ) and POLQ may help predict potential trends in pancreatic cancer risk. MSH6 and POLQ may have slightly different potential cancer-associated trends in women and men, respectively. Given the current limitations of pancreatic cancer risk prediction in the general population, algorithmic considerations for the VUS distribution of cancer-predisposing genes will be important in estimating trends in individual genetic risks in the future. Further studies are required to confirm these findings.
The authors thank all the staff of the Department of Pathology Tokyo Metropolitan Geriatric Hospital Institute of Gerontology for preparing the samples. This study was supported in part by the Smoking Research Foundation (T.A.). The authors thank the following individuals for their contributions: Dr. Julien Legrand, Shizuoka University. Dr. Josvin K. Dr. Aye K.K Minn, Tohoku University. Dr. Moli G. This study was partially funded by the Smoking Research Foundation (T.A.). Conflicts of Interest There are no conflicts of interest.
There are no conflicts of interest.
|
Acute benzyl alcohol intoxication: An autopsy case report | 7d687982-ef9f-455d-a798-9ddfe5f075b2 | 10063254 | Forensic Medicine[mh] | Benzyl alcohol is a metabolite of toluene and is oxidized by an alcohol-depleting enzyme in the liver that produces benzoic acid. Benzoic acid is then combined with glycine and excreted in urine as hippuric acid. Benzyl alcohol is an ingredient in many paint removers and other industrial applications, and is even found in cosmetic products and food additives. There have been few case reports on benzyl alcohol intoxication in adults, and these cases showed impaired consciousness, respiratory depression, hypotension, metabolic acidosis, renal dysfunction, hypothermia, paralytic ileus, and hyperammonemia. However, autopsy reports of cases of benzyl alcohol intoxication have not been published. Therefore, the pathological appearance of cases of benzyl alcohol intoxication has not been fully investigated. We report a rare autopsy case of acute benzyl alcohol intoxication, and discuss the pathogenesis from the pathological appearance of this case.
A 24-year-old man was found collapsed at the workplace where he was painting. He had been performing paint stripping using a paint stripper containing benzyl alcohol to repaint a bridge of an express highway. The time interval between the final confirmation of the man healthy state and the discovery of his collapse was approximately 25 minutes. He suffered from cardiopulmonary arrest. He was transferred to the general hospital approximately 130 minutes after he was found, and manual cardiopulmonary resuscitation was performed up to arrival to the hospital. On arrival to the hospital, blood laboratory data showed that he was in a critical state because of higher concentrations of transaminases, enzymes of the musculature enzyme, and potassium (Table ). Computed tomography showed no abnormal findings. He died 56 minutes after admission despite intensive resuscitation. He did not have any relevant clinical or family history. The postmortem interval at the time of autopsy was 12 hours. An autopsy showed discoloration of the skin with coating of an unknown material in the anterior neck, anterior chest, and both upper arms (Fig. A). Bullous epidermolysis was found on the discoloration area (Fig. B). No considerable macroscopic appearance was found in the internal organs other than congestion. Microscopically, vacuolar degeneration in the epidermis and dermo-epidermal junction was observed (Fig. C). Degeneration of bronchial mucosa was also evident (Fig. D). The brain weighed 1602 g and showed marked edematous swelling, and also showed other pathological changes (Fig. A–C). A histopathological examination showed central chromatolysis in the neurons of the pontine nucleus, nucleus of the solitary tract, and nucleus ambiguus in the medulla oblongata (Fig. D). Additionally, grumose degeneration was observed in the cerebellar dentate nucleus (Fig. E). Amyloid precursor protein-positive axonal bulbs were not detected. Severe erosion with degeneration of the bronchial epithelium was evident (Fig. F), but degeneration of the tubular epithelium of the proximal and distal renal tubules was not found (Fig. G). The cardiac blood benzyl alcohol concentration was analyzed by headspace-gas chromatography-flame ionization detection methods (GC-2010, Shimadzu, Kyoto, Japan). Benzyl alcohol was quantified with a linear curve fit using methyl ethyl ketone as the internal standard. The blood benzyl alcohol concentration was 780.0 μg/mL.
A lethal concentration of blood benzyl alcohol has not been established because of the rarity of such cases. Also, we should consider the possibility that a certain amount of postmortem diffusion from lung to cardiac blood may occur. However, the value in the present case was higher than that of a rescued, severe, symptomatic case (489 ± 5 μg/mL). Therefore, we consider that the cause of death in the present case was acute benzyl alcohol intoxication. To the best of our knowledge, this is the first autopsy report of a case of acute benzyl alcohol intoxication. Unfortunately, the present case was found in the state of cardiopulmonary arrest and died in a short interval. Therefore, the pathological appearance in the present case may have been caused by a direct effect of benzyl alcohol on the human body. A detailed pathological investigation may be useful for determining the possible pathogenesis leading to acute cardiopulmonary arrest in severe benzyl alcohol intoxication. Inappropriate protective procedures might have been the cause of accidental intoxication of benzyl alcohol in the present case. Benzyl alcohol can be absorbed orally, by inhalation, or transdermally. Vacuolar degeneration in the epidermis and dermo-epidermal junction, which was found in the present case, may be consistent with macroscopic bullous epidermolysis. Vacuolar degeneration may be a sign of benzyl alcohol exposure as shown by a previous case report. Additionally, the degeneration of the bronchial mucosa may be pathological evidence of inhalation of benzyl alcohol in the present case. Exposure by inhalation and transdermal absorption may be associated with rapid progression of acute intoxication. Benzyl alcohol is considered to cause direct injury to the central nervous system, but the specific mechanism and pathological appearance have not been fully established. McClosky et al reported a case of altered mental status after exposure to benzyl alcohol, and additionally showed that benzyl alcohol itself was toxic in a mouse model study. Experimental studies on the central neurotoxicity of toluene showed that multiple neurotransmitter and receptor systems, including N-methyl-D-aspartate receptors, dopamine neurons, and the g-aminobutyric acid transmission pathway, were affected by toluene exposure. Impairment of N-methyl-D-aspartate receptors, dopamine neurons, and the g-aminobutyric acid transmission pathway might occur in cases of acute benzyl alcohol intoxication. Central chromatolysis is a consequence of axonal injury, and chromatolysis can be characterized by reorganization of the cell soma and redistribution of Nissl substances to reconstitute injured axons. Grumose degeneration in the cerebellar dentate nucleus, which shows degeneration of the axon terminal of Purkinje cells, is found not only in chronic neurodegenerative disease (e.g., progressive supranuclear palsy and spinocerebellar degeneration), but also in acute ischemic/hypoxic encephalopathy. We recently found central chromatolysis and grumose degeneration in an autopsy case with rapid consciousness disturbance due to acute colchicine intoxication. The findings in the present case suggest that rapid progressive microscopic degeneration can occur in the central nervous system in a short interval after exposure to benzyl alcohol. Cardiac-projecting neurons of the nucleus ambiguus play a critical role in cardiac parasympathetic tone. Therefore, their activation elicits bradycardia via acetylcholine release in cardiac ganglia. Additionally, neurons in the nucleus of the solitary tract are essential for processing and coordinating respiratory and sympathetic responses to hypoxia. These studies indicate that pathological changes in the circulatory and respiratory centers in the medulla oblongata, as observed in the present case, may be strongly associated with the prognosis of acute benzyl alcohol intoxication. Renal tubular dysfunction is considered to be a major complication, and associated symptoms are metabolic acidosis, hypokalemia, hypophosphatemia, and hyperammonemia. We could not evaluate renal tubular function and electrolyte values because the victim was already critical on admission. However, the pathological appearance of the kidney did not show degeneration of the tubular epithelium. The findings in the present case suggest that multiple pathways of exposure may be associated with more rapid progression in acute benzyl alcohol intoxication, and that early and/or severe involvement of the central nervous system may be associated with an early death.
The authors thank Ms. Syuko Matsumori, Ms. Miyuki Maekawa, Ms. Misa Kusaba, and Mr. Osamu Yamamoto for their technical assistance. We thank Ellen Knapp, PhD, from Edanz ( https://jp.edanz.com/ac ) for editing a draft of this manuscript.
Conceptualization: Shojiro Ichimata, Naoki Nishida. Data curation: Shojiro Ichimata, Yukiko Hata, Ryosuke Zaimoku, Naoki Nishida. Investigation: Shojiro Ichimata, Yukiko Hata, Ryosuke Zaimoku, Naoki Nishida. Supervision: Ryosuke Zaimoku, Naoki Nishida. Validation: Shojiro Ichimata, Yukiko Hata. Visualization: Yukiko Hata. Writing – original draft: Shojiro Ichimata. Writing – review & editing: Naoki Nishida.
|
A mixed methods study of Aboriginal health workers’ and exercise physiologists’ experiences of co-designing chronic lung disease ‘yarning’ education resources | ca231128-e443-457c-8147-45d4cf232aba | 10063331 | Patient Education as Topic[mh] | Aboriginal and Torres Strait Islander people, the Indigenous peoples of Australia (hereafter respectfully referred to as Aboriginal people), have a higher burden of chronic obstructive pulmonary disease (COPD) than other Australians [ – ], which could be addressed by improved access to primary care services . Pulmonary rehabilitation (PR) is a key component of effective COPD management . PR is an evidence-based program, typically conducted for one hour, twice a week for 8 weeks. PR programs consist of individually tailored exercise training prescribed by a physiotherapist (PT) or exercise physiologist (EP), and education sessions related to COPD management . Most PR programs in Australia are provided by state-funded local health services in hospital outpatient settings . Access to and uptake of these mainstream PR programs by Aboriginal people is low , largely due to mainstream health services being perceived as culturally unsafe . Aboriginal Community Controlled Health Services (ACCHS) are incorporated organisations, governed by a Board consisting of local Aboriginal community members. ACCHS were established to address the inadequacies of mainstream health services to deliver culturally safe care for Aboriginal people . ACCHS operate in dynamic environments and deliver holistic and effective primary health care in innovative ways that respond to local health priorities . Although ACCHS deliver a range of chronic care services, PR is not typically offered as core business and there is no specific funding mechanism for PR in these settings . There has been little published about the outcomes of PR for Aboriginal people with COPD, with only one study of PR provided by an ACCHS reporting that Aboriginal participants with chronic respiratory or heart disease significantly improved functional exercise capacity and quality of life . The reportedly low level of knowledge and confidence of the Australian rural and remote Aboriginal and non-Aboriginal health workforce around the provision of PR may impact delivery of this service. Enhancing a capable and trained Aboriginal health workforce has shown to improve health care access and outcomes for Aboriginal people and positively contribute to the performance, quality and deliverable outcomes of the Aboriginal primary health sector . ACCHS function at the cultural interface between Western medical delivery and Aboriginal ways of knowing, being and doing. Therefore, building an Aboriginal health workforce that has both professional and cultural competence is a priority [ , , ]. In Australia, Canada, New Zealand and the United States, Indigenous health care workers perform critical functions as cultural brokers in program planning, service delivery and care by applying clinical and socio-cultural skills to improve patients’ access to culturally safe care . An Australian narrative review identified strategies to develop and maintain a skilled rural and remote health workforce and revealed the importance of Aboriginal and non-Aboriginal staff working together, with Aboriginal staff being supported to use their specialised skill sets and knowledge of the local Aboriginal community needs to effectively implement health programs . The importance of designing and delivering training focused on the specific needs of both the Aboriginal workforce and the Aboriginal community receiving the program or care was also highlighted . The ‘8 Ways of learning’ is an Aboriginal pedagogy framework and was an initiative of the NSW Department of Education that engaged Aboriginal knowledge holders in its development. 8 Ways of learning incorporates cultural protocols and Aboriginal perspectives to realign teaching techniques and approaches, and to strengthen engagement and outcomes for Aboriginal learners . Fundamental to this framework is prioritising people and relationships and integrating Aboriginal ways of knowing, being and doing [ – ]. The cultural protocols include the following 8 ways: Deconstruct/reconstruct (dismantling and reassembling approaches using local Aboriginal world views); Community Links (identification of key community stakeholders to be consulted about projects); Learning maps (exploring and visualising concepts and strategies); Story Sharing (clarifying and sharing content verbally); Symbols and images (explaining concepts using artwork and visual methods); Land links (relating learning to spaces, places and the living landscape); Non-verbal (using hands-on activities and body language, including silence); and Non-linear (non-consequential and indirect orientation of learning) . While face-to-face experiential learning and multi-faceted workforce strategies have been shown to be effective in building knowledge, confidence and skills of the Aboriginal health workforce , the impact of the COVID-19 pandemic and travel-related restrictions meant sectors such as education and health needed to pivot to virtual care and online learning [ – ]. The pandemic provided an opportunity to reflect on the design of online training, the learning needs of the Aboriginal health workforce and scaffolding strategies to build knowledge and autonomy as well as teaching approaches , such as the incorporation of 8 Ways of learning and yarning. Yarning for Aboriginal people is a cultural conversational process of sharing stories and experiences, thoughts and ideas in culturally safe environments which centres around Aboriginal ways of knowing, being and doing [ – ]. The aim of this study was to explore the experiences of AHWs, EPs and PTs within ACCHS who attended online education sessions using co-design principles and 8 Ways of learning to enhance workforce capacity and provide culturally safe education as part of a PR program for Aboriginal people with COPD. The study used participatory action research which encourages self-reflective enquiry and shares power and decision making with participants in the design and/or delivery of programs using co-design principles , which aligns well with 8 Ways of learning. Aboriginal health projects using participatory action research, guided by co-design principles, and incorporating Aboriginal perspectives have effectively engaged stakeholders and produced measurable outcomes .
Context and settings This mixed methods study was a component of the overarching Breathe Easy Walk Easy Lungs for Life (BE WELL) project, evaluating the implementation of PR within ACCHS. The study is approved by the Aboriginal Health & Medical Research Council of NSW Human Resource Ethics Committee (HREC 1261/17). For the overall BE WELL project, the research team contacted all NSW-based ACCHS by letter inviting them to join the study. Four ACCHS classified as Rural Zone Codes 3–5, agreed to participate . The BE WELL project protocol is published , and the trial is registered with the Australian New Zealand Clinical Trials Registry (ANZCTR): ACTRN12617001337369 . Participants Study participants were staff from four ACCHS responsible for local BE WELL implementation. These were the EPs or PTs providing the exercise prescription and training and the AHWs providing education about COPD within the BE WELL PR programs. These staff (EPs/PTs and AHWs) were invited to attend a two-day face-to-face BE WELL Workshop which concentrated on patient assessment and the exercise training component of PR. Evaluation of the two-day workshop will be conducted separately. To provide training around the education component of PR, staff were invited to attend additional face-to-face training to build their COPD management knowledge and co-design the education (yarning) material. Due to COVID-19 travel-related restrictions in 2020, the additional face-to-face training was delivered online. Each of the four ACCHS negotiated the timing and frequency of the online education sessions around their existing responsibilities. All participants provided informed consent. Online education Seven topics were included in the online education sessions. These were: How the lungs work; What is COPD; Medications and how to use inhalers and COPD Action Plans; Why exercise is important; Managing breathlessness; Healthy eating; Managing anxiety and depression. All sessions were delivered via Zoom using interactive internet-based video conferencing (also known as Voice over Internet Protocol (VoIP)-mediated technologies) with participant and facilitator discussion . Sessions were between 30–60 min depending on participants’ preference. Sessions commenced with an Acknowledgment of Country and introductions. Each session was designed to incorporate 8 Ways of learning and co-design principles . For each topic, the facilitator prepared and presented brief PowerPoint slides of the main topic points and, depending on the topic, sessions would include story sharing between participants stimulated by videos of patients with lung disease or demonstrations of the correct use of inhalers. Images, symbols, artwork, metaphors and analogies, as well as links to the living landscape, such as trees and river systems and meeting places were used to explain how the lungs work. Resources such as diagrams, booklets, medication charts, PowerPoint slide notes and session planning tools were provided. Examples of questions that the AHWs could ask patients to stimulate yarning around each of the topics were suggested and further developed. Following each session, AHWs with support from EPs/PTs, co-designed education ‘yarnings’ to deliver with their community. The yarnings used 8 Ways of learning and incorporated simple terms, phrases and words from the local Aboriginal languages to ensure each topic was engaging, relevant and culturally safe for each Aboriginal community. The yarning scripts developed by the AHW were used as a tool to clearly outline their yarning approach for each topic, key conversational points, stimulus questions and learning aides, such as pictorial or physical resources to be used. Outcome measures After completing all the online education sessions, participants were invited via email to complete an anonymous online survey. The survey consisted of 19 questions, 16 of which used a 5-point Likert scale and three were free text. See Supplementary Survey, Additional file . Participants were also invited to participate in a semi-structured interview about their experience of the online education. See Supplementary Interview questions, Additional file . Interviews were conducted using VoIP technologies by an investigator (KG) experienced in qualitative health research with Aboriginal participants, and who had not been involved in the design or delivery of the online education sessions. Interviews were conducted between November and December 2020. Recruitment ceased after all available AHW/EPs who participated in the online education sessions attended an interview. Sufficient data were collected to support saturation allowing for broad and deep insights related to the research question. A professional transcription service transcribed the audio recordings verbatim into text. Transcriptions were de-identified to ensure confidentiality of participants. DM and JA reviewed transcriptions against audio recordings for accuracy. Data analysis Survey data were analysed using descriptive statistics within IBM SPSS statistics version 27. The questions with the 5-point Likert scale were given numerical values of -2 (strongly disagree/never/very poor), -1 (disagree/rarely/poor), 0 (unsure/sometimes/average), 1 agree/mostly/good), 2 (strongly agree/always/very good). Interview data were managed using NVivo (QSR International Pty Ltd 2020). Reflexive thematic data analysis was applied within a framework of four central areas of AHW and EP experience participating in the online education sessions and co-designing ‘yarning’ resources, and their perception of patient experience of access to care and health literacy of COPD . Thematic analysis used the six phases outlined by Braun and Clarke consisting of familiarisation, generating codes, searching for themes, reviewing potential themes, naming the themes and producing the report . Familiarisation was facilitated by (DM, JA, SD) listening to the audio recordings and reading and rereading the transcriptions, taking notes and examining the data for meaning. Following familiarisation of the data, codes were defined inductively by (DM, JA, SD). Assumptions about code generation were checked with SD as the experienced qualitative researcher, with the perspectives of the Aboriginal researcher (DM) given deep consideration relating to their Aboriginal socio-cultural meaning. Final themes were generated based on central concepts and consensus reached by authors (DM, JA, SD, JMc). Authors (DM and JA) remained conscious of reflexivity and the impact of their perspectives (professional, personal and cultural) during study implementation, data collection, analysis and presentation of the findings to strengthen rigour, trustworthiness and overall quality of the study. Quality of coding and analysis of the data was strengthened by authors (SD and JMc) through enquiry and verification of the inductive coding framework and coded text with authors (DM and JA) .
This mixed methods study was a component of the overarching Breathe Easy Walk Easy Lungs for Life (BE WELL) project, evaluating the implementation of PR within ACCHS. The study is approved by the Aboriginal Health & Medical Research Council of NSW Human Resource Ethics Committee (HREC 1261/17). For the overall BE WELL project, the research team contacted all NSW-based ACCHS by letter inviting them to join the study. Four ACCHS classified as Rural Zone Codes 3–5, agreed to participate . The BE WELL project protocol is published , and the trial is registered with the Australian New Zealand Clinical Trials Registry (ANZCTR): ACTRN12617001337369 .
Study participants were staff from four ACCHS responsible for local BE WELL implementation. These were the EPs or PTs providing the exercise prescription and training and the AHWs providing education about COPD within the BE WELL PR programs. These staff (EPs/PTs and AHWs) were invited to attend a two-day face-to-face BE WELL Workshop which concentrated on patient assessment and the exercise training component of PR. Evaluation of the two-day workshop will be conducted separately. To provide training around the education component of PR, staff were invited to attend additional face-to-face training to build their COPD management knowledge and co-design the education (yarning) material. Due to COVID-19 travel-related restrictions in 2020, the additional face-to-face training was delivered online. Each of the four ACCHS negotiated the timing and frequency of the online education sessions around their existing responsibilities. All participants provided informed consent.
Seven topics were included in the online education sessions. These were: How the lungs work; What is COPD; Medications and how to use inhalers and COPD Action Plans; Why exercise is important; Managing breathlessness; Healthy eating; Managing anxiety and depression. All sessions were delivered via Zoom using interactive internet-based video conferencing (also known as Voice over Internet Protocol (VoIP)-mediated technologies) with participant and facilitator discussion . Sessions were between 30–60 min depending on participants’ preference. Sessions commenced with an Acknowledgment of Country and introductions. Each session was designed to incorporate 8 Ways of learning and co-design principles . For each topic, the facilitator prepared and presented brief PowerPoint slides of the main topic points and, depending on the topic, sessions would include story sharing between participants stimulated by videos of patients with lung disease or demonstrations of the correct use of inhalers. Images, symbols, artwork, metaphors and analogies, as well as links to the living landscape, such as trees and river systems and meeting places were used to explain how the lungs work. Resources such as diagrams, booklets, medication charts, PowerPoint slide notes and session planning tools were provided. Examples of questions that the AHWs could ask patients to stimulate yarning around each of the topics were suggested and further developed. Following each session, AHWs with support from EPs/PTs, co-designed education ‘yarnings’ to deliver with their community. The yarnings used 8 Ways of learning and incorporated simple terms, phrases and words from the local Aboriginal languages to ensure each topic was engaging, relevant and culturally safe for each Aboriginal community. The yarning scripts developed by the AHW were used as a tool to clearly outline their yarning approach for each topic, key conversational points, stimulus questions and learning aides, such as pictorial or physical resources to be used.
After completing all the online education sessions, participants were invited via email to complete an anonymous online survey. The survey consisted of 19 questions, 16 of which used a 5-point Likert scale and three were free text. See Supplementary Survey, Additional file . Participants were also invited to participate in a semi-structured interview about their experience of the online education. See Supplementary Interview questions, Additional file . Interviews were conducted using VoIP technologies by an investigator (KG) experienced in qualitative health research with Aboriginal participants, and who had not been involved in the design or delivery of the online education sessions. Interviews were conducted between November and December 2020. Recruitment ceased after all available AHW/EPs who participated in the online education sessions attended an interview. Sufficient data were collected to support saturation allowing for broad and deep insights related to the research question. A professional transcription service transcribed the audio recordings verbatim into text. Transcriptions were de-identified to ensure confidentiality of participants. DM and JA reviewed transcriptions against audio recordings for accuracy.
Survey data were analysed using descriptive statistics within IBM SPSS statistics version 27. The questions with the 5-point Likert scale were given numerical values of -2 (strongly disagree/never/very poor), -1 (disagree/rarely/poor), 0 (unsure/sometimes/average), 1 agree/mostly/good), 2 (strongly agree/always/very good). Interview data were managed using NVivo (QSR International Pty Ltd 2020). Reflexive thematic data analysis was applied within a framework of four central areas of AHW and EP experience participating in the online education sessions and co-designing ‘yarning’ resources, and their perception of patient experience of access to care and health literacy of COPD . Thematic analysis used the six phases outlined by Braun and Clarke consisting of familiarisation, generating codes, searching for themes, reviewing potential themes, naming the themes and producing the report . Familiarisation was facilitated by (DM, JA, SD) listening to the audio recordings and reading and rereading the transcriptions, taking notes and examining the data for meaning. Following familiarisation of the data, codes were defined inductively by (DM, JA, SD). Assumptions about code generation were checked with SD as the experienced qualitative researcher, with the perspectives of the Aboriginal researcher (DM) given deep consideration relating to their Aboriginal socio-cultural meaning. Final themes were generated based on central concepts and consensus reached by authors (DM, JA, SD, JMc). Authors (DM and JA) remained conscious of reflexivity and the impact of their perspectives (professional, personal and cultural) during study implementation, data collection, analysis and presentation of the findings to strengthen rigour, trustworthiness and overall quality of the study. Quality of coding and analysis of the data was strengthened by authors (SD and JMc) through enquiry and verification of the inductive coding framework and coded text with authors (DM and JA) .
Twelve participants, (eight AHWs and four EPs) from four NSW-based ACCHS implementing BE WELL attended online education sessions between July and December 2020 (Table ). At the time of the online education, no PTs were employed by ACCHS. The online education sessions were delivered to each of the four ACCHS separately, with timing of sessions based on their existing clinical responsibilities. Session length was between 30–60 min and sessions were delivered over 5 to 11 weeks. Survey Eleven, (7 AHW and 4 EPs) completed the survey (Table ). All (100%) participants Strongly Agreed or Agreed the online sessions were easy to understand (mean (SD)) (1.8 (0.4)), had enough information about each topic (1.5 (0.5)), helped them understand COPD (1.5 (0.5)), and that their cultural perspectives (2.0 (0.0)) and opinions (1.8 (0.4)) were valued and that the resources were helpful for patients (1.5 (0.5)) (Fig. ). Almost all (91%) participants Strongly Agreed or Agreed the yarning sessions enabled them to gain knowledge and skills on how to help patients manage COPD (1.4 (0.7)). Almost all (91%) participants Strongly Agreed that the feedback received on the topics they presented was helpful (1.6 (1.2)). All (100%) participants rated using VoIP technologies for the yarning education as Very Good or Good (1.4 (0.5)). All (100%) participants were Always or Mostly able to ask questions during the online sessions (1.8 (0.4)). Almost all (91%) participants Always and Mostly felt that the yarning sessions they prepared and delivered helped them to understand the topics better (1.5 (0.7)), that they were encouraged to incorporate their cultural knowledge into the topics they presented (1.8 (0.4)), and overall, that the sessions were useful (1.6 (0.7)). The majority (73%) of participants Strongly Disagreed or Disagreed that the sessions were too long (-0.9 (0.7)). Almost half (45%), Strongly Disagreed or Disagreed that the sessions were too often (-0.4 (1.3)), and 36% Agreed (Fig. ). Participants ranked the seven education topics from most to least useful. The most useful topic was How the lungs work and the least useful was Healthy Eating (Table ). Free-text feedback consistently reported that responding to this question was difficult as participants found each topic was equally important. Participants also responded to three open ended questions: What participants liked most; What needed improving; Any additional comments. Participants reported they liked learning about the function of the lung and COPD the most. The sessions increased their knowledge about how to present information and support patients with COPD in a culturally safe way. “It really helped my team become more confident with delivering the content during the yarnings. Culturally I got a lot out of it.” (EP_02) Several participants reported that they valued the structure of the weekly topics and presenting their own yarnings. “ The whole session was great; everything was presented well. Yarning sessions are great, very helpful. Learnt more with the weekly sessions.” (AHW_11) Suggested improvements included strengthening the formal structure of the first few online sessions, facilitators covering each topic over two sessions, further developing the role-playing by incorporating patients with COPD, delivering the education sessions over a whole day or block structure, and closely aligning education sessions with the roll out of BE WELL programs at each ACCHS. Interviews Interviews ranged from 15 to 29 min, with six interviews over 20 min. A total of 235 min with AHWs and EPs were recorded. Four themes were identified: 1. Revealing the Aboriginal lung health landscape; 2. Participating in online learning; 3. Structuring the online education sessions; 4. Co-designing with the facilitators. The themes were similar for AHWs and EPs. Variations mainly related to their differing roles within the broader BE WELL project and that none of the EPs were Aboriginal. Revealing the Aboriginal lung health landscape Professional experience Most of the AHWs reported they had only commenced their role as an AHW in the last year, so had limited to no prior experience providing clinical care or delivering health programs. Four AHWs had completed nationally recognised training in Fitness, Mental Health or Aboriginal Health. All AHWs reported no previous education about COPD or experience providing PR programs before participating in the BE WELL project. Several AHWs stated they had not previously heard about COPD. “I’ve never had no idea, or understanding of COPD, or what it was about; but you know being a part of this BE WELL Program, I’ve learnt so much.” (AHW_11) In contrast with AHWs, EPs had a much longer work experience, but each stated their current employer was the first ACCHS in which they had worked. All EPs reported little to no prior experience providing clinical care or programs for Aboriginal people with COPD. Perspectives of lung health literacy AHWs and EPs consistently perceived there was low lung health literacy within their respective Aboriginal communities. Some AHWs mentioned when lung disease was discussed, the community focused on asthma or smoking. An AHW stated that when attending smoking cessation appointments with Aboriginal patients, many reported a smoking history of 20 + years, but displayed limited lung health knowledge. EPs perceived asthma, smoking and asbestosis were discussed in the community, but thought key information about these conditions may be missing. An EP stated lung health is openly discussed due to asbestos mining in the area, with many Aboriginal people dying from asbestos-related lung disease. EPs also spoke about competing priorities for Aboriginal people's attention such as a focus on other chronic health conditions such as diabetes and heart disease rather than chronic lung diseases. This has resulted in Aboriginal patients often presenting to the ACCHS with breathlessness, unsure of the reason for their symptoms. "They will come into me and say, Look, I can't breathe very well and I don't know what's going on. And you look at their notes and they're smokers or have a diagnosis of COPD". (EP_01) Perspectives of access to lung health services AHWs and EPs perceived Aboriginal people have low access and limited knowledge of lung health services provided by ACCHS and mainstream health services, such as PR. AHWs perceived an ACCHS would be the first point of contact for Aboriginal people with COPD, but acknowledged their ACCHS had low numbers of existing patients diagnosed with COPD. The ACCHS in the BE WELL project had only recently commenced offering PR and most previous clinical discussions related to lung disease focussed on quitting smoking and nicotine replacement therapy. “We actually haven't got many clients with [ACCHS] that I know that are diagnosed, but I think they should be diagnosed since starting the BE WELL project.” (AHW_05) Since participating in the BE WELL project several AHWs reported they were now able to identify existing ACCHS patients at risk of COPD and able to refer them to PR. “And you know, ever since being able to learn more about the lungs and COPD, I'm able to identify the patients when they do come into the clinic.” (AHW_11) EPs mentioned delivering PR by their ACCHS was important, considering mainstream health services struggle to engage Aboriginal people because those services are perceived as culturally unsafe. “I don't think they go too often [to hospital]. It's not their meeting place.” (EP_01) An EP also spoke about needing to work with Aboriginal patients to overcome fear and avoidance of breathlessness during exercise training and maintaining their ongoing participation in an 8-week PR program. “The hardest thing for us now [at ACCHS] is to get participants and then get them to regularly attend.” (EP_01) EPs also spoke of the need to strengthen the cultural component of lung care by delivering programs on Country away from the clinical environment, spending additional time listening to and allowing Aboriginal people to share their ‘story’, and increasing access to care by improving referral processes within their ACCHS for assessment and management. Participating in online learning Experiences of VoIP technologies Overall AHWs reported the education sessions were successfully adjusted to an online delivery mode using VoIP technologies. AHWs positively described their experience using VoIP technologies as it enabled the educational content to be packaged into small manageable topics around other work demands. This approach supported thorough explanation of each of the seven COPD topics and enabled AHWs to ask questions and revisit educational material. "I think the shorter sessions are probably better." (EP_07) AHWs mentioned some technical challenges such as limited Wi-Fi connectivity and using a work computer with a broken camera. “I enjoyed it. – [Facilitator] went through the sessions really well. [Facilitator] explained every bit, you know, every detail, bit by bit, in the way that I was able to understand.” (AHW_ 11) The EPs perception of the use of VoIP technologies varied. Two EPs reported that although online delivery ran smoothly and they got a lot out of the online sessions, their preference would be for face-to-face delivery to enable practical and interactive education. One of these EPs also mentioned they had initially suggested that the education sessions could be provided as a one-day refresher. Positive perspectives about using VoIP technologies were that they were nearly as good as attending in person because information and pictures could be brought up on the computer screen, and questions could always be asked of the facilitators. An EP mentioned using VoIP technologies was better than providing the information quickly face-to-face over two days. From their perspective, the best aspect was having a one week break in between each session to develop yarning scripts. The same EP also reflected on the experience of VoIP technologies for AHWs. “I think particularly for the Aboriginal Health Worker they are a little bit quiet and… felt a bit of shame. I suppose both, you know, when you're talking on video conferencing and particularly in the beginning. In the end, they just opened up completely…I think it was really quite good.” (EP_01) Perspectives of the education sessions and developing yarning scripts All AHWs reported that engaging with the education sessions, preparing their own yarning scripts, applying the stimulus questions and demonstrating how they would deliver their own yarning sessions to colleagues and the facilitators reassured to them they had interpreted the seven COPD topics correctly. The sessions were reported as being a culturally safe and a self-paced learning environment, where AHWs were driving their own knowledge development, enabling them to confidently design their own yarnings. Several AHWs also spoke about the structured, clear and culturally safe way the scripts they developed enabled them to yarn about complex lung health information simply with Aboriginal people. “This isn't like a GP just reading off a piece of paper. Look, we made it into a good discussion, but we [AHWs] made it. So there's feedback, there's input to start more yarnings and kind of branch off each other. Yeah, I think yarning will be a good thing.” (AHW_03) For the EPs, the education sessions were reported as helpful and more as a refresher, complimenting their existing lung health clinical skills and strengthening their Aboriginal cultural knowledge. Consistently EPs expressed professional respect for the AHWs, emphasising the value of collegiality, shared decision-making and autonomy. An EP observed that an AHW found it confronting initially to allocate time and to prepare yarning scripts. As a result, the EP supported the AHW to better prepare and deliver their yarning scripts. EPs also reported that the education and yarning sessions enabled them to have confidence that AHWs were seeing good examples of patients with lung disease and were aware of what to look out for clinically. “We worked together and I let her present it [the yarning script], cause it's her mob [Aboriginal community], her program. So I was really supporting her delivering the program.” (EP_01) Structuring the online education sessions Perspectives of the education topics Delivering information into manageable topics was reported to increase comprehension and reduce complexity. Subsequently this scaffolding approach was reported as enabling the AHW to make greater sense of the COPD information. “Yeah, especially the medications. That was so far out of my scope at the time.” (AHW_03) EPs consistently reported the topics were pitched appropriately and included the right amount and type of information for AHWs, Aboriginal patients and covered everything they needed to know. “I think, medication wise, that's one for me that I still definitely need a little bit more assistance with, so that was good to be able to go over them again”. (EP_07) Perspectives of the resources AHWs provided consistent positive feedback about the range of resources used during the online sessions. Resources were described as ‘clear’ and ‘culturally safe’ . Resources defined as useful were the pictures, diagrams, booklets, medication charts, PowerPoint slides and videos of the correct use of inhalers and of patients with lung disease. An AHW mentioned practical teaching aides, such as placebo inhalers and spacers were resources that should be used more often in future sessions. Some AHWs stated using videos particularly helped them learn about the symptoms of lung disease. “I have not seen anyone with COPD before. Well, I may have, but I didn't know what it is. So, I thought [using videos of patients with lung disease] was great.” (AHW_03) All EPs mentioned the resources were useful in their role as supervisors of AHWs. The EPs also perceived the resources as culturally appropriate and helpful. An EP reported the resources had been valuable to their role and ACCHS beyond the BE WELL project and patients with COPD, as the format can be applied across other health programs. Future sessions could be enhanced by incorporating additional resources such as real-life scenarios and videos of patients at the beginning of each session. Perspectives about the frequency and length of the education sessions AHWs and EPs agreed that the number, length and frequency of each topic and the education sessions were broadly right. Short 30-min sessions were preferred as these allowed staff to maintain focus. An EP mentioned the timing of the education sessions made it hard for them to attend and support AHWs, and that over time attending became more of a chore. Co-designing with the facilitators Perspectives of the facilitators’ attributes AHWs perspectives of the facilitators were overwhelmingly positive. AHWs consistently described the facilitators as ‘ knowledgeable’ and perceived them as ‘culturally competent’ and ‘professionally respectful ’ and ‘understood’ the demands of the AHWs roles and their learning needs. All AHWs felt that their cultural knowledge and perspectives were respected by the facilitators. They also reported the facilitators had created a culturally safe learning environment for Aboriginal organisations and staff to engage in learning by incorporating Aboriginal protocols, such as an Acknowledgement of Country and using resources depicting Aboriginal people. AHWs described the facilitator (JA) as explaining information in ways the AHWs could understand, which increased AHWs self-reported knowledge. The facilitator had engaged the AHWs in relaxed discussions and provided feedback about the yarning scripts that the AHWs had created. This allowed AHWs to share their perspective, knowledge and yarning approaches openly and deeply. An AHW reported one occasion when the facilitators seemed ‘unapproachable’ . The AHW stated this occurred when minimal staff from their ACCHS attended the first online session which made the session confronting for a lone AHW. The experience resulted in the ACCHS regrouping and identifying how all staff could attend future sessions around clinical responsibilities. EPs described the facilitators’ attributes similarly to the AHWs. They felt the facilitators supported their learning and encouraged them to contribute to discussions exploring the applicability and adaptation of the educational content and material and the overall BE WELL program to fit the local Aboriginal community context. The EPs also stated the facilitators were responsive to the EPs and AHWs ideas and ways of being, such as how they and the ACCHS worked with community. “You know, they were really good at facilitating us to create our own yarn, and also our own ideas for how the program should be run.” (EP_01) Perspectives of the facilitators’ expectations Most AHWs described the facilitators’ expectations to attend the online sessions, prepare and present their yarnings as realistic and reasonable. AHWs felt encouraged to make choices and changes to adapt the education sessions. “[The facilitators] pretty much gave us free rein…. Well, not to talk us up, but, you know, they [the facilitators] loved what we came up with. So, yeah, like they literally gave us a lot of wriggle room.” (AHW_ 03) Two AHWs from different ACCHS highlighted the intensity and demands of the AHW role, while participating in the online education sessions. “Because our transport driver called in sick that day, so I had to fill in [and missed a session].” (AHW_08 ) Three of the four EPs mentioned the facilitators were flexible, accommodating and did not expect too much when the EPs were attending the session or helping the AHWs prepare their yarning scripts. The other EP acknowledged the role of an AHW is ‘tough’ and agreed with an AHW at their service that expectations were too high. "The expectation felt too much on a couple of the health workers, but that's not the facilitator’s fault. That's just, once again time management, and just being super busy, and juggling lots of things. It can just feel like another thing to have to do." (EP_06) Perspectives of co-design All AHWs and EPs agreed co-design occurred. AHWs reported feeling supported by the facilitators and were able to transform the COPD educational material into their own words to better engage the local Aboriginal community. “It [using co-design] was just trying to get it more culturally appropriate. So it's not just going to be like the typical GPs talk. Because that’s when you lose people, that's when your gonna lose community. So it's just a yarn.” (AHW_03) AHWs simplified the clinical information and incorporated local knowledge, terms and phrases, as well as their personal experiences of the local Aboriginal community. This approach made the yarning resources their own and was reported as relieving the pressure and expectation they initially felt co-designing the educational material. However, they felt that there were more Aboriginal cultural references that could be included in the future as they developed their knowledge and confidence and began to deliver yarnings sessions within their respective BE WELL programs. “We haven't as of yet [included local Aboriginal stories or language], because there was a lot of new information for someone learning about COPD.” (AHW_05) In addition, EPs commented the facilitators asked enquiring questions of the team about the format, session frequency, timing, topics and whether the local BE WELL teams’ needs were being met, which supported co-design and local adaptation.
Eleven, (7 AHW and 4 EPs) completed the survey (Table ). All (100%) participants Strongly Agreed or Agreed the online sessions were easy to understand (mean (SD)) (1.8 (0.4)), had enough information about each topic (1.5 (0.5)), helped them understand COPD (1.5 (0.5)), and that their cultural perspectives (2.0 (0.0)) and opinions (1.8 (0.4)) were valued and that the resources were helpful for patients (1.5 (0.5)) (Fig. ). Almost all (91%) participants Strongly Agreed or Agreed the yarning sessions enabled them to gain knowledge and skills on how to help patients manage COPD (1.4 (0.7)). Almost all (91%) participants Strongly Agreed that the feedback received on the topics they presented was helpful (1.6 (1.2)). All (100%) participants rated using VoIP technologies for the yarning education as Very Good or Good (1.4 (0.5)). All (100%) participants were Always or Mostly able to ask questions during the online sessions (1.8 (0.4)). Almost all (91%) participants Always and Mostly felt that the yarning sessions they prepared and delivered helped them to understand the topics better (1.5 (0.7)), that they were encouraged to incorporate their cultural knowledge into the topics they presented (1.8 (0.4)), and overall, that the sessions were useful (1.6 (0.7)). The majority (73%) of participants Strongly Disagreed or Disagreed that the sessions were too long (-0.9 (0.7)). Almost half (45%), Strongly Disagreed or Disagreed that the sessions were too often (-0.4 (1.3)), and 36% Agreed (Fig. ). Participants ranked the seven education topics from most to least useful. The most useful topic was How the lungs work and the least useful was Healthy Eating (Table ). Free-text feedback consistently reported that responding to this question was difficult as participants found each topic was equally important. Participants also responded to three open ended questions: What participants liked most; What needed improving; Any additional comments. Participants reported they liked learning about the function of the lung and COPD the most. The sessions increased their knowledge about how to present information and support patients with COPD in a culturally safe way. “It really helped my team become more confident with delivering the content during the yarnings. Culturally I got a lot out of it.” (EP_02) Several participants reported that they valued the structure of the weekly topics and presenting their own yarnings. “ The whole session was great; everything was presented well. Yarning sessions are great, very helpful. Learnt more with the weekly sessions.” (AHW_11) Suggested improvements included strengthening the formal structure of the first few online sessions, facilitators covering each topic over two sessions, further developing the role-playing by incorporating patients with COPD, delivering the education sessions over a whole day or block structure, and closely aligning education sessions with the roll out of BE WELL programs at each ACCHS.
Interviews ranged from 15 to 29 min, with six interviews over 20 min. A total of 235 min with AHWs and EPs were recorded. Four themes were identified: 1. Revealing the Aboriginal lung health landscape; 2. Participating in online learning; 3. Structuring the online education sessions; 4. Co-designing with the facilitators. The themes were similar for AHWs and EPs. Variations mainly related to their differing roles within the broader BE WELL project and that none of the EPs were Aboriginal.
Professional experience Most of the AHWs reported they had only commenced their role as an AHW in the last year, so had limited to no prior experience providing clinical care or delivering health programs. Four AHWs had completed nationally recognised training in Fitness, Mental Health or Aboriginal Health. All AHWs reported no previous education about COPD or experience providing PR programs before participating in the BE WELL project. Several AHWs stated they had not previously heard about COPD. “I’ve never had no idea, or understanding of COPD, or what it was about; but you know being a part of this BE WELL Program, I’ve learnt so much.” (AHW_11) In contrast with AHWs, EPs had a much longer work experience, but each stated their current employer was the first ACCHS in which they had worked. All EPs reported little to no prior experience providing clinical care or programs for Aboriginal people with COPD. Perspectives of lung health literacy AHWs and EPs consistently perceived there was low lung health literacy within their respective Aboriginal communities. Some AHWs mentioned when lung disease was discussed, the community focused on asthma or smoking. An AHW stated that when attending smoking cessation appointments with Aboriginal patients, many reported a smoking history of 20 + years, but displayed limited lung health knowledge. EPs perceived asthma, smoking and asbestosis were discussed in the community, but thought key information about these conditions may be missing. An EP stated lung health is openly discussed due to asbestos mining in the area, with many Aboriginal people dying from asbestos-related lung disease. EPs also spoke about competing priorities for Aboriginal people's attention such as a focus on other chronic health conditions such as diabetes and heart disease rather than chronic lung diseases. This has resulted in Aboriginal patients often presenting to the ACCHS with breathlessness, unsure of the reason for their symptoms. "They will come into me and say, Look, I can't breathe very well and I don't know what's going on. And you look at their notes and they're smokers or have a diagnosis of COPD". (EP_01) Perspectives of access to lung health services AHWs and EPs perceived Aboriginal people have low access and limited knowledge of lung health services provided by ACCHS and mainstream health services, such as PR. AHWs perceived an ACCHS would be the first point of contact for Aboriginal people with COPD, but acknowledged their ACCHS had low numbers of existing patients diagnosed with COPD. The ACCHS in the BE WELL project had only recently commenced offering PR and most previous clinical discussions related to lung disease focussed on quitting smoking and nicotine replacement therapy. “We actually haven't got many clients with [ACCHS] that I know that are diagnosed, but I think they should be diagnosed since starting the BE WELL project.” (AHW_05) Since participating in the BE WELL project several AHWs reported they were now able to identify existing ACCHS patients at risk of COPD and able to refer them to PR. “And you know, ever since being able to learn more about the lungs and COPD, I'm able to identify the patients when they do come into the clinic.” (AHW_11) EPs mentioned delivering PR by their ACCHS was important, considering mainstream health services struggle to engage Aboriginal people because those services are perceived as culturally unsafe. “I don't think they go too often [to hospital]. It's not their meeting place.” (EP_01) An EP also spoke about needing to work with Aboriginal patients to overcome fear and avoidance of breathlessness during exercise training and maintaining their ongoing participation in an 8-week PR program. “The hardest thing for us now [at ACCHS] is to get participants and then get them to regularly attend.” (EP_01) EPs also spoke of the need to strengthen the cultural component of lung care by delivering programs on Country away from the clinical environment, spending additional time listening to and allowing Aboriginal people to share their ‘story’, and increasing access to care by improving referral processes within their ACCHS for assessment and management.
Most of the AHWs reported they had only commenced their role as an AHW in the last year, so had limited to no prior experience providing clinical care or delivering health programs. Four AHWs had completed nationally recognised training in Fitness, Mental Health or Aboriginal Health. All AHWs reported no previous education about COPD or experience providing PR programs before participating in the BE WELL project. Several AHWs stated they had not previously heard about COPD. “I’ve never had no idea, or understanding of COPD, or what it was about; but you know being a part of this BE WELL Program, I’ve learnt so much.” (AHW_11) In contrast with AHWs, EPs had a much longer work experience, but each stated their current employer was the first ACCHS in which they had worked. All EPs reported little to no prior experience providing clinical care or programs for Aboriginal people with COPD.
AHWs and EPs consistently perceived there was low lung health literacy within their respective Aboriginal communities. Some AHWs mentioned when lung disease was discussed, the community focused on asthma or smoking. An AHW stated that when attending smoking cessation appointments with Aboriginal patients, many reported a smoking history of 20 + years, but displayed limited lung health knowledge. EPs perceived asthma, smoking and asbestosis were discussed in the community, but thought key information about these conditions may be missing. An EP stated lung health is openly discussed due to asbestos mining in the area, with many Aboriginal people dying from asbestos-related lung disease. EPs also spoke about competing priorities for Aboriginal people's attention such as a focus on other chronic health conditions such as diabetes and heart disease rather than chronic lung diseases. This has resulted in Aboriginal patients often presenting to the ACCHS with breathlessness, unsure of the reason for their symptoms. "They will come into me and say, Look, I can't breathe very well and I don't know what's going on. And you look at their notes and they're smokers or have a diagnosis of COPD". (EP_01)
AHWs and EPs perceived Aboriginal people have low access and limited knowledge of lung health services provided by ACCHS and mainstream health services, such as PR. AHWs perceived an ACCHS would be the first point of contact for Aboriginal people with COPD, but acknowledged their ACCHS had low numbers of existing patients diagnosed with COPD. The ACCHS in the BE WELL project had only recently commenced offering PR and most previous clinical discussions related to lung disease focussed on quitting smoking and nicotine replacement therapy. “We actually haven't got many clients with [ACCHS] that I know that are diagnosed, but I think they should be diagnosed since starting the BE WELL project.” (AHW_05) Since participating in the BE WELL project several AHWs reported they were now able to identify existing ACCHS patients at risk of COPD and able to refer them to PR. “And you know, ever since being able to learn more about the lungs and COPD, I'm able to identify the patients when they do come into the clinic.” (AHW_11) EPs mentioned delivering PR by their ACCHS was important, considering mainstream health services struggle to engage Aboriginal people because those services are perceived as culturally unsafe. “I don't think they go too often [to hospital]. It's not their meeting place.” (EP_01) An EP also spoke about needing to work with Aboriginal patients to overcome fear and avoidance of breathlessness during exercise training and maintaining their ongoing participation in an 8-week PR program. “The hardest thing for us now [at ACCHS] is to get participants and then get them to regularly attend.” (EP_01) EPs also spoke of the need to strengthen the cultural component of lung care by delivering programs on Country away from the clinical environment, spending additional time listening to and allowing Aboriginal people to share their ‘story’, and increasing access to care by improving referral processes within their ACCHS for assessment and management.
Experiences of VoIP technologies Overall AHWs reported the education sessions were successfully adjusted to an online delivery mode using VoIP technologies. AHWs positively described their experience using VoIP technologies as it enabled the educational content to be packaged into small manageable topics around other work demands. This approach supported thorough explanation of each of the seven COPD topics and enabled AHWs to ask questions and revisit educational material. "I think the shorter sessions are probably better." (EP_07) AHWs mentioned some technical challenges such as limited Wi-Fi connectivity and using a work computer with a broken camera. “I enjoyed it. – [Facilitator] went through the sessions really well. [Facilitator] explained every bit, you know, every detail, bit by bit, in the way that I was able to understand.” (AHW_ 11) The EPs perception of the use of VoIP technologies varied. Two EPs reported that although online delivery ran smoothly and they got a lot out of the online sessions, their preference would be for face-to-face delivery to enable practical and interactive education. One of these EPs also mentioned they had initially suggested that the education sessions could be provided as a one-day refresher. Positive perspectives about using VoIP technologies were that they were nearly as good as attending in person because information and pictures could be brought up on the computer screen, and questions could always be asked of the facilitators. An EP mentioned using VoIP technologies was better than providing the information quickly face-to-face over two days. From their perspective, the best aspect was having a one week break in between each session to develop yarning scripts. The same EP also reflected on the experience of VoIP technologies for AHWs. “I think particularly for the Aboriginal Health Worker they are a little bit quiet and… felt a bit of shame. I suppose both, you know, when you're talking on video conferencing and particularly in the beginning. In the end, they just opened up completely…I think it was really quite good.” (EP_01) Perspectives of the education sessions and developing yarning scripts All AHWs reported that engaging with the education sessions, preparing their own yarning scripts, applying the stimulus questions and demonstrating how they would deliver their own yarning sessions to colleagues and the facilitators reassured to them they had interpreted the seven COPD topics correctly. The sessions were reported as being a culturally safe and a self-paced learning environment, where AHWs were driving their own knowledge development, enabling them to confidently design their own yarnings. Several AHWs also spoke about the structured, clear and culturally safe way the scripts they developed enabled them to yarn about complex lung health information simply with Aboriginal people. “This isn't like a GP just reading off a piece of paper. Look, we made it into a good discussion, but we [AHWs] made it. So there's feedback, there's input to start more yarnings and kind of branch off each other. Yeah, I think yarning will be a good thing.” (AHW_03) For the EPs, the education sessions were reported as helpful and more as a refresher, complimenting their existing lung health clinical skills and strengthening their Aboriginal cultural knowledge. Consistently EPs expressed professional respect for the AHWs, emphasising the value of collegiality, shared decision-making and autonomy. An EP observed that an AHW found it confronting initially to allocate time and to prepare yarning scripts. As a result, the EP supported the AHW to better prepare and deliver their yarning scripts. EPs also reported that the education and yarning sessions enabled them to have confidence that AHWs were seeing good examples of patients with lung disease and were aware of what to look out for clinically. “We worked together and I let her present it [the yarning script], cause it's her mob [Aboriginal community], her program. So I was really supporting her delivering the program.” (EP_01)
Overall AHWs reported the education sessions were successfully adjusted to an online delivery mode using VoIP technologies. AHWs positively described their experience using VoIP technologies as it enabled the educational content to be packaged into small manageable topics around other work demands. This approach supported thorough explanation of each of the seven COPD topics and enabled AHWs to ask questions and revisit educational material. "I think the shorter sessions are probably better." (EP_07) AHWs mentioned some technical challenges such as limited Wi-Fi connectivity and using a work computer with a broken camera. “I enjoyed it. – [Facilitator] went through the sessions really well. [Facilitator] explained every bit, you know, every detail, bit by bit, in the way that I was able to understand.” (AHW_ 11) The EPs perception of the use of VoIP technologies varied. Two EPs reported that although online delivery ran smoothly and they got a lot out of the online sessions, their preference would be for face-to-face delivery to enable practical and interactive education. One of these EPs also mentioned they had initially suggested that the education sessions could be provided as a one-day refresher. Positive perspectives about using VoIP technologies were that they were nearly as good as attending in person because information and pictures could be brought up on the computer screen, and questions could always be asked of the facilitators. An EP mentioned using VoIP technologies was better than providing the information quickly face-to-face over two days. From their perspective, the best aspect was having a one week break in between each session to develop yarning scripts. The same EP also reflected on the experience of VoIP technologies for AHWs. “I think particularly for the Aboriginal Health Worker they are a little bit quiet and… felt a bit of shame. I suppose both, you know, when you're talking on video conferencing and particularly in the beginning. In the end, they just opened up completely…I think it was really quite good.” (EP_01)
All AHWs reported that engaging with the education sessions, preparing their own yarning scripts, applying the stimulus questions and demonstrating how they would deliver their own yarning sessions to colleagues and the facilitators reassured to them they had interpreted the seven COPD topics correctly. The sessions were reported as being a culturally safe and a self-paced learning environment, where AHWs were driving their own knowledge development, enabling them to confidently design their own yarnings. Several AHWs also spoke about the structured, clear and culturally safe way the scripts they developed enabled them to yarn about complex lung health information simply with Aboriginal people. “This isn't like a GP just reading off a piece of paper. Look, we made it into a good discussion, but we [AHWs] made it. So there's feedback, there's input to start more yarnings and kind of branch off each other. Yeah, I think yarning will be a good thing.” (AHW_03) For the EPs, the education sessions were reported as helpful and more as a refresher, complimenting their existing lung health clinical skills and strengthening their Aboriginal cultural knowledge. Consistently EPs expressed professional respect for the AHWs, emphasising the value of collegiality, shared decision-making and autonomy. An EP observed that an AHW found it confronting initially to allocate time and to prepare yarning scripts. As a result, the EP supported the AHW to better prepare and deliver their yarning scripts. EPs also reported that the education and yarning sessions enabled them to have confidence that AHWs were seeing good examples of patients with lung disease and were aware of what to look out for clinically. “We worked together and I let her present it [the yarning script], cause it's her mob [Aboriginal community], her program. So I was really supporting her delivering the program.” (EP_01)
Perspectives of the education topics Delivering information into manageable topics was reported to increase comprehension and reduce complexity. Subsequently this scaffolding approach was reported as enabling the AHW to make greater sense of the COPD information. “Yeah, especially the medications. That was so far out of my scope at the time.” (AHW_03) EPs consistently reported the topics were pitched appropriately and included the right amount and type of information for AHWs, Aboriginal patients and covered everything they needed to know. “I think, medication wise, that's one for me that I still definitely need a little bit more assistance with, so that was good to be able to go over them again”. (EP_07) Perspectives of the resources AHWs provided consistent positive feedback about the range of resources used during the online sessions. Resources were described as ‘clear’ and ‘culturally safe’ . Resources defined as useful were the pictures, diagrams, booklets, medication charts, PowerPoint slides and videos of the correct use of inhalers and of patients with lung disease. An AHW mentioned practical teaching aides, such as placebo inhalers and spacers were resources that should be used more often in future sessions. Some AHWs stated using videos particularly helped them learn about the symptoms of lung disease. “I have not seen anyone with COPD before. Well, I may have, but I didn't know what it is. So, I thought [using videos of patients with lung disease] was great.” (AHW_03) All EPs mentioned the resources were useful in their role as supervisors of AHWs. The EPs also perceived the resources as culturally appropriate and helpful. An EP reported the resources had been valuable to their role and ACCHS beyond the BE WELL project and patients with COPD, as the format can be applied across other health programs. Future sessions could be enhanced by incorporating additional resources such as real-life scenarios and videos of patients at the beginning of each session. Perspectives about the frequency and length of the education sessions AHWs and EPs agreed that the number, length and frequency of each topic and the education sessions were broadly right. Short 30-min sessions were preferred as these allowed staff to maintain focus. An EP mentioned the timing of the education sessions made it hard for them to attend and support AHWs, and that over time attending became more of a chore.
Delivering information into manageable topics was reported to increase comprehension and reduce complexity. Subsequently this scaffolding approach was reported as enabling the AHW to make greater sense of the COPD information. “Yeah, especially the medications. That was so far out of my scope at the time.” (AHW_03) EPs consistently reported the topics were pitched appropriately and included the right amount and type of information for AHWs, Aboriginal patients and covered everything they needed to know. “I think, medication wise, that's one for me that I still definitely need a little bit more assistance with, so that was good to be able to go over them again”. (EP_07)
AHWs provided consistent positive feedback about the range of resources used during the online sessions. Resources were described as ‘clear’ and ‘culturally safe’ . Resources defined as useful were the pictures, diagrams, booklets, medication charts, PowerPoint slides and videos of the correct use of inhalers and of patients with lung disease. An AHW mentioned practical teaching aides, such as placebo inhalers and spacers were resources that should be used more often in future sessions. Some AHWs stated using videos particularly helped them learn about the symptoms of lung disease. “I have not seen anyone with COPD before. Well, I may have, but I didn't know what it is. So, I thought [using videos of patients with lung disease] was great.” (AHW_03) All EPs mentioned the resources were useful in their role as supervisors of AHWs. The EPs also perceived the resources as culturally appropriate and helpful. An EP reported the resources had been valuable to their role and ACCHS beyond the BE WELL project and patients with COPD, as the format can be applied across other health programs. Future sessions could be enhanced by incorporating additional resources such as real-life scenarios and videos of patients at the beginning of each session.
AHWs and EPs agreed that the number, length and frequency of each topic and the education sessions were broadly right. Short 30-min sessions were preferred as these allowed staff to maintain focus. An EP mentioned the timing of the education sessions made it hard for them to attend and support AHWs, and that over time attending became more of a chore.
Perspectives of the facilitators’ attributes AHWs perspectives of the facilitators were overwhelmingly positive. AHWs consistently described the facilitators as ‘ knowledgeable’ and perceived them as ‘culturally competent’ and ‘professionally respectful ’ and ‘understood’ the demands of the AHWs roles and their learning needs. All AHWs felt that their cultural knowledge and perspectives were respected by the facilitators. They also reported the facilitators had created a culturally safe learning environment for Aboriginal organisations and staff to engage in learning by incorporating Aboriginal protocols, such as an Acknowledgement of Country and using resources depicting Aboriginal people. AHWs described the facilitator (JA) as explaining information in ways the AHWs could understand, which increased AHWs self-reported knowledge. The facilitator had engaged the AHWs in relaxed discussions and provided feedback about the yarning scripts that the AHWs had created. This allowed AHWs to share their perspective, knowledge and yarning approaches openly and deeply. An AHW reported one occasion when the facilitators seemed ‘unapproachable’ . The AHW stated this occurred when minimal staff from their ACCHS attended the first online session which made the session confronting for a lone AHW. The experience resulted in the ACCHS regrouping and identifying how all staff could attend future sessions around clinical responsibilities. EPs described the facilitators’ attributes similarly to the AHWs. They felt the facilitators supported their learning and encouraged them to contribute to discussions exploring the applicability and adaptation of the educational content and material and the overall BE WELL program to fit the local Aboriginal community context. The EPs also stated the facilitators were responsive to the EPs and AHWs ideas and ways of being, such as how they and the ACCHS worked with community. “You know, they were really good at facilitating us to create our own yarn, and also our own ideas for how the program should be run.” (EP_01) Perspectives of the facilitators’ expectations Most AHWs described the facilitators’ expectations to attend the online sessions, prepare and present their yarnings as realistic and reasonable. AHWs felt encouraged to make choices and changes to adapt the education sessions. “[The facilitators] pretty much gave us free rein…. Well, not to talk us up, but, you know, they [the facilitators] loved what we came up with. So, yeah, like they literally gave us a lot of wriggle room.” (AHW_ 03) Two AHWs from different ACCHS highlighted the intensity and demands of the AHW role, while participating in the online education sessions. “Because our transport driver called in sick that day, so I had to fill in [and missed a session].” (AHW_08 ) Three of the four EPs mentioned the facilitators were flexible, accommodating and did not expect too much when the EPs were attending the session or helping the AHWs prepare their yarning scripts. The other EP acknowledged the role of an AHW is ‘tough’ and agreed with an AHW at their service that expectations were too high. "The expectation felt too much on a couple of the health workers, but that's not the facilitator’s fault. That's just, once again time management, and just being super busy, and juggling lots of things. It can just feel like another thing to have to do." (EP_06) Perspectives of co-design All AHWs and EPs agreed co-design occurred. AHWs reported feeling supported by the facilitators and were able to transform the COPD educational material into their own words to better engage the local Aboriginal community. “It [using co-design] was just trying to get it more culturally appropriate. So it's not just going to be like the typical GPs talk. Because that’s when you lose people, that's when your gonna lose community. So it's just a yarn.” (AHW_03) AHWs simplified the clinical information and incorporated local knowledge, terms and phrases, as well as their personal experiences of the local Aboriginal community. This approach made the yarning resources their own and was reported as relieving the pressure and expectation they initially felt co-designing the educational material. However, they felt that there were more Aboriginal cultural references that could be included in the future as they developed their knowledge and confidence and began to deliver yarnings sessions within their respective BE WELL programs. “We haven't as of yet [included local Aboriginal stories or language], because there was a lot of new information for someone learning about COPD.” (AHW_05) In addition, EPs commented the facilitators asked enquiring questions of the team about the format, session frequency, timing, topics and whether the local BE WELL teams’ needs were being met, which supported co-design and local adaptation.
AHWs perspectives of the facilitators were overwhelmingly positive. AHWs consistently described the facilitators as ‘ knowledgeable’ and perceived them as ‘culturally competent’ and ‘professionally respectful ’ and ‘understood’ the demands of the AHWs roles and their learning needs. All AHWs felt that their cultural knowledge and perspectives were respected by the facilitators. They also reported the facilitators had created a culturally safe learning environment for Aboriginal organisations and staff to engage in learning by incorporating Aboriginal protocols, such as an Acknowledgement of Country and using resources depicting Aboriginal people. AHWs described the facilitator (JA) as explaining information in ways the AHWs could understand, which increased AHWs self-reported knowledge. The facilitator had engaged the AHWs in relaxed discussions and provided feedback about the yarning scripts that the AHWs had created. This allowed AHWs to share their perspective, knowledge and yarning approaches openly and deeply. An AHW reported one occasion when the facilitators seemed ‘unapproachable’ . The AHW stated this occurred when minimal staff from their ACCHS attended the first online session which made the session confronting for a lone AHW. The experience resulted in the ACCHS regrouping and identifying how all staff could attend future sessions around clinical responsibilities. EPs described the facilitators’ attributes similarly to the AHWs. They felt the facilitators supported their learning and encouraged them to contribute to discussions exploring the applicability and adaptation of the educational content and material and the overall BE WELL program to fit the local Aboriginal community context. The EPs also stated the facilitators were responsive to the EPs and AHWs ideas and ways of being, such as how they and the ACCHS worked with community. “You know, they were really good at facilitating us to create our own yarn, and also our own ideas for how the program should be run.” (EP_01)
Most AHWs described the facilitators’ expectations to attend the online sessions, prepare and present their yarnings as realistic and reasonable. AHWs felt encouraged to make choices and changes to adapt the education sessions. “[The facilitators] pretty much gave us free rein…. Well, not to talk us up, but, you know, they [the facilitators] loved what we came up with. So, yeah, like they literally gave us a lot of wriggle room.” (AHW_ 03) Two AHWs from different ACCHS highlighted the intensity and demands of the AHW role, while participating in the online education sessions. “Because our transport driver called in sick that day, so I had to fill in [and missed a session].” (AHW_08 ) Three of the four EPs mentioned the facilitators were flexible, accommodating and did not expect too much when the EPs were attending the session or helping the AHWs prepare their yarning scripts. The other EP acknowledged the role of an AHW is ‘tough’ and agreed with an AHW at their service that expectations were too high. "The expectation felt too much on a couple of the health workers, but that's not the facilitator’s fault. That's just, once again time management, and just being super busy, and juggling lots of things. It can just feel like another thing to have to do." (EP_06)
All AHWs and EPs agreed co-design occurred. AHWs reported feeling supported by the facilitators and were able to transform the COPD educational material into their own words to better engage the local Aboriginal community. “It [using co-design] was just trying to get it more culturally appropriate. So it's not just going to be like the typical GPs talk. Because that’s when you lose people, that's when your gonna lose community. So it's just a yarn.” (AHW_03) AHWs simplified the clinical information and incorporated local knowledge, terms and phrases, as well as their personal experiences of the local Aboriginal community. This approach made the yarning resources their own and was reported as relieving the pressure and expectation they initially felt co-designing the educational material. However, they felt that there were more Aboriginal cultural references that could be included in the future as they developed their knowledge and confidence and began to deliver yarnings sessions within their respective BE WELL programs. “We haven't as of yet [included local Aboriginal stories or language], because there was a lot of new information for someone learning about COPD.” (AHW_05) In addition, EPs commented the facilitators asked enquiring questions of the team about the format, session frequency, timing, topics and whether the local BE WELL teams’ needs were being met, which supported co-design and local adaptation.
The study aimed to explore the experiences of AHWs and EPs from four ACCHS engaged in online education sessions to build knowledge and skills to co-design education ‘yarning’ material for Aboriginal people with COPD. The quantitative survey highlighted the benefits of the online education sessions to build AHWs and EPs capability to co-design COPD yarning resources. The online sessions were also reported as being helpful in allowing AHWs and EPs to support Aboriginal patients with COPD. Overall, the online learning space that was created in the study allowed AHWs to feel their cultural perspectives and professional opinions were valued and respected. Many findings of the survey were crystalised in the qualitative interviews, supporting triangulation . Participants, particularly AHWs, engaged easily with online education using VoIP and reported that the sessions supported self-paced learning, where they could apply their lived experiences and knowledge of their local Aboriginal community to develop reflective, effective and culturally-centred COPD yarning resources. This study adds to the limited PR literature about building health professionals knowledge and skills around the management of people with COPD [ , , ], as well as limited studies evaluating participants’ experiences of participating in co-design . To our knowledge, this is the first study exploring online education for AHWs and EPs to co-design COPD educational ‘yarning’ material for Aboriginal people with COPD guided by 8 Ways of learning. Online education Advancements in technology-enabled learning are revolutionising face-to-face education towards online platforms and classrooms, which has potential to support learning and self-management of patients with COPD, PR programs broadly, as well as how health professionals access and engage with education . The online education sessions were designed to be participant-centred, responding to AHWs’ and EPs’ training needs and placed value on the use of co-design and 8 Ways of learning. The value of incorporating an Aboriginal pedagogy framework to improve culturally safe and accessible mainstream services is being realised within the NSW public health sector, with the first study published focusing on introducing 8 Ways of learning guided by participatory action research to shape project and program design . Variations were made to the format and structure of the online education sessions to align with AHWs’ and EPs’ socio-cultural, professional and the contexts of each of the four ACCHS. Of note was how positively AHWs perceived online education. The AHWs engaged with the learning material and teaching approaches, which enabled a supportive and flexible learning environment to be created. It has been reported in other studies that Aboriginal people are early adopters of technology and partners in digital health programs and research . In this study it was reported that using VoIP technologies enabled information to be paced and different teaching techniques applied to thoroughly explore specific aspects of COPD. The structure of the online education sessions allowed adequate time from when AHWs and EPs received new COPD knowledge, adapted this information to the local Aboriginal community and presented their own yarning sessions. The structured scaffolding approach of building knowledge across the seven topics enabled complicated information to be easily understood by the AHWs, even in topics, such as medications that they initially believed were outside their scope and ability . Culturally safe learning environment AHWs and EPs highlighted that the online education sessions were culturally safe spaces to ask questions and seek advice and feedback. AHWs expressed they experienced professional independence and autonomy and collaborated with facilitators and EPs, which supported their knowledge development and facilitated local accountability to confidently co-design their own yarning material. EPs specifically found the online education sessions which used 8 Ways of learning culturally helpful, especially for their supervisory role with AHWs. The sessions also further enhanced their existing clinical knowledge and experience in managing people with COPD. The EPs’ cultural competence and support of AHWs was evident. This was important as AHWs within the ACCHS sector have reported negative experiences working with allied health professionals . EPs consistently expressed cultural respect and professional humility towards the value of AHWs role within the ACCHS and their communities. EPs actively encouraged AHWs to apply their professional and personal autonomy, knowledge and skills. EPs listened to AHWs’ perspectives when they explored COPD from an Aboriginal worldview, and openly shared their own clinical interactions working with and understanding Aboriginal patients lived experience of COPD, and reasons they may choose to engage (or not engage) in services provided by ACCHS. As a result, AHWs, supported by EPs, successfully co-designed their own COPD yarning resources which were reflective of the Aboriginal communities they serve. AHWs also self-reported improved knowledge, skills and ability in providing care for Aboriginal people with chronic lung disease. These findings aligned with Browne et al., that highlighted the importance and benefit of an equal partnership and working with culturally competent allied health professionals who display reciprocity of two-way learning and who are committed to building AHWs’ capabilities . These findings also acknowledge the importance of using 8 Ways of learning and investing in similar frameworks that equally value and incorporate Western and Indigenous knowledges and systems in health research and education, such as Both-Ways Seeing or Two-Eyed Seeing [ – ]. While there was an integration of Western and Indigenous knowledges and systems, the topics covered during the online education sessions maintained a Western understanding of health and disease. Aboriginal landscape Overall, the study revealed COPD in Aboriginal communities is poorly understood and could be under-detected and that services for managing COPD, such as PR, are underdeveloped. AHWs and EPs perceived limited COPD health literacy and low access to lung services within ACCHS or mainstream health services by the local Aboriginal community. Although limited studies and no national data are available, it has been reported that Aboriginal people are highly likely to be at risk of poorer health literacy , and low uptake of PR . AHWs and EPs consistently perceived mainstream services were unable to deliver culturally safe lung health services for Aboriginal people, which is reflected in other studies about access to mainstream health care . Low knowledge and awareness of COPD and PR has been found in the mainstream rural and remote health workforce , and was mirrored with AHWs and EPs in this study. Through the online education sessions AHW’s increased their self-reported knowledge, awareness and management of COPD, and their skills to support the early detection and diagnosis of COPD in Aboriginal people. ACCHS have a long and successful history of delivering high quality culturally safe care, with AHWs playing a critical role as the first point of contact and cultural brokers, facilitating Aboriginal peoples’ access to health promotion, education, and clinical services . Whilst there is an imperative to improve the cultural safety of mainstream PR programs, an opportunity exists for the ACCHS sector to lead COPD service provision, workforce development and implement PR that better meets the needs of Aboriginal people in the community and increases access to care . Local adaptation Consistently, cultural adaptation occurred around ways of talking and communicating (‘yarning’) to make COPD information and engagement with the material conversational. Developing research skills of AHWs is particularly important during Aboriginal research . AHWs, even those new to the profession, bring significant knowledge and experience of their community and how the community accesses health care and engages with health education and programs . Our research confirms that in many instances, AHWs drew from their lived experiences and engagement with their local Aboriginal community to adapt complex clinical information to co-design COPD yarning resources. AHWs incorporated their own Aboriginal worldviews, phrasing and experiences to fit the context of the ACCHS and reflect the local Aboriginal community, which allowed AHWs to translate, not replicate the material verbatim. Further, the yarning scripts developed aligned with how the AHWs and the ACCHS engage and communicate health information with the local Aboriginal community. The value of cultural adaptation or Indigenising health care, teaching and education is broadly known [ , , ]. It is anticipated that as AHWs work more closely with patients living with COPD who attend the BE WELL PR program, and they begin delivering yarning sessions, inclusion of Aboriginal language, stories and deeper cultural understanding related to COPD may occur. Challenges The study highlighted some challenges for implementing online education sessions. Some AHWs and EPs reported pressures participating in the study and attending the online sessions while meeting other work responsibilities, staying engaged over time, and preparing the yarning scripts too far in advance of the ACCHS commencing their BE WELL PR Program. These findings are consistent with Farnbach et al. , which revealed the increased pressure experienced by staff involved in the dual roles of clinical practice within the community and engaging with research . This pressure of dual roles was consistently reported by one ACCHS, likely illustrating when co-design principles were poorly applied, and the timing, structure and format of the online education sessions did not entirely align with their needs. This feedback highlights the challenge for participants working within an ACCHS to find time to engage in training while maintaining their clinical roles and community responsibilities. Co-design emphasises the importance of sharing power and maintaining equal partnerships with participants, which can be a delicate balance, considering time pressures of participants . This feedback emphasises the importance of using participatory engagement and shared decision making to mitigate pressures for participants to engage in training, which are more effective than top-down approaches . Strengths and limitations Initial coding was conducted independently by three researchers (DM, JA, SD). The results represent the experiences of AHWs and EPs by using their voices as study participants. The engagement of Aboriginal and non-Aboriginal members of the research team during the design, delivery and analysis strengthens the inclusion of Aboriginal perspectives and meaning of the data. Two members of the research team (JA and DM) facilitated the online education sessions with close and ongoing involvement with participants. They developed deep knowledge about the insights, benefits and strengths of the study, and identified ‘objective distance’ was required during data collection, because both were close to the study and mindful of the potential positive and negative effects on research quality. To reduce researcher bias, an independent member of the research team (KG) with no contact with the participants during the online education sessions conducted the qualitative interviews using VoIP technologies. It is acknowledged additional research is required to validate the impact of VoIP technologies on the quality of participants’ experiences of qualitative data collection . Participants completed the survey and interview between two to eight weeks following the online education sessions, the latter might have led to recall bias. Although staff from four diverse NSW-based ACCHS participated in the online education sessions and contributed data, findings may not be generalisable to other ACCHS. However, the data provide insights that online education can be successful in building ACCHS workforce capacity and enabling co-design of culturally safe education resources. As such the online environment may provide greater potential for scale-up of the education component of PR across the ACCHS sector. Future research should explore whether online education of AHWs and EPs to develop yarnings about COPD management, improves patient health outcomes, enables behavioural changes and increases access to culturally safe PR. Future researchers should also be conscious of the importance of an iterative and flexible process, with a genuine commitment to establishing and maintaining equal partnerships with participants.
Advancements in technology-enabled learning are revolutionising face-to-face education towards online platforms and classrooms, which has potential to support learning and self-management of patients with COPD, PR programs broadly, as well as how health professionals access and engage with education . The online education sessions were designed to be participant-centred, responding to AHWs’ and EPs’ training needs and placed value on the use of co-design and 8 Ways of learning. The value of incorporating an Aboriginal pedagogy framework to improve culturally safe and accessible mainstream services is being realised within the NSW public health sector, with the first study published focusing on introducing 8 Ways of learning guided by participatory action research to shape project and program design . Variations were made to the format and structure of the online education sessions to align with AHWs’ and EPs’ socio-cultural, professional and the contexts of each of the four ACCHS. Of note was how positively AHWs perceived online education. The AHWs engaged with the learning material and teaching approaches, which enabled a supportive and flexible learning environment to be created. It has been reported in other studies that Aboriginal people are early adopters of technology and partners in digital health programs and research . In this study it was reported that using VoIP technologies enabled information to be paced and different teaching techniques applied to thoroughly explore specific aspects of COPD. The structure of the online education sessions allowed adequate time from when AHWs and EPs received new COPD knowledge, adapted this information to the local Aboriginal community and presented their own yarning sessions. The structured scaffolding approach of building knowledge across the seven topics enabled complicated information to be easily understood by the AHWs, even in topics, such as medications that they initially believed were outside their scope and ability .
AHWs and EPs highlighted that the online education sessions were culturally safe spaces to ask questions and seek advice and feedback. AHWs expressed they experienced professional independence and autonomy and collaborated with facilitators and EPs, which supported their knowledge development and facilitated local accountability to confidently co-design their own yarning material. EPs specifically found the online education sessions which used 8 Ways of learning culturally helpful, especially for their supervisory role with AHWs. The sessions also further enhanced their existing clinical knowledge and experience in managing people with COPD. The EPs’ cultural competence and support of AHWs was evident. This was important as AHWs within the ACCHS sector have reported negative experiences working with allied health professionals . EPs consistently expressed cultural respect and professional humility towards the value of AHWs role within the ACCHS and their communities. EPs actively encouraged AHWs to apply their professional and personal autonomy, knowledge and skills. EPs listened to AHWs’ perspectives when they explored COPD from an Aboriginal worldview, and openly shared their own clinical interactions working with and understanding Aboriginal patients lived experience of COPD, and reasons they may choose to engage (or not engage) in services provided by ACCHS. As a result, AHWs, supported by EPs, successfully co-designed their own COPD yarning resources which were reflective of the Aboriginal communities they serve. AHWs also self-reported improved knowledge, skills and ability in providing care for Aboriginal people with chronic lung disease. These findings aligned with Browne et al., that highlighted the importance and benefit of an equal partnership and working with culturally competent allied health professionals who display reciprocity of two-way learning and who are committed to building AHWs’ capabilities . These findings also acknowledge the importance of using 8 Ways of learning and investing in similar frameworks that equally value and incorporate Western and Indigenous knowledges and systems in health research and education, such as Both-Ways Seeing or Two-Eyed Seeing [ – ]. While there was an integration of Western and Indigenous knowledges and systems, the topics covered during the online education sessions maintained a Western understanding of health and disease.
Overall, the study revealed COPD in Aboriginal communities is poorly understood and could be under-detected and that services for managing COPD, such as PR, are underdeveloped. AHWs and EPs perceived limited COPD health literacy and low access to lung services within ACCHS or mainstream health services by the local Aboriginal community. Although limited studies and no national data are available, it has been reported that Aboriginal people are highly likely to be at risk of poorer health literacy , and low uptake of PR . AHWs and EPs consistently perceived mainstream services were unable to deliver culturally safe lung health services for Aboriginal people, which is reflected in other studies about access to mainstream health care . Low knowledge and awareness of COPD and PR has been found in the mainstream rural and remote health workforce , and was mirrored with AHWs and EPs in this study. Through the online education sessions AHW’s increased their self-reported knowledge, awareness and management of COPD, and their skills to support the early detection and diagnosis of COPD in Aboriginal people. ACCHS have a long and successful history of delivering high quality culturally safe care, with AHWs playing a critical role as the first point of contact and cultural brokers, facilitating Aboriginal peoples’ access to health promotion, education, and clinical services . Whilst there is an imperative to improve the cultural safety of mainstream PR programs, an opportunity exists for the ACCHS sector to lead COPD service provision, workforce development and implement PR that better meets the needs of Aboriginal people in the community and increases access to care .
Consistently, cultural adaptation occurred around ways of talking and communicating (‘yarning’) to make COPD information and engagement with the material conversational. Developing research skills of AHWs is particularly important during Aboriginal research . AHWs, even those new to the profession, bring significant knowledge and experience of their community and how the community accesses health care and engages with health education and programs . Our research confirms that in many instances, AHWs drew from their lived experiences and engagement with their local Aboriginal community to adapt complex clinical information to co-design COPD yarning resources. AHWs incorporated their own Aboriginal worldviews, phrasing and experiences to fit the context of the ACCHS and reflect the local Aboriginal community, which allowed AHWs to translate, not replicate the material verbatim. Further, the yarning scripts developed aligned with how the AHWs and the ACCHS engage and communicate health information with the local Aboriginal community. The value of cultural adaptation or Indigenising health care, teaching and education is broadly known [ , , ]. It is anticipated that as AHWs work more closely with patients living with COPD who attend the BE WELL PR program, and they begin delivering yarning sessions, inclusion of Aboriginal language, stories and deeper cultural understanding related to COPD may occur.
The study highlighted some challenges for implementing online education sessions. Some AHWs and EPs reported pressures participating in the study and attending the online sessions while meeting other work responsibilities, staying engaged over time, and preparing the yarning scripts too far in advance of the ACCHS commencing their BE WELL PR Program. These findings are consistent with Farnbach et al. , which revealed the increased pressure experienced by staff involved in the dual roles of clinical practice within the community and engaging with research . This pressure of dual roles was consistently reported by one ACCHS, likely illustrating when co-design principles were poorly applied, and the timing, structure and format of the online education sessions did not entirely align with their needs. This feedback highlights the challenge for participants working within an ACCHS to find time to engage in training while maintaining their clinical roles and community responsibilities. Co-design emphasises the importance of sharing power and maintaining equal partnerships with participants, which can be a delicate balance, considering time pressures of participants . This feedback emphasises the importance of using participatory engagement and shared decision making to mitigate pressures for participants to engage in training, which are more effective than top-down approaches .
Initial coding was conducted independently by three researchers (DM, JA, SD). The results represent the experiences of AHWs and EPs by using their voices as study participants. The engagement of Aboriginal and non-Aboriginal members of the research team during the design, delivery and analysis strengthens the inclusion of Aboriginal perspectives and meaning of the data. Two members of the research team (JA and DM) facilitated the online education sessions with close and ongoing involvement with participants. They developed deep knowledge about the insights, benefits and strengths of the study, and identified ‘objective distance’ was required during data collection, because both were close to the study and mindful of the potential positive and negative effects on research quality. To reduce researcher bias, an independent member of the research team (KG) with no contact with the participants during the online education sessions conducted the qualitative interviews using VoIP technologies. It is acknowledged additional research is required to validate the impact of VoIP technologies on the quality of participants’ experiences of qualitative data collection . Participants completed the survey and interview between two to eight weeks following the online education sessions, the latter might have led to recall bias. Although staff from four diverse NSW-based ACCHS participated in the online education sessions and contributed data, findings may not be generalisable to other ACCHS. However, the data provide insights that online education can be successful in building ACCHS workforce capacity and enabling co-design of culturally safe education resources. As such the online environment may provide greater potential for scale-up of the education component of PR across the ACCHS sector. Future research should explore whether online education of AHWs and EPs to develop yarnings about COPD management, improves patient health outcomes, enables behavioural changes and increases access to culturally safe PR. Future researchers should also be conscious of the importance of an iterative and flexible process, with a genuine commitment to establishing and maintaining equal partnerships with participants.
Online education using co-design and 8 Ways of learning was rated highly by AHWs and EPs for improving COPD knowledge and valuing cultural perspectives. The use of co-design principles and 8 Ways of learning with AHWs and EPs supported the cultural adaptation of COPD resources for Aboriginal people with COPD. The findings demonstrated that building culturally safe learning environments can provide effective online learning spaces that increase knowledge, skills and confidence and promote effective participation and collaboration. Key to these findings was respect for AHWs’ cultural knowledge and lived experiences and deeply valuing Aboriginal ways of learning and communicating through yarning.
Additional file 1. Survey. BE WELL online education participant survey tool. Additional file 2. Interview questions. BE WELL online education participant interview questions.
|
Environmental Investigation and Surveillance for | 254d1410-9ee6-4d59-9307-b90fe379783f | 10063469 | Microbiology[mh] | Legionellosis is caused by the gram-negative bacterium Legionella . This infection is the consequence of environmental exposure to pathogenic legionellae that are ubiquitous in both water and moist soil ecosystems. The severity of the disease varies from a mild febrile illness (Pontiac fever) (incubation period commonly 24–48 h) to a serious and sometimes fatal form of pneumonia (Legionnaires’ disease) (incubation period commonly 2–10 days) . In Aotearoa New Zealand (NZ), legionellosis has been a notifiable disease since June 1980 . Inhalation of aerosolized bacteria from an environmental source is the usual means of Legionella transmission, usually from engineered environments such as wet cooling towers and water distribution systems, which act as reservoirs and amplifiers of the bacteria . Furthermore, the inhalation of contaminated aerosolized dust from the handling of compost and potting mix is likely to be an important transmission route that contributes to the cases of legionellosis caused by Legionella longbeachae , which is the predominant species causing disease in NZ . In NZ sporadic community cases of legionellosis comprise at least 90% of all reported cases compared to travel-associated (6.5%, ) and hospital-acquired (< 1%, ). A total of 2628 cases of legionellosis were notified between 2000 and 2020 (representing an overall mean annual incidence rate of 2.7/100 000 population). The mean annual incidence rate increased from 1.6/100 000 population in 2000–2009 to 3.9/100 000 population in 2010–2020, with much of this increase attributed to improved awareness and testing . NZ has limited public health resources, which means that there is likely to be variability in the intensity of case investigations nationally, that is, whether sampling is deemed appropriate as part of a case follow up. Public health practitioners need to know the sources of sporadic Legionella so they can effectively target control measures . The goal of this study is to provide a finer lens on the environmental sources of Legionella that contribute to human disease in NZ to support improved prevention and control measures. Specific aims are to: (1) describe the environmental sources identified for outbreaks of legionellosis; (2) describe the sources implicated for sporadic cases; and (3) summarize the extensive environmental sampling data collected from a variety of sources. These aims seek to address a knowledge gap identified in the literature on the lack of Legionella studies that support definite and probable ranked sources .
In NZ legionellosis surveillance (including cases of LD and Pontiac fever) comprises both laboratory and disease notification data (including outbreaks) collected for public health purposes . Surveillance data is collated by the Institute of Environmental Science and Research (ESR) on behalf of the NZ Ministry of Health. Outbreaks are defined as two or more cases associated with a single identified site of exposure with dates of onset within 6 months of each other . Data on legionellosis outbreaks reported from 2000–2020 were obtained from ESR. The outbreaks were delineated by monthly distribution, location, outbreak source, Legionella species (spp.) and serogroup (sg), number of confirmed cases, hospitalizations, and deaths. We analysed source data for Legionella bacteria isolated from all environmental sources collected by the ESR Legionella Laboratory between 2000 and 2020. These included all of those identified from case and outbreak investigations and from systematic environmental testing conducted for risk assessment and compliance testing purposes. The data (all environmental isolates) were amalgamated to assess the prevalence of different Legionella strains from the full range of source categories (known reservoirs and exposure sources for this bacteria). A further analysis of Legionella culture isolates by infective organism (species and serogroup) from environmental source category for the 5-year period 2016–2020 was also undertaken. For the purpose of this study, we have used a classification system adapted from Orkis et al., 2018 with three defined categories: a definitive source where there is a molecular match (using sequence-based typing or mip gene sequence analysis between the clinical isolate and the environmental isolate from the source to which the case was exposed during their incubation period; a probable source where there is a species match between the case and their environmental source but identification was made using incongruent methods, such as serology for the clinical case and culture for the linked environmental source; and a suspected source where there is a likely source identified, but no environmental sampling is undertaken; for example, cases of L. longbeachae infection where the interviewed case reports using compost during the incubation period, no other exposure risks were identified, but no sampling/testing of the compost materials was undertaken. We also include data for the clinical cases where no infection source was indicated, and no environmental sampling was undertaken.
Legionella Outbreaks Seventeen outbreaks with an identified source were detected during the 2000–2020 period (Table ). These were mainly traced to either a cooling tower, a spa pool, or a compost source. No outbreaks were identified in years 2001, 2004, 2008–2010, 2012, 2014, and 2016–2019, indicating that these events are not commonly identified in NZ. In most years when outbreaks were reported, there were multiple outbreaks. The 92 outbreak cases represented 3.5% of all notifications ( N = 2628) for this period. The number of legionellosis cases associated with these outbreaks ranged from two to 19. A total of 54 outbreak-associated cases were hospitalized (hospitalization information was not recorded for one outbreak). Five outbreak-associated cases died (Table ). Collectively, L. pneumophila has been the sole causative agent in 9 of 17 outbreaks since 2000. Seven of the outbreaks were attributed to L. pneumophila serogroup 1 (Lpsg1), with the source identified as either cooling towers (three outbreaks), spa pools (3) or rainwater tanks (1). Two further outbreaks were attributed to display spa pools caused by Lpsg2. L. longbeachae caused six outbreaks attributed to compost or potting mix exposure, with five caused by L. longbeachae serogroup 1 (Llsg1) and one caused by Llsg2. A further outbreak in 2015 attributed to cooling towers caused mixed infections with Lpsg1 and Llsg1 (which were isolated from the towers), along with infections with Lpsg5, Llsg2, and L sainthelensi (which weren’t isolated from the towers). A further cooling tower outbreak was caused by L. dumoffii . Eight of the 17 outbreaks occurred in spring, while four each occurred in summer and autumn, compared to only one during winter. The measured Legionella concentration for 10 outbreaks varied widely depending on the identified source, ranging from 47 cfu/mL (public spa pool) to 2400 cfu/mL (cooling tower). Overall, the Legionella level in cooling towers involved in outbreaks was 1-to-2 logs higher than those seen in recreational water sources (Table ). Clinical Case Source Investigation Data Supplementary Table A1 describes individual clinical cases with an epidemiological link to an environmental source for the Legionella infection during the 5-year period from 2016 to 2020. A source was classified as definite, probable, or suspect, as described in the methods. A total of 17 different source categories were identified as contaminated with Legionella bacteria between 2016 and 2020, with 14 definitive, 67 probable, and 541 suspected sources, giving a total of 622 cases. The most commonly identified sources associated with Legionella infections were compost and gardening activities with a total of 571 cases (Supplementary Table A1). Water-based systems only accounted for 40 cases (Supplementary Table A1). Environmental Isolate Data Between 2000 and 2020, a total of 2245 separate environmental isolates of Legionella bacteria were identified from a variety of sources, the majority identified at the species level (Supplementary Table A2). Supplementary Table A2 shows the diversity of species that have been isolated from the environment, since 2000 as part of routine source investigation for sporadic and outbreak cases, as well as for risk assessment and surveillance sampling. Legionella spp. isolated from water sources such as cooling towers ranked the highest ( N = 816) followed by compost/mulch/potting mix ( N = 671) and buildings with complex water distribution systems such as hospitals, aged care homes, and hotels ( N = 483). The most common Legionella species isolated from hot water distribution systems was L. anisa ( N = 191, or 39.5%), while the most common isolate from cooling towers was L. pneumophila ( N = 544, or 66.7%). L. longbeachae isolates made up 55.1% ( N = 370) of the 671 isolates from compost, with L. pneumophila isolates contributing 20.4% ( N = 137) and L. bozemanae being the next most common with 71 isolates (10.6%). Although there was an observed increase in L. longbeachae isolates from compost between the two decades of note, is that an increase in other species, such as L. pneumophila was also detected in the same organic source categories between the two decades, increasing from N = 24, 2000–2009, to N = 133, 2010–2020. Our examination of environmental sources for Legionella infection in NZ shows considerable diversity, with both water and soil-based sources contributing to sporadic and outbreak cases. Isolate testing also detected less commonly encountered pathogenic (e.g., L. gratiana, L. parisiensis, and L. steelei ) and non-pathogenic (e.g., L. moravica and L. tunisiensis ) Legionella species from a limited number of ecological niches. The data reviewed here indicate that cooling towers play a key role as the sources of recognized legionellosis outbreaks in NZ. Environmental sampling confirmed the common association of L. pneumophila with cooling towers and L. longbeachae with compost. However, it also revealed that such associations are not absolute and that some commonly isolated environmental species are not common human pathogens, for example, L. anisa. This finding is likely to be due to the difference in comparative pathogenicity of L. anisa compared to L. pneumophila . This study confirms that recognized outbreaks make only a small contribution to legionellosis case totals in NZ. However, where a source was identified it was always one where aerosolization provides a mechanism for wide dispersion of the contaminated material, be it water or compost. Table records all reported outbreaks since 2000 and shows that although cooling towers were the attributed source of 29.4% (5/17) of outbreaks, they accounted for more than half (58.7 or 54/92) of outbreak-associated legionellosis cases. Spa pools were also attributed as the source of 29.4% (5/17) of outbreaks, but only 14.1% (13/92) of these cases, while compost material was attributed to more than a third (35.3% or 6/17) of all outbreaks, but only 14.1% (13/92) of cases. The significant difference in number of cases associated with each of the major contamination sources is probably due to different dispersal factors and environmental conditions. Cooling towers are common, especially in urban and industrial settings, and capable of discharging aerosolized water contaminated with Legionella bacteria over a wide area, potentially exposing many people in the immediate environs as well as those further afield . Sporadic cases of legionellosis are very rarely linked to cooling towers, primarily because these are not routinely sampled when undertaking source investigation for individual legionellosis cases, and this may indicate a deficiency in current testing and sampling protocols. Of note is that there is no widely accepted definition of dangerous Legionella concentration levels in the water of cooling tower systems internationally . Spa pools have been linked to both outbreaks and sporadic cases of legionellosis, and their preponderance in such cases reflects their widespread use, both privately and publicly. Invariably, cases are restricted to those that have been exposed to aerosols generated either actively by water jets or by passive dispersion from the water surface and are limited to those in the immediate vicinity of the spa pool. The cases do not have to be users of the pool to be infected. Invariably, investigations for both spa pool and cooling tower outbreaks have identified that the major contributing factors to their cause are inadequate maintenance and cleaning regimes. Outbreaks and sporadic cases of legionellosis attributed to compost and other soil conditioners follow the inhalation of aerosolized dust containing Legionella bacteria and often occur during spring months . Aerosols from this material are usually created when bagged material is emptied or mechanically or manually moved . Invariably, close contact (less than two metres) is required and outbreaks in this setting usually involve two or more people moving large volumes of material or sustained exposure for long periods of time (hours, not minutes). Anecdotally, there have been reports of family members moving a ‘trailer load of compost by shovel and wheelbarrow around the garden’ before presenting days later with symptoms and subsequently being diagnosed with legionellosis. Physical exertion may result in deeper or more rapid breathing, increasing the likelihood of infection following exposure. In 2012 a L. pneumophila sg1 cluster of 19 cases with two deaths occurred in the Greater Auckland areas between March and May of that year . Despite epidemiological, laboratory, and geospatial investigations revealing no common source or sources, it was widely suspected that poorly maintained cooling towers were the likely cause and so mass dosing of all such towers with biocide was recommended via media releases . In the absence of a mandatory register of wet cooling towers, a bylaw was introduced in Auckland requiring owners to register their industrial wet cooling tower systems annually, to monitor Legionella bacteria levels, regularly clean and maintain their cooling towers, and send test results to the Council. There have been no recorded outbreaks of legionellosis due to Lpsg1 in the five years since the bylaw was adopted in 2015. The results of this analysis show that there is considerable potential to improve the completeness of source identification, which would provide a stronger evidence base for action to prevent infections. Supplementary Table A1 shows that several reservoirs were classified as a suspected source as the supporting evidence was weak, usually relying on self-reporting with no environmental sampling. The justification for identification as a suspected source is based on a probabilistic scenario with the elimination of all common sources except the one implicated and is based on information obtained from case interviews. However, these reservoirs and sources may play an important role in the transmission of Legionella bacteria, resulting in an underestimation of their importance as a source without physical examination and by attempting to culture legionellae from them. In the absence of regulation aimed at reducing Legionella growth and transmission and to optimize source investigation, it is imperative that knowledge about all potential sources of Legionella is obtained . A more systematic approach to environmental sampling for Legionella could also improve the evidence base to guide prevention measures. Table A2 points to a decrease in active environmental surveillance in contrast to clinical surveillance, which appears to have increased throughout the study period. These findings raise concerns since environmental surveillance is a crucial component of proactive risk assessment and supports actions to reduce exposure to Legionella spp. and prevent infections. Any risk assessment for control measures against Legionella should be performed not only in response to laboratory-identified cases of disease, but also on a regular basis to prevent the disease . For example, it is noteworthy the gradual decline in NZ since 2014 (2014; N = 79, 2020: N = 11) of positive isolates from cooling towers being referred to the centralised laboratory for further confirmatory testing and typing which, as evidenced by this study, is of concern since they have been responsible for three outbreaks since 2010 (Table ). Previous studies have demonstrated that Legionella colonization of cooling towers is common, so it would be prudent to assume that all towers have some degree of colonization . It is noteworthy that, as with other jurisdictions , clinical cases of Lp have been matched with Lp detected in soil/potting mix/compost samples. This reinforces the importance of considering exposures to organic matter (soils, potting mix, and compost) when investigating cases of LD, especially in circumstances where there is an absence of exposure to an artificial or natural water reservoir. The main strength of this study is the use of pooled data on Legionella environmental reservoirs from laboratory and disease notification data, which highlighted the diversity of Legionella species from a variety of environmental sources. Supplementary Tables A1 and A2 list several different artificial or natural reservoirs that support the proliferation and transmission of Legionella species to susceptible human hosts most likely due to the aerosolized material containing legionellae. The sources listed in Supplementary Table A2 invariably include water systems that often contain stagnant or recirculating warm water, or reservoirs composed of rich organic material (compost, potting mixes, sewage effluent, and leaf mulch). However, a limitation of this study is associated with the use of routinely collected surveillance data. It is possible that environmental sources of Legionella have been detected outside of ESR, as direct notification of laboratory-confirmed environmental Legionella samples is not mandatory in NZ.
Outbreaks Seventeen outbreaks with an identified source were detected during the 2000–2020 period (Table ). These were mainly traced to either a cooling tower, a spa pool, or a compost source. No outbreaks were identified in years 2001, 2004, 2008–2010, 2012, 2014, and 2016–2019, indicating that these events are not commonly identified in NZ. In most years when outbreaks were reported, there were multiple outbreaks. The 92 outbreak cases represented 3.5% of all notifications ( N = 2628) for this period. The number of legionellosis cases associated with these outbreaks ranged from two to 19. A total of 54 outbreak-associated cases were hospitalized (hospitalization information was not recorded for one outbreak). Five outbreak-associated cases died (Table ). Collectively, L. pneumophila has been the sole causative agent in 9 of 17 outbreaks since 2000. Seven of the outbreaks were attributed to L. pneumophila serogroup 1 (Lpsg1), with the source identified as either cooling towers (three outbreaks), spa pools (3) or rainwater tanks (1). Two further outbreaks were attributed to display spa pools caused by Lpsg2. L. longbeachae caused six outbreaks attributed to compost or potting mix exposure, with five caused by L. longbeachae serogroup 1 (Llsg1) and one caused by Llsg2. A further outbreak in 2015 attributed to cooling towers caused mixed infections with Lpsg1 and Llsg1 (which were isolated from the towers), along with infections with Lpsg5, Llsg2, and L sainthelensi (which weren’t isolated from the towers). A further cooling tower outbreak was caused by L. dumoffii . Eight of the 17 outbreaks occurred in spring, while four each occurred in summer and autumn, compared to only one during winter. The measured Legionella concentration for 10 outbreaks varied widely depending on the identified source, ranging from 47 cfu/mL (public spa pool) to 2400 cfu/mL (cooling tower). Overall, the Legionella level in cooling towers involved in outbreaks was 1-to-2 logs higher than those seen in recreational water sources (Table ).
Supplementary Table A1 describes individual clinical cases with an epidemiological link to an environmental source for the Legionella infection during the 5-year period from 2016 to 2020. A source was classified as definite, probable, or suspect, as described in the methods. A total of 17 different source categories were identified as contaminated with Legionella bacteria between 2016 and 2020, with 14 definitive, 67 probable, and 541 suspected sources, giving a total of 622 cases. The most commonly identified sources associated with Legionella infections were compost and gardening activities with a total of 571 cases (Supplementary Table A1). Water-based systems only accounted for 40 cases (Supplementary Table A1).
Between 2000 and 2020, a total of 2245 separate environmental isolates of Legionella bacteria were identified from a variety of sources, the majority identified at the species level (Supplementary Table A2). Supplementary Table A2 shows the diversity of species that have been isolated from the environment, since 2000 as part of routine source investigation for sporadic and outbreak cases, as well as for risk assessment and surveillance sampling. Legionella spp. isolated from water sources such as cooling towers ranked the highest ( N = 816) followed by compost/mulch/potting mix ( N = 671) and buildings with complex water distribution systems such as hospitals, aged care homes, and hotels ( N = 483). The most common Legionella species isolated from hot water distribution systems was L. anisa ( N = 191, or 39.5%), while the most common isolate from cooling towers was L. pneumophila ( N = 544, or 66.7%). L. longbeachae isolates made up 55.1% ( N = 370) of the 671 isolates from compost, with L. pneumophila isolates contributing 20.4% ( N = 137) and L. bozemanae being the next most common with 71 isolates (10.6%). Although there was an observed increase in L. longbeachae isolates from compost between the two decades of note, is that an increase in other species, such as L. pneumophila was also detected in the same organic source categories between the two decades, increasing from N = 24, 2000–2009, to N = 133, 2010–2020. Our examination of environmental sources for Legionella infection in NZ shows considerable diversity, with both water and soil-based sources contributing to sporadic and outbreak cases. Isolate testing also detected less commonly encountered pathogenic (e.g., L. gratiana, L. parisiensis, and L. steelei ) and non-pathogenic (e.g., L. moravica and L. tunisiensis ) Legionella species from a limited number of ecological niches. The data reviewed here indicate that cooling towers play a key role as the sources of recognized legionellosis outbreaks in NZ. Environmental sampling confirmed the common association of L. pneumophila with cooling towers and L. longbeachae with compost. However, it also revealed that such associations are not absolute and that some commonly isolated environmental species are not common human pathogens, for example, L. anisa. This finding is likely to be due to the difference in comparative pathogenicity of L. anisa compared to L. pneumophila . This study confirms that recognized outbreaks make only a small contribution to legionellosis case totals in NZ. However, where a source was identified it was always one where aerosolization provides a mechanism for wide dispersion of the contaminated material, be it water or compost. Table records all reported outbreaks since 2000 and shows that although cooling towers were the attributed source of 29.4% (5/17) of outbreaks, they accounted for more than half (58.7 or 54/92) of outbreak-associated legionellosis cases. Spa pools were also attributed as the source of 29.4% (5/17) of outbreaks, but only 14.1% (13/92) of these cases, while compost material was attributed to more than a third (35.3% or 6/17) of all outbreaks, but only 14.1% (13/92) of cases. The significant difference in number of cases associated with each of the major contamination sources is probably due to different dispersal factors and environmental conditions. Cooling towers are common, especially in urban and industrial settings, and capable of discharging aerosolized water contaminated with Legionella bacteria over a wide area, potentially exposing many people in the immediate environs as well as those further afield . Sporadic cases of legionellosis are very rarely linked to cooling towers, primarily because these are not routinely sampled when undertaking source investigation for individual legionellosis cases, and this may indicate a deficiency in current testing and sampling protocols. Of note is that there is no widely accepted definition of dangerous Legionella concentration levels in the water of cooling tower systems internationally . Spa pools have been linked to both outbreaks and sporadic cases of legionellosis, and their preponderance in such cases reflects their widespread use, both privately and publicly. Invariably, cases are restricted to those that have been exposed to aerosols generated either actively by water jets or by passive dispersion from the water surface and are limited to those in the immediate vicinity of the spa pool. The cases do not have to be users of the pool to be infected. Invariably, investigations for both spa pool and cooling tower outbreaks have identified that the major contributing factors to their cause are inadequate maintenance and cleaning regimes. Outbreaks and sporadic cases of legionellosis attributed to compost and other soil conditioners follow the inhalation of aerosolized dust containing Legionella bacteria and often occur during spring months . Aerosols from this material are usually created when bagged material is emptied or mechanically or manually moved . Invariably, close contact (less than two metres) is required and outbreaks in this setting usually involve two or more people moving large volumes of material or sustained exposure for long periods of time (hours, not minutes). Anecdotally, there have been reports of family members moving a ‘trailer load of compost by shovel and wheelbarrow around the garden’ before presenting days later with symptoms and subsequently being diagnosed with legionellosis. Physical exertion may result in deeper or more rapid breathing, increasing the likelihood of infection following exposure. In 2012 a L. pneumophila sg1 cluster of 19 cases with two deaths occurred in the Greater Auckland areas between March and May of that year . Despite epidemiological, laboratory, and geospatial investigations revealing no common source or sources, it was widely suspected that poorly maintained cooling towers were the likely cause and so mass dosing of all such towers with biocide was recommended via media releases . In the absence of a mandatory register of wet cooling towers, a bylaw was introduced in Auckland requiring owners to register their industrial wet cooling tower systems annually, to monitor Legionella bacteria levels, regularly clean and maintain their cooling towers, and send test results to the Council. There have been no recorded outbreaks of legionellosis due to Lpsg1 in the five years since the bylaw was adopted in 2015. The results of this analysis show that there is considerable potential to improve the completeness of source identification, which would provide a stronger evidence base for action to prevent infections. Supplementary Table A1 shows that several reservoirs were classified as a suspected source as the supporting evidence was weak, usually relying on self-reporting with no environmental sampling. The justification for identification as a suspected source is based on a probabilistic scenario with the elimination of all common sources except the one implicated and is based on information obtained from case interviews. However, these reservoirs and sources may play an important role in the transmission of Legionella bacteria, resulting in an underestimation of their importance as a source without physical examination and by attempting to culture legionellae from them. In the absence of regulation aimed at reducing Legionella growth and transmission and to optimize source investigation, it is imperative that knowledge about all potential sources of Legionella is obtained . A more systematic approach to environmental sampling for Legionella could also improve the evidence base to guide prevention measures. Table A2 points to a decrease in active environmental surveillance in contrast to clinical surveillance, which appears to have increased throughout the study period. These findings raise concerns since environmental surveillance is a crucial component of proactive risk assessment and supports actions to reduce exposure to Legionella spp. and prevent infections. Any risk assessment for control measures against Legionella should be performed not only in response to laboratory-identified cases of disease, but also on a regular basis to prevent the disease . For example, it is noteworthy the gradual decline in NZ since 2014 (2014; N = 79, 2020: N = 11) of positive isolates from cooling towers being referred to the centralised laboratory for further confirmatory testing and typing which, as evidenced by this study, is of concern since they have been responsible for three outbreaks since 2010 (Table ). Previous studies have demonstrated that Legionella colonization of cooling towers is common, so it would be prudent to assume that all towers have some degree of colonization . It is noteworthy that, as with other jurisdictions , clinical cases of Lp have been matched with Lp detected in soil/potting mix/compost samples. This reinforces the importance of considering exposures to organic matter (soils, potting mix, and compost) when investigating cases of LD, especially in circumstances where there is an absence of exposure to an artificial or natural water reservoir. The main strength of this study is the use of pooled data on Legionella environmental reservoirs from laboratory and disease notification data, which highlighted the diversity of Legionella species from a variety of environmental sources. Supplementary Tables A1 and A2 list several different artificial or natural reservoirs that support the proliferation and transmission of Legionella species to susceptible human hosts most likely due to the aerosolized material containing legionellae. The sources listed in Supplementary Table A2 invariably include water systems that often contain stagnant or recirculating warm water, or reservoirs composed of rich organic material (compost, potting mixes, sewage effluent, and leaf mulch). However, a limitation of this study is associated with the use of routinely collected surveillance data. It is possible that environmental sources of Legionella have been detected outside of ESR, as direct notification of laboratory-confirmed environmental Legionella samples is not mandatory in NZ.
The implications of this research are that the environmental ecology of Legionella species including species that are infrequently reported, remains pertinent to the prevention of legionellosis. The findings of our study point to the need to put more emphasis on developing control measures in NZ particularly in relation to industrial water systems (cooling towers) and compost exposure. Our findings indicate that Legionella can potentially be a significant hazard in high-risk settings such as hospitals and aged and residential care facilities. For example, 71.2%, N = 341/479 isolates identified from hot-water systems came from healthcare settings. As a potential public health risk, consideration is needed of how Legionella can be managed in such settings and thought should be given to making environmental sources of Legionella notifiable in NZ. A greater understanding of the environmental sources of Legionella will also improve targeted prevention efforts. The value of environmental surveillance data would be enhanced if all test requests included a standard set of key data, including the reasons for testing.
Below is the link to the electronic supplementary material. Supplementary file1 (DOCX 93 KB)
|
Top 10 Histological Mimics of Neuroendocrine Carcinoma You Should Not Miss in the Head and Neck | 250e2a6b-cfec-439b-85dc-887d3c5ec954 | 10063750 | Anatomy[mh] | Neuroendocrine neoplasia (NEN) of the head and neck region are spectacular lesions, often accompanied by unique morphological, immunohistochemical, hormonal, and/or genetic features. While most of these lesions mainly occur sporadically in adult patients, subsets are intimately coupled to genetically inherit syndromic disease and may therefore present in younger patients . In addition to the demographic diversity, neuroendocrine tumors (NETs) may be notoriously difficult to prognosticate—as the metastatic potential may be challenging to assess by morphology alone. Consequently, pathologists have developed risk stratification algorithms for various NENs in order to assess the risk of disease progression, with the proliferation index (as estimated by mitotic index and/or the Ki-67 labeling index) proving particularly important . Indeed, these parameters are nowadays routinely used in grading NENs, with neuroendocrine carcinoma (NEC) displaying the highest proliferation index . Due to the highly proliferative nature of this entity, distant metastases and death due to disease are common outcomes for the NEC patient category . Therefore, it is imperative to correctly identify these lesions in a timely fashion, and care must be taken to not confuse NEC with malignant, non-NEC neoplasms with focal immunoreactivity to neuroendocrine markers, as well as the clinically more indolent well-differentiated NETs.
The most clinically urgent NEN subtype is NEC, a poorly differentiated, highly proliferative, malignant tumor of poor prognosis exhibiting significant tumor necrosis and destructively invasive features . In the head and neck region, NEC may develop within the paranasal sinuses, the nasal cavity, the oro- and hypopharynx, the salivary glands, the oral cavity, and the larynx . NEC of the head and neck region is further divided into small cell neuroendocrine carcinoma (SCNEC), large cell neuroendocrine carcinoma (LCNEC), and mixed NEC with non-neuroendocrine neoplasms . While SCNEC is composed of diffusely arranged small cells with a high nuclear/cytoplasmic ratio, nuclear molding, and necrosis (Fig. A), LCNEC usually exhibits hyperchromatic and pleomorphic tumor cell nuclei (sometimes palisading) with prominent nucleoli (Fig. B) arranged in nests, sheets, or trabeculae. Necrosis is often widespread. Immunohistochemistry (IHC) is usually positive for neuroendocrine markers, such as chromogranin A and synaptophysin (Fig. C, ), but expression of both markers may be absent in some lesions . Inclusion of second-generation neuroendocrine markers, such as INSM1, may be of value in these instances . Both SCNEC and LCNEC express keratins, and the expression may be faint and dot like in the former entity. This feature can be particularly helpful in distinguishing from paragangliomas. The Ki-67 proliferation index is usually high, always > 20%, and frequently between 55 and 100% (Fig. E, ). Subsets of oropharyngeal NECs may be human papillomavirus (HPV)-driven neoplasms, exhibiting strong p16 immunoreactivity . Most head and neck NECs are positive for p53 by immunohistochemistry and negative for retinoblastoma protein (pRb), which is due to frequent somatic mutations in the tumor suppressors TP53 and RB1 . Interestingly, subsets of cases also harbor pathogenic variants in potentially actionable therapeutic target genes associated with the NOTCH and PI3K/AKT/mTOR pathways .
A plethora of entities of diverse lineages must be excluded when considering a diagnosis of NEC. On one hand, the cellular, closely packed monotonous tumor cells in SCNEC bring nearly all small round blue cell tumors into the differential diagnoses, while on the other hand, miscellaneous head and neck epithelial malignancies remain close mimics of LCNEC. The challenges are particularly heightened when dealing with small tissue volumes in biopsy material. Awareness of the histomorphologic spectrum, immunophenotypic nuances, and molecular traits of the diverse mimics and employing a stepwise algorithmic approach can aid in reaching an accurate diagnosis (Fig. , Table ). Given the possible differential diagnostic dilemmas discussed above, we provide a review of the top ten morphological and immunohistochemical NEC mimics that the practicing pathologist should not miss. Well-Differentiated Neuroendocrine Tumors Well-differentiated NETs (WDNETs) encompass well-differentiated neoplasms with neuroendocrine features recognizable on light microscopy, i.e., exhibiting cellular monotony, random anisonucleosis, stippled nuclear chromatin, and varied growth patterns (nests, trabeculae, cords, festoons, rosettes). These tumors are strongly immunoreactive with neuroendocrine markers. WDNETs and NECs are clinically and genetically distinct entities with divergent treatments and outcomes. Therefore, a clear distinction between WDNET and NEC is necessary. This is usually not problematic, however, tends to be challenging in cases with limited or crushed tissue. Pituitary Neuroendocrine Tumor (PitNET) Pituitary neuroendocrine tumors (PitNETs) are well-differentiated adenohypophyseal lesions that may cause a wide variety of symptoms depending on its specific hormone production. PitNETs may be encountered in the sinonasal tract, most frequently in the sphenoid sinus, due to invasion from a sellar tumor or rarely as ectopic PitNET. The previous terminology “pituitary adenoma” and “pituitary carcinoma” are no longer recommended . PitNETs should be subtyped in terms of tumor cell lineage and expression of pituitary hormones by the use of IHC, and tumors are derived from a PIT1 lineage (i.e., somatotroph, lactotroph, and thyrotroph), a TPIT lineage (corticotroph tumors), an SF1 lineage (gonadotroph tumors), or tumors without a distinct lineage (plurihormonal lesions and hormonally silent “null cell tumors”) . Most PitNETs are characterized by a hypercellular and well-differentiated mass composed of monomorphic tumor cells (Fig. A). Although there are morphological clues regarding the PitNET subtype, the distinction is not entirely reliable without IHC and therefore all PitNETs should be assessed for PIT1, TPIT, and SF1 immunoreactivity . Hormone stains are also recommended, including growth hormone (GH), prolactin (PRL), and beta-thyroid-stimulating hormone (β-TSH) for PIT1 lineage tumors; adrenocorticotropic hormone (ACTH) for TPIT lineage tumors; and beta-follicle-stimulating hormone (β-FSH) and beta-luteinizing hormone (β-LH) for SF1 lineage tumors . Keratin IHC can also be useful (CAM5.2, AE1/AE3, and/or CK18) to subtype somatotroph PitNETs as either densely or sparsely granulated, characterized by a perinuclear or globular cytoplasmic stain, respectively . Metastatic PitNETs are rarely encountered, and when metastatic, most commonly affect liver, bone, lung, and lymph nodes . Medullary Thyroid Carcinoma (MTC) Medullary thyroid carcinoma (MTC) is a neuroendocrine tumor derived from the calcitonin-producing C cells of the thyroid gland. Although the majority of tumors arise sporadically, up to 25% are thought to be associated with multiple endocrine neoplasia type 2 (MEN2A or MEN2B) syndrome, with the affected patient demonstrating activating, constitutional pathogenic variants in the RET proto-oncogene . Sporadic tumors are usually driven by somatic RET mutations or mutually exclusive RAS gene mutations . There are numerous morphological MTC patterns reported, but they are not routinely classified on a histological basis as there is no established correlation to either genotype or clinical outcomes . MTCs are characterized by round (sometimes plasmacytoid, polygonal, or spindle shaped) cells in nests with an interdigitating stroma exhibiting various amounts of amyloid deposition (Fig. B). The cytoplasm is usually amphophilic and granular due to their secretory content. MTCs routinely express neuroendocrine markers as well as signs of thyroid differentiation (TTF1), whereas monoclonal PAX8 expression is lacking, as is thyroglobulin (the latter a consequence of the non-follicular cell origin). The hallmark of MTCs is calcitonin immunoreactivity, although the stain can vary in intensity and spatial distribution. In the metastatic setting without an established thyroid lesion, care must be taken not to prematurely assume that a neuroendocrine tumor with focal calcitonin and TTF1 expression is a metastatic MTC—as cases of laryngeal NETs with aberrant expression of these markers have been reported . In terms of prognosis, the 2022 World Health Organization (WHO) classification of endocrine and neuroendocrine tumors recommends that MTCs be graded based on the mitotic index, the presence of tumor necrosis, and the Ki-67 proliferation index, in which high-grade lesions display necrosis, a mitotic count ≥ 5 per 2 mm 2 , and/or a Ki-67 index of ≥ 5% . In terms of metastatic disease, most MTCs spread regionally to neck lymph nodes, but subsets of cases may also spread to the liver, lungs, and bone. It may be worth noting that rare cases of MTC metastatic to the parotid and pituitary glands have been reported, which potentially could constitute differential diagnostic conundrums . Moreover, subsets of MTCs may display a small cell phenotype, further complicating the histological work-up if NEC is suspected . Paraganglioma (PGL) Head and neck paragangliomas (PGLs) are usually parasympathetic, non-functioning neuroendocrine tumors, which sets them apart from their infra-diaphragmatic, norepinephrine, and/or epinephrine-producing counterparts . They are collectively the most inheritable of all human neoplasia with approximately 40% of patients carrying an underlying constitutional genetic event, while the metastatic potential of these lesions is usually low . Patients with constitutional pathogenic variants in succinate dehydrogenase (SDH) subunit A , B , C , D , or AF2 ( SDHA, SDHB, SDHC, SDHD, SDHAF2 ) harbor an increased risk of developing head and neck paraganglioma with a low risk of disseminated disease . The underlying molecular biology is complex, with tumors showing a higher risk of metastases often driven by mutations in tricarboxylic acid (TCA) cycle genes (not only restricted to SDH genes) that will lead to TCA cycle arrest and accumulation of early metabolites, which in turn may activate oncogenic hypoxia-inducible factor (HIF) pathways . However, subsets of cases are driven by mutations in various kinase-associated pathways, and these lesions usually tend to be non-metastatic. Therefore, there is a well-developed genotype–phenotype correlation which can be assessed by histopathology and immunohistochemistry: a positive SDHB immunostain strongly argues against mutational inactivation of either SDHB , SDHC , or SDHD genes—in turn arguing against (although not excluding) the risk of metastatic potential . When presenting in characteristic locations, such as the carotid bifurcation or the jugulotympanic area, a head and neck paraganglioma is quite easily distinguished by morphological assessment. The tumor cells are usually round to oval and arranged in small nests (so-called “zellballen”) embedded in a highly vascular stroma (Fig. C). The cytoplasm is granular with an amphophilic or basophilic appearance, and nuclear pleomorphism is usually limited to absent. Mitoses and tumor necrosis are rarely detected. Tumor cells are positive for chromogranin A, synaptophysin, and GATA3, while consistently keratin negative . An S100 protein (or SOX10) immunohistochemistry identifies the sustentacular network of cells supporting the tumor cells—but the finding of sustentacular cells is not diagnostic for paraganglioma, as other neuroendocrine tumors also may exhibit this feature . Using functional IHC, most head and neck paragangliomas are positive for choline acetyltransferase, an enzyme in the acetylcholine biosynthesis pathway, while often negative for enzymes responsible for catecholamine production, such as tyrosine hydroxylase . Metastatic head and neck paraganglioma usually spread to regional lymph nodes, while distant site involvement is rare . Metastastic Neuroendocrine Neoplasia of Unknown Primary NENs of unknown primary (NEN-UPs) are metastatic lesions without a known primary tumor location, a finding reported in 12–22% of NEN patients . The importance of identifying the primary site cannot be underestimated given that the various clinical and prognostic features of NENs depend on the tumor origin site. There are several morphological clues that can be used to properly identity a NEN-UP, including amyloid deposits in MTC, psammoma bodies in somatostatinoma, a hyalinized stroma in insulinoma, as well as the hyaline globules and basophilic cytoplasm of pheochromocytoma . Even so, it is not unusual for a metastatic NEN-UP to be characterized by nested cells with little or no morphological findings unique to the primary tumor site (Fig. D). From an immunohistochemical perspective, various combinations of neuroendocrine marker and transcription factors results may be useful. For example, TTF1 may help identify pulmonary carcinoids and MTC, PDX1 may assist in recognizing NENs of the upper gastrointestinal tract including the pancreas, whereas CDX2 and SATB2 may highlight NENs of the lower gastrointestinal tract. In addition, testing for various hormones may be useful, including calcitonin for MTC, serotonin for lower gastrointestinal NENs, islet hormones for pancreatic NENs, and GLP1 for rectal NENs, to name just a few . NECs outside of the head and neck area occasionally may metastasize to the jaws and major salivary glands . Indeed, metastatic neuroendocrine tumors to the parotid gland accounted for 22% of all metastatic tumors to this organ in a recent case series, and most cases were either pulmonary NECs, Merkel cell carcinomas (MCCs), or MTCs . If not previously known, a hypothetical IHC panel for NEN-UPs metastatic to the salivary glands would therefore need at least TTF1, calcitonin, and CK20, in addition to neuroendocrine markers and Ki-67. Merkel Cell Carcinoma Merkel cell carcinoma (MCC) is a rare neuroendocrine carcinoma of the skin with an estimated incidence of 2.2 cases per million person-years, afflicting predominantly older patients . The tumors are either driven by UV-induced mutations or by a Merkel cell polyoma virus infection, and the exact proportion of these etiologies varies with geographic distribution . From a morphological perspective, MCCs are composed of solid arrangements of monomorphic tumor cells with a high nuclear/cytoplasmic ratio, smudged nuclear chromatin, indistinct nucleoli, and displaying innumerable mitoses (Fig. A, B). The IHC profile is characteristically neuroendocrine , while a perinuclear, “dot-like” keratin stain (most strikingly with CK20) is characteristic of MCC (Fig. C). The Ki-67 proliferation index is usually exceedingly high (> 90%). Moreover, virus-driven MCCs are positive for the Merkel cell polyoma antigen (Fig. D). When presenting as a primary tumor, the diagnosis is usually quite straight-forward, but metastatic lesions may cause diagnostic difficulties if the primary tumor is not known. To complicate matters even more, subsets of MCCs have been reported to originate from mucosal linings of the upper respiratory and GI tracts and might be clinically silent . NUT Carcinoma NUT carcinoma is a highly aggressive tumor primarily affecting young patients, often presenting in the midline of the thorax and head and neck regions . On the histological level, NUT carcinoma is composed of small to intermediate cells with an undifferentiated, primitive, and monotonous appearance (Fig. A). Mitotic figures and necrosis are easily identified. A significant subset exhibits abrupt keratinization (Fig. B). Using IHC, NUT carcinomas are epithelial neoplasms, reacting with keratins and squamous markers, such as CK5/6, p63, and p40 (Fig. C). CD34 is also positive in approximately 50% of cases . NUT carcinoma is driven by NUTM1 gene rearrangements , and NUT protein immunohistochemistry (Fig. D) is useful to highlight this genetic aberrancy—as NUT protein expression is not normally seen in cells outside of the testis and ciliary ganglion . Interestingly, neuroendocrine differentiation has been reported, with a high level of suspicion required when considering NEC in young patients by incorporating NUT immunohistochemistry . Sinonasal Undifferentiated Carcinoma Sinonasal undifferentiated carcinoma (SNUC) is a rare but highly aggressive epithelial neoplasia lacking morphological and immunohistochemical evidence of lineage (including glandular, squamous, neuroendocrine, or mesenchymal differentiation) . Thus, it is a diagnosis of exclusion in which a broad range of possible differential diagnoses must be considered. Usually presenting as a large mass in the sinonasal tract, these tumors are often invasive at diagnosis . SNUC exhibits high-grade histology with uniform tumor cells growing in sheets, lobules, nests or trabeculae, lacking squamous, or glandular differentiation. Tumor cells express keratins, while p40 is negative and p63 may exhibit weak and unspecific staining (Fig. ). Subsets of SNUC may express patchy chromogranin A and/or synaptophysin immunoreactivity, making them a potential differential diagnosis in the work-up of NEC of the head and neck region . Somatic IDH2 mutations have been identified in large subsets of SNUC and are readily identifiable using sequencing analysis, while IDH immunohistochemistry has proven inconsistent in pinpointing IDH2-mutated SNUC . Basaloid Squamous Cell Carcinoma Squamous cell carcinoma (SCC) is the most common histological type of cancer in the head and neck region. While the diagnosis of a differentiated SCC is not problematic, the basaloid SCC subtype is a common differential diagnosis of SCNEC, particularly in biopsy material. Basaloid SCC remains an uncommon malignancy that is associated with aggressive clinical behavior and poor median survival (18 months) . Closely packed basaloid cells and lack of significant keratinization typically impart a blue cell tumor appearance at low power that resembles SCNEC. Lobules, adenoid/pseudoglandular structures or variably anastomosing islands of tumor cells exhibiting peripheral palisading, thickened basement membrane-like material, and central comedonecrosis are typical histological features (Fig. A) . The neoplastic cells show pleomorphic hyperchromatic nuclei with scanty cytoplasm. The presence of carcinoma in situ or areas of abrupt squamous differentiation (keratin pearl formation) are useful clues to the diagnosis. Mitoses are usually easily identified. There is usually strong and diffuse immunoreactivity for pancytokeratin, p40 (Fig. B), and p63, while neuroendocrine markers are negative. SOX10, CD117, and carcinoembryonic antigen (CEA) are notably positive in a subset of basaloid SCC , features not seen in SCNEC. Adenoid Cystic Carcinoma Solid Pattern While histological features of conventional adenoid cystic carcinoma (ACC) (with its typical cribriform architecture, dual-layered tubules lined by epithelial–myoepithelial cells, and luminal basophilic matrix) are quite characteristic and easy to diagnose, ACC with a solid pattern can resemble SCNEC, especially in a limited or small biopsy. Solid ACC is composed of diffuse sheets of basaloid cells that are largely devoid of the hallmark cribriform glands or tubules (Fig. C). These tumors commonly have increased mitoses and tumor necrosis. However, distinction from SCNEC can be readily achieved with the use of selected IHC. ACC shows positivity for epithelial (CK7, CEA, EMA) and myoepithelial markers (S100 protein, SOX10, SMA, calponin) (Fig. D), while is negative for neuroendocrine markers. CD117 positivity may be seen in both ACC and SCNEC , hence lacks specificity. The majority (60–90%) of ACC reveal a diagnostic fusion involving MYB/MYBL1 with NFIB genes, with MYB::NFIB the most common . Molecular testing is not required routinely, but may be performed to establish an ACC diagnosis in challenging cases. SWI/SNF Complex-Deficient Carcinomas (SMARCB1 & SMARCA4) The differential diagnoses of NEC have expanded to include SWI/SNF complex-deficient sinonasal carcinomas, whether SMARCB1 or SMARCA4 . These tumors predominantly affect adult males , typically arise in the paranasal sinuses (particularly the ethmoids) , and frequently present at an advanced stage . Both are high-grade malignancies histologically characterized by a monotonous population of undifferentiated cells. Similar to NEC, tumors are cellular and composed of islands and sheets of uniformly high-grade cells with brisk mitotic activity and foci of tumor necrosis. SMARCB1- deficient sinonasal carcinomas predominantly exhibit a basaloid (~ 2/3) or rhabdoid (~ 1/3) morphology; the latter may be very focal (Fig. A, B). Additionally, sharp, punched-out vacuoles within tumor sheets, yolk sac-like morphology, and pagetoid spread along the surface epithelium may be seen in SMARCB1 -deficient carcinomas that may aid in diagnosis when present . SMARCA4 -deficient carcinoma is composed of large, epithelioid cells lacking overt differentiation (Fig. D, E) . Rhabdoid and basaloid cells are less frequent. The cytologic appearance mimics LCNEC, requiring exclusion of the SWI/SNF complex-deficient carcinomas. Both entities require IHC to confirm the diagnosis. A complete loss of SMARCB1 (testing INI1]) and SMARCA4 (testing BRG1) reactivity in the tumor nuclei is essential for the diagnosis of SMARCB1 -deficient (Fig. C) and SMARCA4 -deficient sinonasal carcinoma (Fig. F), respectively . Additionally, the tumor cells are positive for pancytokeratin (AE1/AE3, CAM5.2, OSCAR) and variably positive with CK7. Further, there is frequently reactivity with CK5/6, p63, and p40 in SMARCB1 -deficient carcinoma, while these markers are generally negative in SMARCA4 -deficient carcinomas. Tumor cells are negative with NUT and there is no HPV or Epstein Barr virus association . It is noteworthy that both tumor types can focally express neuroendocrine markers (synaptophysin, chromogranin, and INSM1) in most SMARCA4 -deficient carcinomas and up to 18% of SMARCB1 -deficient carcinomas . Thus, INI1 and/or BRG1 must be included in a panel of immunohistochemistry studies when evaluating poorly or undifferentiated carcinomas of the sinonasal tract. Immunohistochemistry is generally sufficient for diagnosis, although FISH or sequencing can be performed to demonstrate biallelic (homozygous) deletions of the SMARCB1 gene or loss-of-function (mostly truncating) mutations in SMARCA4 -deficient carcinomas . Tumors with Neuroectodermal Differentiation Olfactory neuroblastoma (ONB) is a neuroectodermal neoplasm typically arising in the olfactory epithelium centered on the cribriform plate of the ethmoid sinus, composed of lobules of small round cells surrounded by sustentacular cells in a loose fibrovascular stroma. The morphological spectrum of ONB spans from the well-differentiated end (wherein the neoplastic cells display lobular architecture, uniform cells with stippled chromatin, rosettes and/or neurofibrillary stroma, low mitoses, and absence of tumor necrosis) (Fig. A) to the poorly differentiated end (which is characterized by limited lobular architecture, pleomorphism, increased mitoses, karyorrhexis, and tumor necrosis) (Fig. B). These features of diminishing differentiation are assembled into the Hyams tumor grades . Neurons, melanin pigment, or divergent differentiation (glandular, squamous, or rhabdomyoblastic) may be seen . A distinction of high-grade ONB from NEC is challenging and requires additional testing. ONB expresses diffuse neuroendocrine markers (synaptophysin, chromogranin, INSM1), neurofilament, and calretinin (Fig. C); about a third may show focal keratin reactivity . The peripheral rim of sustentacular cells is highlighted by S100 protein and/or SOX10 (Fig. D). Recently, SATB2 and focal GATA3 expression have been demonstrated in grade 1 to 3 ONBs . Tumor cells are negative for CD99, Fli1, NUT, and EBER, while INI1 and BRG1 are retained (intact). In contrast to ONB, NEC is negative for SOX10, S100, calretinin, SATB2, and GATA3 . Teratocarcinosarcoma (TCS) is a unique sinonasal tumor that is composed of a triad of epithelial, mesenchymal, and primitive neuroectodermal components; the three elements are intermixed and any constituent may predominate in a case. A biopsy with a preponderant primitive neuroectodermal component may be mistaken for NEC if the intimately admixed epithelial or mesenchymal components are either sparse or overlooked. Fetal-like (clear) squamous epithelium, immature mesenchyme, benign and/or carcinomatous epithelium, strap cells, or sarcomatous stroma are features that suggest a diagnosis of TCS (Fig. E–G). In contrast to a more uniform histological picture, the varied components of TCS render a very heterogeneous low-power appearance that may serve as an important clue to the diagnosis. Immunohistochemistry can highlight the presence of epithelial and sarcomatous (especially, rhabdomyosarcomatous elements positive for desmin, MyoD1, or Myogenin) apart from neuroendocrine marker expression in the primitive neuroectodermal component. SMARCA4 (BRG1) loss, complete or partial, is identified in up to 80% of the cases , while a subset may reveal nuclear ß-catenin immunoreactivity . Mucosal Melanoma Sometimes referred to as “the great mimicker,” melanoma presents with various morphological appearances and is thus a frequent tumor in many differential categories. Derived from melanocytes of the skin or mucosal linings, up to 25% of melanomas present in the head and neck region, with the scalp and cheek the two most common sites , while oral cavity and sinonasal tract may also be primary sites. Melanoma is usually identifiable using IHC targeting S100 protein, SOX10, MART-1 (Melan A), MITF1, and HMB45. From an embryonic perspective, melanocytes and most neuroendocrine cells both derive from the neural crest, and it is therefore not surprising to find expressional evidence of neuroendocrine differentiation in small subsets of melanoma . Indeed, in a retrospective study of > 300 melanomas, immunoreactivity for chromogranin A and synaptophysin was found in 2% and 8.6% of cases, respectively . Focal or faint expression of at least one of these markers was observed in 37.2% of the tumor cohort, thereby highlighting the need for a careful approach when assessing neuroendocrine markers in melanocytic lesions. Round Cell Sarcomas Ewing Sarcoma Ewing sarcoma (ES) is a primitive small round cell tumor that frequently needs to be distinguished from SCNEC. ES is defined by reciprocal translocations between the FET (encompassing EWSR1, FUS, and TAF15 genes) and the ETS (commonly including FLI1, ERG, ETV1, ETV4, or FEV ) family of genes . Like SCNEC, it is composed of cellular sheets of monotonous small round cells, 1–2 times the size of lymphocytes, with scant cytoplasm, round to oval nuclei, delicate stippled chromatin, and devoid of conspicuous nucleoli; occasional rosettes are identified (Fig. A, B) . Immunohistochemistry can aid in distinguishing ES from SCNEC, in which ES shows diffuse membranous positivity for CD99 (Fig. C) and concurrent nuclear reactivity for NKX2.2 (Fig. D) . Importantly, NKX2.2 can also be seen in SCNECs as can CD99, but the latter is strong and membranous in ES . Importantly, neuroendocrine marker positivity may be observed in ~ 50% of ES cases and ~ 30% of cases show cytokeratin expression . Fli1 and ERG reactivity are seen in cases with the respective fusions. A subtype of ES, adamantinoma-like Ewing sarcoma (ALES) tends to show a nested/lobular architecture, peripheral palisading, hyalinized stroma, and abrupt squamous differentiation; IHC evidence of squamous differentiation in the form of diffuse cytokeratin and p40/p63 reactivity is noted along with CD99 and NKX2.2 and most commonly the EWSR1::FLI1 fusion . Other Undifferentiated Round Cell Sarcomas Rarely, undifferentiated round cell sarcomas other than ES may be encountered that need to be distinguished from SCNEC. These include (1) round cell sarcomas with EWSR1- non- ETS fusions ; (2) CIC -rearranged sarcomas ; and (3) BCOR -rearranged sarcomas . Round Cell Sarcomas with EWSR1–non-ETS Fusions These are round and spindle cell sarcomas with EWSR1 or FUS fusions involving partners unrelated to the ETS gene family. These mainly comprise EWSR1::NFATC2 and FUS::NFATC2 sarcomas and EWSR1::PATZ1 sarcomas . Unlike conventional ES, these tumors exhibit atypical morphological features in the form of scattered enlarged cells, prominent nucleoli, or unusual clinical profiles (older patients). Nonetheless, there is considerable overlap with ES, including membranous CD99 staining. Although the pathologic spectrum is wide, some phenotypic clues to underlying genotypes can be helpful. Sarcomas with NFATC2 fusions tend to exhibit epithelioid features , while PATZ1 sarcomas are composed of largely undifferentiated round to ovoid neoplastic cells in a frequently sclerotic background . NFATC2 sarcomas express diffuse CD99 (like ES) in about 50% of cases; NKX2.2, dot-like keratin, and PAX7 positivity may also be observed . PATZ1 sarcomas do not consistently express CD99, however, may variably express CD34, and show a divergent phenotype with both myogenic (desmin, myogenin, MyoD1) and neurogenic (S100 protein, SOX10) markers , while neuroendocrine markers are usually absent. Identification of the fusion transcripts on molecular testing is the gold standard. CIC-rearranged sarcomas are round cell undifferentiated sarcomas that are defined by CIC -related gene fusions, mostly CIC::DUX4 fusion (about 95%) . CIC sarcomas are composed of undifferentiated round cells, however, tend to show lobulated growth (at least focally), and delicate fibrous septae; cells display mild pleomorphism and possess vesicular chromatin and prominent nucleoli (Fig. G). At times, epithelioid morphology can predominate . By IHC, WT1 (90–95%) (Fig. H) and ETV4 (95–100%) are positive and are extremely useful markers . CD99 is positive albeit patchy and cytoplasmic , rather than membranous. However, NKX2.2 is typically negative . Sarcomas with CIC::NUTM1 fusions are positive for NUT protein . Molecular testing reveals CIC- related fusions. BCOR-related sarcoma is a primitive round cell sarcoma showing BCOR genetic alterations. These tumors typically affect children with > 90% of patients being < 20 years . Histology typically reveals vague nesting, round cells often admixed with focal spindled cells, pale nuclear chromatin, inconspicuous nucleoli, and abundant myxoid stroma with delicate vascularity (Fig. E) . By IHC, tumor cells show diffuse, strong BCOR (Fig. F), SATB2, and cyclin D1 positivity. CD99 is seen in about 50% of cases , but neuroendocrine markers are usually absent. Alveolar Rhabdomyosarcoma (Solid Subtype) Rhabdomyosarcoma (RMS) is a malignant mesenchymal tumor composed of primitive cells exhibiting skeletal muscle differentiation. Head and neck RMS account for about 35–40% of all RMS cases . It encompasses embryonal, alveolar, pleomorphic, and spindle/sclerosing subtypes. Among the subtypes, alveolar rhabdomyosarcoma (ARMS), particularly the solid subtype, most closely mimics SCNEC . In comparison to SCNEC, the patients of ARMS are much younger, with the peak age of ARMS being 10–25-year-old young adults , although cases in adults > 45 years in the sinonasal tract especially are not uncommon . Microscopically, ARMS is characterized by cellular nests of small round cells separated by fibrovascular septae. Toward the center, the tumor cells tend to be dyscohesive conferring an alveolar configuration to the tumor; the latter is a vital diagnostic clue in favor of ARMS. In contrast, the solid subtype of ARMS is composed of diffuse sheets and lacks this nested/alveolar pattern and fibrovascular septae making it morphologically indistinguishable from NEC and small blue round cell tumors (Fig. A) . By IHC, cytoplasmic desmin (Fig. B), diffuse nuclear myogenin (Fig. C), and focal nuclear MyoD1 positivity are diagnostic of ARMS. Notably, neuroendocrine markers and keratins can be expressed in some cases of RMS . Specific neuroendocrine markers (chromogranin A and/or synaptophysin) can be seen in up to 43% of cases (Fig. D) . About 32% of cases can express both cytokeratins and NE markers . This aberrant keratin and neuroendocrine marker expression in RMS can lead to an erroneous diagnosis of NEC if skeletal muscle markers are not employed. Hence, a panel of markers is essential to avoid diagnostic pitfalls. Molecular testing for ARMS diagnosis and prognostication is recommended although not necessary for distinguishing ARMS from NEC. The majority (~ 70–90%) of ARMSs contain PAX3::FOXO1 fusions with the remaining tumors generally PAX7::FOXO1 . Synovial Sarcoma Poorly Differentiated Synovial sarcoma (SS) is a soft tissue sarcoma showing variable epithelial differentiation and is characterized by SS18::SSX1, SSX2 , or SSX4 fusions . Although SS can occur at any age, the majority of patients are adolescents or young adults and < 2% of patients are older than 50 years at diagnosis . Histologically, SS are cellular monophasic or biphasic tumors composed of dense sheets or vague fascicles of uniform appearing small spindle cells with ovoid, hyperchromatic nuclei with regular granular chromatin and inconspicuous nucleoli, and scant cytoplasm (Fig. E, F). A variable proportion of epithelial cells are seen intermixed with spindle components in the biphasic SS , yielding a marbled appearance on low power. The high cellularity and monomorphic appearance frequently place SS in the list of small round cell tumors, especially in limited biopsy material. The poorly differentiated subtype of SS particularly needs distinction from SCNEC. Poorly differentiated SS exhibits areas of increased cellularity, greater nuclear pleomorphism, and a high mitotic rate (> 10 mitoses per 2 mm 2 ) (Fig. F) . The cells may be spindle to round. The tumors with predominantly round cell morphology especially necessitate segregation from SCNEC . Poorly differentiated tumors also tend to be more common in elderly patients . By IHC, SS shows strong diffuse nuclear positivity for TLE1 in nearly all the cases (Fig. G), with variable positivity for CD99 and bcl2, with focal positivity for cytokeratin. Rare reports of neuroendocrine markers (synaptophysin, chromogranin A, and nestin) in FISH-confirmed SS have been reported . Recently, newer antibodies, SS18::SSX fusion-specific antibody (E9X9V, reactive against the breakpoint) and the SSX -specific antibody (E5A2C, reactive against the SSX C-terminus) have shown strong diffuse nuclear staining with excellent sensitivity and specificity (> 95%) for SS . Hematolymphoid Malignancies Lymphoma is a universal differential diagnosis for all small blue round cell tumors. High-grade diffuse large cell lymphomas, B-cell or T-cell lineage, show diffuse sheets of neoplastic cells with high nuclear-to-cytoplasmic ratio, loose chromatin, conspicuous nucleoli, and scanty cytoplasm (Fig. H). Brisk mitoses and apoptotic bodies are frequent. Crushing artifacts are common. Convoluted nuclei, nuclear folds, and grooves are commonly seen in T-cell lineage tumors. Tumor cells infiltrating through fibrotic stroma may simulate clustering similar to carcinomas. Due to overlapping features, distinction from NEC is usually required, especially in a limited biopsy. Immunohistochemistry can readily help in segregating lymphomas from NEC. Lymphomas are positive for hematolymphoid markers including CD45RB, while negative with pancytokeratins and neuroendocrine markers (Fig. ).
Well-differentiated NETs (WDNETs) encompass well-differentiated neoplasms with neuroendocrine features recognizable on light microscopy, i.e., exhibiting cellular monotony, random anisonucleosis, stippled nuclear chromatin, and varied growth patterns (nests, trabeculae, cords, festoons, rosettes). These tumors are strongly immunoreactive with neuroendocrine markers. WDNETs and NECs are clinically and genetically distinct entities with divergent treatments and outcomes. Therefore, a clear distinction between WDNET and NEC is necessary. This is usually not problematic, however, tends to be challenging in cases with limited or crushed tissue. Pituitary Neuroendocrine Tumor (PitNET) Pituitary neuroendocrine tumors (PitNETs) are well-differentiated adenohypophyseal lesions that may cause a wide variety of symptoms depending on its specific hormone production. PitNETs may be encountered in the sinonasal tract, most frequently in the sphenoid sinus, due to invasion from a sellar tumor or rarely as ectopic PitNET. The previous terminology “pituitary adenoma” and “pituitary carcinoma” are no longer recommended . PitNETs should be subtyped in terms of tumor cell lineage and expression of pituitary hormones by the use of IHC, and tumors are derived from a PIT1 lineage (i.e., somatotroph, lactotroph, and thyrotroph), a TPIT lineage (corticotroph tumors), an SF1 lineage (gonadotroph tumors), or tumors without a distinct lineage (plurihormonal lesions and hormonally silent “null cell tumors”) . Most PitNETs are characterized by a hypercellular and well-differentiated mass composed of monomorphic tumor cells (Fig. A). Although there are morphological clues regarding the PitNET subtype, the distinction is not entirely reliable without IHC and therefore all PitNETs should be assessed for PIT1, TPIT, and SF1 immunoreactivity . Hormone stains are also recommended, including growth hormone (GH), prolactin (PRL), and beta-thyroid-stimulating hormone (β-TSH) for PIT1 lineage tumors; adrenocorticotropic hormone (ACTH) for TPIT lineage tumors; and beta-follicle-stimulating hormone (β-FSH) and beta-luteinizing hormone (β-LH) for SF1 lineage tumors . Keratin IHC can also be useful (CAM5.2, AE1/AE3, and/or CK18) to subtype somatotroph PitNETs as either densely or sparsely granulated, characterized by a perinuclear or globular cytoplasmic stain, respectively . Metastatic PitNETs are rarely encountered, and when metastatic, most commonly affect liver, bone, lung, and lymph nodes . Medullary Thyroid Carcinoma (MTC) Medullary thyroid carcinoma (MTC) is a neuroendocrine tumor derived from the calcitonin-producing C cells of the thyroid gland. Although the majority of tumors arise sporadically, up to 25% are thought to be associated with multiple endocrine neoplasia type 2 (MEN2A or MEN2B) syndrome, with the affected patient demonstrating activating, constitutional pathogenic variants in the RET proto-oncogene . Sporadic tumors are usually driven by somatic RET mutations or mutually exclusive RAS gene mutations . There are numerous morphological MTC patterns reported, but they are not routinely classified on a histological basis as there is no established correlation to either genotype or clinical outcomes . MTCs are characterized by round (sometimes plasmacytoid, polygonal, or spindle shaped) cells in nests with an interdigitating stroma exhibiting various amounts of amyloid deposition (Fig. B). The cytoplasm is usually amphophilic and granular due to their secretory content. MTCs routinely express neuroendocrine markers as well as signs of thyroid differentiation (TTF1), whereas monoclonal PAX8 expression is lacking, as is thyroglobulin (the latter a consequence of the non-follicular cell origin). The hallmark of MTCs is calcitonin immunoreactivity, although the stain can vary in intensity and spatial distribution. In the metastatic setting without an established thyroid lesion, care must be taken not to prematurely assume that a neuroendocrine tumor with focal calcitonin and TTF1 expression is a metastatic MTC—as cases of laryngeal NETs with aberrant expression of these markers have been reported . In terms of prognosis, the 2022 World Health Organization (WHO) classification of endocrine and neuroendocrine tumors recommends that MTCs be graded based on the mitotic index, the presence of tumor necrosis, and the Ki-67 proliferation index, in which high-grade lesions display necrosis, a mitotic count ≥ 5 per 2 mm 2 , and/or a Ki-67 index of ≥ 5% . In terms of metastatic disease, most MTCs spread regionally to neck lymph nodes, but subsets of cases may also spread to the liver, lungs, and bone. It may be worth noting that rare cases of MTC metastatic to the parotid and pituitary glands have been reported, which potentially could constitute differential diagnostic conundrums . Moreover, subsets of MTCs may display a small cell phenotype, further complicating the histological work-up if NEC is suspected . Paraganglioma (PGL) Head and neck paragangliomas (PGLs) are usually parasympathetic, non-functioning neuroendocrine tumors, which sets them apart from their infra-diaphragmatic, norepinephrine, and/or epinephrine-producing counterparts . They are collectively the most inheritable of all human neoplasia with approximately 40% of patients carrying an underlying constitutional genetic event, while the metastatic potential of these lesions is usually low . Patients with constitutional pathogenic variants in succinate dehydrogenase (SDH) subunit A , B , C , D , or AF2 ( SDHA, SDHB, SDHC, SDHD, SDHAF2 ) harbor an increased risk of developing head and neck paraganglioma with a low risk of disseminated disease . The underlying molecular biology is complex, with tumors showing a higher risk of metastases often driven by mutations in tricarboxylic acid (TCA) cycle genes (not only restricted to SDH genes) that will lead to TCA cycle arrest and accumulation of early metabolites, which in turn may activate oncogenic hypoxia-inducible factor (HIF) pathways . However, subsets of cases are driven by mutations in various kinase-associated pathways, and these lesions usually tend to be non-metastatic. Therefore, there is a well-developed genotype–phenotype correlation which can be assessed by histopathology and immunohistochemistry: a positive SDHB immunostain strongly argues against mutational inactivation of either SDHB , SDHC , or SDHD genes—in turn arguing against (although not excluding) the risk of metastatic potential . When presenting in characteristic locations, such as the carotid bifurcation or the jugulotympanic area, a head and neck paraganglioma is quite easily distinguished by morphological assessment. The tumor cells are usually round to oval and arranged in small nests (so-called “zellballen”) embedded in a highly vascular stroma (Fig. C). The cytoplasm is granular with an amphophilic or basophilic appearance, and nuclear pleomorphism is usually limited to absent. Mitoses and tumor necrosis are rarely detected. Tumor cells are positive for chromogranin A, synaptophysin, and GATA3, while consistently keratin negative . An S100 protein (or SOX10) immunohistochemistry identifies the sustentacular network of cells supporting the tumor cells—but the finding of sustentacular cells is not diagnostic for paraganglioma, as other neuroendocrine tumors also may exhibit this feature . Using functional IHC, most head and neck paragangliomas are positive for choline acetyltransferase, an enzyme in the acetylcholine biosynthesis pathway, while often negative for enzymes responsible for catecholamine production, such as tyrosine hydroxylase . Metastatic head and neck paraganglioma usually spread to regional lymph nodes, while distant site involvement is rare . Metastastic Neuroendocrine Neoplasia of Unknown Primary NENs of unknown primary (NEN-UPs) are metastatic lesions without a known primary tumor location, a finding reported in 12–22% of NEN patients . The importance of identifying the primary site cannot be underestimated given that the various clinical and prognostic features of NENs depend on the tumor origin site. There are several morphological clues that can be used to properly identity a NEN-UP, including amyloid deposits in MTC, psammoma bodies in somatostatinoma, a hyalinized stroma in insulinoma, as well as the hyaline globules and basophilic cytoplasm of pheochromocytoma . Even so, it is not unusual for a metastatic NEN-UP to be characterized by nested cells with little or no morphological findings unique to the primary tumor site (Fig. D). From an immunohistochemical perspective, various combinations of neuroendocrine marker and transcription factors results may be useful. For example, TTF1 may help identify pulmonary carcinoids and MTC, PDX1 may assist in recognizing NENs of the upper gastrointestinal tract including the pancreas, whereas CDX2 and SATB2 may highlight NENs of the lower gastrointestinal tract. In addition, testing for various hormones may be useful, including calcitonin for MTC, serotonin for lower gastrointestinal NENs, islet hormones for pancreatic NENs, and GLP1 for rectal NENs, to name just a few . NECs outside of the head and neck area occasionally may metastasize to the jaws and major salivary glands . Indeed, metastatic neuroendocrine tumors to the parotid gland accounted for 22% of all metastatic tumors to this organ in a recent case series, and most cases were either pulmonary NECs, Merkel cell carcinomas (MCCs), or MTCs . If not previously known, a hypothetical IHC panel for NEN-UPs metastatic to the salivary glands would therefore need at least TTF1, calcitonin, and CK20, in addition to neuroendocrine markers and Ki-67.
Pituitary neuroendocrine tumors (PitNETs) are well-differentiated adenohypophyseal lesions that may cause a wide variety of symptoms depending on its specific hormone production. PitNETs may be encountered in the sinonasal tract, most frequently in the sphenoid sinus, due to invasion from a sellar tumor or rarely as ectopic PitNET. The previous terminology “pituitary adenoma” and “pituitary carcinoma” are no longer recommended . PitNETs should be subtyped in terms of tumor cell lineage and expression of pituitary hormones by the use of IHC, and tumors are derived from a PIT1 lineage (i.e., somatotroph, lactotroph, and thyrotroph), a TPIT lineage (corticotroph tumors), an SF1 lineage (gonadotroph tumors), or tumors without a distinct lineage (plurihormonal lesions and hormonally silent “null cell tumors”) . Most PitNETs are characterized by a hypercellular and well-differentiated mass composed of monomorphic tumor cells (Fig. A). Although there are morphological clues regarding the PitNET subtype, the distinction is not entirely reliable without IHC and therefore all PitNETs should be assessed for PIT1, TPIT, and SF1 immunoreactivity . Hormone stains are also recommended, including growth hormone (GH), prolactin (PRL), and beta-thyroid-stimulating hormone (β-TSH) for PIT1 lineage tumors; adrenocorticotropic hormone (ACTH) for TPIT lineage tumors; and beta-follicle-stimulating hormone (β-FSH) and beta-luteinizing hormone (β-LH) for SF1 lineage tumors . Keratin IHC can also be useful (CAM5.2, AE1/AE3, and/or CK18) to subtype somatotroph PitNETs as either densely or sparsely granulated, characterized by a perinuclear or globular cytoplasmic stain, respectively . Metastatic PitNETs are rarely encountered, and when metastatic, most commonly affect liver, bone, lung, and lymph nodes .
Medullary thyroid carcinoma (MTC) is a neuroendocrine tumor derived from the calcitonin-producing C cells of the thyroid gland. Although the majority of tumors arise sporadically, up to 25% are thought to be associated with multiple endocrine neoplasia type 2 (MEN2A or MEN2B) syndrome, with the affected patient demonstrating activating, constitutional pathogenic variants in the RET proto-oncogene . Sporadic tumors are usually driven by somatic RET mutations or mutually exclusive RAS gene mutations . There are numerous morphological MTC patterns reported, but they are not routinely classified on a histological basis as there is no established correlation to either genotype or clinical outcomes . MTCs are characterized by round (sometimes plasmacytoid, polygonal, or spindle shaped) cells in nests with an interdigitating stroma exhibiting various amounts of amyloid deposition (Fig. B). The cytoplasm is usually amphophilic and granular due to their secretory content. MTCs routinely express neuroendocrine markers as well as signs of thyroid differentiation (TTF1), whereas monoclonal PAX8 expression is lacking, as is thyroglobulin (the latter a consequence of the non-follicular cell origin). The hallmark of MTCs is calcitonin immunoreactivity, although the stain can vary in intensity and spatial distribution. In the metastatic setting without an established thyroid lesion, care must be taken not to prematurely assume that a neuroendocrine tumor with focal calcitonin and TTF1 expression is a metastatic MTC—as cases of laryngeal NETs with aberrant expression of these markers have been reported . In terms of prognosis, the 2022 World Health Organization (WHO) classification of endocrine and neuroendocrine tumors recommends that MTCs be graded based on the mitotic index, the presence of tumor necrosis, and the Ki-67 proliferation index, in which high-grade lesions display necrosis, a mitotic count ≥ 5 per 2 mm 2 , and/or a Ki-67 index of ≥ 5% . In terms of metastatic disease, most MTCs spread regionally to neck lymph nodes, but subsets of cases may also spread to the liver, lungs, and bone. It may be worth noting that rare cases of MTC metastatic to the parotid and pituitary glands have been reported, which potentially could constitute differential diagnostic conundrums . Moreover, subsets of MTCs may display a small cell phenotype, further complicating the histological work-up if NEC is suspected .
Head and neck paragangliomas (PGLs) are usually parasympathetic, non-functioning neuroendocrine tumors, which sets them apart from their infra-diaphragmatic, norepinephrine, and/or epinephrine-producing counterparts . They are collectively the most inheritable of all human neoplasia with approximately 40% of patients carrying an underlying constitutional genetic event, while the metastatic potential of these lesions is usually low . Patients with constitutional pathogenic variants in succinate dehydrogenase (SDH) subunit A , B , C , D , or AF2 ( SDHA, SDHB, SDHC, SDHD, SDHAF2 ) harbor an increased risk of developing head and neck paraganglioma with a low risk of disseminated disease . The underlying molecular biology is complex, with tumors showing a higher risk of metastases often driven by mutations in tricarboxylic acid (TCA) cycle genes (not only restricted to SDH genes) that will lead to TCA cycle arrest and accumulation of early metabolites, which in turn may activate oncogenic hypoxia-inducible factor (HIF) pathways . However, subsets of cases are driven by mutations in various kinase-associated pathways, and these lesions usually tend to be non-metastatic. Therefore, there is a well-developed genotype–phenotype correlation which can be assessed by histopathology and immunohistochemistry: a positive SDHB immunostain strongly argues against mutational inactivation of either SDHB , SDHC , or SDHD genes—in turn arguing against (although not excluding) the risk of metastatic potential . When presenting in characteristic locations, such as the carotid bifurcation or the jugulotympanic area, a head and neck paraganglioma is quite easily distinguished by morphological assessment. The tumor cells are usually round to oval and arranged in small nests (so-called “zellballen”) embedded in a highly vascular stroma (Fig. C). The cytoplasm is granular with an amphophilic or basophilic appearance, and nuclear pleomorphism is usually limited to absent. Mitoses and tumor necrosis are rarely detected. Tumor cells are positive for chromogranin A, synaptophysin, and GATA3, while consistently keratin negative . An S100 protein (or SOX10) immunohistochemistry identifies the sustentacular network of cells supporting the tumor cells—but the finding of sustentacular cells is not diagnostic for paraganglioma, as other neuroendocrine tumors also may exhibit this feature . Using functional IHC, most head and neck paragangliomas are positive for choline acetyltransferase, an enzyme in the acetylcholine biosynthesis pathway, while often negative for enzymes responsible for catecholamine production, such as tyrosine hydroxylase . Metastatic head and neck paraganglioma usually spread to regional lymph nodes, while distant site involvement is rare .
NENs of unknown primary (NEN-UPs) are metastatic lesions without a known primary tumor location, a finding reported in 12–22% of NEN patients . The importance of identifying the primary site cannot be underestimated given that the various clinical and prognostic features of NENs depend on the tumor origin site. There are several morphological clues that can be used to properly identity a NEN-UP, including amyloid deposits in MTC, psammoma bodies in somatostatinoma, a hyalinized stroma in insulinoma, as well as the hyaline globules and basophilic cytoplasm of pheochromocytoma . Even so, it is not unusual for a metastatic NEN-UP to be characterized by nested cells with little or no morphological findings unique to the primary tumor site (Fig. D). From an immunohistochemical perspective, various combinations of neuroendocrine marker and transcription factors results may be useful. For example, TTF1 may help identify pulmonary carcinoids and MTC, PDX1 may assist in recognizing NENs of the upper gastrointestinal tract including the pancreas, whereas CDX2 and SATB2 may highlight NENs of the lower gastrointestinal tract. In addition, testing for various hormones may be useful, including calcitonin for MTC, serotonin for lower gastrointestinal NENs, islet hormones for pancreatic NENs, and GLP1 for rectal NENs, to name just a few . NECs outside of the head and neck area occasionally may metastasize to the jaws and major salivary glands . Indeed, metastatic neuroendocrine tumors to the parotid gland accounted for 22% of all metastatic tumors to this organ in a recent case series, and most cases were either pulmonary NECs, Merkel cell carcinomas (MCCs), or MTCs . If not previously known, a hypothetical IHC panel for NEN-UPs metastatic to the salivary glands would therefore need at least TTF1, calcitonin, and CK20, in addition to neuroendocrine markers and Ki-67.
Merkel cell carcinoma (MCC) is a rare neuroendocrine carcinoma of the skin with an estimated incidence of 2.2 cases per million person-years, afflicting predominantly older patients . The tumors are either driven by UV-induced mutations or by a Merkel cell polyoma virus infection, and the exact proportion of these etiologies varies with geographic distribution . From a morphological perspective, MCCs are composed of solid arrangements of monomorphic tumor cells with a high nuclear/cytoplasmic ratio, smudged nuclear chromatin, indistinct nucleoli, and displaying innumerable mitoses (Fig. A, B). The IHC profile is characteristically neuroendocrine , while a perinuclear, “dot-like” keratin stain (most strikingly with CK20) is characteristic of MCC (Fig. C). The Ki-67 proliferation index is usually exceedingly high (> 90%). Moreover, virus-driven MCCs are positive for the Merkel cell polyoma antigen (Fig. D). When presenting as a primary tumor, the diagnosis is usually quite straight-forward, but metastatic lesions may cause diagnostic difficulties if the primary tumor is not known. To complicate matters even more, subsets of MCCs have been reported to originate from mucosal linings of the upper respiratory and GI tracts and might be clinically silent .
NUT carcinoma is a highly aggressive tumor primarily affecting young patients, often presenting in the midline of the thorax and head and neck regions . On the histological level, NUT carcinoma is composed of small to intermediate cells with an undifferentiated, primitive, and monotonous appearance (Fig. A). Mitotic figures and necrosis are easily identified. A significant subset exhibits abrupt keratinization (Fig. B). Using IHC, NUT carcinomas are epithelial neoplasms, reacting with keratins and squamous markers, such as CK5/6, p63, and p40 (Fig. C). CD34 is also positive in approximately 50% of cases . NUT carcinoma is driven by NUTM1 gene rearrangements , and NUT protein immunohistochemistry (Fig. D) is useful to highlight this genetic aberrancy—as NUT protein expression is not normally seen in cells outside of the testis and ciliary ganglion . Interestingly, neuroendocrine differentiation has been reported, with a high level of suspicion required when considering NEC in young patients by incorporating NUT immunohistochemistry .
Sinonasal undifferentiated carcinoma (SNUC) is a rare but highly aggressive epithelial neoplasia lacking morphological and immunohistochemical evidence of lineage (including glandular, squamous, neuroendocrine, or mesenchymal differentiation) . Thus, it is a diagnosis of exclusion in which a broad range of possible differential diagnoses must be considered. Usually presenting as a large mass in the sinonasal tract, these tumors are often invasive at diagnosis . SNUC exhibits high-grade histology with uniform tumor cells growing in sheets, lobules, nests or trabeculae, lacking squamous, or glandular differentiation. Tumor cells express keratins, while p40 is negative and p63 may exhibit weak and unspecific staining (Fig. ). Subsets of SNUC may express patchy chromogranin A and/or synaptophysin immunoreactivity, making them a potential differential diagnosis in the work-up of NEC of the head and neck region . Somatic IDH2 mutations have been identified in large subsets of SNUC and are readily identifiable using sequencing analysis, while IDH immunohistochemistry has proven inconsistent in pinpointing IDH2-mutated SNUC .
Squamous cell carcinoma (SCC) is the most common histological type of cancer in the head and neck region. While the diagnosis of a differentiated SCC is not problematic, the basaloid SCC subtype is a common differential diagnosis of SCNEC, particularly in biopsy material. Basaloid SCC remains an uncommon malignancy that is associated with aggressive clinical behavior and poor median survival (18 months) . Closely packed basaloid cells and lack of significant keratinization typically impart a blue cell tumor appearance at low power that resembles SCNEC. Lobules, adenoid/pseudoglandular structures or variably anastomosing islands of tumor cells exhibiting peripheral palisading, thickened basement membrane-like material, and central comedonecrosis are typical histological features (Fig. A) . The neoplastic cells show pleomorphic hyperchromatic nuclei with scanty cytoplasm. The presence of carcinoma in situ or areas of abrupt squamous differentiation (keratin pearl formation) are useful clues to the diagnosis. Mitoses are usually easily identified. There is usually strong and diffuse immunoreactivity for pancytokeratin, p40 (Fig. B), and p63, while neuroendocrine markers are negative. SOX10, CD117, and carcinoembryonic antigen (CEA) are notably positive in a subset of basaloid SCC , features not seen in SCNEC.
While histological features of conventional adenoid cystic carcinoma (ACC) (with its typical cribriform architecture, dual-layered tubules lined by epithelial–myoepithelial cells, and luminal basophilic matrix) are quite characteristic and easy to diagnose, ACC with a solid pattern can resemble SCNEC, especially in a limited or small biopsy. Solid ACC is composed of diffuse sheets of basaloid cells that are largely devoid of the hallmark cribriform glands or tubules (Fig. C). These tumors commonly have increased mitoses and tumor necrosis. However, distinction from SCNEC can be readily achieved with the use of selected IHC. ACC shows positivity for epithelial (CK7, CEA, EMA) and myoepithelial markers (S100 protein, SOX10, SMA, calponin) (Fig. D), while is negative for neuroendocrine markers. CD117 positivity may be seen in both ACC and SCNEC , hence lacks specificity. The majority (60–90%) of ACC reveal a diagnostic fusion involving MYB/MYBL1 with NFIB genes, with MYB::NFIB the most common . Molecular testing is not required routinely, but may be performed to establish an ACC diagnosis in challenging cases.
The differential diagnoses of NEC have expanded to include SWI/SNF complex-deficient sinonasal carcinomas, whether SMARCB1 or SMARCA4 . These tumors predominantly affect adult males , typically arise in the paranasal sinuses (particularly the ethmoids) , and frequently present at an advanced stage . Both are high-grade malignancies histologically characterized by a monotonous population of undifferentiated cells. Similar to NEC, tumors are cellular and composed of islands and sheets of uniformly high-grade cells with brisk mitotic activity and foci of tumor necrosis. SMARCB1- deficient sinonasal carcinomas predominantly exhibit a basaloid (~ 2/3) or rhabdoid (~ 1/3) morphology; the latter may be very focal (Fig. A, B). Additionally, sharp, punched-out vacuoles within tumor sheets, yolk sac-like morphology, and pagetoid spread along the surface epithelium may be seen in SMARCB1 -deficient carcinomas that may aid in diagnosis when present . SMARCA4 -deficient carcinoma is composed of large, epithelioid cells lacking overt differentiation (Fig. D, E) . Rhabdoid and basaloid cells are less frequent. The cytologic appearance mimics LCNEC, requiring exclusion of the SWI/SNF complex-deficient carcinomas. Both entities require IHC to confirm the diagnosis. A complete loss of SMARCB1 (testing INI1]) and SMARCA4 (testing BRG1) reactivity in the tumor nuclei is essential for the diagnosis of SMARCB1 -deficient (Fig. C) and SMARCA4 -deficient sinonasal carcinoma (Fig. F), respectively . Additionally, the tumor cells are positive for pancytokeratin (AE1/AE3, CAM5.2, OSCAR) and variably positive with CK7. Further, there is frequently reactivity with CK5/6, p63, and p40 in SMARCB1 -deficient carcinoma, while these markers are generally negative in SMARCA4 -deficient carcinomas. Tumor cells are negative with NUT and there is no HPV or Epstein Barr virus association . It is noteworthy that both tumor types can focally express neuroendocrine markers (synaptophysin, chromogranin, and INSM1) in most SMARCA4 -deficient carcinomas and up to 18% of SMARCB1 -deficient carcinomas . Thus, INI1 and/or BRG1 must be included in a panel of immunohistochemistry studies when evaluating poorly or undifferentiated carcinomas of the sinonasal tract. Immunohistochemistry is generally sufficient for diagnosis, although FISH or sequencing can be performed to demonstrate biallelic (homozygous) deletions of the SMARCB1 gene or loss-of-function (mostly truncating) mutations in SMARCA4 -deficient carcinomas .
Olfactory neuroblastoma (ONB) is a neuroectodermal neoplasm typically arising in the olfactory epithelium centered on the cribriform plate of the ethmoid sinus, composed of lobules of small round cells surrounded by sustentacular cells in a loose fibrovascular stroma. The morphological spectrum of ONB spans from the well-differentiated end (wherein the neoplastic cells display lobular architecture, uniform cells with stippled chromatin, rosettes and/or neurofibrillary stroma, low mitoses, and absence of tumor necrosis) (Fig. A) to the poorly differentiated end (which is characterized by limited lobular architecture, pleomorphism, increased mitoses, karyorrhexis, and tumor necrosis) (Fig. B). These features of diminishing differentiation are assembled into the Hyams tumor grades . Neurons, melanin pigment, or divergent differentiation (glandular, squamous, or rhabdomyoblastic) may be seen . A distinction of high-grade ONB from NEC is challenging and requires additional testing. ONB expresses diffuse neuroendocrine markers (synaptophysin, chromogranin, INSM1), neurofilament, and calretinin (Fig. C); about a third may show focal keratin reactivity . The peripheral rim of sustentacular cells is highlighted by S100 protein and/or SOX10 (Fig. D). Recently, SATB2 and focal GATA3 expression have been demonstrated in grade 1 to 3 ONBs . Tumor cells are negative for CD99, Fli1, NUT, and EBER, while INI1 and BRG1 are retained (intact). In contrast to ONB, NEC is negative for SOX10, S100, calretinin, SATB2, and GATA3 . Teratocarcinosarcoma (TCS) is a unique sinonasal tumor that is composed of a triad of epithelial, mesenchymal, and primitive neuroectodermal components; the three elements are intermixed and any constituent may predominate in a case. A biopsy with a preponderant primitive neuroectodermal component may be mistaken for NEC if the intimately admixed epithelial or mesenchymal components are either sparse or overlooked. Fetal-like (clear) squamous epithelium, immature mesenchyme, benign and/or carcinomatous epithelium, strap cells, or sarcomatous stroma are features that suggest a diagnosis of TCS (Fig. E–G). In contrast to a more uniform histological picture, the varied components of TCS render a very heterogeneous low-power appearance that may serve as an important clue to the diagnosis. Immunohistochemistry can highlight the presence of epithelial and sarcomatous (especially, rhabdomyosarcomatous elements positive for desmin, MyoD1, or Myogenin) apart from neuroendocrine marker expression in the primitive neuroectodermal component. SMARCA4 (BRG1) loss, complete or partial, is identified in up to 80% of the cases , while a subset may reveal nuclear ß-catenin immunoreactivity .
Sometimes referred to as “the great mimicker,” melanoma presents with various morphological appearances and is thus a frequent tumor in many differential categories. Derived from melanocytes of the skin or mucosal linings, up to 25% of melanomas present in the head and neck region, with the scalp and cheek the two most common sites , while oral cavity and sinonasal tract may also be primary sites. Melanoma is usually identifiable using IHC targeting S100 protein, SOX10, MART-1 (Melan A), MITF1, and HMB45. From an embryonic perspective, melanocytes and most neuroendocrine cells both derive from the neural crest, and it is therefore not surprising to find expressional evidence of neuroendocrine differentiation in small subsets of melanoma . Indeed, in a retrospective study of > 300 melanomas, immunoreactivity for chromogranin A and synaptophysin was found in 2% and 8.6% of cases, respectively . Focal or faint expression of at least one of these markers was observed in 37.2% of the tumor cohort, thereby highlighting the need for a careful approach when assessing neuroendocrine markers in melanocytic lesions.
Ewing Sarcoma Ewing sarcoma (ES) is a primitive small round cell tumor that frequently needs to be distinguished from SCNEC. ES is defined by reciprocal translocations between the FET (encompassing EWSR1, FUS, and TAF15 genes) and the ETS (commonly including FLI1, ERG, ETV1, ETV4, or FEV ) family of genes . Like SCNEC, it is composed of cellular sheets of monotonous small round cells, 1–2 times the size of lymphocytes, with scant cytoplasm, round to oval nuclei, delicate stippled chromatin, and devoid of conspicuous nucleoli; occasional rosettes are identified (Fig. A, B) . Immunohistochemistry can aid in distinguishing ES from SCNEC, in which ES shows diffuse membranous positivity for CD99 (Fig. C) and concurrent nuclear reactivity for NKX2.2 (Fig. D) . Importantly, NKX2.2 can also be seen in SCNECs as can CD99, but the latter is strong and membranous in ES . Importantly, neuroendocrine marker positivity may be observed in ~ 50% of ES cases and ~ 30% of cases show cytokeratin expression . Fli1 and ERG reactivity are seen in cases with the respective fusions. A subtype of ES, adamantinoma-like Ewing sarcoma (ALES) tends to show a nested/lobular architecture, peripheral palisading, hyalinized stroma, and abrupt squamous differentiation; IHC evidence of squamous differentiation in the form of diffuse cytokeratin and p40/p63 reactivity is noted along with CD99 and NKX2.2 and most commonly the EWSR1::FLI1 fusion . Other Undifferentiated Round Cell Sarcomas Rarely, undifferentiated round cell sarcomas other than ES may be encountered that need to be distinguished from SCNEC. These include (1) round cell sarcomas with EWSR1- non- ETS fusions ; (2) CIC -rearranged sarcomas ; and (3) BCOR -rearranged sarcomas . Round Cell Sarcomas with EWSR1–non-ETS Fusions These are round and spindle cell sarcomas with EWSR1 or FUS fusions involving partners unrelated to the ETS gene family. These mainly comprise EWSR1::NFATC2 and FUS::NFATC2 sarcomas and EWSR1::PATZ1 sarcomas . Unlike conventional ES, these tumors exhibit atypical morphological features in the form of scattered enlarged cells, prominent nucleoli, or unusual clinical profiles (older patients). Nonetheless, there is considerable overlap with ES, including membranous CD99 staining. Although the pathologic spectrum is wide, some phenotypic clues to underlying genotypes can be helpful. Sarcomas with NFATC2 fusions tend to exhibit epithelioid features , while PATZ1 sarcomas are composed of largely undifferentiated round to ovoid neoplastic cells in a frequently sclerotic background . NFATC2 sarcomas express diffuse CD99 (like ES) in about 50% of cases; NKX2.2, dot-like keratin, and PAX7 positivity may also be observed . PATZ1 sarcomas do not consistently express CD99, however, may variably express CD34, and show a divergent phenotype with both myogenic (desmin, myogenin, MyoD1) and neurogenic (S100 protein, SOX10) markers , while neuroendocrine markers are usually absent. Identification of the fusion transcripts on molecular testing is the gold standard. CIC-rearranged sarcomas are round cell undifferentiated sarcomas that are defined by CIC -related gene fusions, mostly CIC::DUX4 fusion (about 95%) . CIC sarcomas are composed of undifferentiated round cells, however, tend to show lobulated growth (at least focally), and delicate fibrous septae; cells display mild pleomorphism and possess vesicular chromatin and prominent nucleoli (Fig. G). At times, epithelioid morphology can predominate . By IHC, WT1 (90–95%) (Fig. H) and ETV4 (95–100%) are positive and are extremely useful markers . CD99 is positive albeit patchy and cytoplasmic , rather than membranous. However, NKX2.2 is typically negative . Sarcomas with CIC::NUTM1 fusions are positive for NUT protein . Molecular testing reveals CIC- related fusions. BCOR-related sarcoma is a primitive round cell sarcoma showing BCOR genetic alterations. These tumors typically affect children with > 90% of patients being < 20 years . Histology typically reveals vague nesting, round cells often admixed with focal spindled cells, pale nuclear chromatin, inconspicuous nucleoli, and abundant myxoid stroma with delicate vascularity (Fig. E) . By IHC, tumor cells show diffuse, strong BCOR (Fig. F), SATB2, and cyclin D1 positivity. CD99 is seen in about 50% of cases , but neuroendocrine markers are usually absent. Alveolar Rhabdomyosarcoma (Solid Subtype) Rhabdomyosarcoma (RMS) is a malignant mesenchymal tumor composed of primitive cells exhibiting skeletal muscle differentiation. Head and neck RMS account for about 35–40% of all RMS cases . It encompasses embryonal, alveolar, pleomorphic, and spindle/sclerosing subtypes. Among the subtypes, alveolar rhabdomyosarcoma (ARMS), particularly the solid subtype, most closely mimics SCNEC . In comparison to SCNEC, the patients of ARMS are much younger, with the peak age of ARMS being 10–25-year-old young adults , although cases in adults > 45 years in the sinonasal tract especially are not uncommon . Microscopically, ARMS is characterized by cellular nests of small round cells separated by fibrovascular septae. Toward the center, the tumor cells tend to be dyscohesive conferring an alveolar configuration to the tumor; the latter is a vital diagnostic clue in favor of ARMS. In contrast, the solid subtype of ARMS is composed of diffuse sheets and lacks this nested/alveolar pattern and fibrovascular septae making it morphologically indistinguishable from NEC and small blue round cell tumors (Fig. A) . By IHC, cytoplasmic desmin (Fig. B), diffuse nuclear myogenin (Fig. C), and focal nuclear MyoD1 positivity are diagnostic of ARMS. Notably, neuroendocrine markers and keratins can be expressed in some cases of RMS . Specific neuroendocrine markers (chromogranin A and/or synaptophysin) can be seen in up to 43% of cases (Fig. D) . About 32% of cases can express both cytokeratins and NE markers . This aberrant keratin and neuroendocrine marker expression in RMS can lead to an erroneous diagnosis of NEC if skeletal muscle markers are not employed. Hence, a panel of markers is essential to avoid diagnostic pitfalls. Molecular testing for ARMS diagnosis and prognostication is recommended although not necessary for distinguishing ARMS from NEC. The majority (~ 70–90%) of ARMSs contain PAX3::FOXO1 fusions with the remaining tumors generally PAX7::FOXO1 . Synovial Sarcoma Poorly Differentiated Synovial sarcoma (SS) is a soft tissue sarcoma showing variable epithelial differentiation and is characterized by SS18::SSX1, SSX2 , or SSX4 fusions . Although SS can occur at any age, the majority of patients are adolescents or young adults and < 2% of patients are older than 50 years at diagnosis . Histologically, SS are cellular monophasic or biphasic tumors composed of dense sheets or vague fascicles of uniform appearing small spindle cells with ovoid, hyperchromatic nuclei with regular granular chromatin and inconspicuous nucleoli, and scant cytoplasm (Fig. E, F). A variable proportion of epithelial cells are seen intermixed with spindle components in the biphasic SS , yielding a marbled appearance on low power. The high cellularity and monomorphic appearance frequently place SS in the list of small round cell tumors, especially in limited biopsy material. The poorly differentiated subtype of SS particularly needs distinction from SCNEC. Poorly differentiated SS exhibits areas of increased cellularity, greater nuclear pleomorphism, and a high mitotic rate (> 10 mitoses per 2 mm 2 ) (Fig. F) . The cells may be spindle to round. The tumors with predominantly round cell morphology especially necessitate segregation from SCNEC . Poorly differentiated tumors also tend to be more common in elderly patients . By IHC, SS shows strong diffuse nuclear positivity for TLE1 in nearly all the cases (Fig. G), with variable positivity for CD99 and bcl2, with focal positivity for cytokeratin. Rare reports of neuroendocrine markers (synaptophysin, chromogranin A, and nestin) in FISH-confirmed SS have been reported . Recently, newer antibodies, SS18::SSX fusion-specific antibody (E9X9V, reactive against the breakpoint) and the SSX -specific antibody (E5A2C, reactive against the SSX C-terminus) have shown strong diffuse nuclear staining with excellent sensitivity and specificity (> 95%) for SS .
Ewing sarcoma (ES) is a primitive small round cell tumor that frequently needs to be distinguished from SCNEC. ES is defined by reciprocal translocations between the FET (encompassing EWSR1, FUS, and TAF15 genes) and the ETS (commonly including FLI1, ERG, ETV1, ETV4, or FEV ) family of genes . Like SCNEC, it is composed of cellular sheets of monotonous small round cells, 1–2 times the size of lymphocytes, with scant cytoplasm, round to oval nuclei, delicate stippled chromatin, and devoid of conspicuous nucleoli; occasional rosettes are identified (Fig. A, B) . Immunohistochemistry can aid in distinguishing ES from SCNEC, in which ES shows diffuse membranous positivity for CD99 (Fig. C) and concurrent nuclear reactivity for NKX2.2 (Fig. D) . Importantly, NKX2.2 can also be seen in SCNECs as can CD99, but the latter is strong and membranous in ES . Importantly, neuroendocrine marker positivity may be observed in ~ 50% of ES cases and ~ 30% of cases show cytokeratin expression . Fli1 and ERG reactivity are seen in cases with the respective fusions. A subtype of ES, adamantinoma-like Ewing sarcoma (ALES) tends to show a nested/lobular architecture, peripheral palisading, hyalinized stroma, and abrupt squamous differentiation; IHC evidence of squamous differentiation in the form of diffuse cytokeratin and p40/p63 reactivity is noted along with CD99 and NKX2.2 and most commonly the EWSR1::FLI1 fusion .
Rarely, undifferentiated round cell sarcomas other than ES may be encountered that need to be distinguished from SCNEC. These include (1) round cell sarcomas with EWSR1- non- ETS fusions ; (2) CIC -rearranged sarcomas ; and (3) BCOR -rearranged sarcomas . Round Cell Sarcomas with EWSR1–non-ETS Fusions These are round and spindle cell sarcomas with EWSR1 or FUS fusions involving partners unrelated to the ETS gene family. These mainly comprise EWSR1::NFATC2 and FUS::NFATC2 sarcomas and EWSR1::PATZ1 sarcomas . Unlike conventional ES, these tumors exhibit atypical morphological features in the form of scattered enlarged cells, prominent nucleoli, or unusual clinical profiles (older patients). Nonetheless, there is considerable overlap with ES, including membranous CD99 staining. Although the pathologic spectrum is wide, some phenotypic clues to underlying genotypes can be helpful. Sarcomas with NFATC2 fusions tend to exhibit epithelioid features , while PATZ1 sarcomas are composed of largely undifferentiated round to ovoid neoplastic cells in a frequently sclerotic background . NFATC2 sarcomas express diffuse CD99 (like ES) in about 50% of cases; NKX2.2, dot-like keratin, and PAX7 positivity may also be observed . PATZ1 sarcomas do not consistently express CD99, however, may variably express CD34, and show a divergent phenotype with both myogenic (desmin, myogenin, MyoD1) and neurogenic (S100 protein, SOX10) markers , while neuroendocrine markers are usually absent. Identification of the fusion transcripts on molecular testing is the gold standard. CIC-rearranged sarcomas are round cell undifferentiated sarcomas that are defined by CIC -related gene fusions, mostly CIC::DUX4 fusion (about 95%) . CIC sarcomas are composed of undifferentiated round cells, however, tend to show lobulated growth (at least focally), and delicate fibrous septae; cells display mild pleomorphism and possess vesicular chromatin and prominent nucleoli (Fig. G). At times, epithelioid morphology can predominate . By IHC, WT1 (90–95%) (Fig. H) and ETV4 (95–100%) are positive and are extremely useful markers . CD99 is positive albeit patchy and cytoplasmic , rather than membranous. However, NKX2.2 is typically negative . Sarcomas with CIC::NUTM1 fusions are positive for NUT protein . Molecular testing reveals CIC- related fusions. BCOR-related sarcoma is a primitive round cell sarcoma showing BCOR genetic alterations. These tumors typically affect children with > 90% of patients being < 20 years . Histology typically reveals vague nesting, round cells often admixed with focal spindled cells, pale nuclear chromatin, inconspicuous nucleoli, and abundant myxoid stroma with delicate vascularity (Fig. E) . By IHC, tumor cells show diffuse, strong BCOR (Fig. F), SATB2, and cyclin D1 positivity. CD99 is seen in about 50% of cases , but neuroendocrine markers are usually absent.
These are round and spindle cell sarcomas with EWSR1 or FUS fusions involving partners unrelated to the ETS gene family. These mainly comprise EWSR1::NFATC2 and FUS::NFATC2 sarcomas and EWSR1::PATZ1 sarcomas . Unlike conventional ES, these tumors exhibit atypical morphological features in the form of scattered enlarged cells, prominent nucleoli, or unusual clinical profiles (older patients). Nonetheless, there is considerable overlap with ES, including membranous CD99 staining. Although the pathologic spectrum is wide, some phenotypic clues to underlying genotypes can be helpful. Sarcomas with NFATC2 fusions tend to exhibit epithelioid features , while PATZ1 sarcomas are composed of largely undifferentiated round to ovoid neoplastic cells in a frequently sclerotic background . NFATC2 sarcomas express diffuse CD99 (like ES) in about 50% of cases; NKX2.2, dot-like keratin, and PAX7 positivity may also be observed . PATZ1 sarcomas do not consistently express CD99, however, may variably express CD34, and show a divergent phenotype with both myogenic (desmin, myogenin, MyoD1) and neurogenic (S100 protein, SOX10) markers , while neuroendocrine markers are usually absent. Identification of the fusion transcripts on molecular testing is the gold standard. CIC-rearranged sarcomas are round cell undifferentiated sarcomas that are defined by CIC -related gene fusions, mostly CIC::DUX4 fusion (about 95%) . CIC sarcomas are composed of undifferentiated round cells, however, tend to show lobulated growth (at least focally), and delicate fibrous septae; cells display mild pleomorphism and possess vesicular chromatin and prominent nucleoli (Fig. G). At times, epithelioid morphology can predominate . By IHC, WT1 (90–95%) (Fig. H) and ETV4 (95–100%) are positive and are extremely useful markers . CD99 is positive albeit patchy and cytoplasmic , rather than membranous. However, NKX2.2 is typically negative . Sarcomas with CIC::NUTM1 fusions are positive for NUT protein . Molecular testing reveals CIC- related fusions. BCOR-related sarcoma is a primitive round cell sarcoma showing BCOR genetic alterations. These tumors typically affect children with > 90% of patients being < 20 years . Histology typically reveals vague nesting, round cells often admixed with focal spindled cells, pale nuclear chromatin, inconspicuous nucleoli, and abundant myxoid stroma with delicate vascularity (Fig. E) . By IHC, tumor cells show diffuse, strong BCOR (Fig. F), SATB2, and cyclin D1 positivity. CD99 is seen in about 50% of cases , but neuroendocrine markers are usually absent.
Rhabdomyosarcoma (RMS) is a malignant mesenchymal tumor composed of primitive cells exhibiting skeletal muscle differentiation. Head and neck RMS account for about 35–40% of all RMS cases . It encompasses embryonal, alveolar, pleomorphic, and spindle/sclerosing subtypes. Among the subtypes, alveolar rhabdomyosarcoma (ARMS), particularly the solid subtype, most closely mimics SCNEC . In comparison to SCNEC, the patients of ARMS are much younger, with the peak age of ARMS being 10–25-year-old young adults , although cases in adults > 45 years in the sinonasal tract especially are not uncommon . Microscopically, ARMS is characterized by cellular nests of small round cells separated by fibrovascular septae. Toward the center, the tumor cells tend to be dyscohesive conferring an alveolar configuration to the tumor; the latter is a vital diagnostic clue in favor of ARMS. In contrast, the solid subtype of ARMS is composed of diffuse sheets and lacks this nested/alveolar pattern and fibrovascular septae making it morphologically indistinguishable from NEC and small blue round cell tumors (Fig. A) . By IHC, cytoplasmic desmin (Fig. B), diffuse nuclear myogenin (Fig. C), and focal nuclear MyoD1 positivity are diagnostic of ARMS. Notably, neuroendocrine markers and keratins can be expressed in some cases of RMS . Specific neuroendocrine markers (chromogranin A and/or synaptophysin) can be seen in up to 43% of cases (Fig. D) . About 32% of cases can express both cytokeratins and NE markers . This aberrant keratin and neuroendocrine marker expression in RMS can lead to an erroneous diagnosis of NEC if skeletal muscle markers are not employed. Hence, a panel of markers is essential to avoid diagnostic pitfalls. Molecular testing for ARMS diagnosis and prognostication is recommended although not necessary for distinguishing ARMS from NEC. The majority (~ 70–90%) of ARMSs contain PAX3::FOXO1 fusions with the remaining tumors generally PAX7::FOXO1 .
Synovial sarcoma (SS) is a soft tissue sarcoma showing variable epithelial differentiation and is characterized by SS18::SSX1, SSX2 , or SSX4 fusions . Although SS can occur at any age, the majority of patients are adolescents or young adults and < 2% of patients are older than 50 years at diagnosis . Histologically, SS are cellular monophasic or biphasic tumors composed of dense sheets or vague fascicles of uniform appearing small spindle cells with ovoid, hyperchromatic nuclei with regular granular chromatin and inconspicuous nucleoli, and scant cytoplasm (Fig. E, F). A variable proportion of epithelial cells are seen intermixed with spindle components in the biphasic SS , yielding a marbled appearance on low power. The high cellularity and monomorphic appearance frequently place SS in the list of small round cell tumors, especially in limited biopsy material. The poorly differentiated subtype of SS particularly needs distinction from SCNEC. Poorly differentiated SS exhibits areas of increased cellularity, greater nuclear pleomorphism, and a high mitotic rate (> 10 mitoses per 2 mm 2 ) (Fig. F) . The cells may be spindle to round. The tumors with predominantly round cell morphology especially necessitate segregation from SCNEC . Poorly differentiated tumors also tend to be more common in elderly patients . By IHC, SS shows strong diffuse nuclear positivity for TLE1 in nearly all the cases (Fig. G), with variable positivity for CD99 and bcl2, with focal positivity for cytokeratin. Rare reports of neuroendocrine markers (synaptophysin, chromogranin A, and nestin) in FISH-confirmed SS have been reported . Recently, newer antibodies, SS18::SSX fusion-specific antibody (E9X9V, reactive against the breakpoint) and the SSX -specific antibody (E5A2C, reactive against the SSX C-terminus) have shown strong diffuse nuclear staining with excellent sensitivity and specificity (> 95%) for SS .
Lymphoma is a universal differential diagnosis for all small blue round cell tumors. High-grade diffuse large cell lymphomas, B-cell or T-cell lineage, show diffuse sheets of neoplastic cells with high nuclear-to-cytoplasmic ratio, loose chromatin, conspicuous nucleoli, and scanty cytoplasm (Fig. H). Brisk mitoses and apoptotic bodies are frequent. Crushing artifacts are common. Convoluted nuclei, nuclear folds, and grooves are commonly seen in T-cell lineage tumors. Tumor cells infiltrating through fibrotic stroma may simulate clustering similar to carcinomas. Due to overlapping features, distinction from NEC is usually required, especially in a limited biopsy. Immunohistochemistry can readily help in segregating lymphomas from NEC. Lymphomas are positive for hematolymphoid markers including CD45RB, while negative with pancytokeratins and neuroendocrine markers (Fig. ).
Pathology is often considered a specialty in which experience is measured in case volume or years of practice—and given the rarity of neuroendocrine neoplasms and their mimics, diagnosticians outside of tertiary diagnostic centers may have difficulties in acquiring adequate experience for some of these entities. Moreover, identification of these lesions in the head and neck region usually requires an integration of clinical information, imaging findings, histomorphology, and immunohistochemical assessments and not all hospital settings may provide the latest antibody panels or molecular platforms to assess some of the key features. Even so, when faced with a head and neck tumor in which NEC is a potential differential diagnosis, careful exclusion of the top ten mimickers as highlighted herein will facilitate narrowing the diagnoses to the correct classification.
|
Speleomycology of Air in Stopića Cave (Serbia) | 03035319-b07e-485a-b565-f2536aaaa7d5 | 10064612 | Microbiology[mh] | Caves are defined as subterranean sites, characterized by constant temperatures and relative humidity all year long and with a moderate to low organic carbon input due to the absence of primary production . However, a variety of substrata within the caves are susceptible to microbial colonization, mostly by oligotrophic microorganisms . In general, fungi can colonize any organic matter present in subterranean sites, and among the cave microbiota, they have a significant role as dwellers in different microniches, such as sediments, vermiculations on cave walls, carcasses of troglobites or troglophiles, and bat guano . Fungal cave dwellers play a significant ecological role as decomposers and parasites . According to Wasti et al. , fungi are among the most dominant cave organisms due to their high rate of spores’ dissemination, capability for colonization of various substrata, and tolerance to a wide range of pH values. Unfortunately, some fungal cave inhabitants may be hazardous for mammal health, such as Pseudogymnoascus destructans , a causative agent of white-nose syndrome in Chiroptera , and Histoplasma capsulatum , which triggers systemic histoplasmosis (“Darling’s disease”) in humans and others . Furthermore, in caves with Paleolithic art, as well as in caves repurposed as sacral objects during the history of mankind, fungi can cause biodeterioration of both prehistoric wall paintings as well as murals and other artifacts deposited within [ – ]. In that sense, due to the significance of fungal presence in the caves, Polish scientists introduced a novel term — speleomycology, which encompasses all kinds of research with main focus on cave mycobiota . Fungal propagules of underground sites are suspended as airborne particles, which form air fractions called bioaerosols . Furthermore, fungi are also present as propagules carried by water, bats, arthropods, and humans, although airborne ones are the most abundant . In the caves of touristic significance (show caves), novel amounts of fungal propagules are frequently introduced by lint, hair, and dander of human visitors . Viable propagules settled on different surfaces within the caves, along with favorable growth conditions (humidity, temperature, nutrients), lead to successful colonization of available substrata, enabling fungi to complete their life cycle and to form another generation of spores within the caves’ interior . In that sense, aerobiological analyses have been proved as a suitable approach for investigations of airborne fungal dispersion in subterranean environments in the last decade. These investigations are therefore essential for the detection of the potential hazardous effect of fungi on the visitors’ health, potential detrimental effect on the cave art, and as agents of geological alterations of cave walls and sediments . The aim of this research was to analyze the content of airborne fungal propagules in different parts of “Stopića Cave”, a show cave in Serbia, and to link the fungal presence with different environmental factors and to discuss their features and relation to cave habitat.
Study Area and Sampling Sites Stopića Cave is located on the left bank of the river Prištavica, at an altitude of 711 m, on the eastern edge of the Zlatibor mountain in western parts of Serbia. The entrance to the cave is 30–40 m wide and is located at the bottom of a 50-m high vertical limestone cliff . Stopića Cave consists of five morphological units: Light Hall, Dark Hall, Hall with Tubs, Channel with Tubs, and River Channel . Light Hall — the very beginning of the cave (the entrance zone), Dark Hall (transitional zone), and Hall with Tubs are accessible and adapted for tourists. The most famous tourist attraction is the hall with Tufa tubs (rimstone dams), where many of Tufa tubs are recognized as the biggest and the deepest (over 7 m) of all other caves in Serbia . The cave has been protected by the Serbian government as a natural good of the first category since 2005. Sampling was carried out in three seasons during 2020: Winter (mid-January 2020, prior to the COVID-19 pandemic lockdown), Spring (early June, at the beginning of the tourist season peak, after COVID-19 pandemic lockdown), and Summer (early September, at the end of the tourist season peak), and in four selected sites within the cave: entrance, crossroads, Tufa bathtubs, and waterfall (Fig. ). The opening hours in January are shorter (7 h) than in June and September (8–9 h). Measurement of Microclimate Parameters Environmental parameters, temperature (T, °C), and relative humidity (RH, %) were measured in situ using temperature/humidity meter (VELLEMAN DEM105). Air Sampling and Estimation of Propagule Concentrations Airborne fungi were sampled by the volumetric air sampling method using the SAS Super DUO 360 Air Sampler. Cave indoor air (100 L) was vacuumed and subsequently inoculated on three different mycological media: potato dextrose agar (PDA) and Sabouraud dextrose agar (SDA) for mesophilic fungi and malt yeast 40% sucrose agar (M40Y) for xerophiles and xerotolerants. Before each sampling, the sampler was sterilized with ethanol (70 %) to prevent cross-contamination. Petri dishes with inoculated nutrient media were transported to the laboratory and incubated in a thermostat (UE 500, Memmert) at 25 ± 2 °C for a week. After the incubation period, the Petri dishes were examined, and all visible colonies were counted and labeled. The obtained numbers of the colonies were then converted to probable statistical total values and calculated by formula given by Feller and expressed as CFU m −3 of air. All morphologically different colonies were inoculated in order to obtain pure cultures. Identification of Fungi Isolated fungi were inspected using stereomicroscope (Stemi DV4, Carl Zeiss) and light microscope Zeiss Axio Imager M.1) and identified according to morphological criteria (colony morphology and microscopic characteristics of reproductive structures) using identification keys: Watanabe and Samson et al. . Molecular methods are used to confirm preliminary identification, as well to identify non-sporulating isolates. In that sense primary fungal isolates were reinoculated on PDA, SDA, and M40Y media and incubated at 25 ± 2 °C for a week. For DNA extraction, dry peripheral mycelia (approx. 40 mg) were harvested according to the manufacturer’s instructions of a DNeasy Plant Mini Kit (Qiagen, Valencia, CA, USA). PCR amplification of the ITS region and BenA gene was carried out using selected ITS1/ITS4 and Bt2a/Bt2b primers , respectively, as described previously . The amplified DNA fragments were fractionated in agarose gels (1%) in 0.5 × TBE buffer. Midori Green stain was used for DNA visualization by UV illumination . The obtained PCR products were then shipped for purification and sequencing to Macrogene (the Netherlands). The resulting sequences were then compared with other related sequences deposited to the National Center for Biotechnology Information (NCBI) using the BLAST program (BLAST+ 2.7.1 of the NCBI). Finally, obtained fungal DNA sequences were then deposited in the relevant GenBank database of NCBI. Phylogenetic Analysis of Airborne Fungal Communities Sequence alignment was carried out using the CLUSTALW algorithm in MEGAX software . The phylogenetic tree was built on the basis of the alignment and DNA sequences comparison by employing maximum likelihood phylogeny (1000 bootstrap replicas). Kimura 2 parameter model was determined as the best for estimating genetic distances between tested sequences — measured in the terms of nucleotide substitutions per site. Rhizophydium brooksianum JEL 136 (NR_119550.1) was used as the outgroup. Ecological Indicators In order to estimate indoor air quality in the “Stopića pećina” cave, obtained results of the concentration of fungal propagules in the different cave rooms were expressed in CFU m −3 and compared with ecological indicators proposed by Porca et al. . According to these authors, the true cave atmosphere has been classified into five categories based on fungal propagule concentration: (1) caves with fungal concentration less than 50 CFU m −3 (category I — indicating no problem with fungi); (2) caves with fungal concentration between 50 and 150 CFU m −3 (category II — requires some periodic controls and studies to eliminate fungal problem); (3) caves with fungal concentration between 150 and 500 CFU m −3 (category III — threatened by fungi and requires different cave management and controls); (4) caves with fungal concentration between 500 and 1000 CFU m −3 (Category IV – already affected by fungi as a result of massive visits or spillage); and (5) caves with fungal concentration above 1000 CFU m −3 (category V — have irreversible ecological disturbance). Fungal Communities by Trophic Mode and Ecological Guild All fungal isolates were sorted using FUNGuild v1.0 tool (Guilds_ v1.1.py script, database: http://stbates.org/funguild_db.php ) and literature data [ , – ]. Species were classified on the basis of their trophic mode into ecological categories: pathotroph (P), saprotroph (S), and symbiotroph (Sy), and to corresponding ecological guilds: animal pathogen (ap), endophyte (en), epiphyte (ep), fungal parasite (fp), lichen parasite (lp), litter saprotroph (ls), soil saprotroph (ss) plant pathogen (pp), undefined saprotroph (us), and wood saprotroph (ws). Statistical Analyses Statistical analyses were performed using software Canoco for Windows , Microsoft Excel, and the statistical package XLSTAT . Canonical correspondence analysis (CCA) was done to see the relationship between documented fungal genera (presence/absence) and sampling season that is used as explanatory variable, while temperature, relative air humidity, and CFU m −3 were used as supplementary variables. Documented fungal taxa were observed also against variables referring to cultivation medium and part of the cave where sampling was done, but those analyses showed no significance. Categories (CI–CV) of cave atmosphere based on fungal propagule concentration proposed by Porca et al. were also observed in relation to season, part of the cave and medium using constrained analysis (CCA). Only analysis that shows categories (CI–CV) of cave atmosphere in relation to sampling season was significant.
Stopića Cave is located on the left bank of the river Prištavica, at an altitude of 711 m, on the eastern edge of the Zlatibor mountain in western parts of Serbia. The entrance to the cave is 30–40 m wide and is located at the bottom of a 50-m high vertical limestone cliff . Stopića Cave consists of five morphological units: Light Hall, Dark Hall, Hall with Tubs, Channel with Tubs, and River Channel . Light Hall — the very beginning of the cave (the entrance zone), Dark Hall (transitional zone), and Hall with Tubs are accessible and adapted for tourists. The most famous tourist attraction is the hall with Tufa tubs (rimstone dams), where many of Tufa tubs are recognized as the biggest and the deepest (over 7 m) of all other caves in Serbia . The cave has been protected by the Serbian government as a natural good of the first category since 2005. Sampling was carried out in three seasons during 2020: Winter (mid-January 2020, prior to the COVID-19 pandemic lockdown), Spring (early June, at the beginning of the tourist season peak, after COVID-19 pandemic lockdown), and Summer (early September, at the end of the tourist season peak), and in four selected sites within the cave: entrance, crossroads, Tufa bathtubs, and waterfall (Fig. ). The opening hours in January are shorter (7 h) than in June and September (8–9 h).
Environmental parameters, temperature (T, °C), and relative humidity (RH, %) were measured in situ using temperature/humidity meter (VELLEMAN DEM105).
Airborne fungi were sampled by the volumetric air sampling method using the SAS Super DUO 360 Air Sampler. Cave indoor air (100 L) was vacuumed and subsequently inoculated on three different mycological media: potato dextrose agar (PDA) and Sabouraud dextrose agar (SDA) for mesophilic fungi and malt yeast 40% sucrose agar (M40Y) for xerophiles and xerotolerants. Before each sampling, the sampler was sterilized with ethanol (70 %) to prevent cross-contamination. Petri dishes with inoculated nutrient media were transported to the laboratory and incubated in a thermostat (UE 500, Memmert) at 25 ± 2 °C for a week. After the incubation period, the Petri dishes were examined, and all visible colonies were counted and labeled. The obtained numbers of the colonies were then converted to probable statistical total values and calculated by formula given by Feller and expressed as CFU m −3 of air. All morphologically different colonies were inoculated in order to obtain pure cultures.
Isolated fungi were inspected using stereomicroscope (Stemi DV4, Carl Zeiss) and light microscope Zeiss Axio Imager M.1) and identified according to morphological criteria (colony morphology and microscopic characteristics of reproductive structures) using identification keys: Watanabe and Samson et al. . Molecular methods are used to confirm preliminary identification, as well to identify non-sporulating isolates. In that sense primary fungal isolates were reinoculated on PDA, SDA, and M40Y media and incubated at 25 ± 2 °C for a week. For DNA extraction, dry peripheral mycelia (approx. 40 mg) were harvested according to the manufacturer’s instructions of a DNeasy Plant Mini Kit (Qiagen, Valencia, CA, USA). PCR amplification of the ITS region and BenA gene was carried out using selected ITS1/ITS4 and Bt2a/Bt2b primers , respectively, as described previously . The amplified DNA fragments were fractionated in agarose gels (1%) in 0.5 × TBE buffer. Midori Green stain was used for DNA visualization by UV illumination . The obtained PCR products were then shipped for purification and sequencing to Macrogene (the Netherlands). The resulting sequences were then compared with other related sequences deposited to the National Center for Biotechnology Information (NCBI) using the BLAST program (BLAST+ 2.7.1 of the NCBI). Finally, obtained fungal DNA sequences were then deposited in the relevant GenBank database of NCBI.
Sequence alignment was carried out using the CLUSTALW algorithm in MEGAX software . The phylogenetic tree was built on the basis of the alignment and DNA sequences comparison by employing maximum likelihood phylogeny (1000 bootstrap replicas). Kimura 2 parameter model was determined as the best for estimating genetic distances between tested sequences — measured in the terms of nucleotide substitutions per site. Rhizophydium brooksianum JEL 136 (NR_119550.1) was used as the outgroup.
In order to estimate indoor air quality in the “Stopića pećina” cave, obtained results of the concentration of fungal propagules in the different cave rooms were expressed in CFU m −3 and compared with ecological indicators proposed by Porca et al. . According to these authors, the true cave atmosphere has been classified into five categories based on fungal propagule concentration: (1) caves with fungal concentration less than 50 CFU m −3 (category I — indicating no problem with fungi); (2) caves with fungal concentration between 50 and 150 CFU m −3 (category II — requires some periodic controls and studies to eliminate fungal problem); (3) caves with fungal concentration between 150 and 500 CFU m −3 (category III — threatened by fungi and requires different cave management and controls); (4) caves with fungal concentration between 500 and 1000 CFU m −3 (Category IV – already affected by fungi as a result of massive visits or spillage); and (5) caves with fungal concentration above 1000 CFU m −3 (category V — have irreversible ecological disturbance).
All fungal isolates were sorted using FUNGuild v1.0 tool (Guilds_ v1.1.py script, database: http://stbates.org/funguild_db.php ) and literature data [ , – ]. Species were classified on the basis of their trophic mode into ecological categories: pathotroph (P), saprotroph (S), and symbiotroph (Sy), and to corresponding ecological guilds: animal pathogen (ap), endophyte (en), epiphyte (ep), fungal parasite (fp), lichen parasite (lp), litter saprotroph (ls), soil saprotroph (ss) plant pathogen (pp), undefined saprotroph (us), and wood saprotroph (ws).
Statistical analyses were performed using software Canoco for Windows , Microsoft Excel, and the statistical package XLSTAT . Canonical correspondence analysis (CCA) was done to see the relationship between documented fungal genera (presence/absence) and sampling season that is used as explanatory variable, while temperature, relative air humidity, and CFU m −3 were used as supplementary variables. Documented fungal taxa were observed also against variables referring to cultivation medium and part of the cave where sampling was done, but those analyses showed no significance. Categories (CI–CV) of cave atmosphere based on fungal propagule concentration proposed by Porca et al. were also observed in relation to season, part of the cave and medium using constrained analysis (CCA). Only analysis that shows categories (CI–CV) of cave atmosphere in relation to sampling season was significant.
Fungal Concentration The maximal measured fungal propagule concentrations were recorded during spring sampling on all nutrient media used: entrance (3400 CFU m −3 , 2680 CFU m −3 and 2040 CFU m −3 on PDA, SAB and M40Y respectively) and crossroads (2490 CFU m −3 on PDA). Furthermore, only three sampling points during spring showed fungal propagule concentrations lower than 1000 CFU m −3 : Tufa bathtubs (980 CFU m −3 and 930 CFU m −3 on M40Y and SAB respectively) and waterfall (510 CFU m −3 on M40Y). During all other sampling periods, obtained fungal concentrations were lower than 1000 CFU m −3 and the lowest were detected on crossroads sampling point in the summer on SAB (50 CFU m −3 ). Categories (CI–CV) of cave atmosphere based on fungal propagule concentration were observed in relation to season, growth media and part of the cave where the sampling was done (Fig. ). Category CV was exclusively documented in spring, CIV predominantly in summer, while CIII and CII in winter. Finally, CI was documented only once in the sampling point “crossroads” during the summer on SAB (Figs. and a). CCA analysis showing relationship of categories and seasons confirmed the mentioned distribution and was found significant ( F = 9.7, P = 0.002). Considering part of the cave where the sampling was done, according to performed multivariate analyses, CI was connected only to crossroads, CII to entrance, crossroads and Tufa bathtubs, CIII to Tufa bathtubs and waterfall, CIV and CV to all sections. All were detected on all three media, except CI that was documented only on SAB. Identified Fungi In the study presented here, a total of 29 fungal isolates were identified to species level, apart from one Aspergillus isolate which is identified to the section ( A. sect. Nigri ) and a Stereum isolate which was identified to the genus level. The most diverse genus was Aspergillus with 4 identified species, followed by Cladosporium , Fusarium , Lecanicillium , Mucor , and Penicillium genera with 3 identified species (Table ). A majority of identified fungi were members of the division Ascomycota (79.31%), followed by Mucoromycota (13.79%) and Basidiomycota (6.9%). Aspergilli and Penicillia formed a well-supported Eurotiales clade (bootstrap value, bpv = 96). Likewise, members of genera Lecanicillium , Cordyceps , Trichoderma , and Fusarium formed well-supported Hypocreales clade (bpv = 95), while Alternaria spp. and Ascochyta phacae grouped together as members of Pleosporales clade (bpv = 95). Botrytis cynerea grouped as a member of Leotiales clade (bpv = 98), while Cladosporium spp. were grouped as members of Capnodiales clade (bpv = 96). Within the division Mucoromycota two clades were present: Mucorales (bpv = 99) and Mortieralles (bpv = 100), while two well-supported clades Polyporales (bpv = 100) and Russulales (bpv = 100) were within division Basidiomycota (Fig. ). CCA ( F = 2.2, P = 0.002) representing documented fungal genera in relation to sampling season is shown in Fig. a. Taxa characteristics for only one season were Ascochyta , Bjerkandera , and Stereum (winter), Mortierella and Mucor (spring), and Lecanicillium and Botrytis (summer). Alternaria was recorded in spring and summer and Epicoccum and Trichoderma in winter and summer. The rest of documented taxa were found in all three seasons. T and RH had higher values during spring and summer and had positive correlation with Alternaria , Botrytis , Lecanicillium , Mortierella , Mucor , and Fusarium (Fig. b). The rest were negatively correlated with these two parameters. It is also observed that CFU m −3 was highest in spring and lowest in winter. When part of the cave where sampling was performed and growth medium was used as explanatory variables in relation to fungal taxa (in separate CCA analyses), the analyses were not significant. However, certain taxa were found in certain sections only: entrance, Bjerkandera ; crossroads, Stereum ; Tufa bathtubs, Botrytis ; and waterfall, Ascochyta and Trichoderma . Considering medium, Ascochyta and Trichoderma grew only on PDA medium, Stereum and Bjerkandera on SAB, while M40Y did not have specific taxon growing only on this medium ( Botrytis appeared on M40Y and PDA). Ecological Categories of Cave Mycobiota Most of isolated fungal species were pathotrophs (75.86%) which were followed by saprotrophs (58.62%), while symbiotrophs were the least frequent (27.59%). On the other hand, when regarding ecological guilds, the most dominant were undefined saprobes and animal pathogens (41.38% for each) followed by plant pathogens (34.48%) and endophytes (27.59%). Wood saprobes represented 13.79% of species while all other groups were present with only 3.45%. Pathotrophs were dominated by animal pathogens (54.55% of all pathotrophs) but closely followed by plant pathogens (45.45%). The only fungal pathotroph was Trichoderma harzianum , and the only lichen pathotroph was Epicoccum nigrum . Likewise, saprotrophes were mostly represented by undefined saprotrophes (70.59% of all saprotrophes) which were followed by wood saprotrophes (23.52%). Finally, all symbionts were endophytes, while only Cladosporium cladosporioides also belonged to epiphyte guild.
The maximal measured fungal propagule concentrations were recorded during spring sampling on all nutrient media used: entrance (3400 CFU m −3 , 2680 CFU m −3 and 2040 CFU m −3 on PDA, SAB and M40Y respectively) and crossroads (2490 CFU m −3 on PDA). Furthermore, only three sampling points during spring showed fungal propagule concentrations lower than 1000 CFU m −3 : Tufa bathtubs (980 CFU m −3 and 930 CFU m −3 on M40Y and SAB respectively) and waterfall (510 CFU m −3 on M40Y). During all other sampling periods, obtained fungal concentrations were lower than 1000 CFU m −3 and the lowest were detected on crossroads sampling point in the summer on SAB (50 CFU m −3 ). Categories (CI–CV) of cave atmosphere based on fungal propagule concentration were observed in relation to season, growth media and part of the cave where the sampling was done (Fig. ). Category CV was exclusively documented in spring, CIV predominantly in summer, while CIII and CII in winter. Finally, CI was documented only once in the sampling point “crossroads” during the summer on SAB (Figs. and a). CCA analysis showing relationship of categories and seasons confirmed the mentioned distribution and was found significant ( F = 9.7, P = 0.002). Considering part of the cave where the sampling was done, according to performed multivariate analyses, CI was connected only to crossroads, CII to entrance, crossroads and Tufa bathtubs, CIII to Tufa bathtubs and waterfall, CIV and CV to all sections. All were detected on all three media, except CI that was documented only on SAB.
In the study presented here, a total of 29 fungal isolates were identified to species level, apart from one Aspergillus isolate which is identified to the section ( A. sect. Nigri ) and a Stereum isolate which was identified to the genus level. The most diverse genus was Aspergillus with 4 identified species, followed by Cladosporium , Fusarium , Lecanicillium , Mucor , and Penicillium genera with 3 identified species (Table ). A majority of identified fungi were members of the division Ascomycota (79.31%), followed by Mucoromycota (13.79%) and Basidiomycota (6.9%). Aspergilli and Penicillia formed a well-supported Eurotiales clade (bootstrap value, bpv = 96). Likewise, members of genera Lecanicillium , Cordyceps , Trichoderma , and Fusarium formed well-supported Hypocreales clade (bpv = 95), while Alternaria spp. and Ascochyta phacae grouped together as members of Pleosporales clade (bpv = 95). Botrytis cynerea grouped as a member of Leotiales clade (bpv = 98), while Cladosporium spp. were grouped as members of Capnodiales clade (bpv = 96). Within the division Mucoromycota two clades were present: Mucorales (bpv = 99) and Mortieralles (bpv = 100), while two well-supported clades Polyporales (bpv = 100) and Russulales (bpv = 100) were within division Basidiomycota (Fig. ). CCA ( F = 2.2, P = 0.002) representing documented fungal genera in relation to sampling season is shown in Fig. a. Taxa characteristics for only one season were Ascochyta , Bjerkandera , and Stereum (winter), Mortierella and Mucor (spring), and Lecanicillium and Botrytis (summer). Alternaria was recorded in spring and summer and Epicoccum and Trichoderma in winter and summer. The rest of documented taxa were found in all three seasons. T and RH had higher values during spring and summer and had positive correlation with Alternaria , Botrytis , Lecanicillium , Mortierella , Mucor , and Fusarium (Fig. b). The rest were negatively correlated with these two parameters. It is also observed that CFU m −3 was highest in spring and lowest in winter. When part of the cave where sampling was performed and growth medium was used as explanatory variables in relation to fungal taxa (in separate CCA analyses), the analyses were not significant. However, certain taxa were found in certain sections only: entrance, Bjerkandera ; crossroads, Stereum ; Tufa bathtubs, Botrytis ; and waterfall, Ascochyta and Trichoderma . Considering medium, Ascochyta and Trichoderma grew only on PDA medium, Stereum and Bjerkandera on SAB, while M40Y did not have specific taxon growing only on this medium ( Botrytis appeared on M40Y and PDA).
Most of isolated fungal species were pathotrophs (75.86%) which were followed by saprotrophs (58.62%), while symbiotrophs were the least frequent (27.59%). On the other hand, when regarding ecological guilds, the most dominant were undefined saprobes and animal pathogens (41.38% for each) followed by plant pathogens (34.48%) and endophytes (27.59%). Wood saprobes represented 13.79% of species while all other groups were present with only 3.45%. Pathotrophs were dominated by animal pathogens (54.55% of all pathotrophs) but closely followed by plant pathogens (45.45%). The only fungal pathotroph was Trichoderma harzianum , and the only lichen pathotroph was Epicoccum nigrum . Likewise, saprotrophes were mostly represented by undefined saprotrophes (70.59% of all saprotrophes) which were followed by wood saprotrophes (23.52%). Finally, all symbionts were endophytes, while only Cladosporium cladosporioides also belonged to epiphyte guild.
Aerobiological studies conducted in cave environments could be used for the monitoring of fungal presence and for the prevention of potential fungal outbreaks in the caves of touristic importance. The results of quantitative and qualitative aeromycological analyses obtained in this study correspond to the work of other authors in the terms of fungal propagule concentration and species composition in subterranean sites [ , , ]. The measurement of the fungal propagule concentration in the summer period showed that the majority of tested sampling points belong to categories IV or V. According to Porca et al. , these categories are due to fungal outbreaks caused by frequent visitors. Bearing in mind that sampling of air-borne fungi in “Stopića Cave” has been carried out during COVID-19 pandemic, when due to restrictions to travel abroad, many Serbian citizens were encouraged to travel within Serbian borders, and in that particular time, “Stopića Cave” emerged as novel visiting hotspot for many tourists; high fungal propagule concentration during spring and summer could be connected to more frequent visits. Highest fungal loads in the air were documented during the spring and summer, i.e., after COVID-19 lockdown, which can be attributed to the higher influx of visitors during that time than during winter periods prior to lockdown. Other factors within the “Stopića pećina” cave that could lead to high concentrations of fungal propagules are vegetation period, and big entrance which make “Stopića Cave” more susceptible to climate factors in surroundings (wind, temperature, humidity), in addition to the narrow and short paths for visitors, which could be vectoric carriers of fungal propagules into the deeper parts of the cave. It should be emphasized that there are no official reference standards and limits regarding fungal propagule concentration in indoor air regarding the human health . However, severe impacts on human health are reported only upon exposure to concentrations above 50,000 CFU m −3 . Having in mind that visitors usually do not spend more than few hours during the tour within the cave, the high fungal propagule concentrations could not be regarded as a potential threat for human health. According to the official site https://www.zlatibor.org.rs/sr/sta-videti/atrakcije/stopica-pecina/ , 90,600 tourists visit cave during 2019, but in 2020, that number reached 100,000 which is an absolute record since the opening of the cave to the public (and it was achieved having in mind that the cave was closed for several months due to pandemic). Fernandez-Cortes et al. reported that in Galeria del Calvario room, famous for their paleolithic paintings and engravings in the cave Cueva de Ardales (Spain); fungal propagule concentration (expressed in CFU m −3 ) increased by 100 times after a visit of only 32 people. Similar trend was observed in our study since observed CFU counts were 10 to 30 times higher after the lockdown (in June) compared to the winter (prior to the lockdown). Therefore, most of the rooms during the summer period were classified into the category IV. On the other hand, Martin-Sanchez et al. reported high fungal contamination levels during winter sampling of aeromycobiota in all investigated rooms from famous Lascaux Cave (France) which contains valuable Paleolithic art. All investigated Lascaux rooms belonged to categories III or IV, and this contamination level was explained by convection currents created by the climate system established for preventing condensation of water vapor on the walls which evacuated the airborne bacteria and fungi. Dominguez-Moñino et al. investigated the aerobiology of caves in Southern Spain and reported a high level of fungal contamination for the Cueva del Tesoro cave. All tested rooms in this cave during spring sampling belonged to categories IV and V, as was documented in our study. Relatively small dimensions of the rooms and galleries, which could contribute to the concentration of spores, in addition to the abundance of phototrophic biofilms all over the cave walls are considered to be main factors contributing to the fungal propagule abundance in the air of Cueva del Tesoro Cave. Jurado et al. investigated aeromycobiota of Cueva de Nerja (Spain) and during sampling in winter found two rooms with extremely high fungal propagules concentration: Kitchen Hall and Heracles Hall with 2170 and 1330 CFU m −3 , respectively. However, and contrary to our findings, both mentioned that Cueva de Nerja rooms during previous sampling in summer had low fungal propagules concentration and belonged to category I. Kokurewicz et al. conducted the aeromycological study in Nietoperek Bat Reserve (Western Poland) and detected the highest level of fungal spores reaching the highest number in November and January which placed investigated rooms to category V, but the authors also noticed the number of fungal spores in the air significantly declining in March. According to Kokurewicz et al. , the number of bats in that hibernation site was the primary factor determining the fungal propagule concentration in the indoor air. Duan et al. conducted an aeromycological survey in cave temples Maijishan Grottoes in China and reported the highest fungal concentration of 1389 CFU m −3 (category V) in Upper Seven Buddha Pavilion during summer, which was positively correlated with relative humidity and higher visitor density. According to Bercea et al. , the high YM (yeast and molds) concentration in the cave Meziad (Western Romania) is correlated with the number of tourists and CO 2 level. The structures of airborne fungal communities documented in research presented here showed that no unique fungal composition could be distinguished on any sampling site. The majority of isolated fungi were members of genera: Alternaria , Aspergillus , Cladosporium , Fusarium , Lecanicilium , Mucor , and Penicillium . The results presented in this paper correspond with similar findings presented by other authors, but mostly of those reported by Rafael Ogórek, researcher who investigates air mycobiota from different caves in Slovakia. Ogórek et al. reported the presence of Alternaria alternata , Aspergillus fumigatus , A. niger , Botrytis cinerea , Cladosporium cladosporioides , and Epicoccum nigrum in the indoor air of Harmanecká Cave, the most important subterranean site of bat occurrence in Slovakia. In the air of Driny Cave, which is also open for public in Slovakia, A. alternata , A. fumigatus , C. cladosporioides , C. herbarum , E. nigrum , Fusarium equiseti , Mucor hiemalis , and Trichoderma harzianum were found. Also, the presence of A. niger , B. cinerea , C. cladosporioides , C. herbarum , E. nigrum , M. hiemalis was reported by Ogórek et al. for Demänovska´ Ice Cave, a very popular attraction in Slovakia. Apart from Slovakian caves, Popkova et al. reported the presence of A. alternata , A. fumigatus , A. niger , B. cinerea , C. cladosporioides , C. herbarum , M. hiemalis , Penicillium expansum , and T. harzianum in the indoor air of two caves: Novoafonskaya (Abkhazia, Georgia) and Ali-Sadr (Iran). The similarities in fungal diversity and mycobiota composition between “Stopića Cave” and investigated caves in Slovakia, Georgia, and Iran could be explained by several factors: (1) fungal species in common (members of genera Alternaria, Aspergillus, Cladosporium , Fusarium , Mucor, Penicillium, and Trichoderma ) are according to Vanderwolf et al. recognized as the most characteristic for the subterranean sites; (2) all fungal species in common produce small-sized conidia or sporangiospores which are easily suspended in the air as bioaerosols; and (3) all caves are open for visitors. Apart from ubiquitous, some other fungal isolates are documented less frequently and could be regarded as uncommon due to their narrow ecological valence. The presence of documented entomopathogenic fungi Cordyceps farinosa , and Lecanicilium psaliotae , members of animal pathogens’ guild , could be explained by troglobiotic entomofauna within the Stopića Cave, namely, “Stopića Cave” is a habitat of the stenoendemic subspecies of troglobiotic ground beetle (Carabidae) Rascioduvalius stopicensis . Also, trogloxenic ground beetle Trechus obtusus is also documented in Stopića Cave . It should be noted that Bjerkandera adusta and Stereum sp. are the only two basidiomycetes documented in this research, both of them members of wood saprotrophs’ guild, namely, the members of the division Basidiomycota are seldom reported as constituents of cave aeromycobiota. However, B. adusta and Trametes hirsuta were documented in the air of Demänovska´cave (Slovakia) by Ogórek et al. . Bercea et al. reported the presence of basidiomycetous yeast, Trichosporon sp., an opportunistic pathogen, in the cave “Ursilor” (Western Romania). Potential source of Cladosporia , Fusaria , Penicillia , Trichoderma spp., and Basidiomycota spores could be surrounding vegetation since they are potential plant pathogens, soil and wood decaying fungi. Furthermore, Stopića pećina is located on the slopes of Zlatibor mountain and is surrounded by beech forest and has large entrance, which enables air communication between exterior and interior of the cave and consequently spore flow via air currents. It should be emphasized that animal pathogens were documented with high frequency in the indoor air of Stopića pećina (41.38%), and hence, their presence in indoor environments could be regarded as a potential threat for visitors’ health. Among them, Aspergillus fumigatus is a known animal and human pathogen and a causative agent of pulmonary aspergillosis . However, it should be emphasized that A. fumigatus is in research presented here, documented only once (entrance, winter sampling), so it could be assumed the spores of this pathogen are not frequent inside the cave. Our work presents a risk assessment study which lays foundations for the further ecological investigations which will be carried out in the future. Health risk assessment is very important in the understanding of fungal diversity and is especially significant in assessing a potential impact for the visitors.
Caves of touristic importance are especially vulnerable to human impacts, and the fungal propagules’ concentration within the caves could be good indices for the level of ecological disturbance. In order to prevent fungal outbreaks in such caves, which could affect the health of visitors and cave workers, as well as the biodeterioration of cave art, speleomycological research, including aeromycological studies, is essential and therefore should be introduced to cave management as a routine monitoring.
|
Plant domestication shapes rhizosphere microbiome assembly and metabolic functions | b9c04166-4bf3-4c53-8c45-ec5434673760 | 10064753 | Microbiology[mh] | Plant domestication, a complex evolutionary process, is a crucial accomplishment in human history . It provides a continuous food supply and enhances the establishment of stable human settlements and has far-reaching consequences for human prosperity. The first crop plants were subjected to domestication ~ 13,000 years ago, and a plethora of crop plants, such as rice, wheat, and barley, have been undergoing continuous domestication by anthropogenic intervention in modern agricultural areas . The improvement of phenotypic traits of domesticated cultivars not only is determined by their own genetic features but also depends on their root-associated microbial communities. Pioneering studies have indicated that plant domestication profoundly influences seed and root microbiome assembly , particularly the rhizosphere microbiome . These affected rhizosphere microbiomes have important roles in crop performance and growth . Hence, the dissection of mechanisms underlying the species coexistence of these affected microbiomes and plant-microbiome interactions at the root-soil interface will provide new avenues for harnessing indigenous rhizobiomes to promote soil health and crop production . In addition to being subjected to the selective power of domestication, root-associated microbiomes are largely affected by various root exudates that are released by distinctive genotypes of plants . Root exudates include amino acids, sugars, organic acids, nucleotides, fatty acids, hormones, and secondary metabolites, which have direct and indirect impacts on nutrient availability in the rhizosphere . Increasing evidence highlights that host plants are able to harness root-associated microbiomes by releasing root exudates for plant nutrient acquisition [ , , ] and mitigating abiotic and biotic stresses during plant domestication processes . These studies highlight the importance of root exudation in shaping root-associated microbial communities under field and experiment conditions . However, we still lack comprehensive knowledge of how and to what degree crops under distinct domestication statuses control root exudation to shape rhizosphere microbiomes with specialized metabolic functions. The aforementioned plant-microbiome associations along the soil-root continuum not only are affected by root exudation but also are also profoundly influenced by microbe-microbe interactions. Emerging studies have indicated that these interactions are now recognized to implement crucial functions for promoting plant biodiversity and plant survival and protect hosts against soilborne pathogen infection . Importantly, the putative keystone taxa that strongly interact with other species in the microbial interkingdom network are suggested to play crucial roles in enhancing microbial network complexity , crop production , and ecosystem multi-functionality . However, knowledge is lacking on how these complex microbial interkingdom interactions and putative keystone taxa respond to plant domestication. Domesticated emmer wheat ( Triticum turgidum ssp. dicoccon ) evolved from wild emmer wheat ( Triticum turgidum ssp. dicoccoides ), which was subjected to domestication that began ~ 10,300 years ago in the Fertile Crescent . The domestication of emmer wheat was essential to human prosperity but also led to significant changes in wheat genetic diversity . Understanding the genetic consequences of the evolution and domestication history of tetraploid wheat in root-associated microbial community assembly and functions will pave the way for precise plant breeding strategies and soil health. Nevertheless, studies assessing the mechanisms of species coexistence and functional adaptation in response to the domestication of tetraploid wheat remain scarce. To fill this knowledge gap, the approach of “going back to the roots” was adopted, and 44 accessions of tetraploid wheat, including wild emmer and domesticated emmer, were selected in the current study. Herein, our aims were to (1) uncover the adaptive strategies and interkingdom interaction patterns of the root-associated microbiome during the wheat domestication process, (2) decipher the functional traits of the rhizosphere microbiome related to nutrient acquisition and dissect how the root exudation covaries with aboveground plant phenotypes in response to plant domestication, and (3) disentangle the crucial role of rhizosphere microbial consortium and individual keystone taxa in sustaining root morphology traits.
Selection of plant accessions Grains of 44 accessions of tetraploid wheat, i.e., 22 wild and 22 domesticated wheat, were obtained from the International Wheat Genome Sequencing Consortium (IWGSC) on the basis of our previous study . Seeds of the 44 tetraploid wheat accessions were planted annually for field experiments since 2014 at Caoxingzhuang Agro-Ecosystem Experimental Station of the Northwest Agriculture and Forest University, Shaanxi Province, China (34°17′N, 108°04′E). In the last 9 years, new seeds of the 44 tetraploid wheat accessions from the current year were planted annually for the current planting season. For winter wheat and fallow system, the summer fallow period of winter wheat is from late July to late September. The seeds of the 44 tetraploid wheat accessions were annually collected at the maturity stage (June 15th-June 30th). Then, these collected new seeds are continued to sow in the middle of October of current year. After nine times of planting season, the 44 tetraploid wheats were chosen for further use. At the heading stage of wheat, five replicates of each wheat accession were randomly chosen to measure the plant attributes, including plant height, ear length, subsegment length, and tiller (Table S ). Based on the phylogenetic tree of 44 tetraploid wheats that were captured from whole-genome sequences (Fig. ), the individual accessions with representative branches in the phylogenetic tree and the best plant performance of phenotypes, including three domesticated accessions D1, D2, and D3 from different wheat-growing regions of Kazakhstan, Espana, and Mexicanos and three wild emmer accessions W1, W2, and W3 from different wheat-growing regions of Turkey, Syria, and Jordan in six distinct branches, were further selected for this study (Table S ). More details for experiment design and field site management can be found here (Supplementary Notes: Method S ). Samples collection, DNA extraction, PCR amplification, and amplicon sequencing At the maturity stage on June 15, 2021, 60 soil samples were collected (5 replicates × 6 plant accessions × 2 compartments; Supplementary Notes: Method S ). All soil samples were stored at − 30 °C until further use. Additionally, the harvested seeds of each wheat accession were further collected to measure their phenotypic attributes. The total DNA from collected soil samples was separately extracted using the PowerSoil DNA Isolation Kit (Mo Bio Laboratories Inc., Carlsbad, CA, USA), according to the manufacturer’s instructions. For PCR amplification, the universal 515F and 806R PCR primers with barcodes were applied to amplify the V4 regions of 16S rRNA. The PCR primers ITS1F and ITS2R were used to amplify the ITS1 regions of fungal 18S rRNA. The PCR protocols for constructing 16S rRNA and fungal ITS libraries followed the manufacture’s instructions. After amplification, PCR products were pooled and purified with an AxyPrep DNA Gel Extraction Kit (Axgen, USA). Finally, the generated amplicon libraries were sequenced on an Illumina HiSeq 2500 PE250 platform (Majorbio, Co., Ltd., China). Following sequencing, the demultiplexing sequences, merging reads, and removing non-biological sequences for all raw reads were performed using the QIIME2 (2021.8, https://QIIME2.org)-CentOS 7.6 platform (Supplementary Notes: Method S ). Shotgun metagenomics sequencing The extracted DNA for amplicon sequencing were also used for metagenomics sequencing on an Illumina HiSeq 2500 Platform (150 bp paired-end reads). The raw metagenomics sequences were quality filtered using fastp (v0.23.1) based on a minimum Q Score of 20 and a minimum sequence length of 50 bp . The obtained clean reads were assembled into scaffolds individually using IDBA-UD v1.1.1 with default parameters . Scaffolds longer than 500 bp were used to predict open reading frames (ORFs) using MetaGeneMark (v0.43) . The generated ORFs were then clustered at 95% similarity to construct a nonredundant gene catalog using MMseqs2 . The nonredundant gene catalog was also searched against the eggNOG database (v5.0) and Kyoto Encyclopedia of Genes and Genomes (KEGG) database (release 80.1) using DIAMOND (v0.922.123), with an E -value < 1e−10 to annotate the predicted genes to clusters of orthologous groups (COG) and KEGG Orthology (KO) groups. For taxonomic assignment (Supplementary Notes: Method S ), the nonredundant genes were searched against the NCBI NR database (release March 15, 2020) using DIAMOND (v0.922.123) with an E -value cutoff of 1e−5 . Root exudate collection and LC–MS analysis After sampling of the rhizospheric soil, the intact root systems were used to collect root exudates according to a pioneering study with minor revision . We totally collected 36 root exudate samples (6 replicates × 6 plant accessions). Prior to detection, root exudate powders were redissolved in 100-μL sterilized deionized water for downstream analysis (Supplementary Notes: Method S ). After detection, the metabolites in each sample with a relative standard deviation (RSD) > 30% were removed from the dataset. The mass spectra of remaining metabolites were annotated with the Golm Metabolome Database (GMD, http://gmd.mpimp-golm.mpg.de/ ), METLIN database ( http://metlin.scripps.edu ), and KEGG) database (release 80.1). Finally, the annotated metabolites were log-transformed for further microbiome-wide association study. Rhizospheric microbiota inoculation and plant morphology traits analysis To verify whether rhizospheric microbiota change the root growth, the inoculation treatment with addition of live microbiota suspensions was conducted (Supplementary Notes: Method S ). The control treatment was denoted without addition of rhizospheric microbiota. Secondly, Microbacterium mitrae was selected to a follow-up experiment. Bacterial culture was normalized to OD600 = 0.2 for plant inoculation experiments. Control treatments received Hoagland nutrient solution instead of a bacteria suspension. Wheat seeds were surface sterilized with 75% ethanol for 30 s and 2.5% sodium hypochlorite for 15 min and then germinated on 0.5 × MS agar media for 5 days. Then, a 5-day-old sterile wheat seedlings were transferred to the Hoagland nutrient solution with or without inoculation microbial in the Magenta boxes at 27 °C with 16-h light/8-h dark cycle. The mixed nutrient solutions were changed at intervals of 3 days. Root-related parameters including root length, and number of root forks, etc., were scanned and registered using Microtek ScanMaker i800 plus system (WSeen, Hangzhou, China). The shoot fresh weights and dry weight of each seedling were measured after 15 days. Statistical analyses The α-diversity and β -diversity were determined using the “vegan” R package , unless otherwise indicated. For β -diversity, the ASV table was normalized by the trimmed mean of M -values (TMM) method using the “EdgeR” bioconductor package . The generated Bray–Curtis dissimilarity matrix retrieved from the “vegdist” function was used to perform further analyses, including an analysis of similarity (ANOSIM), principal coordinate analysis (PCoA), and permutational multivariate analysis of variance (PERMANOVA). To further confirm the constrained variables for the bacterial and fungal communities, constrained analysis of principal coordinates (CAP) was performed using the “capscale” function in the “vegan” R package and the “ordinate” function in the “phyloseq” R package . To determine the enriched and depleted ASVs in wild and domesticated wheat plants, differential abundance analysis was conducted using the generalized linear model (GLM) approach in the “EdgeR” package. More details for following analyses can be found in the here (Supplementary Notes: Method S ). Random forest classification and tenfold cross validation To explore the importance of different microbial taxonomy categories in structuring wild/domesticated microbiome assembly, the “random forest (RF)” method was used to perform machine learning classification. This step was implemented using the “importance” function in the “randomForest” R package . Subsequently, tenfold validation was performed to evaluate the accuracy of the RF model and to select the minimum number of ASVs with the lowest prediction error rate by using the “rfcv” function in the “randomForest” package. Finally, the 30 most important ASVs for bacteria and fungi from the RF model were categorized as wild/domesticated enriched and bulk/rhizosphere-enriched ASVs at the genus level depending on the results of the aforementioned differential abundance analysis. Microbial co-occurrence network analysis To explore the microbial hierarchical interactions between bacterial and fungal communities in the rhizosphere of wild and domesticated wheats, we implemented correlation-based network analysis by using the FastSpar algorithm . To minimize the bias of pairwise correlations, the core community for bacteria was generated based on the criteria of mean relative abundance and occurrence frequency of microbial communities in all samples. After that, core bacterial and fungal ASVs were merged to construct multi-kingdom ASV tables for further analysis. Only compositionality-robust (| ρ |> 0.7) and statistically significant ( q < 0.01) correlations were integrated into the hierarchical network analysis. The node-level topological features including the degree, betweenness, closeness, and eigenvectors were calculated using the “igraph” R package . The identification of keystone taxa was based on the criteria of the nodes with high degree, higher node transitivity, and low betweenness centralities. Finally, the generated co-occurrence network with keystone taxa and different modules was visualized using the Gephi platform (v0.92, https://gephi.org). Plant-microbiota-metabolite association analysis We then implemented partial least squares discriminant analysis (PLS-DA) and orthogonal partial least squares discriminant analysis (OPLS-DA) by using the “opls” function in the “ropls” bioconductor package to uncover the mechanisms that underlie the microbial response to plant domestication. Subsequently, ecological associations among plant features, significant metabolites, and enriched microbiota were calculated using the “rcorr” function in the “Hmisc” R package . Correlation matrix visualization was performed using the “ggtree” R package. Based on this association analysis, key relevant metabolites that were significantly enriched in wild or domesticated wheat were further used to evaluate their importance for plant growth. The importance of each metabolite was determined by evaluating the increase in the mean square error (MSE) between predictions and observations. This analysis was performed using the “rfPermute” and “rp. importance” functions in the “rfPermute” R package . Meanwhile, the significance of the importance of each metabolite associated with plant traits was assessed by using the “rfUtilities” R package The significance of this model was cross-validated using the “A3” R package .
Grains of 44 accessions of tetraploid wheat, i.e., 22 wild and 22 domesticated wheat, were obtained from the International Wheat Genome Sequencing Consortium (IWGSC) on the basis of our previous study . Seeds of the 44 tetraploid wheat accessions were planted annually for field experiments since 2014 at Caoxingzhuang Agro-Ecosystem Experimental Station of the Northwest Agriculture and Forest University, Shaanxi Province, China (34°17′N, 108°04′E). In the last 9 years, new seeds of the 44 tetraploid wheat accessions from the current year were planted annually for the current planting season. For winter wheat and fallow system, the summer fallow period of winter wheat is from late July to late September. The seeds of the 44 tetraploid wheat accessions were annually collected at the maturity stage (June 15th-June 30th). Then, these collected new seeds are continued to sow in the middle of October of current year. After nine times of planting season, the 44 tetraploid wheats were chosen for further use. At the heading stage of wheat, five replicates of each wheat accession were randomly chosen to measure the plant attributes, including plant height, ear length, subsegment length, and tiller (Table S ). Based on the phylogenetic tree of 44 tetraploid wheats that were captured from whole-genome sequences (Fig. ), the individual accessions with representative branches in the phylogenetic tree and the best plant performance of phenotypes, including three domesticated accessions D1, D2, and D3 from different wheat-growing regions of Kazakhstan, Espana, and Mexicanos and three wild emmer accessions W1, W2, and W3 from different wheat-growing regions of Turkey, Syria, and Jordan in six distinct branches, were further selected for this study (Table S ). More details for experiment design and field site management can be found here (Supplementary Notes: Method S ).
At the maturity stage on June 15, 2021, 60 soil samples were collected (5 replicates × 6 plant accessions × 2 compartments; Supplementary Notes: Method S ). All soil samples were stored at − 30 °C until further use. Additionally, the harvested seeds of each wheat accession were further collected to measure their phenotypic attributes. The total DNA from collected soil samples was separately extracted using the PowerSoil DNA Isolation Kit (Mo Bio Laboratories Inc., Carlsbad, CA, USA), according to the manufacturer’s instructions. For PCR amplification, the universal 515F and 806R PCR primers with barcodes were applied to amplify the V4 regions of 16S rRNA. The PCR primers ITS1F and ITS2R were used to amplify the ITS1 regions of fungal 18S rRNA. The PCR protocols for constructing 16S rRNA and fungal ITS libraries followed the manufacture’s instructions. After amplification, PCR products were pooled and purified with an AxyPrep DNA Gel Extraction Kit (Axgen, USA). Finally, the generated amplicon libraries were sequenced on an Illumina HiSeq 2500 PE250 platform (Majorbio, Co., Ltd., China). Following sequencing, the demultiplexing sequences, merging reads, and removing non-biological sequences for all raw reads were performed using the QIIME2 (2021.8, https://QIIME2.org)-CentOS 7.6 platform (Supplementary Notes: Method S ).
The extracted DNA for amplicon sequencing were also used for metagenomics sequencing on an Illumina HiSeq 2500 Platform (150 bp paired-end reads). The raw metagenomics sequences were quality filtered using fastp (v0.23.1) based on a minimum Q Score of 20 and a minimum sequence length of 50 bp . The obtained clean reads were assembled into scaffolds individually using IDBA-UD v1.1.1 with default parameters . Scaffolds longer than 500 bp were used to predict open reading frames (ORFs) using MetaGeneMark (v0.43) . The generated ORFs were then clustered at 95% similarity to construct a nonredundant gene catalog using MMseqs2 . The nonredundant gene catalog was also searched against the eggNOG database (v5.0) and Kyoto Encyclopedia of Genes and Genomes (KEGG) database (release 80.1) using DIAMOND (v0.922.123), with an E -value < 1e−10 to annotate the predicted genes to clusters of orthologous groups (COG) and KEGG Orthology (KO) groups. For taxonomic assignment (Supplementary Notes: Method S ), the nonredundant genes were searched against the NCBI NR database (release March 15, 2020) using DIAMOND (v0.922.123) with an E -value cutoff of 1e−5 .
After sampling of the rhizospheric soil, the intact root systems were used to collect root exudates according to a pioneering study with minor revision . We totally collected 36 root exudate samples (6 replicates × 6 plant accessions). Prior to detection, root exudate powders were redissolved in 100-μL sterilized deionized water for downstream analysis (Supplementary Notes: Method S ). After detection, the metabolites in each sample with a relative standard deviation (RSD) > 30% were removed from the dataset. The mass spectra of remaining metabolites were annotated with the Golm Metabolome Database (GMD, http://gmd.mpimp-golm.mpg.de/ ), METLIN database ( http://metlin.scripps.edu ), and KEGG) database (release 80.1). Finally, the annotated metabolites were log-transformed for further microbiome-wide association study.
To verify whether rhizospheric microbiota change the root growth, the inoculation treatment with addition of live microbiota suspensions was conducted (Supplementary Notes: Method S ). The control treatment was denoted without addition of rhizospheric microbiota. Secondly, Microbacterium mitrae was selected to a follow-up experiment. Bacterial culture was normalized to OD600 = 0.2 for plant inoculation experiments. Control treatments received Hoagland nutrient solution instead of a bacteria suspension. Wheat seeds were surface sterilized with 75% ethanol for 30 s and 2.5% sodium hypochlorite for 15 min and then germinated on 0.5 × MS agar media for 5 days. Then, a 5-day-old sterile wheat seedlings were transferred to the Hoagland nutrient solution with or without inoculation microbial in the Magenta boxes at 27 °C with 16-h light/8-h dark cycle. The mixed nutrient solutions were changed at intervals of 3 days. Root-related parameters including root length, and number of root forks, etc., were scanned and registered using Microtek ScanMaker i800 plus system (WSeen, Hangzhou, China). The shoot fresh weights and dry weight of each seedling were measured after 15 days.
The α-diversity and β -diversity were determined using the “vegan” R package , unless otherwise indicated. For β -diversity, the ASV table was normalized by the trimmed mean of M -values (TMM) method using the “EdgeR” bioconductor package . The generated Bray–Curtis dissimilarity matrix retrieved from the “vegdist” function was used to perform further analyses, including an analysis of similarity (ANOSIM), principal coordinate analysis (PCoA), and permutational multivariate analysis of variance (PERMANOVA). To further confirm the constrained variables for the bacterial and fungal communities, constrained analysis of principal coordinates (CAP) was performed using the “capscale” function in the “vegan” R package and the “ordinate” function in the “phyloseq” R package . To determine the enriched and depleted ASVs in wild and domesticated wheat plants, differential abundance analysis was conducted using the generalized linear model (GLM) approach in the “EdgeR” package. More details for following analyses can be found in the here (Supplementary Notes: Method S ).
To explore the importance of different microbial taxonomy categories in structuring wild/domesticated microbiome assembly, the “random forest (RF)” method was used to perform machine learning classification. This step was implemented using the “importance” function in the “randomForest” R package . Subsequently, tenfold validation was performed to evaluate the accuracy of the RF model and to select the minimum number of ASVs with the lowest prediction error rate by using the “rfcv” function in the “randomForest” package. Finally, the 30 most important ASVs for bacteria and fungi from the RF model were categorized as wild/domesticated enriched and bulk/rhizosphere-enriched ASVs at the genus level depending on the results of the aforementioned differential abundance analysis.
To explore the microbial hierarchical interactions between bacterial and fungal communities in the rhizosphere of wild and domesticated wheats, we implemented correlation-based network analysis by using the FastSpar algorithm . To minimize the bias of pairwise correlations, the core community for bacteria was generated based on the criteria of mean relative abundance and occurrence frequency of microbial communities in all samples. After that, core bacterial and fungal ASVs were merged to construct multi-kingdom ASV tables for further analysis. Only compositionality-robust (| ρ |> 0.7) and statistically significant ( q < 0.01) correlations were integrated into the hierarchical network analysis. The node-level topological features including the degree, betweenness, closeness, and eigenvectors were calculated using the “igraph” R package . The identification of keystone taxa was based on the criteria of the nodes with high degree, higher node transitivity, and low betweenness centralities. Finally, the generated co-occurrence network with keystone taxa and different modules was visualized using the Gephi platform (v0.92, https://gephi.org).
We then implemented partial least squares discriminant analysis (PLS-DA) and orthogonal partial least squares discriminant analysis (OPLS-DA) by using the “opls” function in the “ropls” bioconductor package to uncover the mechanisms that underlie the microbial response to plant domestication. Subsequently, ecological associations among plant features, significant metabolites, and enriched microbiota were calculated using the “rcorr” function in the “Hmisc” R package . Correlation matrix visualization was performed using the “ggtree” R package. Based on this association analysis, key relevant metabolites that were significantly enriched in wild or domesticated wheat were further used to evaluate their importance for plant growth. The importance of each metabolite was determined by evaluating the increase in the mean square error (MSE) between predictions and observations. This analysis was performed using the “rfPermute” and “rp. importance” functions in the “rfPermute” R package . Meanwhile, the significance of the importance of each metabolite associated with plant traits was assessed by using the “rfUtilities” R package The significance of this model was cross-validated using the “A3” R package .
Distinct diversity patterns of soil microbial communities from wild and domesticated wheat For all wheat genotypes, the bacterial community from domesticated wheat presented higher Shannon diversity than that from wild wheat, while the fungal community displayed the opposite pattern (Fig. S and Table S ). The multivariate linear regression analysis suggested that plant domestication status had a greater selection influence on both bacterial ( F = 8.388, P = 0.005) and fungal Shannon–Wiener diversities ( F = 7.501, P = 0.008) among different genotypes in the rhizosphere soils. For bulk soils, the analysis revealed that plant domestication status was also the main factor influencing α-diversity of both bacterial ( P = 0.0029) and fungal ( P = 0.0013) communities. Apart from affected by plant domestication and niche compartment, the genotype of host speciation could have potentially broader influences on bacterial ( R 2 = 17.2%, P < 0.001) and fungal communities ( R 2 = 12.8%, P < 0.001; Table S ). Considering β -diversity, bacterial and fungal communities displayed a marked dissimilarities in wild wheat compared with domesticated wheat on the basis of Bray–Curtis matrix (Fig. S ). Analogously, this distinct separation was also found in different niche compartments, such as bulk and rhizosphere soils (Table S ). Notably, CAP results further revealed a marked separation between the soil microbial communities of wild and domesticated wheat for both bacteria and fungi ( P < 0.001; Figs. S and S ). Based on these results, domestication exerted stronger selection influences on the rhizosphere bacterial and fungal community assembly, although other processes, such as niche compartment and plant genotypes, were also significant for the microbial community variations. Domestication altered microbial taxonomic patterns and interkingdom co-occurrences Taxonomic classification showed that both bacterial and fungal communities in the rhizosphere and bulk soils varied distinctly across wild and domesticated wheats (Fig. S ). Specifically, the differential abundance analysis indicated that 326 bacterial and 191 fungal ASVs from the rhizosphere were significantly affected by plant domestication status (Fig. A and B). Random forest modeling analysis further revealed that the top 30 enriched bacterial ASVs, mainly from the phyla Proteobacteria, Bacteroidetes, and Actinobacteria, were the most important predictors of plant domestication status from the rhizosphere (Fig. C). For the fungal community, the top 30 discriminant ASVs consisted of Ascomycota and Basidiomycota (Fig. D). The opposite enrichment patterns in the bulk soils were also observed between wild and domesticated wheat (Fig. S ). The marked differentially distributed ASVs showed asymmetric enrichment patterns in accordance with the plant domestication status and niche differentiation. The microbial co-occurrence network analysis revealed that bacterial-fungal interkingdom interaction patterns evolved clearly across wheat domestication, with discrepant bacterial and fungal roles in the wild and domesticated microbial network. Specifically, bacterial taxa from rhizospheric soils showed higher network connectivity than fungal taxa in the domesticated microbial network, while the pattern diverged greatly in the wild microbial network (Fig. A–D and Table S ). In addition, the network connectedness of bacteria to bacteria and bacteria to fungi were higher in the rhizospheric microbial network in domesticated wheat than in wild wheat, but network connectedness of fungi to fungi were higher in the wild network than in the domesticated network. Compared to the rhizosphere network, the opposite patterns were observed in the microbial network of bulk soils (Fig. S and Table S ). Notably, within the microbial interkingdom association network in the bulk soils, we found that plant domestication reduced the network connectivity, accompanied by lower proportion of associations of bacteria-fungi and fungi-fungi communities. Dominant taxa in different network modules confirmed the discrepant influences of wheat domestication on the multitrophic network (Fig. S ). These results indicated that wheat domestication enhanced the bacteria-bacteria intrakingdom and bacteria-fungi interkingdom associations but decreased the fungi-fungi associations in the rhizosphere. We further defined potential “keystone taxa” based on the criteria of nodes with high values of degree, transitivity, and low betweenness centrality in the microbial interkingdom networks. It was evident that the keystone taxa of the rhizosphere microbial network in domesticated wheat were bacteria, whereas associations within the wild wheat microbial network were mostly clustered around the fungal network. (Fig. A and B). This observed contrast pattern suggests the distinct roles of fungi and bacteria in sustaining the microbial network of domesticated and wild wheat. Collectively, these results indicated that assemblage patterns and ecological associations of the microbial community in the rhizosphere of domesticated wheat were regulated by bacteria, whereas fungi were more important in the wild wheat. Domestication status affects the functional profiles of the rhizosphere microbiome Based on the metagenomics sequencing for rhizosphere samples from domesticated and wild wheats (Table S ), differential abundance analysis revealed that the number of significantly enriched KO functional categories in wild wheat was higher than that in domesticated wheat (Wilcox test, P < 0.05, Fig. S ). Notably, we found that “translation,” “replication and repair,” and “folding, sorting, and degradation” functional categories affiliated to “genetic information processing” pathway were only enriched in the wild wheat (Fig. A and Table S ). In the differential KO functional categories enriched in domesticated wheat, the functional gene (K03406) affiliated with methyl-accepting chemotaxis protein (MCP) was the most abundant KO functional category, followed by K00626 (atoB, acetyl-CoA C-acetyltransferase) and K03070 (secA, preprotein translocase subunit). These functional categories are involved in “two-component system,” “fatty acid degradation,” and “quorum-sensing” pathways (Fig. B). In contrast, KO functional categories involved in “nucleotide excision repair” and “purine metabolism” pathways were labelled as dominant genes in wild wheat, such as K03657 (pcrA, ATP-dependent DNA helicase), K03701 (urvA, excinuclease ABC subunit A), and K03046 (rpoC, DNA-directed RNA polymerase subunit beta). NMDS ordinations of the KEGG Orthology, CAZyome, and COG profiles indicated that the functional position of rhizosphere microbiome of the domesticated wheat significantly differed from that of wild wheat ( P < 0.001), and domestication status also had a significant influence ( R 2 = 26.2%, P = 0.015) on rhizosphere microbiome functions (Fig. S A). Notably, the domesticated wheat possessed higher microbiome functional diversity than that of wild wheat. Moreover, a whole bunch of functional genes associated with C, N, and P cycling were dominantly enriched in wild wheat. In terms of C cycling (Fig. S B), functional genes involved in carbon fixation (e.g., pccA and smtA ) were more abundant in wild wheat. Regarding N cycling (Fig. S A), nitrate reduction genes (e.g., narG and narH ) and nitrogen assimilation gene ( nasA and nasB ), and denitrification genes (e.g., nirS and nosZ ), were more abundant in wild wheat, while nitrogen fixation genes (e.g., nifH and nifK ) and organic nitrogen metabolism gene ( gdhA ) were more abundant in domesticated wheat. For P cycling (Fig. S B), wild wheat enriched more functional genes related to organic P mineralization (e.g., phnX and ugpQ ) and P transport system (e.g., phnC and phnE ), while functional genes affiliated with inorganic P solubilization ( gcd ) were more abundant in domesticated wheat. Additionally, we found that functional genes (i.e., K02314, K01687, and K04066) associated with C, N, and P cycles were important predictors of plant phenotype traits, including plant height, chlorophyll content, and ear length (Fig. S ). Root metabolite traits and their links with the microbial community and plant phenotype profiles Metabolomics analysis revealed that root exudates exhibited significant differences between wheat with different domestication statuses, reflecting dissimilar metabolic profiles in wild and domesticated wheat (Fig. S ). More importantly, metabolites such as oleamide, apiin, and xanthine were more abundant in the wild wheat, while metabolites such as malic acid, leucodopachrome, and postin were enriched in the domesticated wheat (Fig. S ). These results indicated that divergences in metabolites were highly correlated with plant domestication and that this activity led to a decline in metabolite diversity in plant breeding history. We further used a metabolite-wide association study (MWAS) to investigate the specific response of dominant microbial taxa to biomarker metabolites as well as their links with plant phenotype. For domesticated wheat, the top 25 dominant bacterial genera and fungal genera had more negative associations and positive correlations with enriched metabolites, respectively (Fig. ). However, these patterns were reserved in the wild wheat. In regard to the associations between metabolites and plant phenotype, random forest analysis revealed that these associations were robust and some metabolites involved in malic acid, corchorifatty acid, postin, apiin, and octadecanamide were important predictors of plant attributes (Fig. S ). Regarding the links between plant attributes and dominant taxa, plant height of domesticated wheat had more positive correlations with some dominant bacterial and fungal species (Fig. A and C), while the patterns were reserved in wild wheat (Fig. B and D). These results based on the MWAS analysis revealed that distinct domestication statuses corresponded to the divergent ecological associations between bacterial and fungal communities, indicating their contrasting substrate preferences. To further corroborate the links between plant attributes and microbial taxa based on the field data analysis, we performed a microbiota inoculation experiment to assess the impact of the bacterial culture or inoculum on plant growth (Fig. A; Table S ). The results showed that the inoculated rhizosphere microbiota significantly decreased root length but promoted root average diameter compared to the control treatment, regardless of plant domestication status (Fig. B). We further found that biodiversity of keystone phylotypes involved in bacterial genera Nocardioides ( R 2 = 0.529, P = 0.0012) and fungal genera Holtermanniella ( R 2 = 0.671, P = 1.1e-04) was significantly associated with the root length of domesticated wheat. For wild wheat, we also found that the biodiversity of Pedobacter ( R 2 = 0.656, P = 1.5e-04) and Microdochium ( R 2 = 0.746, P = 2.03e-05) was significantly correlated with root length (Fig. S ). Based on the results from rhizosphere microbiota inoculation experiment, we further verified Microbacterium mitrae significantly affected root growth of domesticated and wild wheat (Fig. C, D, and E). Regarding the root morphological traits, the inoculation of M. mitrae exerts inhibition root growth and reduction seedling fresh/dry weight (Fig. F). This inhibition effects can be eliminated with removing M. mitrae . Considering the ecological role of putative keystone taxa (e.g., M. mitrae ) in maintaining the root morphological traits, these results suggest that individual keystone taxa led to a consistent growth regulation of root regardless of wheat domestication status.
For all wheat genotypes, the bacterial community from domesticated wheat presented higher Shannon diversity than that from wild wheat, while the fungal community displayed the opposite pattern (Fig. S and Table S ). The multivariate linear regression analysis suggested that plant domestication status had a greater selection influence on both bacterial ( F = 8.388, P = 0.005) and fungal Shannon–Wiener diversities ( F = 7.501, P = 0.008) among different genotypes in the rhizosphere soils. For bulk soils, the analysis revealed that plant domestication status was also the main factor influencing α-diversity of both bacterial ( P = 0.0029) and fungal ( P = 0.0013) communities. Apart from affected by plant domestication and niche compartment, the genotype of host speciation could have potentially broader influences on bacterial ( R 2 = 17.2%, P < 0.001) and fungal communities ( R 2 = 12.8%, P < 0.001; Table S ). Considering β -diversity, bacterial and fungal communities displayed a marked dissimilarities in wild wheat compared with domesticated wheat on the basis of Bray–Curtis matrix (Fig. S ). Analogously, this distinct separation was also found in different niche compartments, such as bulk and rhizosphere soils (Table S ). Notably, CAP results further revealed a marked separation between the soil microbial communities of wild and domesticated wheat for both bacteria and fungi ( P < 0.001; Figs. S and S ). Based on these results, domestication exerted stronger selection influences on the rhizosphere bacterial and fungal community assembly, although other processes, such as niche compartment and plant genotypes, were also significant for the microbial community variations.
Taxonomic classification showed that both bacterial and fungal communities in the rhizosphere and bulk soils varied distinctly across wild and domesticated wheats (Fig. S ). Specifically, the differential abundance analysis indicated that 326 bacterial and 191 fungal ASVs from the rhizosphere were significantly affected by plant domestication status (Fig. A and B). Random forest modeling analysis further revealed that the top 30 enriched bacterial ASVs, mainly from the phyla Proteobacteria, Bacteroidetes, and Actinobacteria, were the most important predictors of plant domestication status from the rhizosphere (Fig. C). For the fungal community, the top 30 discriminant ASVs consisted of Ascomycota and Basidiomycota (Fig. D). The opposite enrichment patterns in the bulk soils were also observed between wild and domesticated wheat (Fig. S ). The marked differentially distributed ASVs showed asymmetric enrichment patterns in accordance with the plant domestication status and niche differentiation. The microbial co-occurrence network analysis revealed that bacterial-fungal interkingdom interaction patterns evolved clearly across wheat domestication, with discrepant bacterial and fungal roles in the wild and domesticated microbial network. Specifically, bacterial taxa from rhizospheric soils showed higher network connectivity than fungal taxa in the domesticated microbial network, while the pattern diverged greatly in the wild microbial network (Fig. A–D and Table S ). In addition, the network connectedness of bacteria to bacteria and bacteria to fungi were higher in the rhizospheric microbial network in domesticated wheat than in wild wheat, but network connectedness of fungi to fungi were higher in the wild network than in the domesticated network. Compared to the rhizosphere network, the opposite patterns were observed in the microbial network of bulk soils (Fig. S and Table S ). Notably, within the microbial interkingdom association network in the bulk soils, we found that plant domestication reduced the network connectivity, accompanied by lower proportion of associations of bacteria-fungi and fungi-fungi communities. Dominant taxa in different network modules confirmed the discrepant influences of wheat domestication on the multitrophic network (Fig. S ). These results indicated that wheat domestication enhanced the bacteria-bacteria intrakingdom and bacteria-fungi interkingdom associations but decreased the fungi-fungi associations in the rhizosphere. We further defined potential “keystone taxa” based on the criteria of nodes with high values of degree, transitivity, and low betweenness centrality in the microbial interkingdom networks. It was evident that the keystone taxa of the rhizosphere microbial network in domesticated wheat were bacteria, whereas associations within the wild wheat microbial network were mostly clustered around the fungal network. (Fig. A and B). This observed contrast pattern suggests the distinct roles of fungi and bacteria in sustaining the microbial network of domesticated and wild wheat. Collectively, these results indicated that assemblage patterns and ecological associations of the microbial community in the rhizosphere of domesticated wheat were regulated by bacteria, whereas fungi were more important in the wild wheat.
Based on the metagenomics sequencing for rhizosphere samples from domesticated and wild wheats (Table S ), differential abundance analysis revealed that the number of significantly enriched KO functional categories in wild wheat was higher than that in domesticated wheat (Wilcox test, P < 0.05, Fig. S ). Notably, we found that “translation,” “replication and repair,” and “folding, sorting, and degradation” functional categories affiliated to “genetic information processing” pathway were only enriched in the wild wheat (Fig. A and Table S ). In the differential KO functional categories enriched in domesticated wheat, the functional gene (K03406) affiliated with methyl-accepting chemotaxis protein (MCP) was the most abundant KO functional category, followed by K00626 (atoB, acetyl-CoA C-acetyltransferase) and K03070 (secA, preprotein translocase subunit). These functional categories are involved in “two-component system,” “fatty acid degradation,” and “quorum-sensing” pathways (Fig. B). In contrast, KO functional categories involved in “nucleotide excision repair” and “purine metabolism” pathways were labelled as dominant genes in wild wheat, such as K03657 (pcrA, ATP-dependent DNA helicase), K03701 (urvA, excinuclease ABC subunit A), and K03046 (rpoC, DNA-directed RNA polymerase subunit beta). NMDS ordinations of the KEGG Orthology, CAZyome, and COG profiles indicated that the functional position of rhizosphere microbiome of the domesticated wheat significantly differed from that of wild wheat ( P < 0.001), and domestication status also had a significant influence ( R 2 = 26.2%, P = 0.015) on rhizosphere microbiome functions (Fig. S A). Notably, the domesticated wheat possessed higher microbiome functional diversity than that of wild wheat. Moreover, a whole bunch of functional genes associated with C, N, and P cycling were dominantly enriched in wild wheat. In terms of C cycling (Fig. S B), functional genes involved in carbon fixation (e.g., pccA and smtA ) were more abundant in wild wheat. Regarding N cycling (Fig. S A), nitrate reduction genes (e.g., narG and narH ) and nitrogen assimilation gene ( nasA and nasB ), and denitrification genes (e.g., nirS and nosZ ), were more abundant in wild wheat, while nitrogen fixation genes (e.g., nifH and nifK ) and organic nitrogen metabolism gene ( gdhA ) were more abundant in domesticated wheat. For P cycling (Fig. S B), wild wheat enriched more functional genes related to organic P mineralization (e.g., phnX and ugpQ ) and P transport system (e.g., phnC and phnE ), while functional genes affiliated with inorganic P solubilization ( gcd ) were more abundant in domesticated wheat. Additionally, we found that functional genes (i.e., K02314, K01687, and K04066) associated with C, N, and P cycles were important predictors of plant phenotype traits, including plant height, chlorophyll content, and ear length (Fig. S ).
Metabolomics analysis revealed that root exudates exhibited significant differences between wheat with different domestication statuses, reflecting dissimilar metabolic profiles in wild and domesticated wheat (Fig. S ). More importantly, metabolites such as oleamide, apiin, and xanthine were more abundant in the wild wheat, while metabolites such as malic acid, leucodopachrome, and postin were enriched in the domesticated wheat (Fig. S ). These results indicated that divergences in metabolites were highly correlated with plant domestication and that this activity led to a decline in metabolite diversity in plant breeding history. We further used a metabolite-wide association study (MWAS) to investigate the specific response of dominant microbial taxa to biomarker metabolites as well as their links with plant phenotype. For domesticated wheat, the top 25 dominant bacterial genera and fungal genera had more negative associations and positive correlations with enriched metabolites, respectively (Fig. ). However, these patterns were reserved in the wild wheat. In regard to the associations between metabolites and plant phenotype, random forest analysis revealed that these associations were robust and some metabolites involved in malic acid, corchorifatty acid, postin, apiin, and octadecanamide were important predictors of plant attributes (Fig. S ). Regarding the links between plant attributes and dominant taxa, plant height of domesticated wheat had more positive correlations with some dominant bacterial and fungal species (Fig. A and C), while the patterns were reserved in wild wheat (Fig. B and D). These results based on the MWAS analysis revealed that distinct domestication statuses corresponded to the divergent ecological associations between bacterial and fungal communities, indicating their contrasting substrate preferences. To further corroborate the links between plant attributes and microbial taxa based on the field data analysis, we performed a microbiota inoculation experiment to assess the impact of the bacterial culture or inoculum on plant growth (Fig. A; Table S ). The results showed that the inoculated rhizosphere microbiota significantly decreased root length but promoted root average diameter compared to the control treatment, regardless of plant domestication status (Fig. B). We further found that biodiversity of keystone phylotypes involved in bacterial genera Nocardioides ( R 2 = 0.529, P = 0.0012) and fungal genera Holtermanniella ( R 2 = 0.671, P = 1.1e-04) was significantly associated with the root length of domesticated wheat. For wild wheat, we also found that the biodiversity of Pedobacter ( R 2 = 0.656, P = 1.5e-04) and Microdochium ( R 2 = 0.746, P = 2.03e-05) was significantly correlated with root length (Fig. S ). Based on the results from rhizosphere microbiota inoculation experiment, we further verified Microbacterium mitrae significantly affected root growth of domesticated and wild wheat (Fig. C, D, and E). Regarding the root morphological traits, the inoculation of M. mitrae exerts inhibition root growth and reduction seedling fresh/dry weight (Fig. F). This inhibition effects can be eliminated with removing M. mitrae . Considering the ecological role of putative keystone taxa (e.g., M. mitrae ) in maintaining the root morphological traits, these results suggest that individual keystone taxa led to a consistent growth regulation of root regardless of wheat domestication status.
Plant domestication strongly influences the rhizosphere microbial communities and simultaneously decreases the functional diversity related to multiple nutrient cycles Deciphering how plant domestication affects rhizosphere microbiome assembly and functions is of great importance to reinforce the plant-microbiome interactions under the framework of “going back to the roots” . Our results demonstrated that changes in the wheat rhizosphere microbiome were largely influenced by plant domestication status. Moreover, rhizosphere microbiomes were more sensitive to plant domestication than soil microbiomes. This finding is in line with previous studies conducted with Arabidopsis , barley , rice , and soybean , indicating that plant domestication is a crucial factor in shaping the assembly of the plant rhizosphere microbiome. Metagenomic analyses in our study revealed that functional genes, which encode MCPs associated with “signal transduction” pathway, were enriched in the domesticated cultivars. MCPs have been found in plant-growth-promoting rhizobacteria such as the genera of Pseudomonas , which were also markedly enriched in the rhizosphere of wheat in our study. The beneficial bacteria would use MCPs to detect specific concentrations of molecules in the extracellular matrix, enabling directional accumulation of the bacteria to the plant, and in turn protect the plant against agricultural activity stresses. The findings suggest that MCP gene enrichment in domesticated plants is likely associated with the accumulation of beneficial bacteria. These harbored bacteria can deploy plant-release signal molecules in the extracellular matrix, such as volatile organic compounds, oxalic acid, trehalose, glucose, or thiamine , triggering detected changes between rhizospheric bacteria and fungi and promoting interkingdom associations via host-to-microbe, microbe-to-microbe, and microbe-to-host interactions . Furthermore, some functional genes involved in the C, N, and P cycles were the best predictors for the plant phenotype, suggesting that high gene diversity in the rhizosphere microbiome of wild wheat could ensure that host plants have a better phenotype with multiple nutrient cycles. The potential explanation is attributed to the highly functional diversity of microbial communities, which tend to possess greater functional redundancy and more complex interkingdom interactions [ , , ]. In order to adapt the environmental perturbations and selection pressure of anthropological activities, domesticated plants exhibit a trade-off between high yields and the gene diversity of belowground microbial communities . Thus, it would not be surprising to find that plant domestication decreased both microbial and functional diversity. Collectively, these findings suggest that plant domestication has significant implications for soil biogeochemical processes, including carbon fixation, nitrate reduction, and organic phosphate mineralization. Plant domestication shifts rhizosphere microbial communities from slow growing and fungi dominated to fast growing and bacteria dominated Bacteria and fungi likely coevolved and interacted in soils prior to plants colonizing terrestrial ecosystem ~ 450 mya . In this study, the increasing enrichment of the bacteria/fungi ratio suggests that rhizosphere bacteria became the dominant microorganisms in the domesticated wheat, while fungi played a more important ecological role in the wild wheat. This phenomenon is likely attributed to fungi favoring the habitat niches that maintained by the wild wheats and their associated root exudates, even though in the fertile conditions. Additionally, plant domestication was also accompanied by considerable habitat specialization and management practices, with gradually increased reliance on the inputs of agricultural activities (e.g., fertilizers) to acquire higher crop productivity and to mitigate abiotic and biotic stresses on the domesticated crops . The transition of plants from native habitats with low input activities to highly managed agricultural soil also results in substantial changes in microbial community and has further hampered beneficial interactions between plant and microbiome. Network analysis provide comprehensive insights into the stability of microbial communities under environmental stresses, although this approach has some limitations . In this study, domestication decreased the stability of microbial networks. The less stable interkingdom interactions in the rhizosphere microbial community of wheat are likely associated with long-lasting legacies from agricultural management practices , ecological competition at the phylogenetic level , and lower modularity . Consistent with a previous study , the enriched keystone bacterial taxa in the present study mostly belonged to bacterial community, whereas keystone fungal taxa were associated with fungal community. This is likely due to the bacteria and fungi adopt distinct life history strategies to confront natural and anthropogenic stress during plant domestication . Since microbial carbon fixation and carbon degradation processes are regulated by autotrophic and heterotrophic microorganisms, respectively, our results suggest that domestication shifts rhizosphere microbial communities from slow-growing autotrophic microorganisms to fast-growing decomposers or carbon utilizers. Additionally, we found that carbon fixation and degradation genes were both significantly associated with multiple edaphic variables (Table S and Fig. S ). Consistent with pioneer studies , our findings reinforced that soil nutrients, particularly nitrogen, may greatly contribute to carbon flow at the soil-root interface. More importantly, our present study indirectly verified these shifts partly due to the release of distinct root exudates from domesticated and wild wheat. Pioneer studies highlighted how root exudates affect evolution of the crop rhizosphere , indicating that plant hosts can modulate nutrient conditions and energy flow to control an ever-evolving microbial community in the rhizosphere via plants actively release exudates. The root exudates could attract and selectively enrich for specific metabolic functions of microbial species. Collectively, these findings suggest that domestication shifts rhizosphere microbial communities from fungi dominated and slow growing to bacteria dominated and fast growing, thereby resulting in a shift from fungi-dominated membership with enrichment of C fixation genes to bacteria-dominated membership with enrichment of C-degradation genes. This finding supported the notion that bacteria and fungi are relatively more important game-changers in carbon utilization and carbon sink, respectively . Contrasting functional roles of root exudates in shaping rhizosphere microbiome and determining root morphology traits Deciphering the ecological role of root exudates will advance our understanding of belowground economic traits and provide valuable insights into the manipulation of the rhizosphere microbiome for sustainable agriculture . Our results in the present study indicated that plant domestication status determined the type of root exudate release. These root exudates exert selection pressures on rhizosphere microbiome that lead to differential effects on host performance and root morphological traits via host-to-microbe, microbe-to-microbe, and microbe-to-host feedback . However, these positive or negative associations between root exudates and root morphology traits do not hold invariably . The underlying mechanisms by which contrasting roles of plant-derived metabolites shape the assembly of rhizosphere microbiome currently include the following: (i) modulation of root morphology (number and length of roots, root hairs, root diameter, and root surface), which directly determined whether the rhizosphere microbiome is successful in colonization on the root surface ; (ii) regulation of microbial intrakingdom or interkingdom interactions, which determined the complexity and stability of rhizosphere microbiome [ , , ]; and (ii) formation of nutritional interdependencies, which determined the reciprocal exchange of metabolites between metabolic deficiency microbiome and molecular compounds from root exudates . Given plant-soil feedback encompasses host-to-microbiome, microbiome-to-microbiome, and microbiome-to-host interactions , the host plants and rhizosphere microbiome are not an independent evolutionary unit but acting with a common interest, particularly for nutrient resource availability in agricultural intensive environment, which in turn modulates root trait variation via divergent belowground resource-acquisition strategies, such as the nutrient acquisition strategies, and includes nutrient-scavenging and nutrient-mining strategies . Pioneering studies contend that root exudates are a colossal of the carbon pool . From a cost–benefit perspective, the nutrient-scavenging pathways are an energy consumption process that requires significant carbon investment from increased root exudate release. The benefits are numerous, but plants may reallocate the release of root exudates to trade off against root growth and nutrient acquisition. Therefore, it would not be surprising to find that domesticated and wild wheat exhibit substantial variation in root exudation and root morphology. Apart from nutrient-scavenging strategies, root exudate-induced nutrient-mining strategies recruit beneficial microbes to the rhizosphere, which enhance nutrient availability and plant growth. Previous studies contended that P-solubilizing microbes (PSMs) have the capacity to solubilize P from recalcitrant forms of inorganic and organic P and thus enhance plant P acquisition and growth . Metagenomics analysis in our study further corroborated that rhizosphere microbiomes from domesticated wheat enriched more abundant inorganic P solubilization genes (e.g., ppx and ppa ) via releasing of root exudates (Fig. S ). More importantly, N-cycling microbial-dominated mobilizing processes play a pivotal role in plant N acquisition and growth . Metagenomics analyses based on nitrogen metabolism further support that domesticated plants selectively recruit rhizosphere keystone taxa that carry nitrogen fixation genes (e.g., nifH ) and organic nitrogen ammonification genes (e.g., gdhA ) to sustain plant N uptake and plant growth through the attraction of root exudates (Fig. S ). These plant-available N through microbial-dominated mobilizing processes can ultimately affect root morphological traits . In vitro studies on plant-microbiome interactions, such as crop (e.g., wheat, maize, and rice) and model plants (e.g., Arabidopsis), which have provided rapid and in-depth knowledge on how the development of the root morphology traits is regulated by single or culture-dependent synthetic communities . To definitively explore the role of the keystone taxa in determining of plant growth-promoting features and the impacts on root architecture, analysis of associations between nutrient-cycling-related genes and root morphology traits together with in vitro experiments further highlighted that a single bacterial genus or fungi could maintain plant root growth in a complex microbiome, which is in line with a recent study . Notably, we found that individual microbial consortium had inhibition effect on the root growth of wheat and thus led to poor plants. These findings suggest that although domestication-induced keystone taxa had crucial role in promoting plant growth, the ubiquitous of maleficent microbiota during plant domestication should be given more attention to develop correspondingly microbial inhibitors to sniping these harmful microbial consortiums for promoting plant growth. However, the mechanisms that how individual keystone taxa reprogramming root morphology traits needed to be explored using RNA sequencing in the further study, as well as the contribution of root exudation to colonization of individual microbial consortium at the developmental stage of wild and domesticated wheat, require more rigorous assessment in the next growing season. Collectively, this study provides a systematic understanding of the rhizosphere microbial communities and their metabolic functions during plant domestication processes (Fig. ). Our findings presented here demonstrate that domestication exerted stronger selection power on the rhizosphere community assembly, although niche compartment and plant genotypes were also significant for the microbial community variations. We further demonstrated that plant domestication leads to a decline in gene diversity and a shift in microbial functional traits, particularly for functional genes related with multi-nutrients cycling. Furthermore, we found that contrasting functional roles of root exudates shaped rhizosphere microbiome and determined root morphology traits. Taken together, these findings suggest that plant domestication exerts a strong and direct selective power on the rhizosphere microbiome assembly and functional adaptation through the release of root exudates with contrasting roles. The current study significantly underpins our understanding of the species coexistence and functional adaptation of the rhizosphere microbiomes of wheat during plant domestication and paves the way for new plant breeding strategies.
Deciphering how plant domestication affects rhizosphere microbiome assembly and functions is of great importance to reinforce the plant-microbiome interactions under the framework of “going back to the roots” . Our results demonstrated that changes in the wheat rhizosphere microbiome were largely influenced by plant domestication status. Moreover, rhizosphere microbiomes were more sensitive to plant domestication than soil microbiomes. This finding is in line with previous studies conducted with Arabidopsis , barley , rice , and soybean , indicating that plant domestication is a crucial factor in shaping the assembly of the plant rhizosphere microbiome. Metagenomic analyses in our study revealed that functional genes, which encode MCPs associated with “signal transduction” pathway, were enriched in the domesticated cultivars. MCPs have been found in plant-growth-promoting rhizobacteria such as the genera of Pseudomonas , which were also markedly enriched in the rhizosphere of wheat in our study. The beneficial bacteria would use MCPs to detect specific concentrations of molecules in the extracellular matrix, enabling directional accumulation of the bacteria to the plant, and in turn protect the plant against agricultural activity stresses. The findings suggest that MCP gene enrichment in domesticated plants is likely associated with the accumulation of beneficial bacteria. These harbored bacteria can deploy plant-release signal molecules in the extracellular matrix, such as volatile organic compounds, oxalic acid, trehalose, glucose, or thiamine , triggering detected changes between rhizospheric bacteria and fungi and promoting interkingdom associations via host-to-microbe, microbe-to-microbe, and microbe-to-host interactions . Furthermore, some functional genes involved in the C, N, and P cycles were the best predictors for the plant phenotype, suggesting that high gene diversity in the rhizosphere microbiome of wild wheat could ensure that host plants have a better phenotype with multiple nutrient cycles. The potential explanation is attributed to the highly functional diversity of microbial communities, which tend to possess greater functional redundancy and more complex interkingdom interactions [ , , ]. In order to adapt the environmental perturbations and selection pressure of anthropological activities, domesticated plants exhibit a trade-off between high yields and the gene diversity of belowground microbial communities . Thus, it would not be surprising to find that plant domestication decreased both microbial and functional diversity. Collectively, these findings suggest that plant domestication has significant implications for soil biogeochemical processes, including carbon fixation, nitrate reduction, and organic phosphate mineralization.
Bacteria and fungi likely coevolved and interacted in soils prior to plants colonizing terrestrial ecosystem ~ 450 mya . In this study, the increasing enrichment of the bacteria/fungi ratio suggests that rhizosphere bacteria became the dominant microorganisms in the domesticated wheat, while fungi played a more important ecological role in the wild wheat. This phenomenon is likely attributed to fungi favoring the habitat niches that maintained by the wild wheats and their associated root exudates, even though in the fertile conditions. Additionally, plant domestication was also accompanied by considerable habitat specialization and management practices, with gradually increased reliance on the inputs of agricultural activities (e.g., fertilizers) to acquire higher crop productivity and to mitigate abiotic and biotic stresses on the domesticated crops . The transition of plants from native habitats with low input activities to highly managed agricultural soil also results in substantial changes in microbial community and has further hampered beneficial interactions between plant and microbiome. Network analysis provide comprehensive insights into the stability of microbial communities under environmental stresses, although this approach has some limitations . In this study, domestication decreased the stability of microbial networks. The less stable interkingdom interactions in the rhizosphere microbial community of wheat are likely associated with long-lasting legacies from agricultural management practices , ecological competition at the phylogenetic level , and lower modularity . Consistent with a previous study , the enriched keystone bacterial taxa in the present study mostly belonged to bacterial community, whereas keystone fungal taxa were associated with fungal community. This is likely due to the bacteria and fungi adopt distinct life history strategies to confront natural and anthropogenic stress during plant domestication . Since microbial carbon fixation and carbon degradation processes are regulated by autotrophic and heterotrophic microorganisms, respectively, our results suggest that domestication shifts rhizosphere microbial communities from slow-growing autotrophic microorganisms to fast-growing decomposers or carbon utilizers. Additionally, we found that carbon fixation and degradation genes were both significantly associated with multiple edaphic variables (Table S and Fig. S ). Consistent with pioneer studies , our findings reinforced that soil nutrients, particularly nitrogen, may greatly contribute to carbon flow at the soil-root interface. More importantly, our present study indirectly verified these shifts partly due to the release of distinct root exudates from domesticated and wild wheat. Pioneer studies highlighted how root exudates affect evolution of the crop rhizosphere , indicating that plant hosts can modulate nutrient conditions and energy flow to control an ever-evolving microbial community in the rhizosphere via plants actively release exudates. The root exudates could attract and selectively enrich for specific metabolic functions of microbial species. Collectively, these findings suggest that domestication shifts rhizosphere microbial communities from fungi dominated and slow growing to bacteria dominated and fast growing, thereby resulting in a shift from fungi-dominated membership with enrichment of C fixation genes to bacteria-dominated membership with enrichment of C-degradation genes. This finding supported the notion that bacteria and fungi are relatively more important game-changers in carbon utilization and carbon sink, respectively .
Deciphering the ecological role of root exudates will advance our understanding of belowground economic traits and provide valuable insights into the manipulation of the rhizosphere microbiome for sustainable agriculture . Our results in the present study indicated that plant domestication status determined the type of root exudate release. These root exudates exert selection pressures on rhizosphere microbiome that lead to differential effects on host performance and root morphological traits via host-to-microbe, microbe-to-microbe, and microbe-to-host feedback . However, these positive or negative associations between root exudates and root morphology traits do not hold invariably . The underlying mechanisms by which contrasting roles of plant-derived metabolites shape the assembly of rhizosphere microbiome currently include the following: (i) modulation of root morphology (number and length of roots, root hairs, root diameter, and root surface), which directly determined whether the rhizosphere microbiome is successful in colonization on the root surface ; (ii) regulation of microbial intrakingdom or interkingdom interactions, which determined the complexity and stability of rhizosphere microbiome [ , , ]; and (ii) formation of nutritional interdependencies, which determined the reciprocal exchange of metabolites between metabolic deficiency microbiome and molecular compounds from root exudates . Given plant-soil feedback encompasses host-to-microbiome, microbiome-to-microbiome, and microbiome-to-host interactions , the host plants and rhizosphere microbiome are not an independent evolutionary unit but acting with a common interest, particularly for nutrient resource availability in agricultural intensive environment, which in turn modulates root trait variation via divergent belowground resource-acquisition strategies, such as the nutrient acquisition strategies, and includes nutrient-scavenging and nutrient-mining strategies . Pioneering studies contend that root exudates are a colossal of the carbon pool . From a cost–benefit perspective, the nutrient-scavenging pathways are an energy consumption process that requires significant carbon investment from increased root exudate release. The benefits are numerous, but plants may reallocate the release of root exudates to trade off against root growth and nutrient acquisition. Therefore, it would not be surprising to find that domesticated and wild wheat exhibit substantial variation in root exudation and root morphology. Apart from nutrient-scavenging strategies, root exudate-induced nutrient-mining strategies recruit beneficial microbes to the rhizosphere, which enhance nutrient availability and plant growth. Previous studies contended that P-solubilizing microbes (PSMs) have the capacity to solubilize P from recalcitrant forms of inorganic and organic P and thus enhance plant P acquisition and growth . Metagenomics analysis in our study further corroborated that rhizosphere microbiomes from domesticated wheat enriched more abundant inorganic P solubilization genes (e.g., ppx and ppa ) via releasing of root exudates (Fig. S ). More importantly, N-cycling microbial-dominated mobilizing processes play a pivotal role in plant N acquisition and growth . Metagenomics analyses based on nitrogen metabolism further support that domesticated plants selectively recruit rhizosphere keystone taxa that carry nitrogen fixation genes (e.g., nifH ) and organic nitrogen ammonification genes (e.g., gdhA ) to sustain plant N uptake and plant growth through the attraction of root exudates (Fig. S ). These plant-available N through microbial-dominated mobilizing processes can ultimately affect root morphological traits . In vitro studies on plant-microbiome interactions, such as crop (e.g., wheat, maize, and rice) and model plants (e.g., Arabidopsis), which have provided rapid and in-depth knowledge on how the development of the root morphology traits is regulated by single or culture-dependent synthetic communities . To definitively explore the role of the keystone taxa in determining of plant growth-promoting features and the impacts on root architecture, analysis of associations between nutrient-cycling-related genes and root morphology traits together with in vitro experiments further highlighted that a single bacterial genus or fungi could maintain plant root growth in a complex microbiome, which is in line with a recent study . Notably, we found that individual microbial consortium had inhibition effect on the root growth of wheat and thus led to poor plants. These findings suggest that although domestication-induced keystone taxa had crucial role in promoting plant growth, the ubiquitous of maleficent microbiota during plant domestication should be given more attention to develop correspondingly microbial inhibitors to sniping these harmful microbial consortiums for promoting plant growth. However, the mechanisms that how individual keystone taxa reprogramming root morphology traits needed to be explored using RNA sequencing in the further study, as well as the contribution of root exudation to colonization of individual microbial consortium at the developmental stage of wild and domesticated wheat, require more rigorous assessment in the next growing season. Collectively, this study provides a systematic understanding of the rhizosphere microbial communities and their metabolic functions during plant domestication processes (Fig. ). Our findings presented here demonstrate that domestication exerted stronger selection power on the rhizosphere community assembly, although niche compartment and plant genotypes were also significant for the microbial community variations. We further demonstrated that plant domestication leads to a decline in gene diversity and a shift in microbial functional traits, particularly for functional genes related with multi-nutrients cycling. Furthermore, we found that contrasting functional roles of root exudates shaped rhizosphere microbiome and determined root morphology traits. Taken together, these findings suggest that plant domestication exerts a strong and direct selective power on the rhizosphere microbiome assembly and functional adaptation through the release of root exudates with contrasting roles. The current study significantly underpins our understanding of the species coexistence and functional adaptation of the rhizosphere microbiomes of wheat during plant domestication and paves the way for new plant breeding strategies.
Additional file 1: Fig. S1. Alpha diversity included Shannon index, Simpson index, ACE index, and Chao index for bacteria and fungi communities in different accession of domesticated and wild wheats. Errors bars represent standard errors ( n = 5). Different lowcase letters above the bars indicate significant differences ( P < 0.05), based on Kruskal-Wallis test. Fig. S2. The Bray-Curtis dissimilarity of bacteria and fungi communities in domesticated and wild wheats. DB1~DB3 and WB1-WB3 indicates that samples are affiliated with bulk soil in the domesticated and wild wheats, respectively. DT1~DT3 and WT1-WT3 indicates that samples are affiliated with rhizosphere soil in the domesticated and wild wheats, respectively. The dots represent the values of Bray-Curtis dissimilarity between replicates in the any of groups. The details on the D1-D3 and W1-W3 groups are available in Table S . Fig. S3. Constrained Analysis of Principal Coordinates analysis (CAP) ordination constraned to domestication status (left panel), habitat type (middle panel), and genome group (right panel) based on Bray-Curtis metric of bacteria and fungi communities. Variance of community dissimilarity among five treatment were draw from ANOVA-like permutation analysis. Fig. S4. Principal coordinate analysis ordinations (PCoA) for bacteria and fungi communities in domesticated and wild wheats. Fig. S5. Distribution of dominant phyla and genus in the bacteria and fungi communities in domesticated and wild wheats. Fig. S6. Amplicon sequences variants (ASVs) form bulk soils responsible for the community differences in the wild wheats and domesticated wheats that are calculated by a differential abundance test and random forest classification. (A) and (B) The volcano plot illustrating the enrichment and depletion patterns of rhizosphere bacterial and fungal microbiomes in the three wild wheats compared with three domesticated wheat accessions. DI, depleted index; DSI, dissimilarity index. (C) and (D) Joyplots showing the relative abundance profiles of top 20 ASVs in bacterial and fungal communities that are revealed by a Random forest (RF) classifier. The top ASVs on the genus level are listed along the y-axis represent their importance in contributing to the accuracy of domesticated and wild wheats prediction by calculating their mean decrease accuracy in the RF model. Fig. S7. Microbial interkingdom association networks and node-level topological features for bulk soils. (A) and (B) Interkingdom co-occurrence networks in the domesticated and wild wheats. Only compositionality-robust (p > 0.7) and statistically significant ( q < 0.01) correlations were shown. The size of each node indicates the relative abundance of each ASV. The color of each node represents the bacteria or fungi taxa. Blue solid lines represent co-presence associations and red line represent mutual exclusive correlations. The thickness of each link line is proportional to the correlation coefficients of the connections. The keystone taxa and dominant modules for each networks were also shown. (C) and (D) Box graphs illustrating the node-level topological features of each networks, including betweenness and degree. Comparison of these two features demonstrating the high degree and low betweenness for the keystone taxa. Bar diagrams showing the proportion of inter-and intra-kingdom edges of positive or negative correlations in the rhizosphere network. The significance of differences between domesticated and wild wheats were determined by Kruskal-Wallis test. Fig. S8. Network modularity profiles of rhizosphere soil and bulk soil microbial communities in the domesticated and wild wheats. The eight bar plots in the left and middle columns represents microbial taxa on the class level in the module I and module II. The four bar plots in the right columns indicates the proportion of ASVs that affiliated with bacteria and fungi communities in the module I and module II. Fig. S9. The volcano plot illustrating the enrichment and depletion patterns of KO functional categories and KO pathway in the wild wheats compared with domesticated wheat accessions. The KOs were colored by their categorization as “wild-enriched”, “domesticated-enriched”, and “non-differential” according to their values of Log2(count per million) and Log2(fold change). Fig. S10. (A) Nonmetric multidimensional scaling (NMDS) ordination of KEGG Orthology, CAZyome, and COG based on Bray-Curtis distances. The significance of differences between domesticated and wild wheats were determined by Kruskal-Wallis test. (B) Heat map illustrating the relative abundance (Z-score) of functional genes (based on KO) affiliated with carbon cycling in different accession of wheats. Fig. S11. Heat map illustrating the relative abundance (Z-score) of functional genes (based on KO) affiliated with nitrogen (A) and phosphorus (B) cycling in different accession of wheats. Fig. S12. (A-H) Random forest (RF) mean predictor importance of KO functional categories as drivers for the plant phenotypes, including (A) plant height (PH), (B) chlorophyll content (CC), (C) ear length (EL), (D) subsegment length (SSL), (E) Tiller, (F) thousand grain weight (TGW), (G) seed length (SL), and (H) seed width (SW), respectively. The accuracy importance measure was calculated for each tree and averaged over the forest (2000 trees). Percentage increase in the mean squared error (MSE) of variables were applied to evaluate the importance of these predictors, and higher MSE% values represent more important predictors. Significance levels of each predictor are as follows: * P < 0.05, ** P < 0.01, and *** P < 0.001. Fig. S13. Partial least squares discriminant analysis (PLS-DA) and orthogonal partial least squares discriminant analysis (OPLS-DA) for root exudates that extracted from roots of wild and domesticated wheats. Fig. S14. (A) The average relative abundance of enriched root exudates in the wild and domesticated wheats. (B) The fold changes of enriched metabolites in domesticated and wild wheats. The error bars represent standard errors of sample replicates and asterisks (*) indicate metabolic categories that are significantly more predominant in wild or domesticated wheats (P value <0.05, Wilcoxon test). (C) Heat map illustrating the relative abundance of root exudates in different accession of wheats. Fig. S15. (A-H) Random forest (RF) mean predictor importance of enriched metabolites as drivers for the plant phenotypes, including (A) plant height (PH), (B) chlorophyll content (CC), (C) ear length (EL), (D) subsegment length (SSL), (E) Tiller, (F) thousand grain weight (TGW), (G) seed length (SL), and (H) seed width (SW), respectively. The accuracy importance measure was calculated for each tree and averaged over the forest (2000 trees). Percentage increase in the mean squared error (MSE) of variables were applied to evaluate the importance of these predictors, and higher MSE% values represent more important predictors. Significance levels of each predictor are as follows: * P < 0.05, ** P < 0.01, and *** P < 0.001. Fig. S16. (A-D) Ecological associations between root morphology and rhizosphere bacteria and fungi in domesticated wheat and wild wheats. Root morphology traits including root volume (RV), root average diameter (RAD), stem thickness (ST), root length (RL), fresh weight of root (FWR), were scanned and registered using Microtek ScanMaker i800 plus system. Significance levels of each association are as follows: * P < 0.05, ** P < 0.01, and *** P < 0.001. Fig. S17. (A) Pearson correlation relationships between edaphic variables and carbon-cycling-related functional gene. (B) Pearson correlation relationships between dominant metabolites and carbon-cycling-related functional gene. Significance levels of each association are as follows: * P < 0.05, ** P < 0.01, and *** P < 0.001. Fig. S18. Pearson correlation relationships between dominant metabolites and phosphorus-cycling-related functional gene. Significance levels of each association are as follows: * P < 0.05, ** P < 0.01, and *** P < 0.001. Fig. S19. Pearson correlation relationship between dominant metabolites and nitrogen-cycling-related functional gene. Significance levels of each association are as follows: * P < 0.05, ** P < 0.01, and *** P < 0.001. Additional file 2: Table S1. Plant phenotypes of 44 Tetraploid wheat species after nine years cultivation. Table S2. Plant phenotypes of six selected Tetraploid wheat species, involving T. turgidum Espana (labelled “D1”), T. turgidum Mexicanos (labelled “D2”), T. turgidum Kazakhstan (labelled “D3”), T. turgidum Turkey III (labelled “W1”), T. turgidum Syria VI (labelled “W2”), and T. turgidum Jordan III (labelled “W3”). Five replicates of each wheat accessions were randomly chosen to measure the plant attributes, including plant height, ear length, sub-segment length, and tiller. Table S3. Effect of domestication status, genome type and niche compartment on the bacterial and fungal communities based on PERMANOVA analysis. Table S4. Analysis of ANOSIM testing microbial communities from rhizosphere and bulk soils based on the Bray-Curtis across different treatments. Significance levels of each predictor are as follows: * P < 0.05, ** P < 0.01, and *** P < 0.001. Table S5. Topological features of microbial interkingdom association networks in different domesticated status and their corresponding random networks. Table S6. Characteristics of metagenomics sequencing of rhizosphere microbiomes in wild and domesticated wheats. Table S7. Relative abundance of the metagenomics microbial function profiling (KEGG Orthology function category) in wild and domesticated wheats. Table S8. Root morphology trait of wild and domesticated wheat grown in Hogland nutrient solution and inoculation with rhizosphere microbiota. Table S9. Characteristics of soil samples in different plots for domesticated and whild wheats. Soil nutrients include total carbon (TC), total nitrogen (TN), total phosphorus (TP), total organic carbon (TOC), dissolved organic carbon (DOC), dissolved organic nitrogen (DON), nitrate (NO3-), ammonium (NH4+), soil available phosphorus (AP), soil available potassium (AK). Additional file 3: Method S1. Experiment design and field site management. Method S2. Samples collection, DNA extraction, PCR amplification and amplicon sequencing. Method S3. Shotgun metagenomics sequencing. Method S4. Root exudate collection and LC-MS analysis. Method S5. Microbiota inoculation experiment. Method S6. Statistical analyses.
|
Including sexual orientation and gender identity data to advance nephrology care | 3b8425dd-d4b9-4c5a-9e30-ae7e834b4db8 | 10064956 | Internal Medicine[mh] | |
Longitudinal evaluation of the Ophthalmology residents in Brazil: an
observational prospective study | 67abec9e-cb16-445f-ad77-c4ed651f5e47 | 10065116 | Ophthalmology[mh] | Knowledge assessment plays an important role in medical education since professional
expertise development appears to be strongly connected to knowledge. Research has shown that assessment may be used in different ways. For
example, studies have demonstrated that assessment drives and stimulates learning, , provides educational efficacy information to institutions and teachers, and
protects patients. The definitions of “to test” in the dictionary are as follows: to discover the worth
of something by trial, to obtain more information about the object of assessment,
and to improve the quality of something by trial. Thus, assessment in the broader sense involves testing, measuring,
collecting, combining information, and providing feedback. In many medical residency programs, modular, intermediate, or final tests have been
used to measure the knowledge level of trainees. , However, these types of tests are associated with the promotion of short-term memorization. In addition, residents' performance may not correspond to the real knowledge
level since it is merely a one-point measurement, not allowing any extrapolation to
the maintained knowledge level over time. To benefit students' long-term retention, longitudinal testing in the form of
the progress test, the most known and established kind of longitudinal test, has
been suggested. , Progress testing aims to measure students' knowledge at the end level and allows the
measurement of knowledge growth. , In addition, progress testing forces students to study over time, encouraging
more profound and deep learning since it is impossible for students to cram before the test. Alternatively,
students must acquire information continuously in such a way that it is available
when required. Progress tests allow for individual learning pathways, which may provide
clues for future performance. Finally, progress testing can be organized at a
national level and can be used to compare the results of candidates from different countries. Progress tests have been used in different ways, such as for providing feedback to students, , understanding knowledge growth on questions requiring lower and higher order
of cognitive processing, , comparing national and international curricula, and the effectiveness of educational strategies. , Many medical schools worldwide have already adopted this progress testing as
part of their curricula, such as the Netherlands, Canada, Germany, Indonesia, South Africa, the United States, and Brazil. , Despite being a well-established assessment tool in the undergraduate context,
progress testing is much less widespread in the postgraduate context, where the best
test format remains controversial. Some authors believe that, at least in theory, longitudinal tests would also
be an interesting approach to knowledge assessment in postgraduate medical education. So far, only a few residency programs have already included the progress test
in their curricula, such as in obstetrics and gynecology, radiology, and in general practice, , demonstrating promising results. The World Reference Institution in ophthalmology residency programs is the
International Council of Ophthalmology (ICO). According to the ICO, medical
knowledge is one of the general core competencies expected from ophthalmic
specialists (besides patient care, practice-based learning and improvement,
communication skills, professionalism, and systems-based practice). , Progress testing during residency could play an important role in monitoring the
competence progress. Besides, it could be useful for the quality control of
residency programs in Brazil, by allowing interventions during the course. In
addition, the tests can serve as self-learning tools for residents. Finally, it can
be useful to predict residents' results in the specialist test of the Brazilian
Opthalmology Council.
This study aimed to investigate residents' knowledge growth during their residency
training. This study also describes the implementation of a progress test in
ophthalmological residency training across several medical schools in Brazil.
Finally, this study aimed to investigate whether there was a correlation between the
performance of the progress test and the specialist title test.
This was a prospective observational study carried out through an online
platform. This study was approved by the ethics committee of Universidade Estadual de Campinas
on December 17, 2018 (CAAE number:02613718.9.0000.5404). Participants: The study was conducted in 2019. All participants were ophthalmology
residents who agreed to participate voluntarily in the study and signed a consent
form. Ophthalmology Residency in Brazil In Brazil, the ophthalmology residency consists of a 3 years program. The institution that represents the Brazilian Ophthalmology is the Brazilian
Ophthalmology Council (Conselho Brasileiro de Oftomologia, CBO). According to the CBO, the minimum pedagogic program required for the
ophthalmology specialization consists of the following content: Basic sciences: 100% in the 1 st year and 0% in the
2 nd and 3 rd year Propaedeutics: 60% in the 1 st year, 30% in the 2 nd ,
and 10% in the 3 rd year Optometry: 50% in the 1 st year, 50% in the 2 nd , and
0% in the 3 rd year Surgical techniques: 50% in the 1 st year, 50% in the
2 nd , and 0% in the 3 rd year Clinics and surgery: 25% in the 1 st year, 50% in the
2 nd , and 25% in the 3 rd year Besides this mandatory content, there may be complementary activities, such as
clinical case discussions, pathological anatomy sections, and scientific article discussions. Progress test construction and application The progress test consisted of 125 multiple-choice questions on clinical and
surgical issues in ophthalmology. The blueprint followed the same pattern as the
Brazilian Ophthalmology Council specialist test: uveitis: 9 questions; neuro ophthalmology: 7 questions; orbit: 4 questions; lacrimal system: 4 questions; ocular plastics: 8 questions; ocular tumors: 5 questions; cornea: 14 questions; contact lenses: 4 questions; refractive surgery: 2 questions; retina: 13 questions; cataract: 10 questions; glaucoma: 11 questions; refraction: 23 questions; strabismus: 7 questions; low vision: 4 questions. The shows the division of the test
questions. As the tests consisted of 125 multiple-choice questions, for the statistical
analysis, a 0.08 value corresponds to 10/125 for each correct answer; thus, it
was attributed to a score that could vary from 0 to 10 for each test. The questions in the tests were taken from the following books: Review Questions
in Opthalmology, Clinical Optics and Refraction, and Self-tests in Optic and Refraction. They were chosen according to the issue and level of difficulty (judged
by the authors), in a way that there were questions of different issues and
levels of difficulty. As there were residents from many parts of the country, the tests were online,
and all the residents from the 1 st to the 3 rd year of the
ophthalmology residency programs performed the tests simultaneously. Therefore,
all residents were enrolled in the same test, regardless of whether they were in
their 1 st , 2 nd , or 3 rd year of residency. The
tests were conducted once a year at the end of the school year. Each service organized the implementation of the tests, and the only requirement
was that all residents sat on the test simultaneously. Some services used their
own informatic lab rooms, while those that did not have one allowed their
residents to use their own computers, either at the service or at home, at a
predetermined schedule, as long as there was one computer for each resident. Site First, participants had to create an account. Once completed, they were able to
access the site. – show a small portion of the site. The presentation page contains some important advice to read and the consent term
that had to be signed before the test itself . The test page contained the test itself. Once completed, participants had to
submit their answers. Immediately after the submission, the participant received
feedback . The feedback page shows the number of correct and incorrect answers, the time
spent performing the test, and the score (Figure 3). shows the correct answers and
explanations. Data analysis Frequency tables were used for the descriptive analysis of categorical variables.
Positions and dispersion measures were used for numeric variables. The
Kruskal–Wallis test was used to compare the differences between years, followed
by Dunn's test to identify significant differences. The Friedman or Wilcoxon test was used to compare students' knowledge growth. To investigate the relationship between the progress test and CBO scores, the
Spearman linear correlation coefficient and Wilcoxon test for related samples
were conducted. A statistical level of 0.05 was considered significant. Data were analyzed using the Statistical Analysis Software (SAS) System for
Windows (Statistical Analysis System), version 9.4. SAS Institute Inc,
2002-2012, Cary, North Carolina, United States.
In Brazil, the ophthalmology residency consists of a 3 years program. The institution that represents the Brazilian Ophthalmology is the Brazilian
Ophthalmology Council (Conselho Brasileiro de Oftomologia, CBO). According to the CBO, the minimum pedagogic program required for the
ophthalmology specialization consists of the following content: Basic sciences: 100% in the 1 st year and 0% in the
2 nd and 3 rd year Propaedeutics: 60% in the 1 st year, 30% in the 2 nd ,
and 10% in the 3 rd year Optometry: 50% in the 1 st year, 50% in the 2 nd , and
0% in the 3 rd year Surgical techniques: 50% in the 1 st year, 50% in the
2 nd , and 0% in the 3 rd year Clinics and surgery: 25% in the 1 st year, 50% in the
2 nd , and 25% in the 3 rd year Besides this mandatory content, there may be complementary activities, such as
clinical case discussions, pathological anatomy sections, and scientific article discussions.
The progress test consisted of 125 multiple-choice questions on clinical and
surgical issues in ophthalmology. The blueprint followed the same pattern as the
Brazilian Ophthalmology Council specialist test: uveitis: 9 questions; neuro ophthalmology: 7 questions; orbit: 4 questions; lacrimal system: 4 questions; ocular plastics: 8 questions; ocular tumors: 5 questions; cornea: 14 questions; contact lenses: 4 questions; refractive surgery: 2 questions; retina: 13 questions; cataract: 10 questions; glaucoma: 11 questions; refraction: 23 questions; strabismus: 7 questions; low vision: 4 questions. The shows the division of the test
questions. As the tests consisted of 125 multiple-choice questions, for the statistical
analysis, a 0.08 value corresponds to 10/125 for each correct answer; thus, it
was attributed to a score that could vary from 0 to 10 for each test. The questions in the tests were taken from the following books: Review Questions
in Opthalmology, Clinical Optics and Refraction, and Self-tests in Optic and Refraction. They were chosen according to the issue and level of difficulty (judged
by the authors), in a way that there were questions of different issues and
levels of difficulty. As there were residents from many parts of the country, the tests were online,
and all the residents from the 1 st to the 3 rd year of the
ophthalmology residency programs performed the tests simultaneously. Therefore,
all residents were enrolled in the same test, regardless of whether they were in
their 1 st , 2 nd , or 3 rd year of residency. The
tests were conducted once a year at the end of the school year. Each service organized the implementation of the tests, and the only requirement
was that all residents sat on the test simultaneously. Some services used their
own informatic lab rooms, while those that did not have one allowed their
residents to use their own computers, either at the service or at home, at a
predetermined schedule, as long as there was one computer for each resident.
First, participants had to create an account. Once completed, they were able to
access the site. – show a small portion of the site. The presentation page contains some important advice to read and the consent term
that had to be signed before the test itself . The test page contained the test itself. Once completed, participants had to
submit their answers. Immediately after the submission, the participant received
feedback . The feedback page shows the number of correct and incorrect answers, the time
spent performing the test, and the score (Figure 3). shows the correct answers and
explanations.
Frequency tables were used for the descriptive analysis of categorical variables.
Positions and dispersion measures were used for numeric variables. The
Kruskal–Wallis test was used to compare the differences between years, followed
by Dunn's test to identify significant differences. The Friedman or Wilcoxon test was used to compare students' knowledge growth. To investigate the relationship between the progress test and CBO scores, the
Spearman linear correlation coefficient and Wilcoxon test for related samples
were conducted. A statistical level of 0.05 was considered significant. Data were analyzed using the Statistical Analysis Software (SAS) System for
Windows (Statistical Analysis System), version 9.4. SAS Institute Inc,
2002-2012, Cary, North Carolina, United States.
Among the many ophthalmology residents all around Brazil invited to join the study,
24 accepted the invitation. A total of 297 residents participated in the progress
test. Of these, 100 (33.7%) were from the 1 st year, 108 (36.4%) from the
2 nd year, and 89 (30.0%) from the 3 rd year. Descriptive analysis and comparison of the scores for each residency
year The mean score of the 1 st year residents was 4.3, that of the
2 nd year residents was 5.1, and that of the 3 rd year
residents was 5.4. and show the descriptive analysis and
comparison of scores for each residency year. The Kruskal–Wallis test was used to compare the mean scores across the three
years of residency. The P value was < 0.0001, which was considered
statistically significant. Therefore, it is possible that there was a difference
between the mean scores. The Wilcoxon test was used for multiple comparisons of the mean scores for each
pair of the residency years (1 st versus 2 nd ,
1 st versus 3 rd , and 2 nd versus
3 rd ) to check the difference between the pairs. There was a
significant difference between the 1 st and 2 nd years and
the 1 st and 3 rd years of residency (P < 0.0001 in both
cases). However, the difference between the 2 nd and 3 rd years of residency was not significant (P = 0.0619). This may be because of the
pedagogic program itself since, if we look at it, we can see that almost all the
theoretical content was taught in the first two years of residency, with only a
small percentage remaining in the 3 rd year of residency. Relationship between the progress test and Brazilian Ophthalmology Council
(CBO) scores ( and ) For this analysis, we had only eight residents from the 3 rd year.
Correlation analysis demonstrated an association between the progress test and
CBO scores. Spearman correlation showed a positive and significant correlation between these two scores
(which was 0.61), which means that the higher the score on the progress test,
the higher the score on the CBO test.
The mean score of the 1 st year residents was 4.3, that of the
2 nd year residents was 5.1, and that of the 3 rd year
residents was 5.4. and show the descriptive analysis and
comparison of scores for each residency year. The Kruskal–Wallis test was used to compare the mean scores across the three
years of residency. The P value was < 0.0001, which was considered
statistically significant. Therefore, it is possible that there was a difference
between the mean scores. The Wilcoxon test was used for multiple comparisons of the mean scores for each
pair of the residency years (1 st versus 2 nd ,
1 st versus 3 rd , and 2 nd versus
3 rd ) to check the difference between the pairs. There was a
significant difference between the 1 st and 2 nd years and
the 1 st and 3 rd years of residency (P < 0.0001 in both
cases). However, the difference between the 2 nd and 3 rd years of residency was not significant (P = 0.0619). This may be because of the
pedagogic program itself since, if we look at it, we can see that almost all the
theoretical content was taught in the first two years of residency, with only a
small percentage remaining in the 3 rd year of residency.
and ) For this analysis, we had only eight residents from the 3 rd year.
Correlation analysis demonstrated an association between the progress test and
CBO scores. Spearman correlation showed a positive and significant correlation between these two scores
(which was 0.61), which means that the higher the score on the progress test,
the higher the score on the CBO test.
In this study, we demonstrated that progress tests could be used for ophthalmology
residency training. They helped to detect the residents' knowledge growth over time
and had a moderate relationship with the CBO test. Our findings are aligned with
previous studies in both undergraduate , , and residency training. , – , For example, in a study by Tomic et al., 4 years of progress testing were evaluated
in a medical school in Brazil and positive results were found, with a continuum of
cognitive gain during medical training. Similarly, previous studies with longitudinal tests on the residency program , , found that the progress test was able to detect the difference , among residency years. Taken together, the knowledge scores increased over
the years. , Concerning the relationship between the progress and CBO tests, our results were
partially in concordance with those of previous studies. For example, in an
undergraduate context, a study by Hamamoto Filho et al. found a correlation between
students' progress testing scores and their performance in a residency selection
process in Brazil. In the residency context, a descriptive study by Al-Mohammed A et al. compared the residents' performance on the American College of Physicians
(ACP) Internal Medicine In-Training Examination (IM-ITE) results and the
certification examination of the American Board of Internal Medicine (CABIM) and
American Board of Surgery Qualifying Examinations in Qatar, found that the
performance on the ITE could accurately predict the performance on both qualifying exams, which is in concordance with our results. Therefore, our study is in concordance with previous studies performed by residents.
What makes our study exclusive is that besides being performed in a country where
there are almost no similar studies, it is, as far as we are concerned, the only one
performed with ophthalmology residents. For the future Two more different tests will be developed, and each test will be used at the end
of the school year by all the residents from the 1 st to the
3 rd year of the ophthalmology residency programs. All the tests will have the same number of questions (125). They will follow the
same division of national testing issues; however, the questions will be
completely different from one test to another. In other words, all questions
will be changed from the 1 st year to another. Thus, at the end of the
3 years of residency, each resident performed three different tests. After the end of the tests, the tests will be revised, and each resident will
receive individual performance feedback through an online program developed with
personal login and password. Limitations of the study In some services, the residents were allowed to do the test at home because the
service did not have informatics labs or an appropriate classroom for them to
perform the tests. This can be biased because we cannot guarantee they did not
cheat on the test. In addition, as participation in the study was voluntary and
the progress test score was not part of the official residency program, some
residents did not take it seriously. Finally, our sample size for comparison of
the progress and CBO tests was small. However, even with such a small sample
size, we found a moderate and significant correlation.
Two more different tests will be developed, and each test will be used at the end
of the school year by all the residents from the 1 st to the
3 rd year of the ophthalmology residency programs. All the tests will have the same number of questions (125). They will follow the
same division of national testing issues; however, the questions will be
completely different from one test to another. In other words, all questions
will be changed from the 1 st year to another. Thus, at the end of the
3 years of residency, each resident performed three different tests. After the end of the tests, the tests will be revised, and each resident will
receive individual performance feedback through an online program developed with
personal login and password.
In some services, the residents were allowed to do the test at home because the
service did not have informatics labs or an appropriate classroom for them to
perform the tests. This can be biased because we cannot guarantee they did not
cheat on the test. In addition, as participation in the study was voluntary and
the progress test score was not part of the official residency program, some
residents did not take it seriously. Finally, our sample size for comparison of
the progress and CBO tests was small. However, even with such a small sample
size, we found a moderate and significant correlation.
Based on the data obtained, it is possible to see that the scores of the residents
improved over the years, which means that their knowledge increased. In other words,
there was progress along the residency course. Residents approved the longitudinal test as a self-learning tool and as a tool for
improving residency programs. Therefore, we can say that the implementation of a
longitudinal evaluation system in ophthalmological residency schools in Brazil was
successful and could be implemented in other medical subspecialties.
|
Safety and efficacy of probiotic supplements as adjunctive therapies in patients with COVID-19: A systematic review and meta-analysis | a0d8ae41-92e5-40f3-ba6b-3e01d2045b0f | 10065254 | Dental[mh] | Over 628,694,943 COVID-19 cases and over 6,576,088 associated deaths have been confirmed since the coronavirus disease 2019 (COVID-19) pandemic began . Despite 12,961,382,558 vaccine doses administered globally and COVID-19-specific treatments, disease burden, morbidity, and mortality remain significant, resulting in ongoing pandemic waves in many regions . Symptomatic illness rate remains high at 37,764 per 100,000 across all ages, and hospitalization rate and mortality have been rising from time to time, disproportionally impacting those with weaker immune systems and older populations above 65 . Respiratory complaints and complications from gastrointestinal symptoms are among the frequent causes of emergency room visits and hospitalization in patients with COVID-19 . GI symptoms such as diarrhea, abdominal pain, loss of appetite, nausea or vomiting can emerge in parallel or earlier than a dry cough, fever or dyspnea . More than 50% of hospitalized patients reported at least one gastrointestinal (GI) symptom. Up to 20% of all COVID-19 cases were manifested by GI symptoms alone and may be associated with longer disease duration and/or severity . This has prompted investigations for well-tolerated and cost-effective therapies that effectively reduce disease severity and symptom burden among patients and potentially decrease healthcare resource utilization. Probiotic therapy or bacteriotherapy is an oral supplement of live microorganisms with multiple health benefits. It exerts beneficial properties by enhancing intestinal microbial homeostasis and modulating the host’s immune response to pathogens . Altered gut homeostasis, or dysbiosis, is associated with various intestinal and extra-intestinal chronic diseases such as rheumatoid arthritis, osteoporosis and diabetes [ – ]. Furthermore, the immunomodulating roles of gut microflora in lung homeostasis and lung disease underscore the gut-lung axis crosslinked by pathogen-associated molecular patterns (PAMPs), lipopolypsaccharide (LPS) and migration of immune cells from the gut to the lungs . Probiotic use is associated with fewer episodes of acute upper respiratory tract infections with no significant side effects . Several strains of probiotics studied in randomized controlled trials effectively prevented ventilator-associated bacterial pneumonia in critically ill patients on mechanical ventilation . Recently published systematic review and meta-analysis in critically ill patients demonstrated a significant role of probiotics in reducing ventilator and healthcare-associated pneumonia to decrease intensive care unit (ICU), hospital length of stay and utilization of invasive mechanical ventilation based on low-certainty evidence . From the gastrointestinal aspect, several studies showed probiotics prevented and decreased the onset of antibiotic-associated diarrhea (AAD) and antibiotic-associated C. difficile colitis, potentially reducing healthcare resource utilization [ – ]. The antiviral properties of probiotics were demonstrated by in vitro studies for their ability to inhibit viral infection and replication, and suppress proinflammatory cascade . SARS-Cov-2 infection of human epithelial cells was attenuated by Lacticaseibacilus, and pro-inflammatory gene expression was similarly suppressed. In another in vitro study, Lactiplantibacillus effectively suppressed the replications of SARS-Cov-2 in human respiratory epithelial cells . The clinical application probiotics was studied in a propensity-score matched retrospective study, Bifidobacterium, Lactobacillus and Enterococcus associated with shorter recovery time and reduced hospitalization days in COVID-19 patients with moderate to severe symptoms . More compelling evidence came from several randomized controlled trials involving probiotic supplements during the coronavirus disease 2019 (COVID-19) pandemic. However, there is currently a lack of higher-level evidence evaluating probiotic therapy’s potential impact on COVID-19 symptoms and disease course. We aimed to conduct a systematic review and meta-analysis of randomized controlled trials to investigate the safety and efficacy of probiotic therapy in patients diagnosed with COVID-19 on their symptom development and overall clinical outcomes.
We pre-registered a protocol on PROSPERO (International prospective register of systematic reviews): CRD42022328256. We report our results according to the PRISMA checklist . Eligibility criteria We included all randomized controlled trials (RCTs) that randomized hospitalized or outpatient individuals with symptomatic COVID-19 infection with a confirmed diagnosis via a positive COVID-19 test (i.e., Real Time-Polymerase chain reaction or point of care testing). In addition, we included RCTs that randomized patients to oral probiotic supplementation immediately before or after the study period, compared to placebo or standard care. The probiotic agents included are prescribed at a standard, recommended, therapeutic dosage or adaptive dosing with appropriate clinical reasoning. The probiotic agent does not need to be licensed by the Food and Drug Administration (FDA) to be included. On top of the probiotic agent, other adjuncts, such as nutritional supplements or medication, can be present if they are in all arms of the study. Primary outcomes included reported adverse events related to treatments, any COVID-19 symptom, its duration, and severity with a focus on gastrointestinal and respiratory symptoms and change in COVID-19-related biomarkers. Secondary outcomes included escalation of care (requiring oxygen support by nasal cannula, non-invasive mask, invasive mechanical ventilation, circulatory support, or vasopressor use) and/or deaths. We excluded studies examining non-COVID-19 coronavirus-related respiratory illnesses such as SARS-CoV/ MERS-CoV and studies whereby the COVID-19 population is strictly pediatric (< 18 years old). With the assistance of an experienced librarian, we searched PubMed Central, Embase, CINAHL, and Cochrane Library from inception to July 31, 2022, for randomized trials meeting our inclusion criteria. Search string details are provided in ( ). In addition, trial registers were searched at ClinicalTrials.gov, the Cochrane Central Register of Controlled Trials and PROSPERO for ongoing trial reports. presents our database search strategy. There was no year or language restriction. Study selection Following training and calibration exercises to ensure sufficient agreement, pairs of reviewers, working independently and in duplicate, screened titles, and abstracts of search records and, subsequently, the full texts of records deemed potentially eligible at the title and abstract screening stage. Reviewers resolved discrepancies by discussion and, when necessary, by adjudication with a third party. Data collection Following training and calibration exercises to ensure sufficient agreement, pairs of reviewers, working independently and in duplicate, collected data on general information (first author, publication year, the country in which the study was conducted), trial characteristics in PICO format (participant setting; details of study design, probiotic microorganisms, and their frequency, route of administration and dosage, as well as the duration of treatment; control via placebo and/or standard or care treatment; primary and secondary outcomes, the total length of follow-up), patient characteristics (average age, sex, COVID-19 severity classification according to WHO COVID-19 Clinical Progression Scale ), and outcomes of interest (as previously described). In trials with information on clinical escalation and deaths, we defined a single composite outcome by tallying the number of both endpoint events while subtracting the number of reported deaths from the clinical escalation subgroups to avoid concomitant events or double counting. In addition, we sought clarifications from the corresponding authors regarding patient characteristics, protocols, and other unpublished data whenever necessary. Risk of bias Following training and calibration exercises to ensure sufficient agreement, pairs of reviewers, working independently and in duplicate, assessed the risk of bias with the Cochrane tool for assessing the risk of bias in randomized trials (RoB 2.0). We assessed the risk of bias across five domains: bias arising from the randomization process, bias due to departures from the intended intervention; bias from missing outcome data; bias in the measurement of the outcome; and bias in the selection of the reported result. Reviewers resolved discrepancies by discussion and, when not necessary, with adjudication by a third-party reviewer. Data analysis We performed a random-effects pairwise meta-analysis for all outcomes, using the inverse variance method with a restricted maximum likelihood estimator (REML). When convergence was not possible, we used the maximal likelihood estimator. We also planned to perform a fixed-effect analysis for estimates with less than four trials. We reported dichotomous outcomes using relative risk (RR) with a 95% confidence interval (CI) and continuous outcomes as mean differences with 95% CI. To optimize interpretability, for dichotomous outcomes, we calculated absolute effects by multiplying relative effects with the median baseline risk calculated from the placebo arm of the randomized trials. As a result, we presented an absolute risk difference and associated 95% CI. We performed subgroup analysis based on patient setting (outpatient versus inpatient). We hypothesized there may be differences in the effect of probiotics on clinical symptoms based on whether probiotics were administered in the outpatient versus inpatient setting. For statistically significant subgroups, we used the Instrument for assessing the Credibility of Effect Modification Analyses (ICEMAN) tool to assess the credibility of the subgroups . For outcomes with ten or more trials, we assessed for publication bias by visual inspection of funnel plots and Egger’s test. All data were analyzed using STATA version 17.0 Certainty of the evidence Two reviewers, working independently and in duplicate, assessed the certainty of the evidence using the GRADE approach. We made judgments of imprecision using the minimally contextualized approach. A minimally contextualized approach considers whether confidence intervals include the minimally important difference and thus does not consider whether it includes both minimally important and large effects. We sourced minimally important differences (MID) from available literature and by consensus when formal minimally important differences were not available in the literature. Whether a probiotic treatment is effective depends on whether the point estimate includes the MID. We report our results using the standard language guidance for GRADE. This involves referring to the evidence’s certainty by using specific modifiers. For example, the effectiveness of treatment could be rated by saying: Treatment X reduces mortality (high certainty), Treatment X probably reduces mortality (moderate certainty), Treatment X may reduce mortality (low certainty), and the Effect of Treatment X on mortality is very uncertain. We rated the certainty for each comparison and outcome as high, moderate, low, or very low based on the risk of bias, inconsistency, indirectness, publication bias, and imprecision.
We included all randomized controlled trials (RCTs) that randomized hospitalized or outpatient individuals with symptomatic COVID-19 infection with a confirmed diagnosis via a positive COVID-19 test (i.e., Real Time-Polymerase chain reaction or point of care testing). In addition, we included RCTs that randomized patients to oral probiotic supplementation immediately before or after the study period, compared to placebo or standard care. The probiotic agents included are prescribed at a standard, recommended, therapeutic dosage or adaptive dosing with appropriate clinical reasoning. The probiotic agent does not need to be licensed by the Food and Drug Administration (FDA) to be included. On top of the probiotic agent, other adjuncts, such as nutritional supplements or medication, can be present if they are in all arms of the study. Primary outcomes included reported adverse events related to treatments, any COVID-19 symptom, its duration, and severity with a focus on gastrointestinal and respiratory symptoms and change in COVID-19-related biomarkers. Secondary outcomes included escalation of care (requiring oxygen support by nasal cannula, non-invasive mask, invasive mechanical ventilation, circulatory support, or vasopressor use) and/or deaths. We excluded studies examining non-COVID-19 coronavirus-related respiratory illnesses such as SARS-CoV/ MERS-CoV and studies whereby the COVID-19 population is strictly pediatric (< 18 years old). With the assistance of an experienced librarian, we searched PubMed Central, Embase, CINAHL, and Cochrane Library from inception to July 31, 2022, for randomized trials meeting our inclusion criteria. Search string details are provided in ( ). In addition, trial registers were searched at ClinicalTrials.gov, the Cochrane Central Register of Controlled Trials and PROSPERO for ongoing trial reports. presents our database search strategy. There was no year or language restriction.
Following training and calibration exercises to ensure sufficient agreement, pairs of reviewers, working independently and in duplicate, screened titles, and abstracts of search records and, subsequently, the full texts of records deemed potentially eligible at the title and abstract screening stage. Reviewers resolved discrepancies by discussion and, when necessary, by adjudication with a third party.
Following training and calibration exercises to ensure sufficient agreement, pairs of reviewers, working independently and in duplicate, collected data on general information (first author, publication year, the country in which the study was conducted), trial characteristics in PICO format (participant setting; details of study design, probiotic microorganisms, and their frequency, route of administration and dosage, as well as the duration of treatment; control via placebo and/or standard or care treatment; primary and secondary outcomes, the total length of follow-up), patient characteristics (average age, sex, COVID-19 severity classification according to WHO COVID-19 Clinical Progression Scale ), and outcomes of interest (as previously described). In trials with information on clinical escalation and deaths, we defined a single composite outcome by tallying the number of both endpoint events while subtracting the number of reported deaths from the clinical escalation subgroups to avoid concomitant events or double counting. In addition, we sought clarifications from the corresponding authors regarding patient characteristics, protocols, and other unpublished data whenever necessary.
Following training and calibration exercises to ensure sufficient agreement, pairs of reviewers, working independently and in duplicate, assessed the risk of bias with the Cochrane tool for assessing the risk of bias in randomized trials (RoB 2.0). We assessed the risk of bias across five domains: bias arising from the randomization process, bias due to departures from the intended intervention; bias from missing outcome data; bias in the measurement of the outcome; and bias in the selection of the reported result. Reviewers resolved discrepancies by discussion and, when not necessary, with adjudication by a third-party reviewer.
We performed a random-effects pairwise meta-analysis for all outcomes, using the inverse variance method with a restricted maximum likelihood estimator (REML). When convergence was not possible, we used the maximal likelihood estimator. We also planned to perform a fixed-effect analysis for estimates with less than four trials. We reported dichotomous outcomes using relative risk (RR) with a 95% confidence interval (CI) and continuous outcomes as mean differences with 95% CI. To optimize interpretability, for dichotomous outcomes, we calculated absolute effects by multiplying relative effects with the median baseline risk calculated from the placebo arm of the randomized trials. As a result, we presented an absolute risk difference and associated 95% CI. We performed subgroup analysis based on patient setting (outpatient versus inpatient). We hypothesized there may be differences in the effect of probiotics on clinical symptoms based on whether probiotics were administered in the outpatient versus inpatient setting. For statistically significant subgroups, we used the Instrument for assessing the Credibility of Effect Modification Analyses (ICEMAN) tool to assess the credibility of the subgroups . For outcomes with ten or more trials, we assessed for publication bias by visual inspection of funnel plots and Egger’s test. All data were analyzed using STATA version 17.0
Two reviewers, working independently and in duplicate, assessed the certainty of the evidence using the GRADE approach. We made judgments of imprecision using the minimally contextualized approach. A minimally contextualized approach considers whether confidence intervals include the minimally important difference and thus does not consider whether it includes both minimally important and large effects. We sourced minimally important differences (MID) from available literature and by consensus when formal minimally important differences were not available in the literature. Whether a probiotic treatment is effective depends on whether the point estimate includes the MID. We report our results using the standard language guidance for GRADE. This involves referring to the evidence’s certainty by using specific modifiers. For example, the effectiveness of treatment could be rated by saying: Treatment X reduces mortality (high certainty), Treatment X probably reduces mortality (moderate certainty), Treatment X may reduce mortality (low certainty), and the Effect of Treatment X on mortality is very uncertain. We rated the certainty for each comparison and outcome as high, moderate, low, or very low based on the risk of bias, inconsistency, indirectness, publication bias, and imprecision.
Study selection We identified 323 unique studies to screen for eligibility, and 15 were eligible for full-text review. Nine studies met the inclusion criteria for a systematic review and underwent data extraction [ – ]; eight were eligible to be included in the final meta-analysis [ – ]. presents more detail on the study selection process. Study characteristics We included nine studies that randomized COVID-19 patients to probiotics and eight reported outcomes of interest. In the eight studies that were included in the meta-analysis, there were a total of 1027 participants. The mean age was 57.2 years, and 49.2% were male. Probiotics were administered by an oral route. Frequency, dosage, and duration of probiotic administration are summarized in the supplementary ( ). Five studies involved multi-strain bacterial formulations, two involved a single bacterial strain and one involved a single-strain yeast formulation. They were all conducted on patients with confirmed COVID-19 diagnoses. In addition, two studies were conducted in outpatient settings in patients with mild symptoms. In comparison, six studies took place in hospitals and recruited patients with moderate to severe symptoms according to the WHO classification . In all studies, probiotics were given immediately or within 48 hours of COVID-19 diagnosis. The duration of probiotic treatment was between 6 to 30 days, with once or split daily dosing regimens. and Table present more details on the characteristics of the included studies (systematic review N = 9; meta-analysis N = 8). Risk of bias in studies Two trials were at risk of bias due to concerns with allocation concealment, and two were at risk of bias due to concerns around the measurement of the outcome; no trial was at risk of bias due to missing data or selective reporting of results. presents more details on the risk of bias in included trials. Outcomes Cough or dyspnea-like respiratory symptoms Five trials reported respiratory symptoms of cough or dyspnea-like reparatory symptoms, including 658 patients with a median follow of 14 days. Probiotics probably improve cough or dyspnea compared to placebo/standard care (RR 0.37 [0.19 to 0.73]; moderate certainty). We rated down once for imprecision. There was unimportant heterogeneity (I 2 = 38.11%). reports the summary of the findings. presents the forest plot. Gastrointestinal symptoms Seven trials reported gastrointestinal symptoms of diarrhea, including 958 patients with a median follow of 26 days. Probiotics probably improve the risk of diarrhea in COVID-19 patients (RR 0.61 [0.43 to 0.87]; moderate certainty). We rated the down once for imprecision. There was no heterogeneity (I 2 = 0.00%) reports the summary of the findings. presents the forest plot Change in inflammatory biomarkers Four trials reported c-reactive protein levels, including 630 patients with a median follow-up of 14 days. The effect of probiotics on c-reactive protein levels compared to placebo/standard care is very uncertain (mean difference -9.27 mg/L [-28.10 mg/L to 9.56 mg/L]; very low certainty). We rated down twice for imprecision and once for inconsistency. There was important heterogeneity (I 2 = 99.89%). reports the summary of the findings. presents the forest plot. Adverse events Seven trials reported treatment-associated adverse events, including 938 patients with a median follow-up of 24 days. Probiotics probably have reduced adverse events compared to placebo/standard care (RR 0.62 [0.46 to 0.83]; moderate certainty). We rated down once for imprecision. There was no heterogeneity (I 2 = 0.00%). reports the summary of the findings. presents the forest plot. Subgroup analysis We found a statistically significant subgroup difference for respiratory symptoms with a patient setting as a modifier (p<0.001). There was a substantially larger effect in hospitalized patients (RR 0.13 [95% CI 0.05 to 0.39]) as compared to outpatients (RR 0.65 [95% CI 0.53 to 0.79]). We found this subgroup to have moderate to low credibility using the ICEMAN tool. presents the forest plots. We did not find statistically significant subgroups for mortality, clinical escalation, composite of the two, gastrointestinal symptoms, serious adverse events or change in CRP (mg/L). – Figs present the forest plots for these analyses. Clinical escalation Six trials reported clinical escalation due to COVID-19, including 823 patients with a median follow of 25 days. In the probiotic arms, 12.7% of patients underwent escalation of care versus 20.5% in the placebo arms. Probiotics may not reduce clinical escalation compared to placebo (RR 0.57 [0.31 to 1.07]; low certainty). We rated down twice for imprecision and once for risk of bias. There was unimportant heterogeneity (I 2 = 13.4%) reports the summary of the findings. presents the forest plot. Mortality Six trials reported deaths from COVID-19, including 823 patients with a median follow of 25 days. The mean mortality rates were 1.1% in the probiotic arm and 5.6% in the placebo arm. Probiotics may have no significant effect on mortality compared to placebo (RR 0.50 [0.20 to 1.29]; low certainty). We rated down twice for imprecision and once for risk of bias. There was no heterogeneity (I 2 = 0.00%) reports the summary of the findings. presents the forest plot. Composite clinical escalation and mortality Six trials reported clinical escalation and deaths from COVID-19, including 823 patients with a median follow of 25 days. Probiotics may reduce composite clinical escalation or mortality compared to placebo (RR 0.41 [0.18 to 0.93]; low certainty). We rated down twice for imprecision. There was significant heterogeneity (I 2 = 64.58%). reports the summary of the findings. presents the forest plot.
We identified 323 unique studies to screen for eligibility, and 15 were eligible for full-text review. Nine studies met the inclusion criteria for a systematic review and underwent data extraction [ – ]; eight were eligible to be included in the final meta-analysis [ – ]. presents more detail on the study selection process.
We included nine studies that randomized COVID-19 patients to probiotics and eight reported outcomes of interest. In the eight studies that were included in the meta-analysis, there were a total of 1027 participants. The mean age was 57.2 years, and 49.2% were male. Probiotics were administered by an oral route. Frequency, dosage, and duration of probiotic administration are summarized in the supplementary ( ). Five studies involved multi-strain bacterial formulations, two involved a single bacterial strain and one involved a single-strain yeast formulation. They were all conducted on patients with confirmed COVID-19 diagnoses. In addition, two studies were conducted in outpatient settings in patients with mild symptoms. In comparison, six studies took place in hospitals and recruited patients with moderate to severe symptoms according to the WHO classification . In all studies, probiotics were given immediately or within 48 hours of COVID-19 diagnosis. The duration of probiotic treatment was between 6 to 30 days, with once or split daily dosing regimens. and Table present more details on the characteristics of the included studies (systematic review N = 9; meta-analysis N = 8).
Two trials were at risk of bias due to concerns with allocation concealment, and two were at risk of bias due to concerns around the measurement of the outcome; no trial was at risk of bias due to missing data or selective reporting of results. presents more details on the risk of bias in included trials.
Cough or dyspnea-like respiratory symptoms Five trials reported respiratory symptoms of cough or dyspnea-like reparatory symptoms, including 658 patients with a median follow of 14 days. Probiotics probably improve cough or dyspnea compared to placebo/standard care (RR 0.37 [0.19 to 0.73]; moderate certainty). We rated down once for imprecision. There was unimportant heterogeneity (I 2 = 38.11%). reports the summary of the findings. presents the forest plot.
Five trials reported respiratory symptoms of cough or dyspnea-like reparatory symptoms, including 658 patients with a median follow of 14 days. Probiotics probably improve cough or dyspnea compared to placebo/standard care (RR 0.37 [0.19 to 0.73]; moderate certainty). We rated down once for imprecision. There was unimportant heterogeneity (I 2 = 38.11%). reports the summary of the findings. presents the forest plot.
Seven trials reported gastrointestinal symptoms of diarrhea, including 958 patients with a median follow of 26 days. Probiotics probably improve the risk of diarrhea in COVID-19 patients (RR 0.61 [0.43 to 0.87]; moderate certainty). We rated the down once for imprecision. There was no heterogeneity (I 2 = 0.00%)
presents the forest plot Change in inflammatory biomarkers Four trials reported c-reactive protein levels, including 630 patients with a median follow-up of 14 days. The effect of probiotics on c-reactive protein levels compared to placebo/standard care is very uncertain (mean difference -9.27 mg/L [-28.10 mg/L to 9.56 mg/L]; very low certainty). We rated down twice for imprecision and once for inconsistency. There was important heterogeneity (I 2 = 99.89%). reports the summary of the findings. presents the forest plot. Adverse events Seven trials reported treatment-associated adverse events, including 938 patients with a median follow-up of 24 days. Probiotics probably have reduced adverse events compared to placebo/standard care (RR 0.62 [0.46 to 0.83]; moderate certainty). We rated down once for imprecision. There was no heterogeneity (I 2 = 0.00%). reports the summary of the findings. presents the forest plot. Subgroup analysis We found a statistically significant subgroup difference for respiratory symptoms with a patient setting as a modifier (p<0.001). There was a substantially larger effect in hospitalized patients (RR 0.13 [95% CI 0.05 to 0.39]) as compared to outpatients (RR 0.65 [95% CI 0.53 to 0.79]). We found this subgroup to have moderate to low credibility using the ICEMAN tool. presents the forest plots. We did not find statistically significant subgroups for mortality, clinical escalation, composite of the two, gastrointestinal symptoms, serious adverse events or change in CRP (mg/L). – Figs present the forest plots for these analyses.
Four trials reported c-reactive protein levels, including 630 patients with a median follow-up of 14 days. The effect of probiotics on c-reactive protein levels compared to placebo/standard care is very uncertain (mean difference -9.27 mg/L [-28.10 mg/L to 9.56 mg/L]; very low certainty). We rated down twice for imprecision and once for inconsistency. There was important heterogeneity (I 2 = 99.89%). reports the summary of the findings. presents the forest plot.
Seven trials reported treatment-associated adverse events, including 938 patients with a median follow-up of 24 days. Probiotics probably have reduced adverse events compared to placebo/standard care (RR 0.62 [0.46 to 0.83]; moderate certainty). We rated down once for imprecision. There was no heterogeneity (I 2 = 0.00%). reports the summary of the findings. presents the forest plot.
We found a statistically significant subgroup difference for respiratory symptoms with a patient setting as a modifier (p<0.001). There was a substantially larger effect in hospitalized patients (RR 0.13 [95% CI 0.05 to 0.39]) as compared to outpatients (RR 0.65 [95% CI 0.53 to 0.79]). We found this subgroup to have moderate to low credibility using the ICEMAN tool. presents the forest plots. We did not find statistically significant subgroups for mortality, clinical escalation, composite of the two, gastrointestinal symptoms, serious adverse events or change in CRP (mg/L). – Figs present the forest plots for these analyses.
Six trials reported clinical escalation due to COVID-19, including 823 patients with a median follow of 25 days. In the probiotic arms, 12.7% of patients underwent escalation of care versus 20.5% in the placebo arms. Probiotics may not reduce clinical escalation compared to placebo (RR 0.57 [0.31 to 1.07]; low certainty). We rated down twice for imprecision and once for risk of bias. There was unimportant heterogeneity (I 2 = 13.4%) reports the summary of the findings. presents the forest plot.
Six trials reported deaths from COVID-19, including 823 patients with a median follow of 25 days. The mean mortality rates were 1.1% in the probiotic arm and 5.6% in the placebo arm. Probiotics may have no significant effect on mortality compared to placebo (RR 0.50 [0.20 to 1.29]; low certainty). We rated down twice for imprecision and once for risk of bias. There was no heterogeneity (I 2 = 0.00%) reports the summary of the findings. presents the forest plot.
Six trials reported clinical escalation and deaths from COVID-19, including 823 patients with a median follow of 25 days. Probiotics may reduce composite clinical escalation or mortality compared to placebo (RR 0.41 [0.18 to 0.93]; low certainty). We rated down twice for imprecision. There was significant heterogeneity (I 2 = 64.58%). reports the summary of the findings. presents the forest plot.
Main findings We found that probiotics were effective in treating COVID-19 symptoms, particularly respiratory and gastrointestinal symptoms. Early probiotic supplements and standard therapy decrease gastrointestinal and respiratory symptoms and are associated with a lower risk for adverse events, confirming their safety profiles with moderate certainty of evidence. Plausible explanations for these findings are decreased COVID-19 symptom duration and severity, and reduced hospital-acquired infections such as ventilator-associated pneumonia and antibiotic-associated diarrhea in the probiotic-treated groups. The potential implications are probiotics may lead to reduced emergency room visits and hospitalization rates, reduced length of hospital or ICU stays, and other healthcare resource utilization rates among patients with COVID-19. Furthermore, reduced healthcare resource utilization is supported by lower composite end-point care escalation or death from COVID-19. Relation to previous findings Currently, several COVID-19-specific treatments exist and are prioritized for selected patients with selected patient demographics and clinical profiles . Potential barriers and contraindications to these therapies are relatively common and can negatively affect patient access, especially in the more vulnerable sub-populations with known adverse outcomes . In outpatient settings, access to COVID-19 therapies, delayed contact with health professionals, and drug-drug interactions are significant obstacles to administering COVID-19-specific therapies . At the same time, the majority of these therapies require parenteral administration. Oral probiotics, being an oral formulation, are safe, convenient and accessible options for patients of all severity types. In the inpatient settings, there were no reported adverse events related to the concomitant use of probiotics and other COVID-19 treatments suggesting probiotics were well tolerated by various inpatient regimens and caused no strain-specific nosocomial infections across the disease spectrum. Before this review, there has yet to be a systematic review and meta-analysis generated from randomized controlled trials on the effects of probiotics on COVID-19 symptoms and disease course. Previous reviews were qualitative and retrospective . However, there were recent systematic reviews and meta-analyses on probiotics in hospital-acquired and ventilator-associated pneumonia . These reviews have provided evidence for reduced ventilator-associated pneumonia (VAP) in non- COVID-19 patients treated with probiotics. Patients with COVID-19 had an increased risk of VAP and hospital-acquired pneumonia compared to non-COVID patients admitted to ICU or hospitals and overall experienced considerable morbidity and mortality . Speculated intrinsic factors are more severe parenchymal lung damage and poorer lung compliance from COVID-19 than non-COVID patients, and extrinsic factors included increased use of systematic immunosuppressants. This systematic review provides the best-updated evidence for the effects of probiotics on symptomatic COVID-19 cases, among them those with respiratory symptoms. Patient settings modified the respiratory effect. Hospitalized patients with moderate to severe respiratory symptoms experienced greater risk reduction than those treated in outpatient settings. In addition to antiviral properties, the results suggested probiotics may effectively reduce ventilator-associated pneumonia and hospital-acquired pneumonia among COVID-19 populations. Lower gastrointestinal symptoms such as diarrhea may be present in 34% of COVID-19 cases, and abdominal pain or nausea in 26% to 35% . SARS-CoV-2 RNA can be readily detected in stool. Although the mechanisms are still poorly understood, disruption of the intestinal microbial composition by SaRS-Cov-2 and antibiotics used to treat COVID-19 may worsen dysbiosis and inflammation of the respiratory and gastrointestinal systems . The results from this review are consistent with previous systematic reviews and meta-analyses, whereby probiotic therapies were shown to prevent antibiotic-associated diarrhea and reduce acute viral-associated diarrhea symptoms . The lack of effect on mortality is consistent with reports from retrospective and non-randomized prospective studies and results from non-COVID critically ill patients . This review focuses on probiotic safety as an adjunctive therapy to existing therapies such as dexamethasone, antibiotics, anticoagulants, and other types of COVID-19-specific treatments, including remdesivir and immunomodulators. There were significantly fewer adverse event rates from patients on probiotics; in particular, there were no reports of bacterial infections from probiotic therapy, regardless of comorbidity and disease severity. Implications and future directions One of the most robust findings was the reduction in COVID-19 symptoms in the probiotic groups. Probiotics may be recommended as a cost-effective strategy for patients with COVID-19 symptoms to prevent symptom progression and decrease symptom duration. The mechanism of action for probiotics and their effectiveness in long COVID symptoms warrant investigation by future studies; there may be differences in response rate and response magnitude in specific subpopulations that necessitate future investigations. Strengths and limitations Our review has several strengths. First, our protocol was prospectively registered on PROSPERO. Second, the meta-analysis is generated from randomized controlled trials involving patients with a confirmed diagnosis of COVID-19 by nasopharyngeal PCR, with a low risk of bias for the majority; this strengthens its internal validity. Third, we have included populations from outpatient and inpatient settings with a spectrum of disease severity. Fourth, the populations examined by our study are diverse in countries of origin, age, and ethnicity, thus lending support to its external validity. Finally, we performed sub-group analysis based on patient settings to explore how sub-populations with different clinical characteristics may respond to probiotics. Our review has a few limitations. First, probiotic supplementations are somewhat diverse, and their mechanism of action needs to be better understood; externalizing our findings to other types of probiotics may be too ambitious. Second, there is limited data on older populations more vulnerable to symptomatic COVID-19 infections and sequelae; sub-group analysis based on patient demographics, such as age, may clarify who will benefit the most from probiotic therapy. Third, primary or secondary preventative roles are beyond this review’s scope since only confirmed cases are included. Lastly, given that some trials were published during the early-to-mid pandemic era, there could be selection and publication bias.
We found that probiotics were effective in treating COVID-19 symptoms, particularly respiratory and gastrointestinal symptoms. Early probiotic supplements and standard therapy decrease gastrointestinal and respiratory symptoms and are associated with a lower risk for adverse events, confirming their safety profiles with moderate certainty of evidence. Plausible explanations for these findings are decreased COVID-19 symptom duration and severity, and reduced hospital-acquired infections such as ventilator-associated pneumonia and antibiotic-associated diarrhea in the probiotic-treated groups. The potential implications are probiotics may lead to reduced emergency room visits and hospitalization rates, reduced length of hospital or ICU stays, and other healthcare resource utilization rates among patients with COVID-19. Furthermore, reduced healthcare resource utilization is supported by lower composite end-point care escalation or death from COVID-19.
Currently, several COVID-19-specific treatments exist and are prioritized for selected patients with selected patient demographics and clinical profiles . Potential barriers and contraindications to these therapies are relatively common and can negatively affect patient access, especially in the more vulnerable sub-populations with known adverse outcomes . In outpatient settings, access to COVID-19 therapies, delayed contact with health professionals, and drug-drug interactions are significant obstacles to administering COVID-19-specific therapies . At the same time, the majority of these therapies require parenteral administration. Oral probiotics, being an oral formulation, are safe, convenient and accessible options for patients of all severity types. In the inpatient settings, there were no reported adverse events related to the concomitant use of probiotics and other COVID-19 treatments suggesting probiotics were well tolerated by various inpatient regimens and caused no strain-specific nosocomial infections across the disease spectrum. Before this review, there has yet to be a systematic review and meta-analysis generated from randomized controlled trials on the effects of probiotics on COVID-19 symptoms and disease course. Previous reviews were qualitative and retrospective . However, there were recent systematic reviews and meta-analyses on probiotics in hospital-acquired and ventilator-associated pneumonia . These reviews have provided evidence for reduced ventilator-associated pneumonia (VAP) in non- COVID-19 patients treated with probiotics. Patients with COVID-19 had an increased risk of VAP and hospital-acquired pneumonia compared to non-COVID patients admitted to ICU or hospitals and overall experienced considerable morbidity and mortality . Speculated intrinsic factors are more severe parenchymal lung damage and poorer lung compliance from COVID-19 than non-COVID patients, and extrinsic factors included increased use of systematic immunosuppressants. This systematic review provides the best-updated evidence for the effects of probiotics on symptomatic COVID-19 cases, among them those with respiratory symptoms. Patient settings modified the respiratory effect. Hospitalized patients with moderate to severe respiratory symptoms experienced greater risk reduction than those treated in outpatient settings. In addition to antiviral properties, the results suggested probiotics may effectively reduce ventilator-associated pneumonia and hospital-acquired pneumonia among COVID-19 populations. Lower gastrointestinal symptoms such as diarrhea may be present in 34% of COVID-19 cases, and abdominal pain or nausea in 26% to 35% . SARS-CoV-2 RNA can be readily detected in stool. Although the mechanisms are still poorly understood, disruption of the intestinal microbial composition by SaRS-Cov-2 and antibiotics used to treat COVID-19 may worsen dysbiosis and inflammation of the respiratory and gastrointestinal systems . The results from this review are consistent with previous systematic reviews and meta-analyses, whereby probiotic therapies were shown to prevent antibiotic-associated diarrhea and reduce acute viral-associated diarrhea symptoms . The lack of effect on mortality is consistent with reports from retrospective and non-randomized prospective studies and results from non-COVID critically ill patients . This review focuses on probiotic safety as an adjunctive therapy to existing therapies such as dexamethasone, antibiotics, anticoagulants, and other types of COVID-19-specific treatments, including remdesivir and immunomodulators. There were significantly fewer adverse event rates from patients on probiotics; in particular, there were no reports of bacterial infections from probiotic therapy, regardless of comorbidity and disease severity.
One of the most robust findings was the reduction in COVID-19 symptoms in the probiotic groups. Probiotics may be recommended as a cost-effective strategy for patients with COVID-19 symptoms to prevent symptom progression and decrease symptom duration. The mechanism of action for probiotics and their effectiveness in long COVID symptoms warrant investigation by future studies; there may be differences in response rate and response magnitude in specific subpopulations that necessitate future investigations.
Our review has several strengths. First, our protocol was prospectively registered on PROSPERO. Second, the meta-analysis is generated from randomized controlled trials involving patients with a confirmed diagnosis of COVID-19 by nasopharyngeal PCR, with a low risk of bias for the majority; this strengthens its internal validity. Third, we have included populations from outpatient and inpatient settings with a spectrum of disease severity. Fourth, the populations examined by our study are diverse in countries of origin, age, and ethnicity, thus lending support to its external validity. Finally, we performed sub-group analysis based on patient settings to explore how sub-populations with different clinical characteristics may respond to probiotics. Our review has a few limitations. First, probiotic supplementations are somewhat diverse, and their mechanism of action needs to be better understood; externalizing our findings to other types of probiotics may be too ambitious. Second, there is limited data on older populations more vulnerable to symptomatic COVID-19 infections and sequelae; sub-group analysis based on patient demographics, such as age, may clarify who will benefit the most from probiotic therapy. Third, primary or secondary preventative roles are beyond this review’s scope since only confirmed cases are included. Lastly, given that some trials were published during the early-to-mid pandemic era, there could be selection and publication bias.
Our study found favorable effects from the early probiotic supplements in patients treated for COVID-19 in hospital or outpatient settings. An early probiotic supplement is safe in mild, moderate, or severe disease types and is associated with reduced symptom progression and duration. In addition, the evidence suggests that probiotic supplements may play a role in reducing overall health costs associated with COVID-19 by decreasing its disease burden. Registration : Protocol was registered on PROSPERO (International prospective register of systematic reviews): CRD42022328256.
S1 Checklist PRISMA 2020 checklist. (DOCX) Click here for additional data file. S1 Fig Subgroup analysis of clinical escalation based on patient setting. (TIFF) Click here for additional data file. S2 Fig Subgroup analysis of composite end point based on patient setting. (TIFF) Click here for additional data file. S3 Fig Subgroup analysis of CRP level (mg/L) based on patient setting. (TIFF) Click here for additional data file. S4 Fig Subgroup analysis of gastrointestinal symptoms based on patient setting. (TIFF) Click here for additional data file. S5 Fig Subgroup analysis of mortality based on patient setting. (TIFF) Click here for additional data file. S6 Fig Subgroup analysis of adverse events based on patient setting. (TIFF) Click here for additional data file. S7 Fig C-reactive protein levels (mg/L). (TIFF) Click here for additional data file. S8 Fig Adverse events. (TIFF) Click here for additional data file. S9 Fig Subgroup analysis of respiratory symptoms based on patient setting. (TIFF) Click here for additional data file. S1 Table Database search strategies. (TIFF) Click here for additional data file. S2 Table Outcome summary. (TIFF) Click here for additional data file. S3 Table Study characteristics. (TIFF) Click here for additional data file.
|
Management of hypopituitarism: a perspective from the Brazilian Society of Endocrinology and Metabolism | ff4d2119-9423-4c34-a576-61c2e02c5fe2 | 10065316 | Physiology[mh] | Hypopituitarism is a heterogeneous disease characterized by insufficient secretion of one or more pituitary hormones due to genetic or acquired causes . The only available epidemiological data estimating the frequency of hypopituitarism in the adult population derived from a Spanish study published in 2001 showing a prevalence of 455 cases per million inhabitants and an incidence of 42.1 cases per million inhabitants per year . If we consider these numbers to be true for our population, Brazil has approximately 100,000 patients with hypopituitarism and roughly 8,500 new cases per year. However, these numbers certainly underestimate the frequency of hypopituitarism, which has increased over the last years due to recognition of new etiologies, such as cerebral hemorrhage and head trauma, emergence of new etiologies, such as drug-induced hypophysitis, and improvements in diagnostic tools . Hypopituitarism has been associated with increased mortality, particularly due to cardiovascular and cerebrovascular diseases . Two recent meta-analyses involving observational studies have confirmed such increased mortality, with higher rates observed especially in women and in patients of younger age at diagnosis . Nevertheless, new concepts in the pathophysiology of hormonal deficiencies, recent advances in diagnostic tools, and the emergence of new formulations for hormone replacement have significantly contributed to a reduction in morbidity and mortality rates in these patients . This narrative review provides a guide for the management of patients with hypopituitarism in Brazil, considering as much as possible regional differences and health care disparities.
Hypopituitarism is a consequence of disorders that compromise the secretory function of the anterior pituitary or interfere with the hypothalamic secretion of anterior pituitary-releasing hormones. Hypopituitarism may be secondary to genetic defects, congenital abnormalities, or acquired lesions, such as tumors, vascular abnormalities, trauma, or inflammatory, infiltrative, and infectious diseases. In the pediatric population, the most frequent causes of hypopituitarism are genetic or congenital disorders affecting the hypothalamic-pituitary region, usually associated with mutations or low expression of transcription factors responsible for pituitary development, alteration in hypothalamic hormone receptors, structural defects, or mutations in pituitary hormones or their subunits . On the other hand, acquired causes are more prevalent in adulthood. Despite that, pituitary tumors and their treatment with surgery and/or radiotherapy account for up to two-thirds of the cases in most literature series. Indeed, in a recent single-center Brazilian study involving 99 adult patients with hypopituitarism, around half of the patients had nontumoral causes of hypopituitarism, while FSH/LH, GH, TSH, and ACTH deficiencies were present in 99%, 98.6%, 96%, and 81.8% of them, respectively . Of note, Sheehan’s syndrome – necrosis of the pituitary gland occurring during or soon after labor – continues to be a common cause of hypopituitarism in women in developing regions of the world . Aneurysmal subarachnoid hemorrhage is another nontumoral etiology associated with pituitary deficiency . A study carried out in Belo Horizonte, Brazil, observed some degree of pituitary dysfunction in 59% of 66 consecutive patients evaluated in the first 15 days after aneurysmal subarachnoid hemorrhage . However, the prevalence of hypopituitarism in this group of patients decreases in the long term, suggesting that some deficiencies are temporary and may improve over time . The prevalence of hypopituitarism following traumatic brain injury (TBI) is extremely variable across studies, with percentages ranging from 16% to 69% (owing mainly to differences in study populations, severity of trauma, time of evaluation, and laboratory criteria used to define pituitary deficiency), while some authors consider these rates to be overestimated . Brazilian investigators have studied the association between low LH and testosterone levels with morbidity and mortality during the acute phase of severe TBI, but the role of these hormones as prognostic factors is still uncertain . Similarly, sports-related TBI, comprising continuous or acute trauma in professional athletes, amateur sporting, or even during recreational activities, may also result in pituitary dysfunction that is commonly neglected and undiagnosed . Hypophysitis, another frequently underdiagnosed cause of hypopituitarism, is related to an inflammatory process of the pituitary and may be classified as primary or secondary, depending on its etiology. Primary hypophysitis has a prevalence in the population of approximately 0.2–0.88% and an annual incidence of 1/9,000,000 . The most common form of hypophysitis is lymphocytic or autoimmune, corresponding to approximately 72% of the cases . Secondary hypophysitis is related to inflammatory (sarcoidosis, granulomatosis with polyangiitis), infectious (tuberculosis, syphilis, fungal infections), or infiltrative (hemochromatosis, amyloidosis, Langerhans-cell histiocytosis) diseases, or may be drug-related, as observed with two classes of immune checkpoint inhibitors, namely, cytotoxic T lymphocyte associated antigen (CTLA-4) and programmed cell death protein 1 receptor (PD-1) inhibitors . Hypophysitis affects 11–12% of the patients treated with CTLA-4 inhibitors (ipilimumab and tremelimumab) and is more prevalent in men. In contrast, hypophysitis is less common during therapy with PD-1 inhibitors (nivolumab and pembrolizumab) and is rarely observed with PD-L1-targeted drugs (atezolizumab, avelumab) . PD-1/PD-L1 interaction may regulate apoptosis, while PD-1/PD-L1 blockade may result in positive activation of the immune system with consequent inhibition of tumor growth . Early recognition and appropriate management of immune-mediated hypophysitis are important to initiate pituitary hormone replacement without interrupting cancer treatment ( and ) .
Clinical symptoms of hypopituitarism vary greatly depending on the cause, the age of the patient and speed of onset, affected pituitary hormones, and magnitude of hormone deficiency. The symptoms usually develop insidiously in adults, begin up to several years before diagnosis, and are generally nonspecific, including weakness, tiredness, lethargy, increased sensitivity to cold, discomfort, appetite loss, and weight loss or gain. Most patients with hypopituitarism have multiple pituitary hormone deficiencies, and it is challenging to assign specific signs and symptoms to a single hormone deficiency . Hypopituitarism has a variable dynamic throughout its development and follow-up, characterized by a complete or sequential loss of pituitary function, which is usually permanent, although transient deficiencies with recovery years after the initial event may occur . The sensitivity of the different pituitary hormones to pathological damage is variable. The usual sequential pattern of hormonal failure is loss of secretion of GH followed by gonadotropins, TSH, and ACTH; this order is mainly seen in patients with tumors and after radiation therapy, while hypopituitarism due to other etiologies may present with a different sequence of deficiency, for example, ACTH deficiency may be the first manifestation in hypophysitis . Importantly, some patients may present with acute onset of pituitary hormone deficiency, while a dangerous presentation is the sudden emergence of ACTH deficiency . Hypopituitarism may be associated with several metabolic and cardiovascular comorbidities, such as hypertension, unfavorable changes in body composition (with increased total and abdominal fat associated with decreased lean mass), and decreased exercise capacity, which may be accompanied by dyslipidemia, insulin resistance, premature atherosclerosis, and cardiac dysfunction . Insulin sensitivity varies greatly among patients with hypopituitarism, depending on numerous factors such as the etiology and treatment of the underlying cause, obesity, severity of pituitary deficiency, and inadequate hormone replacement . Metabolic syndrome, as defined by the National Cholesterol Education Program (NCEP) Adult Treatment Panel III criteria, has been observed in roughly 40% of the adult patients with hypopituitarism, and is associated with an increased risk of diabetes and vascular events . Aside from the symptoms resulting from specific hormone deficiencies in hypopituitarism, the symptoms related to the underlying cause of hypopituitarism can dominate the clinical presentation. This is the case of pituitary tumors that induce visual changes, such as bitemporal hemianopsia due to compression of the optic chiasm, or diplopia due to invasion of the cavernous sinus and involvement of the cranial nerves crossing this sinus. Other symptoms associated with tumor growth and local invasion are headache and cerebrospinal fluid leakage. Also, in the case of functioning pituitary tumors, symptoms resulting from increased hormone secretion may coexist and predominate in the clinical presentation .
Central hypothyroidism (CH) is defined by a decrease in thyroid hormone secretion secondary to insufficient TSH stimulation of a normal thyroid gland . Mechanisms responsible for CH include decreased number of functioning thyrotrophs, decreased synthesis and/or secretion of hypothalamic TRH, decreased TRH release to the pituitary, and decreased biologic activity of TSH. In most cases, CH is associated with other pituitary deficiencies. The signs and symptoms of CH are similar to those present in primary hypothyroidism, including somnolence, tiredness, mild weight gain, cold intolerance, constipation, dry skin, and bradycardia. The main differences of CH compared with primary hypothyroidism are usually the absence of goiter and reduced severity of the symptoms . The laboratory diagnosis of CH is based on low serum free T4 (FT4) levels concomitant with a low or inappropriately normal TSH level . Less commonly, serum TSH may be mildly elevated, usually below 10 IU/L. The measurement of total T4 may replace the measurement of FT4 in the diagnosis of childhood-onset CH, but the accuracy of total T4 is poorer in adulthood-onset CH, as approximately 45% of these patients have total T4 within the normal reference range. Of note, FT4 levels remain within the low-normal range in approximately 18% of the patients with adulthood onset CH . The TRH test and serum T3 levels are not useful for diagnosing CH. In longitudinal follow-up, a decrease of more than 20% in FT4 levels should alert for an increased risk of CH, even when the values remain within the low-normal range . The diagnosis of subclinical CH (normal FT4) is cumbersome and difficult to confirm by serum thyroid function markers. On Doppler echocardiography, abnormal presystolic time measurements have been demonstrated in patients with subclinical CH, but further studies are needed before recommendation of widespread use of this parameter . Other markers of thyroid hormone peripheral actions, such as serum cholesterol, sex–hormone-binding globulin (SHBG), carboxy-terminal telopeptide of type I collagen, osteocalcin (also known as bone gamma-carboxyglutamic acid [Gla]-containing protein), and IL-2 soluble receptor (sIL-2R) have also been used to better define subclinical CH, but thus far, they have not shown good accuracy . The mainstay of hormone replacement in CH is levothyroxine (LT4). Replacement with T3 is currently not recommended due to lack of evidence showing clinical superiority to LT4 replacement and safety . In adult patients, the average dose of LT4 is 1.6 µg/kg/day but varies depending on the duration of the disease, number of additional pituitary deficiencies, and concomitant replacement with GH and/or estrogens . In most cases, LT4 treatment can be started at full dose, except in older individuals or in patients with cardiac or neurological diseases, in whom therapy should be initiated at lower doses and titrated up with caution. In these cases, symptoms and serum FT4 levels should be assessed at 6- to 8-week intervals during titration and every 6–12 months thereafter. Of note, evaluation of the adrenal axis is recommended, and if concomitant ACTH deficiency is present, LT4 replacement should only be initiated after glucocorticoid replacement due to the risk of adrenal crisis . Dose adjustments of LT4 may be necessary during concomitant treatment with GH in children, since GH replacement increases T4 to T3 conversion . Levels of FT4 in the lowest tertile of the reference range have been associated with unfavorable metabolic profile in patients with CH. Thus, targeted serum FT4 levels during LT4 replacement should be into the upper half of the reference range, with blood samples collected before the daily LT4 intake . On the other hand, higher LT4 doses should also be avoided due to the increased risk of thyrotoxicosis symptoms, adrenal crisis, and osteoporosis . Measurement of TSH and T3 levels are usually unnecessary and not recommended during LT4 therapy in CH . However, low/undetectable TSH is expected in patients with CH on adequate thyroid hormone replacement, whereas a high T3 level indicates excessive thyroid hormone replacement requiring dose adjustments. LT4 requirements increase by 20% to 50% during pregnancy in patients with primary hypothyroidism, but patients with CH generally do not require a similar increase due to preservation of the thyroid response to the thyroid-stimulating effect of β-hCG .
Secondary adrenal insufficiency (SAI) may occur due to deficiency of ACTH or CRH in disorders of the pituitary or hypothalamus, resulting in decreased secretion of adrenal cortex steroids, mainly cortisol and dehydroepiandrosterone (DHEA) . SAI is one of the least frequent pituitary function alterations in most patients with hypopituitarism due to pituitary tumors, but it may be the initial presentation in other etiologies, such as in drug-induced hypophysitis . The prevalence of SAI after surgery for treatment of pituitary adenomas varies widely, while this condition may also develop years after radiotherapy . SAI increases the risk of morbidity and mortality in patients with hypopituitarism, since it predisposes patients to adrenal crisis in situations of acute stress and intercurrent illness . SAI manifests with nonspecific symptoms, such as nausea, dizziness, fatigue, anorexia, weight loss, and hypotension, thus requiring a high index of suspicion. Mild ACTH deficiency may only be clinically significant under concurrent stress or illness. The presence of normochromic normocytic anemia, eosinophilia, hyponatremia, hypoglycemia, and eventually hyperkalemia, should serve as a warning sign of increased risk of SAI in a patient with hypopituitarism . Nevertheless, long delays in diagnosing SAI are common, and in many patients, this condition is only established after hospitalization due to adrenal crisis, a potentially life-threatening medical condition requiring immediate emergency treatment. Adrenal crisis should be suspected in patients presenting with acute shock refractory to adequate fluid resuscitation and vasopressors . Estimates of yearly rates in patients with adrenal insufficiency in Europe project an incidence of adrenal crises of 5–10 per 100 patients and a mortality rate of 0.5 per 100 patient-years, corresponding to a number of deaths between 5,000–10,000 . Although a correct and prompt diagnosis of acute SAI is crucial for indicating appropriate therapy and preventing complications, the diagnostic investigation can be challenging in many cases, such as in critically ill patients . Importantly, diagnostic tests should never delay the prompt start of life-saving hydrocortisone treatment in suspected adrenal crisis . In the clinical context of hypopituitarism, the demonstration of morning cortisol levels collected at 8–9 AM lower than 100 nmol/L (3 μg/dL) are strongly predictive of SAI, whereas values greater than 15 μg/dL exclude this diagnosis. Of note, severely ill patients with SAI may present with morning cortisol levels greater than 500 nmol/L (18 μg/dL). Patients with morning cortisol values between 3–15 μg/dL require additional hormonal evaluation ; in recommending such assessment, it is important to consider that normal individuals have morning cortisol values ranging from 6–15 μg/dL by most immunoassays . Despite considered the gold standard for diagnosis of SAI, the insulin tolerance test (ITT) requires adequate site and trained staff and is contraindicated in patients with epilepsy, cardiac arrhythmias, and cerebrovascular diseases. SAI is diagnosed if peak cortisol during ITT is lower than 500 nmol/L (18 μg/dL), although new immunoassay methods standardized against mass spectrometry can show lower concentrations, around 350 mmol/L (12 μg/dL). An alternative to ITT is the glucagon stimulation test (GST), which should be interpreted similarly to the ITT, but exhibits lower sensitivity and specificity . The simplest strategy to assess SAI is the short Synacthen test (or Cosyntropin test), in which serum cortisol levels are measured before and 30 and 60 minutes after stimulation with 250 μg of synthetic ACTH ; a cortisol peak value below 500 nmol/L (< 18 µg/dL) confirms SAI. Unfortunately, Synacthen and Cosyntropin are not easily available in Brazil, hindering the use of these tests in clinical practice. The test may present false-negative results in SAI, especially within 4 weeks from a pituitary insult or surgery or in partial forms of the disease . A low-dose test (1 µg of synthetic ACTH) has been proposed, but has not shown any clear advantages compared with the traditional test . In patients receiving glucocorticoid replacement, the exogenous glucocorticoid may suppress the hypothalamic-pituitary-adrenal axis and interfere with cortisol measurement. Patients using hydrocortisone can undergo serum cortisol assessment 24 hours after the last dose, but in those using prednisone or prednisolone, this period may be longer, requiring individual assessment based on the dose and the time of the last dose. These drugs may cross-react with cortisol measurement assays at different extents, and the duration of their suppression on the hypothalamic-pituitary-adrenal axis lacks consensus in the literature . Moreover, interference of oral estrogen should also be considered in the interpretation of the test results, since oral estrogen can increase serum cortisol levels by 20–50% due to increase in CBG concentrations, but does not affect salivary or urinary free cortisol measurements . There is no consensus in the literature about the diagnosis of complete or partial forms of SAI, which makes the diagnostic scenario even more complex. In this context, clinical judgment in each case is fundamental, and the general rule is to avoid excessive treatment. Plasma ACTH levels have no value in the diagnosis of SAI, and undetectable ACTH alone is unable to establish this diagnosis; this contrasts with a finding of elevated ACTH levels associated with low cortisol levels, which defines the diagnosis of primary adrenal insufficiency (Addison’s disease) . Other hormone alterations in patients with SAI include decreased or undetectable levels of DHEA and S-DHEA, particularly in younger patients. Women with panhypopituitarism and SAI have severe androgen deficiency, although the clinical importance of this finding is still unclear . Once the diagnosis of SAI is confirmed, glucocorticoid replacement with one of the commercially available synthetic derivatives is recommended. Mineralocorticoid therapy is not necessary in SAI . Ideally, hydrocortisone should be the drug of choice for long-term SAI management, but the type of therapy and dosing regimen varies widely worldwide. In an international survey, 80% of the patients reported use of hydrocortisone, followed by prednisone/prednisolone (10%), cortisone acetate (3%), dexamethasone (3%), and other therapies (3%) . The daily dosage of hydrocortisone ranges between 15–20 mg divided into two to three doses; the largest dose should be taken upon awakening, followed by another dose close to lunchtime and, if needed, a third dose in the afternoon no later than 4–6 PM. Hydrocortisone and cortisone acetate are not widely available in Brazil, and the prescription of these salts by compounding pharmacies raises concerns about the origin, preservation, storage, packaging, and bioequivalence of these products and, therefore, are not recommended . Consequently, other formulations such as prednisone and prednisolone are more frequently prescribed. Prednisone is usually administered once daily in the morning, in doses ranging from 2.5–5.0 mg per day. Dose equivalence of hydrocortisone and prednisone can be calculated by the formula 1 mg prednisone = 4 mg hydrocortisone . Regardless of glucocorticoid type, replacement therapy in SAI should start with lower-range dosage regimens with up-titration according to the patient’s clinical status while avoiding overreplacement. During stress or intercurrent illness, patients should be instructed to double or triple the dose until resolution of the underlying condition. Patients, family members, and other health professionals should be educated about the correct adjustment of glucocorticoid dosage and informed on how to identify the precipitating factors and symptoms of adrenal crisis. In addition, patients should carry an identification and emergency card and be encouraged to use medical alert bracelets . In Brazil, educational material and emergency kits for patients with adrenal insufficiency are provided by the Associação Brasileira Addisoniana through their website at www.abaddison.org.br . Modified-release hydrocortisone preparations have been developed to provide a more physiological replacement therapy, replicating the circadian rhythm of normal cortisol secretion . These compounds are formed by a dual-release formulation, consisting of one extended-release core surrounded by an immediate-release coating. One of these drugs, marketed as Plenadren, has been approved in Europe and seems to have some metabolic advantages over the conventional hydrocortisone . However, treatment cost is much higher with these drugs, and they have no perspective of receiving approval in Brazil. There is no specific and accurate biomarker to guide glucocorticoid replacement, which leads clinicians to rely on subjective clinical assessments to adjust therapy. Not surprisingly, a real-world study of glucocorticoid replacement involving patients from the European Adrenal Insufficiency Registry (EU-AIR) identified 25 different regimens being used to deliver a daily hydrocortisone dose of 20 mg in clinical practice . Conflicting evidence has been shown regarding the effects of multiple treatment strategies on adverse health outcomes, such as quality of life (QoL), bone density, metabolic profile, and risk of adrenal crisis or death . However, prolonged use of doses higher than 20 mg of hydrocortisone or equivalent per day must be avoided as they have been associated with unfavorable metabolic profile, Cushing-like morbidities, and increased mortality in patients with hypopituitarism .
Hypogonadotropic hypogonadism (HH), also known as central or secondary hypogonadism, can be congenital or, more often, acquired. A detailed description of isolated congenital forms of HH is beyond the scope of this article (for a review on this topic, see Reference 63). Acquired HH is very common in patients with hypopituitarism due to pituitary adenomas and/or their treatment with surgery or radiotherapy, usually accompanied by other pituitary hormone deficiencies . Anorexia nervosa, excessive exercise, and psychological distress are common causes of acquired HH in women, and medications such as opiates, psychotropic agents, and glucocorticoids, can induce HH in both sexes. In adults, clinical features of HH in men include decreased libido, erectile dysfunction, and infertility, and in women include oligomenorrhea or amenorrhea, decreased libido, dyspareunia, and infertility that can be aggravated by concomitant hyperprolactinemia and adrenal/ovarian androgen deficiency . Prolonged estrogen deficiency may cause regression of secondary sexual characteristics, urinary symptoms, reduction of muscle mass and bone density, and dyslipidemia. Similarly, prolonged testosterone deficiency has been associated with an increased risk of metabolic disorders . Untreated HH is an independent factor related to high mortality due to vascular complications in patients with hypopituitarism, particularly women, and adequate replacement with sex steroids has been shown to normalize or reduce mortality rates in these patients . In women, amenorrhea in the clinical context of a hypothalamic-pituitary disease indicates the presence of HH, which is confirmed by low estradiol levels associated with low or inappropriately normal FSH/LH levels . Similarly, low total testosterone levels associated with low or inappropriately normal FSH/LH levels establish the diagnosis of HH in men. Measurement of total testosterone should be performed before 10 AM after overnight fasting, in the absence of acute/subacute illness and medications (glucocorticoids, opiates, ketoconazole, barbiturates, cocaine, etc.) known to affect testosterone levels. Two measurements with the same type of assay are necessary to establish the diagnosis of low testosterone in cases of borderline values. There is no indication for stimulation tests . Sex steroid replacement therapy Treatment of male HH involves primarily testosterone replacement, with the goal of improving sexual function, libido, energy levels, bone mineral density (BMD), sense of well-being, muscle mass, and strength. Current guidelines recommend the use of minimal doses necessary to maintain testosterone levels in the range of 450–600 ng/dL . Different testosterone formulations are available, and an individualized therapeutic approach is needed considering effectiveness, patient compliance and preference, drug cost and availability, and potential side effects . Long-acting testosterone undecanoate is often considered the best option among injectable preparations due to its more convenient administration (intramuscular, every 10 to 14 weeks). This regimen usually maintains plasma testosterone levels in the expected range with minor fluctuations . Intramuscular formulations containing testosterone enanthate, cypionate, or the combination of four esters (propionate, phenylpropionate, isocaproate, and decanoate) are usually administered every 2 to 3 weeks, and are efficacious and generally well tolerated . Although these formulations are much less expensive than testosterone undecanoate, they are associated with supraphysiologic peak values shortly after the injection and to subphysiologic levels on the days before the new injection. This often leads to fluctuations in symptoms, mood swings, and emotional instability . Transdermal gel formulations are commonly prescribed, as they provide flexibility of dosing, ease of application, good skin tolerability, and possibly less erythrocytosis than injectable testosterone . The disadvantages of transdermal gel formulations include the potential of transferring of testosterone to women or children by direct skin-to-skin contact, and skin irritation in a small proportion of treated men . Testosterone undecanoate is the only safe testosterone ester for oral therapy, since it is well tolerated and not hepatotoxic, but it must be taken twice daily with fatty meals . It has been approved in the US market, but it is still not available in Brazil. Additionally, a nasal gel formulation and a subcutaneous testosterone enanthate formulation for weekly administration have been recently approved by the US Food and Drug Administration (FDA) . Testosterone therapy should not be started in men with breast or prostate cancer, palpable prostate nodule or induration, and prostate-specific antigen (PSA) > 4 ng/mL or > 3 ng/mL in those at increased risk of prostate cancer (e.g., men with a first-degree relative with diagnosed prostate cancer) without further urological evaluation, elevated hematocrit, untreated severe obstructive sleep apnea, severe lower urinary tract symptoms, uncontrolled heart failure, myocardial infarction or stroke within the last 6 months, or thrombophilia. After starting testosterone therapy, patients should be monitored with measurement of hematocrit and PSA levels in the first year of treatment to investigate possible side effects . Sex steroid replacement should be started in all women with HH who are at premenopausal age and be individualized for older women . Noteworthy, hypopituitarism removes the natural survival advantage that women have over men in terms of vascular complications; therefore, outcomes observed in natural menopause should not be extrapolated to women with hypopituitarism presenting with frank hypogonadism . In this regard, data of 203,767 postmenopausal women from 10 observational studies included in the International Collaboration for a Life Course Approach to Reproductive Health and Chronic Disease Events (InterLACE) indicated that primary hypogonadism due to surgical menopause was associated with over 20% higher risk of cardiovascular disease compared with natural menopause, with an increased risk in women younger than 40 years. Women who experienced surgical menopause at an earlier age (<50 years) and received hormone replacement therapy had a lower risk of incident heart disease than those who were not treated . Assuming that untreated FSH/LH insufficiency may have a negative rather than a beneficial effect on survival, it is unfortunate that HH is significantly more frequently underdiagnosed and undertreated in women compared with men . While the goal of treatment of HH in older women is to control hot flushes, in young women, the goal is to prevent damage induced by estrogen deficiency. Adult women with HH should receive combined estrogen-progestogen regimen or unopposed estrogens (if prior hysterectomy) for improving symptoms of hypoestrogenism such as vaginal atrophy, dysuria, dry skin/hair, hot flushes, and night sweats, and to reduce the risk of cardiovascular disease, osteopenia, and mortality . The choice of estrogen and progestin formulations, route (oral or transdermal), and regimen depends on the risk of adverse effects, patient’s convenience/preference, and cost. Transdermal administration of estrogens either by patch or gel provides more stable plasma concentrations of estradiol and reduces its conversion to estrone compared with oral formulations. By eliminating the first pass through the liver, transdermal estrogen prevents undesirable increases in coagulation factors, blood pressure, and serum triglycerides. In addition, by interfering less with SHBG levels, transdermal estrogen allows a greater fraction of hormone to circulate in free and biologically active forms . The usual doses of transdermal estrogen are 50–100 µg of estradiol daily, applied as a patch twice weekly, or 1.5 mg daily of estradiol hemihydrate gel. The addition of progestogens to the patch does not affect the absorption of estrogen. Some application systems use patches containing varying amounts of estradiol that mimic the physiological variations of estradiol levels throughout the menstrual cycle. Some patients may develop allergy at the patch application site or exhibit a drop in estrogen levels earlier than expected . Furthermore, the transdermal route of estrogen administration does not block the hepatic generation of IGF-1 induced by GH . For use in oral preparations, natural estradiol such as 17-beta-estradiol should be micronized or administered in derivative compounds such as ethinyl estradiol, or estradiol valerate, enanthate, or cypionate, to improve absorption. Benefits of oral estradiol therapy include relatively low cost, long accumulated experience, convenience, increased HDL-cholesterol, and reduced LDL-cholesterol . Another alternative to oral estradiol replacement are conjugated estrogens, a combination of estrogens obtained from urine of pregnant mares. The induction of protein synthesis in the liver by conjugated estrogens is even higher than that observed with the use of estradiol formulations . Daily doses of estrogen in adult women with HH range from 1.0 to 2.0 mg of micronized estradiol or an equivalent dose of other estrogens. In general, these doses lead to clinical improvement and serum estradiol levels between 30–50 pg/mL, which correspond to the levels observed in the early follicular phase . A Brazilian study evaluating bone mass in patients with premature ovarian failure has shown that the doses of estrogen normally used in postmenopause were not adequate to reduce impaired spinal and femoral bone mass in younger women , although no consensus in the literature has established the required estrogen dose in this age group. Cyclic progesterone replacement combined with continuous estrogen prevents endometrial hyperplasia and induces menstruation. In this regimen, 2.5 to 10 mg of medroxyprogesterone acetate (MPA) or synthetic progestogenic derivative is added during 7 to 10 days of the month. Alternatively, estrogens and progestogens can be used simultaneously and continuously, which induces a higher frequency of amenorrhea . Alternatives to MPA are micronized progesterone (100–200 mg daily) and progestogens such as norethindrone (0.35 mg), gestodene (0.75 mg), or levonorgestrel (0.075 mg), which exhibit different affinities for the progesterone, testosterone, estradiol, and aldosterone receptors, and consequently have variable side effects . Endometrial proliferation should be controlled periodically by transvaginal ultrasonography. If endometrial thickening greater than 5 mm is observed, an endometrial biopsy should be performed . Combined estrogen-progestin contraceptive pill may be more acceptable for younger women, but studies comparing the effects of different dose regimens in HH are lacking . After the age of physiological menopause, the prescription of estrogens should be based on current postmenopausal guidelines. Nevertheless, a recent systematic review comprising 2,588,327 postmenopausal women from six clinical trials and 27 prospective observational studies has observed that doses of 0.3–0.625 mg/day of oral conjugated equine estrogen or equivalent have been associated with cardioprotective effects, while higher doses increase the risk of venous thromboembolism and stroke in a dose-dependent manner. The authors concluded that hormone therapy should be used in the lowest effective dose to prevent adverse cardiovascular effects and that the dose should be reduced with advancing age . They also observed that transdermal estrogen may be safer in relation to cardiovascular and thrombotic risk than oral estrogen and that preparations with MPA are associated with increased thrombotic risk . Women with HH have androgen deficiency. Although few studies have shown advantages with treatment , recent guidelines recommended against the routine use of DHEA and testosterone in women due to limited data concerning efficacy and safety . An intravaginal form of DHEA has been approved by the FDA to treat genitourinary syndrome of menopause, but this formulation is not available in Brazil . Moreover, testosterone has been carefully suggested in the guidelines to treat selected women with hypoactive sexual desire disorder . In this clinical context, which might be present in women with hypopituitarism, a trial of low-dose testosterone therapy may be considered, which seems to be safe in the short term. However, efficacy is variable and long-term effects on cardiovascular risk and breast cancer incidence are not known. Undesirable side effects, such as alopecia, changes in vocal timbre, hirsutism, clitoromegaly, acne, elevated hematocrit, and liver function and lipid profile abnormalities, must be balanced against the potential benefits . Fertility issues Fertility issues in patients harboring aggressive pituitary tumors have been recently revisited . Importantly, fertility preservation strategies need to be discussed with the patient desiring conception before or during treatment. Additional pituitary deficiencies should be appropriately treated, particularly GH deficiency, which has been associated in some studies with poor pregnancy rates in patients with hypopituitarism . In men and women with HH, the approach to restoring fertility may include clomiphene citrate, human chorionic gonadotropin (hCG), human menopausal gonadotropin, or purified or recombinant FSH . Pulsatile GnRH administration via mini-infusion pump is only delivered in some research settings worldwide . In men with fertility issues, testosterone replacement treatment should be discontinued. In men with prolactinomas and persistent hypogonadism under dopamine agonist therapy, clomiphene restores normal testosterone levels and improves sperm motility, independent of prolactin levels . In contrast, clomiphene treatment fails to restore normal testosterone levels in men with hypogonadism harboring nonfunctioning pituitary adenomas . A study reported that 67% of men with HH and acromegaly treated with clomiphene achieved normal testosterone levels, but fertility restoration was not evaluated . Administration of hCG is done intramuscularly or preferably subcutaneously in the thigh at an initial dose of 3,000–5,000 IU per week, divided into at least two injections to ensure relatively stable serum levels . Serum testosterone concentration is measured every 2 or 3 months, and the dose should be increased if the values are not between 400 and 800 ng/dL. Very rarely, serum testosterone concentration fails to respond to hCG, even at a very high dose, due to antibodies to hCG . If the sperm count does not reach 5–10 million/mL and/or pregnancy has not occurred within 6 months after serum testosterone is between 400–800 ng/dL, purified or recombinant FSH should be added to the regimen with a starting dose of 75–150 IU every other day (or three times per week). This dose is adjusted upwards if necessary to achieve serum FSH concentration in the physiologic range of 4–8 IU/L . These regimens induce testicular growth in almost all patients, spermatogenesis in approximately 80%, and pregnancy in 50% of the cases after 12–24 months of treatment . In women, clomiphene citrate may induce ovulation in some infertile patients. Therapy with gonadotropin initiates in the follicular phase of the menstrual cycle with FSH, followed by administration of hCG. Ovulation is expected between 36 and 48 hours after hCG injection. If fertilization occurs, progesterone is administered to ensure implantation and is maintained until approximately 10 weeks of gestation, when placental hormone production becomes sufficient. Assisted reproduction technologies are alternative options when therapy with gonadotropins fails . In a recent systematic literature review, the authors suggested that assisted fertility in women hypopituitarism has a good outcome both in terms of achieving pregnancy and infant outcome .
Treatment of male HH involves primarily testosterone replacement, with the goal of improving sexual function, libido, energy levels, bone mineral density (BMD), sense of well-being, muscle mass, and strength. Current guidelines recommend the use of minimal doses necessary to maintain testosterone levels in the range of 450–600 ng/dL . Different testosterone formulations are available, and an individualized therapeutic approach is needed considering effectiveness, patient compliance and preference, drug cost and availability, and potential side effects . Long-acting testosterone undecanoate is often considered the best option among injectable preparations due to its more convenient administration (intramuscular, every 10 to 14 weeks). This regimen usually maintains plasma testosterone levels in the expected range with minor fluctuations . Intramuscular formulations containing testosterone enanthate, cypionate, or the combination of four esters (propionate, phenylpropionate, isocaproate, and decanoate) are usually administered every 2 to 3 weeks, and are efficacious and generally well tolerated . Although these formulations are much less expensive than testosterone undecanoate, they are associated with supraphysiologic peak values shortly after the injection and to subphysiologic levels on the days before the new injection. This often leads to fluctuations in symptoms, mood swings, and emotional instability . Transdermal gel formulations are commonly prescribed, as they provide flexibility of dosing, ease of application, good skin tolerability, and possibly less erythrocytosis than injectable testosterone . The disadvantages of transdermal gel formulations include the potential of transferring of testosterone to women or children by direct skin-to-skin contact, and skin irritation in a small proportion of treated men . Testosterone undecanoate is the only safe testosterone ester for oral therapy, since it is well tolerated and not hepatotoxic, but it must be taken twice daily with fatty meals . It has been approved in the US market, but it is still not available in Brazil. Additionally, a nasal gel formulation and a subcutaneous testosterone enanthate formulation for weekly administration have been recently approved by the US Food and Drug Administration (FDA) . Testosterone therapy should not be started in men with breast or prostate cancer, palpable prostate nodule or induration, and prostate-specific antigen (PSA) > 4 ng/mL or > 3 ng/mL in those at increased risk of prostate cancer (e.g., men with a first-degree relative with diagnosed prostate cancer) without further urological evaluation, elevated hematocrit, untreated severe obstructive sleep apnea, severe lower urinary tract symptoms, uncontrolled heart failure, myocardial infarction or stroke within the last 6 months, or thrombophilia. After starting testosterone therapy, patients should be monitored with measurement of hematocrit and PSA levels in the first year of treatment to investigate possible side effects . Sex steroid replacement should be started in all women with HH who are at premenopausal age and be individualized for older women . Noteworthy, hypopituitarism removes the natural survival advantage that women have over men in terms of vascular complications; therefore, outcomes observed in natural menopause should not be extrapolated to women with hypopituitarism presenting with frank hypogonadism . In this regard, data of 203,767 postmenopausal women from 10 observational studies included in the International Collaboration for a Life Course Approach to Reproductive Health and Chronic Disease Events (InterLACE) indicated that primary hypogonadism due to surgical menopause was associated with over 20% higher risk of cardiovascular disease compared with natural menopause, with an increased risk in women younger than 40 years. Women who experienced surgical menopause at an earlier age (<50 years) and received hormone replacement therapy had a lower risk of incident heart disease than those who were not treated . Assuming that untreated FSH/LH insufficiency may have a negative rather than a beneficial effect on survival, it is unfortunate that HH is significantly more frequently underdiagnosed and undertreated in women compared with men . While the goal of treatment of HH in older women is to control hot flushes, in young women, the goal is to prevent damage induced by estrogen deficiency. Adult women with HH should receive combined estrogen-progestogen regimen or unopposed estrogens (if prior hysterectomy) for improving symptoms of hypoestrogenism such as vaginal atrophy, dysuria, dry skin/hair, hot flushes, and night sweats, and to reduce the risk of cardiovascular disease, osteopenia, and mortality . The choice of estrogen and progestin formulations, route (oral or transdermal), and regimen depends on the risk of adverse effects, patient’s convenience/preference, and cost. Transdermal administration of estrogens either by patch or gel provides more stable plasma concentrations of estradiol and reduces its conversion to estrone compared with oral formulations. By eliminating the first pass through the liver, transdermal estrogen prevents undesirable increases in coagulation factors, blood pressure, and serum triglycerides. In addition, by interfering less with SHBG levels, transdermal estrogen allows a greater fraction of hormone to circulate in free and biologically active forms . The usual doses of transdermal estrogen are 50–100 µg of estradiol daily, applied as a patch twice weekly, or 1.5 mg daily of estradiol hemihydrate gel. The addition of progestogens to the patch does not affect the absorption of estrogen. Some application systems use patches containing varying amounts of estradiol that mimic the physiological variations of estradiol levels throughout the menstrual cycle. Some patients may develop allergy at the patch application site or exhibit a drop in estrogen levels earlier than expected . Furthermore, the transdermal route of estrogen administration does not block the hepatic generation of IGF-1 induced by GH . For use in oral preparations, natural estradiol such as 17-beta-estradiol should be micronized or administered in derivative compounds such as ethinyl estradiol, or estradiol valerate, enanthate, or cypionate, to improve absorption. Benefits of oral estradiol therapy include relatively low cost, long accumulated experience, convenience, increased HDL-cholesterol, and reduced LDL-cholesterol . Another alternative to oral estradiol replacement are conjugated estrogens, a combination of estrogens obtained from urine of pregnant mares. The induction of protein synthesis in the liver by conjugated estrogens is even higher than that observed with the use of estradiol formulations . Daily doses of estrogen in adult women with HH range from 1.0 to 2.0 mg of micronized estradiol or an equivalent dose of other estrogens. In general, these doses lead to clinical improvement and serum estradiol levels between 30–50 pg/mL, which correspond to the levels observed in the early follicular phase . A Brazilian study evaluating bone mass in patients with premature ovarian failure has shown that the doses of estrogen normally used in postmenopause were not adequate to reduce impaired spinal and femoral bone mass in younger women , although no consensus in the literature has established the required estrogen dose in this age group. Cyclic progesterone replacement combined with continuous estrogen prevents endometrial hyperplasia and induces menstruation. In this regimen, 2.5 to 10 mg of medroxyprogesterone acetate (MPA) or synthetic progestogenic derivative is added during 7 to 10 days of the month. Alternatively, estrogens and progestogens can be used simultaneously and continuously, which induces a higher frequency of amenorrhea . Alternatives to MPA are micronized progesterone (100–200 mg daily) and progestogens such as norethindrone (0.35 mg), gestodene (0.75 mg), or levonorgestrel (0.075 mg), which exhibit different affinities for the progesterone, testosterone, estradiol, and aldosterone receptors, and consequently have variable side effects . Endometrial proliferation should be controlled periodically by transvaginal ultrasonography. If endometrial thickening greater than 5 mm is observed, an endometrial biopsy should be performed . Combined estrogen-progestin contraceptive pill may be more acceptable for younger women, but studies comparing the effects of different dose regimens in HH are lacking . After the age of physiological menopause, the prescription of estrogens should be based on current postmenopausal guidelines. Nevertheless, a recent systematic review comprising 2,588,327 postmenopausal women from six clinical trials and 27 prospective observational studies has observed that doses of 0.3–0.625 mg/day of oral conjugated equine estrogen or equivalent have been associated with cardioprotective effects, while higher doses increase the risk of venous thromboembolism and stroke in a dose-dependent manner. The authors concluded that hormone therapy should be used in the lowest effective dose to prevent adverse cardiovascular effects and that the dose should be reduced with advancing age . They also observed that transdermal estrogen may be safer in relation to cardiovascular and thrombotic risk than oral estrogen and that preparations with MPA are associated with increased thrombotic risk . Women with HH have androgen deficiency. Although few studies have shown advantages with treatment , recent guidelines recommended against the routine use of DHEA and testosterone in women due to limited data concerning efficacy and safety . An intravaginal form of DHEA has been approved by the FDA to treat genitourinary syndrome of menopause, but this formulation is not available in Brazil . Moreover, testosterone has been carefully suggested in the guidelines to treat selected women with hypoactive sexual desire disorder . In this clinical context, which might be present in women with hypopituitarism, a trial of low-dose testosterone therapy may be considered, which seems to be safe in the short term. However, efficacy is variable and long-term effects on cardiovascular risk and breast cancer incidence are not known. Undesirable side effects, such as alopecia, changes in vocal timbre, hirsutism, clitoromegaly, acne, elevated hematocrit, and liver function and lipid profile abnormalities, must be balanced against the potential benefits .
Fertility issues in patients harboring aggressive pituitary tumors have been recently revisited . Importantly, fertility preservation strategies need to be discussed with the patient desiring conception before or during treatment. Additional pituitary deficiencies should be appropriately treated, particularly GH deficiency, which has been associated in some studies with poor pregnancy rates in patients with hypopituitarism . In men and women with HH, the approach to restoring fertility may include clomiphene citrate, human chorionic gonadotropin (hCG), human menopausal gonadotropin, or purified or recombinant FSH . Pulsatile GnRH administration via mini-infusion pump is only delivered in some research settings worldwide . In men with fertility issues, testosterone replacement treatment should be discontinued. In men with prolactinomas and persistent hypogonadism under dopamine agonist therapy, clomiphene restores normal testosterone levels and improves sperm motility, independent of prolactin levels . In contrast, clomiphene treatment fails to restore normal testosterone levels in men with hypogonadism harboring nonfunctioning pituitary adenomas . A study reported that 67% of men with HH and acromegaly treated with clomiphene achieved normal testosterone levels, but fertility restoration was not evaluated . Administration of hCG is done intramuscularly or preferably subcutaneously in the thigh at an initial dose of 3,000–5,000 IU per week, divided into at least two injections to ensure relatively stable serum levels . Serum testosterone concentration is measured every 2 or 3 months, and the dose should be increased if the values are not between 400 and 800 ng/dL. Very rarely, serum testosterone concentration fails to respond to hCG, even at a very high dose, due to antibodies to hCG . If the sperm count does not reach 5–10 million/mL and/or pregnancy has not occurred within 6 months after serum testosterone is between 400–800 ng/dL, purified or recombinant FSH should be added to the regimen with a starting dose of 75–150 IU every other day (or three times per week). This dose is adjusted upwards if necessary to achieve serum FSH concentration in the physiologic range of 4–8 IU/L . These regimens induce testicular growth in almost all patients, spermatogenesis in approximately 80%, and pregnancy in 50% of the cases after 12–24 months of treatment . In women, clomiphene citrate may induce ovulation in some infertile patients. Therapy with gonadotropin initiates in the follicular phase of the menstrual cycle with FSH, followed by administration of hCG. Ovulation is expected between 36 and 48 hours after hCG injection. If fertilization occurs, progesterone is administered to ensure implantation and is maintained until approximately 10 weeks of gestation, when placental hormone production becomes sufficient. Assisted reproduction technologies are alternative options when therapy with gonadotropins fails . In a recent systematic literature review, the authors suggested that assisted fertility in women hypopituitarism has a good outcome both in terms of achieving pregnancy and infant outcome .
A nationwide study carried out in Denmark has reported an annual incidence of adult GH deficiency (AGHD) of 17.6 per 1,000,000 people, with higher rates in men . Applying these numbers to the Brazilian population, we would expect approximately 3,700 new cases of AGHD yearly in our country. According to the age at onset, patients may present with childhood-onset (CO) AGHD or adulthood-onset AGHD . Isolated idiopathic GHD diagnosed in children based on biochemical GH tests must be reevaluated during transition into adult life. A Brazilian study reported that only 31% of the children with GHD persisted with GHD when retested, with an IGF-I cutoff value of 110 ng/mL presenting 94.5% sensitivity and 100% specificity for diagnosis of AGHD in the transition period . Noteworthy, short children treated with GH for non-GHD pediatric indications (e.g., Turner syndrome, idiopathic short stature, small for gestational age, etc.) and adults with functional GHD (obesity, metabolic syndrome, aging) without evidence of hypothalamic-pituitary disease must not be investigated for AGHD . The clinical presentation of AGHD is characterized by multiple, nonspecific clinical findings involving changes in body composition, increased fracture risk, altered physical capacity, presence of cardiovascular risk factors, and poor QoL . These manifestations vary according to age, time of onset, and presence of other hormone deficiencies. For instance, low bone mass is more frequent in young adults with GHD, while cardiovascular risk factors are more common among older patients with GHD . In the United Kingdom, perceived impairment of QoL is required to initiate GH replacement in adults, while lack of improvement in QoL is used for discontinuation of GH therapy . QoL can be evaluated by general questionnaires or by the specific questionnaire Quality of Life Assessment of Growth Hormone Deficiency in Adults, which has been translated into Brazilian Portuguese and validated for the Brazilian population . In an appropriate clinical context, the diagnosis of AGHD must be biochemically confirmed by serum IGF-I measurements alone or in association with GH stimulating tests. Basal serum GH levels have no diagnostic value due to the pulsatile nature of pituitary GH secretion. ITT is considered the gold-standard test, and severe GHD is diagnosed when the peak GH value is below 3 µg/L or 5 µg/L in the transition period . As mentioned before, ITT requires medical supervision and is contraindicated in elderly individuals and in patients with cardiac or neurologic diseases. An alternative to ITT is the GST, with a cutoff value of 3 µg/L recommended to discriminate between normal response and severe AGHD, or 1 µg/L in overweight/obese subjects . Macimorelin acetate is an orally active ghrelin mimetic that binds to the GH secretagogues receptors and stimulates GH secretion. The macimorelin test for AGHD has a sensitivity and specificity of 92% and 96%, respectively, with a diagnostic accuracy similar to that of ITT, without risk of hypoglycemia and less false-positive results. However, macimorelin acetate is an expensive medication and is not available in Brazil . Other GH tests commonly used in children to diagnose GHD are not recommended in adults due to very low accuracy . In patients with a high probability of AGHD, such as those with multiple pituitary hormone deficiencies associated with well-documented structural lesions in the hypothalamic-pituitary region, a low serum IGF-I level (below the lower limit of reference for age and sex) is enough to diagnose AGHD and precludes the need for a provocative GH test . However, factors interfering with IGF-I measurement should be considered, such as the use of oral estrogens, uncontrolled hypothyroidism, malnutrition, hepatic insufficiency, and uncontrolled diabetes, since these conditions may lead to false results. On the other hand, it is important to emphasize that normal IGF-I levels do not rule out the diagnosis of AGHD ( and ). GH replacement therapy is indicated to all patients with a diagnosis of AGHD for whom treatment is intended . The main contraindications to GH therapy include active malignancy, uncontrolled diabetes, diabetic retinopathy, and intracranial hypertension . In Brazil, the Ministry of Health Ordinance number 28 of November 30, 2018, contemplates the “Clinical Protocol and Therapeutic Guidelines for Hypopituitarism”, a document that regulates in the public health care system (SUS) the therapeutic use of GH in children and adults . In contrast with GH treatment in children, GH replacement doses in adults are not calculated based on body weight, because this approach results in a high frequency of adverse events, mainly related to fluid retention, including paresthesias, joint stiffness, peripheral edema, arthralgia, myalgia, and carpal tunnel syndrome. Thus, the starting dose of GH is 0.1 mg for individuals 60 years or older, 0.2 mg for young men, 0.3 mg for young women, and 0.4–0.7 mg in the transition period, administered in daily subcutaneous injections at bedtime. The dose should be titrated every month until attaining IGF-I levels between the median and upper limit of normal for age . The maintenance dose generally varies between 0.4–0.6 mg/day, with a large individual variability depending on sex, age, body mass index (BMI), and route of estrogen replacement in women, among other factors . SAI and CH can develop or worsen during GH therapy; therefore, cortisol and FT4 should be closely monitored with proper adjustments in glucocorticoid and LT4 replacement doses . Anthropometry (BMI, waist circumference), blood pressure, and QoL should be assessed at each visit during follow-up for dose titration and every 6 months thereafter. Lipid profile, glycemia, and hemoglobin A1C can be evaluated every 6–12 months, while evaluation of body composition and BMD by dual-energy x-ray absorptiometry (DXA) and cardiovascular parameters is recommended yearly or every 2 years . GH replacement therapy improves or normalizes several abnormalities related to AGHD and may promote health benefits in patients with hypopituitarism . In the transition period, continuation or reintroduction of GH therapy after attaining final height promotes full somatic development of bone and muscles and prevents the development of abnormal metabolic features observed in older patients with AGHD . Meta-analyses of placebo-controlled trials have shown that GH treatment in adults reduces total and visceral fat mass, increases lean body mass and bone mass, improves cardiovascular risk markers (diastolic blood pressure, total cholesterol, and LDL-cholesterol), and increases anaerobic and exercise capacity . However, individual responses to GH therapy concerning different therapeutic endpoints vary greatly, and not all patients show the same pattern of improvement . It is still unclear whether GH replacement improves QoL, but in cohorts of AGHD followed up for up to 10 years, sustained improvement in QoL scores toward normal values has been demonstrated, especially in women and patients with low QoL at baseline . GH replacement is generally safe and well tolerated in adult patients with hypopituitarism. However, special attention should be given to older, heavier, and female AGHD patients, as well as to those with a family history of type 2 diabetes, because they are more susceptible to adverse events, especially related to glucose homeostasis . In these individuals, close monitoring with fasting blood glucose and hemoglobin A1C is recommended during GH therapy. GH therapy should be initiated when pituitary tumor growth is under control. In this context, there is no evidence that long-term GH replacement in adults affects the progression of pituitary tumors. Additionally, there is no evidence of tumor recurrence or increased risk of neoplasia in AGHD; therefore, cancer surveillance in adult patients on GH therapy should follow the same standard practice as those in the general population . The impact of untreated GHD on increasing mortality in patients with hypopituitarism is unclear. GH therapy does not change the prevalence of metabolic syndrome over time but, intriguingly, has been associated with reduction or even normalization of mortality in AGHD . A possible reason for this apparent paradox is the positive GH effect on endothelial and cardiovascular parameters . Analysis of a Brazilian kindred with congenital isolated GHD due to a homozygous mutation in the GH-releasing hormone (GHRH) receptor gene showed no serious deleterious effect of congenital isolated GHD on risks of vascular events and survival . However, this finding should not be extended to patients with hypopituitarism and AGHD, who present a distinct phenotype . Unfortunately, randomized clinical trials evaluating the effect of GH therapy on mortality in AGHD are unavailable. Thus, the primary goal of GH replacement therapy should be oriented to improve the associated morbidities in the short and long term.
Pituitary hypofunction usually requires replacement of several hormones, and it is essential to understand the main interactions between the different pituitary axes during hormone replacement to optimize the therapeutic outcomes. Prescription of thyroid hormones without previous replacement of glucocorticoids may precipitate an adrenal crisis in patients with ACTH deficiency. The mechanisms involved in cortisol reduction after LT4 treatment are not completely understood but may involve accelerated cortisol catabolism induced by an increase in renal 11β-hydroxysteroid dehydrogenase type 2 or increased urinary cortisol clearance due to augmented urinary flow caused by hypothyroidism treatment . Therefore, a careful investigation of SAI should be performed before initiating thyroid hormone therapy in patients with hypopituitarism, while the daily glucocorticoid dose may require adjustment in patients on glucocorticoid replacement who are started on LT4 treatment. Similarly, cortisol status must be monitored before and during GH replacement therapy, since GH reduces the activity of the 11β-hydroxysteroid dehydrogenase type 1 in the liver and adipose tissue, an enzyme that converts inactive cortisone to active cortisol . Consequently, GH therapy can unmask subclinical SAI, requiring the initiation or adequation of glucocorticoid replacement . In addition, patients replaced with GH can develop a variable reduction in FT4 levels and a slight increase in free T3 levels . The underlying mechanisms likely involve enhancement of peripheral deiodination of T4 to T3 due to activation of deiodinase type 1 by GH, reduction of residual TSH secretion due to an increase in somatostatinergic tonus determined by IGF-I action on hypothalamus, and/or reduction of residual TSH secretion due T3 feedback on the pituitary . Accordingly, FT4 levels should be evaluated 6 weeks after starting GH therapy, and LT4 initiation or dose adjustment may be necessary in patients with established CH to maintain FT4 levels in the adequate range . Estrogen therapy may increase thyroxin binding protein (TBG) levels, reducing FT4 levels ; consequently, the mean LT4 dose is usually higher in women on estrogen therapy than in those not taking estrogen . As previously mentioned, oral estrogen may antagonize GH action in the liver. This effect is induced by the expression of an intracellular protein named SOCS-2, which blocks GH signaling and reduces IGF-I generation . In addition, oral estrogens can interfere with IGFBPs levels during GH treatment. . Like oral estrogens, selective estrogen receptor modulators (SERMs) can also interfere with the hepatic IGF-I production . Thus, as previously mentioned, women with hypopituitarism and on GH replacement should receive estrogen via transdermal route. Taken together, all these interactions highlight the importance of constant clinical and laboratory monitoring of patients with hypopituitarism on replacement therapies to improve compliance, make appropriate adjustments, and optimize therapeutic outcomes .
Management of hypopituitarism is one of the greatest challenges in clinical endocrinology, and several fundamental aspects for adequate control of this condition are presented in this document. Patients and family members must be educated to understand the particularities of hypopituitarism treatment and hormonal interactions. This is particularly important during clinical complications, surgeries, and any procedure that the patient may require. Physicians working in other specialties must also become knowledgeable in hormone replacement to be able to manage patients with hypopituitarism as needed.
|
Choosing Wisely for Thyroid Conditions: Recommendations of the Thyroid Department of the Brazilian Society of Endocrinology and Metabolism | d82088ec-ca5b-465a-b7ac-82f87731ec59 | 10065321 | Physiology[mh] | The overuse of low-value interventions is a global problem. A growing body of knowledge highlights that it leads to a waste of precious health care resources and poses patients at risk of harm . Choosing Wisely is a worldwide health care professional-led initiative that aims to advance the dialogue between physicians and patients about low-value health interventions, avoiding wasteful or unnecessary medical tests, treatments, and procedures . The Choosing Wisely initiative started in the United States in 2012 led by the American Board of Internal Medicine (ABIM) and is now present in many countries, involving more than one hundred medical associations/societies, hospitals, and universities. Given that thyroid conditions are widespread and that low-value interventions for these conditions are frequent in clinical practice, we aimed to develop a list of recommendations of measures not to be done when approaching patients with thyroid conditions. This article aims to summarize the top five recommendations in the list developed by the Thyroid Department of the Brazilian Society of Endocrinology and Metabolism (SBEM) for the Choosing Wisely project.
In 2017 the Thyroid Department of the Brazilian Society of Endocrinology and Metabolism (SBEM) and the Choosing Wisely Brasil agreed to work towards the development of a list of Choosing Wisely recommendations for thyroid conditions. The Thyroid Department named a Chair (JMD) and a Co-chair (ALM) for the initiative. Additional members of SBEM, representing thyroid groups from 10 tertiary care, teaching-based Brazilian institutions, were invited to compose the Task Force. Of note, all members of the Task Force were clinical thyroid specialists, not surgeons. The Task Force work consisted of 3 phases: brainstorming and ideation was conducted electronically through a Delphi approach in which all Task Force members were invited to submit proposals of recommendations; all proposals of recommendations were reviewed, and duplicates removed; the Task Force graded each recommendation on a 0 (lowest agreement) to 10 (highest agreement) scale. The 10 recommendations that received the highest scores by the Task Force were submitted for voting by e-mail to all SBEM associates, who graded each recommendation on the 0 (lowest agreement) to 10 (highest agreement) scale. A final list of the 5 recommendations that received the highest scores by the SBEM associates composes the Choosing Wisely – Thyroid SBEM recommendations.
The Task Force was composed of 14 thyroidologists from 10 different Brazilian institutions. The brainstorming and ideation phase resulted in 69 recommendations, thus each member of the Task Force contributed with a median of 3 (min.-max. 1-11) recommendations. After the removal of duplicates and recommendations that did not adhere to the scope of the initiative, 35 recommendations remained. Then the 14 members of the Task Force graded each recommendation. The mean grades for the 35 recommendations ranged from 9.4 ± 1.2 to 6.0 ± 2.3. The 10 recommendations that received the highest scores by the Task Force were then submitted to all SBEM associates by e-mail . A total of 683 associates graded each recommendation electronically, attributing a grade (0 to 10) for each recommendation. The grades for the 10 recommendations ranged from 9.8 ± 1.2 to 7.6 ± 2.7, and the 5 recommendations that received the highest scores by the SBEM associates compose our final list of recommendations . The Portuguese version of the 5 recommendations is available as .
In this report, we described the development of the Thyroid Department of the Brazilian Society of Endocrinology and Metabolism (SBEM) Choosing Wisely initiative list. A set of recommendations to avoid unnecessary medical tests, treatments, or procedures for thyroid conditions is offered with a transparent methodology. This initiative aims to foster productive interactions between physicians and patients, stimulating shared decision-making and elevating the standards of care for thyroid conditions.
|
First osteological evidence of severed hands in Ancient Egypt | 0c35c732-e3fb-49b5-b136-aed4e3a3fe80 | 10066219 | Anatomy[mh] | The reliability of information must always be questioned, both now and in the past. The more sources available to confirm a story, the more likely it is to be true. Unfortunately, in history and even more so in prehistory, we are limited to a few sources, sometimes only to one. Much information on the lives, habits, and history of the Ancient Egyptians is depicted on temple and tomb walls, as well as recorded on papyri, etc. Like today, information can create certain ideas, exert political influence, and also present facts in a different and not necessarily realistic light. Iconographic and literary sources from Ancient Egypt depict and praise the pharaoh as a victorious military leader. A recurring propagandist motives refers to soldiers presenting the severed right hands of foes to the Pharaoh in order to garner the “gold of honour” – , a prestigious reward, primarily in the form of a collar of golden beads . Until now, this practice is known only from tomb inscriptions of prominent warriors and from inscriptions and temple reliefs, all dating from the start of the New Kingdom (18th–20th Dynasties) onwards. The bioarchaeological analysis of human hands, found in 2011 at Tell el-Dab‘a (ancient Avaris) in the eastern Nile delta region , – (Fig. A), provides a third source of evidence (additional to iconographic – , – and archaeological , ), offering unique insights and, to date, unknown details of this practice. Taphonomic and biological analyses carried out on the bones reveal information regarding the act of mutilation and preparation of these body parts, as well as about the individuals to whom they originally belonged. From a wider transdisciplinary scenario, the results presented in this article address questions about the embodiment of violence in the context of war, and, specifically, trophy taking as a structured language of dominance.
Materials and archaeological context In the forecourt of a Hyksos Period Middle Bronze Age-style palace (c. 1640–1530 BC) – , three pits with severed hands were found , , , (Fig. ). The palace was built on top of a similar 14 th Dynasty palace and had a longer lifespan, covering the major part of the Hyksos Period. One of its main occupants seems to have been the Hyksos Khayan (c. 1700 BC–1580 BC), whose numerous seal impressions were found in offering pits belonging to its earlier phase – . Judging from its later offering pits and the filling of a well from the later phase, the palace may have been used until the late Hyksos Period but may have lost its purpose when a new palatial compound was built further north in the late Hyksos Period. The smallest of the three pits, Pit L1777, in front of the throne room, contained a single fully articulated hand sealed beneath the south wall of a later added broad-room building, most probably a temple built against the western enclosure wall of the palace’s forecourt. The pit seems to have been dug into the temple’s open foundation trenches because it cut the continuous foundation layer of loam-mortar at the base of the trench (Fig. B,D,F). It can, therefore, be dated in relative terms between the early and the late phase of the palace. Two more pits were discovered c. 7 m north-east of the broad-room building near the western enclosure wall, just below the modern agricultural fields (Fig. B,C,E). These two pits were aligned to the enclosure wall of the palace’s forecourt and no ceramic material later than the Hyksos Period was found inside them. Pit L1542 contained the remains of three and L1543 the remains of eight articulated hands, thus the right hands of 12 individuals. In both pits there were also disarticulated fingers, their attribution will be discussed below. Bioarchaeological evidence Number of individuals The superficial position of the hands in the ground led to heavy erosion, flaking and cracking of the bone tissue. The high humidity and soil composition make them soft and brittle and very difficult to excavate (see chapter limitations). Referring only to the quantitative state of preservation, the hands will be grouped in three classes. Complete: at least 75% of the hand bones are present and articulated; Almost complete: 75–50% are present and articulated; Single digits: when a complete digit (including metacarpal bone, proximal to distal phalanges) or parts of them are present. Anatomical markers identify all hands and single phalanges from the three pits as being from right hands (for more detail, see ). L1777 comprised one complete right hand, while L1542 comprised two complete right hands together with an incomplete single second right digit belonging to a third person. Pit L1543 contained eight complete and almost complete right hands and eleven additional phalanges that partly belong together, leaving six additional complete or incomplete digits. Therefore, the evidence suggests a minimum of 12 right hands, hence at least 12 individuals, from the three pits in front of the throne room and along its wall (L1543: 8 hands, L1542: 3 hands, L1777: 1 hand). Assuming that the six single digits each represent an extra hand, that would result in a maximum of 18 right hands in the three pits. Deposition Of eleven complete right hands (see Table in supplementary information), eight are placed on their palmar surface and three on their dorsal surface (Fig. ). The fingers of six hands were splayed wide; the fingers of four hands were lying close together; and in one hand, the position of the fingers could not be determined. There is no correlation between the position of the fingers and the placement on the dorsal or palmar side. Because the single phalanges might have been moved after the placement of the hands, the informative value of their position is low. The metacarpophalangeal joints of the two intact hands found in L1542 were hyper-extended, so were the hands in L1543-7 and L1543-8. L1543-8 also showed a misplacement of the first digit, which was hyper-abducted (98°), exaggerating the maximum value of 45° between the first and second digits. L1777 had a displaced first digit, with the proximal metacarpal and the radial carpals positioned below the rest of the hand. Following the deposition of the hands, some post-mortem displacements might have occurred: the first digit of one complete hand (L1543-8) was either disarticulated before or after the hand was placed in the ground. Six single fingers/phalanges were scattered between the other, complete and almost complete, hands. Either these elements were moved from the intact hands or they represent the remnants of additional hands that were displaced by rodent activity or originated from other such pits that were disturbed by the later interments of hands. Post-mortem displacement of these elements would have required leaving the hands in the pit open for a period of time, at least until the soft tissue decomposed and elements could be moved. However, there are no animal gnawing marks on the bones. Therefore, it seems unlikely that the pit was left open for an extended period of time. The position of the hands, mainly on their palmar surfaces with splayed fingers, might have been caused either by taphonomic reasons, or it might have been due to their deliberate placement. In the first case, when the hands would have been thrown into the pits, soil pressure would have flattened them, by pressing the arch of the hand into the ground, possibly leading to splaying of the fingers and hyperextension of the metacarpophalangeal joints . If we assume a deliberate placement, this arrangement could have been done to make the hands look more impressive, possibly larger, and to better match the prototype of a hand. Yet, there is no pattern in the placement of the hands; some were single, some lay on top others in a smaller group. Severing and preparation Only six hands have preserved proximal row carpal bones, and none exhibit cut marks or any evidence of soft tissue removal. Because no fragments of lower arm bones were attached or found in the pit, it implies that these hands were precisely severed from the lower arm. One technique of severing hands is to cut the joint capsule and open it by intersecting the tendons spanning the wrist joint . If done correctly, there are no cut marks on the bones. If done unprofessionally, however, cut marks are to be expected. Mutilating people without regard to their survival is often done by severing the arm at any anatomical position. This method is faster and easier, but it leaves a section of the lower arm attached to the hand. If this was the case with these hands, the people offering them, or those in charge of the ceremony, cared enough about their proper presentation to detach parts of the lower arm. Two main distinctions can be made regarding the procedure of hand detachment: collecting them from the recently deceased or mutilating living people. In both cases, the hands must have been soft and flexible when they were placed into the pit. That is, either before rigor mortis sets in or after it has resolved. Rigor mortis of the hands commonly begins 6–8 h after death (there are different times for different body parts). This means that living victims were mutilated during or shortly before the ceremony. It seems, however, much more likely that the hands were placed after rigor mortis ended, between 24 and 48 h after death. This indicates that the hands were collected and kept for a period of time before being placed in the pit. The hands were buried while they were still intact, at least with the tendons and ligaments holding the skeletal elements in their original place and remaining supple enough to flex passively under appropriate stress. This capacity is affected by surrounding environmental factors like humidity and temperature. Biological profile of the hands The closed epiphyseal lines indicate that all the specimens belonged to adult individuals older than 14–21 years . The absence of even incipient bone changes owing to age-related degenerative processes, e.g., DJD and osteoporosis, rules out individuals reaching the old adult age class. The large size and robustness of the hands point to the male rather than female sex of the individuals. However, the size of hands varies between males and females. Because genetic analyses to determine the sex could not be applied due to the very poor preservation of the bones (see “ ”), an estimation of the 2D:4D ratio was employed to determine the sex of the individuals – . The typical male 2D:4D ratio is that the fourth digit is longer than the second. The proportion of these two fingers is different in males and females due to prenatal exposure to androgen . The measurements of the phalanges of all the hands show the fourth digit to be longer than the second, indicating male sex (see ). The sole possible exception to the 2D:4D ratio is the L1543-2 hand, which macroscopically appears smaller than the others. This hand’s 2D:4D ratio suggests a female (SDS of 2D:4D ration is beyond the normal range). However, in this case the phalanges of the second and fourth finger were partially incomplete, thus measurements could only be estimated. Interpretation and contextualisation The bioarchaeological evidence from Tell el-Dab‘a addresses the question, crucial to its interpretation, of whether the mutilation occurred as a form of punishment or as an accounting and reward system following military victories. Thus far, the severing of hands as a punishment is not attested in Egyptian texts. Nevertheless, the removal of right hands is mentioned by papyrus Salt 124, l, 7, from the 20th Dynasty , , . This deals with the act of plundering Sety II-Merenptah’s royal tomb. During the plunder, the pharaoh’s hand was removed, apparently by the tomb robbers, perhaps to obtain quickly the rings from the fingers. Indeed, the right hand of the mummy of Sety II is missing, and hand removal is also attested from other royal mummies, most likely for this reason , . The pits containing the hands were located in the palace’s forecourt, in front of the throne room. Their position points to the widespread visibility conferred by the practice that generated the deposits as part of a public ceremony. The later attested ‘Window of Appearances,’ through which New Kingdom kings offered the ‘gold of honour,’ may have already existed in this palace . The absence of the distal parts of the lower arm and the lack of cut marks indicate the hands underwent a careful pre-depositional preparation phase, aimed at removing all elements deemed to be unrelated to the anatomy of the hand. At the moment of their deposition in the pits, the hands might have been arranged, predominantly on their palmar faces (n = 8/11), in most cases with the fingers spread out (n = 6/10). Assuming an intentional positioning, this appears aimed at facilitating the identification of the body part in the pit as a hand, thus conferring on it a defined semiotic function. In the iconicity of this position , each hand represents an individual: pars pro toto . It appears that all the hands belonged to adult individuals who may not have reached late adulthood. The results also suggest that all the individuals were likely males, except for a possible female. If we refer to the male warrior hypothesis, long debated in behavioural sciences , this data may support the scenario presented above, because the severed hands offered in the “gold of honour” ceremony belonged to foes, generally male individuals of fighting age, killed in battle. At the same time, the presence of a female individual advocates for a less gender-rigid approach to the reconstruction of this procedure. Throughout history, women have played various roles in military societies. Women and warfare did not exist in separate worlds. On the contrary, they were inextricably linked to the political, social and religious spheres , . Consequently, we cannot exclude that the specific hand attested at Tell el-Dab‘a belonged to a woman. Although the hands cannot be attributed to a specific ethnic or cultural group, the custom of severing the right hands of foes appears to have been introduced to Egypt by the Hyksos , , approximately 50–80 years earlier than the inscriptional and pictorial evidence. The Egyptians adopted this custom at the latest in King Ahmose’s reign, as shown by a relief of a pile of hands at his temple in Abydos and in the autobiographies of Ahmose, son of Ibana , , and Ahmose, son of Pennekhbet in El-Kab , . Ahmose was the one who conquered Avaris and defeated the Hyksos, and thus was likely familiar with this practice. Early 18th Dynasty tomb inscriptions and temple reliefs from the 18th to the 20th Dynasties consistently depict hand counts on the battlefield following major battles – , – . The idea that the custom of severing enemy hands originated in the Near East may be supported linguistically . During the early 18th Dynasty, a specific new hieroglyph appears with the first inscriptional mention of the cutting of hands in warfare in the inscription of Ahmose, son of Ibana. It is not the typical logogram-sign for hand d.t in side view (Gardiner list, D46) but a very realistic representation of an outstretched palm, showing five spread fingers (Fig. ). This new pictogram may signal a new word instead d.t ‘hand’—which now also refers to ‘severed hand.’ In Egyptian inscriptions beginning with the 19th Dynasty (Merenptah), a Semitic loanword kp כף ‘hand’ ‘palm’ is introduced to refer specifically to ‘severed hands’ , . In the hieroglyphic inscriptions of the 20th Dynasty kings, the new word only refers to ‘severed hands’ . It seems possible that this semitic loanword entered already in the 18th Dynasty into military language. The outstretched hand symbol is a compelling parallel to the bioarchaeological evidence of this paper. Although it is beyond the scope of this article to discuss the possible origins of this practice, it should be noted that corporal mutilation of enemies has been known in Egypt since the time of King Narmer in the Early Dynastic period, but the severing of hands appears in Egyptian records only after Hyksos rule. Seals with rows of possible severed hands and heads, together with animal heads, also appear in Middle Bronze Age Syrian glyptic art . Although bioarchaeological literature offers abundant evidence of trophy-taking practices from all over the world , , the case presented here documents a marked visual component, as the body parts, i.e., severed hands, were prepared and arranged for presentation in a public ceremony in the pharaoh's palace. In this politically structured frame, the severed hands served as symbolic currency for status acquisition, within a system of values celebrating warfare and dominance, as argued by other bioarchaeological studies in geographically diverse contexts . Iconographic sources from several archaeological sites in Egypt deliver a large corpus of evidence for the practice of body dismemberment and mutilation, particularly related to war contexts , . The representation of piles of severed heads, ears and genitals follows codified and publicly recognisable ethics of violence as it ultimately conveys a message of political stability: the pharaoh, representation and personification of the gods, maintains the universal order by defeating the forces of chaos, personified by the enemies . The public enactment was the necessary step to deliver this message. Here, the act of dominance is conveyed by maiming defeated enemies, depriving them of their right hands, hampering their capacity to carry out future attacks and essential daily-life activities. The likelihood that hands were taken from captives is low, since this would limit their potential as future slaves . Because corporeal integrity was vital for survival in the Ancient Egyptian view of the afterlife , the victim’s impairment adds a deeper dimension to this act of dominance.
In the forecourt of a Hyksos Period Middle Bronze Age-style palace (c. 1640–1530 BC) – , three pits with severed hands were found , , , (Fig. ). The palace was built on top of a similar 14 th Dynasty palace and had a longer lifespan, covering the major part of the Hyksos Period. One of its main occupants seems to have been the Hyksos Khayan (c. 1700 BC–1580 BC), whose numerous seal impressions were found in offering pits belonging to its earlier phase – . Judging from its later offering pits and the filling of a well from the later phase, the palace may have been used until the late Hyksos Period but may have lost its purpose when a new palatial compound was built further north in the late Hyksos Period. The smallest of the three pits, Pit L1777, in front of the throne room, contained a single fully articulated hand sealed beneath the south wall of a later added broad-room building, most probably a temple built against the western enclosure wall of the palace’s forecourt. The pit seems to have been dug into the temple’s open foundation trenches because it cut the continuous foundation layer of loam-mortar at the base of the trench (Fig. B,D,F). It can, therefore, be dated in relative terms between the early and the late phase of the palace. Two more pits were discovered c. 7 m north-east of the broad-room building near the western enclosure wall, just below the modern agricultural fields (Fig. B,C,E). These two pits were aligned to the enclosure wall of the palace’s forecourt and no ceramic material later than the Hyksos Period was found inside them. Pit L1542 contained the remains of three and L1543 the remains of eight articulated hands, thus the right hands of 12 individuals. In both pits there were also disarticulated fingers, their attribution will be discussed below.
Number of individuals The superficial position of the hands in the ground led to heavy erosion, flaking and cracking of the bone tissue. The high humidity and soil composition make them soft and brittle and very difficult to excavate (see chapter limitations). Referring only to the quantitative state of preservation, the hands will be grouped in three classes. Complete: at least 75% of the hand bones are present and articulated; Almost complete: 75–50% are present and articulated; Single digits: when a complete digit (including metacarpal bone, proximal to distal phalanges) or parts of them are present. Anatomical markers identify all hands and single phalanges from the three pits as being from right hands (for more detail, see ). L1777 comprised one complete right hand, while L1542 comprised two complete right hands together with an incomplete single second right digit belonging to a third person. Pit L1543 contained eight complete and almost complete right hands and eleven additional phalanges that partly belong together, leaving six additional complete or incomplete digits. Therefore, the evidence suggests a minimum of 12 right hands, hence at least 12 individuals, from the three pits in front of the throne room and along its wall (L1543: 8 hands, L1542: 3 hands, L1777: 1 hand). Assuming that the six single digits each represent an extra hand, that would result in a maximum of 18 right hands in the three pits. Deposition Of eleven complete right hands (see Table in supplementary information), eight are placed on their palmar surface and three on their dorsal surface (Fig. ). The fingers of six hands were splayed wide; the fingers of four hands were lying close together; and in one hand, the position of the fingers could not be determined. There is no correlation between the position of the fingers and the placement on the dorsal or palmar side. Because the single phalanges might have been moved after the placement of the hands, the informative value of their position is low. The metacarpophalangeal joints of the two intact hands found in L1542 were hyper-extended, so were the hands in L1543-7 and L1543-8. L1543-8 also showed a misplacement of the first digit, which was hyper-abducted (98°), exaggerating the maximum value of 45° between the first and second digits. L1777 had a displaced first digit, with the proximal metacarpal and the radial carpals positioned below the rest of the hand. Following the deposition of the hands, some post-mortem displacements might have occurred: the first digit of one complete hand (L1543-8) was either disarticulated before or after the hand was placed in the ground. Six single fingers/phalanges were scattered between the other, complete and almost complete, hands. Either these elements were moved from the intact hands or they represent the remnants of additional hands that were displaced by rodent activity or originated from other such pits that were disturbed by the later interments of hands. Post-mortem displacement of these elements would have required leaving the hands in the pit open for a period of time, at least until the soft tissue decomposed and elements could be moved. However, there are no animal gnawing marks on the bones. Therefore, it seems unlikely that the pit was left open for an extended period of time. The position of the hands, mainly on their palmar surfaces with splayed fingers, might have been caused either by taphonomic reasons, or it might have been due to their deliberate placement. In the first case, when the hands would have been thrown into the pits, soil pressure would have flattened them, by pressing the arch of the hand into the ground, possibly leading to splaying of the fingers and hyperextension of the metacarpophalangeal joints . If we assume a deliberate placement, this arrangement could have been done to make the hands look more impressive, possibly larger, and to better match the prototype of a hand. Yet, there is no pattern in the placement of the hands; some were single, some lay on top others in a smaller group. Severing and preparation Only six hands have preserved proximal row carpal bones, and none exhibit cut marks or any evidence of soft tissue removal. Because no fragments of lower arm bones were attached or found in the pit, it implies that these hands were precisely severed from the lower arm. One technique of severing hands is to cut the joint capsule and open it by intersecting the tendons spanning the wrist joint . If done correctly, there are no cut marks on the bones. If done unprofessionally, however, cut marks are to be expected. Mutilating people without regard to their survival is often done by severing the arm at any anatomical position. This method is faster and easier, but it leaves a section of the lower arm attached to the hand. If this was the case with these hands, the people offering them, or those in charge of the ceremony, cared enough about their proper presentation to detach parts of the lower arm. Two main distinctions can be made regarding the procedure of hand detachment: collecting them from the recently deceased or mutilating living people. In both cases, the hands must have been soft and flexible when they were placed into the pit. That is, either before rigor mortis sets in or after it has resolved. Rigor mortis of the hands commonly begins 6–8 h after death (there are different times for different body parts). This means that living victims were mutilated during or shortly before the ceremony. It seems, however, much more likely that the hands were placed after rigor mortis ended, between 24 and 48 h after death. This indicates that the hands were collected and kept for a period of time before being placed in the pit. The hands were buried while they were still intact, at least with the tendons and ligaments holding the skeletal elements in their original place and remaining supple enough to flex passively under appropriate stress. This capacity is affected by surrounding environmental factors like humidity and temperature. Biological profile of the hands The closed epiphyseal lines indicate that all the specimens belonged to adult individuals older than 14–21 years . The absence of even incipient bone changes owing to age-related degenerative processes, e.g., DJD and osteoporosis, rules out individuals reaching the old adult age class. The large size and robustness of the hands point to the male rather than female sex of the individuals. However, the size of hands varies between males and females. Because genetic analyses to determine the sex could not be applied due to the very poor preservation of the bones (see “ ”), an estimation of the 2D:4D ratio was employed to determine the sex of the individuals – . The typical male 2D:4D ratio is that the fourth digit is longer than the second. The proportion of these two fingers is different in males and females due to prenatal exposure to androgen . The measurements of the phalanges of all the hands show the fourth digit to be longer than the second, indicating male sex (see ). The sole possible exception to the 2D:4D ratio is the L1543-2 hand, which macroscopically appears smaller than the others. This hand’s 2D:4D ratio suggests a female (SDS of 2D:4D ration is beyond the normal range). However, in this case the phalanges of the second and fourth finger were partially incomplete, thus measurements could only be estimated.
The superficial position of the hands in the ground led to heavy erosion, flaking and cracking of the bone tissue. The high humidity and soil composition make them soft and brittle and very difficult to excavate (see chapter limitations). Referring only to the quantitative state of preservation, the hands will be grouped in three classes. Complete: at least 75% of the hand bones are present and articulated; Almost complete: 75–50% are present and articulated; Single digits: when a complete digit (including metacarpal bone, proximal to distal phalanges) or parts of them are present. Anatomical markers identify all hands and single phalanges from the three pits as being from right hands (for more detail, see ). L1777 comprised one complete right hand, while L1542 comprised two complete right hands together with an incomplete single second right digit belonging to a third person. Pit L1543 contained eight complete and almost complete right hands and eleven additional phalanges that partly belong together, leaving six additional complete or incomplete digits. Therefore, the evidence suggests a minimum of 12 right hands, hence at least 12 individuals, from the three pits in front of the throne room and along its wall (L1543: 8 hands, L1542: 3 hands, L1777: 1 hand). Assuming that the six single digits each represent an extra hand, that would result in a maximum of 18 right hands in the three pits.
Of eleven complete right hands (see Table in supplementary information), eight are placed on their palmar surface and three on their dorsal surface (Fig. ). The fingers of six hands were splayed wide; the fingers of four hands were lying close together; and in one hand, the position of the fingers could not be determined. There is no correlation between the position of the fingers and the placement on the dorsal or palmar side. Because the single phalanges might have been moved after the placement of the hands, the informative value of their position is low. The metacarpophalangeal joints of the two intact hands found in L1542 were hyper-extended, so were the hands in L1543-7 and L1543-8. L1543-8 also showed a misplacement of the first digit, which was hyper-abducted (98°), exaggerating the maximum value of 45° between the first and second digits. L1777 had a displaced first digit, with the proximal metacarpal and the radial carpals positioned below the rest of the hand. Following the deposition of the hands, some post-mortem displacements might have occurred: the first digit of one complete hand (L1543-8) was either disarticulated before or after the hand was placed in the ground. Six single fingers/phalanges were scattered between the other, complete and almost complete, hands. Either these elements were moved from the intact hands or they represent the remnants of additional hands that were displaced by rodent activity or originated from other such pits that were disturbed by the later interments of hands. Post-mortem displacement of these elements would have required leaving the hands in the pit open for a period of time, at least until the soft tissue decomposed and elements could be moved. However, there are no animal gnawing marks on the bones. Therefore, it seems unlikely that the pit was left open for an extended period of time. The position of the hands, mainly on their palmar surfaces with splayed fingers, might have been caused either by taphonomic reasons, or it might have been due to their deliberate placement. In the first case, when the hands would have been thrown into the pits, soil pressure would have flattened them, by pressing the arch of the hand into the ground, possibly leading to splaying of the fingers and hyperextension of the metacarpophalangeal joints . If we assume a deliberate placement, this arrangement could have been done to make the hands look more impressive, possibly larger, and to better match the prototype of a hand. Yet, there is no pattern in the placement of the hands; some were single, some lay on top others in a smaller group.
Only six hands have preserved proximal row carpal bones, and none exhibit cut marks or any evidence of soft tissue removal. Because no fragments of lower arm bones were attached or found in the pit, it implies that these hands were precisely severed from the lower arm. One technique of severing hands is to cut the joint capsule and open it by intersecting the tendons spanning the wrist joint . If done correctly, there are no cut marks on the bones. If done unprofessionally, however, cut marks are to be expected. Mutilating people without regard to their survival is often done by severing the arm at any anatomical position. This method is faster and easier, but it leaves a section of the lower arm attached to the hand. If this was the case with these hands, the people offering them, or those in charge of the ceremony, cared enough about their proper presentation to detach parts of the lower arm. Two main distinctions can be made regarding the procedure of hand detachment: collecting them from the recently deceased or mutilating living people. In both cases, the hands must have been soft and flexible when they were placed into the pit. That is, either before rigor mortis sets in or after it has resolved. Rigor mortis of the hands commonly begins 6–8 h after death (there are different times for different body parts). This means that living victims were mutilated during or shortly before the ceremony. It seems, however, much more likely that the hands were placed after rigor mortis ended, between 24 and 48 h after death. This indicates that the hands were collected and kept for a period of time before being placed in the pit. The hands were buried while they were still intact, at least with the tendons and ligaments holding the skeletal elements in their original place and remaining supple enough to flex passively under appropriate stress. This capacity is affected by surrounding environmental factors like humidity and temperature.
The closed epiphyseal lines indicate that all the specimens belonged to adult individuals older than 14–21 years . The absence of even incipient bone changes owing to age-related degenerative processes, e.g., DJD and osteoporosis, rules out individuals reaching the old adult age class. The large size and robustness of the hands point to the male rather than female sex of the individuals. However, the size of hands varies between males and females. Because genetic analyses to determine the sex could not be applied due to the very poor preservation of the bones (see “ ”), an estimation of the 2D:4D ratio was employed to determine the sex of the individuals – . The typical male 2D:4D ratio is that the fourth digit is longer than the second. The proportion of these two fingers is different in males and females due to prenatal exposure to androgen . The measurements of the phalanges of all the hands show the fourth digit to be longer than the second, indicating male sex (see ). The sole possible exception to the 2D:4D ratio is the L1543-2 hand, which macroscopically appears smaller than the others. This hand’s 2D:4D ratio suggests a female (SDS of 2D:4D ration is beyond the normal range). However, in this case the phalanges of the second and fourth finger were partially incomplete, thus measurements could only be estimated.
The bioarchaeological evidence from Tell el-Dab‘a addresses the question, crucial to its interpretation, of whether the mutilation occurred as a form of punishment or as an accounting and reward system following military victories. Thus far, the severing of hands as a punishment is not attested in Egyptian texts. Nevertheless, the removal of right hands is mentioned by papyrus Salt 124, l, 7, from the 20th Dynasty , , . This deals with the act of plundering Sety II-Merenptah’s royal tomb. During the plunder, the pharaoh’s hand was removed, apparently by the tomb robbers, perhaps to obtain quickly the rings from the fingers. Indeed, the right hand of the mummy of Sety II is missing, and hand removal is also attested from other royal mummies, most likely for this reason , . The pits containing the hands were located in the palace’s forecourt, in front of the throne room. Their position points to the widespread visibility conferred by the practice that generated the deposits as part of a public ceremony. The later attested ‘Window of Appearances,’ through which New Kingdom kings offered the ‘gold of honour,’ may have already existed in this palace . The absence of the distal parts of the lower arm and the lack of cut marks indicate the hands underwent a careful pre-depositional preparation phase, aimed at removing all elements deemed to be unrelated to the anatomy of the hand. At the moment of their deposition in the pits, the hands might have been arranged, predominantly on their palmar faces (n = 8/11), in most cases with the fingers spread out (n = 6/10). Assuming an intentional positioning, this appears aimed at facilitating the identification of the body part in the pit as a hand, thus conferring on it a defined semiotic function. In the iconicity of this position , each hand represents an individual: pars pro toto . It appears that all the hands belonged to adult individuals who may not have reached late adulthood. The results also suggest that all the individuals were likely males, except for a possible female. If we refer to the male warrior hypothesis, long debated in behavioural sciences , this data may support the scenario presented above, because the severed hands offered in the “gold of honour” ceremony belonged to foes, generally male individuals of fighting age, killed in battle. At the same time, the presence of a female individual advocates for a less gender-rigid approach to the reconstruction of this procedure. Throughout history, women have played various roles in military societies. Women and warfare did not exist in separate worlds. On the contrary, they were inextricably linked to the political, social and religious spheres , . Consequently, we cannot exclude that the specific hand attested at Tell el-Dab‘a belonged to a woman. Although the hands cannot be attributed to a specific ethnic or cultural group, the custom of severing the right hands of foes appears to have been introduced to Egypt by the Hyksos , , approximately 50–80 years earlier than the inscriptional and pictorial evidence. The Egyptians adopted this custom at the latest in King Ahmose’s reign, as shown by a relief of a pile of hands at his temple in Abydos and in the autobiographies of Ahmose, son of Ibana , , and Ahmose, son of Pennekhbet in El-Kab , . Ahmose was the one who conquered Avaris and defeated the Hyksos, and thus was likely familiar with this practice. Early 18th Dynasty tomb inscriptions and temple reliefs from the 18th to the 20th Dynasties consistently depict hand counts on the battlefield following major battles – , – . The idea that the custom of severing enemy hands originated in the Near East may be supported linguistically . During the early 18th Dynasty, a specific new hieroglyph appears with the first inscriptional mention of the cutting of hands in warfare in the inscription of Ahmose, son of Ibana. It is not the typical logogram-sign for hand d.t in side view (Gardiner list, D46) but a very realistic representation of an outstretched palm, showing five spread fingers (Fig. ). This new pictogram may signal a new word instead d.t ‘hand’—which now also refers to ‘severed hand.’ In Egyptian inscriptions beginning with the 19th Dynasty (Merenptah), a Semitic loanword kp כף ‘hand’ ‘palm’ is introduced to refer specifically to ‘severed hands’ , . In the hieroglyphic inscriptions of the 20th Dynasty kings, the new word only refers to ‘severed hands’ . It seems possible that this semitic loanword entered already in the 18th Dynasty into military language. The outstretched hand symbol is a compelling parallel to the bioarchaeological evidence of this paper. Although it is beyond the scope of this article to discuss the possible origins of this practice, it should be noted that corporal mutilation of enemies has been known in Egypt since the time of King Narmer in the Early Dynastic period, but the severing of hands appears in Egyptian records only after Hyksos rule. Seals with rows of possible severed hands and heads, together with animal heads, also appear in Middle Bronze Age Syrian glyptic art . Although bioarchaeological literature offers abundant evidence of trophy-taking practices from all over the world , , the case presented here documents a marked visual component, as the body parts, i.e., severed hands, were prepared and arranged for presentation in a public ceremony in the pharaoh's palace. In this politically structured frame, the severed hands served as symbolic currency for status acquisition, within a system of values celebrating warfare and dominance, as argued by other bioarchaeological studies in geographically diverse contexts . Iconographic sources from several archaeological sites in Egypt deliver a large corpus of evidence for the practice of body dismemberment and mutilation, particularly related to war contexts , . The representation of piles of severed heads, ears and genitals follows codified and publicly recognisable ethics of violence as it ultimately conveys a message of political stability: the pharaoh, representation and personification of the gods, maintains the universal order by defeating the forces of chaos, personified by the enemies . The public enactment was the necessary step to deliver this message. Here, the act of dominance is conveyed by maiming defeated enemies, depriving them of their right hands, hampering their capacity to carry out future attacks and essential daily-life activities. The likelihood that hands were taken from captives is low, since this would limit their potential as future slaves . Because corporeal integrity was vital for survival in the Ancient Egyptian view of the afterlife , the victim’s impairment adds a deeper dimension to this act of dominance.
The location, treatment, and possibly the positioning of the severed hands argues against the hypothesis of law-enforcing punishment as the motivation for these acts. When contextualised in a transdisciplinary approach to the archaeological and historic sources, the bioarchaeological evidence presented here suggests that the severed hands were offered as trophies as part of a public event that took place in the palace. They belonged to at least eleven males and possibly one female, which may indicate that women and warfare were not worlds apart. To the best of the authors’ knowledge, the results put forward in this paper provide the first direct bioarchaeological evidence for the ‘gold of honour’ ceremony performed in front of the king’s palace and contribute significantly to the debate over the reconstruction of this ceremony.
Osteology At the excavation site, the hands were hardened with acetone-soluble glue and recovered en bloc in a plaster cast. In the lab, the very fragile bones remained in the plaster casts; only the surfaces were carefully cleaned of adhering sand. Unfortunately, a thin layer of calcareous deposits has covered some surfaces. This could not be removed since proper cleaning would have resulted in surface destruction. The bones were examined in detail using a magnifying glass. The surfaces of all bones, but particularly the proximal carpal bones, were inspected for cut marks. The hands were measured in situ using a sliding calliper (maximum length of phalanges, distances between the fingers; maximum width and length of the hands). The minimum number of individuals (MNI) was calculated by excluding skeletal elements that could belong to other hands. Age estimation was carried out on the epiphyseal closure of the metacarpals and phalanges . Additionally, when the bones were broken, the state of osteoporosis of the compact and spongy bones was estimated, and the joint surfaces were assessed for degenerative changes. Osteometry The 2D:4D ratio is sex-based, with the fourth digit typically being longer than the second in males , . In living people, digit length is “measured on the ventral surface from the basal crease to the tip of the digit” . By adapting measurements from living people to archaeological skeletons, we proposed that the ratio of the length of the single phalanges of the fourth and second digits should differ as well. The measurements of all complete phalanges were taken (Table in supplementary information). The 2D:4D ratio for each phalanx of the second and fourth digit was determined, and the individual mean and the standard deviation score (SDS) of the 2D:4D ratio were calculated (Table in supplementary information). The SDS is the individual value subtracted by the mean of a reference divided by the standard deviation of a reference. In our case, we calculated the ratio from each existing pair of phalanges of the second and fourth finger, and calculated the mean (individual value) for each individual. The sum of individuals is our reference. As a result, we have standardised dimensionless values (SDS) that enable us to compare individuals. Long bones are always distributed normally in a population , . Measurements between the ± 1.28 SDS range are significant (Fig. suppl). Limitations The excavation of human remains in this particular region of Egypt is very difficult due to the high humidity of the ground and the soil chemistry . The humid and dark soil affects the bones, which stains them and giving them a similar colour to the soil. It also makes them soft and brittle. As a result, they are extremely difficult to detect and to fully excavate without causing damage. Small body parts, like hands, are especially vulnerable to being overlooked, and by the time they become visible, parts of them may have already been lost. The 2D:4D ratio for sex estimation applies to living people. We transferred the method to archaeological skeletons, which is not exactly comparable to the data obtained from living people. However, the measurements taken from the bones should be even more exact due to the missing tissue, which can confound the measurements of finger lengths. There is no reason to assume that the influence of soft tissue on fingers is sex specific.
At the excavation site, the hands were hardened with acetone-soluble glue and recovered en bloc in a plaster cast. In the lab, the very fragile bones remained in the plaster casts; only the surfaces were carefully cleaned of adhering sand. Unfortunately, a thin layer of calcareous deposits has covered some surfaces. This could not be removed since proper cleaning would have resulted in surface destruction. The bones were examined in detail using a magnifying glass. The surfaces of all bones, but particularly the proximal carpal bones, were inspected for cut marks. The hands were measured in situ using a sliding calliper (maximum length of phalanges, distances between the fingers; maximum width and length of the hands). The minimum number of individuals (MNI) was calculated by excluding skeletal elements that could belong to other hands. Age estimation was carried out on the epiphyseal closure of the metacarpals and phalanges . Additionally, when the bones were broken, the state of osteoporosis of the compact and spongy bones was estimated, and the joint surfaces were assessed for degenerative changes.
The 2D:4D ratio is sex-based, with the fourth digit typically being longer than the second in males , . In living people, digit length is “measured on the ventral surface from the basal crease to the tip of the digit” . By adapting measurements from living people to archaeological skeletons, we proposed that the ratio of the length of the single phalanges of the fourth and second digits should differ as well. The measurements of all complete phalanges were taken (Table in supplementary information). The 2D:4D ratio for each phalanx of the second and fourth digit was determined, and the individual mean and the standard deviation score (SDS) of the 2D:4D ratio were calculated (Table in supplementary information). The SDS is the individual value subtracted by the mean of a reference divided by the standard deviation of a reference. In our case, we calculated the ratio from each existing pair of phalanges of the second and fourth finger, and calculated the mean (individual value) for each individual. The sum of individuals is our reference. As a result, we have standardised dimensionless values (SDS) that enable us to compare individuals. Long bones are always distributed normally in a population , . Measurements between the ± 1.28 SDS range are significant (Fig. suppl).
The excavation of human remains in this particular region of Egypt is very difficult due to the high humidity of the ground and the soil chemistry . The humid and dark soil affects the bones, which stains them and giving them a similar colour to the soil. It also makes them soft and brittle. As a result, they are extremely difficult to detect and to fully excavate without causing damage. Small body parts, like hands, are especially vulnerable to being overlooked, and by the time they become visible, parts of them may have already been lost. The 2D:4D ratio for sex estimation applies to living people. We transferred the method to archaeological skeletons, which is not exactly comparable to the data obtained from living people. However, the measurements taken from the bones should be even more exact due to the missing tissue, which can confound the measurements of finger lengths. There is no reason to assume that the influence of soft tissue on fingers is sex specific.
Supplementary Information.
|
Morphological and histochemical identification of telocytes in adult yak epididymis | d060c115-79cb-44d3-9a10-b618f4fb9f4b | 10066225 | Anatomy[mh] | Telocytes (TCs) are a newly discovered type of mesenchymal cells with unique morphological characteristics that have a long cytoplasmic extensions called telopodes (TPs). TPs are slender, long, and varied in number. Their shapes are mainly irregular ellipsoid, pear, and spindle; they are rich in mitochondria, the endoplasmic reticulum, and have a secretory function – . The unique morphological characteristics of TCs make them different from other mesenchymal cells. The TPs are in close contact with blood vessels, nerve bundles, and local immune system cells through organ matrix distribution, forming a network between tissues; this network was considered to be the structural basis for cell communication , . Furthermore, TCs may also establish unique spatial relationships with a variety of cells, including adjacent parenchymal cells and other cells in the interstitial compartment, and are considered to regulate the dynamic balance of the local microenvironment by contacting or releasing secretory vesicles between cells – . These secretory vesicles are considered to be “messengers” of substance exchange and signal transduction between TCs and other cells , , and are also an important reason why TCs are thought to be secretory cells. As early as a century ago, a special cell group in the muscular layer of the human intestinal tract was discovered by Cajal and named as “interstitial neurons” . Telocytes were named by Popescu and Faussone-Pellegrini at the beginning of this century . For many years, research on telocytes was based on imaging their ultrastructure under a transmission electron microscope, so transmission electron microscopy is considered the “gold standard” for research into TC morphology. TCs have been found in the human myocardium and gallbladder; CD34, c-kit (CD117), and vimentin were co-expression in TCs, and involved in stem cell differentiation, the coordination of new angiogenesis, and the regulation of paracrine function in the interstitial tissue , . Furthermore, TCs were abundant in the mouse lung and rat kidney , . TCs were found in the fish brain and in poultry skin . TCs have been discovered to be involved in many functional aspects in recent years, including cell regeneration , inhibition of apoptosis , inflammation repair , cell communication , angiogenesis , , and stem cell function . In previous studies, CD34, CD117, and vimentin were widely considered effective markers of TCs – . The proteins are also widely used to locate and screen TCs. CD34 is a marker receptor found on the surface of mesenchymal stem cells; it is also a highly glycosylated type I transmembrane glycoprotein expressed on the surface of hematopoietic stem/progenitor cells of humans and other mammals – . CD117 (C-kit) is a marker of stem cells/progenitor cells in the heart and is considered one of the effective markers of TCs . Vimentin, a conserved type III intermediate filament protein, is often found in fibroblasts, vascular endothelial cells, neutrophils, and macrophages, and is abundantly expressed in glomeruli, tubules, and renal interstitial cells . Vimentin is a marker of mesenchymal phenotype and an important cytoskeletal protein. It is generally expressed only in mesenchymal cells and is closely related to the growth, invasion, and metastasis of tumor cells . For a long time, studies of TCs have been based on some stem cell characteristics, with the stem cell surface markers such as CD34 and CD117 generally considered the marker proteins of TCs. Vimentin is a phenotypic marker of interstitial tissue, which is often used to locate TCs in the interstitial tissue. Yak has a reputation as a ‘plateau boat’ and ‘omnipotent livestock’ in the Plateau pastoral area , . The internal environment and functions of epididymis affected by high altitude hypoxia environment, while the epididymis is an important place for sperm maturation, processing, and storage , . For example, the activity of the enzyme acrosome and its change in reactivity in hypoxic conditions damages sperm fertilization ability – . Therefore, our study aimed to determine the distribution location of TCs in the epididymis of yak and analyzed the ultrastructure of TCs, to provide new clues or information as to the function of TCs in plateau animals.
Ultrastructural characteristics of yak epididymis TCs under TEM TEM is the most effective method to identify TCs. TEM observation showed that the TCs in the epididymis of yak contained a large nucleus of indefinite shape; the most typical were ellipsoid, serrated, and pear-shaped. The nucleus, with obvious chromatin, was surrounded by a small amount of cytoplasm, rich in secretory vesicles and mitochondria. There were a large number of TPs composed of long cytoplasmic fragments (Fig. A,B). Most of the TCs in the epididymis of yak are distributed around the blood vessels and can be clearly observed to contact the blood vessels through TPs and may form special cell connections (Fig. C,D,E, and H). In addition, some of the TCs distributed in the interstitium of the epididymis were in contact with peritubular myoid cells and fibroblasts (Fig. F,G). The morphology of TCs in the corpus epididymis was similar to that in caput epididymis, which is characterized by a large body, thick cytoplasm, full nucleus, clear nucleolus, and with abundant secretory vesicles distributed around the nucleus of TCs (Fig. A–F). The nucleoli of TCs were clearly visible at high magnification and were also rich in extranuclear secretory vesicles and rough endoplasmic reticulum in Fig. G. Furthermore, there was also a slight difference in morphology between TCs distributed near the basal membrane and stroma in the cauda (Fig. A–G). TCs in the stroma had more plump cells, longer TPs, and numerous secretory vesicles, while the TCs outside the basement membrane had smaller bodies, larger nuclei, irregular strip shapes, and shorter TPs, they were also closely connected with many epithelial cells (Fig. A, B, and F). TPs as a signal communication tool extend to a variety of cells, forming a special network structure (Fig. B and F). Morphological structure of yak epididymis TCs under SEM The SEM photographs were colored using Adobe Photoshop 2020 software. It was found that TCs had complete cell morphology with the presence of obvious cytoplasmic processes on TPs. TCs usually attached to the epithelium or connected to other cells through TPs, but exist alone in the epididymis stroma. Moreover, the vast interactions can be used to make a complex TPs network between adjacent epithelial cells, and cell secretions are attached to TPs (F g. A–I). Morphological model of telocytes The morphological structure and the morphological model diagram of TCs in yak epididymis were proposed based on the results of TEM and SEM (Figs. and ). TCs special staining results The toluidine blue staining results showed that a few TCs were distributed in the interstitium and more were distributed near the microvascular of the epididymis. However, they had different structures, adopting ellipsoid, spindle, pear, or other forms (Fig. A–C). Mercury–bromophenol blue staining showed that TPs were more obvious and the cytoplasmic processes on TPs were stained dark blue (Fig. D–F). Immunohistochemical and immunofluorescence analysis of TC surface markers The immunohistochemical results showed that thick epithelial cells, mesenchymal cells, and capillaries were observed in the epididymis of yak. CD34 was strongly expressed in the interstitium and distributed mainly in the interstitium and near the epithelium (Fig. A–C). Compared with the caput and cauda, the CD34-positive intensity in the epididymis corpus was higher (Fig. B). Vimentin staining was strong positive in the caput, corpus, and cauda of yak epididymis. It is worth noting that vimentin has intensely positive expression in the stroma, epithelium, and microvascular wall of the epididymis (Fig. D–F). CD117 had strong positive expression in the caput, corpus, and cauda of yak epididymis, and the specificity was stronger than that of CD34 and vimentin. CD117 immunopositive cells were mainly distributed in the epithelial cytoplasm and mesenchymal cells (F g. G–I). The immunofluorescence results showed that vimentin was widely expressed in the epididymis of yak, including fibroblasts, perivascular muscle-like cells and vascular endothelial cells, with strong positive expression (Fig. ). These cells are present in the loose connective tissue around the tubulointerstitium. CD34 was positively distributed in the stroma and epithelium (Fig. ). The CD34-positive oval cells near the epithelium were of a short shape, with cytoplasmic processes, which may be dendritic cells or TCs, and the positive expression was stronger in the caput. The co-expression of vimentin/CD34 was observed in the interstitium, capillary, and epithelium of yak epididymis (Fig. ). There was strong positive expression of CD117 in the stroma and epithelium, with a large number of CD117-positive cells around the blood vessels of mesenchymal hair cells, and relatively few in the epithelium (Fig. ). The cells with strong positive CD34 expression and weakly positive CD117 expression appeared in the yak epididymis corpus; they may be phagocytes or lymphocytes (Fig. ). CD117/CD34 co-expression was observed in the interstitial epididymis, as shown by yellow fluorescence (Fig. ). The common features were proved by the co-expression of vimentin/CD34 and CD117/CD34 and showed that the cells had a long cytoplasmic extension and a large nucleus, and that most of them were distributed around the interstitial capillaries of yak epididymis and outside the epididymal epithelium. The obtained morphology conformed to the basic characteristics expected for TCs. The mRNA and protein expression of effective markers of telocytes in yak epididymis The expression of CD34, Vimentin, and CD117 mRNA in yak epididymis was detected by qRT-PCR. The expression of CD34 mRNA in caput epididymis tissues of yaks compared to corpus and cauda showed an elevated trend ( p < 0.01). Although there was no significant difference in the expression trend in vimentin mRNA in caput and corpus, it was significantly higher than that in the cauda epididymis ( p < 0.01). Similarly, CD117 mRNA expression in yak caput epididymis was also significantly higher than that in the corpus and cauda ( p < 0.01) (Fig. ). Western blotting results showed that the expression levels of the proteins of CD34, vimentin, and CD117 in the caput epididymis of yak were significantly higher than those in the corpus and cauda (Fig. ); moreover, protein expression trends patterns were consistent with mRNA expression. In summary, the relatively high gene transcription and protein translation levels of TC surface markers in the caput epididymis of yaks were probably closely related to its physiological function.
TEM is the most effective method to identify TCs. TEM observation showed that the TCs in the epididymis of yak contained a large nucleus of indefinite shape; the most typical were ellipsoid, serrated, and pear-shaped. The nucleus, with obvious chromatin, was surrounded by a small amount of cytoplasm, rich in secretory vesicles and mitochondria. There were a large number of TPs composed of long cytoplasmic fragments (Fig. A,B). Most of the TCs in the epididymis of yak are distributed around the blood vessels and can be clearly observed to contact the blood vessels through TPs and may form special cell connections (Fig. C,D,E, and H). In addition, some of the TCs distributed in the interstitium of the epididymis were in contact with peritubular myoid cells and fibroblasts (Fig. F,G). The morphology of TCs in the corpus epididymis was similar to that in caput epididymis, which is characterized by a large body, thick cytoplasm, full nucleus, clear nucleolus, and with abundant secretory vesicles distributed around the nucleus of TCs (Fig. A–F). The nucleoli of TCs were clearly visible at high magnification and were also rich in extranuclear secretory vesicles and rough endoplasmic reticulum in Fig. G. Furthermore, there was also a slight difference in morphology between TCs distributed near the basal membrane and stroma in the cauda (Fig. A–G). TCs in the stroma had more plump cells, longer TPs, and numerous secretory vesicles, while the TCs outside the basement membrane had smaller bodies, larger nuclei, irregular strip shapes, and shorter TPs, they were also closely connected with many epithelial cells (Fig. A, B, and F). TPs as a signal communication tool extend to a variety of cells, forming a special network structure (Fig. B and F).
The SEM photographs were colored using Adobe Photoshop 2020 software. It was found that TCs had complete cell morphology with the presence of obvious cytoplasmic processes on TPs. TCs usually attached to the epithelium or connected to other cells through TPs, but exist alone in the epididymis stroma. Moreover, the vast interactions can be used to make a complex TPs network between adjacent epithelial cells, and cell secretions are attached to TPs (F g. A–I).
The morphological structure and the morphological model diagram of TCs in yak epididymis were proposed based on the results of TEM and SEM (Figs. and ).
The toluidine blue staining results showed that a few TCs were distributed in the interstitium and more were distributed near the microvascular of the epididymis. However, they had different structures, adopting ellipsoid, spindle, pear, or other forms (Fig. A–C). Mercury–bromophenol blue staining showed that TPs were more obvious and the cytoplasmic processes on TPs were stained dark blue (Fig. D–F).
The immunohistochemical results showed that thick epithelial cells, mesenchymal cells, and capillaries were observed in the epididymis of yak. CD34 was strongly expressed in the interstitium and distributed mainly in the interstitium and near the epithelium (Fig. A–C). Compared with the caput and cauda, the CD34-positive intensity in the epididymis corpus was higher (Fig. B). Vimentin staining was strong positive in the caput, corpus, and cauda of yak epididymis. It is worth noting that vimentin has intensely positive expression in the stroma, epithelium, and microvascular wall of the epididymis (Fig. D–F). CD117 had strong positive expression in the caput, corpus, and cauda of yak epididymis, and the specificity was stronger than that of CD34 and vimentin. CD117 immunopositive cells were mainly distributed in the epithelial cytoplasm and mesenchymal cells (F g. G–I). The immunofluorescence results showed that vimentin was widely expressed in the epididymis of yak, including fibroblasts, perivascular muscle-like cells and vascular endothelial cells, with strong positive expression (Fig. ). These cells are present in the loose connective tissue around the tubulointerstitium. CD34 was positively distributed in the stroma and epithelium (Fig. ). The CD34-positive oval cells near the epithelium were of a short shape, with cytoplasmic processes, which may be dendritic cells or TCs, and the positive expression was stronger in the caput. The co-expression of vimentin/CD34 was observed in the interstitium, capillary, and epithelium of yak epididymis (Fig. ). There was strong positive expression of CD117 in the stroma and epithelium, with a large number of CD117-positive cells around the blood vessels of mesenchymal hair cells, and relatively few in the epithelium (Fig. ). The cells with strong positive CD34 expression and weakly positive CD117 expression appeared in the yak epididymis corpus; they may be phagocytes or lymphocytes (Fig. ). CD117/CD34 co-expression was observed in the interstitial epididymis, as shown by yellow fluorescence (Fig. ). The common features were proved by the co-expression of vimentin/CD34 and CD117/CD34 and showed that the cells had a long cytoplasmic extension and a large nucleus, and that most of them were distributed around the interstitial capillaries of yak epididymis and outside the epididymal epithelium. The obtained morphology conformed to the basic characteristics expected for TCs.
The expression of CD34, Vimentin, and CD117 mRNA in yak epididymis was detected by qRT-PCR. The expression of CD34 mRNA in caput epididymis tissues of yaks compared to corpus and cauda showed an elevated trend ( p < 0.01). Although there was no significant difference in the expression trend in vimentin mRNA in caput and corpus, it was significantly higher than that in the cauda epididymis ( p < 0.01). Similarly, CD117 mRNA expression in yak caput epididymis was also significantly higher than that in the corpus and cauda ( p < 0.01) (Fig. ). Western blotting results showed that the expression levels of the proteins of CD34, vimentin, and CD117 in the caput epididymis of yak were significantly higher than those in the corpus and cauda (Fig. ); moreover, protein expression trends patterns were consistent with mRNA expression. In summary, the relatively high gene transcription and protein translation levels of TC surface markers in the caput epididymis of yaks were probably closely related to its physiological function.
Homeostasis of the epididymal microenvironment guarantees male animal reproductive ability. As a type of interstitial cell, TCs play an important role in the epididymis immune microenvironment homeostasis and blood–epididymis barrier function. TCs have been found in many animal tissues, such as the human colon , fish brain , and poultry skin, and TCs have common features: slender cytoplasmic extensions and a large nucleus. In addition, TCs were also found in the testis of rats , rabbits , and camels . TCs found in our study have common morphological characteristics with previous studies. The large nucleus is one of the features of TCs, and serrated nuclei of TCs have so far been described only in camel testes, whereas the same was present in the serrated nuclei of the TCs in yak cauda. TEM has been considered to be the most effective way to differentiate TCs from other mesenchymal and epithelial cells . In this study, the TEM ultrastructure of TCs in different locations of the yak epididymis were identified. The TCs distributed around the capillaries or near the basement membrane had differences in morphology. The former has more full cell bodies and TPs are relatively long, whereas the latter had contrasting features. Some scholars have confirmed that TCs around the capillaries may be associated with angiogenesis and material exchange . TCs near the epididymis basement membrane were closely connected with the peritubular myoid cells, which provide structural support for the epididymal duct, and may participate in the contraction of the epididymal duct, providing an impetus for the transport of sperm. The TEM analysis of the ultrastructure revealed that there were multitudinous secretory vesicles in the TPs of yak epididymis TCs; these were released to the surrounding environment of TCs in the form of exocytosis and were easily observed by TEM. Our findings show that secretory vesicles also exist in TCs found in the human testis, myocardium, ovary, and other tissues. Considered the “messengers” of TCs that communicate with the outside world , this feature is regarded as one of the most important conditions for distinguishing TCs from other interstitial cells in ultrastructure. We found TCs in the interstitium of the yak caput epididymis by SEM were mainly distributed in the interstitium and around the blood vessels. The differently sized TPs from TC cell bodies, as well as the network structure composed of dense TPs, can be observed by SEM, which is also one of the important characteristics of TCs. In this study, we explored multiple special staining methods and found that mercury–bromophenol blue stained TPs and the cytoplasmic processes dark blue. The cytoplasmic processes of TCs have the function of secreted proteins, and TCs in the yak epididymis of during the estrous season may be involved in the reproductive regulation of the epididymis through the secretion of related proteins by the cytoplasmic processes of TPs; thus, this suggests the need for further study of the seasonal variation characteristics of TCs. Studies have shown that immunohistochemistry can be used as a basic method to localize TCs . Because there was strong expression of vimentin in the stroma and epithelium of epididymis, the immunohistochemistry and double immunofluorescence co-localization were designed to determine the location of TCs. We found that the cells with double-positive expression of vimentin/CD34 and CD117/CD34 were similar to the TCs phenotype in the stroma and capillary of yak epididymis. These cells had a long cytoplasmic extension and oval nuclei, which were consistent with the basic characteristics of TCs. Similar results were found in testis of humans , dove , and Pelodiscus sinensis . We found that the mRNA and protein expression of CD34, vimentin, and CD117 were relatively high in the yak caput epididymis. The caput epididymis plays an important role in sperm maturation and processing, especially in sperm concentration, maturation, and transport , . Compared with the corpus and cauda, the caput is more like a sperm “processing workshop”, suggesting that high expression of TCs markers in the epididymis caput may be related to sperm processing. Via double immunofluorescence staining, we also found that TCs were associated with dendritic cells, peritubular myoid cells, and lymphocytes through TPs. We believe that there is a specific network structure between TCs and the epithelium and stroma in the epididymis of yak, namely the TCs network. Following a previous discovery that the network structure of TCs in human testis may be related to the blood–testicular barrier , the TCs network of yak epididymis TCs may play an important role in the formation of yak blood–testicular barrier. Studies have found that TCs present in the epididymis of camels are positive for vascular endothelial growth factors , indicating that TCs are probably involved in angiogenesis in the epididymis of the camel. TCs promote angiogenesis by secreting extracellular vesicles containing microRNA , . Angiogenesis is a process by which monolayer endothelial cells (ECs) control the permeability of blood cells through material exchange, and maintain homeostasis of the vascular environment and a series of physiological phenomena leading to the formation of new blood vessels . Microvessels in the epididymis are considered to be the prerequisite for sperm maturation and transport . Interestingly, we found that TPs of TCs were always extremely close to the blood vessel and extended to the vascular wall. Combined with TEM evidence, we speculated that TCs may be a material exchange pump that play an important role in the process of material and energy exchange between interstitial and nutrient vessels in the epididymis. At present, more function of TCs are emerging , , such as injury repair , , vascular regeneration , and cell communication . However, there are few studies of TCs in animal reproduction. Studies have found that the secretory vesicles of TCs in camel testis are affected by seasonal changes, with more secretion in spring and less secretion in summer . There is some relationship between TCs secretion vesicles and camel estrus. However, there is insufficient evidence to prove TCs are involved with the regulation of animal estrus. Some researchers believe that TCs may indirectly affect the secretion and release of androgen by establishing intercellular connections with other interstitial cells to regulate the reproductive activities of male animals . Furthermore, studies reported that TCs express progesterone and estrogen receptors in the female gonadal axis – . The emergence of such studies is constantly drawing TCs into the field of reproductive study. Our study revealed the morphological structure and distribution location of TCs in yak reproductive organs and provided a reference for the study of TCs in animal reproduction in a hypoxic plateau environment.
Animals and sample acquisition The epididymal tissue of adult healthy yak (n = 10; ≥ 3 years) was collected from designated slaughterhouses from July to August in Xining City, Qinghai Province (at an average altitude of 3100 m) China. Based on their anatomical characteristics, epididymis samples were divided into three parts: caput, corpus, and cauda. A portion of each sample was quickly frozen in liquid nitrogen, transported to the laboratory and then stored at −80°C for RNA and protein extraction; the remaining samples were stored in 4% paraformaldehyde and 2.5% glutaraldehyde separately for histological and ultrastructural study. All experimental animals were approved by the Animal Care and Use Committee of the Veterinary College of Gansu Agricultural University (Ratification number: GSAU-Eth-VMC-2021-010), and all methods were performed in accordance with the relevant guidelines and regulations. Drugs and reagents All experimental antibodies were purchased from commercial suppliers. Rabbit polyclonal antibody CD34 (bs-8996R), vimentin (bs-8533R), and CD117 (bs-1005R) were purchased from Beijing BIOSS Antibodies Co., Ltd, China. Goat anti-rabbit IgG H&L (ab150077, Alexa Fluor® 488; ab150079, Alexa Fluor® 647; ab150080, Alexa Fluor® 594) were provided by Abcam,Cambridge, UK. The DAB color reagent kit (PA110) was provided by Beijing TIANGEN Biotechnology Co., Ltd. The immunohistochemical staining kit (SP-0023) used was produced by ZYMED USA, Beijing BIOSS Biotechnology Co., Ltd. ECL Plus ultrasensitive luminescent solution (PE0010) was purchased from Solebao Biotechnology Co., Ltd. Sample preparation and observation Preparation of ordinary samples: Epididymal tissue samples (0.5 × 0.5 × 0.5 cm) were fixed with 4% paraformaldehyde solution and rinsed in running water for 24 h before gradient ethanol dehydration. Subsequently, samples were made transparent with xylene, embedded using an Epon 812 paraffin embedding machine, and 4-μm-thick serial sections were cut. Adjacent slices were used for toluidine blue staining, mercury–bromophenol blue staining, immunohistochemistry and immunofluorescence staining, respectively. SEM sample preparation: The epididymal tissue of yak was cut into 0.2 cm × 0.2 cm × 0.2 cm pieces and fixed with 2.5% glutaraldehyde for 2 days. The tissue was washed four times with 0.1 mol/L phosphate buffer; each wash was 15 min. The samples were treated with 1% OsO4 for 1 h and washed with double-distilled water six times (10 min each wash). Then, the tissue was treated with 2% tannic acid for 30 min and washed with double-distilled water six times (each wash 10 min), and then subjected to gradient ethanol dehydration (30%, 50%, 70%, 80%, 90%, 95%, and 100%; each stage 30 min) and then soaked in isoamyl acetate for 30 min. Tissues were dried by critical point drying, and then the samples were sprayed with gold and observed using a scanning electron microscope. TEM sample preparation: The epididymal tissue of yak fixed in 2.5% glutaraldehyde was cut into small pieces (0.2 cm × 0.2 cm × 0.2 cm) and fixed in 2% osmium tetraoxide at 4°C for 3 h. The pieces were dehydrated with a gradient acetone series (30%, 50%, 70%, 80%, 90%, 95%, and 100%) and then embedded in epoxy resin. Ultrathin sections were prepared and affixed to the copper mesh, stained with uranium acetate and lead citrate, and then examined using a JEM-100CX electron microscope (Japan NEC). Immunohistochemistry and immunofluorescence The epididymal tissues were embedded in paraffin and cut into 4-µm-thick sections. Paraffin sections were dewaxed and dehydrated, repaired with microwave oven antigen retrieval, blocked with 3% H2O2 solution for 10 min, and incubated with goat serum albumin for 15 min. Subsequently, 50 µL rabbit polyclonal antibody (CD34, Vimentin, and CD117) diluted to 1:300 was added to each slide; the negative control consisted of 0.01 mol/L PBS instead of the first antibody. The slides were incubated at 37 °C for 4 h, washed three times with phosphate buffer solution (PBS) (each wash 5 min) and then 50 µL of biotin-labeled goat anti-rabbit IgG working solution was added and the sections were incubated at 37 °C for 15 min and washed three times with PBS (each wash 5 min). Horseradish enzyme-labeled streptavidin solution was added and washed with PBS three times (each wash, 5 min). The DAB color developing solution was added for 5–20 min. Hematoxylin counterstaining was performed for 5 min; then, sections were dehydrated by an alcohol gradient, made transparent with xylene, and sealed with neutral gum. The sections were observed under a microscope. Immunofluorescence staining was performed with the primary antibody; sections were incubated at 37°C for 4 h and rinsed with PBS three times. Subsequent steps were completed in a dark room. Anti-Rabbit IgG H&L AF488 or AF594 (dilution ratio 1:1000) was added, incubated at 37°C for 1 h, and washed five times with PBS (each wash 5 min). Then, the second antibody was added and the sections were incubated at 37°C for 4 h, washed five times in PBS (each wash 5 min). Rabbit Anti-PHD2/AF647 (dilution ratio 1:1000) was added dropwise and incubated at 37°C for 1.5 h. After washing with PBS, DAPI was added dropwise and incubated in a dark room for 10 min. After further washing with PBS, the patch was sealed with a capping agent and the sections were observed under a laser confocal microscope. The negative control consisted of 0.01 mol/L PBS instead of the first antibody. The remaining conditions and steps were the same. qRT-PCR analysis The caput, corpus, and cauda tissues of yak epididymis stored at − 80 °C were removed from storage, and 0.1 g was weighed and placed into a mortar. Liquid nitrogen was added to the grind, and 1 mL Transzol was added to the shock treatment. Then 0.2 mL of chloroform was used to extract RNA and confirm its purity. cDNA was synthesized by RNA reverse transcription, and stored in a refrigerator at − 80 °C for further use. Primer Premier 5.0 software was used (Primer Biosoft International, Palo Alto, USA) was used to design primers; the primer sequence was obtained with reference to the NCBI database ( www.ncbi.nlm.nih.gov ), and the primer information is shown in Table . The β-actin gene was used as the internal reference. qRT-PCR was performed using a Light Cycler 480 thermocycler (Roche, Mannheim, Germany) in a final reaction volume of 20 μL, comprising 1 μL of cDNA, 1 μL of forward primer, 1 μL of reverse primer, 10 μL of 2 × SYBR Green II PCR mix (TaKaRa, Dalian, China), 0.4 μL of ROX reference dye, and 6.6 μL of nuclease-free H2O. The cycling reaction conditions were 95 °C for 30 s; followed by 95 °C, for 5 s, and 60 °C for 30 s each, for a total of 45 cycles. Three replicates were performed for each sample to ensure relative expression accuracy of the target genes. Western blotting analysis From the epididymal tissue of yak stored at − 80 °C, 0.1 g was collected and placed into a mortar. Liquid nitrogen was added and the tissue was ground into a fine powder with a pestle. Then, protein cracking liquid was added and samples were cracked on ice for 3 h after eddy shock. Tissue and cell lysates were centrifuged at 12,000 rpm at 4 °C for 15 min and the supernatant was stored at − 80 °C. Determination of protein concentration by BCA protein assay kit (PC0020, Solarbio Biotechnology Co., Ltd., Beijing, China), and all proteins were diluted to the same concentration. Protein samples (25 μg) were separated by SDS–polyacrylamide gel electrophoresis (SDS-PAGE) using a 5% stacking gel and 12% separating gel. After electrophoresis, the separation gel was cut according to the size of the target protein and referred to the Marker, and the cut target bands were then wet-transferred to the support membrane, and the membrane was incubated with primary antibodies (1:800) at 4 °C overnight and washed with Tris-buffered saline + Tween 20 (TBST). Horseradish peroxidase-labeled goat anti-rabbit IgG was used as the secondary antibody and the incubation was performed for 2 h at 37 °C in TBST buffer for 10 min. The polyvinylidene fluoride membrane was then subjected to chemiluminescence detection. Chemiluminescent substrate solutions A and B were mixed at a ratio of 1:1, and the reaction proceeded at 25 °C. The transfer membrane was photographed for analysis, with β-actin used as the internal reference. Statistical analysis Western blotting data were quantified by ImageJ software (National Institutes of Health, Maryland, USA). The qRT-PCR data were analyzed by the 2 − ΔΔCT method; the obtained results were subjected to the dominance test by SPSS 17.0 statistical software, and the histogram was plotted by GraphPad 9.0 software. Institutional review board statement The study is reported in accordance with ARRIVE guidelines. And all experimental animals were approved by the Animal Care and Use Committee of the Veterinary College of Gansu Agricultural University (Ratification number: GSAU-Eth-VMC-2020-016w).
The epididymal tissue of adult healthy yak (n = 10; ≥ 3 years) was collected from designated slaughterhouses from July to August in Xining City, Qinghai Province (at an average altitude of 3100 m) China. Based on their anatomical characteristics, epididymis samples were divided into three parts: caput, corpus, and cauda. A portion of each sample was quickly frozen in liquid nitrogen, transported to the laboratory and then stored at −80°C for RNA and protein extraction; the remaining samples were stored in 4% paraformaldehyde and 2.5% glutaraldehyde separately for histological and ultrastructural study. All experimental animals were approved by the Animal Care and Use Committee of the Veterinary College of Gansu Agricultural University (Ratification number: GSAU-Eth-VMC-2021-010), and all methods were performed in accordance with the relevant guidelines and regulations.
All experimental antibodies were purchased from commercial suppliers. Rabbit polyclonal antibody CD34 (bs-8996R), vimentin (bs-8533R), and CD117 (bs-1005R) were purchased from Beijing BIOSS Antibodies Co., Ltd, China. Goat anti-rabbit IgG H&L (ab150077, Alexa Fluor® 488; ab150079, Alexa Fluor® 647; ab150080, Alexa Fluor® 594) were provided by Abcam,Cambridge, UK. The DAB color reagent kit (PA110) was provided by Beijing TIANGEN Biotechnology Co., Ltd. The immunohistochemical staining kit (SP-0023) used was produced by ZYMED USA, Beijing BIOSS Biotechnology Co., Ltd. ECL Plus ultrasensitive luminescent solution (PE0010) was purchased from Solebao Biotechnology Co., Ltd.
Preparation of ordinary samples: Epididymal tissue samples (0.5 × 0.5 × 0.5 cm) were fixed with 4% paraformaldehyde solution and rinsed in running water for 24 h before gradient ethanol dehydration. Subsequently, samples were made transparent with xylene, embedded using an Epon 812 paraffin embedding machine, and 4-μm-thick serial sections were cut. Adjacent slices were used for toluidine blue staining, mercury–bromophenol blue staining, immunohistochemistry and immunofluorescence staining, respectively. SEM sample preparation: The epididymal tissue of yak was cut into 0.2 cm × 0.2 cm × 0.2 cm pieces and fixed with 2.5% glutaraldehyde for 2 days. The tissue was washed four times with 0.1 mol/L phosphate buffer; each wash was 15 min. The samples were treated with 1% OsO4 for 1 h and washed with double-distilled water six times (10 min each wash). Then, the tissue was treated with 2% tannic acid for 30 min and washed with double-distilled water six times (each wash 10 min), and then subjected to gradient ethanol dehydration (30%, 50%, 70%, 80%, 90%, 95%, and 100%; each stage 30 min) and then soaked in isoamyl acetate for 30 min. Tissues were dried by critical point drying, and then the samples were sprayed with gold and observed using a scanning electron microscope. TEM sample preparation: The epididymal tissue of yak fixed in 2.5% glutaraldehyde was cut into small pieces (0.2 cm × 0.2 cm × 0.2 cm) and fixed in 2% osmium tetraoxide at 4°C for 3 h. The pieces were dehydrated with a gradient acetone series (30%, 50%, 70%, 80%, 90%, 95%, and 100%) and then embedded in epoxy resin. Ultrathin sections were prepared and affixed to the copper mesh, stained with uranium acetate and lead citrate, and then examined using a JEM-100CX electron microscope (Japan NEC).
The epididymal tissues were embedded in paraffin and cut into 4-µm-thick sections. Paraffin sections were dewaxed and dehydrated, repaired with microwave oven antigen retrieval, blocked with 3% H2O2 solution for 10 min, and incubated with goat serum albumin for 15 min. Subsequently, 50 µL rabbit polyclonal antibody (CD34, Vimentin, and CD117) diluted to 1:300 was added to each slide; the negative control consisted of 0.01 mol/L PBS instead of the first antibody. The slides were incubated at 37 °C for 4 h, washed three times with phosphate buffer solution (PBS) (each wash 5 min) and then 50 µL of biotin-labeled goat anti-rabbit IgG working solution was added and the sections were incubated at 37 °C for 15 min and washed three times with PBS (each wash 5 min). Horseradish enzyme-labeled streptavidin solution was added and washed with PBS three times (each wash, 5 min). The DAB color developing solution was added for 5–20 min. Hematoxylin counterstaining was performed for 5 min; then, sections were dehydrated by an alcohol gradient, made transparent with xylene, and sealed with neutral gum. The sections were observed under a microscope. Immunofluorescence staining was performed with the primary antibody; sections were incubated at 37°C for 4 h and rinsed with PBS three times. Subsequent steps were completed in a dark room. Anti-Rabbit IgG H&L AF488 or AF594 (dilution ratio 1:1000) was added, incubated at 37°C for 1 h, and washed five times with PBS (each wash 5 min). Then, the second antibody was added and the sections were incubated at 37°C for 4 h, washed five times in PBS (each wash 5 min). Rabbit Anti-PHD2/AF647 (dilution ratio 1:1000) was added dropwise and incubated at 37°C for 1.5 h. After washing with PBS, DAPI was added dropwise and incubated in a dark room for 10 min. After further washing with PBS, the patch was sealed with a capping agent and the sections were observed under a laser confocal microscope. The negative control consisted of 0.01 mol/L PBS instead of the first antibody. The remaining conditions and steps were the same.
The caput, corpus, and cauda tissues of yak epididymis stored at − 80 °C were removed from storage, and 0.1 g was weighed and placed into a mortar. Liquid nitrogen was added to the grind, and 1 mL Transzol was added to the shock treatment. Then 0.2 mL of chloroform was used to extract RNA and confirm its purity. cDNA was synthesized by RNA reverse transcription, and stored in a refrigerator at − 80 °C for further use. Primer Premier 5.0 software was used (Primer Biosoft International, Palo Alto, USA) was used to design primers; the primer sequence was obtained with reference to the NCBI database ( www.ncbi.nlm.nih.gov ), and the primer information is shown in Table . The β-actin gene was used as the internal reference. qRT-PCR was performed using a Light Cycler 480 thermocycler (Roche, Mannheim, Germany) in a final reaction volume of 20 μL, comprising 1 μL of cDNA, 1 μL of forward primer, 1 μL of reverse primer, 10 μL of 2 × SYBR Green II PCR mix (TaKaRa, Dalian, China), 0.4 μL of ROX reference dye, and 6.6 μL of nuclease-free H2O. The cycling reaction conditions were 95 °C for 30 s; followed by 95 °C, for 5 s, and 60 °C for 30 s each, for a total of 45 cycles. Three replicates were performed for each sample to ensure relative expression accuracy of the target genes.
From the epididymal tissue of yak stored at − 80 °C, 0.1 g was collected and placed into a mortar. Liquid nitrogen was added and the tissue was ground into a fine powder with a pestle. Then, protein cracking liquid was added and samples were cracked on ice for 3 h after eddy shock. Tissue and cell lysates were centrifuged at 12,000 rpm at 4 °C for 15 min and the supernatant was stored at − 80 °C. Determination of protein concentration by BCA protein assay kit (PC0020, Solarbio Biotechnology Co., Ltd., Beijing, China), and all proteins were diluted to the same concentration. Protein samples (25 μg) were separated by SDS–polyacrylamide gel electrophoresis (SDS-PAGE) using a 5% stacking gel and 12% separating gel. After electrophoresis, the separation gel was cut according to the size of the target protein and referred to the Marker, and the cut target bands were then wet-transferred to the support membrane, and the membrane was incubated with primary antibodies (1:800) at 4 °C overnight and washed with Tris-buffered saline + Tween 20 (TBST). Horseradish peroxidase-labeled goat anti-rabbit IgG was used as the secondary antibody and the incubation was performed for 2 h at 37 °C in TBST buffer for 10 min. The polyvinylidene fluoride membrane was then subjected to chemiluminescence detection. Chemiluminescent substrate solutions A and B were mixed at a ratio of 1:1, and the reaction proceeded at 25 °C. The transfer membrane was photographed for analysis, with β-actin used as the internal reference.
Western blotting data were quantified by ImageJ software (National Institutes of Health, Maryland, USA). The qRT-PCR data were analyzed by the 2 − ΔΔCT method; the obtained results were subjected to the dominance test by SPSS 17.0 statistical software, and the histogram was plotted by GraphPad 9.0 software.
The study is reported in accordance with ARRIVE guidelines. And all experimental animals were approved by the Animal Care and Use Committee of the Veterinary College of Gansu Agricultural University (Ratification number: GSAU-Eth-VMC-2020-016w).
Supplementary Information 1. Supplementary Information 2.
|
Frailty and Cancer: Current Perspectives on Assessment and Monitoring | d14f68a5-f5ae-471f-aa93-dc4dccdb4293 | 10066705 | Internal Medicine[mh] | Every year, 20 million new cancer cases occur worldwide. Incidence rates are low in younger people, but show a steep increase in older adults. In the United States, for example, the incidence in people up to the age of 50 years is less than 500 per 100,000 inhabitants a year. For those over 70 years, however, it is four times higher. Approximately 30% of US patients newly diagnosed with cancer are 65 to 74 years old. Another 25% are 75 years or older. Comparable incidence rates across age groups have been reported for other regions and countries. Unlike younger patients with cancer, vulnerable older subjects are more susceptible to unfavorable health events and medical complications during the clinical course. “Frailty” is an established term to describe aging-associated vulnerability, and it has been recognized as a main obstacle of cancer therapy in patients of advanced age. , With frailty, longer lasting therapeutic success is more difficult to achieve. For example, frailty increases the risk of chemotherapy intolerance and of poorer treatment response. , Patients with frailty undergoing cancer surgery have an increased likelihood of post-operative complications. , Advanced frailty may also pose competing risks of morbidity and mortality independent of cancer and its treatment. The prevalence of frailty in older adults with cancer is around 40–50% with a wide range from 5% to 90% depending on the patient population and the method used to assess frailty. , Over the past two decades, huge efforts have been made in order to optimize the detection and quantification of frailty in such patients. The basic underlying idea was to determine a patient’s individual degree of frailty at baseline (ie, before the start of cancer therapy) and to use this information to adjust the oncological treatment. This includes the decision whether the patient should receive tumor therapy or not as well as the choice of the most adequate treatment modality and regimen (eg, standard versus gentler therapy). The principle of using information from a frailty evaluation to therapeutically target this condition with suitable interventions has been established in geriatric medicine for a long time but just recently adopted to the oncological context. International and national medical societies (eg, International Society of Geriatric Oncology [SIOG], American Society of Clinical Oncology [ASCO]) have developed detailed recommendations for the assessment of frailty in older adults prior to the initiation of cancer therapy. There is a growing understanding that frailty in older patients with cancer is not a static biomarker that just needs to be recorded at a single point in time to make final treatment decisions. , Instead, frailty in such individuals is subject to dynamic changes throughout a patient’s remaining lifespan. This raises the question whether and for what specific purposes frailty should be recorded repeatedly during cancer therapy. Compared to the amount of guidance that is available for the initial frailty assessment in older adults with cancer, there is surprisingly little advice on frailty monitoring so far. This narrative review makes the effort to summarize the latest advances in the conceptualization of frailty in the context of cancer. The focus is on the increasingly important link between frailty evaluation and frailty interventions as well as the re-evaluation of frailty during cancer treatment. A PubMed search was performed by using variations of the following global search term: [cancer OR tumor] AND [frailty OR geriatric] AND [screening OR assessment OR management OR evaluation OR intervention]. Articles published from January 2005 to January 2023 considered relevant to the topic were examined in greater detail. Additionally, we examined guidelines and consensus recommendations that have been published by SIOG and ASCO or other medical societies (eg, National Comprehensive Cancer Network [NCCN]) , on the assessment and management of frailty in older adults with cancer. This literature served as the basis for preparing this review.
Frailty is generally defined as an age-related clinical condition of increased vulnerability to acute endogenous or exogenous stressors. Older adults with frailty are at increased risk to experience worsening of their overall health status due to adverse health events emerging from interactions between existing frailty features and new stressor events . Frailty arises primarily from normal aging and age-related diseases. , , Ordinary aging processes at the molecular and cellular level including senescence and stem cell exhaustion cause a physiological decrease of the reserve capacity of organs and organ systems. , Coincidently, these processes may promote the creation of an environment that fosters the development of pathological tissue degeneration (eg, vascular, musculoskeletal) and subclinical inflammation which lead to common aging-associated diseases (eg, coronary heart disease, osteoarthritis) followed by a possible decline in physical and mental function and disability. Later, the cancer disease and damage caused by its treatment may contribute to frailty. , In older patients with cancer, comorbidity (ie, the burden of chronic illness), disability (ie, the loss of function and autonomy), and frailty often co-exist. They are considered as overlapping although not identical phenomena. There is also significant overlap between the concepts of frailty and “intrinsic capacity”. From a simplified view, intrinsic capacity and resiliency can be understood as the opposite to vulnerability and frailty . Two methods are generally recognized as the gold standard to identify frailty in older people. , The “Fried frailty criteria” are based on a phenotype model. In this model, presence of at least 3 of the following 5 criteria in a patient indicate frailty: low physical activity, poor endurance (self-reported exhaustion), weakness (reduced grip strength), slowness (decreased walking speed), and unintentional weight loss. However, the use of this approach to identify frailty in routine clinical care has remained uncommon, because there is no generally accepted and clinically easily applicable operationalization of the individual criteria. The “Rockwood frailty index” is based on a deficit-accumulation model. The index is calculated by dividing the number of deficits diagnosed by the total number of 70 pre-defined deficits. Deficit items include various diseases, signs from clinical examinations, and impairments of activities of daily living. In contrast to the phenotype model, deficit-accumulation models not just allow to determine whether frailty is present or not (categorical variable), but also to quantify the extent of frailty in a patient (continuous variable). With the 70-item model, an index of 0.25 and above may indicate frailty. In routine care, however, an index calculation with 70 or even with fewer deficit items (eg, 50 or 20) has proven to be too cumbersome to capture frailty. This approach has therefore not become more widely established in clinical practice. In the oncology context, a minority of studies have used Fried criteria or the Rockwood index to identify and measure frailty in older cancer patients. , Techniques other than Fried criteria or the Rockwood index have been accepted as appropriate for detecting frailty in older people. These include frailty screenings and geriatric assessment (GA). Both methods are easier to implement in everyday clinical care. Among numerous frailty screening tools (eg, Identification of Seniors at Risk [ISAR], Groningen Frailty Indicator [GFI], Vulnerable Elders Survey-13 [VES-13], Triage Risk Screening Tool [TRST]), the Clinical Frailty Scale (CFS) has recently received greater attention and increasingly been used in clinical settings during the Covid pandemic. This pictogram-driven screening tool summarizes the overall level of frailty of an older person and is easy for clinicians to use as part of their medical history taking and physical examination. Due to its simple structure, CFS is also suitable in situations with acute or new illness to record the previous level of frailty before the new stressor disease has occurred (eg, a symptomatic Covid-19 infection). Such information is highly relevant and helps to avoid under-treatment or over-treatment of older patients when it comes to far-reaching treatment decisions (eg, for or against ventilation therapy in the example of Covid-19). Independent of the pandemic, this principle can also be applied to the oncological context. In old patients with newly diagnosed cancer who present in poor general condition, knowledge of frailty before the onset of the tumor disease is very important when anticipating the prospect of tumor-specific therapy of re-improving the condition. The number of studies examining CFS in the oncology setting has increased over the past 1–2 years. A majority was conducted in the context of tumor surgery. Results were promising regarding the usefulness of this tool to predict outcomes . GA is a core methodology in geriatric medicine. , Its use for a systematic and comprehensive recording of vulnerabilities in older patients across geriatric domains (social support - activities of daily living - mobility and falls - nutrition - cognition - emotion - sleep - vision and hearing - pain and wounds - co-morbidity - polypharmacy) is firmly anchored in routine geriatric care. GA makes use of traditional assessment tools that have been tried and tested over many years. Among these are scores and scales (eg, Lawton scale for instrumental activities of daily living, Katz scale for basal activities of daily living, Charlson score for comorbidities) as well as performance tests (eg, Timed-Up&Go test for mobility, Mini Mental State Exam for cognition). Notably, new approaches are currently emerging to perhaps replace parts of a GA using modern sensor-based diagnostics (eg, wearable sensors, apps). Over recent years, medical disciplines other than geriatrics, such as cardiology or trauma surgery, have begun to discover GA as a potentially useful technique specific to their field. , Oncology is at the forefront of this development.
A fundamental goal of frailty assessment and management in older adults with cancer is to protect them from adverse health outcomes as well as possible . Unfavorable outcomes to be prevented in general must be distinguished from those that are specific to the oncological context . Practical recommendations on frailty evaluation and interventions in older patients with cancer are based on clinical studies examining whether an assessment was able to predict the occurrence or reduced the incidence of such outcomes. Approaches that were subject to these studies included performance scores, frailty screening, and geriatric assessment. The validity of these approaches can be summarized as follows. Validity of Performance Scores In routine care, oncologists often use Eastern Cooperative Oncology Group or Karnofsky performance score (ECOG PS, KPS) to roughly estimate the general condition of their patients. , ECOG PS and KPS are tools to describe the overall health status and global activity level of cancer patients. However, poor ECOG PS or KPS numbers may be observed in already chronically ill patients with severe pre-existing diseases (eg, terminal COPD, progressive Parkinson disease) as well as in otherwise healthy patients but with acute symptomatic cancer illness (eg, cancer fatigue, cancer pain, acute infection). However, good ECOG PS or KPS numbers do not rule out the existence of chronic and clinically relevant vulnerabilities in older patients (eg, tendency to fall, mild cognitive impairment, inappropriate polypharmacy). The capacity of ECOG PS to predict chemotherapy toxicity in older cancer patients is low. Performance scores such as ECOG PS or KPS are therefore insufficient to comprehensively surrogate frailty in older adults with cancer. Neither do these tools allow for differentiated oncological treatment decisions nor for the selection of meaningful frailty interventions. Validity of Frailty Screenings Most recommendations of international and national medical societies have in common that frailty assessment in older adults with cancer should start with a quick screening in order to identify those patients who are presumably vulnerable and could benefit from a comprehensive GA. lists the screening tools proposed by SIOG and ASCO. The advice is based on study results. The available evidence has been compiled in several systematic reviews . The most frequently applied tool is the so-called G8 (Geriatrics 8) screening. Numerous studies in older patients with cancer demonstrated associations between abnormal G8 scores and poor frailty outcomes such as shortened survival, increased treatment toxicity and complications after tumor surgery . , The G8 tool queries key vulnerabilities (mobility issues, nutritional issues, cognitive issues and mood problems, and polypharmacy) together with age and subjective health perception. During routine oncological work-ups, such frailty features are usually not checked systematically. Meanwhile, a self-reported version of the G8 has been made available to facilitate its implementation in busy clinics. A G8 screening score of ≤14 identifies patients with possibly increased vulnerability and should prompt for a comprehensive GA. Importantly, abnormal results of G8 or other frailty screenings such as VES-13 should not be used alone to mark a patient as being “frail”, because false positive screening may occur. G8 has a high sensitivity albeit lower specificity. In contrast, VES-13 has lower sensitivity but higher specificity than G8. Of note, frailty must not be considered as an absolute measure, but always judged in relation to the level of stress burdened on the patient by an offered cancer treatment. The level of stress may vary depending on the type of the tumor therapy (eg, major surgery, high-dose chemotherapy, hematopoietic stem cell transplantation vs mild chemotherapy, immunotherapy, oral hormone therapy etc.). Therefore, G8 or VES-13 are well suited in distinguishing between presumably robust and vulnerable patients. However, these tools have not been sufficiently validated regarding their utility to tailor final oncological treatment or frailty intervention plans. For these purposes, greater knowledge about single vulnerabilities is required including information on their severeness, underlying causes, possible consequences in the course of the cancer disease and treatment, and intervenability. Utility of Geriatric Assessment SIOG and ASCO strongly recommend to perform comprehensive GA in cancer patients ≥ 65–70 years who were identified as presumably vulnerable by prior frailty screening. The SIOG recommendations are not targeted to a specific subset of older cancer patients. , The ASCO guideline refers to the subset of older adults receiving chemotherapy. Both guidelines uniformly recommend that the GA should cover essential geriatric domains. The minimum is physical function including instrumental and basic activities of daily living (IADL, ADL), mobility, nutrition, cognition, mood, co-morbidity, and co-medications. For each geriatric domain, the guidelines propose a set of GA instruments . To date, a plethora of retrospective and prospective, non-comparative and comparative GA studies in older patients with cancer has been published. The accumulated study evidence has been summarized in several systematic reviews and is the basis for the current recommendations made by SIOG and ASCO or for other country-specific guidelines (eg, NCCN). In general, studies of GA in older adults with cancer were highly heterogenous regarding oncological settings, patient populations, sizes, and endpoints . Many of these studies examined one of the following aspects: Feasibility of GA in older cancer patients. Prevalence of geriatric impairments (as assessed by GA) in older cancer patients. Association of geriatric impairments with treatment complications such as toxicity, dose modifications, treatment discontinuation, length of hospital stay, or unplanned hospitalization. Association of geriatric impairments with cancer treatment efficacy endpoints such as response rates or progression-free survival. Association of geriatric impairments with overall survival (mostly all-cause mortality). Impact of performing a GA (vs not performing GA) on communication with patients and their caregivers. Impact of performing a GA (vs not performing GA) on oncological or non-oncological treatment decisions. Impact of performing GA (vs not performing GA) on outcomes such as treatment complications or survival. Impact of performing GA with vs without geriatric management (frailty interventions) on outcomes. Overall, it can be concluded from these studies that GA uncovers geriatric impairments, predicts treatment tolerability and feasibility, predicts (all-cause) mortality, facilitates communication about treatment goals and preferences, results in changes of oncological treatment plans, and enables targeted non-oncological treatment of geriatric impairments in older cancer patients . The ability of GA to predict treatment complications was exploited by developing chemotoxicity risk calculators. These tools incorporated GA elements and allow to calculate the likelihood of grade 3–5 toxicity during chemotherapy of older patients with cancer. Use of such calculators is strongly recommended by the ASCO guideline addressing the subset of chemotherapy-treated older patients. The CARG (Cancer Aging Research Group) and the CRASH (Chemotherapy Risk Assessment Score for High Age Patients) tool are easily accessible online. , Both tools take functional impairments of the patient into account and thus require careful geriatric examination of the patient. Results of selected studies investigating CARG or CRASH in older patients with cancer are outlined in . It should be noted that neither the CARG nor the CRASH score have been validated in greater detail for their capacity to predict toxicity of non-chemotherapeutic agents (eg, kinase inhibitors, immune checkpoint inhibitors). Recent pivotal randomized-controlled trials (RCTs) investigating the impact of GA with and without subsequent geriatric management added very compelling evidence that GA is a powerful frailty assessment for older patients with cancer. , These RCTs demonstrated that GA-directed management of vulnerabilities is able to reduce the risk of these patients to experience toxicity or premature discontinuation of systemic cancer treatment. In the two largest RCTs (GAP70+ and GAIN study) with more than 600 patients each, rates of grade 3–5 toxicity were reduced by 20% and 10%, respectively. , The smaller INTEGERATE trial reported lower chemotherapy discontinuation rates with integrated GA-guided oncogeriatric care compared with usual care (33% vs 53%). In a RCT in colorectal cancer patients with tumor surgery followed by chemotherapy (GERICO), more patients in the oncogeriatric intervention arm completed scheduled chemotherapy compared with patients of the control arm (45% vs 28%). More details for RCTs examining GA with or without geriatric management are shown in . Of note, there have also been trials which did not meet their primary endpoint. Comprehensive delivery of GA-guided frailty interventions ideally happens within a multidisciplinary approach involving social workers, nurses/nurse practitioners, physiotherapists, occupational therapists, dieticians, psychologists, pharmacists, and geriatricians in addition to oncologists, radiotherapists, and surgeons. , shows a list of interventions used in GAP70+, GAIN, INTEGERATE, and GERICO with the intention to improve single vulnerabilities and hence to modify the overall frailty level of older cancer patients over time. Next to the measurement of oncological outcome improvements (eg, decreased treatment toxicity), successful frailty intervention may also be validated by measuring whether vulnerabilities captured at baseline improve during further follow-up. However, except for the GAP70+ trial, none of the randomized frailty intervention trials shown in included a geriatric re-assessment during or after the cancer treatment.
In routine care, oncologists often use Eastern Cooperative Oncology Group or Karnofsky performance score (ECOG PS, KPS) to roughly estimate the general condition of their patients. , ECOG PS and KPS are tools to describe the overall health status and global activity level of cancer patients. However, poor ECOG PS or KPS numbers may be observed in already chronically ill patients with severe pre-existing diseases (eg, terminal COPD, progressive Parkinson disease) as well as in otherwise healthy patients but with acute symptomatic cancer illness (eg, cancer fatigue, cancer pain, acute infection). However, good ECOG PS or KPS numbers do not rule out the existence of chronic and clinically relevant vulnerabilities in older patients (eg, tendency to fall, mild cognitive impairment, inappropriate polypharmacy). The capacity of ECOG PS to predict chemotherapy toxicity in older cancer patients is low. Performance scores such as ECOG PS or KPS are therefore insufficient to comprehensively surrogate frailty in older adults with cancer. Neither do these tools allow for differentiated oncological treatment decisions nor for the selection of meaningful frailty interventions.
Most recommendations of international and national medical societies have in common that frailty assessment in older adults with cancer should start with a quick screening in order to identify those patients who are presumably vulnerable and could benefit from a comprehensive GA. lists the screening tools proposed by SIOG and ASCO. The advice is based on study results. The available evidence has been compiled in several systematic reviews . The most frequently applied tool is the so-called G8 (Geriatrics 8) screening. Numerous studies in older patients with cancer demonstrated associations between abnormal G8 scores and poor frailty outcomes such as shortened survival, increased treatment toxicity and complications after tumor surgery . , The G8 tool queries key vulnerabilities (mobility issues, nutritional issues, cognitive issues and mood problems, and polypharmacy) together with age and subjective health perception. During routine oncological work-ups, such frailty features are usually not checked systematically. Meanwhile, a self-reported version of the G8 has been made available to facilitate its implementation in busy clinics. A G8 screening score of ≤14 identifies patients with possibly increased vulnerability and should prompt for a comprehensive GA. Importantly, abnormal results of G8 or other frailty screenings such as VES-13 should not be used alone to mark a patient as being “frail”, because false positive screening may occur. G8 has a high sensitivity albeit lower specificity. In contrast, VES-13 has lower sensitivity but higher specificity than G8. Of note, frailty must not be considered as an absolute measure, but always judged in relation to the level of stress burdened on the patient by an offered cancer treatment. The level of stress may vary depending on the type of the tumor therapy (eg, major surgery, high-dose chemotherapy, hematopoietic stem cell transplantation vs mild chemotherapy, immunotherapy, oral hormone therapy etc.). Therefore, G8 or VES-13 are well suited in distinguishing between presumably robust and vulnerable patients. However, these tools have not been sufficiently validated regarding their utility to tailor final oncological treatment or frailty intervention plans. For these purposes, greater knowledge about single vulnerabilities is required including information on their severeness, underlying causes, possible consequences in the course of the cancer disease and treatment, and intervenability.
SIOG and ASCO strongly recommend to perform comprehensive GA in cancer patients ≥ 65–70 years who were identified as presumably vulnerable by prior frailty screening. The SIOG recommendations are not targeted to a specific subset of older cancer patients. , The ASCO guideline refers to the subset of older adults receiving chemotherapy. Both guidelines uniformly recommend that the GA should cover essential geriatric domains. The minimum is physical function including instrumental and basic activities of daily living (IADL, ADL), mobility, nutrition, cognition, mood, co-morbidity, and co-medications. For each geriatric domain, the guidelines propose a set of GA instruments . To date, a plethora of retrospective and prospective, non-comparative and comparative GA studies in older patients with cancer has been published. The accumulated study evidence has been summarized in several systematic reviews and is the basis for the current recommendations made by SIOG and ASCO or for other country-specific guidelines (eg, NCCN). In general, studies of GA in older adults with cancer were highly heterogenous regarding oncological settings, patient populations, sizes, and endpoints . Many of these studies examined one of the following aspects: Feasibility of GA in older cancer patients. Prevalence of geriatric impairments (as assessed by GA) in older cancer patients. Association of geriatric impairments with treatment complications such as toxicity, dose modifications, treatment discontinuation, length of hospital stay, or unplanned hospitalization. Association of geriatric impairments with cancer treatment efficacy endpoints such as response rates or progression-free survival. Association of geriatric impairments with overall survival (mostly all-cause mortality). Impact of performing a GA (vs not performing GA) on communication with patients and their caregivers. Impact of performing a GA (vs not performing GA) on oncological or non-oncological treatment decisions. Impact of performing GA (vs not performing GA) on outcomes such as treatment complications or survival. Impact of performing GA with vs without geriatric management (frailty interventions) on outcomes. Overall, it can be concluded from these studies that GA uncovers geriatric impairments, predicts treatment tolerability and feasibility, predicts (all-cause) mortality, facilitates communication about treatment goals and preferences, results in changes of oncological treatment plans, and enables targeted non-oncological treatment of geriatric impairments in older cancer patients . The ability of GA to predict treatment complications was exploited by developing chemotoxicity risk calculators. These tools incorporated GA elements and allow to calculate the likelihood of grade 3–5 toxicity during chemotherapy of older patients with cancer. Use of such calculators is strongly recommended by the ASCO guideline addressing the subset of chemotherapy-treated older patients. The CARG (Cancer Aging Research Group) and the CRASH (Chemotherapy Risk Assessment Score for High Age Patients) tool are easily accessible online. , Both tools take functional impairments of the patient into account and thus require careful geriatric examination of the patient. Results of selected studies investigating CARG or CRASH in older patients with cancer are outlined in . It should be noted that neither the CARG nor the CRASH score have been validated in greater detail for their capacity to predict toxicity of non-chemotherapeutic agents (eg, kinase inhibitors, immune checkpoint inhibitors). Recent pivotal randomized-controlled trials (RCTs) investigating the impact of GA with and without subsequent geriatric management added very compelling evidence that GA is a powerful frailty assessment for older patients with cancer. , These RCTs demonstrated that GA-directed management of vulnerabilities is able to reduce the risk of these patients to experience toxicity or premature discontinuation of systemic cancer treatment. In the two largest RCTs (GAP70+ and GAIN study) with more than 600 patients each, rates of grade 3–5 toxicity were reduced by 20% and 10%, respectively. , The smaller INTEGERATE trial reported lower chemotherapy discontinuation rates with integrated GA-guided oncogeriatric care compared with usual care (33% vs 53%). In a RCT in colorectal cancer patients with tumor surgery followed by chemotherapy (GERICO), more patients in the oncogeriatric intervention arm completed scheduled chemotherapy compared with patients of the control arm (45% vs 28%). More details for RCTs examining GA with or without geriatric management are shown in . Of note, there have also been trials which did not meet their primary endpoint. Comprehensive delivery of GA-guided frailty interventions ideally happens within a multidisciplinary approach involving social workers, nurses/nurse practitioners, physiotherapists, occupational therapists, dieticians, psychologists, pharmacists, and geriatricians in addition to oncologists, radiotherapists, and surgeons. , shows a list of interventions used in GAP70+, GAIN, INTEGERATE, and GERICO with the intention to improve single vulnerabilities and hence to modify the overall frailty level of older cancer patients over time. Next to the measurement of oncological outcome improvements (eg, decreased treatment toxicity), successful frailty intervention may also be validated by measuring whether vulnerabilities captured at baseline improve during further follow-up. However, except for the GAP70+ trial, none of the randomized frailty intervention trials shown in included a geriatric re-assessment during or after the cancer treatment.
Following the initial frailty assessment at the start of a cancer therapy, the overall frailty level as well as single vulnerabilities may undergo significant changes during treatment and throughout a patient’s further life . Over time, alterations of the social situation, the physical and mental functionality, and co-morbidities may occur. Observations from studies suggest that both deterioration and improvements are possible. For example, in a study of 144 over 50 years old breast cancer patients examined for frailty by using modified Fried criteria before and after chemotherapy, the proportion of subjects with a Fried score of 3 or 4 increased from 13% to 46%. Another study with 439 older adults with cancer (≥70 years) found functional declines in IADL and ADL in about one third of the patients during chemotherapy. The large GOSAFE study, which included more than 1000 participants, explored functional recovery after cancer surgery. In individual patients of this study, both loss and gain of function were observed at 3 and 6 months follow-up. There have also been some studies suggesting loss of cognitive capacity (memory function) in patients after chemotherapy. Other analyses demonstrated that cancer survivors develop frailty earlier and more frequently compared to non-cancer survivors. In the GAP70+ trial, delivery of targeted frailty interventions to older patients receiving chemotherapy resulted in lower numbers of falls and lower numbers of prescribed drugs while physical functioning (IADL, mobility) and mood remained unchanged. However, the total number of studies exploring frailty over time in older cancer patients have remained rather low and the available data are of descriptive nature. Therefore, underlying mechanisms of worsening or improvement of frailty must be inferred from clinical observations rather than from systematic study evidence. Clinical experience suggests that progressive tumor disease, toxicity by the tumor treatment, and exacerbation or progression of chronic conditions or acute intercurrent diseases independent of the cancer could drive deterioration of frailty, whereas improvements may occur in response to the remission of a tumor or targeted frailty interventions. Although frailty is subject to dynamic changes and actively modifiable the currently available recommendations by SIOG and ASCO on frailty assessment in older cancer patients are lacking further advice on how to monitor this condition during or after the tumor treatment. At present, this topic must be discussed in the absence of broader evidence or guidance. Potential Indications for Frailty Monitoring Frailty monitoring may play a role in clinical research as well as in clinical practice: In RCTs investigating cancer treatments, oncology-specific outcomes such as response rates, progression-free and overall survival, adverse events, and quality of life are typical study endpoints. In contrast, re-evaluation of geriatric domains such as physical or cognitive functioning or nutritional status post-treatment have remained uncommon in such trials, even when studying a population of older patients. Nevertheless, the inclusion of these endpoints in future RCTs is essential to better understand the risks and benefits of oncological therapies in vulnerable older patients. In routine care, re-performing frailty assessments after the start of an oncological treatment together with non-oncological frailty interventions would allow an evaluation of whether and to what degree a patient is responsive to such management. This would enable a multidisciplinary team to decide whether frailty interventions should be continued, escalated, de-escalated, or stopped, or whether the focus of frailty interventions must be shifted towards other vulnerabilities than those considered crucial at the beginning of the tumor treatment. Furthermore, frailty monitoring over time may guide oncologists to increase or decrease the intensity (eg, dosing) of the cancer treatment. To date, however, there has been no study exploring the utility of repeated frailty assessments to guide continuous adaptation of cancer treatment after its initiation. In routine practice, re-doing a frailty assessment also appears useful if the overall health status of an old patient suddenly deteriorates during systemic cancer treatment and the patient gets hospitalized in an unplanned manner . In addition to the standard investigations of the primary cause of the hospitalization (eg, an infection), a careful holistic view on vulnerabilities at this time point could support decision-making about the need of early rehabilitation measures in the hospital, for example on an acute geriatric ward. Finally, re-assessment of frailty after the end of cancer treatment could inform rehabilitation measures as well as reasonable rehabilitation goals. , This approach seems particularly useful after tumor surgery, radiotherapy, or adjuvant systemic therapy. Traditional Tools for Frailty Monitoring Various tools may be used to monitor a patient’s general frailty level as well as his or her individual vulnerabilities . Performance scores (ECOG PS, KPS) are commonly used in oncology to follow the general condition and activity level of patients. , However, these tools do not provide deeper insight into the course of single vulnerabilities. The same applies to frailty screenings such as G8 that was not designed to monitor frailty. , However, some screening tools (eg, CFS) may be usable and easy to implement in workflows. Repeated GA using standard assessments are another option for frailty monitoring in older cancer patients. There is no consensus whether the entire GA should be repeated or monitoring of selected GA domains (eg, those that showed abnormal results on initial assessment) is sufficient. Of note, neither the CARG nor the CRASH tool have been studied with regards to monitoring the risk of chemotherapy toxicity over time (eg, between treatment cycles or prior and after rehabilitation). Smart Digital Tools for Frailty Monitoring Frailty monitoring in older adults with cancer could be an application for wearable sensors and other digital assessment technology . Meanwhile, the technological progress allows for the collection of a multitude of data by portable devices such as wristwatches, foot pods, breast belts, and smartphones worn on the body and equipped with apps. Numerous manufacturers offer such devices fully configured and ready to use for end users. Using commercial activity trackers, physical activity and movement behavior including number of steps, falls, etc. can be tracked in real time. Moreover, these devices are increasingly capable of recording sleeping behavior as well as circulatory and respiratory parameters as for example heart rate, oxygen saturation, and body temperature. Surrogates of cardiopulmonary capacity (eg, VO2max) can be calculated and monitored over time. Downloadable apps for smart phones and tablets offer new opportunities to follow frailty domains other than physical activity and cardiopulmonary reserve. For example, users can repeatedly enter information about their drinking and eating habits, calorie intake or body weight into apps offered commercially in app stores. In the future, apps might also allow us to remotely query mood and drive, or to follow cognitive function through repeated app-embedded cognitive testing. A growing number of studies examine whether these features of new smart technologies can help to measure frailty in patients at least as well or even better than standard GA instruments or than the Fried criteria or the Rockwood index. , The majority of these studies is aimed to develop digital biomarkers for physical frailty, but there have also been some studies exploring the role of digital sensors to assess other aspects of frailty (eg, cognitive frailty). Unfortunately, the feasibility and usefulness of such approaches have only been little investigated in oncological settings. So far, there is only rudimentary data available. For example, in a study with 84 older cancer patients (median age 71 years), gait and balance parameters assessed by wearables were degraded when compared with age-matched non-cancer patients as well as in patients with chemotherapy-induced peripheral neuropathy (CIPN) versus without CIPN. Another trial explored a wearable activity tracker in somewhat younger patients with cancer undergoing chemotherapy. This study reported a correlation between unplanned health encounters and tracker-recorded activity data, but not ECOG PS. Advantages and disadvantages must be weighed against each other when using wearable sensors and apps to assess and monitor frailty in older patients with cancer. Although many products might be available off-the-shelf and consumer-ready (eg, smart watches, app stores), there is no broader accepted standard device and no consented protocol regarding the processing, safe storage and transmission of these digital data. For the moment, the lack of technical standards as well as data protection rules may limit a broader application of these tools for assessing and monitoring frailty. However, if such hurdles are overcome, many new possibilities open up. For example, the data might be transmitted to oncological practitioners or a multiprofessional team responsible for frailty interventions. These caregivers could be alerted in real-time if a patient’s frailty level deteriorates or improves, and may enable them to immediately adjust therapeutic approaches to the new frailty situation.
Frailty monitoring may play a role in clinical research as well as in clinical practice: In RCTs investigating cancer treatments, oncology-specific outcomes such as response rates, progression-free and overall survival, adverse events, and quality of life are typical study endpoints. In contrast, re-evaluation of geriatric domains such as physical or cognitive functioning or nutritional status post-treatment have remained uncommon in such trials, even when studying a population of older patients. Nevertheless, the inclusion of these endpoints in future RCTs is essential to better understand the risks and benefits of oncological therapies in vulnerable older patients. In routine care, re-performing frailty assessments after the start of an oncological treatment together with non-oncological frailty interventions would allow an evaluation of whether and to what degree a patient is responsive to such management. This would enable a multidisciplinary team to decide whether frailty interventions should be continued, escalated, de-escalated, or stopped, or whether the focus of frailty interventions must be shifted towards other vulnerabilities than those considered crucial at the beginning of the tumor treatment. Furthermore, frailty monitoring over time may guide oncologists to increase or decrease the intensity (eg, dosing) of the cancer treatment. To date, however, there has been no study exploring the utility of repeated frailty assessments to guide continuous adaptation of cancer treatment after its initiation. In routine practice, re-doing a frailty assessment also appears useful if the overall health status of an old patient suddenly deteriorates during systemic cancer treatment and the patient gets hospitalized in an unplanned manner . In addition to the standard investigations of the primary cause of the hospitalization (eg, an infection), a careful holistic view on vulnerabilities at this time point could support decision-making about the need of early rehabilitation measures in the hospital, for example on an acute geriatric ward. Finally, re-assessment of frailty after the end of cancer treatment could inform rehabilitation measures as well as reasonable rehabilitation goals. , This approach seems particularly useful after tumor surgery, radiotherapy, or adjuvant systemic therapy.
Various tools may be used to monitor a patient’s general frailty level as well as his or her individual vulnerabilities . Performance scores (ECOG PS, KPS) are commonly used in oncology to follow the general condition and activity level of patients. , However, these tools do not provide deeper insight into the course of single vulnerabilities. The same applies to frailty screenings such as G8 that was not designed to monitor frailty. , However, some screening tools (eg, CFS) may be usable and easy to implement in workflows. Repeated GA using standard assessments are another option for frailty monitoring in older cancer patients. There is no consensus whether the entire GA should be repeated or monitoring of selected GA domains (eg, those that showed abnormal results on initial assessment) is sufficient. Of note, neither the CARG nor the CRASH tool have been studied with regards to monitoring the risk of chemotherapy toxicity over time (eg, between treatment cycles or prior and after rehabilitation).
Frailty monitoring in older adults with cancer could be an application for wearable sensors and other digital assessment technology . Meanwhile, the technological progress allows for the collection of a multitude of data by portable devices such as wristwatches, foot pods, breast belts, and smartphones worn on the body and equipped with apps. Numerous manufacturers offer such devices fully configured and ready to use for end users. Using commercial activity trackers, physical activity and movement behavior including number of steps, falls, etc. can be tracked in real time. Moreover, these devices are increasingly capable of recording sleeping behavior as well as circulatory and respiratory parameters as for example heart rate, oxygen saturation, and body temperature. Surrogates of cardiopulmonary capacity (eg, VO2max) can be calculated and monitored over time. Downloadable apps for smart phones and tablets offer new opportunities to follow frailty domains other than physical activity and cardiopulmonary reserve. For example, users can repeatedly enter information about their drinking and eating habits, calorie intake or body weight into apps offered commercially in app stores. In the future, apps might also allow us to remotely query mood and drive, or to follow cognitive function through repeated app-embedded cognitive testing. A growing number of studies examine whether these features of new smart technologies can help to measure frailty in patients at least as well or even better than standard GA instruments or than the Fried criteria or the Rockwood index. , The majority of these studies is aimed to develop digital biomarkers for physical frailty, but there have also been some studies exploring the role of digital sensors to assess other aspects of frailty (eg, cognitive frailty). Unfortunately, the feasibility and usefulness of such approaches have only been little investigated in oncological settings. So far, there is only rudimentary data available. For example, in a study with 84 older cancer patients (median age 71 years), gait and balance parameters assessed by wearables were degraded when compared with age-matched non-cancer patients as well as in patients with chemotherapy-induced peripheral neuropathy (CIPN) versus without CIPN. Another trial explored a wearable activity tracker in somewhat younger patients with cancer undergoing chemotherapy. This study reported a correlation between unplanned health encounters and tracker-recorded activity data, but not ECOG PS. Advantages and disadvantages must be weighed against each other when using wearable sensors and apps to assess and monitor frailty in older patients with cancer. Although many products might be available off-the-shelf and consumer-ready (eg, smart watches, app stores), there is no broader accepted standard device and no consented protocol regarding the processing, safe storage and transmission of these digital data. For the moment, the lack of technical standards as well as data protection rules may limit a broader application of these tools for assessing and monitoring frailty. However, if such hurdles are overcome, many new possibilities open up. For example, the data might be transmitted to oncological practitioners or a multiprofessional team responsible for frailty interventions. These caregivers could be alerted in real-time if a patient’s frailty level deteriorates or improves, and may enable them to immediately adjust therapeutic approaches to the new frailty situation.
The evidence for benefits of a frailty assessment in older adults with cancer has significantly increased in recent years. Most importantly, recent prospective, randomized-controlled studies have demonstrated that frailty assessment improves the outcome of such patients. Frailty assessment followed by frailty interventions significantly enhances the treatment tolerability and feasibility, particularly in elderly patients receiving systemic cancer therapy. The number needed to treat is relatively low at around 5–10. , These new data underscore that frailty assessment is not meant to exclude patients with pre-identified vulnerabilities from cancer therapy, but to make oncological treatments in these patients as safe as possible through additional supportive measures. Based on these new data, performing a comprehensive GA (ie, GA with GA-guided interventions) is at the edge of becoming mandatory in older adults with cancer. However, despite the high level of evidence, only a minority of cancer centers worldwide have integrated GA and GA-led interventions into the routine care of elderly cancer patients so far. The implementation barriers are diverse and include lack of knowledge, limited human, temporal and spatial resources, and billing and reimbursement problems. In addition to a broad implementation of frailty assessments before starting cancer therapy, there is also a growing need to follow-up frailty in older cancer patients in the course of their disease and treatment. Modern digital technologies such as wearable sensors and apps may offer new ways to simplify and advance frailty assessment and monitoring in this patient population. However, evidence in the oncology context remains low. In the future, such approaches could perhaps replace or supplement parts of a GA, thereby reducing the need of resources. This review is the first to address the issue of continuous frailty assessment in elderly patients with cancer in more detail. Further studies are needed to expand the evidence base. In such studies, the following key questions should be examined as a matter of priority: Which frailty trajectories are particularly common and typical in older cancer patients receiving a particular treatment? How do frailty interventions modify such trajectories and how can the success or failure of these interventions be predicted in individual patients? What tools should be used as a standard to determine changes in frailty in individual patients during cancer treatment and frailty interventions? How can frailty assessment and monitoring be improved by new smart technologies in older patients with cancer?
|
Learning in a real-life escape room: an explorative study on the supervisory relationship in GP residency during the COVID-19 pandemic | ad808f4d-38c2-45eb-8d52-9574d63b678d | 10066972 | Family Medicine[mh] | The effects of war, climate change and emerging epidemics can create disruptive situations in healthcare and medical education . When these disruptions occur, it is necessary to consider what this means for the training of the new generation of doctors and determine whether residents training is still satisfactory . In this study, we explored the nature of the supervisory relationship during COVID-19. General practice residents make the transition from novice doctors to fully qualified general practitioners (GPs) through participation in clinical care under the supervision of senior GPs . Throughout resident training, the relationship with the supervisor is fundamental to residents’ acquisition of knowledge and skills, and the development of their professional identity . This relationship enables them to manage clinical uncertainty and risk, maintain patient confidence, and take initial clinical responsibility for patient care, which is fundamental to functioning as autonomous GPs . The supervisory relationship also plays an important role in balancing the tensions between patient safety and resident learning and autonomy in workplace learning . Workplace learning can be understood, according to Billet’s co-participation theory, as the interrelationship between workplace affordances and residents’ responses to these learning opportunities . Supervisors largely determine what residents are afforded to take on in clinical care , and residents choose how they engage in these learning opportunities . Supervisors’ provision of opportunities and residents’ engagement with opportunities are influenced by trust and supervisors’ and residents’ alignment on supervisory goals . Recently, however, there has been a growing recognition of the benefits of encouraging the two-way nature of the supervisor-resident relationship. Not only to value the knowledge and skills that residents bring, yet also to acknowledge the lifelong learning journey of professionals . Nevertheless, our current understanding of how the supervisory relationship supports residents’ development and ensures patient safety is still underpinned by the premise that supervisors are medical experts and experienced doctors who combine clinical and educational activities . However, in disruptive situations, such as disasters and unknown infectious disease epidemics , supervisors and residents face new and overwhelming professional challenges. Yet supervisors still need to guide residents’ professional development. The COVID-19 pandemic was such a disruption. During COVID, supervisors were no longer the medical experts in pathology and had to reinvent their practices and patient care overnight . Learning opportunities for residents were suddenly limited and supervision went remote due the restraints on physical contact . This had implications for the relationship between supervisors and residents . To understand the nature of the supervisory relationship during COVID-19, it is pertinent to consider the interactions between supervisors and residents in the workplace, the challenges they experienced and how this affected their learning processes while collaborating in this disrupted situation full of unknowns. To fully appreciate the changing dynamics of workplace learning, we used the work of Wiese et al. (2018) who created a comprehensive framework to understand learning processes and underlying learning mechanisms of workplace learning of residents and their supervisors .
The framework of Wiese et al. (2018) is embedded in sociocultural orientations to learning. This family of theories recognises that learning is inherently social. We learn from and through interaction with others and our environment. Learning is considered a transformative process. Learners transform their understandings, roles and responsibilities as they participate in the activities of a community . Specific theories that underpin the work of Wiese et al.’s are cognitive apprenticeship, Billet’s theory of co-participation, Communities of Practice and experience-based learning . All theories consider learners’ increasing access to practice, where progress is achieved by reducing risk or increasing supervision . Cognitive apprenticeship supports learning by enabling learners to acquire, develop, and use cognitive tools in authentic activity . Central to Communities of Practice is that individuals adopt and acquire the roles, skills, norms and values of the culture and community through their participation, active engagement and increasing responsibility . In experience-based learning learners engage in meaningful collaborative activities that contribute to patient care and personal and professional development as a doctor . Although Wiese et al. (2018) consider workplace learning between supervisors and residents in hospital settings, the three learning processes and underlying mechanisms they identified correspond to the interplay between supervisors and residents in general practice training. The first process, supervised participation in the workplace, involves the mechanisms of supervisors’ entrustment of residents and residents’ support-seeking behaviour. Studies of supervisor- resident encounters in general practice show that supervisors need to be confident residents are able to provide safe patient care . Whether and how supervisors’ confidence in residents is affected by a disruptive situation in clinical care is currently unknown. In the absence of disruptive situations, we know that this trust is built by gaining a sense of residents’ competence. This is influenced by residents’ help-seeking behaviour . To ensure patient safety, residents seek help from supervisors when they feel uncertain or uncomfortable performing patient care independently . An important factor in residents’ help seeking behaviours is supervisors’ credibility . What is still unknown is how residents seek help in disruptive situations, as the credibility of supervisors may be affected as they are no longer more experienced than residents. In short, it is unknown how the mechanisms of entrustment and support -seeking are affected by a disruptive situation and, therefore, what supervised participation in the workplace looks like under these conditions. The second learning process in the normal course of workplace learning is mutual observation. This is based on the mechanisms of monitoring of the resident by the supervisor and modelling behaviour by the resident. For the latter, residents observe the senior GP to integrate the senior’s behaviour in their own practice. When residents observe supervisors during a patient encounter, it supports them to make their own clinical judgements and manage uncertainty . Supervisors monitor residents during the initial weeks of their placement through direct and indirect observation to maintain clinical oversight and establish confidence in residents . At this point it is unknown how a disruptive situation which creates new challenges for supervisors, influences monitoring and modelling. The third learning process between supervisors and residents in the workplace is dialogue about practice with the underlying mechanisms of meaning-making and feedback . Meaning- making aims to stimulate critical thinking and uncover underlying presumptions through shared clinical reasoning. Supervisors and residents do this by iteratively asking and answering questions . In this context, feedback is seen as a mutual construction of residents’ performance and the means to improve it. This includes informal comments about their work, intertwined with discussions about patient care . Supervisors’ ability to adapt their supportive style and feedback to residents’ learning needs , their receptiveness to residents’ knowledge and their ability to stimulate residents learning through follow-up questions affect the quality of the supervisory relationship and the learning opportunities for both resident and supervisor. At present, it is not known how these mechanisms of meaning- making and feedback are affected by new challenges presented to supervisors and residents in a disruptive situation. Therefore, this study aims to contribute to the understanding of the supervisory relationship in a disruptive learning situation. An insight into the supervisory relationship will help to understand how residents’ learning is affected in a disruptive situation. This will also provide guidance on how to strengthen the supervisory relationship now and in future disruptive situations, and will contribute to workplace learning. The COVID-19 pandemic provided us with a good opportunity to investigate the supervisory relationship in a disruptive situation. Our guiding research question is: What characterizes the supervisory relationship in general practice residency during the COVID-19 pandemic?
Design We took a qualitative approach to our research question, guided by principles of a constructivist paradigm, which means that we view reality as context-specific, socially constructed and experience-based, and therefore subjective . We chose a case study design because case studies generally delve deeply into relationships and processes, with the aim of unravelling the complexity of a given situation . We took inspiration from Stake’s instrumental case study which he described as having “ a research question, a puzzlement, a need for general understanding, and feel that we may get insight into the question by studying a particular case” . Our case consisted of the supervisory relationship between pairs of GP supervisors and residents during COVID-19, which is a naturally occurring and bounded situation, typical of a case study . Consistent with the chosen research approach and design, we sought to understand residents’ and supervisors’ perceptions of the supervisory relationship in regard to workplace learning during the COVID-19 pandemic. We first explored their workplace learning experiences during the outbreak of the pandemic. To recognise both the subjective experiences and insights of residents and supervisors and the importance of the interaction between supervisor and resident in constructing meaning , we chose to conduct two rounds of semi-structured interviews. Due to restrictions on human contact at the time this study was conducted, we were unable to use observation as a method of data collection. In the first round of interviews, we asked residents and supervisors individually about their meaningful learning experiences related to the COVID-19 outbreak. In the second round, we asked resident and supervisor pairs about the activities undertaken and their roles and contributions to these learning experiences. By conducting joint interviews and having resident and supervisor pairs respond to each other’s experiences and insights, we expected to gain an in-depth understanding of supervisory relationships during COVID-19 . We received ethical approval to conduct this study from the Educational Research Review Board of the Leiden University Medical Centre. The first interview round was held between September and November 2020 and the second between January and February 2021. Setting and participants This exploratory study is part of a larger research project aimed at designing an evidence-informed, innovative postgraduate programme in general practice. In general, this project encouraged the development of medical leadership skills for residents. It included several cohort studies in a design-based research approach. Residents registered themselves for the innovative GP programme. They were free to choose not to participate in the different sub-studies. All residents in the third cohort of the innovation project participated in the present study. Participants were seven general practice residents and their 10 supervisors. The residents were aged between 31 and 35 years, four of them were women and three were men. The supervisors were aged between 37 and 62 years. Six of them were women and four were men. The supervisors had between three to seven years of experience in training residents. Participants came from one University Medical Centre in the Netherlands. The residents had completed the first phase of general practitioner training, which consists of 12 months of general practice placement and 6 months of emergency training. As their final phase of training, the 18-month general practice placement, coincided with the start of the COVID-19 pandemic, these residents and their supervisors could provide us with an insight into the supervisory relationship during the outbreak of COVID-19. Research team The research team consisted of an educationalist with no prior background in medical education and PhD student (IM), an educationalist and specialist in medical education (IAS) an educationalist specialized in STEM (MvdB), a health scientist (VN) and a professor in general practice focused on medical education and initiator of the innovative GP program (AK). None of the researchers were directly involved in the training of residents or supervisors. At every stage of the study - design, execution, analysis, writing - everyone’s expertise was brought to bear on the findings. Data collection An overview of the data collection is provided in Fig. . Prior to each interview round the main researcher held a trial session with volunteer(s). These trial sessions were observed by the second author and discussed afterwards to improve the main researcher’s interview skills and the interview guide. In the first round we gathered information about what residents and supervisors had learned concerning the COVID-19 outbreak (see Appendix 1 for the individual interview guide). We asked participants for their three most important learning experiences. We defined learning experiences as: the knowledge and skills that someone has gained, and the change in someone’s attitude, from doing something for a period of time and the process of gaining them . For this first round, we chose to interview one on one to create a safe environment in which participants would feel free to elaborate on their experiences. In the second round, we used the results of the first round to explore how participants had learned. We were specifically interested in whether and how the supervisory relationship had played a role (see Appendix 2 for the joint interview guide). Hence, we chose to interview resident-supervisor pairs and encouraged them to react to each other’s reflections . All interviews were held online, were recorded, and transcribed verbatim. Data analysis First round Since learning experiences concerning the COVID outbreak were novel and not previously described we used an inductive approach to data analysis. Using thematic analysis ,we constructed basic themes from the data, which were grouped into organising themes, and finally aggregated into global themes . We used an Atlas.ti software package to support data management and coding. The analysis was conducted in the following way. IM, MG, and VN familiarised themselves with the data before starting independent coding. Basic themes were derived from the data using In Vivo codes to accurately capture participants views. IM coded all 17 transcripts and MG and VN independently coded 13 transcripts in total. Dilemmas concerning coding were discussed with MG and VN until agreement was reached. After coding basic themes, IM, VN and IAS clustered codes representing similar issues into groups of organizing themes. Through additional rounds of discussion IM, IAS, VN, and AK constructed global themes capturing common learning experiences pertaining to COVID-19. We selected global themes that reflected learning experiences relating to the interplay between supervisors and residents for the second interview protocol. Second round The transcripts of the 10 joint interviews were analysed through template analysis . Template analysis can be seen as a form of thematic analysis. It is a method for identifying, analysing and reporting patterns or themes within data. In template analysis, it is possible to start with some a priori themes identified in advance as likely to be useful and relevant to the analysis . To better our understanding of the supervisory relationship during COVID-19, we found it particularly important to look for the three learning processes between supervisors and residents – supervised participation, mutual observation, dialogue about practice - and their six underlying mechanisms – entrustment & support seeking, monitoring & modelling, meaning-making & feedback - described by Wiese et al. . We used these as a priori themes. This allowed us to combine the use of deductive, theory based, and inductive, data driven, coding . Initially,the apriori codes were used as a template for data analysis. However, we did not restrict ourselves to these codes and kept an open mind to identify new patterns and codes to describe best the interplay between resident-supervisor pairs under disruptive circumstances. IM coded all 10 joint interviews. Seven interviews also coded by either VN or MH. After 2 transcripts were independently double -coded, codes were compared and dilemmas were discussed until consensus was reached. The initial template was adjusted iteratively, and coded transcripts were adapted. The final themes were created through additional rounds of discussion between IM, IAS, VN, and AK.
We took a qualitative approach to our research question, guided by principles of a constructivist paradigm, which means that we view reality as context-specific, socially constructed and experience-based, and therefore subjective . We chose a case study design because case studies generally delve deeply into relationships and processes, with the aim of unravelling the complexity of a given situation . We took inspiration from Stake’s instrumental case study which he described as having “ a research question, a puzzlement, a need for general understanding, and feel that we may get insight into the question by studying a particular case” . Our case consisted of the supervisory relationship between pairs of GP supervisors and residents during COVID-19, which is a naturally occurring and bounded situation, typical of a case study . Consistent with the chosen research approach and design, we sought to understand residents’ and supervisors’ perceptions of the supervisory relationship in regard to workplace learning during the COVID-19 pandemic. We first explored their workplace learning experiences during the outbreak of the pandemic. To recognise both the subjective experiences and insights of residents and supervisors and the importance of the interaction between supervisor and resident in constructing meaning , we chose to conduct two rounds of semi-structured interviews. Due to restrictions on human contact at the time this study was conducted, we were unable to use observation as a method of data collection. In the first round of interviews, we asked residents and supervisors individually about their meaningful learning experiences related to the COVID-19 outbreak. In the second round, we asked resident and supervisor pairs about the activities undertaken and their roles and contributions to these learning experiences. By conducting joint interviews and having resident and supervisor pairs respond to each other’s experiences and insights, we expected to gain an in-depth understanding of supervisory relationships during COVID-19 . We received ethical approval to conduct this study from the Educational Research Review Board of the Leiden University Medical Centre. The first interview round was held between September and November 2020 and the second between January and February 2021.
This exploratory study is part of a larger research project aimed at designing an evidence-informed, innovative postgraduate programme in general practice. In general, this project encouraged the development of medical leadership skills for residents. It included several cohort studies in a design-based research approach. Residents registered themselves for the innovative GP programme. They were free to choose not to participate in the different sub-studies. All residents in the third cohort of the innovation project participated in the present study. Participants were seven general practice residents and their 10 supervisors. The residents were aged between 31 and 35 years, four of them were women and three were men. The supervisors were aged between 37 and 62 years. Six of them were women and four were men. The supervisors had between three to seven years of experience in training residents. Participants came from one University Medical Centre in the Netherlands. The residents had completed the first phase of general practitioner training, which consists of 12 months of general practice placement and 6 months of emergency training. As their final phase of training, the 18-month general practice placement, coincided with the start of the COVID-19 pandemic, these residents and their supervisors could provide us with an insight into the supervisory relationship during the outbreak of COVID-19.
The research team consisted of an educationalist with no prior background in medical education and PhD student (IM), an educationalist and specialist in medical education (IAS) an educationalist specialized in STEM (MvdB), a health scientist (VN) and a professor in general practice focused on medical education and initiator of the innovative GP program (AK). None of the researchers were directly involved in the training of residents or supervisors. At every stage of the study - design, execution, analysis, writing - everyone’s expertise was brought to bear on the findings.
An overview of the data collection is provided in Fig. . Prior to each interview round the main researcher held a trial session with volunteer(s). These trial sessions were observed by the second author and discussed afterwards to improve the main researcher’s interview skills and the interview guide. In the first round we gathered information about what residents and supervisors had learned concerning the COVID-19 outbreak (see Appendix 1 for the individual interview guide). We asked participants for their three most important learning experiences. We defined learning experiences as: the knowledge and skills that someone has gained, and the change in someone’s attitude, from doing something for a period of time and the process of gaining them . For this first round, we chose to interview one on one to create a safe environment in which participants would feel free to elaborate on their experiences. In the second round, we used the results of the first round to explore how participants had learned. We were specifically interested in whether and how the supervisory relationship had played a role (see Appendix 2 for the joint interview guide). Hence, we chose to interview resident-supervisor pairs and encouraged them to react to each other’s reflections . All interviews were held online, were recorded, and transcribed verbatim.
First round Since learning experiences concerning the COVID outbreak were novel and not previously described we used an inductive approach to data analysis. Using thematic analysis ,we constructed basic themes from the data, which were grouped into organising themes, and finally aggregated into global themes . We used an Atlas.ti software package to support data management and coding. The analysis was conducted in the following way. IM, MG, and VN familiarised themselves with the data before starting independent coding. Basic themes were derived from the data using In Vivo codes to accurately capture participants views. IM coded all 17 transcripts and MG and VN independently coded 13 transcripts in total. Dilemmas concerning coding were discussed with MG and VN until agreement was reached. After coding basic themes, IM, VN and IAS clustered codes representing similar issues into groups of organizing themes. Through additional rounds of discussion IM, IAS, VN, and AK constructed global themes capturing common learning experiences pertaining to COVID-19. We selected global themes that reflected learning experiences relating to the interplay between supervisors and residents for the second interview protocol. Second round The transcripts of the 10 joint interviews were analysed through template analysis . Template analysis can be seen as a form of thematic analysis. It is a method for identifying, analysing and reporting patterns or themes within data. In template analysis, it is possible to start with some a priori themes identified in advance as likely to be useful and relevant to the analysis . To better our understanding of the supervisory relationship during COVID-19, we found it particularly important to look for the three learning processes between supervisors and residents – supervised participation, mutual observation, dialogue about practice - and their six underlying mechanisms – entrustment & support seeking, monitoring & modelling, meaning-making & feedback - described by Wiese et al. . We used these as a priori themes. This allowed us to combine the use of deductive, theory based, and inductive, data driven, coding . Initially,the apriori codes were used as a template for data analysis. However, we did not restrict ourselves to these codes and kept an open mind to identify new patterns and codes to describe best the interplay between resident-supervisor pairs under disruptive circumstances. IM coded all 10 joint interviews. Seven interviews also coded by either VN or MH. After 2 transcripts were independently double -coded, codes were compared and dilemmas were discussed until consensus was reached. The initial template was adjusted iteratively, and coded transcripts were adapted. The final themes were created through additional rounds of discussion between IM, IAS, VN, and AK.
Since learning experiences concerning the COVID outbreak were novel and not previously described we used an inductive approach to data analysis. Using thematic analysis ,we constructed basic themes from the data, which were grouped into organising themes, and finally aggregated into global themes . We used an Atlas.ti software package to support data management and coding. The analysis was conducted in the following way. IM, MG, and VN familiarised themselves with the data before starting independent coding. Basic themes were derived from the data using In Vivo codes to accurately capture participants views. IM coded all 17 transcripts and MG and VN independently coded 13 transcripts in total. Dilemmas concerning coding were discussed with MG and VN until agreement was reached. After coding basic themes, IM, VN and IAS clustered codes representing similar issues into groups of organizing themes. Through additional rounds of discussion IM, IAS, VN, and AK constructed global themes capturing common learning experiences pertaining to COVID-19. We selected global themes that reflected learning experiences relating to the interplay between supervisors and residents for the second interview protocol.
The transcripts of the 10 joint interviews were analysed through template analysis . Template analysis can be seen as a form of thematic analysis. It is a method for identifying, analysing and reporting patterns or themes within data. In template analysis, it is possible to start with some a priori themes identified in advance as likely to be useful and relevant to the analysis . To better our understanding of the supervisory relationship during COVID-19, we found it particularly important to look for the three learning processes between supervisors and residents – supervised participation, mutual observation, dialogue about practice - and their six underlying mechanisms – entrustment & support seeking, monitoring & modelling, meaning-making & feedback - described by Wiese et al. . We used these as a priori themes. This allowed us to combine the use of deductive, theory based, and inductive, data driven, coding . Initially,the apriori codes were used as a template for data analysis. However, we did not restrict ourselves to these codes and kept an open mind to identify new patterns and codes to describe best the interplay between resident-supervisor pairs under disruptive circumstances. IM coded all 10 joint interviews. Seven interviews also coded by either VN or MH. After 2 transcripts were independently double -coded, codes were compared and dilemmas were discussed until consensus was reached. The initial template was adjusted iteratively, and coded transcripts were adapted. The final themes were created through additional rounds of discussion between IM, IAS, VN, and AK.
Before we explain what the supervisory relationship during the disruptive period of COVID-19 looked like, we first consider the learning experiences deemed significant for this period. The insights gained from these learning experiences into the nature of the clinical workplace during the pandemic are essential for understanding the interplay between residents and supervisors during this period. From the data, we constructed four global themes of learning experiences related to the supervisory relationship : (1) disruptive uncertainty; (2) patient care; (3) learning opportunities of the residents; (4) supervision. In relation to the fourth theme, supervisors reported they learned much about providing supervision and residents about receiving supervision. We included these results in the second part of the results section, where the emphasis is on the supervisory relationship. Learning experiences Disruptive uncertainty The outbreak of COVID-19 immediately and abruptly confronted supervising general practitioners and residents with a high extent of uncertainty in the workplace: “This was something no one had ever experienced, not even my 7 colleagues who are in their 60s” (Supervisor 1). Supervisors and residents perceived high levels of unpredictability in their work. This manifested in two ways. First, the evident lack of medical knowledge about the clinical symptoms, the course of the disease and effective treatment. “You learn about the medical content from your patients because it is a new disease, so you don’t know what the recovery process will be like. In the beginning, you are not very clear about how it presents itself; we didn’t know the first symptom of infection could be the loss of smell and taste” (Supervisor 7). Second, the contamination rate of COVID-19 was unknown and unpredictable. Consequently, some of the supervisors and residents became anxious for their own or their family’s health and either wore extensive protective equipment or worked from home. In addition, the risk of contamination affected the way patient care was organized, because common routines to see patients became potentially risky. However, national policies and local guidelines were constantly updated, and protective equipment was insufficient, especially at the start of the outbreak. As a result, each GP practice had to evaluate the situation and determine its policies and procedures. One protective measure taken in all GP practices was the restriction of physical examinations. One resident voiced the confusion and uncertainty they experienced as follows: “To examine patients, when do you use face masks, gloves? And when do you not use them? Who do you invite to the consultation hours? Who is not invited? These are really tricky things. I think that goes for everyone” (Resident 1). .The profound uncertainty disrupted workplace routines. This created novel learning opportunities as supervisors and residents faced new challenges. “In the beginning you still had to come up with creative solutions” (Resident 5). Patient care Much of the delivery of patient care had to be reinvented as physical consultations were largely replaced by telephone triages and consultations. Telephone triage became an essential part of the daily work of residents and supervisors. They had to assess whether the physical examination of patients could be carried out safely. If a patient was infected with COVID, they decided whether the patient should be referred to the hospital or whether they could check in with the patient regularly via telephone consultations. Residents commented that they initially found it “difficult” and “burdensome” to decide how ill a person was over the phone, with little medical knowledge about the symptoms of COVID-19. This discomfort was expressed by a resident: “suddenly every tool you had was knocked out of your hands” (Resident 2). However, residents and supervisors learned to make these assessments over time: “I learned a lot from assessing patients’ complaints and symptoms over the telephone and acting on the findings, and it pays off when you do telephone consultations” (Resident 5). Over time, as residents and supervisors gained insight into the symptoms of COVID-19, they learned to accept the limited role they had in the care for COVID patients: “If they [patients] are not sick enough then there is not much you can do for them, and it is really a matter of waiting to see how things develop, and that is a process of learning to let go” (Resident 7). As supervisors and residents became accustomed to providing patient care over the phone, they began to recognize some advantages for patient care. Some supervisors who were initially sceptical about telephone consultations learned that “while we were thinking ‘we can’t do anything anymore because everything has to be done by telephone’, we noticed that 95% of things do work out using the telephone. That the world does not end when you are severely limited when it comes to seeing patients physically” (Supervisor 9). In addition, residents and supervisors described the advantages of digital tools for patient care: “We mainly use e-consulting, we make use of pictures [of symptoms sent by patients], and we do much more by phone. Covid brought these developments to the practice, and we still notice the benefits. We can easily take care of little things by using photos or by providing telephone advice” (Supervisor 8). Learning opportunities for residents The uncertainty and the challenges in patient care blocked routine learning opportunities, such as chronic elderly care. Yet, it also allowed for new learning opportunities. Unable to revert to how they usually started the training of a new resident, supervisors had to find new ways of training residents. Many focussed on trying to create sufficiently diverse learning opportunities: “My resident worked in the COVID facility [in a hospital] several times from the beginning. I felt she should learn the things that can be specifically learned right now, so I tried to seize really the moment” (Supervisor 1). Most of the residents learned to accept that the final part of their training period was different than they had envisioned it to be. They learned to utilize effectively the new opportunities: “I’m sure there are things I have not been able to learn as a result. However, it really has been a unique opportunity to experience this pandemic as a resident instead of a qualified GP or locum” (Resident 7). In sum, we described three themes of learning experiences that were significant during the pandemic as they specify the unique challenges faced by supervisors and residents. We used these themes to determine characteristics of the supervisory relationship during the COVID-19 outbreak. The supervisory relationship Collaboration between supervisors and residents proved to be a key element in their relationship during the outbreak of the pandemic. We identified three types of collaboration, each focussed on different workplace challenges and achieving different goals. We label these three types of collaboration as: getting the job done, residents’ learning, and collective learning . For each type of collaboration, we describe the characteristics of the supervisory relationship. Collaboration focussed on getting the job done The aim of this type of collaboration between supervisors and residents was to continue patient care under disruptive circumstances, or ‘ getting the job done’. To this end, supervisors transferred part of patient care to their residents as if they already were qualified colleagues. Residents characterized their learning process as ‘learning by doing’. Supervisors seemed to entrust them more readily with the care than they would have under normal circumstances. “When a resident joins our practice, at first you need to find out who this person is and how things go. Yet, because I was working elsewhere [in the emergency organization for local Covid care], the resident was in charge right away and I think that actually got the training off to a flying start” (Supervisor joint interview (ji) 9). Supervisors did acknowledge that residents were not yet proficient colleagues as they continued to monitor residents’ ways of thinking, their actions, and decisions. For instance, supervisors continued to monitor their residents by keeping track of their calendars. “You see in your calendar what is happening you see the patient’s name in italics and when I click on it, I see what happened. And 9 times out of 10 it is just a second, and for some patients you ask the resident why they handled it in a certain way and how they think we should continue care. So, the difficult cases do come up along the way” (Supervisor ji 3). In addition to monitoring, the supervisors also provided a safety net for their residents. As experienced colleagues, they continued to be available for support or to discuss immediate patient care issues. “The fact that you can always call to consult [with the supervisor about telephone consultations], and the possibility to discuss afterwards what went well or what didn’t, gives you peace of mind” (Resident ji 3). Residents’ support-seeking behaviour when faced with challenges they did not feel comfortable dealing with themselves was also an important element in this type of collaboration. The combination of supervisors’ entrustment, monitoring and provision of a safety net, and residents’ support-seeking behaviour allowed residents to take on the role of new yet qualified colleague right from the beginning. Collaboration focussed on residents’ learning In GP practices where physical consultations were restricted for months, learning opportunities for residents were not self-evident. In these circumstances residents and supervisors were aware of the challenge and the need to find sufficient learning opportunities for residents. We found three ways in which supervisors and residents collaborated to promote residents’ development. In general, supervisors initiated learning opportunities, yet not always, and residents chose how to utilize these opportunities. First, supervisors and residents tried to use existing possibilities optimally . For example, supervisors invited their residents to participate structurally in discussions about management. “I tried to involve my resident in many issues, such as how we make decisions about the practice” (Supervisor ji 8). Residents used the available opportunities to their advantage. They used their position as learners to actively participate in team deliberations without the burden of accountability: “Now something really has to change, and then you see what happens when […] a walk-in consultation hour suddenly stops and 11,000 patients have to be informed. […] It is great to be able to observe this, especially since you are not responsible for it, you can have opinions and think about it without it having many consequences” (Resident ji 3). Second, supervisors and residents created additional learning opportunities . In some cases, supervisors transferred their consultation hours to their residents in order to increase their exposure to patients : “You gave me some of your consultation hours. Then you did the telephone consultations with the assistants so that I could see more patients” [resident addresses supervisor during the interview] (Resident ji 8). In other cases, residents initiated a conversation with their supervisor about the need for more patient consultations. “I did say that I needed to see more patients. I want to see more patients so I can develop professionally. I was worried that I might miss opportunities to learn. And my supervisors agreed that they had to arrange more patient contact for me” (Resident ji 5). Third, supervisors created new learning opportunities specifically related to COVID-19 . Residents made use of these opportunities. “(My supervisor) involved me in what was going on with the whole organisation (of COVID care in the area). I was able to work at the COVID facility soon after it became a thing, which was exciting, and also very cool in the beginning. We were in those COVID suits with large masks straight away. […] those are advantages and I think we really used them” (Resident ji 9). In addition, amidst the hectic workplace, supervisors and residents had to make time for reflection and discussion about residents’ learning. Whereas supervisors and residents used to have dedicated reflection sessions to discuss issues ranging from patient care to residents’ professional development, these ‘reflective learning conversations’ were no longer self-evident. Residents and supervisors placed different levels of importance on these learning conversations. The amount of time allocated to these conversations varied from practice to practice. Some supervisor-resident pairs used these conversations as an anchor point to structure the residents’ training in an otherwise chaotic time. In other practices, the ‘reflective learning conversations’ were abandoned due to the hectic start of the training and they were not implemented later. Finally, this type of collaboration did not include mutual observation. Nor did we find signs of modelling. Collaboration focussed on collective learning This type of collaboration occurred between all team members of the GP practice, not just between supervisors and residents. The aim was to gain a better understanding of the disease and its impact on patient care and the organisation of patient care. Residents, supervisors, non-supervising GPs and other team members shared information and developed strategies to adjust effectively the organisation of patient care. All team members had equal positions in this collective enterprise. The following quote from a supervisor illustrates the collective team spirit at that time: “We were absolutely no better informed than, well, no more than the doctor’s assistants. I mean everyone started from zero and that also made it enjoyable. It is as if you have to solve a problem together in an escape room. Yes, everyone contributes something […] it is a kind of new problem that everyone is totally focussed on” (Supervisor ji 8). As supervisors, residents, and non-supervising GPs experienced similar challenges in patient care, including telephone triage and consultation, they shared experiences and exchanged information to improve their diagnostic skills over the telephone. The following quote from a resident illustrates that residents discussed their doubts concerning clinical activities with several GP colleagues, rather than restricting themselves to their supervisor: “Because everyone was learning (triage over the phone), […] so we talked a lot in the coffee room […] you just discussed a lot with each other like: this was difficult for me. We are a huge team […] and we discussed so much in the early corona meetings, things like (following up on) ‘how did that go?’ and sharing tips and tricks with each other (Resident ji 1). As residents engaged in collective meaning-making activities, they quickly became a part of the team. “I think you […] become part of a team much faster if a new situation is created for everyone, because then everyone has to find a new position, like OK, how are we going to do this. So you definitely feel more at home and you get to know everyone quicker” (Resident ji 8). Our data do not indicate whether the information shared, and the knowledge created collaboratively with the team circled back to the supervisory relationship, for example, in reflective learning conversations. Nor did we find clues about feedback on residents’ performance. In conclusion, collaboration between residents and supervisors early in the pandemic was manifested in three types, each containing distinctive characteristics of the supervisory relationship. In the first type getting the job done , residents took on the role of starting yet qualified colleague, while supervisors acted as senior colleagues providing a safety net. In the second type residents’ learning , the resident’s role was that of the learner who needed enough learning opportunities to develop competencies, while the supervisor acted as a facilitator of learning. The third type of collaboration, collective learning , went beyond the supervisor-resident dyad. Residents and supervisors were both team members who learned about the pandemic in the same way as the other team members. All supervisor-resident pairs used all three types of collaboration depending on the challenges they faced.
Disruptive uncertainty The outbreak of COVID-19 immediately and abruptly confronted supervising general practitioners and residents with a high extent of uncertainty in the workplace: “This was something no one had ever experienced, not even my 7 colleagues who are in their 60s” (Supervisor 1). Supervisors and residents perceived high levels of unpredictability in their work. This manifested in two ways. First, the evident lack of medical knowledge about the clinical symptoms, the course of the disease and effective treatment. “You learn about the medical content from your patients because it is a new disease, so you don’t know what the recovery process will be like. In the beginning, you are not very clear about how it presents itself; we didn’t know the first symptom of infection could be the loss of smell and taste” (Supervisor 7). Second, the contamination rate of COVID-19 was unknown and unpredictable. Consequently, some of the supervisors and residents became anxious for their own or their family’s health and either wore extensive protective equipment or worked from home. In addition, the risk of contamination affected the way patient care was organized, because common routines to see patients became potentially risky. However, national policies and local guidelines were constantly updated, and protective equipment was insufficient, especially at the start of the outbreak. As a result, each GP practice had to evaluate the situation and determine its policies and procedures. One protective measure taken in all GP practices was the restriction of physical examinations. One resident voiced the confusion and uncertainty they experienced as follows: “To examine patients, when do you use face masks, gloves? And when do you not use them? Who do you invite to the consultation hours? Who is not invited? These are really tricky things. I think that goes for everyone” (Resident 1). .The profound uncertainty disrupted workplace routines. This created novel learning opportunities as supervisors and residents faced new challenges. “In the beginning you still had to come up with creative solutions” (Resident 5). Patient care Much of the delivery of patient care had to be reinvented as physical consultations were largely replaced by telephone triages and consultations. Telephone triage became an essential part of the daily work of residents and supervisors. They had to assess whether the physical examination of patients could be carried out safely. If a patient was infected with COVID, they decided whether the patient should be referred to the hospital or whether they could check in with the patient regularly via telephone consultations. Residents commented that they initially found it “difficult” and “burdensome” to decide how ill a person was over the phone, with little medical knowledge about the symptoms of COVID-19. This discomfort was expressed by a resident: “suddenly every tool you had was knocked out of your hands” (Resident 2). However, residents and supervisors learned to make these assessments over time: “I learned a lot from assessing patients’ complaints and symptoms over the telephone and acting on the findings, and it pays off when you do telephone consultations” (Resident 5). Over time, as residents and supervisors gained insight into the symptoms of COVID-19, they learned to accept the limited role they had in the care for COVID patients: “If they [patients] are not sick enough then there is not much you can do for them, and it is really a matter of waiting to see how things develop, and that is a process of learning to let go” (Resident 7). As supervisors and residents became accustomed to providing patient care over the phone, they began to recognize some advantages for patient care. Some supervisors who were initially sceptical about telephone consultations learned that “while we were thinking ‘we can’t do anything anymore because everything has to be done by telephone’, we noticed that 95% of things do work out using the telephone. That the world does not end when you are severely limited when it comes to seeing patients physically” (Supervisor 9). In addition, residents and supervisors described the advantages of digital tools for patient care: “We mainly use e-consulting, we make use of pictures [of symptoms sent by patients], and we do much more by phone. Covid brought these developments to the practice, and we still notice the benefits. We can easily take care of little things by using photos or by providing telephone advice” (Supervisor 8). Learning opportunities for residents The uncertainty and the challenges in patient care blocked routine learning opportunities, such as chronic elderly care. Yet, it also allowed for new learning opportunities. Unable to revert to how they usually started the training of a new resident, supervisors had to find new ways of training residents. Many focussed on trying to create sufficiently diverse learning opportunities: “My resident worked in the COVID facility [in a hospital] several times from the beginning. I felt she should learn the things that can be specifically learned right now, so I tried to seize really the moment” (Supervisor 1). Most of the residents learned to accept that the final part of their training period was different than they had envisioned it to be. They learned to utilize effectively the new opportunities: “I’m sure there are things I have not been able to learn as a result. However, it really has been a unique opportunity to experience this pandemic as a resident instead of a qualified GP or locum” (Resident 7). In sum, we described three themes of learning experiences that were significant during the pandemic as they specify the unique challenges faced by supervisors and residents. We used these themes to determine characteristics of the supervisory relationship during the COVID-19 outbreak.
The outbreak of COVID-19 immediately and abruptly confronted supervising general practitioners and residents with a high extent of uncertainty in the workplace: “This was something no one had ever experienced, not even my 7 colleagues who are in their 60s” (Supervisor 1). Supervisors and residents perceived high levels of unpredictability in their work. This manifested in two ways. First, the evident lack of medical knowledge about the clinical symptoms, the course of the disease and effective treatment. “You learn about the medical content from your patients because it is a new disease, so you don’t know what the recovery process will be like. In the beginning, you are not very clear about how it presents itself; we didn’t know the first symptom of infection could be the loss of smell and taste” (Supervisor 7). Second, the contamination rate of COVID-19 was unknown and unpredictable. Consequently, some of the supervisors and residents became anxious for their own or their family’s health and either wore extensive protective equipment or worked from home. In addition, the risk of contamination affected the way patient care was organized, because common routines to see patients became potentially risky. However, national policies and local guidelines were constantly updated, and protective equipment was insufficient, especially at the start of the outbreak. As a result, each GP practice had to evaluate the situation and determine its policies and procedures. One protective measure taken in all GP practices was the restriction of physical examinations. One resident voiced the confusion and uncertainty they experienced as follows: “To examine patients, when do you use face masks, gloves? And when do you not use them? Who do you invite to the consultation hours? Who is not invited? These are really tricky things. I think that goes for everyone” (Resident 1). .The profound uncertainty disrupted workplace routines. This created novel learning opportunities as supervisors and residents faced new challenges. “In the beginning you still had to come up with creative solutions” (Resident 5).
Much of the delivery of patient care had to be reinvented as physical consultations were largely replaced by telephone triages and consultations. Telephone triage became an essential part of the daily work of residents and supervisors. They had to assess whether the physical examination of patients could be carried out safely. If a patient was infected with COVID, they decided whether the patient should be referred to the hospital or whether they could check in with the patient regularly via telephone consultations. Residents commented that they initially found it “difficult” and “burdensome” to decide how ill a person was over the phone, with little medical knowledge about the symptoms of COVID-19. This discomfort was expressed by a resident: “suddenly every tool you had was knocked out of your hands” (Resident 2). However, residents and supervisors learned to make these assessments over time: “I learned a lot from assessing patients’ complaints and symptoms over the telephone and acting on the findings, and it pays off when you do telephone consultations” (Resident 5). Over time, as residents and supervisors gained insight into the symptoms of COVID-19, they learned to accept the limited role they had in the care for COVID patients: “If they [patients] are not sick enough then there is not much you can do for them, and it is really a matter of waiting to see how things develop, and that is a process of learning to let go” (Resident 7). As supervisors and residents became accustomed to providing patient care over the phone, they began to recognize some advantages for patient care. Some supervisors who were initially sceptical about telephone consultations learned that “while we were thinking ‘we can’t do anything anymore because everything has to be done by telephone’, we noticed that 95% of things do work out using the telephone. That the world does not end when you are severely limited when it comes to seeing patients physically” (Supervisor 9). In addition, residents and supervisors described the advantages of digital tools for patient care: “We mainly use e-consulting, we make use of pictures [of symptoms sent by patients], and we do much more by phone. Covid brought these developments to the practice, and we still notice the benefits. We can easily take care of little things by using photos or by providing telephone advice” (Supervisor 8).
The uncertainty and the challenges in patient care blocked routine learning opportunities, such as chronic elderly care. Yet, it also allowed for new learning opportunities. Unable to revert to how they usually started the training of a new resident, supervisors had to find new ways of training residents. Many focussed on trying to create sufficiently diverse learning opportunities: “My resident worked in the COVID facility [in a hospital] several times from the beginning. I felt she should learn the things that can be specifically learned right now, so I tried to seize really the moment” (Supervisor 1). Most of the residents learned to accept that the final part of their training period was different than they had envisioned it to be. They learned to utilize effectively the new opportunities: “I’m sure there are things I have not been able to learn as a result. However, it really has been a unique opportunity to experience this pandemic as a resident instead of a qualified GP or locum” (Resident 7). In sum, we described three themes of learning experiences that were significant during the pandemic as they specify the unique challenges faced by supervisors and residents. We used these themes to determine characteristics of the supervisory relationship during the COVID-19 outbreak.
Collaboration between supervisors and residents proved to be a key element in their relationship during the outbreak of the pandemic. We identified three types of collaboration, each focussed on different workplace challenges and achieving different goals. We label these three types of collaboration as: getting the job done, residents’ learning, and collective learning . For each type of collaboration, we describe the characteristics of the supervisory relationship. Collaboration focussed on getting the job done The aim of this type of collaboration between supervisors and residents was to continue patient care under disruptive circumstances, or ‘ getting the job done’. To this end, supervisors transferred part of patient care to their residents as if they already were qualified colleagues. Residents characterized their learning process as ‘learning by doing’. Supervisors seemed to entrust them more readily with the care than they would have under normal circumstances. “When a resident joins our practice, at first you need to find out who this person is and how things go. Yet, because I was working elsewhere [in the emergency organization for local Covid care], the resident was in charge right away and I think that actually got the training off to a flying start” (Supervisor joint interview (ji) 9). Supervisors did acknowledge that residents were not yet proficient colleagues as they continued to monitor residents’ ways of thinking, their actions, and decisions. For instance, supervisors continued to monitor their residents by keeping track of their calendars. “You see in your calendar what is happening you see the patient’s name in italics and when I click on it, I see what happened. And 9 times out of 10 it is just a second, and for some patients you ask the resident why they handled it in a certain way and how they think we should continue care. So, the difficult cases do come up along the way” (Supervisor ji 3). In addition to monitoring, the supervisors also provided a safety net for their residents. As experienced colleagues, they continued to be available for support or to discuss immediate patient care issues. “The fact that you can always call to consult [with the supervisor about telephone consultations], and the possibility to discuss afterwards what went well or what didn’t, gives you peace of mind” (Resident ji 3). Residents’ support-seeking behaviour when faced with challenges they did not feel comfortable dealing with themselves was also an important element in this type of collaboration. The combination of supervisors’ entrustment, monitoring and provision of a safety net, and residents’ support-seeking behaviour allowed residents to take on the role of new yet qualified colleague right from the beginning. Collaboration focussed on residents’ learning In GP practices where physical consultations were restricted for months, learning opportunities for residents were not self-evident. In these circumstances residents and supervisors were aware of the challenge and the need to find sufficient learning opportunities for residents. We found three ways in which supervisors and residents collaborated to promote residents’ development. In general, supervisors initiated learning opportunities, yet not always, and residents chose how to utilize these opportunities. First, supervisors and residents tried to use existing possibilities optimally . For example, supervisors invited their residents to participate structurally in discussions about management. “I tried to involve my resident in many issues, such as how we make decisions about the practice” (Supervisor ji 8). Residents used the available opportunities to their advantage. They used their position as learners to actively participate in team deliberations without the burden of accountability: “Now something really has to change, and then you see what happens when […] a walk-in consultation hour suddenly stops and 11,000 patients have to be informed. […] It is great to be able to observe this, especially since you are not responsible for it, you can have opinions and think about it without it having many consequences” (Resident ji 3). Second, supervisors and residents created additional learning opportunities . In some cases, supervisors transferred their consultation hours to their residents in order to increase their exposure to patients : “You gave me some of your consultation hours. Then you did the telephone consultations with the assistants so that I could see more patients” [resident addresses supervisor during the interview] (Resident ji 8). In other cases, residents initiated a conversation with their supervisor about the need for more patient consultations. “I did say that I needed to see more patients. I want to see more patients so I can develop professionally. I was worried that I might miss opportunities to learn. And my supervisors agreed that they had to arrange more patient contact for me” (Resident ji 5). Third, supervisors created new learning opportunities specifically related to COVID-19 . Residents made use of these opportunities. “(My supervisor) involved me in what was going on with the whole organisation (of COVID care in the area). I was able to work at the COVID facility soon after it became a thing, which was exciting, and also very cool in the beginning. We were in those COVID suits with large masks straight away. […] those are advantages and I think we really used them” (Resident ji 9). In addition, amidst the hectic workplace, supervisors and residents had to make time for reflection and discussion about residents’ learning. Whereas supervisors and residents used to have dedicated reflection sessions to discuss issues ranging from patient care to residents’ professional development, these ‘reflective learning conversations’ were no longer self-evident. Residents and supervisors placed different levels of importance on these learning conversations. The amount of time allocated to these conversations varied from practice to practice. Some supervisor-resident pairs used these conversations as an anchor point to structure the residents’ training in an otherwise chaotic time. In other practices, the ‘reflective learning conversations’ were abandoned due to the hectic start of the training and they were not implemented later. Finally, this type of collaboration did not include mutual observation. Nor did we find signs of modelling. Collaboration focussed on collective learning This type of collaboration occurred between all team members of the GP practice, not just between supervisors and residents. The aim was to gain a better understanding of the disease and its impact on patient care and the organisation of patient care. Residents, supervisors, non-supervising GPs and other team members shared information and developed strategies to adjust effectively the organisation of patient care. All team members had equal positions in this collective enterprise. The following quote from a supervisor illustrates the collective team spirit at that time: “We were absolutely no better informed than, well, no more than the doctor’s assistants. I mean everyone started from zero and that also made it enjoyable. It is as if you have to solve a problem together in an escape room. Yes, everyone contributes something […] it is a kind of new problem that everyone is totally focussed on” (Supervisor ji 8). As supervisors, residents, and non-supervising GPs experienced similar challenges in patient care, including telephone triage and consultation, they shared experiences and exchanged information to improve their diagnostic skills over the telephone. The following quote from a resident illustrates that residents discussed their doubts concerning clinical activities with several GP colleagues, rather than restricting themselves to their supervisor: “Because everyone was learning (triage over the phone), […] so we talked a lot in the coffee room […] you just discussed a lot with each other like: this was difficult for me. We are a huge team […] and we discussed so much in the early corona meetings, things like (following up on) ‘how did that go?’ and sharing tips and tricks with each other (Resident ji 1). As residents engaged in collective meaning-making activities, they quickly became a part of the team. “I think you […] become part of a team much faster if a new situation is created for everyone, because then everyone has to find a new position, like OK, how are we going to do this. So you definitely feel more at home and you get to know everyone quicker” (Resident ji 8). Our data do not indicate whether the information shared, and the knowledge created collaboratively with the team circled back to the supervisory relationship, for example, in reflective learning conversations. Nor did we find clues about feedback on residents’ performance. In conclusion, collaboration between residents and supervisors early in the pandemic was manifested in three types, each containing distinctive characteristics of the supervisory relationship. In the first type getting the job done , residents took on the role of starting yet qualified colleague, while supervisors acted as senior colleagues providing a safety net. In the second type residents’ learning , the resident’s role was that of the learner who needed enough learning opportunities to develop competencies, while the supervisor acted as a facilitator of learning. The third type of collaboration, collective learning , went beyond the supervisor-resident dyad. Residents and supervisors were both team members who learned about the pandemic in the same way as the other team members. All supervisor-resident pairs used all three types of collaboration depending on the challenges they faced.
The aim of this type of collaboration between supervisors and residents was to continue patient care under disruptive circumstances, or ‘ getting the job done’. To this end, supervisors transferred part of patient care to their residents as if they already were qualified colleagues. Residents characterized their learning process as ‘learning by doing’. Supervisors seemed to entrust them more readily with the care than they would have under normal circumstances. “When a resident joins our practice, at first you need to find out who this person is and how things go. Yet, because I was working elsewhere [in the emergency organization for local Covid care], the resident was in charge right away and I think that actually got the training off to a flying start” (Supervisor joint interview (ji) 9). Supervisors did acknowledge that residents were not yet proficient colleagues as they continued to monitor residents’ ways of thinking, their actions, and decisions. For instance, supervisors continued to monitor their residents by keeping track of their calendars. “You see in your calendar what is happening you see the patient’s name in italics and when I click on it, I see what happened. And 9 times out of 10 it is just a second, and for some patients you ask the resident why they handled it in a certain way and how they think we should continue care. So, the difficult cases do come up along the way” (Supervisor ji 3). In addition to monitoring, the supervisors also provided a safety net for their residents. As experienced colleagues, they continued to be available for support or to discuss immediate patient care issues. “The fact that you can always call to consult [with the supervisor about telephone consultations], and the possibility to discuss afterwards what went well or what didn’t, gives you peace of mind” (Resident ji 3). Residents’ support-seeking behaviour when faced with challenges they did not feel comfortable dealing with themselves was also an important element in this type of collaboration. The combination of supervisors’ entrustment, monitoring and provision of a safety net, and residents’ support-seeking behaviour allowed residents to take on the role of new yet qualified colleague right from the beginning.
In GP practices where physical consultations were restricted for months, learning opportunities for residents were not self-evident. In these circumstances residents and supervisors were aware of the challenge and the need to find sufficient learning opportunities for residents. We found three ways in which supervisors and residents collaborated to promote residents’ development. In general, supervisors initiated learning opportunities, yet not always, and residents chose how to utilize these opportunities. First, supervisors and residents tried to use existing possibilities optimally . For example, supervisors invited their residents to participate structurally in discussions about management. “I tried to involve my resident in many issues, such as how we make decisions about the practice” (Supervisor ji 8). Residents used the available opportunities to their advantage. They used their position as learners to actively participate in team deliberations without the burden of accountability: “Now something really has to change, and then you see what happens when […] a walk-in consultation hour suddenly stops and 11,000 patients have to be informed. […] It is great to be able to observe this, especially since you are not responsible for it, you can have opinions and think about it without it having many consequences” (Resident ji 3). Second, supervisors and residents created additional learning opportunities . In some cases, supervisors transferred their consultation hours to their residents in order to increase their exposure to patients : “You gave me some of your consultation hours. Then you did the telephone consultations with the assistants so that I could see more patients” [resident addresses supervisor during the interview] (Resident ji 8). In other cases, residents initiated a conversation with their supervisor about the need for more patient consultations. “I did say that I needed to see more patients. I want to see more patients so I can develop professionally. I was worried that I might miss opportunities to learn. And my supervisors agreed that they had to arrange more patient contact for me” (Resident ji 5). Third, supervisors created new learning opportunities specifically related to COVID-19 . Residents made use of these opportunities. “(My supervisor) involved me in what was going on with the whole organisation (of COVID care in the area). I was able to work at the COVID facility soon after it became a thing, which was exciting, and also very cool in the beginning. We were in those COVID suits with large masks straight away. […] those are advantages and I think we really used them” (Resident ji 9). In addition, amidst the hectic workplace, supervisors and residents had to make time for reflection and discussion about residents’ learning. Whereas supervisors and residents used to have dedicated reflection sessions to discuss issues ranging from patient care to residents’ professional development, these ‘reflective learning conversations’ were no longer self-evident. Residents and supervisors placed different levels of importance on these learning conversations. The amount of time allocated to these conversations varied from practice to practice. Some supervisor-resident pairs used these conversations as an anchor point to structure the residents’ training in an otherwise chaotic time. In other practices, the ‘reflective learning conversations’ were abandoned due to the hectic start of the training and they were not implemented later. Finally, this type of collaboration did not include mutual observation. Nor did we find signs of modelling.
This type of collaboration occurred between all team members of the GP practice, not just between supervisors and residents. The aim was to gain a better understanding of the disease and its impact on patient care and the organisation of patient care. Residents, supervisors, non-supervising GPs and other team members shared information and developed strategies to adjust effectively the organisation of patient care. All team members had equal positions in this collective enterprise. The following quote from a supervisor illustrates the collective team spirit at that time: “We were absolutely no better informed than, well, no more than the doctor’s assistants. I mean everyone started from zero and that also made it enjoyable. It is as if you have to solve a problem together in an escape room. Yes, everyone contributes something […] it is a kind of new problem that everyone is totally focussed on” (Supervisor ji 8). As supervisors, residents, and non-supervising GPs experienced similar challenges in patient care, including telephone triage and consultation, they shared experiences and exchanged information to improve their diagnostic skills over the telephone. The following quote from a resident illustrates that residents discussed their doubts concerning clinical activities with several GP colleagues, rather than restricting themselves to their supervisor: “Because everyone was learning (triage over the phone), […] so we talked a lot in the coffee room […] you just discussed a lot with each other like: this was difficult for me. We are a huge team […] and we discussed so much in the early corona meetings, things like (following up on) ‘how did that go?’ and sharing tips and tricks with each other (Resident ji 1). As residents engaged in collective meaning-making activities, they quickly became a part of the team. “I think you […] become part of a team much faster if a new situation is created for everyone, because then everyone has to find a new position, like OK, how are we going to do this. So you definitely feel more at home and you get to know everyone quicker” (Resident ji 8). Our data do not indicate whether the information shared, and the knowledge created collaboratively with the team circled back to the supervisory relationship, for example, in reflective learning conversations. Nor did we find clues about feedback on residents’ performance. In conclusion, collaboration between residents and supervisors early in the pandemic was manifested in three types, each containing distinctive characteristics of the supervisory relationship. In the first type getting the job done , residents took on the role of starting yet qualified colleague, while supervisors acted as senior colleagues providing a safety net. In the second type residents’ learning , the resident’s role was that of the learner who needed enough learning opportunities to develop competencies, while the supervisor acted as a facilitator of learning. The third type of collaboration, collective learning , went beyond the supervisor-resident dyad. Residents and supervisors were both team members who learned about the pandemic in the same way as the other team members. All supervisor-resident pairs used all three types of collaboration depending on the challenges they faced.
Principal findings In this study we explored the impact of a disruptive situation, i.e. the COVID-19 pandemic, on the relationship between GP residents and their supervisors. We found that the disruptive uncertainty, the extensive consequences for the patient care and the impact of learning opportunities for residents resulted in three types of collaboration. Within the supervisory relationship, supervisors and residents collaborated to get the job done and to facilitate residents’ learning. Beyond the supervisory relationship, collective learning emerged as supervisors, residents and other team members collaborated in information sharing and joint meaning-making to enhance their understanding of the situation and to improve patient care. While the first two types of collaboration, getting the job done and residents’ learning, are typical for workplace learning, the extent to which they occurred was different. Getting the job done became more prominent as residents were encouraged to quickly work independently. In contrast, creating learning opportunities for residents required more attention and inventiveness since the regular training structure fell away. The third type of collaboration, collective learning, i.e. learning with and from each other, including non-supervising GPs and the entire team, has received little attention in the literature on general practice training. Our study contributes to the understanding of the supervisory relationship in GP training, by providing insight into the characteristics of this relationship during a period of major and unprecedented change. As we discuss our findings in the light of other literature, we will begin with the impact of uncertainty on the supervisory relationship during the COVID-19 pandemic, before we continue with the types of collaboration and the underlying interaction patterns that we found. Uncertainty Clinical uncertainty is a key aspect of general practice as the entry point to the healthcare system . Residents learn to manage and tolerate this uncertainty by observing how their supervisors deal with uncertainty. Residents develop the same strategies for responding to uncertainty as their supervisors . However, the type of uncertainty we addressed in this study is pervasive and disruptive - causing problems so that something cannot proceed normally . The scope of the uncertainty we found was not limited to clinical uncertainty, yet was much broader. Supervisors, residents and other team members were faced with the unpredictability of the disease and its impact on care delivery. As a result, even supervisors had to learn to manage this type of uncertainty. Uncertainty was therefore a general and shared lack of clarity about clinical activities and the organisation of patient care, rather than a ‘state of mind’ of an individual, often the resident . Interestingly, residents and supervisors used strategies to cope with uncertainty similar to those described by Han and colleagues (2021), namely , i.e. (1) ignorance-focussed; (2) uncertainty-focussed; (3) response-focussed; and (4) relationship-focussed. They reduced their ignorance by, for example, seeking medical information. They changed their response to uncertainty by learning to accept their limited influence on patients with COVID disease. Finally, they shared doubts or discomfort they felt with their team, a ‘relationship’-focussed strategy. Our study adds collective strategies, such as shared meaning-making, to the individual strategies used by supervisors and residents to deal with disruptive uncertainty. Collaboration and interaction patterns The influence of disruptive uncertainty is reflected in the types of collaboration and underlying patterns of interaction between supervisor and resident pairs. The first type of collaboration, ‘getting the job done’, involved the underlying interaction patterns as described by Wiese et al. (2018) of supervisors’ monitoring and entrustment, and residents’ help seeking behaviour. We found that supervisors immediately entrusted residents with patient care. While entrustment in general practice residency develops rather rapidly, holistically and presumptively , the pace by which entrustment was awarded here was remarkable. The residents, in turn, accepted the invitation to patient care as though they were already qualified colleagues. This type of collaboration facilitates residents’ participation and active engagement in the professional community as they took on increasing responsibility for patient care . Sfard’s metaphor of learning as ‘participation’ is appropriate here. It conceives of learning as a process of meaning construction and identity formation through participation in community activities . The second type of collaboration, residents’ learning, involved supervisors creating sufficient learning opportunities for residents to develop the competencies required for the diverse range of patients in primary care. This resembles Sfard’s metaphor of learning as ‘acquisition’ . In this metaphor learning is seen as acquisition of knowledge, skills, and attributes that are ‘owned’ by the individual . We did not find any signs of supervisors demonstrating their expertise nor of residents integrating their supervisors’ modelled skills. This might be because the supervisors did not have more specific expertise than the residents, or because of the limitations in interpersonal contact, which prevented supervisors and residents from seeing patients together. The third type of collaboration, collective learning, involved residents, supervisors and other staff creating an understanding of the impact of the pandemic and developing new knowledge for patient care and policy. They did this through a collaborative process of meaning making and problem solving. This type of collaboration facilitated the joint creation of knowledge. This resembles Paavola and Hakkarainen’s metaphor of learning as ‘knowledge creation’ . In this metaphor, learning is seen as the collaborative development of activity in response to challenges. In short, our findings show how supervisors and residents collaborate in workplace learning during a disruptive situation. Although we found some similarities with the learning processes and interaction patterns described by Wiese et al. , other features of the supervisory relationship were highlighted during the pandemic. This adds to our understanding of the relationship between supervisor and resident in workplace learning. Collective learning The most notable finding for us is that of collective learning, as it is a clear complement to workplace learning between supervisors and residents and the supervisory relationship in times of disruption. However, it is not a surprising finding. Collective learning appears to be an intuitive response to new or complex challenges , and during the first outbreak of COVID-19, no one had more expertise than anyone else. Also, the shared challenges supervisors, residents and other staff faced during the pandemic created a sense of belonging, eliminating the sense of hierarchy that could hinder collective learning . Collective learning is internally generated by a team who are willing to learn and who are aware of the importance of finding solutions to practice problems . Because it is often implicit, collective learning is frequently unnoticed by professionals . The disruptive situation of COVID-19 made collective learning explicit. Strengths and limitations This study has several methodological strengths and some limitations. First, to our knowledge, this is one of the few studies to explore the supervisory relationship under conditions where the supervisor is not the expert. Second, our research design allowed us to build on participants’ actual learning experiences and as such we developed a rich understanding of the impact of COVID-19 on the workplace and of the supervisory relationship therein. The limitations of this study are that we had to narrow our methods to online interviews. Observations could not be included due to the restrictions on interpersonal contact at the time of our data collection. Our study might have benefited from observations as a means for triangulation of the data that could potentially lead to even richer data. Implications for practice and further research This study provides insights into the supervisory relationship in relation to workplace learning in disruptive times and has clues to improve the training of future general practitioners. First, collective learning in teams occurs in response to major common challenges. It is ad hoc and implicit . To further enrich the learning experiences, collective learning can be complemented by reflection. Reflection creates a better understanding of oneself and the situation, which helps to inform future actions . Reflection can be promoted in the training institution during a release day by encouraging residents, supervisors and, possibly, other staff to discuss and reflect together on situations and to formulate lessons learnt. This may be done retrospectively, after the turmoil in the GP practice has subsided. Secondly, it is important that residents have sufficient and varied learning opportunities to develop as GPs. Whether this is possible in a period of disruption depends in part on the learning opportunities available in the workplace. The training institution can play a mediating and facilitating role by bringing together different supervisors to share or create opportunities. We also see several avenues for further research. We need a better understanding of less established forms of clinical workplace learning, including collective learning, bi-directional learning and reverse mentoring - which we anticipate will become more common in the future. Situations where the supervisor is no longer the expert are expected to become more common. We expect residents will be more familiar with eHealth applications and ways to interact with patients through social media than their supervisors, and for residents to take on a mentoring role. To further our understanding of learning between supervisors and residents on such topics, an observational study could be conducted to explore what kind of learning occurs, its nature, and its barriers and facilitators.
In this study we explored the impact of a disruptive situation, i.e. the COVID-19 pandemic, on the relationship between GP residents and their supervisors. We found that the disruptive uncertainty, the extensive consequences for the patient care and the impact of learning opportunities for residents resulted in three types of collaboration. Within the supervisory relationship, supervisors and residents collaborated to get the job done and to facilitate residents’ learning. Beyond the supervisory relationship, collective learning emerged as supervisors, residents and other team members collaborated in information sharing and joint meaning-making to enhance their understanding of the situation and to improve patient care. While the first two types of collaboration, getting the job done and residents’ learning, are typical for workplace learning, the extent to which they occurred was different. Getting the job done became more prominent as residents were encouraged to quickly work independently. In contrast, creating learning opportunities for residents required more attention and inventiveness since the regular training structure fell away. The third type of collaboration, collective learning, i.e. learning with and from each other, including non-supervising GPs and the entire team, has received little attention in the literature on general practice training. Our study contributes to the understanding of the supervisory relationship in GP training, by providing insight into the characteristics of this relationship during a period of major and unprecedented change. As we discuss our findings in the light of other literature, we will begin with the impact of uncertainty on the supervisory relationship during the COVID-19 pandemic, before we continue with the types of collaboration and the underlying interaction patterns that we found.
Clinical uncertainty is a key aspect of general practice as the entry point to the healthcare system . Residents learn to manage and tolerate this uncertainty by observing how their supervisors deal with uncertainty. Residents develop the same strategies for responding to uncertainty as their supervisors . However, the type of uncertainty we addressed in this study is pervasive and disruptive - causing problems so that something cannot proceed normally . The scope of the uncertainty we found was not limited to clinical uncertainty, yet was much broader. Supervisors, residents and other team members were faced with the unpredictability of the disease and its impact on care delivery. As a result, even supervisors had to learn to manage this type of uncertainty. Uncertainty was therefore a general and shared lack of clarity about clinical activities and the organisation of patient care, rather than a ‘state of mind’ of an individual, often the resident . Interestingly, residents and supervisors used strategies to cope with uncertainty similar to those described by Han and colleagues (2021), namely , i.e. (1) ignorance-focussed; (2) uncertainty-focussed; (3) response-focussed; and (4) relationship-focussed. They reduced their ignorance by, for example, seeking medical information. They changed their response to uncertainty by learning to accept their limited influence on patients with COVID disease. Finally, they shared doubts or discomfort they felt with their team, a ‘relationship’-focussed strategy. Our study adds collective strategies, such as shared meaning-making, to the individual strategies used by supervisors and residents to deal with disruptive uncertainty.
The influence of disruptive uncertainty is reflected in the types of collaboration and underlying patterns of interaction between supervisor and resident pairs. The first type of collaboration, ‘getting the job done’, involved the underlying interaction patterns as described by Wiese et al. (2018) of supervisors’ monitoring and entrustment, and residents’ help seeking behaviour. We found that supervisors immediately entrusted residents with patient care. While entrustment in general practice residency develops rather rapidly, holistically and presumptively , the pace by which entrustment was awarded here was remarkable. The residents, in turn, accepted the invitation to patient care as though they were already qualified colleagues. This type of collaboration facilitates residents’ participation and active engagement in the professional community as they took on increasing responsibility for patient care . Sfard’s metaphor of learning as ‘participation’ is appropriate here. It conceives of learning as a process of meaning construction and identity formation through participation in community activities . The second type of collaboration, residents’ learning, involved supervisors creating sufficient learning opportunities for residents to develop the competencies required for the diverse range of patients in primary care. This resembles Sfard’s metaphor of learning as ‘acquisition’ . In this metaphor learning is seen as acquisition of knowledge, skills, and attributes that are ‘owned’ by the individual . We did not find any signs of supervisors demonstrating their expertise nor of residents integrating their supervisors’ modelled skills. This might be because the supervisors did not have more specific expertise than the residents, or because of the limitations in interpersonal contact, which prevented supervisors and residents from seeing patients together. The third type of collaboration, collective learning, involved residents, supervisors and other staff creating an understanding of the impact of the pandemic and developing new knowledge for patient care and policy. They did this through a collaborative process of meaning making and problem solving. This type of collaboration facilitated the joint creation of knowledge. This resembles Paavola and Hakkarainen’s metaphor of learning as ‘knowledge creation’ . In this metaphor, learning is seen as the collaborative development of activity in response to challenges. In short, our findings show how supervisors and residents collaborate in workplace learning during a disruptive situation. Although we found some similarities with the learning processes and interaction patterns described by Wiese et al. , other features of the supervisory relationship were highlighted during the pandemic. This adds to our understanding of the relationship between supervisor and resident in workplace learning.
The most notable finding for us is that of collective learning, as it is a clear complement to workplace learning between supervisors and residents and the supervisory relationship in times of disruption. However, it is not a surprising finding. Collective learning appears to be an intuitive response to new or complex challenges , and during the first outbreak of COVID-19, no one had more expertise than anyone else. Also, the shared challenges supervisors, residents and other staff faced during the pandemic created a sense of belonging, eliminating the sense of hierarchy that could hinder collective learning . Collective learning is internally generated by a team who are willing to learn and who are aware of the importance of finding solutions to practice problems . Because it is often implicit, collective learning is frequently unnoticed by professionals . The disruptive situation of COVID-19 made collective learning explicit.
This study has several methodological strengths and some limitations. First, to our knowledge, this is one of the few studies to explore the supervisory relationship under conditions where the supervisor is not the expert. Second, our research design allowed us to build on participants’ actual learning experiences and as such we developed a rich understanding of the impact of COVID-19 on the workplace and of the supervisory relationship therein. The limitations of this study are that we had to narrow our methods to online interviews. Observations could not be included due to the restrictions on interpersonal contact at the time of our data collection. Our study might have benefited from observations as a means for triangulation of the data that could potentially lead to even richer data.
This study provides insights into the supervisory relationship in relation to workplace learning in disruptive times and has clues to improve the training of future general practitioners. First, collective learning in teams occurs in response to major common challenges. It is ad hoc and implicit . To further enrich the learning experiences, collective learning can be complemented by reflection. Reflection creates a better understanding of oneself and the situation, which helps to inform future actions . Reflection can be promoted in the training institution during a release day by encouraging residents, supervisors and, possibly, other staff to discuss and reflect together on situations and to formulate lessons learnt. This may be done retrospectively, after the turmoil in the GP practice has subsided. Secondly, it is important that residents have sufficient and varied learning opportunities to develop as GPs. Whether this is possible in a period of disruption depends in part on the learning opportunities available in the workplace. The training institution can play a mediating and facilitating role by bringing together different supervisors to share or create opportunities. We also see several avenues for further research. We need a better understanding of less established forms of clinical workplace learning, including collective learning, bi-directional learning and reverse mentoring - which we anticipate will become more common in the future. Situations where the supervisor is no longer the expert are expected to become more common. We expect residents will be more familiar with eHealth applications and ways to interact with patients through social media than their supervisors, and for residents to take on a mentoring role. To further our understanding of learning between supervisors and residents on such topics, an observational study could be conducted to explore what kind of learning occurs, its nature, and its barriers and facilitators.
The supervisory relationship was characterized by three types of collaboration during the outbreak of COVID-19: (1) getting the job done, (2) residents’ learning and (3) collective learning. Collective learning complements our general understanding of the supervisory relationship and workplace learning between supervisors and residents. Collective learning occurred when supervisors of residents and other staff were faced with the disruptive effects of COVID-19, where supervisors were no longer the experts. Situations where supervisors are no longer the experts are likely to become more common with the introduction of e-health applications and interaction with patients through social media. Therefore, a better understanding of collective learning and other less established forms of clinical learning in the workplace is needed.
Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 2
|
Argyrophilic grain disease is common in older adults and may be a risk factor for suicide: a study of Japanese forensic autopsy cases | 062f8d5a-c0a5-4b59-bde6-273f2c1c8866 | 10067165 | Forensic Medicine[mh] | Argyrophilic grains (AGs) were first reported by Braak and Braak in 1987. Their pathological features are punctate or filiform structures in the neuropil, which can be visualized using Gallyas–Braak silver staining . Previous studies have shown a high frequency of AGs among older adult patients with dementia , which has led to AGs or argyrophilic grain disease (AGD) being proposed as an isolated neuropathological entity [ – ]. Neuropathological diagnosis of AGD is currently based on a specific pathological appearance: AGs consisting of four-repeat isoform tau protein , neuronal cytoplasmic tau-positive inclusions referred to as “pretangles”, and oligodendroglial coiled bodies . Tau-positive granular/fuzzy astrocytes , which were initially reported by Botez et al. as bush-like astrocytes, as well as ballooned neurons, are also considered as associated lesions of AGD (Fig. ). However, granular/fuzzy astrocytes are now defined as age-related tau astrogliopathy and thus, there is currently no AGD-specific tau-positive astrocytic lesion. Earlier studies have established that AGs are commonly distributed in various areas of the limbic system [ – ]. Saito et al. proposed the staging paradigm of AGs: AGs start with the involvement of the ambient gyrus and amygdala (stage 1: ambient stage), then extend into the posterior parahippocampal gyrus and medial anterior temporal pole (stage 2: medial temporal stage), and finally extend to the basal forebrain and anterior cingulate gyrus (stage 3: frontal stage) . Although AGD has four-repeat tauopathy, similar to progressive supranuclear palsy (PSP) and corticobasal degeneration (CBD), AGs are rarely found in the basal ganglia, brain stem nuclei, or cerebellum . However, rare cases with an extensive distribution of AGs in the neocortex have been reported . A clinical definition of AGD has not yet been established. Although AGD is considered a distinct clinicopathological entity and a cause of dementia , this is likely because of the heterogeneous clinical presentation and the frequent overlap with other neurodegenerative diseases . The relatively high frequency of AGD in cognitively normal older adults over 60 years old prompts the question of whether AGD is a distinct neurological disease entity or a by-product of the aging process [ – ]. AGD has been reported in subjects with a relatively late onset of prominent psychiatric symptoms, which may be associated with the anatomical distribution of AGs [ – ]. Although neuropathological investigations targeting neuropsychiatric disorders, including suicide, and neurodegenerative diseases are scarce, they may be valuable [ , – ]. In our previous investigation in younger subjects aged under 40 years, Alzheimer’s disease (AD)-related pathology was not shown to be a significant accelerating factor for suicide , and we previously observed AGD in a suicide victim aged 40 years . In the present study, we investigated a large number of serial forensic autopsy cases to reveal the frequency and comorbid pathology of AGD and examine the clinical appearance by comparing AGD cases with non-AGD cases.
Subjects We reviewed the archives of all 1982 medicolegal autopsy cases in our department between 2007 and 2020. Of these cases, brain specimens from 1449 people aged over 40 years (881 males and 568 females, aged 40–101 years, mean age 70.0 ± 14.1 years) without severe injury or severe postmortem degeneration (e.g., liquefaction preventing histological analysis), large chronic infarcts (> 1 cm), and/or multiple lacunas over three were examined. A total of 359 cases died of natural causes, and 669 cases had an accidental traumatic death, such as a fall, traffic accident, burning, drowning, and hypothermia. Suicide accounted for 395 cases, and 23 cases died by homicide. There were 1103 cases (625 males and 478 females) aged over 60 years and 438 cases (208 males and 230 females) aged over 80 years. The clinical history of patients was obtained from the family and police examination records. For cases with a history of hospital visits, medical records were provided by the primary physician. Tissue sampling and pathological assessment All brains were fixed in 20% buffered formalin for a minimum of 2 weeks before sampling. Specimens that are routinely sampled at our department have been described in a previous report and are shown in Additional file : Figure S1. All sections were cut and stained with luxol fast blue-hematoxylin and eosin. Gallyas–Braak and Holzer staining was also performed. Routine immunohistochemistry was performed for samples of the frontal lobe, temporal lobes including the amygdala and hippocampus, basal ganglia, and midbrain of all cases to detect phosphorylated tau (clone AT8, 1:1000; Endogen, Woburn, MA), phosphorylated α-synuclein (clone LB508, 1:500; Zymed, San Francisco, CA), TAR DNA binding protein-43 (TDP-43; 1:5000; Protein Tech Group, Chicago, IL), glial fibrillary acidic protein (clone ZCG 29, 1:1000, Nichirei Tokyo, Japan), and β-amyloid (Aβ; clone 6F/3D, 1:50; Novocastra Vector Labs, Burlingame, CA). Antibody binding was detected using a biotin-streptavidin detection system (Nichilei, Tokyo, Japan) using 3,3′-diaminobenzidine as the chromogenic substrate. If positive findings were detected in the preliminary immunohistochemistry, an additional staining procedure was performed for subsequent sections. Staining for three- and four-repeat tau (Merck-Millipore, Billerica, MA) was also performed in cases positive for AT8. The pathological staging of neurofibrillary tangles (NFTs) was evaluated according to the modified Braak stages of NFT burden using AT8, Gallyas–Braak . The density of neuritic plaques was evaluated in accordance with the Consortium to Establish a Registry for Alzheimer’s disease (CERAD) criteria using thioflavin-S and Aβ immunostaining . The extent of senile plaques in the brain was evaluated using the criteria of Thal et al. . Based on these results, the degree of AD-related neuropathological change was divided into four categories based on the National Institute on Aging-Alzheimer’s Association (NIA-AA) guidelines : none, low, intermediate, and high. The pathology of Lewy body (LBs) disease was assessed according to the Third Consensus Guidelines for Dementia with LBs and the Braak stages for the development of Parkinson’s disease-related pathology using α-synuclein immunohistochemistry [ – ]. AGs were detected using Gallyas–Braak staining, and the pathological staging of AGs was assessed following the AGD system proposed by Saito et al. . We used the National Institute of Neuronal Disorders and Stroke criteria to neuropathologically diagnose PSP . The pathological type of TDP-43 proteinopathy was assessed according to the stages of AD and the classification system for frontotemporal lobar degeneration-TDP pathology . Investigation of predominant AGD subjects Because AGD cases often have comorbid pathology, we extracted predominant AGD cases who had no comorbid pathology to examine the clinical significance of AGD. Predominant AGD cases were defined as cases with only mild AD pathology (Braak tau stage 2 or below, Thal phase 2 or below, or CERAD A or below) and without LBs or TDP-43 pathology. Controls were defined as cases with only mild AD lesions of the same severity as the above. Statistical analysis Differences in continuous variables were analyzed using Student’s t -tests, and differences in ordinal variables were analyzed using chi-square tests. Spearman’s rank correlation test was used to analyze correlations. Although pathological stage was an ordinal variable, it was analyzed as a continuous variable. A propensity score was used to adjust for bias. Bonferroni correction was performed for multiple comparisons, and significance was set at P < 0.05. SPSS Statistics version 26 (IBM Corporation, Armonk, NY) and JMP 14.3 (JMP Statistical Discovery LLC, Cary, NC) were used for all analyses.
We reviewed the archives of all 1982 medicolegal autopsy cases in our department between 2007 and 2020. Of these cases, brain specimens from 1449 people aged over 40 years (881 males and 568 females, aged 40–101 years, mean age 70.0 ± 14.1 years) without severe injury or severe postmortem degeneration (e.g., liquefaction preventing histological analysis), large chronic infarcts (> 1 cm), and/or multiple lacunas over three were examined. A total of 359 cases died of natural causes, and 669 cases had an accidental traumatic death, such as a fall, traffic accident, burning, drowning, and hypothermia. Suicide accounted for 395 cases, and 23 cases died by homicide. There were 1103 cases (625 males and 478 females) aged over 60 years and 438 cases (208 males and 230 females) aged over 80 years. The clinical history of patients was obtained from the family and police examination records. For cases with a history of hospital visits, medical records were provided by the primary physician.
All brains were fixed in 20% buffered formalin for a minimum of 2 weeks before sampling. Specimens that are routinely sampled at our department have been described in a previous report and are shown in Additional file : Figure S1. All sections were cut and stained with luxol fast blue-hematoxylin and eosin. Gallyas–Braak and Holzer staining was also performed. Routine immunohistochemistry was performed for samples of the frontal lobe, temporal lobes including the amygdala and hippocampus, basal ganglia, and midbrain of all cases to detect phosphorylated tau (clone AT8, 1:1000; Endogen, Woburn, MA), phosphorylated α-synuclein (clone LB508, 1:500; Zymed, San Francisco, CA), TAR DNA binding protein-43 (TDP-43; 1:5000; Protein Tech Group, Chicago, IL), glial fibrillary acidic protein (clone ZCG 29, 1:1000, Nichirei Tokyo, Japan), and β-amyloid (Aβ; clone 6F/3D, 1:50; Novocastra Vector Labs, Burlingame, CA). Antibody binding was detected using a biotin-streptavidin detection system (Nichilei, Tokyo, Japan) using 3,3′-diaminobenzidine as the chromogenic substrate. If positive findings were detected in the preliminary immunohistochemistry, an additional staining procedure was performed for subsequent sections. Staining for three- and four-repeat tau (Merck-Millipore, Billerica, MA) was also performed in cases positive for AT8. The pathological staging of neurofibrillary tangles (NFTs) was evaluated according to the modified Braak stages of NFT burden using AT8, Gallyas–Braak . The density of neuritic plaques was evaluated in accordance with the Consortium to Establish a Registry for Alzheimer’s disease (CERAD) criteria using thioflavin-S and Aβ immunostaining . The extent of senile plaques in the brain was evaluated using the criteria of Thal et al. . Based on these results, the degree of AD-related neuropathological change was divided into four categories based on the National Institute on Aging-Alzheimer’s Association (NIA-AA) guidelines : none, low, intermediate, and high. The pathology of Lewy body (LBs) disease was assessed according to the Third Consensus Guidelines for Dementia with LBs and the Braak stages for the development of Parkinson’s disease-related pathology using α-synuclein immunohistochemistry [ – ]. AGs were detected using Gallyas–Braak staining, and the pathological staging of AGs was assessed following the AGD system proposed by Saito et al. . We used the National Institute of Neuronal Disorders and Stroke criteria to neuropathologically diagnose PSP . The pathological type of TDP-43 proteinopathy was assessed according to the stages of AD and the classification system for frontotemporal lobar degeneration-TDP pathology .
Because AGD cases often have comorbid pathology, we extracted predominant AGD cases who had no comorbid pathology to examine the clinical significance of AGD. Predominant AGD cases were defined as cases with only mild AD pathology (Braak tau stage 2 or below, Thal phase 2 or below, or CERAD A or below) and without LBs or TDP-43 pathology. Controls were defined as cases with only mild AD lesions of the same severity as the above.
Differences in continuous variables were analyzed using Student’s t -tests, and differences in ordinal variables were analyzed using chi-square tests. Spearman’s rank correlation test was used to analyze correlations. Although pathological stage was an ordinal variable, it was analyzed as a continuous variable. A propensity score was used to adjust for bias. Bonferroni correction was performed for multiple comparisons, and significance was set at P < 0.05. SPSS Statistics version 26 (IBM Corporation, Armonk, NY) and JMP 14.3 (JMP Statistical Discovery LLC, Cary, NC) were used for all analyses.
Epidemiology Of the 1449 cases, we detected 342 AGD cases (23.6%; 177 males and 165 females), with a mean age of 79.7 years. The frequency of AGD was higher than that in numerous other studies (Table ). The 1107 cases without AGD (704 males and 405 females) had a mean age of 67.0 years. There were significant differences in age and sex between the two groups (Table ). The youngest AGD case was a 46-year-old male reported previously , and the oldest AGD case was 100 years old. The frequency and stage of AGD increased with age (Spearman’s rank correlation test P < 0.001), and 1.2% of patients aged 40–49 years and 45.7% of patients aged over 80 years had AGD. Pathological AGD stage also increased with age, and 23.7% of patients aged over 80 years were classified as AGD stage 3 (Fig. ). Neither frequency nor AGD stage differed by sex in any age group (Fig. ). Comorbid pathology In the AGD cases, 51 (14.9%) had a high degree of AD pathology according to NIA-AA criteria, 92 (26.9%) had LB pathology, 46 (13.5%) had TDP-43 pathology, and 44 (12.9%) had PSP pathology. These conditions were more frequently observed in AGD patients than in AGD-negative cases. After matching for age and sex, the comorbidity of these conditions remained significantly higher in the AGD group than in the AGD-negative group, except for Aβ pathology. AD-related tau comorbidity was more severe in AGD cases than in non-AGD cases; however, the comorbidity rate of advanced-stage AD did not differ between the AGD and non-AGD groups (Table ). The severity and comorbidity of these pathologies tended to increase with the stage of AGD, which was not observed in the age-matched AGD-negative group (Table ). In addition, there was a sex difference in comorbidity; TDP-43 was significantly more common in women than in men, and PSP lesions were significantly more common in men than in women (Additional file : Tables S1 and S2). Clinical appearance of AGD In the AGD cases, 80 (23.4%) patients had dementia, although the difference was not significant when controlled for age, sex, and comorbid pathology. Fifty-one subjects (15.2%) had a history of psychiatric visits. The number of cases for each cause of death was 175 (51.2%) for accidental death, 109 (31.9%) for suicide, and 52 (15.2%) for natural causes. The rates of suicide and medical history of psychiatric disorders were significantly higher in these AGD cases than in the AGD-negative cases matched for age, sex, and comorbid pathology (Table ). The suicide rate of AGD cases was higher than that of AGD-negative cases in all age groups, although the difference was significant in only the 60 years and older age group. Moreover, the relative risk of suicide was 2.61 (1.01–6.73) in the 90 years and older group (Fig. ). The suicide rates for each AGD stage were 34.9% for stage 1, 29.4% for stage 2, and 32.9% for stage 3. Rates did not differ between stages; however, they were significantly higher in the AGD cases than in the AGD-negative cases across all stages (Table ). The relative risk of suicide was 1.72 (1.30–2.26) and was statistically significant in female cases, with a relative risk of 1.80 (1.28–2.52), but not in male cases (Additional file : Tables S1 and S2). The relative risk of suicide increased to 2.27 (1.20–4.30) and 6.50 (1.58–26.76) in AGD cases with Lewy body pathology and PSP, respectively, and decreased to 0.88 (0.38–2.10) in AGD cases with advanced AD pathology. Dementia cases were significantly more common in the AGD group (23.4%) than in the non-AGD group. The relative risk of dementia was 1.38 (1.00–1.93) in patients with only AGD and increased to 1.73 (0.99–3.02) and 3.75 (1.93–7.29) in AGD cases with Lewy body pathology and severe AD pathology, respectively (Table ). Predominant AGD cases Thirty-nine cases (11.4%) were predominant AGD cases, of whom 16 (41.0%) were suicidal and four (10.3%) had dementia. A comparison of cases between these AGD cases and those with the same level of mild AD pathology only (i.e., the control group), matched for age, sex, and AD pathology, showed a significantly higher suicide rate and a significantly lower rate of natural death in the predominant AGD group (Table ).
Of the 1449 cases, we detected 342 AGD cases (23.6%; 177 males and 165 females), with a mean age of 79.7 years. The frequency of AGD was higher than that in numerous other studies (Table ). The 1107 cases without AGD (704 males and 405 females) had a mean age of 67.0 years. There were significant differences in age and sex between the two groups (Table ). The youngest AGD case was a 46-year-old male reported previously , and the oldest AGD case was 100 years old. The frequency and stage of AGD increased with age (Spearman’s rank correlation test P < 0.001), and 1.2% of patients aged 40–49 years and 45.7% of patients aged over 80 years had AGD. Pathological AGD stage also increased with age, and 23.7% of patients aged over 80 years were classified as AGD stage 3 (Fig. ). Neither frequency nor AGD stage differed by sex in any age group (Fig. ).
In the AGD cases, 51 (14.9%) had a high degree of AD pathology according to NIA-AA criteria, 92 (26.9%) had LB pathology, 46 (13.5%) had TDP-43 pathology, and 44 (12.9%) had PSP pathology. These conditions were more frequently observed in AGD patients than in AGD-negative cases. After matching for age and sex, the comorbidity of these conditions remained significantly higher in the AGD group than in the AGD-negative group, except for Aβ pathology. AD-related tau comorbidity was more severe in AGD cases than in non-AGD cases; however, the comorbidity rate of advanced-stage AD did not differ between the AGD and non-AGD groups (Table ). The severity and comorbidity of these pathologies tended to increase with the stage of AGD, which was not observed in the age-matched AGD-negative group (Table ). In addition, there was a sex difference in comorbidity; TDP-43 was significantly more common in women than in men, and PSP lesions were significantly more common in men than in women (Additional file : Tables S1 and S2).
In the AGD cases, 80 (23.4%) patients had dementia, although the difference was not significant when controlled for age, sex, and comorbid pathology. Fifty-one subjects (15.2%) had a history of psychiatric visits. The number of cases for each cause of death was 175 (51.2%) for accidental death, 109 (31.9%) for suicide, and 52 (15.2%) for natural causes. The rates of suicide and medical history of psychiatric disorders were significantly higher in these AGD cases than in the AGD-negative cases matched for age, sex, and comorbid pathology (Table ). The suicide rate of AGD cases was higher than that of AGD-negative cases in all age groups, although the difference was significant in only the 60 years and older age group. Moreover, the relative risk of suicide was 2.61 (1.01–6.73) in the 90 years and older group (Fig. ). The suicide rates for each AGD stage were 34.9% for stage 1, 29.4% for stage 2, and 32.9% for stage 3. Rates did not differ between stages; however, they were significantly higher in the AGD cases than in the AGD-negative cases across all stages (Table ). The relative risk of suicide was 1.72 (1.30–2.26) and was statistically significant in female cases, with a relative risk of 1.80 (1.28–2.52), but not in male cases (Additional file : Tables S1 and S2). The relative risk of suicide increased to 2.27 (1.20–4.30) and 6.50 (1.58–26.76) in AGD cases with Lewy body pathology and PSP, respectively, and decreased to 0.88 (0.38–2.10) in AGD cases with advanced AD pathology. Dementia cases were significantly more common in the AGD group (23.4%) than in the non-AGD group. The relative risk of dementia was 1.38 (1.00–1.93) in patients with only AGD and increased to 1.73 (0.99–3.02) and 3.75 (1.93–7.29) in AGD cases with Lewy body pathology and severe AD pathology, respectively (Table ).
Thirty-nine cases (11.4%) were predominant AGD cases, of whom 16 (41.0%) were suicidal and four (10.3%) had dementia. A comparison of cases between these AGD cases and those with the same level of mild AD pathology only (i.e., the control group), matched for age, sex, and AD pathology, showed a significantly higher suicide rate and a significantly lower rate of natural death in the predominant AGD group (Table ).
Although we identified AGD cases using Gallyas–Braak staining, which is considered less sensitive than immunohistochemistry, the frequency of AGD was higher than that in most other studies. With a few exceptions, most forensic autopsies in Japan are performed under the criminal code and are performed when the cause of death is suspected to be unnatural or linked to a crime . Although our previous study of a large sample of forensic autopsy cases may not represent the general Japanese population because of a lower frequency of bedridden cases in the terminal phase of various diseases, we showed that the frequency of PSP and CBD may be higher than expected in a forensic autopsy sample . In PSP patients, in particular, a high incidence of lethal traumatic injury and suicide may be associated with the high frequency of PSP in our previous study. Therefore, a high frequency of unnatural death cases, especially suicide cases, may contribute to the high frequency of AGD in the present study. In contrast, an autopsy study conducted in Japan by Saito et al. showed a high frequency of AGD (36.2%) in hospital autopsy cases, despite the exclusion of suicide cases . Although the mean age of participants in the study by Saito et al. (80.6 ± 8.9 years) was higher than that of the participants in the present study , the frequency of AGD in each age group was similar across both studies. The difference in the frequency of AGD may be associated with the difference in the age of the sampled cases. It is worth noting that the frequency of neurodegenerative diseases as investigated by autopsy may vary depending on the investigated population. Although studies have shown that AGD is associated with a higher rate of mild cognitive impairment , several studies have shown a high frequency of AGD pathology in control cases [ , , ]. Rodriguez et al. revealed that in AGD cases without other neuropathological findings (8 of 152 AGD cases), half of them had some degree of cognitive decline, whereas 59% of their 152 AGD subjects were cognitively normal . In addition, Davis et al. identified AGD in 23% of 59 cognitively normal older adult subjects , and Knopman et al. identified 12 cases of AGD among 59 cognitively normal older adult subjects . Wurms et al. showed varied clinical manifestations in younger AGD subjects (aged under 75 years) without significant comorbid neuropathological lesions . In our study, we used a propensity score to compare AGD subjects with age-, sex-, and comorbid pathology-matched control cases and showed that AGD was not a distinct neuropathological lesion of dementia. This result is consistent with the conclusion of several previous studies [ , , , , ]. However, we cannot conclude from our findings that AGD is not entirely associated with the development of dementia because the neurological findings based on clinical evaluations in many of our forensic autopsy cases tended to be less comprehensive than previous studies targeting hospitalized patients or brain donation program participants. In summary, we assumed that the effect of AGD on dementia may be relatively weak and that other pathologies, such as AD or LB pathology, may be more strongly associated with dementia in subjects with AGD. We showed that AGD may be a distinct contributing factor to psychiatric disorders in the older adult population. Several psychiatric conditions, such as aggression, irritability, depression, psychosis, and mild dementia, have been shown to be associated with AGD [ , , , ], and several studies have revealed that these psychological changes tend to precede the appearance of dementia in patients with AGD, whereas dementia tends to appear before psychological changes in patients with AD [ , , ]. Of these various psychiatric symptoms, late-life depression (LLD) has been shown to cause cognitive dysfunction and increase the risk of dementia and AD by two folds; however, the nature of the relationship between LLD and dementia has not yet been explored . Togo et al. also showed that amnesia and emotional disorders are frequently present in patients with AGD, whereas other cognitive functions tend to be spared relative to the severity of amnesia. Shioya et al. investigated the pathological appearance of 11 older adult cases with bipolar disorder and showed that AGD may be associated with late-onset bipolar disorder in middle-aged or older subjects . Furthermore, Jellinger showed that personality changes and frontal lobe signs are much more prominent in AGD patients than in dementia patients. Nagao et al. showed that late-onset delusions occur significantly more frequently in patients with AGD than in those with minimal AD pathology alone and suggested that AGD is associated with the occurrence of late-onset schizophrenia and delusional disorders. Although the specific mechanism underlying psychiatric symptoms in AGD patients has not yet been established, psychiatric disorders tend to occur in various conditions after the limbic region and temporal cortex become affected [ – ]. Investigations in subjects with Parkinson’s disease (PD) have shown that dysfunctions of the limbic system and brainstem nuclei are associated with depression . Moreover, other selective and morphological limbic dysfunctions, such as amnesia and behavioral changes, have been reported to be caused by cerebrovascular disease, traumatic brain injury, epilepsy , herpes simplex encephalitis, and paraneoplastic syndrome . Specifically, the amygdala, which is a crucial hub of the emotional processing neural system, has been implicated in LLD pathophysiology . In addition to the initial and predominant distribution of AGs in patients with AGD, we found that subjects in the early pathological stage of AGD are also at risk of developing a psychiatric disorder. This suggests that the involvement of the amygdala is an important mechanism in the progression of psychiatric disorders in subjects with AGD. Numerous functional magnetic resonance imaging (fMRI) studies have also shown the involvement of the amygdala in various psychiatric disorders, including depression . However, morphometric analyses of the amygdala of patients with depression using clinical imaging data have provided inconsistent results, including an increase, a decrease, and no change . Surdhar et al. reported that amygdala volumes are significantly smaller in PD patients with mild depressive symptoms than in healthy controls. Leal et al. investigated older adults with and without depressive symptoms using fMRI to discern signals in the hippocampal subfields and amygdala nuclei and showed that the disruption of the amygdala–entorhinal–hippocampal network is associated with LLD . Davey et al. suggested that the amygdala is involved in the generation of negative affect that characterizes depression . These radiological findings suggest an association between AGD and late-onset psychiatric disorders, such as LLD, due to the morphological alteration of the amygdala caused by the accumulation of AGs. A key result of the current study is that AGD is a distinct risk factor for suicide in the older adult population. As an individual ages, quality of life may decrease because of a decline in physical and cognitive abilities or illnesses. Chronic diseases and reduced strength lead to a sense of worthlessness and anxiety, which can eventually result in depression [ – ]. Epidemiological studies have identified various risk factors for suicidal thoughts and behaviors; however, whether these findings can generalize to other cultures remains unclear . Obuobi-Donker et al. revealed that adults aged 60 years and above are at a risk of developing LLD, which can expose them to suicidal behaviors. Psychiatric and physical illnesses, functional impairments, and social/economic factors may also contribute to suicidal behavior in older adults . A study of 538 suicide cases indicated that suicidal behavior is a product of the interaction of numerous factors; however, no single independent risk factor was found . In the present study, suicide rates did not differ between AGD stages, but the suicide rate was significantly higher in AGD cases, even at stage-1 AGD, than in non-AGD cases. Therefore, the involvement of the amygdala and the accumulation of AGs may be associated with suicidal behavior. Detailed neuropathological examinations of suicide victims investigating the correlation between suicide and neurodegenerative diseases are scarce. To date, three studies have investigated AD pathology in victims of suicide, although conclusions varied. Rubio et al. examined the autopsies of 28 individuals who had died of suicide and found that the frequency of AD pathology was higher than that in the control group. In contrast, Peisah et al. did not find a positive correlation between suicide and AD pathology. Moreover, Matschke et al. found no difference in the frequency of AD-type pathology in the brain between suicide victims and control cases . However, these studies were all limited by the lack of standard neuropathological investigations and evaluations of AD pathology. Furthermore, the amygdala was not studied. We showed that impaired function of the limbic system, including the amygdala, may be a neuropathological substrate for both psychiatric disorders and suicidal behavior. However, further clinicopathological investigations are required to explore why there was no significant difference in the rate of suicide between different stages of AGD. We found a significantly higher frequency of suicide in women in our sample, although men also showed a similar tendency. Several epidemiological studies have focused on sex difference in risk factors for LLD and/or suicide [ – ]. Although older adult women are more likely to experience depression than older adult men , men are more likely to die by suicide, whereas women are more likely to attempt suicide . The mechanism underlying sex difference in suicidal behavior remains unclear because only a few studies have been conducted, and such studies have not found a significant interaction between various risk factors and suicidal behavior . It is uncertain why the suicide rate was significantly higher in the AGD group than in the non-AGD group in women only, especially because the frequency of AGD did not differ between men and women. Possible causes are the difference in the sensitivity of the limbic system to AGs between men and women or the presence of other potent accelerating factors for suicidal behavior in men. Crestani et al. investigated 538 suicide cases and showed that chronic and debilitating diseases, often accompanied by profound psychological symptoms, are powerful stimuli to prompt suicide among men, whereas mental state is a significant risk factor for women, as the majority of them had depression . In older adult subjects, mixed pathology (i.e., mixed neurodegenerative pathology or mixed neurodegenerative and other pathologies) is thought to lower the threshold for developing cognitive impairment and dementia; moreover, the severity of mixed pathologies increases with age and correlates with the severity of clinical symptoms . We revealed that the overlap of AGD with PSP or LB pathology may increase the risk of suicide in the present study. Indeed, we previously reported that AGD may be a significant risk factor for suicide attempts in older adults with a clinical history of acute post-stroke depression or incipient PSP lesions . Neuropsychiatric changes are common in PSP patients, especially apathy, and depression is less common , although another study suggested that depression is highly common in patients with PSP . Although we did not examine the detailed neuropsychiatric appearance of subjects with overlapping AGD and PSP or LB pathology, the overlap of AGD with other pathologies may increase the amount of pathological substrate in the limbic system, which may lead to the development of psychiatric disorders, including suicide attempts. On the other hand, although it remains unclear why the overlapping AD-related tau pathology decreased the risk of suicide in the present study, we speculate that the ability to attempt suicide and/or complete suicide is diminished as cognitive impairment progresses owing to the deterioration of AD pathology. A recent clinical study reported that subjects younger than 65 years in the earlier stages of dementia had a 6.69-times (95% CI 1.49–30.12) higher suicide risk compared with subject without dementia . Further investigations of neuropsychiatric symptoms, including those in the earlier stages of dementia, and their association with suicidal behavior in elderly subjects with autopsy-proven neurodegenerative disease are essential. In addition to a certain level of bias in our study population, our study was also limited by the lack of clinical information of some cases, which was mainly because they lacked severe clinical symptoms or had low rates of neurologist consultation. In particular, the level of education, the standard of living and other possible confounding factors such as debilitating illness, overall weakness, and the financial situation of each patient were not evaluated. In addition, we identified AGD subjects by examining the left hemisphere of the brain. However, neuropathological asymmetry in AGD subjects as revealed by Adachi et al. has been shown to be associated with the progression of psychiatric symptoms, including suicide and/or dementia . Finally, we did not examine whether a larger amount of AGs are associated with the development of clinical symptoms. Thus, additional clinicopathological investigations, especially more detailed clinical information and investigation of the association between suicide and other abnormal protein accumulations, may be required.
Although the frequency of AGD in autopsy studies varies depending on the population studied, we show here that the frequency of AGD in a Japanese forensic autopsy series is 25.2% in individuals aged over 40 years. The frequency increases with age, where 1.2% of patients in their 40s have AGD, and 45.7% of patients aged over 80 years have AGD. AGD is identified as a significant and distinct risk factor for psychiatric hospital visits and completion of suicide but not for dementia. Overlapping PSP and LB pathology increases the risk of suicide, whereas overlapping advanced AD pathology decreases the risk of completing suicide. Morphological changes in the limbic system, including an increase in AGs in the amygdala, may be associated with the development of psychiatric disorders and suicide attempts in older adults with AGD. The development of methods to detect AGs in clinical practice may help prevent suicide in older adults.
Additional file 1: Fig. S1. Low power view of the histological specimen (luxol fast blue-hematoxylin and eosin). Table S1. Clinicopathological features of male argyrophilic grain disease cases. Table S2. Clinicopathological features of female argyrophilic grain disease cases.
|
Systematic analysis of levels of evidence supporting American Academy of Ophthalmology Preferred Practice Pattern guidelines, 2012–2021 | 6f575744-d323-4f7c-a1d0-c58c291c716e | 10067168 | Ophthalmology[mh] | Over the past four decades, there has been an increased emphasis on evidence-based medicine across all medical specialties, particularly in the development of clinical practice guidelines. However, prior analyses of clinical guidelines in cardiology suggest that randomized controlled trials , typically considered the highest level of evidence, are not cited in a majority of clinical guidelines developed by key professional societies . Additionally, the proportion of recommendations citing the highest level of evidence has not increased significantly over time . Prior studies in ophthalmology have evaluated the types of evidence published in ophthalmology journals ; however, there is limited literature analyzing the levels of evidence present in ophthalmology clinical guidelines. The American Academy of Ophthalmology Preferred Practice Patterns guidelines designate recommended diagnostic and treatment approaches for various ophthalmic conditions and are typically revised every five years . The purpose of this systematic analysis of the American Academy of Ophthalmology Preferred Practice Pattern guidelines was to understand the evidence behind current guidelines, assess changes over time in the levels of evidence used to generate recommendations, and compare levels of evidence utilized across guidelines from different ophthalmology subspecialties. Review of Guidelines Current American Academy of Ophthalmology (AAO) Preferred Practice Pattern (PPP) guidelines were identified as those posted on the AAO website ( https://www.aao.org/guidelines-browse ) as of March 20, 2022. Only full-text PPP guidelines documents were included. Summary Benchmarks and PPP Clinical Questions were not included. Since PPP guidelines are typically valid for five years, the immediate predecessors of current guidelines were identified to assess changes over time. Prior guidelines were either identified on the AAO website ( https://www.aaojournal.org/content/preferred-practice-pattern ) or requested from the AAO if they were issued prior to 2015 . No human subjects, human-derived materials, or human medical records were involved in this study to necessitate review by an Institutional Review Board. The guidelines report levels of evidence (LOE) based on the Scottish Intercollegiate Guidelines Network (SIGN) scale (Table ) . The guidelines also report quality of evidence and strengths of recommendation defined by the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) scale (Table ) . Current PPPs report SIGN and GRADE ratings throughout the PPP main texts. Prior PPPs report the ratings in a centralized appendix. For each guideline, recommendations with reported LOE were abstracted by one of two reviewers (either A.S. or J.B.L.). Any statement with a reported LOE was considered a recommendation. Further details of LOE reporting are presented in Table . For each recommendation with LOE reported, we recorded the LOE, the quality of evidence, and the recommendation strength. Each recommendation was categorized as a recommendation for diagnosis, management, or both. We also recorded the subspecialty associated with each PPP guideline following the subspecialty categories listed on the AAO PPP website (cataract/anterior segment, comprehensive ophthalmology, cornea/external disease, glaucoma, neuro-ophthalmology, ocular pathology/oncology, oculoplastics/orbit, pediatric ophthalmology/strabismus, refractive management/intervention, retina/vitreous, and uveitis). An additional comprehensive review of all sentences in the Care Process section, where recommendations are typically located, of the 2021 edition of the Cataract in the Adult Eye PPP was performed . A single reviewer (A.S.) abstracted all sentences from the document. Two reviewers (A.S. and J.B.L) independently determined whether each sentence constituted a recommendation statement—a sentence was considered a recommendation statement if it addressed how patients should be diagnosed or managed clinically. For each recommendation statement, the two reviewers examined all references in the document associated with the statement to determine their LOE based on the SIGN scale and recorded the highest LOE (LOE I > LOE II > LOE III). When disagreement existed, the two reviewers had a discussion to reach a consensus decision. Further, a third reviewer (A.N.K.), a board-certified anterior segment ophthalmologist and cataract surgeon, validated all the findings. All current PPP guidelines were also reviewed to determine whether or not cost-effectiveness or cost/value factors were explicitly mentioned as part of the justification for each recommendation. Additional review of each guideline was performed to determine the presence of cost/value statements, broadly defined as any statement in which cost or value was mentioned, and whether such statements were used to: 1) report a gap in cost/value evidence; 2) highlight economic impact of disease or care; and 3) advocate for cost/value-related issues, consistent with a framework previously described in the cardiology literature . Data analysis We calculated the total number of recommendations with LOE reported for each current and prior PPP and calculated the change in number of recommendations over time. The median number of recommendations per guideline was determined. The number of recommendations with reported LOE were also summarized by subspecialty and care process category (diagnosis, management, or both). Additionally, the numbers and proportions of recommendations classified as LOE I, II, and III among all current and prior PPPs were determined. To further assess differences across subspecialties, we compared the number and proportion of LOE I, II, and III recommendations in current PPPs with those in prior PPPs by subspecialty. We also reported the quality of evidence according to the GRADE scale, stratified by LOE. For the current Cataract in the Adult Eye PPP, agreement between the two reviewers for whether a sentence constituted a recommendation statement was measured by percent agreement and kappa statistics. If references were provided in the document, the numbers and proportions of validated recommendation statements that were classified as LOE I, II, and III were determined. If recommendation statements also had LOE explicitly reported in the document, we compared the reported LOE with the study team-determined LOE. To evaluate the role of cost/value in PPPs, the proportion of current PPPs that had any recommendation supported by cost-effectiveness or cost/value considerations was determined. Further analysis was done to assess the proportions of current PPPs that contain statements addressing each of the specific areas related to cost/value considerations. Current American Academy of Ophthalmology (AAO) Preferred Practice Pattern (PPP) guidelines were identified as those posted on the AAO website ( https://www.aao.org/guidelines-browse ) as of March 20, 2022. Only full-text PPP guidelines documents were included. Summary Benchmarks and PPP Clinical Questions were not included. Since PPP guidelines are typically valid for five years, the immediate predecessors of current guidelines were identified to assess changes over time. Prior guidelines were either identified on the AAO website ( https://www.aaojournal.org/content/preferred-practice-pattern ) or requested from the AAO if they were issued prior to 2015 . No human subjects, human-derived materials, or human medical records were involved in this study to necessitate review by an Institutional Review Board. The guidelines report levels of evidence (LOE) based on the Scottish Intercollegiate Guidelines Network (SIGN) scale (Table ) . The guidelines also report quality of evidence and strengths of recommendation defined by the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) scale (Table ) . Current PPPs report SIGN and GRADE ratings throughout the PPP main texts. Prior PPPs report the ratings in a centralized appendix. For each guideline, recommendations with reported LOE were abstracted by one of two reviewers (either A.S. or J.B.L.). Any statement with a reported LOE was considered a recommendation. Further details of LOE reporting are presented in Table . For each recommendation with LOE reported, we recorded the LOE, the quality of evidence, and the recommendation strength. Each recommendation was categorized as a recommendation for diagnosis, management, or both. We also recorded the subspecialty associated with each PPP guideline following the subspecialty categories listed on the AAO PPP website (cataract/anterior segment, comprehensive ophthalmology, cornea/external disease, glaucoma, neuro-ophthalmology, ocular pathology/oncology, oculoplastics/orbit, pediatric ophthalmology/strabismus, refractive management/intervention, retina/vitreous, and uveitis). An additional comprehensive review of all sentences in the Care Process section, where recommendations are typically located, of the 2021 edition of the Cataract in the Adult Eye PPP was performed . A single reviewer (A.S.) abstracted all sentences from the document. Two reviewers (A.S. and J.B.L) independently determined whether each sentence constituted a recommendation statement—a sentence was considered a recommendation statement if it addressed how patients should be diagnosed or managed clinically. For each recommendation statement, the two reviewers examined all references in the document associated with the statement to determine their LOE based on the SIGN scale and recorded the highest LOE (LOE I > LOE II > LOE III). When disagreement existed, the two reviewers had a discussion to reach a consensus decision. Further, a third reviewer (A.N.K.), a board-certified anterior segment ophthalmologist and cataract surgeon, validated all the findings. All current PPP guidelines were also reviewed to determine whether or not cost-effectiveness or cost/value factors were explicitly mentioned as part of the justification for each recommendation. Additional review of each guideline was performed to determine the presence of cost/value statements, broadly defined as any statement in which cost or value was mentioned, and whether such statements were used to: 1) report a gap in cost/value evidence; 2) highlight economic impact of disease or care; and 3) advocate for cost/value-related issues, consistent with a framework previously described in the cardiology literature . We calculated the total number of recommendations with LOE reported for each current and prior PPP and calculated the change in number of recommendations over time. The median number of recommendations per guideline was determined. The number of recommendations with reported LOE were also summarized by subspecialty and care process category (diagnosis, management, or both). Additionally, the numbers and proportions of recommendations classified as LOE I, II, and III among all current and prior PPPs were determined. To further assess differences across subspecialties, we compared the number and proportion of LOE I, II, and III recommendations in current PPPs with those in prior PPPs by subspecialty. We also reported the quality of evidence according to the GRADE scale, stratified by LOE. For the current Cataract in the Adult Eye PPP, agreement between the two reviewers for whether a sentence constituted a recommendation statement was measured by percent agreement and kappa statistics. If references were provided in the document, the numbers and proportions of validated recommendation statements that were classified as LOE I, II, and III were determined. If recommendation statements also had LOE explicitly reported in the document, we compared the reported LOE with the study team-determined LOE. To evaluate the role of cost/value in PPPs, the proportion of current PPPs that had any recommendation supported by cost-effectiveness or cost/value considerations was determined. Further analysis was done to assess the proportions of current PPPs that contain statements addressing each of the specific areas related to cost/value considerations. Current PPP guidelines Overall, LOE from 24 current PPP guidelines published between 2017 and 2021 were abstracted. Across the 24 guidelines, the LOE (SIGN) was provided for 94 recommendations. The median number of recommendations with LOE per guideline was 1.5 (interquartile range [IQR]: 1.0—5.0). 83 (88%) recommendations had LOE I, 2 (2%) had LOE II, and 9 (10%) had LOE III. All LOE II and III recommendations were in the 2017 Refractive Errors and Refractive Surgery PPP. Among the 94 recommendations with LOE, the vast majority (98%) were for management. The remaining 2% were for diagnosis. The number of guidelines per subspecialty area ranged from 0 to 7 (Table ). The guidelines also reported recommendation strengths (GRADE) for 104 recommendations. 86 (83%) recommendations were strong recommendations, and 18 (17%) were discretionary recommendations. These recommendations included the 94 recommendations with reported LOE, of which 78 (83%) were strong recommendations, and 16 (17%) were discretionary recommendations. Importantly, aside from indication of recommendation strengths, the current guidelines did not clearly designate statements as recommendations and thus differentiate them from background or evidence synthesis information in the documents. The 2021 Cataract in the Adult Eye PPP was the guideline with the greatest number of recommendations with reported LOE ( n = 18). In our systematic review of this guideline, we identified 510 statements that could be considered recommendation statements. The two reviewers both identified 386 (76%) of these recommendation statements independently. Their overall percent agreement was 82%, and the kappa statistic was 0.64. Ninety-five additional statements were identified as recommendation statements by consensus after initially being included by only one reviewer. Twenty-nine statements were additionally considered recommendations by the third reviewer. We found that 267 (52%) recommendation statements did not have any reference associated with them. For the recommendations that did have references, we reviewed the references and categorized 92 (18%) recommendations as supported by LOE I, 95 (19%) as supported by LOE II, and 56 (11%) recommendations as supported by LOE III. For the 18 statements that had reported LOE, our categorization had 100% agreement with the reported LOE. Changes from prior to current guidelines Twenty-three current PPP guidelines had a prior edition (published between 2012 and 2016) available for comparison. The current Adult Strabismus PPP was published for the first time in 2019. Across the 23 prior PPPs, LOE were provided for 1254 recommendations, compared with 94 (93% decrease) in current editions. Table shows the changes in the number of recommendations with LOE by subspeciality and topic. Overall, the number of recommendations with LOE has decreased for a majority of specialties and topics. Among the 23 PPPs with both prior and current editions available, the number of recommendations with LOE I decreased from 114 to 83, the number of recommendations with LOE II decreased from 147 to 2, and the number of recommendations with LOE III decreased from 993 to 9. The proportion of LOE I recommendations rose from 9 to 88%, driven by a disproportionate decrease in reporting of evidence lower than LOE I. The median number of LOE I recommendations per PPP was 2 (IQR: 0–7), compared with a median of 1.5 LOE I recommendations (IQR: 1.0—5.0) per PPP in current guidelines. A subgroup analysis by subspecialty revealed similar findings (Fig. ). In addition to LOE based on the SIGN scale (LOE I, II and III), both current and prior PPP guidelines reported quality of evidence based on the GRADE scale (good quality, moderate quality, and insufficient quality). Figure shows the proportions of recommendations by level (SIGN) and quality (GRADE) of evidence in current and prior PPP guidelines. Among all recommendations with quality of evidence ratings, current guidelines rated the evidence for 61 (64.9%) recommendations as good quality, 22 (23.4%) as moderate quality, and 11 (11.7%) as insufficient quality. By contrast, prior guidelines rated 761 (57.2%) recommendations as good quality, 109 (8.2%) as moderate quality, and 461 (34.6%) as insufficient quality. An analysis of all recommendations with insufficient quality evidence in the current guidelines showed that though these recommendations had level I evidence ( n = 11), the evidence base either had a high risk of bias (i.e., LOE I- based on SIGN) or was not rated for risk of bias. By contrast, in prior guidelines, only 3 recommendations with insufficient quality evidence had level I evidence. The role of cost/value in PPP guidelines Among the 24 current PPP guidelines, 21 (88%) guidelines contained cost/value statements. A majority (75%) used cost/value statements to highlight the economic impact of disease or care, and 58% used cost/value statements to report gaps in cost/value evidence. None of them used cost/value considerations to justify specific recommendations or advocated for cost/value-related issues (Fig. ). Overall, LOE from 24 current PPP guidelines published between 2017 and 2021 were abstracted. Across the 24 guidelines, the LOE (SIGN) was provided for 94 recommendations. The median number of recommendations with LOE per guideline was 1.5 (interquartile range [IQR]: 1.0—5.0). 83 (88%) recommendations had LOE I, 2 (2%) had LOE II, and 9 (10%) had LOE III. All LOE II and III recommendations were in the 2017 Refractive Errors and Refractive Surgery PPP. Among the 94 recommendations with LOE, the vast majority (98%) were for management. The remaining 2% were for diagnosis. The number of guidelines per subspecialty area ranged from 0 to 7 (Table ). The guidelines also reported recommendation strengths (GRADE) for 104 recommendations. 86 (83%) recommendations were strong recommendations, and 18 (17%) were discretionary recommendations. These recommendations included the 94 recommendations with reported LOE, of which 78 (83%) were strong recommendations, and 16 (17%) were discretionary recommendations. Importantly, aside from indication of recommendation strengths, the current guidelines did not clearly designate statements as recommendations and thus differentiate them from background or evidence synthesis information in the documents. The 2021 Cataract in the Adult Eye PPP was the guideline with the greatest number of recommendations with reported LOE ( n = 18). In our systematic review of this guideline, we identified 510 statements that could be considered recommendation statements. The two reviewers both identified 386 (76%) of these recommendation statements independently. Their overall percent agreement was 82%, and the kappa statistic was 0.64. Ninety-five additional statements were identified as recommendation statements by consensus after initially being included by only one reviewer. Twenty-nine statements were additionally considered recommendations by the third reviewer. We found that 267 (52%) recommendation statements did not have any reference associated with them. For the recommendations that did have references, we reviewed the references and categorized 92 (18%) recommendations as supported by LOE I, 95 (19%) as supported by LOE II, and 56 (11%) recommendations as supported by LOE III. For the 18 statements that had reported LOE, our categorization had 100% agreement with the reported LOE. Twenty-three current PPP guidelines had a prior edition (published between 2012 and 2016) available for comparison. The current Adult Strabismus PPP was published for the first time in 2019. Across the 23 prior PPPs, LOE were provided for 1254 recommendations, compared with 94 (93% decrease) in current editions. Table shows the changes in the number of recommendations with LOE by subspeciality and topic. Overall, the number of recommendations with LOE has decreased for a majority of specialties and topics. Among the 23 PPPs with both prior and current editions available, the number of recommendations with LOE I decreased from 114 to 83, the number of recommendations with LOE II decreased from 147 to 2, and the number of recommendations with LOE III decreased from 993 to 9. The proportion of LOE I recommendations rose from 9 to 88%, driven by a disproportionate decrease in reporting of evidence lower than LOE I. The median number of LOE I recommendations per PPP was 2 (IQR: 0–7), compared with a median of 1.5 LOE I recommendations (IQR: 1.0—5.0) per PPP in current guidelines. A subgroup analysis by subspecialty revealed similar findings (Fig. ). In addition to LOE based on the SIGN scale (LOE I, II and III), both current and prior PPP guidelines reported quality of evidence based on the GRADE scale (good quality, moderate quality, and insufficient quality). Figure shows the proportions of recommendations by level (SIGN) and quality (GRADE) of evidence in current and prior PPP guidelines. Among all recommendations with quality of evidence ratings, current guidelines rated the evidence for 61 (64.9%) recommendations as good quality, 22 (23.4%) as moderate quality, and 11 (11.7%) as insufficient quality. By contrast, prior guidelines rated 761 (57.2%) recommendations as good quality, 109 (8.2%) as moderate quality, and 461 (34.6%) as insufficient quality. An analysis of all recommendations with insufficient quality evidence in the current guidelines showed that though these recommendations had level I evidence ( n = 11), the evidence base either had a high risk of bias (i.e., LOE I- based on SIGN) or was not rated for risk of bias. By contrast, in prior guidelines, only 3 recommendations with insufficient quality evidence had level I evidence. Among the 24 current PPP guidelines, 21 (88%) guidelines contained cost/value statements. A majority (75%) used cost/value statements to highlight the economic impact of disease or care, and 58% used cost/value statements to report gaps in cost/value evidence. None of them used cost/value considerations to justify specific recommendations or advocated for cost/value-related issues (Fig. ). This systematic analysis of LOE supporting AAO PPP guidelines evaluated the proportion of guidelines with a LOE listed, as well as changes in reporting patterns over time and across specialties. Overall, this study demonstrated that while current guidelines report LOE for substantially fewer recommendations, a much higher proportion of recommendations are supported by evidence from randomized controlled trials. Eighty-eight percent of current recommendations with reported LOE had LOE I. Subgroup analysis by subspecialty showed similar trends. These results suggest that while current AAO PPPs emphasize evidence from randomized controlled trials, LOE from other types of studies may not be formally rated or reported. Although analyses of guidelines have been performed in other specialties, such as cardiology , prior investigation of evidence supporting guidelines in ophthalmology is limited. A 2015 study examined the LOE of papers published in four major ophthalmology journals and concluded that lower LOE publications would continue to play a large role in guiding the field of ophthalmology . At first glance, the findings from our study do not appear to suggest this same trend among the reported LOE supporting AAO PPPs, as the vast majority of recommendations with reported LOE had the highest level of evidence. However, our comprehensive review of the 2021 Cataract in the Adult Eye PPP and independent rating of LOE of the citations show that 30% of recommendations rely on level II and III evidence (vs. 18% level I), but the LOE was simply not reported in the PPP. The majority (52%) of recommendations did not have any citations, consistent with a prior study investigating the relationship between findings from systematic reviews and the 2015 AAO PPP on interventions for age-related macular degeneration . The study found that only 1 out of 35 treatment recommendations in the PPP cited a reliable intervention systematic review . Our study complements the existing literature, highlighting that there may be areas to include additional supporting evidence in AAO PPPs. In evidence-based medicine, randomized controlled trials (RCTs) and systematic reviews/meta-analyses synthesizing their results are the pinnacle of evidence as randomization reduces bias and allows for investigation of causal relationships. A study conducted in 2019 found that only 2% of all publications in the field of ophthalmology were RCTs . In our study, while the proportion of LOE I (ie, meta-analysis, systematic reviews of RCTs, or RCTs) recommendations has increased from prior PPPs to current PPPs, this increase is primarily driven by a disproportionate underreporting of lower-level evidence. In fact, the number of LOE I recommendations has not increased. On the one hand, this trend suggests guideline authors may have attempted to highlight LOE I recommendations in the current PPPs. On the other hand, fully reporting both level I and lower-level evidence could help to expand the evidence base highlighted in ophthalmology guidelines. The Institute of Medicine’s landmark reports on clinical practice guidelines were the impetus for the initial development of many guidelines in effect today . In 2011, the Institute of Medicine recommended standards for developing trustworthy clinical practice guidelines . The standards state that for each recommendation in a guideline, “a rating of the level of confidence in the evidence underpinning the recommendation” should be provided . Our results suggest that substantial underreporting of LOE may exist in current PPPs, as the number of recommendations with reported LOE fell from 1254 in prior PPPs to 94 in current PPPs. These results suggest that there is significant opportunity to include level II and III evidence, which, despite risk of bias, is nonetheless often critically important data . Use of data sources such as insurance claims or multi-institutional registries can provide information about real-world clinical practice that cannot be generated by randomized clinical trials . Furthermore, there are clinical questions for which a randomized trial is infeasible, such as for rare conditions or for procedures where shams are not possible, and in these circumstances, lower levels of evidence ought to be weighted more heavily . Since many ophthalmologic diseases have a low incidence and a heavy reliance on surgical management in certain subspecialties, performing randomized controlled trials may be especially challenging . By acknowledging this and including varied levels of evidence in ophthalmology PPPs, authors may be able to more easily adopt the Institute of Medicine’s recommendation about LOE reporting in clinical practice guidelines. The same report from the Institute of Medicine also proposed that “recommendations should be articulated in a standardized form detailing precisely what the recommendation action is, and under what circumstances it should be performed” . Our findings demonstrated that aside from indication of recommendation strengths ( n = 104 recommendations across 24 PPPs), the current guidelines do not articulate recommendations in a standardized form. Without such standardization, our comprehensive review of the 2021 Cataract in the Adult Eye PPP identified 510 statements that addressed how patients should be diagnosed or managed clinically and thus could be considered recommendations. While the agreement between our two reviewers was good (k statistic = 0.64), this result suggests that interpretations of potential recommendation statements in the PPP can be variable. Standardized articulation of recommendations would help clinicians clearly identify recommended actions for clinical practice. For example, American Heart Association guidelines list all recommendations in visually distinctive boxes, which stand out from the surrounding text and include levels of evidence . Clearly articulating recommendations could also facilitate the creation and assessment of programs to improve the quality of care. As health care usage and expenditure continue to rise in the United States, value-based care has become an increasingly important concept . A recent systematic review found that between 75.7 and 101.2 billion was spent on low-value care in the United States . Clinical practice guidelines play an important role in shaping practice patterns and thus may be well-suited to promote high-value care. In this area, prior work in cardiology has evaluated cost and value considerations in contemporary heart failure clinical guidelines . The study concluded that although most contemporary heart failure guidelines contained cost/value statements, they were rarely used to support clinical guidance recommendations. In the ophthalmology guidelines, a majority (88%) of PPPs included cost/value statements. In particular, the high economic impact of disease or care was frequently highlighted (75% of PPPs). However, cost/value considerations have yet to be incorporated into the development of specific recommendations, representing an avenue for future work in ophthalmology guideline development. More than half of the PPPs also reported gaps in cost/value evidence—ongoing efforts in the field such as the IRIS® Registry, which includes performance metrics, may facilitate real-world evidence generation in this area and help to provide needed data for guideline development . The strengths of this study include analysis of all PPPs spanning a 10-year period, including all contemporary PPPs and their immediate predecessors. This thorough analysis allowed us to assess evolutions of PPPs over time and trends in all the specialties and topics that PPPs cover. Furthermore, we reported levels of evidence exactly as described in the guidelines. Additional strengths of this study included independent two-party grading, with validation by a board-certified anterior segment specialist, of levels of evidence for our review of the 2021 Cataract in the Adult Eye PPP. The limitations include the potential underreporting of LOE in current PPPs, which prohibits us from drawing conclusions about all evidence supporting PPPs. This is partially addressed by our comprehensive review of the 2021 Cataract in the Adult Eye PPP, including its references, which suggests substantial underreporting of LOE across all LOE and disproportionate underreporting of lower-level evidence. In conclusion, we performed a systematic analysis of reported LOE supporting AAO PPP guidelines. Compared with prior PPPs, current PPPs emphasize evidence from randomized controlled trials. While underreporting of LOE across all LOE exists, there appears to be a disproportionate underreporting of lower-level evidence. Future guideline development may consider clearly defining recommendations, explicitly reporting LOE associated with each recommendation, and integrating cost/value considerations in recommendations. |
Experiences of older people, healthcare providers and caregivers on implementing person-centered care for community-dwelling older people: a systematic review and qualitative meta-synthesis | 19b0542d-e317-4d54-b7f4-3a61029d1923 | 10067217 | Patient-Centered Care[mh] | The aged population is rapidly growing worldwide. Approximately 16% of the global population will be over 65 by 2050 . Home and community-based services (HCBSs) embody the core care concept of "aging in place" and combine the humanistic concept using the cost-effectiveness principle . The community-based and person-centered care (PCC) model benefits aged care. PCC is considered a proxy for quality care and has been demonstrated to improve health outcomes and satisfaction of older people . Older people often have multiple care needs with complex health conditions, making them an ideal group to benefit from PCC . Therefore, identifying the enablers and barriers to implementing PCC for community-dwelling older people is crucial. The term 'PCC' is seen as an umbrella concept that covers the same meaning in this study, such as ‘individualized-centered care’, ‘client-centered care’, ‘resident-focused care’, etc. According to the World Health Organization (WHO), the PCC implies that the individual is viewed as a whole with many needs and goals in caring practice . According to the American Geriatrics Society panel, PCC requires individuals to be motivated to express their values and preferences . Person-centered practice framework of McCormack pointed out “best practices” for PCC included four dimensions: prerequisites (focus on attributes of the care worker); care environment; person-centered processes; outcomes (e.g., satisfaction; feeling of well-being) . Although there is general consensus on the elements of the person-centered, it is necessary to translate the PCC framework into practice . Person-centered care has attracted immense attention in recent years. The WHO has called for person-centered policies to address the complex challenges that individuals face in their communities . A previous systematic review explored the content and essential components of implementing PCC for non-hospitalized older people (65 +) , such as treating patients as a whole, shared decision-making, teamwork, and building a PCC foundation. The implementation is also hindered by some factors, such as insufficient educational or institutional help with PCC assessment and care skills , resource constraints, less positive attitudes of community doctors , limited professional autonomy of HCPs, imbalanced interpersonal contact with older people , unprofessional personal qualities of HCPs , and the challenge of older people participating in PCC processes . Enablers of PCC implementation include leadership, professional training, organizational support, and appropriate incentives . Furthermore, recent research has shown that family caregivers can benefit from education and support while implementing PCC for people with dementia, which can help those people to be independent . To personalize care, HCPs need to tailor the care plans to meet the needs and preferences of recipients . Due to the complexity of community settings and PCC interventions (involving individuals, organizations, and society), it is difficult to draw accurate conclusions about the influencing factors of implementing PCC in the community based on a single research article. Previous reviews focused on the concept, elements, key intervention categories, effects of PCC, and whether or not PCC has a relational ethics perspective [ , – ]. However, there is no systematic elaboration on the factors influencing the successful implementation of PCC. Qualitative research offers many advantages for an in-depth understanding of the PCC experiences of different stakeholders in the community. Besides, capturing different perspectives of older people, HCPs, and caregivers is more productive for mutual understanding and interactions . The capacity, opportunity, and motivation-behavior model (COM-B model) have been widely used in the medical field to explain and guide various behavioral interventions that can comprehensively and systematically understand the influencing factors in the behavior change process . This study aimed to explore stakeholders' experiences regarding the implementation of PCC in the community and to identify the enablers and barriers based on the COM-B model.
Research design This systematic review with qualitative meta-synthesis was performed by the guidelines of the Joanna Briggs Institute (JBI) . A meta-aggregative approach to the synthesis of qualitative evidence was used. We used the Procedure PROSPERO ( https://www.crd.york.ac.uk/prospero/ ) to identify published or ongoing projects relevant to the topic. This review was registered with PROSPERO (CRD42022314924). In addition, the reporting of this review was guided by the Enhancing Transparency in Reporting the Synthesis of Qualitative Research (ENTREQ) Statement . Search strategy The search strategy aimed to identify peer-reviewed published studies. We conducted a broad bibliographic search using the CINAHL (EBSCOhost), PubMed (OvidSP), Embase (Ovid), Cochrane Database, and PsycINFO (Ovid) from the earliest available date till February 2023. A three-step search strategy was used to locate the literature in this systematic review. First, two researchers undertook an initial limited search of CINAHL and PubMed, followed by a structured analysis of text words contained in titles, and of index terms used to describe the article. With this initial step, we would ensure that our search strategy is sufficiently sensitive, precise, and specific regarding our research objectives as well as the population, concepts, and context of interest. Then, we undertook a structured search using all identified keywords and index terms for a second extensive search. Finally, the reference lists of all included articles were searched manually for additional sources. The initial search included the key search terms of PCC, person-centered care, older people, and community care. Only English language studies were included. Additional file lists the full search strategy. Eligibility criteria and study selection To ensure the correct identification and selection of relevant studies, we developed inclusion/exclusion criteria to aid the selection of relevant papers. The inclusion criteria of the systematic review are as follows: a). all included studies were qualitative or mixed methods studies; b). the context was community home caring organizations providing professional home healthcare visits. Hospitals or nursing homes were excluded; c). the focus was on stakeholders’ experiences participating in PCC intervention programs, including but not limited to older people over the age of 60, HCPs, and family caregivers. These HCPs may have been working in any sector in community health services organizations. Studies only using quantitative methods to analyze data were excluded. Quality assessment Studies were assessed for quality using the JBI qualitative research appraisal tool . Each item could be answered with "yes" (1 point), "no" (0 point), "not applicable" (0 point), or "unclear" (0 point). If the criteria are fully met, the possibility of bias is minimum, which is grade A; if partially meeting the assessment criteria, the possibility of bias is moderate, which is grade B; those which did not meet the assessment criteria at all and had a high possibility of bias is classified as Grade C. To avoid possible biases, lower-quality studies were excluded, similar to previous studies . Original studies had to score more than five assessment criteria to be selected in this review synthesis. Any disagreements were resolved after a discussion between the two reviewers (LLL and GCY) or following consultation with a third reviewer (FMJ) on the team. Data extraction and synthesis All included papers in the systematic review were analyzed independently by two authors. Relevant details for each study were extracted using a standardized data extraction tool from Joanna Briggs Institute Qualitative Assessment and Review Instrument (JBI-QARI). The findings are extracted verbatim from the studies, as are the illustrative quotes usually recorded directly from the participants. Results were graded based on the reader's level of confidence in the findings based on the published data. Two reviewers (LLL and GCY) critically appraised each article independently and attributed a level of credibility to each one. The JBI-QARI qualitative criteria have three levels of credibility: unequivocal (U)—refers to findings that are a matter of fact, beyond a reasonable doubt; credible (C)—refers to findings that are plausible interpretations of the primary data within the theoretical framework; and unsupported (Un)—relates to findings that are unsupported by the data. There were no disagreements between the reviewers in this process. Based on the similarity of meanings, those findings were combined to form different categories; these categories were then subjected to a meta-synthesis to generate comprehensive synthesized findings by meta-aggregation . The first author led a systematic process of data organization and synthesis. After consulting the primary literature, a group discussion to reach an agreement was held in case of a disagreement. We assessed the final synthesized findings based on the JBI approach for rating the confidence of synthesized qualitative results (ConQual) to determine the confidence level . The summary of findings table was created using the following major elements: population, phenomena of interest, context, synthesized finding, type of research, and the final ConQual scores.
This systematic review with qualitative meta-synthesis was performed by the guidelines of the Joanna Briggs Institute (JBI) . A meta-aggregative approach to the synthesis of qualitative evidence was used. We used the Procedure PROSPERO ( https://www.crd.york.ac.uk/prospero/ ) to identify published or ongoing projects relevant to the topic. This review was registered with PROSPERO (CRD42022314924). In addition, the reporting of this review was guided by the Enhancing Transparency in Reporting the Synthesis of Qualitative Research (ENTREQ) Statement .
The search strategy aimed to identify peer-reviewed published studies. We conducted a broad bibliographic search using the CINAHL (EBSCOhost), PubMed (OvidSP), Embase (Ovid), Cochrane Database, and PsycINFO (Ovid) from the earliest available date till February 2023. A three-step search strategy was used to locate the literature in this systematic review. First, two researchers undertook an initial limited search of CINAHL and PubMed, followed by a structured analysis of text words contained in titles, and of index terms used to describe the article. With this initial step, we would ensure that our search strategy is sufficiently sensitive, precise, and specific regarding our research objectives as well as the population, concepts, and context of interest. Then, we undertook a structured search using all identified keywords and index terms for a second extensive search. Finally, the reference lists of all included articles were searched manually for additional sources. The initial search included the key search terms of PCC, person-centered care, older people, and community care. Only English language studies were included. Additional file lists the full search strategy.
To ensure the correct identification and selection of relevant studies, we developed inclusion/exclusion criteria to aid the selection of relevant papers. The inclusion criteria of the systematic review are as follows: a). all included studies were qualitative or mixed methods studies; b). the context was community home caring organizations providing professional home healthcare visits. Hospitals or nursing homes were excluded; c). the focus was on stakeholders’ experiences participating in PCC intervention programs, including but not limited to older people over the age of 60, HCPs, and family caregivers. These HCPs may have been working in any sector in community health services organizations. Studies only using quantitative methods to analyze data were excluded.
Studies were assessed for quality using the JBI qualitative research appraisal tool . Each item could be answered with "yes" (1 point), "no" (0 point), "not applicable" (0 point), or "unclear" (0 point). If the criteria are fully met, the possibility of bias is minimum, which is grade A; if partially meeting the assessment criteria, the possibility of bias is moderate, which is grade B; those which did not meet the assessment criteria at all and had a high possibility of bias is classified as Grade C. To avoid possible biases, lower-quality studies were excluded, similar to previous studies . Original studies had to score more than five assessment criteria to be selected in this review synthesis. Any disagreements were resolved after a discussion between the two reviewers (LLL and GCY) or following consultation with a third reviewer (FMJ) on the team.
All included papers in the systematic review were analyzed independently by two authors. Relevant details for each study were extracted using a standardized data extraction tool from Joanna Briggs Institute Qualitative Assessment and Review Instrument (JBI-QARI). The findings are extracted verbatim from the studies, as are the illustrative quotes usually recorded directly from the participants. Results were graded based on the reader's level of confidence in the findings based on the published data. Two reviewers (LLL and GCY) critically appraised each article independently and attributed a level of credibility to each one. The JBI-QARI qualitative criteria have three levels of credibility: unequivocal (U)—refers to findings that are a matter of fact, beyond a reasonable doubt; credible (C)—refers to findings that are plausible interpretations of the primary data within the theoretical framework; and unsupported (Un)—relates to findings that are unsupported by the data. There were no disagreements between the reviewers in this process. Based on the similarity of meanings, those findings were combined to form different categories; these categories were then subjected to a meta-synthesis to generate comprehensive synthesized findings by meta-aggregation . The first author led a systematic process of data organization and synthesis. After consulting the primary literature, a group discussion to reach an agreement was held in case of a disagreement. We assessed the final synthesized findings based on the JBI approach for rating the confidence of synthesized qualitative results (ConQual) to determine the confidence level . The summary of findings table was created using the following major elements: population, phenomena of interest, context, synthesized finding, type of research, and the final ConQual scores.
Search results We found 4,944 papers after searching relevant databases and grey literature. EndNote X20 software was used to import all search results. After removing 323 duplicates, a total of 4,621 articles were identified. Following the inclusion and exclusion criteria, two trained reviewers independently screened the titles and abstracts, and 4,464 papers were deleted. After a full-text review, 144 of the 157 articles that met the inclusion criteria were eliminated, and 13 met the eligibility criteria. One study was excluded due to quality issues . Figure shows the results of the search and screening strategies. Articles were independently screened by two researchers, and there were no disagreements. Characteristics of the studies Twelve research studies met the eligibility criteria for inclusion in the meta-synthesis, all of which were qualitative studies. The methods applied in the twelve articles were primarily one-to-one interviews and focus group interviews. The articles are spread over time (from 2006 to 2023), mostly concentrated in 2018–2022. Among the twelve papers, four were from Canada, two were from America, two were from Australia, and four were from Ireland, the United Kingdom, the Netherlands, and England, respectively. Table depicts further detailed results of the included studies. Methodological quality All included studies scored 7–9. The quality appraisal revealed that twelve included articles were rated as B. One of the original studies received only five scores in the quality assessment process due to a lack of adequate details and was therefore excluded from this review (Additional file ). Table presents the results of the quality assessment. Meta-synthesis of qualitative data We extracted a total of 122 findings from the 12 included studies: 106 unequivocal and 16 credible. Those findings were aggregated into 12 categories based on the similarity of meanings, which were then meta-aggregated into three synthesized findings. Figure depicts the final conceptual map. The conceptual map demonstrates that PCC behaviors can be affected in three domains based on the COM-B framework. The capacities of stakeholders and opportunities in implementing PCC can influence the motivation for implementing PCC and as result PCC behaviors. These three elements can also affect the PCC behavior separately. The results obtained from this study are listed in Additional file , whereas the results of the meta-synthesis process are shown in Additional file . In addition, Table shows the included study from which each theme is derived. Synthesized finding 1: Capacities of older people, HCPs and caregivers It is crucial to recognize that the capacities of older people, HCPs, and caregivers affect the implementation of PCC, including lack of person-centered knowledge and skills, negative attitudes toward shared decision-making, and lack of formal training. Lack of person-centered knowledge and skills among older people, HCPs and caregivers (barrier) Older people, HCPs, and caregivers lack professional knowledge and skills to implement effective PCC. They have an inadequate understanding of PCC and lack communication skills, and implement aged care through their experiences. Communication can be number one, no matter who you are looking after. (McKenzie and Brown, 2021, P.278). Negative attitudes toward shared decision-making among older people and HCPs (barrier) Shared decision-making is the process of older people and HCPs making health decisions together. Some participants stated the HCPs lacked emphasis on shared decision-making. Most HCPs made decisions instead of older people and did not involve older people in developing care plans. I had a surgery and the doctor that operated on me I met the day of the surgery. They didn't even give me an appointment to meet him or for me to be more informed of the surgery. After the surgery I never saw the surgeon again. (Manalili et al., 2021, P.10). Lack of formal training to enhance capabilities among HCPs (barrier) Meanwhile, some participants stated the community managers should carry out targeted training to improve professional competence among HCPs. …training increased their knowledge of Major Neurocognitive Disorders, responsive behaviors, and strategies to respond to clients’ unique needs. This information was consistent with the community documents regarding professional development through training and assessing the quality of care. (Zarshenas S et al., 2023, P.8). Synthesized finding 2: Opportunities in the implementation of PCC It is essential to note that opportunities play a significant role in implementing PCC programs. Factors hindering the implementation of PCC include a lack of coordination in resource allocation and time constraints. Furthermore, strengthening a multidisciplinary team facilitates the development of tailored and comprehensive care plans. Establishing a safe and friendly environment can also facilitate the implementation of PCC. Lack of coordination in resource allocation (barrier) Lack of coordination in resource allocation (human and fiscal) can impact the implementation of PCC and, in particular, the choice of intervention measures. Despite these various resources being available to nurse assistants, they noted that the comprehensive application of resources could be challenging in the short term since these skills are achieved through practice and experience. (Zarshenas S et al., 2023, P.8). Strengthening multidisciplinary teamwork (enabler) Reinforcing multidisciplinary teams and collaboration are an important part of implementing successful PCC program. Older people with diverse diseases often need support in multiple domains, which also facilitates continuity of care for them over time. One of the greatest barriers is the lack of understanding of person- centeredness. People think that the nurses are the only people to be person-centered but everybody from the maintenance man, cook, household staff and the team all have to be person-centered. (Doody, C et al., 2013, P.1117). Establishing a safe and friendly environment (enabler) Some HCPs stated that establishing a safe living environment is important. Furthermore, the person-centered approach that guided communication might create trust between older people, family caregivers, and HCPs. Significant time spent deliberately building trust and rapport makes sense for implementing PCC programs. It is important because it’s also a part of building a relationship of trust. Clients apparently like the social aspect, having a nice time. Well, I do think that this is an important component, but it’s certainly not my main reason for coming. (Uittenbroek, R. J et al., 2018, P.6). Time constraints (barrier) Time constraints are opposed to implementing successful PCC programs. This could be partly due to a lack of health staff and bureaucratic overload (lack of a whole service approach, staff turnover, sharp targets, and complex procedures). The situation resulted in high work pressure, overtime, and reduced quality of care. We’re only touching the tip of the iceberg in relation to person-centeredness. We do our utmost in choice, in documentation, in family involvement, but we would need ten times more staff to do what possibly could be done for each service user to fulfil their dreams, we do the best we can with person centeredness at the core. (Doody, C et al., 2013, P.1117). Synthesized finding 3: Motivation in the implementation of PCC Motivation in PCC programs includes reflexive and automatic motivation , which guides one to produce positive or negative emotions toward behavioral goals by increasing knowledge and experiences. Encouragement of self-reflection and regulation in practice leads to self-improvement and provide better care services for older people. Respecting the autonomy of older people and maintaining resilient and positive attitudes contribute to the engagement of all stakeholders in the process of PCC. Furthermore, the lack of clear reward and empowerment mechanisms can reduce staff motivation. Encouraging self-reflection and regulation (enabler) The HCPs can promote self-directed learning through critical thinking skills and engage in self-correction and reflection. Self-reflection is observing and evaluating self-perceptions, emotions, and behaviors . Self-regulation is a motivational mechanism by which an individual's cognitive development moves from a state of disequilibrium to equilibrium . This review refers to how HCPs can reinforce, maintain, or change their behavior based on available rewards or punishments. I have probably psychoanalyzed a lot of the experiences that I’ve been through to learn from my mistakes. (Narayan MC and Mallinson RK, 2022, P.5). Respecting the autonomy of older people (enabler) Participants reported that HCPs and family caregivers should consider proactive and compassionate care at the heart of their practice and respect the autonomy of older people. They also stated that older people should be treated with respect and dignity while maintaining their autonomy and equality. Actually, whenever I go to doctor, they call me by name. Once [they] call me by my name I feel close, attached to them. Otherwise, I'm going to feel bad. my relationship with my doctor is really good. (Manalili, K et al., 2021, P.8). Lack of clear reward and empowerment mechanisms (barrier) Establishing rewards and accountability mechanisms can help foster active HCPs engagement. Some participants stated that the role and task of case managers should be clarified. They should maintain close communication with older people, provide feedback on care plans, and keep an open mind. Doing the right thing is quality, right thing is a standard. So, if you are diagnosed with particular disease for a patient, then you have to do the right things, what you need to do, so quality, in my opinion he's doing the right things. (Manalili, K et al., 2021, P.10). Having a resilient and optimistic attitude (enabler) Being resilient and optimistic toward the caring management of older people is a vital step toward activation in PCC programs. The HCPs can overcome the barriers with a resilient attitude that embrace the positive aspects of implementing PCC. I make it work for me and for the patient. And I can make it work for the agency as well. I’m not afraid to think out of the box, change the way I do things. I am always thinking is there something that I could do differently to be more successful” (Narayan MC and Mallinson RK, 2022, P.6).
We found 4,944 papers after searching relevant databases and grey literature. EndNote X20 software was used to import all search results. After removing 323 duplicates, a total of 4,621 articles were identified. Following the inclusion and exclusion criteria, two trained reviewers independently screened the titles and abstracts, and 4,464 papers were deleted. After a full-text review, 144 of the 157 articles that met the inclusion criteria were eliminated, and 13 met the eligibility criteria. One study was excluded due to quality issues . Figure shows the results of the search and screening strategies. Articles were independently screened by two researchers, and there were no disagreements.
Twelve research studies met the eligibility criteria for inclusion in the meta-synthesis, all of which were qualitative studies. The methods applied in the twelve articles were primarily one-to-one interviews and focus group interviews. The articles are spread over time (from 2006 to 2023), mostly concentrated in 2018–2022. Among the twelve papers, four were from Canada, two were from America, two were from Australia, and four were from Ireland, the United Kingdom, the Netherlands, and England, respectively. Table depicts further detailed results of the included studies.
All included studies scored 7–9. The quality appraisal revealed that twelve included articles were rated as B. One of the original studies received only five scores in the quality assessment process due to a lack of adequate details and was therefore excluded from this review (Additional file ). Table presents the results of the quality assessment.
We extracted a total of 122 findings from the 12 included studies: 106 unequivocal and 16 credible. Those findings were aggregated into 12 categories based on the similarity of meanings, which were then meta-aggregated into three synthesized findings. Figure depicts the final conceptual map. The conceptual map demonstrates that PCC behaviors can be affected in three domains based on the COM-B framework. The capacities of stakeholders and opportunities in implementing PCC can influence the motivation for implementing PCC and as result PCC behaviors. These three elements can also affect the PCC behavior separately. The results obtained from this study are listed in Additional file , whereas the results of the meta-synthesis process are shown in Additional file . In addition, Table shows the included study from which each theme is derived.
It is crucial to recognize that the capacities of older people, HCPs, and caregivers affect the implementation of PCC, including lack of person-centered knowledge and skills, negative attitudes toward shared decision-making, and lack of formal training.
Older people, HCPs, and caregivers lack professional knowledge and skills to implement effective PCC. They have an inadequate understanding of PCC and lack communication skills, and implement aged care through their experiences. Communication can be number one, no matter who you are looking after. (McKenzie and Brown, 2021, P.278).
Shared decision-making is the process of older people and HCPs making health decisions together. Some participants stated the HCPs lacked emphasis on shared decision-making. Most HCPs made decisions instead of older people and did not involve older people in developing care plans. I had a surgery and the doctor that operated on me I met the day of the surgery. They didn't even give me an appointment to meet him or for me to be more informed of the surgery. After the surgery I never saw the surgeon again. (Manalili et al., 2021, P.10).
Meanwhile, some participants stated the community managers should carry out targeted training to improve professional competence among HCPs. …training increased their knowledge of Major Neurocognitive Disorders, responsive behaviors, and strategies to respond to clients’ unique needs. This information was consistent with the community documents regarding professional development through training and assessing the quality of care. (Zarshenas S et al., 2023, P.8).
It is essential to note that opportunities play a significant role in implementing PCC programs. Factors hindering the implementation of PCC include a lack of coordination in resource allocation and time constraints. Furthermore, strengthening a multidisciplinary team facilitates the development of tailored and comprehensive care plans. Establishing a safe and friendly environment can also facilitate the implementation of PCC.
Lack of coordination in resource allocation (human and fiscal) can impact the implementation of PCC and, in particular, the choice of intervention measures. Despite these various resources being available to nurse assistants, they noted that the comprehensive application of resources could be challenging in the short term since these skills are achieved through practice and experience. (Zarshenas S et al., 2023, P.8).
Reinforcing multidisciplinary teams and collaboration are an important part of implementing successful PCC program. Older people with diverse diseases often need support in multiple domains, which also facilitates continuity of care for them over time. One of the greatest barriers is the lack of understanding of person- centeredness. People think that the nurses are the only people to be person-centered but everybody from the maintenance man, cook, household staff and the team all have to be person-centered. (Doody, C et al., 2013, P.1117).
Some HCPs stated that establishing a safe living environment is important. Furthermore, the person-centered approach that guided communication might create trust between older people, family caregivers, and HCPs. Significant time spent deliberately building trust and rapport makes sense for implementing PCC programs. It is important because it’s also a part of building a relationship of trust. Clients apparently like the social aspect, having a nice time. Well, I do think that this is an important component, but it’s certainly not my main reason for coming. (Uittenbroek, R. J et al., 2018, P.6).
Time constraints are opposed to implementing successful PCC programs. This could be partly due to a lack of health staff and bureaucratic overload (lack of a whole service approach, staff turnover, sharp targets, and complex procedures). The situation resulted in high work pressure, overtime, and reduced quality of care. We’re only touching the tip of the iceberg in relation to person-centeredness. We do our utmost in choice, in documentation, in family involvement, but we would need ten times more staff to do what possibly could be done for each service user to fulfil their dreams, we do the best we can with person centeredness at the core. (Doody, C et al., 2013, P.1117).
Motivation in PCC programs includes reflexive and automatic motivation , which guides one to produce positive or negative emotions toward behavioral goals by increasing knowledge and experiences. Encouragement of self-reflection and regulation in practice leads to self-improvement and provide better care services for older people. Respecting the autonomy of older people and maintaining resilient and positive attitudes contribute to the engagement of all stakeholders in the process of PCC. Furthermore, the lack of clear reward and empowerment mechanisms can reduce staff motivation.
The HCPs can promote self-directed learning through critical thinking skills and engage in self-correction and reflection. Self-reflection is observing and evaluating self-perceptions, emotions, and behaviors . Self-regulation is a motivational mechanism by which an individual's cognitive development moves from a state of disequilibrium to equilibrium . This review refers to how HCPs can reinforce, maintain, or change their behavior based on available rewards or punishments. I have probably psychoanalyzed a lot of the experiences that I’ve been through to learn from my mistakes. (Narayan MC and Mallinson RK, 2022, P.5).
Participants reported that HCPs and family caregivers should consider proactive and compassionate care at the heart of their practice and respect the autonomy of older people. They also stated that older people should be treated with respect and dignity while maintaining their autonomy and equality. Actually, whenever I go to doctor, they call me by name. Once [they] call me by my name I feel close, attached to them. Otherwise, I'm going to feel bad. my relationship with my doctor is really good. (Manalili, K et al., 2021, P.8). Lack of clear reward and empowerment mechanisms (barrier) Establishing rewards and accountability mechanisms can help foster active HCPs engagement. Some participants stated that the role and task of case managers should be clarified. They should maintain close communication with older people, provide feedback on care plans, and keep an open mind. Doing the right thing is quality, right thing is a standard. So, if you are diagnosed with particular disease for a patient, then you have to do the right things, what you need to do, so quality, in my opinion he's doing the right things. (Manalili, K et al., 2021, P.10). Having a resilient and optimistic attitude (enabler) Being resilient and optimistic toward the caring management of older people is a vital step toward activation in PCC programs. The HCPs can overcome the barriers with a resilient attitude that embrace the positive aspects of implementing PCC. I make it work for me and for the patient. And I can make it work for the agency as well. I’m not afraid to think out of the box, change the way I do things. I am always thinking is there something that I could do differently to be more successful” (Narayan MC and Mallinson RK, 2022, P.6).
Establishing rewards and accountability mechanisms can help foster active HCPs engagement. Some participants stated that the role and task of case managers should be clarified. They should maintain close communication with older people, provide feedback on care plans, and keep an open mind. Doing the right thing is quality, right thing is a standard. So, if you are diagnosed with particular disease for a patient, then you have to do the right things, what you need to do, so quality, in my opinion he's doing the right things. (Manalili, K et al., 2021, P.10).
Being resilient and optimistic toward the caring management of older people is a vital step toward activation in PCC programs. The HCPs can overcome the barriers with a resilient attitude that embrace the positive aspects of implementing PCC. I make it work for me and for the patient. And I can make it work for the agency as well. I’m not afraid to think out of the box, change the way I do things. I am always thinking is there something that I could do differently to be more successful” (Narayan MC and Mallinson RK, 2022, P.6).
The ConQual approach was used to assess the confidence level of synthesized findings, including credibility and dependability . The evidence quality of synthesized findings is initially assumed to be high. However, the final confidence for all synthesized findings was low, downgraded by two levels (limitation of included studies). Additional file depicts the ConQual Scores and the summary of the synthesized findings.
This systematic review identified 12 qualitative studies from diverse countries on implementing PCC in the community and included the perspectives of different stakeholders: older people, HCPs, and caregivers. We derived three synthesized findings: capacities of older people, HCPs, and caregivers; opportunities and motivation for implementing PCC. These findings identify the barriers and enablers to implementing effective PCC interventions. This study is timely, considering successful PCC programs might benefit from the quality improvement of care among community-dwelling older people. The capacities of older people, HCPs, and caregivers are essential factors affecting the successful implementation of PCC programs Most participants emphasized the significance of person-centered communication skills and shared decision-making capabilities for implementing effective PCC interventions in the community. Through shared decision-making, older people's values and preferences are combined with the expertise and knowledge of their caring teams . Older people should express their needs and preferences clearly to translate them into professional actions that result in person-centered outcomes . However, it is noteworthy that there is a fine line between negotiating care goals with older people and providing the care they desire . Therefore, appropriate social relationships should be established among older people, family caregivers, and HCPs. Mutual communication must be improved to meet person-centered service imperatives and improve quality care. The findings of this review also suggest that it would also be useful to provide evidence-based training by combining experience with literature to enhance the professional capabilities of HCPs . The currently available evidence on PCC interventions is not necessarily used in daily clinical practice . Professional training should be designed to integrate theoretical knowledge with practice wisdom. In the process of training, the practicability of theoretical knowledge should be highlighted, trainees' self-thinking and adaptability should be improved . Improving the professional knowledge and skills of HCPs will also, in turn, improve their motivation to participate in PCC programs. Opportunities are crucial factors affecting the successful implementation of PCC programs Most challenges and obstacles exist in the field of opportunity. This review has shown that establishing multidisciplinary teamwork would increase the opportunity to implement individualized care plans. According to the plan-do-check-act (PDCA) cycling management mode, multidisciplinary teamwork and dynamic self-monitoring management are essential for continuous quality improvement . Other studies have also drawn a similar conclusion. For example, Chenoweth and Wu et al. found that collaboration between team members was favorable for fostering a positive environment for meaningful interactions between HCPs and older people. Establishing a safe and friendly environment is an enabler in implementing PCC. The Institute of Medicine (IOM) put forward that "PCC" means that HCPs treat older people as equal partners rather than recipients, establish a trustful and respectful relationship with older people and their families, ensure that older people receive education and support they need, and develop effective care plans . These show the importance of a trusting and welcoming environment for effective PCC implementation. Time constraint is a barrier to successfully implementing PCC in practice. The essence of time constraints is significantly understaffed and low working efficiency, reflected in lower HCPs to older people ratios and insufficient capacity. The increased health needs and workloads add to the work burden and time pressure. It has been suggested that the shortage of HCPs can be addressed by training more health personnel and by improving working efficiency and quality . Overall, an adequate supply of resources is more conducive to PCC programs; the community infrastructure and positive environment enable older people to experience relative well-being . Motivation is a necessary component for the successful implementation of PCC programs Motivation is a mental process that directs behavior . The real action is motivated by the desires of the participants , and in the context of implementing PCC interventions, encouraging self-reflection and regulation are crucial for preserving motivation. The present findings suggest that HCPs can learn from their mistakes through self-correction and reflection to create customized interventions more effectively. Ennis showed critical self-reflection, which is the ability to evaluate, analyze and synthesize outside information that can influence our beliefs and actions . Therefore, HCPs can better adapt to changes in unforeseen circumstances by fostering their reflective skills, self-correction, and reflection, which impact effective behavior change regarding PCC interventions. The findings of McKenzie and Brown in this review also affirm the impact and importance of reflective practice on the motivation of HCPs to actively participate in PCC programs . Consequently, it is critical to help HCPs in developing self-reflective skills in PCC interventions. We found that older people are more willing to cooperate if HCPs respect their autonomy. Clinicians mentioned that older people should exercise their rights, and they should also convey respect and empathy for older people . Jean Watson’s theory of human caring mentioned that each individual should be treated as a whole, their rights respected and treated equally, and the care recipient's self-identity supported . Furthermore, some participants focused on the availability and appropriateness of care for older people. These results are consistent with a previous study that found considerations for access and equity in the healthcare system . Implementing effective PCC interventions for older people is hampered by a lack of clear reward and empowerment mechanisms. The community managers should appropriately authorize HCPs, create a fair, caring, and rule-oriented ethical atmosphere, and establish a firm and reasonable reward-and-punishment mechanism . Empowerment (knowledge, competence, values, impact) and establishment of reward and punishment mechanisms are crucial for promoting the likelihood of sustained lean efforts and improving the desired health outcomes . Maintaining resilience and optimism toward PCC programs was found to increase the motivation of the participants. Some participants expressed concern that HCPs be uncontrollably biased against the patient, which affects the person's perception . Besides, some participants reported that HCPs could adjust their work roles according to their situation and overcome work with a resilient attitude . A previous study also found a resilient attitude capable of adapting to changing situations . Therefore, measures should be taken (e.g., articulating the benefits of PCC and enhancing the practice skills of HCPs) to enhance the positive and resilient attitude of HCPs toward changes. Qualitative research emphasizes subjectivity and individuality. Meta-synthesis is a systematic review of original qualitative research that can comprehensively interpret the phenomenon. This review provides researchers with the perspectives and experiences of stakeholders regarding PCC interventions as well as an understanding of potential factors that may influence the implementation of PCC programs. However, there are several limitations in this review. First, the included literature is primarily regarded as having moderate dependability because most studies do not provide a statement locating the researcher theoretically or culturally. Second, the number of included studies was too small, and available information may be insufficient. Third, given the limitations of included articles, we did not stratify the analysis by types of HCPs. Fourth, we only included research reported in English, and a potential publication bias may be triggered, which may result in the potential to miss relevant articles. Furthermore, quantitative studies may provide barriers in limitations sections rather than in the results section, so they are also excluded. Finally, the present findings have limited generalizability because the included studies were conducted in developed countries.
Most participants emphasized the significance of person-centered communication skills and shared decision-making capabilities for implementing effective PCC interventions in the community. Through shared decision-making, older people's values and preferences are combined with the expertise and knowledge of their caring teams . Older people should express their needs and preferences clearly to translate them into professional actions that result in person-centered outcomes . However, it is noteworthy that there is a fine line between negotiating care goals with older people and providing the care they desire . Therefore, appropriate social relationships should be established among older people, family caregivers, and HCPs. Mutual communication must be improved to meet person-centered service imperatives and improve quality care. The findings of this review also suggest that it would also be useful to provide evidence-based training by combining experience with literature to enhance the professional capabilities of HCPs . The currently available evidence on PCC interventions is not necessarily used in daily clinical practice . Professional training should be designed to integrate theoretical knowledge with practice wisdom. In the process of training, the practicability of theoretical knowledge should be highlighted, trainees' self-thinking and adaptability should be improved . Improving the professional knowledge and skills of HCPs will also, in turn, improve their motivation to participate in PCC programs.
Most challenges and obstacles exist in the field of opportunity. This review has shown that establishing multidisciplinary teamwork would increase the opportunity to implement individualized care plans. According to the plan-do-check-act (PDCA) cycling management mode, multidisciplinary teamwork and dynamic self-monitoring management are essential for continuous quality improvement . Other studies have also drawn a similar conclusion. For example, Chenoweth and Wu et al. found that collaboration between team members was favorable for fostering a positive environment for meaningful interactions between HCPs and older people. Establishing a safe and friendly environment is an enabler in implementing PCC. The Institute of Medicine (IOM) put forward that "PCC" means that HCPs treat older people as equal partners rather than recipients, establish a trustful and respectful relationship with older people and their families, ensure that older people receive education and support they need, and develop effective care plans . These show the importance of a trusting and welcoming environment for effective PCC implementation. Time constraint is a barrier to successfully implementing PCC in practice. The essence of time constraints is significantly understaffed and low working efficiency, reflected in lower HCPs to older people ratios and insufficient capacity. The increased health needs and workloads add to the work burden and time pressure. It has been suggested that the shortage of HCPs can be addressed by training more health personnel and by improving working efficiency and quality . Overall, an adequate supply of resources is more conducive to PCC programs; the community infrastructure and positive environment enable older people to experience relative well-being .
Motivation is a mental process that directs behavior . The real action is motivated by the desires of the participants , and in the context of implementing PCC interventions, encouraging self-reflection and regulation are crucial for preserving motivation. The present findings suggest that HCPs can learn from their mistakes through self-correction and reflection to create customized interventions more effectively. Ennis showed critical self-reflection, which is the ability to evaluate, analyze and synthesize outside information that can influence our beliefs and actions . Therefore, HCPs can better adapt to changes in unforeseen circumstances by fostering their reflective skills, self-correction, and reflection, which impact effective behavior change regarding PCC interventions. The findings of McKenzie and Brown in this review also affirm the impact and importance of reflective practice on the motivation of HCPs to actively participate in PCC programs . Consequently, it is critical to help HCPs in developing self-reflective skills in PCC interventions. We found that older people are more willing to cooperate if HCPs respect their autonomy. Clinicians mentioned that older people should exercise their rights, and they should also convey respect and empathy for older people . Jean Watson’s theory of human caring mentioned that each individual should be treated as a whole, their rights respected and treated equally, and the care recipient's self-identity supported . Furthermore, some participants focused on the availability and appropriateness of care for older people. These results are consistent with a previous study that found considerations for access and equity in the healthcare system . Implementing effective PCC interventions for older people is hampered by a lack of clear reward and empowerment mechanisms. The community managers should appropriately authorize HCPs, create a fair, caring, and rule-oriented ethical atmosphere, and establish a firm and reasonable reward-and-punishment mechanism . Empowerment (knowledge, competence, values, impact) and establishment of reward and punishment mechanisms are crucial for promoting the likelihood of sustained lean efforts and improving the desired health outcomes . Maintaining resilience and optimism toward PCC programs was found to increase the motivation of the participants. Some participants expressed concern that HCPs be uncontrollably biased against the patient, which affects the person's perception . Besides, some participants reported that HCPs could adjust their work roles according to their situation and overcome work with a resilient attitude . A previous study also found a resilient attitude capable of adapting to changing situations . Therefore, measures should be taken (e.g., articulating the benefits of PCC and enhancing the practice skills of HCPs) to enhance the positive and resilient attitude of HCPs toward changes. Qualitative research emphasizes subjectivity and individuality. Meta-synthesis is a systematic review of original qualitative research that can comprehensively interpret the phenomenon. This review provides researchers with the perspectives and experiences of stakeholders regarding PCC interventions as well as an understanding of potential factors that may influence the implementation of PCC programs. However, there are several limitations in this review. First, the included literature is primarily regarded as having moderate dependability because most studies do not provide a statement locating the researcher theoretically or culturally. Second, the number of included studies was too small, and available information may be insufficient. Third, given the limitations of included articles, we did not stratify the analysis by types of HCPs. Fourth, we only included research reported in English, and a potential publication bias may be triggered, which may result in the potential to miss relevant articles. Furthermore, quantitative studies may provide barriers in limitations sections rather than in the results section, so they are also excluded. Finally, the present findings have limited generalizability because the included studies were conducted in developed countries.
Additional file depicts recommendations from the review, which has been assigned a level of recommendation based on guidelines from the JBI . Three grades of recommendation are used: Grade "A" (strong recommendation), Grade "B" (intermediate recommendation), and Grade "C” (weak recommendation). In order to ensure the success of PCC interventions, stakeholders (HCPs, older people, and caregivers) must have adequate knowledge and competence as well as access to education programs. The community managers should establish and integrate the multidisciplinary team and conduct rational coordination of resource allocation. Therefore, management structures and processes for developing care plans may need to be realigned.
This study comprehensively synthesized related qualitative evidence that can guide the implementation of PCC intervention programs. A re-conceptualization process was used in the meta-synthesis to understand enablers and barriers in order to provide PCC for community-dwelling older people. The lack of person-centered communication skills, negative attitudes toward shared decision-making, lack of coordination in resource allocation, lack of clear reward and empowerment mechanisms, and time constraints limited the effective implementation of PCC. A supportive environment, positive motivation, and professional educational training would facilitate the implementation of PCC. Considering the above factors as the entry point, community-based interventions could be implemented to improve the practical level of PCC.
Additional file 1. Additional file 2. Additional file 3. Additional file 4. Additional file 5. Additional file 6.
|
Nudging General Practitioners to explore suicidal thoughts among depressed patients | 0a8db1e7-e977-413c-8e75-2d4d0050c220 | 10067310 | Family Medicine[mh] | Worldwide, over 700.000 people die every year as a result of suicide and approximately 20 times as many attempt suicide . In addition, an estimated 8% of people experience suicidal thoughts at some point in their live . Suicidal thoughts are an important risk factor for suicidal acts; over a quarter of people who experiences these thoughts attempts suicide later in life . Early recognition and intervention of patients at risk for suicide is therefore crucial. In countries with a primary care network, such as the Netherlands, General practitioners (GPs) are the core of the health care system. They function as gatekeepers in referring patients to specialized care and often have enduring relationships with patients . Therefore, GPs have a large exposure to patients who are potentially at risk for suicide. In fact, people who died by suicide were more often in contact with a GP than with mental health services prior to death . A Dutch study reported that in the month before their suicide or attempted suicide, about half of the patients had consulted a GP . Supporting primary care is therefore seen as one of the most effective elements of suicide prevention strategies . In primary care, suicide prevention practices are mostly delivered by physicians, and their behaviour serves as proxy for quality of care. Improving the quality of care often requires physicians to change behaviour in accordance with evidence-based practice or clinical guidelines . Various clinical guidelines recommend assessing the presence of suicidal thoughts among patients with an elevated risk, such as patients diagnosed with depression . Regardless, in a previous study we found that assessment of suicidal thoughts occurs in about two third of depression-related consultations . Other studies have also shown that suicide exploration rates in primary care are suboptimal . Many strategies have been developed to influence professional behaviour, although most consist of providing education or training . Suicide prevention interventions in primary care also mostly rely on the provision of education and training . Although widely recommended, most studies were only able to report short-term effects on confidence, knowledge or skills, and fail to report long-term effects on behaviour change . Changing behaviour is complex and often requires more than influencing knowledge or skills . An important overview study of systematic reviews on changing provider behaviour also emphasized that education provides small and short-term effects. More passive methods, such as the delivery of information and materials, are only effective for creating awareness. Issuing reminders was the most promising individual intervention to substantially change medical practice, especially for the provision of preventive care . Besides, developing a reminder system is a relatively inexpensive intervention that is easy to implement . Issuing reminders may thus be a potential strategy to change provider behaviour and increase suicide exploration rates in general practice. To our knowledge, this has not been tested before. This study aims to assess whether a pop-up reminder system nudges GPs to more frequently explore suicidal thoughts over the course of two years. We implemented a pop-up intervention in GPs’ Electronic Health Record (EHR) to alert GPs to explore suicide risk in patients consulting for depression. Because of the great variability in suicide exploration of GPs we found in a previous study , we also tested if general practices, as hypothesized, show a different development over time.
Design and setting This pre-post study was performed among the network of Dutch Sentinel general practices of the Nivel Primary Care database . The sentinel network consists of approximately 40 general practices, which provide in-depth health care information on illnesses and procedures that cannot be obtained from electronic medical records. In 2018, the network of sentinel general practices had almost 130.000 registered patients and was representative in terms of patients’ age, gender, geographical distribution, and population density . We created an automated pop-up screen that was implemented in GPs’ EHR. The pop-up screen was activated after recording a new episode of depression during a consultation. This pop-up screen referred to a questionnaire designed to gather data about GPs’ suicide exploration behaviour, described previously . Within this infrastructure, it was not possible to create a control group because the pop-up functioned as both intervention and measurement instrument. Effect of the pop-up screen was therefore only assessed in relation to consultation year. Instrument and measures The pop-up screen itself included some control questions about the consultation to make sure it concerned (1) a regular patient of the GP, (2) a patient with depression, (3) a new episode of depression, and (4) a face-to-face consultation. If these were answered affirmatively, the follow-up questionnaire was activated. GPs could access the questionnaire straight away or complete it at a later moment. The full questionnaire included 19 items, but not all questions were applicable during all consultations. On average, it took 66 seconds to complete the questionnaire. Our dichotomous main outcome measure ‘suicide exploration’ was assessed with the question: “have you asked the patient if he/she experiences suicidal thoughts?”. Data collection Data was collected from January 2017 until December 2018. General practices that did not provide data over the full study period were excluded, yielding a total of 35 general practices. The present study is based on consultations with patients who were diagnosed with a new episode of depression. Analyses We analysed our data using multilevel logistic regression techniques. The first model only includes year of consultation and was adjusted for clustered data of patients within general practices (two level structure). Since we found in a previous study that suicide exploration occurred more frequently in male and younger patients , we created a new model where we adjusted for these variables. The Intraclass Correlation Coefficient (ICC) is a measure to calculate the between group variance as proportion of the total variance and is used to estimate variation between general practices . We added a random slope for consultation year to the model and used the likelihood ratio test to assess whether the general practices showed different development over time. We computed and plotted the random effects to specify the variation between general practices. The analyses were performed using RStudio Statistical Software (version 2021.09.1) using lme4 and ggeffects packages . The significance level was set at < 0.05.
This pre-post study was performed among the network of Dutch Sentinel general practices of the Nivel Primary Care database . The sentinel network consists of approximately 40 general practices, which provide in-depth health care information on illnesses and procedures that cannot be obtained from electronic medical records. In 2018, the network of sentinel general practices had almost 130.000 registered patients and was representative in terms of patients’ age, gender, geographical distribution, and population density . We created an automated pop-up screen that was implemented in GPs’ EHR. The pop-up screen was activated after recording a new episode of depression during a consultation. This pop-up screen referred to a questionnaire designed to gather data about GPs’ suicide exploration behaviour, described previously . Within this infrastructure, it was not possible to create a control group because the pop-up functioned as both intervention and measurement instrument. Effect of the pop-up screen was therefore only assessed in relation to consultation year.
The pop-up screen itself included some control questions about the consultation to make sure it concerned (1) a regular patient of the GP, (2) a patient with depression, (3) a new episode of depression, and (4) a face-to-face consultation. If these were answered affirmatively, the follow-up questionnaire was activated. GPs could access the questionnaire straight away or complete it at a later moment. The full questionnaire included 19 items, but not all questions were applicable during all consultations. On average, it took 66 seconds to complete the questionnaire. Our dichotomous main outcome measure ‘suicide exploration’ was assessed with the question: “have you asked the patient if he/she experiences suicidal thoughts?”.
Data was collected from January 2017 until December 2018. General practices that did not provide data over the full study period were excluded, yielding a total of 35 general practices. The present study is based on consultations with patients who were diagnosed with a new episode of depression.
We analysed our data using multilevel logistic regression techniques. The first model only includes year of consultation and was adjusted for clustered data of patients within general practices (two level structure). Since we found in a previous study that suicide exploration occurred more frequently in male and younger patients , we created a new model where we adjusted for these variables. The Intraclass Correlation Coefficient (ICC) is a measure to calculate the between group variance as proportion of the total variance and is used to estimate variation between general practices . We added a random slope for consultation year to the model and used the likelihood ratio test to assess whether the general practices showed different development over time. We computed and plotted the random effects to specify the variation between general practices. The analyses were performed using RStudio Statistical Software (version 2021.09.1) using lme4 and ggeffects packages . The significance level was set at < 0.05.
Table presents the descriptive statistics per year and in total. In total, we included 625 completed questionnaires about consultations with patients consulting for a new episode of depression. The majority (60%) were women, mean age of the patients was 49.9 years old (sd 19.0). Overall, in 69% of the consultations, suicidal feelings were explored and in 45% of those, patients reported suicidal feelings. Table shows the results of the multilevel analyses with a random intercept on general practice level. The first model shows the influence of contact year on suicide exploration while adjusted for nested data of consultations within general practices. This model shows that the number of consultations during which suicide was explored was significantly higher in 2018 compared to 2017 (OR 1.48; 95%CI 1.01–2.16). The ICC that was calculated for this model is 0.24, meaning that 24% of the variation in suicide exploration is explained by general practice. In the second model, which is adjusted for patients’ gender and age, the effect of consultation year on suicide exploration of GPs attenuated (OR 1.33; 95%CI 0.90–1.97). Female gender and older age are both negatively correlated with suicide exploration (respectively OR 0.64; 95% CI 0.43–0.98 and OR 0.97; 95% CI 0.96–0.98 per year older). The ICC for this model is 0.26, indicating that 26% of the variation in suicide exploration among patients is explained by general practices. We also created a model with a random slope for consultation year to assess whether suicide exploration of general practices developed differently over time. According to the log likelihood ratio test, this was not the case (X 2 = 1.212, P = 0.545). This model is therefore not presented here. In Fig. the random effects (intercepts) of general practices are plotted to show how suicide exploration differs per general practice. The lowest value represents a general practice where suicidality was explored in 1/12 depressed patients and the highest value represents a general practice where this was explored in 17/17 patients.
This study shows that patient-specific pop-up reminders were not effective in nudging GPs to more frequently explore suicidal thoughts over a period of two years. GPs were 50% more likely to explore suicidal thoughts in the second year compared to the first, however, this effect disappeared after adjusting for patients’ gender and age. Though there was considerable practice variation in suicide exploration, general practices showed no different development over time. Issuing reminders effectively changed professional behaviour in previous studies across a range of settings . Shea at al. reported that issuing reminders was effective for improving preventive services with 77% overall. However, not all preventive services were improved; screening for cervical cancer, for instance, was not. The authors speculate this is related to contextual factors, specifically the time it takes to perform a pelvic examination and patients’ resistance for this procedure . The reason why we did not find evidence for the effectiveness of the reminder system in the present study, may be because of the suicide prevention context, which is perceived difficult and complex by many primary care professionals . In order for interventions to effectively change professionals’ behaviour, it is important to adjust them to the context and consider barriers that are in place. For suicide prevention, important barriers are lack of knowledge, skills and/or confidence, lack of time, and limited access to mental health care . Most interventions to change suicide prevention practices of professionals include providing education or training . This addresses the individual barriers lack of knowledge, skills and/or confidence. However, in order to effectuate true behaviour change, the environmental barriers should be addressed too. These are much harder to influence because they often require changes on system level. Effective interventions in primary care that address these include the institution of specialized nurses or other health care professionals, organizational changes to increase collaboration within primary care professionals, or collaborative or shared care practices . Not all these interventions have been specifically evaluated for suicide prevention. The mental health support staff was introduced in primary care about a decade ago. This positively influenced suicide prevention practices and was highly valued by GPs . A collaborative care model for suicide prevention was recently implemented and tested. Although we are awaiting the long-term results, short term results are promising and stakeholders especially valued the chain of care element to support collaboration . The Behaviour Change Wheel (BCW) is a contemporary model for behaviour change that can be used for understanding and influencing behaviour in context. In the core of the wheel is the COM-B system, containing the three constructs of behaviour (B): Capability, Opportunity and Motivation. Surrounding the core are nine intervention functions, such as education, training, enablement, and environmental restructuring. These can be used to address deficits in one or more of these behaviour constructs. Education and training influence professionals’ capability and reflective motivation . Providing pop-ups in the EHR system to remind clinicians to engage in a certain behaviour is a type of environmental restructuring. Prompting these questions restructures the physical context, and influences the opportunity and automatic motivation conditions of clinicians’ behaviour . This increases the chance that newly adopted behaviour is structurally applied. Combining these with other interventions to overcome the environmental barriers for suicide prevention in primary care modifies behaviour more sustainably. This is in line with recommendations from other studies, arguing that multifaceted interventions targeting various barriers and influencing multiple behaviour constructs simultaneously are more likely to effectuate behaviour change and improve patient-level outcomes . Strengths & limitations The naturalistic design of this study was a major strength. We developed a feature that was inexpensive and easy to incorporate in the GPs’ natural workflow. Especially for GPs this is important, because they are known to have restricted time, so developing interventions that add to their high workload will only increase their burden further. The fact that we collected over 600 completed questionnaires in two years shows we were able to adapt to their workflow and create engagement among these GPs. However, the design also caused some important limitations. First, it was not possible to compare the effects of the intervention group with a control group. Therefore, we cannot rule out whether any of the results are influenced by other extraneous variables. Second, only a limited number of variables were available for this study. Variables we expected to be of influence, such as work experience or previous mental health training , are missing. This limited our possibilities to adjust for confounding or effect modification. Finally, implementing pop-up reminders in the EHR system has become more prevalent in clinical practice. Even to the point that it may result in ‘alert fatigue’, causing health care professionals to ignore alerts due to the overload of prompts and reminders . Unfortunately, this data was not available so we could not take it into consideration. We recommend researchers to test whether implementing these nudges as part of a multifaceted approach will lead to a stronger effect on GPs’ suicide exploration. Further, we recommend adopting a design that allows for comparison with a control group and to include more GP- en general practice variables to develop better understanding of the effect of the reminder system. Previous studies among the sentinel practices have indicated some clinical factors that are associated with suicide attempts and suicides, such as high consultation frequency and other psychological complaints or disorders . Furthermore, suicidal thoughts should also be assessed during later consultations concerning depression, and, preferably, during consultations concerning depressive mood or other psychological complaints. Future studies are therefore recommended to broaden their scope and include more illnesses and complaints. Finally, these insights and insights about sociodemographic variables should be used to improve the accuracy of the reminders and determine whether this affects the effectiveness of the pop-up system.
The naturalistic design of this study was a major strength. We developed a feature that was inexpensive and easy to incorporate in the GPs’ natural workflow. Especially for GPs this is important, because they are known to have restricted time, so developing interventions that add to their high workload will only increase their burden further. The fact that we collected over 600 completed questionnaires in two years shows we were able to adapt to their workflow and create engagement among these GPs. However, the design also caused some important limitations. First, it was not possible to compare the effects of the intervention group with a control group. Therefore, we cannot rule out whether any of the results are influenced by other extraneous variables. Second, only a limited number of variables were available for this study. Variables we expected to be of influence, such as work experience or previous mental health training , are missing. This limited our possibilities to adjust for confounding or effect modification. Finally, implementing pop-up reminders in the EHR system has become more prevalent in clinical practice. Even to the point that it may result in ‘alert fatigue’, causing health care professionals to ignore alerts due to the overload of prompts and reminders . Unfortunately, this data was not available so we could not take it into consideration. We recommend researchers to test whether implementing these nudges as part of a multifaceted approach will lead to a stronger effect on GPs’ suicide exploration. Further, we recommend adopting a design that allows for comparison with a control group and to include more GP- en general practice variables to develop better understanding of the effect of the reminder system. Previous studies among the sentinel practices have indicated some clinical factors that are associated with suicide attempts and suicides, such as high consultation frequency and other psychological complaints or disorders . Furthermore, suicidal thoughts should also be assessed during later consultations concerning depression, and, preferably, during consultations concerning depressive mood or other psychological complaints. Future studies are therefore recommended to broaden their scope and include more illnesses and complaints. Finally, these insights and insights about sociodemographic variables should be used to improve the accuracy of the reminders and determine whether this affects the effectiveness of the pop-up system.
Developing a pop-up intervention and implementing it in the GPs’ information system is a low-cost and easy to administer intervention. Unfortunately, we were not able to report a robust effect of the intervention on GPs’ suicide exploration behaviour. We encourage studies to test whether implementing these nudges as part of a multifaceted approach to improve suicide prevention practices of GPs is more effective. In addition, we recommend researchers to include more GP- and general practice variables to better understand the effect of the intervention on GPs’ behaviour.
|
Highlighting the immunohistochemical differences of malignant mesothelioma subtypes via case presentations | 467339f9-42d6-4087-a7f4-7acde60d269d | 10067356 | Anatomy[mh] | The first case of malignant mesothelioma (MM) was described in 1767 by Joseph Lieutand. He characterized it as “pleural tumor”, while in 1931, Rabin and Klemperer recommended the use of the term mesothelioma. MM is a rare tumor of mesothelial cells, with an increasing incidence both in developed, and developing countries. Males are 3–4 times more likely to be affected, and the average age for patients is 70 years. Few cases have been described in children, albeit in those cases, no etiological connection has been found with asbestos exposure. , This type of malignancy has high mortality due to its aggressive growth, unspecific symptoms, and difficulties in surgical removal. The pleura is by far the most commonly affected area, followed by the peritoneum, and the pericardium. Even though it is stated that MM is caused by industrial pollutants, most of all, asbestos has been defined as a causative agent, and it has also been associated with prior ionizing radiation. , Symptoms of MM are nonspecific, including dyspnea, chest pain, and general tumor manifestations, such as cachexia, fever, and fatigue. Therefore, the diagnosis is often encumbered, and delayed. Primary peritoneal mesothelioma often presents as abdominal pain, and is first misdiagnosed as cholecystitis. The first‐line diagnostic tool for MM are imaging techniques, including computed tomography (CT), magnetic resonance imaging (MRI), positron emission tomography/computed tomography (PET/CT), and ultrasonography. Despite the development of radiology, a definitive diagnosis can only be facilitated with histological evaluation. According to the World Health Organization (WHO) Classification of 2021, MM has three major histological subtypes, namely epithelioid, biphasic, and sarcomatoid. , , MM has to be distinguished from primary or secondary lung tumors, and then evidently the MM subtypes have to be differentiated from each other. Proper distinction may be a challenging task for the pathologist, due to the unspecific morphology. The preferred treatment option for all MM subtypes is surgical resection, and a favourable outcome has been reported when combined with chemo‐ or radiotherapy, although relapse is still fairly common. , , , Regardless of the therapeutic options, the prognosis of diffuse MM remains dismal. Amin and coauthors analyzed 888 cases of pleural and peritoneal MM, and in their study, the median overall survival of these patients was 15 months, with better outcomes in patients with peritoneal involvement. Favorable prognostic factors have been identified, namely female gender, younger age (less than 45 years), epithelioid histological subtype, stage I category, peritoneal presence, and combined surgical and chemotherapeutical treatment. Sarcomatoid MM represents an even poorer prognostic group. This subtype also mainly arises from the pleura, and an association with asbestos was not found in the majority of cases examined. Two cases of MM are presented here to emphasize the clinicopathological features of this tumor focusing on immunohistochemical (IHC) characteristics, and to facilitate the establishment of correct diagnosis.
MM A 78‐year‐old male patient with a history of ischemic heart disease, type 2 diabetes, and atrial fibrillation had been treated for months with therapy‐resistant hydrothorax. Even though several thoracenteses were carried out, the evaluation of the drained fluid did not confirm malignancy. In April 2021, he was admitted to the University of Szeged to surgically manage the recurring hydrothorax. During the surgery, the preliminary diagnosis was pyothorax or disseminated tumor. Chest X‐ray examination of the patient showed signs of congestion of the pulmonary circulation, cardiomegaly, and fluid accumulation in the left sinus (Figure ). Macroscopic examination of the surgical specimen revealed firm, gray thickening of the pleura. Histological evaluation showed that the entire extent of the pleura had been infiltrated by relatively monomorphic, epithelioid neoplastic cells, forming solid nests or trabeculae between collagen bundles (Figure ). IHC was applied to identify the nature of the malignant neoplasm. The neoplastic cells showed cytokeratin 5/6 (CK5/6; Figure ), calretinin (Figure ) and Wilms‐tumor‐1 (WT1) expression, and were negative for thyroid transcription factor‐1 (TTF‐1), and BerEP4. BRCA1‐associated protein 1 (BAP1) negativity was seen in the nucleus of neoplastic cells, reflecting the loss of BAP1 tumor suppressor genes (Figure ). Ki‐67 proliferation fraction was approximately 10% (Figure ). The case was concluded to be epithelioid MM. Although since the surgery, the patient did not cooperate with the medical team and did not appear in the control examinations, he is still alive (overall survival: OS = 20 months).
MM A 69‐year‐old male patient with a history of smoking, chronic obstructive pulmonary disease, prostatic hyperplasia, and cataract was admitted to the hospital, due to an accident at home. The patient complained of severe thoracic pain, localized specifically to the ribs, and his left shoulder. During the exploration of the patient's medical history, it was discovered that he had been working as a mechanic, and although he was not known to have been exposed to asbestos, he had been heating his home with coal for a decade. The first CT scan revealed tumorous thickening of the left sixth and seventh ribs, nearly 8 cm in largest diameter. The sixth, seventh, and eighth ribs and the intercostal muscles were surgically resected, and a GORE‐TEX patch was applied for the reconstruction of the chest wall defect. Histological examination of the specimen showed tumorous infiltration of a biphasic neoplasm, consisting of both epithelioid, and spindle cell components. The former component formed solid nests, the latter created irregular fascicles. The atypical spindle cells demonstrated expression of epithelial membrane antigen (EMA), cytokeratin AE1/AE3 (CK AE1/AE3), and mesothelin, while WT1, epithelial cell adhesion molecule (EpCAM), thrombomodulin, TTF1, p63, CD31, and calretinin remained negative. According to the histomorphology, and immunophenotype, a rib destructing biphasic MM was diagnosed. Complete resection of the tumor could not be confirmed from the surgical specimen. The patient received four cycles of cisplatin, and pemetrexed combined chemotherapy. At the end of 2020, the PET/CT examination reported recurring tumorous involvement of the pleura, ribs, and also the lungs (Figure ). The metastatic foci located in the left upper, and lower lobes of the lung, and the infiltrated chest wall including the residual sixth rib, were removed in a second surgical procedure. The chest wall defect was covered with a GORE‐TEX patch. Microscopic examination described tumor cells with decisively spindle cell morphology, surrounded by abundant hyalinised stroma. Focally, extreme pleomorphism, and multinucleated tumorous cells, and a large number of mitotic figures were also seen (Figure ). Signs of vascular, lymphovascular or perineural invasion were not present. The IHC examination revealed WT1, and mesothelin positivity in the tumor cells, while CK5/6 and calretinin remained negative (Figure ). Loss of BAP1 expression was also described, corresponding with the presence of mutant BAP gene (Figure ). Mitotic rate was high (21 mitoses/10 high power fields). Ki‐67 proliferation marker was expressed in 60% of tumor cells (Figure ). The results of IHC examination, and the microscopic morphology ratified the diagnosis of sarcomatoid MM. Alongside the GORE‐TEX patch, severe foreign body reaction developed, with chronic inflammation, and numerous giant cells. Novel tumorous infiltration of the basis of the left lung has been reported in the most recent control PET/CT examination. The third surgical procedure will be performed with video‐assisted thoracic surgery, and is due in the near future. The OS of the patient is currently 24 months. Because complete resection of the tumor from the first surgery was not proven, it can be stated that the tumor developed as biphasic MM from the beginning, and later the more aggressive part with sarcomatoid morphology recidivated. The OS also supports this hypothesis. Even though the patient was not known to have been exposed to asbestos, the literature already describes the association of MM with coal.
Albeit histologically MM can show diverse morphology, prognosis correlates well with the epithelioid, biphasic, and sarcomatoid classification. According to the study by Amin et al., the median survival of the epithelioid subtype was 18 months, while it proved to be 10 months for the biphasic subtype, and remained only 7 months for the sarcomatoid subtype. In a large scale series by Brustugun and coworkers that examined 1509 MM cases in a 20‐year‐long period, an even worse prognosis was observed, with median survival of 5.1 months for nonepithelioid subtypes. The chemotherapeutical response of MM subtypes has been investigated in some studies. In the meta‐analysis by Mansfield et al., the results of 41 trials were analyzed and revealed that the rate of response to chemotherapy was only 21.9%, and 13.9% of patients with epithelioid, and sarcomatoid MM, respectively. A primary diagnosis of MM is still demanding. Differential diagnosis includes primary lung adenocarcinoma, squamous cell carcinoma, sarcomatoid carcinoma, vascular tumors, melanoma, and metastatic origin (breast, gastrointestinal, prostate, kidney, ovary, thyroid cancer etc) also has to be excluded. Less frequent, but possible challenging diagnosis constitute lymphomas, SMARCA4‐ deficient thoracic tumors, desmoplastic small round cell tumor, monophasic synovial sarcoma, and CIC ‐translocated sarcomas. Regarding nontumorous conditions, inflammation, chronic pyothorax, reactive mesothelial hyperplasia, pleuritis, and callus must be considered. Regarding differential diagnosis of diffuse MM, Ali and coauthors introduced a pattern‐based approach in 2018. Regarding reactive pleural changes versus diffuse MM, the application of the following IHC markers are favored. Desmin and glucose transporter 1 (GLUT1) remains generally positive in mesothelial hyperplasia. In cases of p53, aberrant or nonwild‐type expression could serve as a clue in comprehending malignant versus benign lesions. EMA positivity has been linked both to reactive and neoplastic lesions, although its combined use with desmin could serve as a solution, while EMA positivity alongside with desmin negativity favors diffuse MM. Its opposite, EMA negativity and desmin positivity facilitates reactive processes. , Positivity of insulin‐like growth factor II messenger ribonucleic acid binding protein‐3 (IMP‐3), and thrombomodulin IHC markers tend to be observed more in diffuse MM cases, rather than those which are reactive. , Differentiation of chronic, active, fibrosing pleuritis may be difficult, as a result of the misinterpretation of fatlike spaces being present in organizing pneumonia, and pleuritis cases to real fat tissue infiltration (stromal infiltration) of desmoplastic MM. In such scenarios, S100 can help in discerning actual fat tissue, and fatlike structures. In most cases, the discrimination is mainly based on examination of the hematoxylin and eosin (HE) staining, because a laminar appearance has to be present in fibrosing pleuritis. From inside to outside, several layers have to be defined, including fibrin, neutrophil granulocytes, mononuclear inflammatory cells, granulation tissue, and connective tissue composed of hyalinised collagen bundles. Somatic mutation of tumor suppressor gene, BAP1 has been described as fairly common in diffuse MM. The loss of BAP1 can be observed in the majority of epithelioid, and mixed (60–70%), while it is present in 15% of sarcomatoid MM cases. Since the mutation results in protein loss, during IHC examination, BAP1 negativity could be seen. The lack of BAP1 expression has a low sensitivity (20%–53%), but approximately a 100% specificity as a marker of diffuse MM, therefore BAP1 can serve as a useful tool for distinguishing MM from reactive lesions. , According to the results of Ali et al., fluorescent in situ hybridization (FISH) could be useful in selected cases in order to differentiate benign, and malignant lesions. Since CDKN2A gene codes two proteins via alternative splicing (p16/INK4A and p14/ARF), its loss is detectable. Although this examination has 100% specificity for the diagnosis of MM, it is not sufficient for differentiating epithelioid, and sarcomatoid subtypes. Further molecular diagnostic procedures have not yet been described. Methylthioadenosine phosphorylase (MTAP) is a newly described IHC surrogate of FISH. According to the recommendations of the current WHO, in cases of distinction of carcinoma versus epithelioid, and mixed MM subtypes, at least two carcinoma and two mesothelial IHC markers are required, due to their low sensitivity. , , Spindle cell malignancies can be differentiated from sarcomatoid mesothelioma, with calretinin and D2‐40. Even after finally agreeing upon a diagnosis of MM, the histological evaluation of MM subtypes could also be a challenging task for pathologists because of their nonspecific morphology; therefore, conducting IHC can help in confirming the final diagnosis. In compliance with the above mentioned sections, the following diagnostic algorithm can be applied in cases of epithelioid MM. After exclusion of reactive processes, carcinomas, and mesenchymal neoplasms, additional EMA, desmin, IMP‐3, and thrombomodulin positivity can be observed in the majority of cases, alongside with BAP1 loss. On the other hand, sarcomatoid MM tends to be negative with WT1, B cell lymphoma‐2 (Bcl‐2), CD34, and desmin. In light of the results of Chirieac et al., the majority of sarcomatoid MM cases showed either negativity, or focal positivity of keratin markers, including CKAE1/AE3, CAM 5.2, and MNF 116. Solely, one fourth of cases were positive with calretinin. The review by Rossi et al. highlights the possible aberrant expression of several markers including p40 (5,5%) and p63 positivity in epithelioid MMs, as well as the positivity of the TTF1 SP141clone in 42% of sarcomatoid MM cases. Husain and coauthors emphasize that there is currently no useful IHC recommendation on this matter, furthermore, in some cases, no positivity could be observed, due to the overfixation of the surgical specimen. We would like to further illustrate the diagnostic challenges of MM by mentioning the reproducibility examinations previously reported. The first dates back to 1997 when five pathologists evaluated 77 cases of HE staining, and later evaluated the cases with IHC markers, including cytokeratins, vimentin, HMFG‐2, CD15, BerEP4, B72.3, and carcinoembyonic antigen (CEA). The results reflect that IHC did not change the diagnosis of MM in most cases. Brčić et al. focused on the differentiation of MM subtypes. Three pathologists assessed 200 MM cases, one representative HE slide from each, and moderate agreement (κ = 0.36) was achieved at the first round, while substantial agreement (κ = 0.63) was observed in the second round, after a consensus meeting. The authors emphasize the use of a strict, consensus based diagnosis. A diagnosis of biphasic mesothelioma possibly remains the hardest task after all. Based on the reproducibility examination of by the International Mesothelioma Panel from the MESOPATH Reference Center, moderate interobserver correlation was achieved (weighted κ = 0.45), with 14 examiners evaluating 544 cases by using only BAP1 and p16 IHC stainings. Our two cases and Table summarize the most commonly used and worldworld widely available IHC markers for the differentiation of MM subtypes. Mutual positivity was observed with WT1, and mutual negativity was seen with TTF‐1 or napsin‐A, excluding the possibility of primary lung cancer. In both subtypes, BAP1 was negative, reflecting the loss of gene expression. The most helpful markers in our cases proved to be CK5/6, mesothelin, calretinin, and Ki‐67. The epithelioid subtype showed positivity with all of them, and Ki‐67 proliferation marker was 10%. On the other hand, the sarcomatoid subtype remained negative with CK5/6 and calretinin, had focal cytoplasmic positivity with mesothelin, and Ki‐67 proliferation marker was 50%–60%. We recommend the use of these widely available markers.
The differentiation between MM subtypes could be a challenging task, due to the lack of specific histological features. IHC may be the optimal method in distinction. WT1, TTF‐1, BAP1 markers help setting the diagnosis of MM, while CK5/6, mesothelin, calretinin, and Ki‐67 are helpful in the establishment of subclassification.
Concept and design – Anita Sejben, Tamás Zombori, Tamás Pancsa. Search and evaluation of references – all authors. Drafting the manuscript – Anita Sejben, Tamás Zombori. Approval of final manuscript – all authors.
The authors declare no conflict of interest.
Ethics approval statement was provided from the Institutional Committee of the Albert Szent‐Györgyi Medical University.
|
Biphenotypic sinonasal sarcoma with | 3b3dcb43-7f53-40ea-b591-72c1848c9c1a | 10067655 | Anatomy[mh] | Biphenotypic sinonasal sarcoma (BSNS) was initially described by Lewis et al. in 2012 as a low-grade sarcoma with neural and myogenic features . Clinically, these tumors usually follow an indolent course with frequent recurrences but no reported metastases and only exceptional disease-related mortality . Morphologically, BSNS typically consists of an infiltrative hypercellular proliferation of uniform spindle cells arranged in fascicles, often with a herringbone pattern and frequent invaginations of hyperplastic surface mucosa. Mitotic activity is usually minimal. The vast majority of cases co-express smooth muscle actin (SMA) and S100 protein, while a minority of cases may also exhibit morphological and/or immunohistochemical signs of skeletal muscle differentiation . Fusions of PAX3 gene are the molecular hallmark of BSNS, with MAML3 being the fusion partner in the majority of cases . Herein, we present a unique case of high-grade rhabdomyosarcoma (RMS) emerging from a typical BSNS with an aggressive clinical course.
The patient was a 67-year-old male with Parkinson’s disease and a history of septoplasty, bilateral frontal sinusotomy, and removal of right middle turbinate concha bullosa 3 years before presenting with nasal congestion and epiphora. MRI showed a soft tissue mass involving the right ethmoid, maxillary, and frontal sinuses and invading the extraconal orbit causing mild asymmetric right-sided ocular proptosis (Fig. A,B). Endoscopic evaluation revealed a soft tissue mass measuring 4.4 × 3.4 × 2.2 cm which obscured the nasal airway and centered around the middle turbinate. Biopsy revealed a high-grade sarcoma with myogenic differentiation by immunohistochemistry, diagnosed as embryonal RMS. Staging PET was negative for metastatic disease. The patient initiated neoadjuvant chemotherapy with vincristine-dactinomycin-cytoxan. Repeated imaging after three cycles showed local progression and increased FDG avidity without metastatic disease. The resection specimen showed a proliferation of uniform spindled cells with moderate to high cellularity, occasional fascicular arrangement, minimal atypia, and no mitotic activity (Fig. B). In some areas, invaginations of the hyperplastic surface mucosa were enveloped by the tumor cells (Fig. A). No rhabdoid differentiation was noted by morphological examination in this area. However, in some sections, this typical BSNS morphology sharply transitioned to a high-grade sarcoma with rhabdoid features, very high mitotic activity, and areas of necrosis (Fig. A,B). Immunohistochemically, the conventional BSNS areas showed patchy SMA (Fig. C) and diffuse S100 protein (Fig. D) expression. There was also a diffuse strong positivity with PAX7 (Fig. E) and patchy expression of MyoD1 (Fig. F), while desmin and myogenin were completely negative. In contrast, the high-grade areas were completely negative for S100 protein and SMA, and there was only patchy positivity with PAX7 (Fig. D), whereas the expression of desmin (Fig. C), MyoD1 (Fig. E), and myogenin (Fig. F) was diffuse in these parts. Based on the presence of the typical BSNS areas, molecular studies were performed using Archer FusionPlex assay as described previously . This analysis revealed PAX3 (exon7) ::MAML3 (exon2) fusion which was confirmed by FISH, using MAML3 (4q31.1 ) and PAX3 (2q36.1) break-apart probes (both from SureFISH, Agilent). Of note, the FISH analysis confirmed the presence of the rearrangement (with a cut-off defined as at least 10% cells with a break) in both the conventional and high-grade components. Postoperatively, the patient received 5040 cGy in 28 fractions. Imaging studies 4 months after resection demonstrated recurrence along the right medial orbital wall and orbital floor. He was initiated on vincristine-urinotecan-temozolamide. Imaging after three cycles again showed local progression. Clinically, the tumor caused total vision loss in the right eye and started to protrude from the right nare. Due to continued progression on maximal therapy, the patient transitioned to hospice and died 15 months after his initial diagnosis.
BSNS with morphologically evident rhabdomyoblastic differentiation has been described in approximately 10% of cases . However, as the 2 largest studies have shown, at least focal immunohistochemical expression of desmin or MyoD1 is common, with the percentage of immunopositive cases ranging from 35 to 66% and 26 to 91%, respectively. Myogenin expression is the least frequent and is encountered in only 20% of cases . PAX7 expression in BSNS has not been extensively studied but was noted previously as well . The skeletal muscle phenotype of BSNS is in line with its molecular background, which is most commonly characterized by fusions of PAX3 with either MAML3 or (less commonly) with FOXO1 or NCOA1 . Very rarely, other fusions partners may be involved, including RREB::MRTFB ( MKL2 ) . However, the nosological classification of the latter as BSNS is somewhat controversial as identical gene fusions have been described in other head and neck mesenchymal tumors . During development, PAX3 determines the cell fate of melanocytic, neuronal, and skeletal muscle differentiation and regulates normal myogenesis and postnatal muscular regeneration , while MAML3 has been shown to function as a potent transactivator of PAX3 response elements . Gene expression profiling of BSNS with PAX3::MAML3 fusion also showed altered expression of several genes and signaling networks involved in neural crest, skeletal system, and general embryonic development, including the myogenic genes MYOCD and MYOD1 . Interestingly, the PAX3::FOXO1 and PAX3::NCOA1 fusions were also described in rare cases of alveolar RMS . It has been speculated that the differences in clinicopathological features between BSNS and alveolar RMS are probably determined by the cell of origin and cellular environment and by additional genetic aberrations . Nevertheless, the case we are presenting herein shows that a very small subset of BSNS may progress towards a high-grade RMS. After review of the literature, we found 2 recently reported cases of molecularly confirmed BSNS with high-grade transformation, their clinicopathological features together with our case are summarized in Table . From the available description and figures, it seems that both cases showed a high-grade spindle cell morphology without the rhabdoid features detected in our case. Unfortunately, the first case was not tested for skeletal muscle markers at all , and the other was tested only for desmin and myogenin, both of which were focally positive in the high-grade areas of this case, suggesting differentiation into spindle cell RMS . The differential diagnosis of ordinary BSNS includes a list of other neoplasms with uniform spindle cell morphology occurring in this anatomic area such as malignant peripheral nerve sheath tumor (including malignant Triton tumor when rhabdomyoblastic differentiation is present), cellular schwannoma, monophasic fibrous synovial sarcoma, sinonasal glomangiopericytoma, and solitary fibrous tumor. However, the distinction is usually possible using a carefully selected panel of immunohistochemical stains . In contrast, the rhabdomyosarcomatous component in our case could be mistaken for embryonal RMS or pleomorphic RMS. If this area is sampled without the conventional BSNS component (as happened in the initial probatory biopsy in our case), the distinction is impossible without molecular genetic methods. Since the clinical behavior and/or response to treatment of embryonal and pleomorphic RMS might differ from RMS arising from BSNS, we believe it is reasonable that molecular investigation of such cases in the sinonasal area with either PAX3 FISH probe or preferably using an adequate RNA-sequencing panel is carried out. Significant prognostic differences have already been noted between several molecular subgroups of spindle cell RMS. For example, spindle cell RMS with VGLL2 and VGLL3 fusions has a relatively favorable prognosis which is in contrast to the very aggressive subset harboring MYOD1 mutations . Importantly, spindle cell RMS with VGLL3 fusions and MYOD1 mutations has a predilection for the head and neck area . As the other published case of BSNS with high-grade transformation featured spindle cell RMS pattern, a comprehensive molecular investigation for this morphological variant of RMS is warranted as well. Lastly, the herein presented case emphasizes the importance of careful follow-up of patients with BSNS and a throughout sampling of every case to prevent a delayed detection of high-grade transformation. With regard to the case presented herein, it is interesting to note the significantly different expression of skeletal muscle markers in the conventional BSNS areas compared to the high-grade areas. Even though the former areas were diffusely positive for PAX7 and MyoD1, they were completely negative for desmin and myogenin. In contrast, the high-grade areas were moderately to diffusely positive for all these markers. Both PAX7 and MyoD1 regulate earlier stages of mammalian myogenesis than myogenin , and all these 3 transcriptional factors are expressed significantly earlier in myogenesis than the structural protein desmin . This suggests that the cells in conventional BSNS areas were arrested at earlier stages of myogenesis which would be also in line with a generally more frequent expression of MyoD1 compared to the expression of desmin and myogenin in the previous studies . In contrast, possibly due to additional molecular aberrations, the cells in the high-grade areas, although being highly anaplastic, were apparently able to differentiate further along the myogenic pathway as also evidenced by their rhabdoid morphology. In conclusion, we presented a unique case of BSNS with transformation into high-grade RMS which together with 2 previously published cases of BSNS with high-grade transformation helps to better understand this novel phenomenon. Although the risk for high-grade transformation of BSNS appears low, it has important clinical and diagnostic implications. Besides advocating for a careful follow-up of patients with BSNS and a throughout sampling of every case, we believe that molecular profiling of sinonasal RMS of any type is warranted.
|
Genetically determined cardiomyopathies at autopsy: the pivotal role of the pathologist in establishing the diagnosis and guiding family screening | 28599036-7fc7-440f-8fa1-ede1eaf32e07 | 10067659 | Forensic Medicine[mh] | Cardiomyopathies (CMP) comprise a heterogenous group of diseases affecting primarily the myocardium, with marked variation in clinical presentation, underlying cardiac pathophysiology and etiological background. These backgrounds can be either genetic and/or acquired diseases. As Association for European Cardiovascular Pathology (AECVP), we feel we need a document on autopsy diagnosis of inherited CMP because these entities have undergone major advances in the past 20 years and more precise autopsy diagnosis is essential to move forward with these advances. The complexity of the pathologic backgrounds requires proper insight and expertise, even more since several CMP definitions and diagnosis are still evolving with increased and rapid whole genome testing requiring expert input from many medical professionals. While many classification systems have been proposed in the clinical setting [ – ], and morphologic criteria for the pathological diagnosis of the various CMP are available , there is no internationally agreed pathological consensus concerning the diagnostic approach to genetically determined CMP at autopsy. Sudden cardiac death (SCD) is not the only opportunity for a pathologist to establish the true nosology of CMP from cardiac specimens. Cardiac transplantation is also a source of heart specimens. Although usually the diagnosis is already established clinically in this case, sometimes the pathologist can contribute to either alter or refine the diagnosis and the recommendations herein reported apply as well. In cases presenting with cardiac hypertrophy and/or dilatation/scarring with normal coronary arteries, a suspicion of CMP must be considered, and a histological examination is essential. Establishing the actual cause of the disease may require a number of tissue-based and/or fluid-based investigations, be it histological, ultrastructural, or molecular. A history of illicit drug use must be looked for. The process can also be complicated by the fact that information about disease during life, for example pathogenic genotype, family or drug history may not be available at the time of autopsy. SCD is frequently the first manifestation of disease in case of CMP, especially in the young. Also, during routine clinical or forensic autopsies at any age, a suspicion of CMP may arise based on clinical data or pathological findings at autopsy. It is thus a challenge to make a diagnosis of a potentially genetically determined CMP at autopsy. It is very important that the pathology report provides the relevant data and a cardiac diagnosis which can help the family in furthering investigations, including genetic testing in case of inherited forms of CMP [ – ]. With the explosion in molecular testing and the concept of the molecular autopsy, the pathologist should use strict criteria in the diagnosis of CMP, which should be helpful for clinical geneticists and cardiologists who advise the family as to the possibility of a genetic disease.
At present, there are three major clinical classifications for diagnosis of CMP, two of them commonly referred to as the American and the European classification, introduced by the American Heart Association (AHA) in 2006 and the European Society of Cardiology (ESC) in 2008 , respectively. Later on, the MOGE(S) classification forwarded by World Heart Federation (WHF) in 2013 attempted to provide a phenotype-genotype and etiology based nomenclature. These classifications deal with all the important aspects concerning variations in cardiac pathology and etiological backgrounds, including genetics, but with different approaches. All agree that the term CMP refers to a myocardial disorder in which the heart muscle is structurally and functionally abnormal, in the absence of coronary artery disease, hypertension, valvular disease and congenital heart disease sufficient to cause the observed myocardial abnormality. The AHA defined CMP as a heterogenous group of diseases of the myocardium associated with mechanical and/or electrical dysfunction which usually exhibit inappropriate ventricular hypertrophy and dilatation. Cardiomyopathies were divided into two major groups: primary CMP (which can have a genetic or nongenetic cause) where the pathology is predominantly limited to the heart muscle; and secondary CMP, which have cardiac involvement as part of a large variety of generalized systemic disorders (mostly acquired). The expert panel of the AHA states that CMP are due to a variety of diseases that are frequently genetic, and their classification is primarily based on pathophysiological mechanisms and genetic background . The ESC working group on myocardial and pericardial diseases defined CMP as a myocardial disorder in which the heart muscle is structurally and functionally abnormal in the absence of coronary artery disease, hypertension, valvular disease, and congenital heart disease . Their classification is based on specific morphological and functional phenotypes, i.e., dilated CMP (DCM), hypertrophic CMP (HCM), restrictive CMP (RCM), arrhythmogenic right ventricular CMP (ACM), and provisionally still unclassified CMP. Both documents allude to important distinct features of the various types, but limitations may come up since they mix up entities that are either based on anatomic features (i.e., HCM and DCM) and pathophysiologic features (i.e., RCM and ACM). In these classifications, the same condition can appear clinically in two categories such as cases with dilated and hypertrophic ventricles (mixed phenotype). Subjects with HCM may progress clinically to a dilated phenotype, due to extensive cardiac remodeling during the longstanding course of the disease, and this may be the phenotype seen at autopsy. It should be noted also that in the ESC classification, the so-called channelopathies, a heterogenous group of heritable cardiac diseases characterized by a still-increasing spectrum of ion channel gene mutations with variable but significant risk of arrhythmic SD, are not included because the hearts of these patients usually appear otherwise morphologically normal, clinically as well as at autopsy. The AHA classification, being primarily based on genetics, has included these channelopathies, as primary CMP . For the same reason also conduction defects such as the Lenègre disease (also known as progressive cardiac conduction defect) and sick sinus syndromes are included. The most recent classification from the WHF proposed a nomenclature system addressing five attributes of the CMP, i.e., morpho-functional characteristics (M), organ involvement (O), genetic or familial inheritance pattern (G), and etiological details (E), with the optional addition of functional class/status (S) (i.e., the so-called MOGE(S) classification) .
In the autopsy setting when pathologists suspect the existence of a CMP, either as the cause of death or as an incidental finding, they should follow a protocol that will establish the underlying phenotype of the disease, thus adopting the “morpho-functional” phenotype approach. This phenotypic diagnosis can be based on the pathological investigations alone (the full autopsy including microscopic investigations in the laboratory). However, in other cases, where a final conclusion on the nature of the disease cannot be made or still remains uncertain, the pathology report should include the advice to clinicians, cardiologists, and/or geneticists on how to proceed to a final diagnosis. The etiology of CMP is diverse and includes genetic, toxic, infectious, or autoimmune causes, for which the pathology findings need to be integrated by further investigations. If a genetic etiology is suspected, a sample will be provided to test for mutations in genes encoding for sarcomeric, cytoskeletal, nucleoskeletal, mitochondrial, desmosomal, ion channel, and calcium handling proteins which requires the expertise of multidisciplinary cardiogenetic centers (see below). Patho-morphologically, these mutations are most commonly associated with HCM, DCM and ACM phenotypes but overlapping phenotypes do exist . This implies that the most practical approach to a suspected CMP at autopsy will be to apply the ESC classification, which is based on initial grouping of morphological and functional phenotypes, and to follow the proposed flowchart towards sub-classification in familial/genetic forms and nonfamilial diseases with the exclusion of ischemic, valvar, and hypertensive heart disease. Basically, this is the method proposed by Michael J Davies in 1995 and then updated [ – ]. The workup presented in this article aims to follow this approach. Many CMP patients die of SCD, indicating that pathologists are often in the first line to evaluate a cardiac diagnosis with potential genetic background, which can be crucial for the family members of the deceased [ – ]. It is therefore important to perform an autopsy following the AECVP guidelines, applying the right methodology of dissection, tissue sampling for histology, toxicological and potential genetic analysis . Knowledge of the clinical history and death circumstances (when available) is of crucial importance to raise the possibility of an underlying CMP. A questionnaire distributed to the family of the deceased and/or a team of specifically trained professionals such as nurses could help in obtaining a suitably detailed history from family members. A previous history of cardiac disease and medications or procedures will be vital. Previous episodes of syncope or pre-syncope, palpitations, or chest pain can be cardiac related. Where there is a possible history of CMP, details concerning this should be obtained and we should proceed with the autopsy with this diagnosis in mind. Also, a family history of SD, sudden infant deaths, unexplained drownings, or accidents will make an underlying genetic cause a possibility. When a patient dies suddenly with no previous history, extra-cardiac causes of death must be excluded. After a careful examination, a significant number of hearts from patients with SCD can appear structurally normal at autopsy (sudden arrhythmic death syndrome-SADS) [ – ]. Such a finding might underlie the heterogenic group of (familial) ion channel mutation related diseases which include the long QT, short QT, and Brugada syndromes and polymorphic catecholaminergic ventricular tachycardia. As mentioned above, they are not included in the ESC classification of CMP and cannot be diagnosed at autopsy due to lack of structural changes. Therefore, the identification of a structurally normal heart in case of arrhythmic SCD is of equal importance as it does not exclude the presence of a genetically determined and potentially hereditary cardiac disease and a further evaluation with a multidisciplinary team approach is needed. Clear-cut criteria to define a normal heart are crucial [ , , – ]. Obviously, a detailed drug history is important and use of cocaine, steroids and methamphetamine sought as these can cause both hypertrophy and dilatation of the heart. Cases with uncertain pathology findings and drug concentration below the lethal threshold should be carefully interpreted [ , , ]. The finding of a pathogenic or probably pathogenic mutation in genes associated with CMP in cases with normal or near normal heart pathology raises the possibility of a SCD in an electrically unstable “early phase” of the disease, without enough structural disease to receive a postmortem diagnosis of CMP. Since still a large proportion of SADS victims exhibits no mutation in ion channel genes, these cases must prompt a comprehensive cardiological family screening looking for channelopathies as well as for CMP features in at risk relatives and an appropriate cascade genetic study should be offered if the CMP-related mutation is considered pathogenic/probably pathogenic.
The pathologist should examine the heart according to a protocol already set out in guidelines on the investigation of SCD and also on cardiac hypertrophy . Once diseases of coronary arteries, great vessels and valves are excluded, then the pathologist must suspect that the morphological cause could be within the myocardium. No significant coronary artery disease and hypertension are very common especially in older patients and should not be given as the cause of death without performing cardiac histology. Histological sampling of both ventricles including the interventricular septum (IVS) must be done, usually at mid ventricular level as previously reported. The pathologist should be aware that a cardiac diagnosis made during life may not necessarily be in agreement with the autopsy findings in the heart. When another cause of death is found, but the heart looks potentially “cardiomyopathic,” tissue sampling of the heart is advised as indicated and material for genetic testing should be taken (if possible, blood in EDTA tube or a small piece of fresh spleen). In some medicolegal settings, this may be discouraged in view of costs but we believe the pathologist has a duty of care to inform the coroner/ district attorney and family of this possibility. Due to widespread occurrence of coronary artery disease in the population, co-occurrence of obstructive atherosclerotic plaques in the coronary arteries of hearts with CMP is not uncommon, particularly in the older age groups. Pathologists should be aware of possible co-existence of highly prevalent acquired disease in these patients in order to avoid misdiagnosis or under-diagnosis. The same is also true for hypertension, especially in males over 40 years As recommended in the guidelines for autopsy investigation of SCD, in the absence of local expertise in cardiovascular pathology, the best practice is that the entire heart is retained and sent to a specialized center for an expert opinion . The referring pathologist should complete the initial steps, including making a transverse apical section of the heart and emptying the heart of blood before referral. If the heart cannot be retained, it is essential that extensive photographic documentation is made, indicating where individual blocks are taken. At least one transverse section of the heart including the LV and RV should be retained for further examination.
The following paragraphs will address the more distinctive gross and histologic features of the four main genetically determined CMP phenotypes (see also Table and Fig. ). For definitions of terms, please refer to Table from Basso et al. . Hypertrophic cardiomyopathy At autopsy, there is usually a clear and appropriate cause for left ventricular (LV) hypertrophy such as aortic stenosis or hypertension . When there is no clear cause then HCM should be considered and the diagnosis is often made first at autopsy . Obviously, if the pathologist is already aware of the diagnosis, confirmation should be obtained at autopsy. According to ESC guidelines on HCM, the definition of HCM applies to children and adults and makes no a priori assumptions about etiology or myocardial pathology . HCM occurs in approximately 1:500 of the general population . It is the most common inherited monogenic cardiac condition. Most HCM have an autosomal dominant pattern of inheritance caused by mutations in genes that encode different proteins of the cardiac sarcomere. Currently, there are over 500 known mutations in 11 genes (beta-myosin heavy chain; cardiac myosin-binding protein C; cardiac troponin-T; troponin I; alpha-tropomyosin; essential and regulatory myosin light chains; actin; alpha-myosin heavy chain; titin; and muscle LIM protein). In most patients, HCM is caused by the first three genes while the other genes each account for only a small fraction. The diagnostic yield of genetic testing is 60–70% [ , , ]. Up to 10% of adult cases are due to other genetic/nongenetic disorders (phenocopies), such as metabolic disorders (Anderson-Fabry, Danon, and Pompe diseases), mitochondrial CMP, neuromuscular disease (Friedreich’s ataxia), malformation syndromes (Noonan, LEOPARD, and Costello syndromes), infiltrative disease (amyloidosis), and endocrine disorders (infants of mothers with diabetes, phaeochromocytoma, and acromegaly). Clinical history, metabolic studies, autopsy findings in other organs, histology, and genetic testing are essential in the diagnosis of these entities . Macroscopic changes in sarcomeric HCM (Fig. ) Heart size . The heart is usually enlarged with increase in transverse and longitudinal dimensions and may have obvious bulging of the LV externally. Heart weight is often but not necessarily increased. See normal values for age and body weight . Wall thickness . IVS and LV thickness are usually increased (see referral normal values) ranging from 15 to > 50 mm. Normal weight and wall thickness are reported in up to 30% of cases and particularly in patients with troponin T and some mutations of the myosin gene. Right ventricular (RV) wall thickness can be increased > 5 mm. Ventricular cavity size is usually reduced. LV cavity enlargement can be observed in advanced end stage forms. Atria . Left atrial (LA) cavity can be increased. Valves . The anterior leaflet of the mitral valve can be thickened due to friction lesion. There can be thickening with ballooning and elongation of both anterior and posterior mitral leaflet with elongation also of the attached cords . Papillary muscle hypertrophy and displacement may occur with insertion of the anterolateral papillary muscle directly into the mid-anterior mitral leaflet without intervening cords. Coronary arteries . Myocardial bridge, i.e., a deep intramyocardial course of the left anterior descending coronary artery, is reported in one fourth of HCM cases but is only significant if associated with ischemia or fibrosis in the anteroseptal wall of the LV . Myocardium . The majority of patients with sarcomeric protein gene mutations have an asymmetrical pattern of hypertrophy. Basal anterior IVS and anterior LV free wall are the most common location. Any pattern and distribution of LV wall thickening, including apical forms, as well as biventricular involvement can be observed. Bundle disarrangement/whorling can be seen macroscopically in areas of hypertrophy. Irregular pale scars can be seen usually in the IVS and the free wall, usually in the most hypertrophied regions. When transmural fibrosis is present, it could lead to wall thinning which can mimic DCM or chronic ischemic heart disease. Hypertrabeculation with increased thickness of the trabecular layer can occur with deep crypt formation. Aneurysms are rare and can occur usually in the apical forms of the disease. Endocardium . Impact/friction lesion is a patch of pale fibrotic endocardial thickening due to the impact of the anterior mitral valve leaflet on the upper IVS. This is associated with basal IVS hypertrophy and LV outflow tract obstruction but may not be present in all cases. Pericardium . No disease-related abnormalities. Microscopic changes in sarcomeric HCM Both hypertrophied and no hypertrophied areas should be sampled. The septal areas and outer free wall are the best areas to look for myocyte hypertrophy and disarray. Myocyte disarray is the most important diagnostic histological feature of HCM. One should avoid interpreting histology from RV as there is prominent trabeculation in this chamber and myocyte disarray is a normal feature around these trabecular regions. The same is for the anterior and posterior septal walls and their junction with the free wall and the subendocardium around the trabeculae as myocyte disarray and increased collagen will be seen here normally. Look also for myocyte hypertrophy with disorganized cellular architecture, myocardial scarring and expanded interstitial collagen. Cardiac myocytes . Hypertrophy is always present. The myocytes show bizarre shapes, nuclear pleomorphism and hyperchromasia. Myocyte disarray is always present. The hypertrophied cardiac myocytes show multiple intercellular connections, often arranged in chaotic alignment at oblique and perpendicular angles to each other or around capillaries as whorls of myocytes. Individual cellular and bundle disarray can be widely distributed, occupying substantial portions of the IVS and LV free wall [ , – ]. Look in areas away from significant replacement fibrosis to search for disarray. It is also common to find sections of myocardium which do not show myocyte hypertrophy or disarray even in hypertrophied areas macroscopically. Fibrosis . General increase in interstitial fibrosis is always present. Foci of loose connective tissue in the center of whorls of myocytes are classical findings . Replacement-type fibrosis is common and is found in the subendocardium as well as scattered large confluent areas in the IVS and free wall of the LV. This fibrosis is due to ischemic myocardial injury. Acute infarction or healing granulation tissue is also seen but is rare. Inflammatory cells . No disease-related inflammatory changes are reported. Intramural small vessels . Abnormal intramural coronary arteries, usually labelled as small vessel disease, characterized by thickened walls with increased intimal and medial collagen and narrowed lumen can be observed especially in areas with replacement-type fibrosis but this feature is not present in all cases [ – ]. Diagnostic challenges Myocyte disarray When the myocyte disarray is very focal, we would advise to repeat sampling of the LV including IVS and free wall which often will show more myocyte disarray to come to a definitive diagnosis. There will always be cases where the myocyte disarray is so subtle that a definite diagnosis cannot be made histologically even with extensive cardiac sampling at autopsy. In these cases, it is best to leave the diagnosis as inconclusive and possibility of HCM must be still included, and advise family screening with cascade genetic testing to solve the dilemma. Myocardial disarray in the RV is not diagnostic. Several systemic diseases (sometimes with isolated cardiac involvement) may present as HCM phenocopies (such as glycogen storage diseases, Fabry disease, and mitochondrial diseases). The histological findings of abnormal cardiomyocyte vacuolization and/or marked inflammation may raise this suspicion, especially if personal or family history include red flags of those entities, and prompt to ask for other diagnostics tests . The presence of accessory pathways should raise the suspicion of PRKAG2 and LAMP2-related HCM (see Basso et al. for more details) . In desmin myopathy, the disease is mainly expressed in skeletal muscle, but cardiac involvement does occur and rarely cardiac involvement is predominant. Myocytes contain eosinophilic conglomerates of fibrillary material staining by immunohistochemistry for desmin. Desmin accumulation may be difficult to be recognized with the light microscopy even at endomyocardial biopsy and the gold standard for the diagnosis of desmin accumulation remains the ultrastructural study . In children, HCM phenotype with disarray histologically is often associated with congenital syndromes, inherited metabolic disorders and neuromuscular diseases, unrelated to sarcomere gene mutations. In patients with systemic hypertension, extreme circumferential LV hypertrophy can be present. Histologically there is marked myocyte hypertrophy with separation of myocytes and expansion of the interstitium . Hypercontraction of the LV may occur in SD with decreased chamber size, but there will be no increase in heart weight and histologically there will hypercontracted myocytes but not myocyte disarray or fibrosis. Arrhythmogenic cardiomyopathy (Fig. ) The ACM phenotype is characterized by fibro-fatty or fibrous replacement of the subepicardial myocardium. Fibro-fatty infiltration with a subepicardial or mid-mural location in both RV and LV are therefore the most common reasons to suspect ACM at gross examination [ – ]. The process starts in the outer layers and extends toward the endocardium and can become transmural. Three variants are described depending on the ventricular chamber mostly affected: classical RV (ACM R), biventricular (ACM B), and LV dominant (ACM L) variant . The normal RV has prominent epicardial fat without fibrosis in the anterior and lateral wall while fat plus fibrosis in the posterobasal RV wall points to ACM R. In ACM L, epicardial fat with linear scarring is more pronounced in the posterobasal wall. ACM occurs in approximately 1:2000–1:5000 of the general population . It is a well-known cause of SCD in the young and athletes often as the first presentation of the disease. ACM phenotype is reported in childhood only in recessive variants such as Carvajal and Naxos syndromes. Genetic background It is mainly an inherited CMP, mostly with an autosomal dominant pattern of inheritance caused by pathogenic mutations in genes encoding structural proteins of intercellular junctions, mostly desmosomes, such as plakophilin ( PKP2 ), desmoplakin ( DSP ), desmoglein ( DSG2 ) and desmocollin ( DSC2 ), and rarely (< 1%) for the “area composita,” such as α-T-catenin ( CTNNA3 ) and N-cadherin ( CDH2 ). ACM-causing mutations have also been found in nondesmosomal genes such as phospholamban ( PLN ), filamin-C ( FLNC ), desmin ( DES ), titin ( TTN ), and lamin A/C ( LMNA ), which are associated with other CMP, such as DCM and neuromuscular CMPs, and may lead to overlapping phenotypes. Causal mutations in nondesmosomal genes, such as transmembrane protein 43 ( TMEM43 ), SCN5A, and transforming growth factor beta-3 ( TGFß-3 ) genes among the others, have been rarely identified. The diagnostic yield of genetic testing in ACM is approximately 50% . While the ACM-R is often linked to desmosomal gene mutations, ACM-L is more problematic as the phenotype of fibro-fatty replacement with prevalent fibrous tissue in the LV may also be the result of healed infarction, myocarditis or toxic agents so care is needed with family follow-up and genetic testing to help in the possible genetic etiology. Macroscopic changes Heart size . The heart can be normal in size or enlarged with increase in transverse and longitudinal size. At external view, yellow or whitish appearance can be found on the epicardial surface. Aneurysms, at autopsy usually present as local thinning of the ventricular wall, are typically seen in the classical ACM R variant, in the so-called triangle of dysplasia (RV inflow, outflow, and apex). Heart weight . The heart weight can be normal or increased. See normal values for age and body weight . Wall thickness . LV and IVS thickness can be either normal or decreased in ACM B or ACM L. RV wall thickness either normal or decreased in classical ACM R; pseudo-hypertrophy due to increase in epicardial fibro-fatty tissue is also described. Ventricular cavity size can be either normal or enlarged. Valves . ACM R with RV chamber dilatation and annulus enlargement, with or without sub-tricuspid RV inferior wall thinning, can underly tricuspid valve incompetence. Coronary arteries . No disease-related abnormalities. Atria . Right atrial (RA) enlargement is described in advanced ACM R with appendage thrombi. Myocardium . A normal appearance of the heart at macroscopic examination does not exclude ACM. Multiple cross sections of the hearts are required to rule out ACM at autopsy due to focal disease distribution. The classical ACM R is often characterized by transmural fibro-fatty replacement with residual myocardium only in the trabeculae, accounting for free wall thinning, translucency, and aneurysm formation. There can be compensatory hypertrophy of the residual myocardium in the trabeculae giving impression of hyper-trabeculation. The ACM L variant is characterized by preserved RV wall thickness and either normal size/increased thickness of LV. There can be enlarged LV cavity with thinned free wall particularly the posterobasal wall; compensatory hypertrophy of the trabeculae is reported in advanced forms; aneurysms are exceptional. The replacing fibrous or fibro-fatty tissue is usually confined to the outer layers (epicardial/midmural). Sometimes the abnormal subepicardial band is so thin that only a fine band-like depression is seen at the macroscopic examination and can easily be missed. The ACM B variants are characterized by both ACM R and ACM L features. The replacing tissue is mostly fibro-fatty tissue in the RV and fibrous tissue in the LV. There can be prominent epicardial fat particularly in the RV anterior and lateral wall in older often female obese patients, so histology is essential for differential diagnosis. Prominent epicardial fat especially infiltrating the myocardium is very unusual in the LV, so is more suspicious of the diagnosis of ACM L but again histology is essential. The IVS is rarely involved and if so, it is mostly affected at its right side. Circumferential LV involvement suggests FLNC or DSP mutations as first genotype options . Endocardium . Fibrous thickening with white appearance is described in advanced stages, with or without mural thrombosis. Pericardium . No disease-related abnormalities. Microscopic changes Both normal and abnormal areas should be sampled. Look for replacing fibrous tissue and fibro-fatty tissue and cardiac myocyte degenerative features. Cardiac myocytes . Histological features can vary from mild to severe degenerative changes with nuclear abnormalities and cytoplasmic vacuolation to cell death. Fibrosis . ACM is characterized by replacement fibrosis which starts in the epicardium or mid-mural layers. Both replacement and interstitial fibrosis can be observed. In the RV it is admixed with fat. In the LV, it may be mainly interstitial and replacement fibrosis with little fat. Endocardial fibrosis can be present. Fibro-fatty tissue replacement in the RV is always present, at the intersection of the free wall with epicardial fat; in the LV, it can be present but usually there is more fibrous tissue. Isolated fatty tissue without fibrosis is not a diagnostic feature of ACM . Inflammatory cells . Focal sparse lympho-monocyte infiltrates can be present similarly to DCM. Sometimes diffuse inflammatory infiltrates preferentially affecting the LV in the subepicardial and/or mid-mural layers are observed which makes difficult or impossible the differential diagnosis between ACM (acute/hot phase) and myocarditis . Granulation tissue and loose connective tissue have been also described in subacute stages of disease evolution. Intramural small vessels . No disease-related abnormalities. Diagnostic challenges and mimics An increased amount of epicardial fat ( adipositas cordis ) and isolated fatty infiltration in the RV are different entities and should not be confused with ACM. They can be observed in obese people, people with metabolic diseases, elderly or even normal people, with or without increased epicardial fat, particularly in the anterior and lateral RV free walls . Fat also accompanies the intramural vessels in both ventricles. A certain amount of fatty tissue is always present within the myocardium of the RV free wall so that strands of myocardial fibers are separated by fatty tissue, without cardiac myocytes abnormalities and replacement-type fibrosis. The boundary between the myocardium and the outer epicardial fat is usually distinct. Focal areas of endocardial fat with no fibrosis are also a nonspecific finding that should not led to a diagnosis of ACM. Myocardial fibrosis, with a nonischemic distribution in the LV free wall, either patchy or confluent in the mid-mural or sub-epicardial layers, can be due to a previous myocarditis, toxic insult or infarction. A descriptive pathology report of this to include all possibilities should be provided and advise molecular autopsy with cardiogenetic screening of first degree family members . Differential diagnosis of ACM also includes DCM, isolated cardiac sarcoidosis, Chagas diseases, previous myocardial infarction, muscular dystrophies, and chronic effects of drug toxicity. Abnormal pulmonary venous drainages or other congenital heart defect with RV overload and dilatation should be excluded. Histological features and/or clinical history and other cardiac or extracardiac findings are of help in differential diagnosis. Dilated cardiomyopathy (Fig. ) It is the most common CMP and includes genetic and nongenetic forms . Its prevalence has been recently estimated in the range of 1:250, and the percentage of genetically determined forms ranges from 30 to 50%. The presence of cardiomegaly with increased cardiac mass and LV or biventricular dilatation with decreased mass/volume ratio, in the absence of coronary artery and valve diseases and hypertension, should led to the suspicion of DCM. Familial DCM accounts for 30–50% of DCM. In patients with familial DCM, approximately 40–50% has an identifiable genetic cause . The genetic background of DCM is wide and complex with pathogenic mutations in more than 50 genes encoding for cytoskeleton, sarcomeric proteins, sarcolemma, nuclear envelope, ion channels, and intercellular junctions. Although mutations in titin remain the most common identifiable cause, there is growing evidence for mutations in lamin A, desmosomal genes, and filamin C gene underlying an arrhythmogenic form of DCM. X-linked, autosomal recessive, and mitochondrial inheritance of DCM are less common. Phospholamban (PLN) CMP has clinical and histological characteristics of both DCM and ACM. Acquired causes of DCM include alcohol, pregnancy, post-myocarditis, hemochromatosis, chronic anemia, anthracycline medications, sarcoidosis, stimulant drugs (e.g., cocaine), etc. Idiopathic DCM is eventually a diagnosis of exclusion. The recent demonstration that up to 20% of DCM patients with an established acquired risk factor or a nonfamilial disease still carries a pathogenic gene variant suggests a broader role for genetic testing in DCM . From the pathological viewpoint, there is no evidence that particular patterns or amount of fibrosis and cardiac myocyte changes will favor either acquired or genetic cause. A diagnosis of DCM should always trigger familial investigation and genetic testing for an etiologic diagnosis. Macroscopic changes Heart size. The heart is enlarged with increased transverse size . Heart . The heart weight is usually increased. See normal values for age and body weight . Wall thickness . IVS and LV wall thickness are usually normal values or even decreased due to wall thinning (see referral normal values) . RV wall thickness is normal or decreased. Ventricular cavity size . LV cavity enlargement is typically observed, internal diameter can help for qualitative assessment (eccentric hypertrophy) . RV cavity can be also enlarged. If chamber dilatation is the only finding especially with delayed autopsy, one should look for other macroscopic and microscopic findings to support the suspicion of DCM. Valves . DCM with chamber dilatation and mitro-tricuspid annulus enlargement could explain the in vivo finding of functional regurgitation. Mild increase in leaflet thickening can be observed. Coronary arteries . No disease-related abnormalities. Atria . RA and LA cavity can be dilated. Endocardial thrombosis can be present, particularly in the appendages. Myocardium . Fibrous scar may be present, of variable size and distribution typically with a nonischemic pattern. Sometimes distinctive patterns, such as midwall fibrosis particularly in the IVS, are visible. Hypertrabeculation can occur in advanced forms with wall thinning and prominent cavity dilatation. Endocardium . Mural thrombosis is frequent due to blood stagnation in the poorly contracting LV. With time organizing thrombosis can lead to whitish appearance of the endocardium with fibrous thickening. Endocardial fibrosis can also be observed in the absence of previous thrombus. Fibroelastosis of the LV endocardium is common in the first years of life, usually associated with viral infection or congenital anomalies . Pericardium . Usually no disease-related abnormalities. The presence of pericardial adhesion could indicate that DCM is a result of previous myo-pericarditis. Microscopic changes The histologic abnormalities are nonspecific. Cardiac myocytes . Histological features can vary from normal to slight to marked degenerative changes with increased cell diameter, nuclear abnormalities and cytoplasmic attenuation with perinuclear halo. Hyperchromatic and asymmetric nuclei are often observed. Fibrosis . Interstitial fibrosis of variable degree is present; replacement-type fibrosis can be present with either patchy or confluent non-ischemic distribution. Inflammatory cells . Focal sparse lympho-monocyte infiltrates can be present. Intramural small vessel . No disease-related abnormalities. Diagnostic challenges and mimics Differential diagnosis of DCM includes myocarditis, isolated cardiac sarcoidosis, end-stage HCM, ACM, muscular dystrophies, and chronic ischemic heart disease with LV remodeling. Specific histological features and/or clinical history and other cardiac or extracardiac findings are of help in differential diagnosis. Pathologists should take in mind that chamber dilatation can be observed as an artefact of postmortem autolytic changes as already stated. Therefore, if chamber dilatation is the only finding, one should look for other macroscopic and microscopic findings that support the diagnosis of DCM. Restrictive cardiomyopathy Primary restrictive CMP (RCM) is a poorly recognized entity characterized by nondilated, nonhypertrophied ventricles with dilated atria. The heart may be more rigid with cutting. In comparison with other forms of CMP, RCM is uncommon and a rare a cause of SCD. The pathology diagnosis is challenging in the absence of clinical findings which are essential to support a certain diagnosis of RCM . Many of the RCM probands had pathologic mutations in either beta-myosin heavy chain ( MYH7 ) or the cardiac troponin I gene ( TNNI3 ) . Mutations in other sarcomeric genes including troponin T ( TTNT2 ), myosin-binding protein C ( MYBPC3 ), myosin light chains ( MYL 2 and 3 ), and alpha cardiac actin ( ACTC ) have also been described. RCM can coexist with HCM in the same family. Most sarcomeric RCM mutations appear to be de novo and associated with a severe disease expression and an early onset. Nonsarcomeric mutations have also been identified in RCM and include mutations in myopalladin ( MYPN ), titin ( TTN ), and filamin-C (FLNC) . Genetic causes cannot be established from macroscopic or microscopic features and family investigation with appropriate genetic testing is essential. Pathologists should be aware that there are several other causes of RCM, including infiltrative diseases (amyloidosis in which the amyloid may be mistaken for fibrosis), storage diseases, mitochondrial disease and a variety of systemic diseases such as hemochromatosis. In these cases, there may be overlap with the other CMP phenotypes such as HCM. Loeffler endocarditis and endomyocardial fibrosis can also present with a restrictive pattern. Histological features and/or clinical history and other cardiac or extracardiac findings are essential in making these diagnoses. Interstitial fibrosis with or without myocyte disarray, without evidence of storage or infiltrative disease, are highly suggestive histologic features of primary RCM. Macroscopic changes Heart size . The heart is usually normal in size with normal transverse and longitudinal size. Heart weight usually normal. See normal values for age and body weight . Wall thickness . IVS, LV, and RV wall thickness usually normal values (see referral normal values). Ventricular cavity size . Normal. Valves . No disease-related abnormalities. Coronary arteries . No disease-related abnormalities. Atria . RA and LA cavity can be dilated. Endocardial thrombosis can be present, particularly in the appendages . Myocardium . No disease-related abnormalities. Endocardium . No disease-related abnormalities. Pericardium . No disease-related abnormalities. Microscopic changes Cardiac myocytes . They are usually of normal size and morphology or may show degenerative changes. Myocyte disarray can be observed as part of HCM spectrum . Fibrosis . Interstitial fibrosis is present in all cases. Replacement-type fibrosis can be present. Inflammatory cells . No disease-related changes are reported. Intramural small vessel . No disease-related abnormalities. Uncertain categories/findings The entity reported as noncompaction CMP, which is also known as LV noncompaction (LVNC), is characterized by excessive trabeculations of the LV and deep intertrabecular recesses communicating with the ventricular cavity, with a > twofold thickening of the endocardial noncompacted (NC) layer compared with the epicardial compacted (C) layer of the myocardium (NC/C > 2) . LVNC has been considered a genetic CMP by the AHA, and an unclassified CMP by the ESC. This phenotype is often associated with congenital heart diseases or other CMP, in which genetic analysis follows the patterns of the underlying genetic CMP. There is evidence for wide genetic heterogeneity in LVNC with mutations in sarcomeric, cytoskeletal, Z-line, and mitochondrial proteins. Therefore, it is controversial whether to consider it as a separate entity and it is very rarely encountered in autopsy practice . There is a risk of misnaming a CMP as LVNC when the underlying genetic cause is in fact a HCM, DCM or ACM. We consider LVNC as a phenotypic trait rather than a CMP, that can occur either in isolation or in association with other either congenital heart defects or CMP. An entity labelled idiopathic LV fibrosis has been put forward as a cause of SCD especially during sport activities . The heart is normal in weight, both ventricles are not dilated and there is no hypertrophy or thinning of the ventricular walls. Fibrosis may be macroscopically visible or not. Microscopically there is significant replacement fibrosis in the LV wall with no myocyte disarray. The distribution of the fibrosis is focal and does not have the patterns seen in other CMP from the histological point of view. This entity, which is also called non ischemic scarring, could be an acquired condition due to healed myocarditis, toxic or drug related but may also be a genetic CMP as familial cases are reported. Thus, the finding of no specific LV scarring should rise the possibility of an inherited CMP in which the phenotype is not well defined.
At autopsy, there is usually a clear and appropriate cause for left ventricular (LV) hypertrophy such as aortic stenosis or hypertension . When there is no clear cause then HCM should be considered and the diagnosis is often made first at autopsy . Obviously, if the pathologist is already aware of the diagnosis, confirmation should be obtained at autopsy. According to ESC guidelines on HCM, the definition of HCM applies to children and adults and makes no a priori assumptions about etiology or myocardial pathology . HCM occurs in approximately 1:500 of the general population . It is the most common inherited monogenic cardiac condition. Most HCM have an autosomal dominant pattern of inheritance caused by mutations in genes that encode different proteins of the cardiac sarcomere. Currently, there are over 500 known mutations in 11 genes (beta-myosin heavy chain; cardiac myosin-binding protein C; cardiac troponin-T; troponin I; alpha-tropomyosin; essential and regulatory myosin light chains; actin; alpha-myosin heavy chain; titin; and muscle LIM protein). In most patients, HCM is caused by the first three genes while the other genes each account for only a small fraction. The diagnostic yield of genetic testing is 60–70% [ , , ]. Up to 10% of adult cases are due to other genetic/nongenetic disorders (phenocopies), such as metabolic disorders (Anderson-Fabry, Danon, and Pompe diseases), mitochondrial CMP, neuromuscular disease (Friedreich’s ataxia), malformation syndromes (Noonan, LEOPARD, and Costello syndromes), infiltrative disease (amyloidosis), and endocrine disorders (infants of mothers with diabetes, phaeochromocytoma, and acromegaly). Clinical history, metabolic studies, autopsy findings in other organs, histology, and genetic testing are essential in the diagnosis of these entities . Macroscopic changes in sarcomeric HCM (Fig. ) Heart size . The heart is usually enlarged with increase in transverse and longitudinal dimensions and may have obvious bulging of the LV externally. Heart weight is often but not necessarily increased. See normal values for age and body weight . Wall thickness . IVS and LV thickness are usually increased (see referral normal values) ranging from 15 to > 50 mm. Normal weight and wall thickness are reported in up to 30% of cases and particularly in patients with troponin T and some mutations of the myosin gene. Right ventricular (RV) wall thickness can be increased > 5 mm. Ventricular cavity size is usually reduced. LV cavity enlargement can be observed in advanced end stage forms. Atria . Left atrial (LA) cavity can be increased. Valves . The anterior leaflet of the mitral valve can be thickened due to friction lesion. There can be thickening with ballooning and elongation of both anterior and posterior mitral leaflet with elongation also of the attached cords . Papillary muscle hypertrophy and displacement may occur with insertion of the anterolateral papillary muscle directly into the mid-anterior mitral leaflet without intervening cords. Coronary arteries . Myocardial bridge, i.e., a deep intramyocardial course of the left anterior descending coronary artery, is reported in one fourth of HCM cases but is only significant if associated with ischemia or fibrosis in the anteroseptal wall of the LV . Myocardium . The majority of patients with sarcomeric protein gene mutations have an asymmetrical pattern of hypertrophy. Basal anterior IVS and anterior LV free wall are the most common location. Any pattern and distribution of LV wall thickening, including apical forms, as well as biventricular involvement can be observed. Bundle disarrangement/whorling can be seen macroscopically in areas of hypertrophy. Irregular pale scars can be seen usually in the IVS and the free wall, usually in the most hypertrophied regions. When transmural fibrosis is present, it could lead to wall thinning which can mimic DCM or chronic ischemic heart disease. Hypertrabeculation with increased thickness of the trabecular layer can occur with deep crypt formation. Aneurysms are rare and can occur usually in the apical forms of the disease. Endocardium . Impact/friction lesion is a patch of pale fibrotic endocardial thickening due to the impact of the anterior mitral valve leaflet on the upper IVS. This is associated with basal IVS hypertrophy and LV outflow tract obstruction but may not be present in all cases. Pericardium . No disease-related abnormalities. Microscopic changes in sarcomeric HCM Both hypertrophied and no hypertrophied areas should be sampled. The septal areas and outer free wall are the best areas to look for myocyte hypertrophy and disarray. Myocyte disarray is the most important diagnostic histological feature of HCM. One should avoid interpreting histology from RV as there is prominent trabeculation in this chamber and myocyte disarray is a normal feature around these trabecular regions. The same is for the anterior and posterior septal walls and their junction with the free wall and the subendocardium around the trabeculae as myocyte disarray and increased collagen will be seen here normally. Look also for myocyte hypertrophy with disorganized cellular architecture, myocardial scarring and expanded interstitial collagen. Cardiac myocytes . Hypertrophy is always present. The myocytes show bizarre shapes, nuclear pleomorphism and hyperchromasia. Myocyte disarray is always present. The hypertrophied cardiac myocytes show multiple intercellular connections, often arranged in chaotic alignment at oblique and perpendicular angles to each other or around capillaries as whorls of myocytes. Individual cellular and bundle disarray can be widely distributed, occupying substantial portions of the IVS and LV free wall [ , – ]. Look in areas away from significant replacement fibrosis to search for disarray. It is also common to find sections of myocardium which do not show myocyte hypertrophy or disarray even in hypertrophied areas macroscopically. Fibrosis . General increase in interstitial fibrosis is always present. Foci of loose connective tissue in the center of whorls of myocytes are classical findings . Replacement-type fibrosis is common and is found in the subendocardium as well as scattered large confluent areas in the IVS and free wall of the LV. This fibrosis is due to ischemic myocardial injury. Acute infarction or healing granulation tissue is also seen but is rare. Inflammatory cells . No disease-related inflammatory changes are reported. Intramural small vessels . Abnormal intramural coronary arteries, usually labelled as small vessel disease, characterized by thickened walls with increased intimal and medial collagen and narrowed lumen can be observed especially in areas with replacement-type fibrosis but this feature is not present in all cases [ – ]. Diagnostic challenges Myocyte disarray When the myocyte disarray is very focal, we would advise to repeat sampling of the LV including IVS and free wall which often will show more myocyte disarray to come to a definitive diagnosis. There will always be cases where the myocyte disarray is so subtle that a definite diagnosis cannot be made histologically even with extensive cardiac sampling at autopsy. In these cases, it is best to leave the diagnosis as inconclusive and possibility of HCM must be still included, and advise family screening with cascade genetic testing to solve the dilemma. Myocardial disarray in the RV is not diagnostic. Several systemic diseases (sometimes with isolated cardiac involvement) may present as HCM phenocopies (such as glycogen storage diseases, Fabry disease, and mitochondrial diseases). The histological findings of abnormal cardiomyocyte vacuolization and/or marked inflammation may raise this suspicion, especially if personal or family history include red flags of those entities, and prompt to ask for other diagnostics tests . The presence of accessory pathways should raise the suspicion of PRKAG2 and LAMP2-related HCM (see Basso et al. for more details) . In desmin myopathy, the disease is mainly expressed in skeletal muscle, but cardiac involvement does occur and rarely cardiac involvement is predominant. Myocytes contain eosinophilic conglomerates of fibrillary material staining by immunohistochemistry for desmin. Desmin accumulation may be difficult to be recognized with the light microscopy even at endomyocardial biopsy and the gold standard for the diagnosis of desmin accumulation remains the ultrastructural study . In children, HCM phenotype with disarray histologically is often associated with congenital syndromes, inherited metabolic disorders and neuromuscular diseases, unrelated to sarcomere gene mutations. In patients with systemic hypertension, extreme circumferential LV hypertrophy can be present. Histologically there is marked myocyte hypertrophy with separation of myocytes and expansion of the interstitium . Hypercontraction of the LV may occur in SD with decreased chamber size, but there will be no increase in heart weight and histologically there will hypercontracted myocytes but not myocyte disarray or fibrosis.
) Heart size . The heart is usually enlarged with increase in transverse and longitudinal dimensions and may have obvious bulging of the LV externally. Heart weight is often but not necessarily increased. See normal values for age and body weight . Wall thickness . IVS and LV thickness are usually increased (see referral normal values) ranging from 15 to > 50 mm. Normal weight and wall thickness are reported in up to 30% of cases and particularly in patients with troponin T and some mutations of the myosin gene. Right ventricular (RV) wall thickness can be increased > 5 mm. Ventricular cavity size is usually reduced. LV cavity enlargement can be observed in advanced end stage forms. Atria . Left atrial (LA) cavity can be increased. Valves . The anterior leaflet of the mitral valve can be thickened due to friction lesion. There can be thickening with ballooning and elongation of both anterior and posterior mitral leaflet with elongation also of the attached cords . Papillary muscle hypertrophy and displacement may occur with insertion of the anterolateral papillary muscle directly into the mid-anterior mitral leaflet without intervening cords. Coronary arteries . Myocardial bridge, i.e., a deep intramyocardial course of the left anterior descending coronary artery, is reported in one fourth of HCM cases but is only significant if associated with ischemia or fibrosis in the anteroseptal wall of the LV . Myocardium . The majority of patients with sarcomeric protein gene mutations have an asymmetrical pattern of hypertrophy. Basal anterior IVS and anterior LV free wall are the most common location. Any pattern and distribution of LV wall thickening, including apical forms, as well as biventricular involvement can be observed. Bundle disarrangement/whorling can be seen macroscopically in areas of hypertrophy. Irregular pale scars can be seen usually in the IVS and the free wall, usually in the most hypertrophied regions. When transmural fibrosis is present, it could lead to wall thinning which can mimic DCM or chronic ischemic heart disease. Hypertrabeculation with increased thickness of the trabecular layer can occur with deep crypt formation. Aneurysms are rare and can occur usually in the apical forms of the disease. Endocardium . Impact/friction lesion is a patch of pale fibrotic endocardial thickening due to the impact of the anterior mitral valve leaflet on the upper IVS. This is associated with basal IVS hypertrophy and LV outflow tract obstruction but may not be present in all cases. Pericardium . No disease-related abnormalities.
Both hypertrophied and no hypertrophied areas should be sampled. The septal areas and outer free wall are the best areas to look for myocyte hypertrophy and disarray. Myocyte disarray is the most important diagnostic histological feature of HCM. One should avoid interpreting histology from RV as there is prominent trabeculation in this chamber and myocyte disarray is a normal feature around these trabecular regions. The same is for the anterior and posterior septal walls and their junction with the free wall and the subendocardium around the trabeculae as myocyte disarray and increased collagen will be seen here normally. Look also for myocyte hypertrophy with disorganized cellular architecture, myocardial scarring and expanded interstitial collagen. Cardiac myocytes . Hypertrophy is always present. The myocytes show bizarre shapes, nuclear pleomorphism and hyperchromasia. Myocyte disarray is always present. The hypertrophied cardiac myocytes show multiple intercellular connections, often arranged in chaotic alignment at oblique and perpendicular angles to each other or around capillaries as whorls of myocytes. Individual cellular and bundle disarray can be widely distributed, occupying substantial portions of the IVS and LV free wall [ , – ]. Look in areas away from significant replacement fibrosis to search for disarray. It is also common to find sections of myocardium which do not show myocyte hypertrophy or disarray even in hypertrophied areas macroscopically. Fibrosis . General increase in interstitial fibrosis is always present. Foci of loose connective tissue in the center of whorls of myocytes are classical findings . Replacement-type fibrosis is common and is found in the subendocardium as well as scattered large confluent areas in the IVS and free wall of the LV. This fibrosis is due to ischemic myocardial injury. Acute infarction or healing granulation tissue is also seen but is rare. Inflammatory cells . No disease-related inflammatory changes are reported. Intramural small vessels . Abnormal intramural coronary arteries, usually labelled as small vessel disease, characterized by thickened walls with increased intimal and medial collagen and narrowed lumen can be observed especially in areas with replacement-type fibrosis but this feature is not present in all cases [ – ].
Myocyte disarray When the myocyte disarray is very focal, we would advise to repeat sampling of the LV including IVS and free wall which often will show more myocyte disarray to come to a definitive diagnosis. There will always be cases where the myocyte disarray is so subtle that a definite diagnosis cannot be made histologically even with extensive cardiac sampling at autopsy. In these cases, it is best to leave the diagnosis as inconclusive and possibility of HCM must be still included, and advise family screening with cascade genetic testing to solve the dilemma. Myocardial disarray in the RV is not diagnostic. Several systemic diseases (sometimes with isolated cardiac involvement) may present as HCM phenocopies (such as glycogen storage diseases, Fabry disease, and mitochondrial diseases). The histological findings of abnormal cardiomyocyte vacuolization and/or marked inflammation may raise this suspicion, especially if personal or family history include red flags of those entities, and prompt to ask for other diagnostics tests . The presence of accessory pathways should raise the suspicion of PRKAG2 and LAMP2-related HCM (see Basso et al. for more details) . In desmin myopathy, the disease is mainly expressed in skeletal muscle, but cardiac involvement does occur and rarely cardiac involvement is predominant. Myocytes contain eosinophilic conglomerates of fibrillary material staining by immunohistochemistry for desmin. Desmin accumulation may be difficult to be recognized with the light microscopy even at endomyocardial biopsy and the gold standard for the diagnosis of desmin accumulation remains the ultrastructural study . In children, HCM phenotype with disarray histologically is often associated with congenital syndromes, inherited metabolic disorders and neuromuscular diseases, unrelated to sarcomere gene mutations. In patients with systemic hypertension, extreme circumferential LV hypertrophy can be present. Histologically there is marked myocyte hypertrophy with separation of myocytes and expansion of the interstitium . Hypercontraction of the LV may occur in SD with decreased chamber size, but there will be no increase in heart weight and histologically there will hypercontracted myocytes but not myocyte disarray or fibrosis.
When the myocyte disarray is very focal, we would advise to repeat sampling of the LV including IVS and free wall which often will show more myocyte disarray to come to a definitive diagnosis. There will always be cases where the myocyte disarray is so subtle that a definite diagnosis cannot be made histologically even with extensive cardiac sampling at autopsy. In these cases, it is best to leave the diagnosis as inconclusive and possibility of HCM must be still included, and advise family screening with cascade genetic testing to solve the dilemma. Myocardial disarray in the RV is not diagnostic. Several systemic diseases (sometimes with isolated cardiac involvement) may present as HCM phenocopies (such as glycogen storage diseases, Fabry disease, and mitochondrial diseases). The histological findings of abnormal cardiomyocyte vacuolization and/or marked inflammation may raise this suspicion, especially if personal or family history include red flags of those entities, and prompt to ask for other diagnostics tests . The presence of accessory pathways should raise the suspicion of PRKAG2 and LAMP2-related HCM (see Basso et al. for more details) . In desmin myopathy, the disease is mainly expressed in skeletal muscle, but cardiac involvement does occur and rarely cardiac involvement is predominant. Myocytes contain eosinophilic conglomerates of fibrillary material staining by immunohistochemistry for desmin. Desmin accumulation may be difficult to be recognized with the light microscopy even at endomyocardial biopsy and the gold standard for the diagnosis of desmin accumulation remains the ultrastructural study . In children, HCM phenotype with disarray histologically is often associated with congenital syndromes, inherited metabolic disorders and neuromuscular diseases, unrelated to sarcomere gene mutations. In patients with systemic hypertension, extreme circumferential LV hypertrophy can be present. Histologically there is marked myocyte hypertrophy with separation of myocytes and expansion of the interstitium . Hypercontraction of the LV may occur in SD with decreased chamber size, but there will be no increase in heart weight and histologically there will hypercontracted myocytes but not myocyte disarray or fibrosis.
) The ACM phenotype is characterized by fibro-fatty or fibrous replacement of the subepicardial myocardium. Fibro-fatty infiltration with a subepicardial or mid-mural location in both RV and LV are therefore the most common reasons to suspect ACM at gross examination [ – ]. The process starts in the outer layers and extends toward the endocardium and can become transmural. Three variants are described depending on the ventricular chamber mostly affected: classical RV (ACM R), biventricular (ACM B), and LV dominant (ACM L) variant . The normal RV has prominent epicardial fat without fibrosis in the anterior and lateral wall while fat plus fibrosis in the posterobasal RV wall points to ACM R. In ACM L, epicardial fat with linear scarring is more pronounced in the posterobasal wall. ACM occurs in approximately 1:2000–1:5000 of the general population . It is a well-known cause of SCD in the young and athletes often as the first presentation of the disease. ACM phenotype is reported in childhood only in recessive variants such as Carvajal and Naxos syndromes. Genetic background It is mainly an inherited CMP, mostly with an autosomal dominant pattern of inheritance caused by pathogenic mutations in genes encoding structural proteins of intercellular junctions, mostly desmosomes, such as plakophilin ( PKP2 ), desmoplakin ( DSP ), desmoglein ( DSG2 ) and desmocollin ( DSC2 ), and rarely (< 1%) for the “area composita,” such as α-T-catenin ( CTNNA3 ) and N-cadherin ( CDH2 ). ACM-causing mutations have also been found in nondesmosomal genes such as phospholamban ( PLN ), filamin-C ( FLNC ), desmin ( DES ), titin ( TTN ), and lamin A/C ( LMNA ), which are associated with other CMP, such as DCM and neuromuscular CMPs, and may lead to overlapping phenotypes. Causal mutations in nondesmosomal genes, such as transmembrane protein 43 ( TMEM43 ), SCN5A, and transforming growth factor beta-3 ( TGFß-3 ) genes among the others, have been rarely identified. The diagnostic yield of genetic testing in ACM is approximately 50% . While the ACM-R is often linked to desmosomal gene mutations, ACM-L is more problematic as the phenotype of fibro-fatty replacement with prevalent fibrous tissue in the LV may also be the result of healed infarction, myocarditis or toxic agents so care is needed with family follow-up and genetic testing to help in the possible genetic etiology. Macroscopic changes Heart size . The heart can be normal in size or enlarged with increase in transverse and longitudinal size. At external view, yellow or whitish appearance can be found on the epicardial surface. Aneurysms, at autopsy usually present as local thinning of the ventricular wall, are typically seen in the classical ACM R variant, in the so-called triangle of dysplasia (RV inflow, outflow, and apex). Heart weight . The heart weight can be normal or increased. See normal values for age and body weight . Wall thickness . LV and IVS thickness can be either normal or decreased in ACM B or ACM L. RV wall thickness either normal or decreased in classical ACM R; pseudo-hypertrophy due to increase in epicardial fibro-fatty tissue is also described. Ventricular cavity size can be either normal or enlarged. Valves . ACM R with RV chamber dilatation and annulus enlargement, with or without sub-tricuspid RV inferior wall thinning, can underly tricuspid valve incompetence. Coronary arteries . No disease-related abnormalities. Atria . Right atrial (RA) enlargement is described in advanced ACM R with appendage thrombi. Myocardium . A normal appearance of the heart at macroscopic examination does not exclude ACM. Multiple cross sections of the hearts are required to rule out ACM at autopsy due to focal disease distribution. The classical ACM R is often characterized by transmural fibro-fatty replacement with residual myocardium only in the trabeculae, accounting for free wall thinning, translucency, and aneurysm formation. There can be compensatory hypertrophy of the residual myocardium in the trabeculae giving impression of hyper-trabeculation. The ACM L variant is characterized by preserved RV wall thickness and either normal size/increased thickness of LV. There can be enlarged LV cavity with thinned free wall particularly the posterobasal wall; compensatory hypertrophy of the trabeculae is reported in advanced forms; aneurysms are exceptional. The replacing fibrous or fibro-fatty tissue is usually confined to the outer layers (epicardial/midmural). Sometimes the abnormal subepicardial band is so thin that only a fine band-like depression is seen at the macroscopic examination and can easily be missed. The ACM B variants are characterized by both ACM R and ACM L features. The replacing tissue is mostly fibro-fatty tissue in the RV and fibrous tissue in the LV. There can be prominent epicardial fat particularly in the RV anterior and lateral wall in older often female obese patients, so histology is essential for differential diagnosis. Prominent epicardial fat especially infiltrating the myocardium is very unusual in the LV, so is more suspicious of the diagnosis of ACM L but again histology is essential. The IVS is rarely involved and if so, it is mostly affected at its right side. Circumferential LV involvement suggests FLNC or DSP mutations as first genotype options . Endocardium . Fibrous thickening with white appearance is described in advanced stages, with or without mural thrombosis. Pericardium . No disease-related abnormalities. Microscopic changes Both normal and abnormal areas should be sampled. Look for replacing fibrous tissue and fibro-fatty tissue and cardiac myocyte degenerative features. Cardiac myocytes . Histological features can vary from mild to severe degenerative changes with nuclear abnormalities and cytoplasmic vacuolation to cell death. Fibrosis . ACM is characterized by replacement fibrosis which starts in the epicardium or mid-mural layers. Both replacement and interstitial fibrosis can be observed. In the RV it is admixed with fat. In the LV, it may be mainly interstitial and replacement fibrosis with little fat. Endocardial fibrosis can be present. Fibro-fatty tissue replacement in the RV is always present, at the intersection of the free wall with epicardial fat; in the LV, it can be present but usually there is more fibrous tissue. Isolated fatty tissue without fibrosis is not a diagnostic feature of ACM . Inflammatory cells . Focal sparse lympho-monocyte infiltrates can be present similarly to DCM. Sometimes diffuse inflammatory infiltrates preferentially affecting the LV in the subepicardial and/or mid-mural layers are observed which makes difficult or impossible the differential diagnosis between ACM (acute/hot phase) and myocarditis . Granulation tissue and loose connective tissue have been also described in subacute stages of disease evolution. Intramural small vessels . No disease-related abnormalities. Diagnostic challenges and mimics An increased amount of epicardial fat ( adipositas cordis ) and isolated fatty infiltration in the RV are different entities and should not be confused with ACM. They can be observed in obese people, people with metabolic diseases, elderly or even normal people, with or without increased epicardial fat, particularly in the anterior and lateral RV free walls . Fat also accompanies the intramural vessels in both ventricles. A certain amount of fatty tissue is always present within the myocardium of the RV free wall so that strands of myocardial fibers are separated by fatty tissue, without cardiac myocytes abnormalities and replacement-type fibrosis. The boundary between the myocardium and the outer epicardial fat is usually distinct. Focal areas of endocardial fat with no fibrosis are also a nonspecific finding that should not led to a diagnosis of ACM. Myocardial fibrosis, with a nonischemic distribution in the LV free wall, either patchy or confluent in the mid-mural or sub-epicardial layers, can be due to a previous myocarditis, toxic insult or infarction. A descriptive pathology report of this to include all possibilities should be provided and advise molecular autopsy with cardiogenetic screening of first degree family members . Differential diagnosis of ACM also includes DCM, isolated cardiac sarcoidosis, Chagas diseases, previous myocardial infarction, muscular dystrophies, and chronic effects of drug toxicity. Abnormal pulmonary venous drainages or other congenital heart defect with RV overload and dilatation should be excluded. Histological features and/or clinical history and other cardiac or extracardiac findings are of help in differential diagnosis.
It is mainly an inherited CMP, mostly with an autosomal dominant pattern of inheritance caused by pathogenic mutations in genes encoding structural proteins of intercellular junctions, mostly desmosomes, such as plakophilin ( PKP2 ), desmoplakin ( DSP ), desmoglein ( DSG2 ) and desmocollin ( DSC2 ), and rarely (< 1%) for the “area composita,” such as α-T-catenin ( CTNNA3 ) and N-cadherin ( CDH2 ). ACM-causing mutations have also been found in nondesmosomal genes such as phospholamban ( PLN ), filamin-C ( FLNC ), desmin ( DES ), titin ( TTN ), and lamin A/C ( LMNA ), which are associated with other CMP, such as DCM and neuromuscular CMPs, and may lead to overlapping phenotypes. Causal mutations in nondesmosomal genes, such as transmembrane protein 43 ( TMEM43 ), SCN5A, and transforming growth factor beta-3 ( TGFß-3 ) genes among the others, have been rarely identified. The diagnostic yield of genetic testing in ACM is approximately 50% . While the ACM-R is often linked to desmosomal gene mutations, ACM-L is more problematic as the phenotype of fibro-fatty replacement with prevalent fibrous tissue in the LV may also be the result of healed infarction, myocarditis or toxic agents so care is needed with family follow-up and genetic testing to help in the possible genetic etiology.
Heart size . The heart can be normal in size or enlarged with increase in transverse and longitudinal size. At external view, yellow or whitish appearance can be found on the epicardial surface. Aneurysms, at autopsy usually present as local thinning of the ventricular wall, are typically seen in the classical ACM R variant, in the so-called triangle of dysplasia (RV inflow, outflow, and apex). Heart weight . The heart weight can be normal or increased. See normal values for age and body weight . Wall thickness . LV and IVS thickness can be either normal or decreased in ACM B or ACM L. RV wall thickness either normal or decreased in classical ACM R; pseudo-hypertrophy due to increase in epicardial fibro-fatty tissue is also described. Ventricular cavity size can be either normal or enlarged. Valves . ACM R with RV chamber dilatation and annulus enlargement, with or without sub-tricuspid RV inferior wall thinning, can underly tricuspid valve incompetence. Coronary arteries . No disease-related abnormalities. Atria . Right atrial (RA) enlargement is described in advanced ACM R with appendage thrombi. Myocardium . A normal appearance of the heart at macroscopic examination does not exclude ACM. Multiple cross sections of the hearts are required to rule out ACM at autopsy due to focal disease distribution. The classical ACM R is often characterized by transmural fibro-fatty replacement with residual myocardium only in the trabeculae, accounting for free wall thinning, translucency, and aneurysm formation. There can be compensatory hypertrophy of the residual myocardium in the trabeculae giving impression of hyper-trabeculation. The ACM L variant is characterized by preserved RV wall thickness and either normal size/increased thickness of LV. There can be enlarged LV cavity with thinned free wall particularly the posterobasal wall; compensatory hypertrophy of the trabeculae is reported in advanced forms; aneurysms are exceptional. The replacing fibrous or fibro-fatty tissue is usually confined to the outer layers (epicardial/midmural). Sometimes the abnormal subepicardial band is so thin that only a fine band-like depression is seen at the macroscopic examination and can easily be missed. The ACM B variants are characterized by both ACM R and ACM L features. The replacing tissue is mostly fibro-fatty tissue in the RV and fibrous tissue in the LV. There can be prominent epicardial fat particularly in the RV anterior and lateral wall in older often female obese patients, so histology is essential for differential diagnosis. Prominent epicardial fat especially infiltrating the myocardium is very unusual in the LV, so is more suspicious of the diagnosis of ACM L but again histology is essential. The IVS is rarely involved and if so, it is mostly affected at its right side. Circumferential LV involvement suggests FLNC or DSP mutations as first genotype options . Endocardium . Fibrous thickening with white appearance is described in advanced stages, with or without mural thrombosis. Pericardium . No disease-related abnormalities.
Both normal and abnormal areas should be sampled. Look for replacing fibrous tissue and fibro-fatty tissue and cardiac myocyte degenerative features. Cardiac myocytes . Histological features can vary from mild to severe degenerative changes with nuclear abnormalities and cytoplasmic vacuolation to cell death. Fibrosis . ACM is characterized by replacement fibrosis which starts in the epicardium or mid-mural layers. Both replacement and interstitial fibrosis can be observed. In the RV it is admixed with fat. In the LV, it may be mainly interstitial and replacement fibrosis with little fat. Endocardial fibrosis can be present. Fibro-fatty tissue replacement in the RV is always present, at the intersection of the free wall with epicardial fat; in the LV, it can be present but usually there is more fibrous tissue. Isolated fatty tissue without fibrosis is not a diagnostic feature of ACM . Inflammatory cells . Focal sparse lympho-monocyte infiltrates can be present similarly to DCM. Sometimes diffuse inflammatory infiltrates preferentially affecting the LV in the subepicardial and/or mid-mural layers are observed which makes difficult or impossible the differential diagnosis between ACM (acute/hot phase) and myocarditis . Granulation tissue and loose connective tissue have been also described in subacute stages of disease evolution. Intramural small vessels . No disease-related abnormalities.
An increased amount of epicardial fat ( adipositas cordis ) and isolated fatty infiltration in the RV are different entities and should not be confused with ACM. They can be observed in obese people, people with metabolic diseases, elderly or even normal people, with or without increased epicardial fat, particularly in the anterior and lateral RV free walls . Fat also accompanies the intramural vessels in both ventricles. A certain amount of fatty tissue is always present within the myocardium of the RV free wall so that strands of myocardial fibers are separated by fatty tissue, without cardiac myocytes abnormalities and replacement-type fibrosis. The boundary between the myocardium and the outer epicardial fat is usually distinct. Focal areas of endocardial fat with no fibrosis are also a nonspecific finding that should not led to a diagnosis of ACM. Myocardial fibrosis, with a nonischemic distribution in the LV free wall, either patchy or confluent in the mid-mural or sub-epicardial layers, can be due to a previous myocarditis, toxic insult or infarction. A descriptive pathology report of this to include all possibilities should be provided and advise molecular autopsy with cardiogenetic screening of first degree family members . Differential diagnosis of ACM also includes DCM, isolated cardiac sarcoidosis, Chagas diseases, previous myocardial infarction, muscular dystrophies, and chronic effects of drug toxicity. Abnormal pulmonary venous drainages or other congenital heart defect with RV overload and dilatation should be excluded. Histological features and/or clinical history and other cardiac or extracardiac findings are of help in differential diagnosis.
) It is the most common CMP and includes genetic and nongenetic forms . Its prevalence has been recently estimated in the range of 1:250, and the percentage of genetically determined forms ranges from 30 to 50%. The presence of cardiomegaly with increased cardiac mass and LV or biventricular dilatation with decreased mass/volume ratio, in the absence of coronary artery and valve diseases and hypertension, should led to the suspicion of DCM. Familial DCM accounts for 30–50% of DCM. In patients with familial DCM, approximately 40–50% has an identifiable genetic cause . The genetic background of DCM is wide and complex with pathogenic mutations in more than 50 genes encoding for cytoskeleton, sarcomeric proteins, sarcolemma, nuclear envelope, ion channels, and intercellular junctions. Although mutations in titin remain the most common identifiable cause, there is growing evidence for mutations in lamin A, desmosomal genes, and filamin C gene underlying an arrhythmogenic form of DCM. X-linked, autosomal recessive, and mitochondrial inheritance of DCM are less common. Phospholamban (PLN) CMP has clinical and histological characteristics of both DCM and ACM. Acquired causes of DCM include alcohol, pregnancy, post-myocarditis, hemochromatosis, chronic anemia, anthracycline medications, sarcoidosis, stimulant drugs (e.g., cocaine), etc. Idiopathic DCM is eventually a diagnosis of exclusion. The recent demonstration that up to 20% of DCM patients with an established acquired risk factor or a nonfamilial disease still carries a pathogenic gene variant suggests a broader role for genetic testing in DCM . From the pathological viewpoint, there is no evidence that particular patterns or amount of fibrosis and cardiac myocyte changes will favor either acquired or genetic cause. A diagnosis of DCM should always trigger familial investigation and genetic testing for an etiologic diagnosis. Macroscopic changes Heart size. The heart is enlarged with increased transverse size . Heart . The heart weight is usually increased. See normal values for age and body weight . Wall thickness . IVS and LV wall thickness are usually normal values or even decreased due to wall thinning (see referral normal values) . RV wall thickness is normal or decreased. Ventricular cavity size . LV cavity enlargement is typically observed, internal diameter can help for qualitative assessment (eccentric hypertrophy) . RV cavity can be also enlarged. If chamber dilatation is the only finding especially with delayed autopsy, one should look for other macroscopic and microscopic findings to support the suspicion of DCM. Valves . DCM with chamber dilatation and mitro-tricuspid annulus enlargement could explain the in vivo finding of functional regurgitation. Mild increase in leaflet thickening can be observed. Coronary arteries . No disease-related abnormalities. Atria . RA and LA cavity can be dilated. Endocardial thrombosis can be present, particularly in the appendages. Myocardium . Fibrous scar may be present, of variable size and distribution typically with a nonischemic pattern. Sometimes distinctive patterns, such as midwall fibrosis particularly in the IVS, are visible. Hypertrabeculation can occur in advanced forms with wall thinning and prominent cavity dilatation. Endocardium . Mural thrombosis is frequent due to blood stagnation in the poorly contracting LV. With time organizing thrombosis can lead to whitish appearance of the endocardium with fibrous thickening. Endocardial fibrosis can also be observed in the absence of previous thrombus. Fibroelastosis of the LV endocardium is common in the first years of life, usually associated with viral infection or congenital anomalies . Pericardium . Usually no disease-related abnormalities. The presence of pericardial adhesion could indicate that DCM is a result of previous myo-pericarditis. Microscopic changes The histologic abnormalities are nonspecific. Cardiac myocytes . Histological features can vary from normal to slight to marked degenerative changes with increased cell diameter, nuclear abnormalities and cytoplasmic attenuation with perinuclear halo. Hyperchromatic and asymmetric nuclei are often observed. Fibrosis . Interstitial fibrosis of variable degree is present; replacement-type fibrosis can be present with either patchy or confluent non-ischemic distribution. Inflammatory cells . Focal sparse lympho-monocyte infiltrates can be present. Intramural small vessel . No disease-related abnormalities. Diagnostic challenges and mimics Differential diagnosis of DCM includes myocarditis, isolated cardiac sarcoidosis, end-stage HCM, ACM, muscular dystrophies, and chronic ischemic heart disease with LV remodeling. Specific histological features and/or clinical history and other cardiac or extracardiac findings are of help in differential diagnosis. Pathologists should take in mind that chamber dilatation can be observed as an artefact of postmortem autolytic changes as already stated. Therefore, if chamber dilatation is the only finding, one should look for other macroscopic and microscopic findings that support the diagnosis of DCM.
Heart size. The heart is enlarged with increased transverse size . Heart . The heart weight is usually increased. See normal values for age and body weight . Wall thickness . IVS and LV wall thickness are usually normal values or even decreased due to wall thinning (see referral normal values) . RV wall thickness is normal or decreased. Ventricular cavity size . LV cavity enlargement is typically observed, internal diameter can help for qualitative assessment (eccentric hypertrophy) . RV cavity can be also enlarged. If chamber dilatation is the only finding especially with delayed autopsy, one should look for other macroscopic and microscopic findings to support the suspicion of DCM. Valves . DCM with chamber dilatation and mitro-tricuspid annulus enlargement could explain the in vivo finding of functional regurgitation. Mild increase in leaflet thickening can be observed. Coronary arteries . No disease-related abnormalities. Atria . RA and LA cavity can be dilated. Endocardial thrombosis can be present, particularly in the appendages. Myocardium . Fibrous scar may be present, of variable size and distribution typically with a nonischemic pattern. Sometimes distinctive patterns, such as midwall fibrosis particularly in the IVS, are visible. Hypertrabeculation can occur in advanced forms with wall thinning and prominent cavity dilatation. Endocardium . Mural thrombosis is frequent due to blood stagnation in the poorly contracting LV. With time organizing thrombosis can lead to whitish appearance of the endocardium with fibrous thickening. Endocardial fibrosis can also be observed in the absence of previous thrombus. Fibroelastosis of the LV endocardium is common in the first years of life, usually associated with viral infection or congenital anomalies . Pericardium . Usually no disease-related abnormalities. The presence of pericardial adhesion could indicate that DCM is a result of previous myo-pericarditis.
The histologic abnormalities are nonspecific. Cardiac myocytes . Histological features can vary from normal to slight to marked degenerative changes with increased cell diameter, nuclear abnormalities and cytoplasmic attenuation with perinuclear halo. Hyperchromatic and asymmetric nuclei are often observed. Fibrosis . Interstitial fibrosis of variable degree is present; replacement-type fibrosis can be present with either patchy or confluent non-ischemic distribution. Inflammatory cells . Focal sparse lympho-monocyte infiltrates can be present. Intramural small vessel . No disease-related abnormalities.
Differential diagnosis of DCM includes myocarditis, isolated cardiac sarcoidosis, end-stage HCM, ACM, muscular dystrophies, and chronic ischemic heart disease with LV remodeling. Specific histological features and/or clinical history and other cardiac or extracardiac findings are of help in differential diagnosis. Pathologists should take in mind that chamber dilatation can be observed as an artefact of postmortem autolytic changes as already stated. Therefore, if chamber dilatation is the only finding, one should look for other macroscopic and microscopic findings that support the diagnosis of DCM.
Primary restrictive CMP (RCM) is a poorly recognized entity characterized by nondilated, nonhypertrophied ventricles with dilated atria. The heart may be more rigid with cutting. In comparison with other forms of CMP, RCM is uncommon and a rare a cause of SCD. The pathology diagnosis is challenging in the absence of clinical findings which are essential to support a certain diagnosis of RCM . Many of the RCM probands had pathologic mutations in either beta-myosin heavy chain ( MYH7 ) or the cardiac troponin I gene ( TNNI3 ) . Mutations in other sarcomeric genes including troponin T ( TTNT2 ), myosin-binding protein C ( MYBPC3 ), myosin light chains ( MYL 2 and 3 ), and alpha cardiac actin ( ACTC ) have also been described. RCM can coexist with HCM in the same family. Most sarcomeric RCM mutations appear to be de novo and associated with a severe disease expression and an early onset. Nonsarcomeric mutations have also been identified in RCM and include mutations in myopalladin ( MYPN ), titin ( TTN ), and filamin-C (FLNC) . Genetic causes cannot be established from macroscopic or microscopic features and family investigation with appropriate genetic testing is essential. Pathologists should be aware that there are several other causes of RCM, including infiltrative diseases (amyloidosis in which the amyloid may be mistaken for fibrosis), storage diseases, mitochondrial disease and a variety of systemic diseases such as hemochromatosis. In these cases, there may be overlap with the other CMP phenotypes such as HCM. Loeffler endocarditis and endomyocardial fibrosis can also present with a restrictive pattern. Histological features and/or clinical history and other cardiac or extracardiac findings are essential in making these diagnoses. Interstitial fibrosis with or without myocyte disarray, without evidence of storage or infiltrative disease, are highly suggestive histologic features of primary RCM. Macroscopic changes Heart size . The heart is usually normal in size with normal transverse and longitudinal size. Heart weight usually normal. See normal values for age and body weight . Wall thickness . IVS, LV, and RV wall thickness usually normal values (see referral normal values). Ventricular cavity size . Normal. Valves . No disease-related abnormalities. Coronary arteries . No disease-related abnormalities. Atria . RA and LA cavity can be dilated. Endocardial thrombosis can be present, particularly in the appendages . Myocardium . No disease-related abnormalities. Endocardium . No disease-related abnormalities. Pericardium . No disease-related abnormalities. Microscopic changes Cardiac myocytes . They are usually of normal size and morphology or may show degenerative changes. Myocyte disarray can be observed as part of HCM spectrum . Fibrosis . Interstitial fibrosis is present in all cases. Replacement-type fibrosis can be present. Inflammatory cells . No disease-related changes are reported. Intramural small vessel . No disease-related abnormalities.
Heart size . The heart is usually normal in size with normal transverse and longitudinal size. Heart weight usually normal. See normal values for age and body weight . Wall thickness . IVS, LV, and RV wall thickness usually normal values (see referral normal values). Ventricular cavity size . Normal. Valves . No disease-related abnormalities. Coronary arteries . No disease-related abnormalities. Atria . RA and LA cavity can be dilated. Endocardial thrombosis can be present, particularly in the appendages . Myocardium . No disease-related abnormalities. Endocardium . No disease-related abnormalities. Pericardium . No disease-related abnormalities.
Cardiac myocytes . They are usually of normal size and morphology or may show degenerative changes. Myocyte disarray can be observed as part of HCM spectrum . Fibrosis . Interstitial fibrosis is present in all cases. Replacement-type fibrosis can be present. Inflammatory cells . No disease-related changes are reported. Intramural small vessel . No disease-related abnormalities.
The entity reported as noncompaction CMP, which is also known as LV noncompaction (LVNC), is characterized by excessive trabeculations of the LV and deep intertrabecular recesses communicating with the ventricular cavity, with a > twofold thickening of the endocardial noncompacted (NC) layer compared with the epicardial compacted (C) layer of the myocardium (NC/C > 2) . LVNC has been considered a genetic CMP by the AHA, and an unclassified CMP by the ESC. This phenotype is often associated with congenital heart diseases or other CMP, in which genetic analysis follows the patterns of the underlying genetic CMP. There is evidence for wide genetic heterogeneity in LVNC with mutations in sarcomeric, cytoskeletal, Z-line, and mitochondrial proteins. Therefore, it is controversial whether to consider it as a separate entity and it is very rarely encountered in autopsy practice . There is a risk of misnaming a CMP as LVNC when the underlying genetic cause is in fact a HCM, DCM or ACM. We consider LVNC as a phenotypic trait rather than a CMP, that can occur either in isolation or in association with other either congenital heart defects or CMP. An entity labelled idiopathic LV fibrosis has been put forward as a cause of SCD especially during sport activities . The heart is normal in weight, both ventricles are not dilated and there is no hypertrophy or thinning of the ventricular walls. Fibrosis may be macroscopically visible or not. Microscopically there is significant replacement fibrosis in the LV wall with no myocyte disarray. The distribution of the fibrosis is focal and does not have the patterns seen in other CMP from the histological point of view. This entity, which is also called non ischemic scarring, could be an acquired condition due to healed myocarditis, toxic or drug related but may also be a genetic CMP as familial cases are reported. Thus, the finding of no specific LV scarring should rise the possibility of an inherited CMP in which the phenotype is not well defined.
The potential genetic origin of heart diseases, including CMP, involves pathologists as part of a multidisciplinary investigation on surviving family members according to AECVP autopsy guidelines and European recommendations established by the Public and Professional Policy Committee of the European Society of Human Genetics (ESHG) . The ESHG recommendations summarize specific procedural, ethical, legal, and practical challenges for post-mortem genetic testing after SCD and indicate how to include postmortem genetic testing in the context of SCD in order to contribute to a better identification of the cause of death, and to a better management of relatives by optimizing screening strategies and treatments of preventable disorders. Their aim is also to stimulate the development of standardized postmortem disclosure policy and procedure at national and international levels for SCD cases and relatives. The role of pathologists in the multidisciplinary team is to perform an autopsy following the AECVP guidelines, to diagnose correctly the case with the help of an expert cardiovascular pathologist if necessary, to store postmortem samples in dedicated biobanks according to legal and ethical guidelines and to help passing on the information either by themselves or via the general practitioner or the cardiologist to the relatives of SCD victims about specific pathway available in their region/country. A complete autopsy is recommended for SCD victims, as mandatory for deaths under the age of 40 and should be considered for deaths between 40 and 65. These recommendations are also in line with the recent document of the Asia Pacific Heart Rhythm Society and the Heart Rhythm Society . However, adherence to guidelines/recommendations is still suboptimal in many European countries. In up to 40% of cases, autopsies are not performed in subjects less than 50 years who may have died from cardiac disease and only 50% of pathologists declared to follow a standard protocol for autopsy examination, apparently due to lack of expertise and/or training .
While clinical diagnoses of inherited CMP have progressed rapidly with imaging and genetic advances, pathological diagnoses have not progressed at the same pace. There has been as a matter of fact no previous guidelines on these entities. Establishing pathological guidelines is essential as these entities can present with SCD and it is up to the pathologist to establish the diagnosis and raise the possibility of a genetic CMP within the family, thus enabling screening in other family members. The pathologist must also take material for genetic investigation in the deceased (the postmortem genetic testing) when an inherited disease is suspected. The pathologist thus has a pivotal role in the diagnosis and prevention of further deaths within the family and is an essential member of the multidisciplinary team dealing with inherited CMP. Guidelines will change and evolve with advances in knowledge and hopefully uncertain entities will be clarified further.
Below is the link to the electronic supplementary material. Supplementary file1 (DOCX 25 KB)
|
Editorial: Special 2021 Frontiers in Endocrinology collection for the 100 | 6492a381-6f3c-47b7-b943-1d42057c510a | 10068784 | Physiology[mh] | All authors listed have made a substantial, direct, and intellectual contribution to the work and approved it for publication. |
Early death after palliative radiation treatment: 30-, 35- and 40-day mortality data and statistically robust predictors | 373d87e2-f610-48ad-a127-4337d7e77ea3 | 10069056 | Internal Medicine[mh] | Topics such as value-based care, quality-of-care indicators, cost-effectiveness and overtreatment have received considerable attention in the oncological literature . Special consideration is necessary in the palliative and terminal phase of anti-cancer treatment, where mismatch between side effects, cost and other disadvantages of interventions on one hand and expected benefit on the other hand should be minimized. Among factors to consider are an intervention’s aim, e.g., life-prolonging versus symptom-directed, and time-frame aspects such as remaining life time and duration of treatment. Palliative radiotherapy is among the most effective and cost-effective interventions and can be tailored to individual patients’ need and preferences . Extreme hypofractionation cuts treatment duration into a fraction of what is needed to complete traditional regimens, e.g. 3 Gy × 10 . As suggested by a recent large meta-analysis , there is room for improvement of physicians’ prescription habits or ability to decipher prognosis, because the authors found that 16% of patients with advanced cancer receiving palliative radiotherapy died within 30 days of treatment. In other words, the remaining life time with, e.g. reduced pain if this was the goal of treatment, may have been too short to outweigh the burden or side effects of radiotherapy in a proportion of patients. Typically, decision regret analyses are not performed near the end of life and it is thus difficult to estimate how many patients would have consented to radiotherapy in the final phase of cancer progression, had they been able to judge outcomes in advance. There are different ways of measuring radiotherapy utilization near the end of life, e.g. 30-day mortality calculated from start of treatment, 30-day mortality calculated from end of treatment, or treatment in the last 30 days of life. In addition, one might be tempted to ask why radiotherapy performed, e.g., in the last 28 days of life is fundamentally different from treatment one or two weeks later. Does the arbitrary 30-day cut-off represent a sound definition, because the early death rate is highest, e.g., at 20–30 days and patients living beyond that mark often survive for another 2–3 months? Or is death a continuous event necessitating a broader evaluation of alternative time frames? In principle, a peak might exist just outside the 30-day time period. These considerations and open questions led us to study death rates and predictors of 30-, 35- and 40-day mortality in an already established database with many baseline parameters that are lacking in large registries such as the National Cancer Database (NCDB) or the Surveillance, Epidemiology, and End Results (SEER) program.
Our single-institution database (2014–2019) includes 219 consecutive patients with bone metastases managed with standard palliative external beam radiotherapy regimens such as a single fraction of 8 Gy, 5 fractions of 4 Gy or 10 fractions of 3 Gy (3-D conformal or intensity-modulated; no stereotactic ablative body radiotherapy). Fractionation was at the discretion of the treating oncologist. Additional lesions were treated as indicated, e.g., soft tissue or lung metastases. In other words, a proportion of patients received radiotherapy to several target volumes at the same time. Interrupted or permanently discontinued radiotherapy series were included to comply with the intention-to-treat principle. Standard-of-care systemic anticancer treatment was given as indicated (tailored to organ function, frailty etc.). Patients who returned for a new treatment course (re-irradiation or new target volume) in the time period of the study were counted twice, resulting in a total number of 287 evaluable treatment courses. In returning patients, actual blood test results, imaging reports, Karnofsky performance status (KPS), weight and other baseline data, as well as survival were registered for each individual treatment course. Imaging and blood tests were part of standard oncological assessment and typically no older than 3 weeks before radiotherapy. Most patients had blood tests taken at the day of treatment planning. All blood test results were dichotomized (normal/abnormal) according to the institutional upper and lower limits of normal. The review-board approved database is regularly updated for survival and has been utilized for different quality-of-care projects before . Overall survival (time to death) from the first day of radiotherapy was calculated employing the Kaplan–Meier method for all 287 treatment courses (SPSS 28, IBM Corp., Armonk, NY, USA). In 27 cases, survival was censored after median 36 months of follow-up (minimum 28 months). Outcomes of interest (30-, 35-, 40-day mortality from start; death within 30 days of last radiation treatment) were dichotomized (alive/dead) and the chi-square test (2-sided) was utilized for further analyses. A multi-nominal logistic regression analysis was also employed. P -values ≤ 0.05 were considered statistically significant. The methods employed by Rades et al. were utilized to calculate a point sum reflective of 30-day mortality . For example, a risk factor associated with 50% 30-day mortality was assigned 5 points, while 3 points were assigned for a factor associated with 30% 30-day mortality.
Regarding all 287 treatment courses, 42 (15%) took place in the last month of life. Mortality from start of radiotherapy was 13% (30-day), 15% (35-day) and 18% (40-day), respectively. As indicated in Fig. , the 30-day landmark is not particularly representative for early death. Death rates were lower in the first 15 days and increased between day 16 and 45. None of the 5-day intervals can be characterized as outlier. Median actuarial overall survival was 6 months (1-year rate 32%). Table describes the patient-, tumor- and treatment-related baseline characteristics. The impact of all these baseline characteristics on 30-, 35- and 40-day mortality was examined and Table shows that a large number of significant correlations was identified in univariate analyses. All predictors of 30-day mortality were also associated with both, 35- and 40-day mortality. Predictors with p ≤ 0.05 in univariate analyses were included in multi-nominal logistic regression analyses. The one for 30-day mortality confirmed KPS (≤ 50 with hazard ratio (HR) 3.7 and 60–70 with HR 1.8, p < 0.001), weight loss (HR 1.8, p = 0.01) and presence of pleural effusion (HR 7.5, p = 0.006) as independent predictors, whereas, e.g., cancer type, blood test results and treatment-related parameters lost their significance. All three significant predictors of 30-day mortality maintained their impact in an exploratory analysis of 40-day mortality with p = 0.001–0.003. Additional predictors emerged, albeit with different p -values. These included adrenal gland metastases ( p = 0.02), progressive disease outside of the irradiated region(s) ( p = 0.03), and serum creatinine (normal versus abnormal, p = 0.02). Interestingly, all three additional predictors were also identified in the earlier analyses displayed in Table (bold text), because of disproportional increase of % mortality over time. Finally, the three significant predictors of 30-day mortality were employed to construct a predictive model based on the methodology developed by Rades et al. . Table shows how the point sum can be calculated and Fig. displays the corresponding 30-day mortality rates of 0–75%.
This study compared death rates during different time intervals in the early phase after radiotherapy and identified variables that impact on, e.g., 30-day mortality. Early death was not limited to the first 30 days after start of radiotherapy. Relatively similar death rates were seen between day 16 and 45. Focusing on 30-day mortality, a widely used endpoint in the literature (radiotherapy and other approaches), is thus an arbitrary decision (some sort of cut-off is needed) and not necessarily data-driven, as shown in the present example. Furthermore, palliative radiotherapy is not normally associated with procedure-related mortality, in contrast to, e.g. surgery. The present results also demonstrate that mortality rates depend on the method of evaluation. 30-day mortality from start of treatment was 13%, while 15% of courses were administered in the last 30 days of life. Modest increase of the cut-off, from 30 to 40 days from start of radiotherapy, increased the rate from 13 to 18%. In a recent large meta-analysis , 16% of patients with advanced cancer who had received palliative radiotherapy died within 30 days of treatment. In contrast to several previous studies, the present one included an unusually large number of baseline parameters, both traditional predictors of survival such as KPS, and less well-studied variables such as presence of pleural effusion and numerous blood test results. All predictors of 30-day mortality were also associated with both, 35- and 40-day mortality, and thus robust. Nevertheless, with increasing time interval and number of events (higher death rate after 40 days), additional predictors of early death emerged, albeit with clearly different p -values. These dynamics suggest that an increasing number of co-variates impact on death rates in analyses that cover a longer time period. KPS, weight loss and pleural effusion maintained their highly significant role and were therefore employed to construct a predictive model, which performed well (Fig. ). Pleural effusion, which was observed, e.g., in patients with lung, breast and prostate cancer, was not necessarily symptomatic and did not always necessitate intervention. Our study did not include patient-reported dyspnea, which in previous studies was associated with poor prognosis . These two factors might be interrelated, an issue that can only be clarified in prospective studies. The present results are in line with numerous prognostic models that include KPS as a main and indisputable driver of poor prognosis . However, additional factors are important to fully elucidate the likelihood of survival at different time points. Their role requires further study in larger databases. Besides number of patients, limitations of the present work include its retrospective single-institution design and selection bias, because a proportion of poor-prognosis patients referred to palliative radiotherapy might die already before planned start. On the other hand, the study cohort represents a real-world patient population of often elderly patients with highly variable disease burden and survival. Furthermore, we had access to a broad set of baseline parameters and were therefore able to extend the knowledge provided by previous, otherwise similar studies. Patients with brain metastases, a small subgroup in the present study, might represent a special population, if treated to the brain rather than skeletal metastases after previous brain-directed therapy. Our group’s previous work resulted in different predictors of 30-day mortality after treatment for brain metastases (n = 100 patients) than those identified in the present bone metastases study, e.g., number of brain metastases and primary tumor control. Despite progress in prognostic stratification, survival predictions in oncology tend to be overly optimistic . Not all patients initially thought to represent suitable candidates for radiotherapy are able to complete their treatment. In a recent study by Vázquez et al., 30-day mortality after palliative radiotherapy was 17.5% . In the multivariate analysis, male gender, ECOG PS 2–3, gastrointestinal and lung cancer were found to be independent factors related to this endpoint. Weight loss and other parameters available in the present study were not included. The large meta-analysis by Kutzko et al. identified multiple treatment sites, hepatobiliary primary, inpatient status, and ECOG PS 3–4 as predictors of 30-day mortality . In contrast to these results, Wu et al. performed a multivariate analysis suggesting that breast or prostate primary tumor, ECOG PS, body mass index, liver metastases, more than 5 active metastases (dichotomized, radiographically identified), albumin level, and hospitalization within 3 months of radiotherapy consult were associated with 30-day survival . Harmonization efforts and cooperation are needed to arrive at generally accepted and widely implemented predictive models, or a single consensus model. So far, it seems that PS and primary tumor type are common and well-established predictors, while contradictory results were obtained for other variables. Ideally, prospective comparisons should be attempted to clarify the role of potentially redundant variables such as patient-reported dyspnea, radiological presence of lung metastases or pleural effusion, and blood test results such as anemia, which might impact on dyspnea. Patients at high risk of early death should preferably be managed with single-fraction radiotherapy for bone metastases , if they prefer radiotherapy over other palliative and supportive measures aiming at pain control. Even patients with longer survival can often achieve satisfactory pain control with such simple treatment, if uncomplicated bone metastases are present, and sometimes additional re-irradiation is able to “boost” and prolong the effect of initial treatment. Special scenarios such as impending fractures, post-operative radiotherapy, large extra-osseous infiltration or ablation of oligometastases require thorough assessment of advantages and disadvantages of prolonged courses of radiotherapy or stereotactic body radiotherapy.
Early death was not limited to the first 30 days after start of palliative radiotherapy for bone metastases. For different cut-off points (30-, 35-, 40-day mortality), similar predictive factors emerged. A model based on three robust predictors was developed, which is easily applicable in clinical practice. External validation by other institutions is warranted.
|
Executive Summary: British Society for Rheumatology guideline on prescribing drugs in pregnancy and breastfeeding: comorbidity medications used in rheumatology practice | ed6a2ea6-caf0-4dbf-8bc7-4581654a03df | 10070061 | Internal Medicine[mh] | Background The rationale behind this update on the 2016 British Society for Rheumatology (BSR) guidelines on prescribing anti-rheumatic drugs in pregnancy and breastfeeding was described in detail in the guideline scope . In brief, despite the existence of additional evidence-based guidelines on prescribing/managing rheumatic disease in pregnancy the information contained within them requires continual review to include emerging information on the safety of new and existing drugs in pregnancy. Chronic disease adversely affects pregnancy. Data from Mothers and Babies: Reducing Risk through Audits and Confidential Enquiries across the UK (MBRRACE-UK) reports regularly from a national programme of work conducting surveillance and investigating the causes of maternal deaths, stillbirths and infant deaths . Data from 2017–19 found that 8.8 women per 100 000 died during pregnancy or up to six weeks after childbirth or the end of pregnancy, and most women who died had multiple health problems or other vulnerabilities . In all decisions regarding medication choices and changes, it is also important to consider the potential for deterioration in the mother's wellbeing through side effects or reduced disease control (and its adverse impact on the baby). Therefore, the exposure of the foetus to different drugs when switches are made must be balanced against possible foetal gains and understanding the potential impact of reduced control of the medical disorder on a pregnancy is vital . Need for guideline Patients with inflammatory rheumatic disease (IRD) should be counselled to achieve and then maintain remission or low disease activity before/during pregnancy to reduce the risk of adverse pregnancy outcomes . This goal is primarily achieved through adjustment of therapy to ensure disease control with disease modifying anti-rheumatic drugs (DMARDs) and/or immunosuppressive drugs that are compatible with pregnancy. These medications are reviewed in the BSR guideline on prescribing drugs in pregnancy and breastfeeding: immunomodulatory anti-rheumatic drugs and corticosteroids . Many patients with IRD, however, have an additional burden of pain and comorbid illness that require treatment with other medications. The compatibility of various comorbidity medications relevant to rheumatic disease will be covered in this update. This updated information will provide advice for healthcare professionals and patients to ensure more confident prescribing in these scenarios and will highlight any medications that should be stopped and/or avoided in the reproductive age group unless highly effective contraception is used, in line with guidance issued by the Medicines and Healthcare Products Regulatory Agency (MHRA) and Faculty of Sexual and Reproductive Healthcare . Objectives of guideline To update the previous BSR guidelines on prescribing in pregnancy in rheumatic disease for the following drug categories: pain management; non-steroidal anti-inflammatory drugs (NSAIDs); low-dose aspirin (LDA); anticoagulants; colchicine; dapsone; bisphosphonates; anti-hypertensives; and pulmonary vasodilators. This revised guideline was produced by consensus review of current evidence to answer specific questions in relation to each drug as follows. Should it be stopped pre-conception? Is it compatible with pregnancy? Is it compatible with breastmilk exposure? Where possible, recommendations are made regarding compatibility with paternal exposure. Target audience The primary audience consists of health professionals in the UK directly involved in managing patients with rheumatic disease who are (or are planning to become) pregnant and/or breastfeeding, men planning to conceive, and patients who have accidentally conceived while taking these medications. This audience includes rheumatologists, rheumatology nurses/allied health professionals, rheumatology speciality trainees and pharmacists, as well as the patients themselves. The guideline will also be useful to obstetricians, obstetric physicians, renal physicians, dermatologists and general practitioners who may prescribe these medications to patients in pregnancy. This guideline uses the terms ‘woman’, ‘maternal’ or ‘mother’ throughout. These should be taken to include people who do not identify as women but are pregnant or have given birth . The areas the guideline does not cover This guideline does not cover the management of infertility or acute pain relief during labour, hence morphine was excluded. Other drug categories: antimalarials; corticosteroids; disease modifying anti-rheumatic and immunosuppressive therapies; and biologic drugs are considered in another guideline . Stakeholder involvement This guideline was commissioned by the BSR Standards, Audit and Guidelines Working Group. A Guideline Working group (GWG) was created, consisting of a chair (I.G.), alongside representatives from relevant stakeholders ( ). In accordance with BSR policy, all members of the GWG made declarations of interest, available on the BSR website. Involvement and affiliations of stakeholder groups involved in guideline development The GWG consisted of rheumatologists from a range of clinical backgrounds, various allied health professionals, other specialists in women’s health, lay members and representatives from the United Kingdom Tetralogy Information Service (UKTIS). All members of the working group contributed to the process for agreeing key questions, guideline content, recommendations and strength of agreement.
The rationale behind this update on the 2016 British Society for Rheumatology (BSR) guidelines on prescribing anti-rheumatic drugs in pregnancy and breastfeeding was described in detail in the guideline scope . In brief, despite the existence of additional evidence-based guidelines on prescribing/managing rheumatic disease in pregnancy the information contained within them requires continual review to include emerging information on the safety of new and existing drugs in pregnancy. Chronic disease adversely affects pregnancy. Data from Mothers and Babies: Reducing Risk through Audits and Confidential Enquiries across the UK (MBRRACE-UK) reports regularly from a national programme of work conducting surveillance and investigating the causes of maternal deaths, stillbirths and infant deaths . Data from 2017–19 found that 8.8 women per 100 000 died during pregnancy or up to six weeks after childbirth or the end of pregnancy, and most women who died had multiple health problems or other vulnerabilities . In all decisions regarding medication choices and changes, it is also important to consider the potential for deterioration in the mother's wellbeing through side effects or reduced disease control (and its adverse impact on the baby). Therefore, the exposure of the foetus to different drugs when switches are made must be balanced against possible foetal gains and understanding the potential impact of reduced control of the medical disorder on a pregnancy is vital .
Patients with inflammatory rheumatic disease (IRD) should be counselled to achieve and then maintain remission or low disease activity before/during pregnancy to reduce the risk of adverse pregnancy outcomes . This goal is primarily achieved through adjustment of therapy to ensure disease control with disease modifying anti-rheumatic drugs (DMARDs) and/or immunosuppressive drugs that are compatible with pregnancy. These medications are reviewed in the BSR guideline on prescribing drugs in pregnancy and breastfeeding: immunomodulatory anti-rheumatic drugs and corticosteroids . Many patients with IRD, however, have an additional burden of pain and comorbid illness that require treatment with other medications. The compatibility of various comorbidity medications relevant to rheumatic disease will be covered in this update. This updated information will provide advice for healthcare professionals and patients to ensure more confident prescribing in these scenarios and will highlight any medications that should be stopped and/or avoided in the reproductive age group unless highly effective contraception is used, in line with guidance issued by the Medicines and Healthcare Products Regulatory Agency (MHRA) and Faculty of Sexual and Reproductive Healthcare .
To update the previous BSR guidelines on prescribing in pregnancy in rheumatic disease for the following drug categories: pain management; non-steroidal anti-inflammatory drugs (NSAIDs); low-dose aspirin (LDA); anticoagulants; colchicine; dapsone; bisphosphonates; anti-hypertensives; and pulmonary vasodilators. This revised guideline was produced by consensus review of current evidence to answer specific questions in relation to each drug as follows. Should it be stopped pre-conception? Is it compatible with pregnancy? Is it compatible with breastmilk exposure? Where possible, recommendations are made regarding compatibility with paternal exposure.
The primary audience consists of health professionals in the UK directly involved in managing patients with rheumatic disease who are (or are planning to become) pregnant and/or breastfeeding, men planning to conceive, and patients who have accidentally conceived while taking these medications. This audience includes rheumatologists, rheumatology nurses/allied health professionals, rheumatology speciality trainees and pharmacists, as well as the patients themselves. The guideline will also be useful to obstetricians, obstetric physicians, renal physicians, dermatologists and general practitioners who may prescribe these medications to patients in pregnancy. This guideline uses the terms ‘woman’, ‘maternal’ or ‘mother’ throughout. These should be taken to include people who do not identify as women but are pregnant or have given birth .
This guideline does not cover the management of infertility or acute pain relief during labour, hence morphine was excluded. Other drug categories: antimalarials; corticosteroids; disease modifying anti-rheumatic and immunosuppressive therapies; and biologic drugs are considered in another guideline .
This guideline was commissioned by the BSR Standards, Audit and Guidelines Working Group. A Guideline Working group (GWG) was created, consisting of a chair (I.G.), alongside representatives from relevant stakeholders ( ). In accordance with BSR policy, all members of the GWG made declarations of interest, available on the BSR website.
The GWG consisted of rheumatologists from a range of clinical backgrounds, various allied health professionals, other specialists in women’s health, lay members and representatives from the United Kingdom Tetralogy Information Service (UKTIS). All members of the working group contributed to the process for agreeing key questions, guideline content, recommendations and strength of agreement.
Statement of scope of literature search and strategy employed Most medications covered in this guideline have been comprehensively and systematically reviewed in multiple other documents, since the first BSR guideline on this topic. Therefore, a consensus-based approach was taken to compile and assess most significant evidence published since 2013 to December 2020 through a comprehensive search of MEDLINE, PubMed and EMBASE databases with specific search terms ( , available at Rheumatology online). Filters were applied to capture National Institute for Health and Care Excellence (NICE) guidance, international guidelines, systematic reviews, cohort studies or case-series. Information was preferentially selected from NICE guidance and/or largest/most recent systematic reviews and where lacking was extracted from largest cohort, case-series or abstract. Findings were cross-referenced with the previous BSR guideline , as well as the Cochrane, Lactmed (a National Library of Medicine database on drugs and lactation) and UKTIS databases. Two independent reviewers screened the title and abstract of 2997 articles, identified 130 and selected the most recent/largest systematic reviews or largest cohort study or case-series as well any NICE guidance and international guidelines. Thirty-six studies ( ) met the inclusion criteria and relevant information was extracted into data-extraction tables. Statement of methods used to formulate the recommendations (levels of evidence) The working group met regularly to formalise search strategy, review evidence, resolve disagreements and finally to determine recommendations. This guideline was developed in line with BSR’s Guideline Protocol using Grading of Recommendations, Assessment, Development and Evaluations (GRADE) methodology to determine quality of evidence and strength of recommendation. Accompanying each recommendation in this guideline, in brackets, is the strength of recommendation, quality of evidence and strength of agreement (SOA). Strength of recommendation Using GRADE, recommendations were categorized as either strong (denoted by 1) or weak (denoted by 2), according to the balance between benefits and risks. A strong recommendation was made when the benefits clearly outweigh the risks (or vice versa). A weak recommendation denotes that the benefits are more closely balanced with the risk or more uncertain. Quality of evidence Using the GRADE approach, the quality of evidence was determined as either high (A), moderate (B) or low/very low (C) reflecting the confidence in the estimates of benefits or harm. Strength of agreement The wording of each recommendation was discussed until all members were satisfied they would score at least 80 on a scale of 1 (no agreement) to 100 (complete agreement) and then 20/23 members with full voting rights scored each recommendation on the same scale and the average was calculated to generate a strength of agreement (SOA) score. Two patient representatives and a data-analyst expressed concern that they did not have sufficient medical knowledge of all drugs reviewed to score all recommendations, so while they fully agreed with each, they did not wish to score each one and did not contribute to the final SOA score. Statement of any limits of search and when guideline will be updated The search was conducted in December 2020. Limits were placed for English language and filters as described above. The guideline will be updated in five years.
Most medications covered in this guideline have been comprehensively and systematically reviewed in multiple other documents, since the first BSR guideline on this topic. Therefore, a consensus-based approach was taken to compile and assess most significant evidence published since 2013 to December 2020 through a comprehensive search of MEDLINE, PubMed and EMBASE databases with specific search terms ( , available at Rheumatology online). Filters were applied to capture National Institute for Health and Care Excellence (NICE) guidance, international guidelines, systematic reviews, cohort studies or case-series. Information was preferentially selected from NICE guidance and/or largest/most recent systematic reviews and where lacking was extracted from largest cohort, case-series or abstract. Findings were cross-referenced with the previous BSR guideline , as well as the Cochrane, Lactmed (a National Library of Medicine database on drugs and lactation) and UKTIS databases. Two independent reviewers screened the title and abstract of 2997 articles, identified 130 and selected the most recent/largest systematic reviews or largest cohort study or case-series as well any NICE guidance and international guidelines. Thirty-six studies ( ) met the inclusion criteria and relevant information was extracted into data-extraction tables.
The working group met regularly to formalise search strategy, review evidence, resolve disagreements and finally to determine recommendations. This guideline was developed in line with BSR’s Guideline Protocol using Grading of Recommendations, Assessment, Development and Evaluations (GRADE) methodology to determine quality of evidence and strength of recommendation. Accompanying each recommendation in this guideline, in brackets, is the strength of recommendation, quality of evidence and strength of agreement (SOA).
Using GRADE, recommendations were categorized as either strong (denoted by 1) or weak (denoted by 2), according to the balance between benefits and risks. A strong recommendation was made when the benefits clearly outweigh the risks (or vice versa). A weak recommendation denotes that the benefits are more closely balanced with the risk or more uncertain.
Using the GRADE approach, the quality of evidence was determined as either high (A), moderate (B) or low/very low (C) reflecting the confidence in the estimates of benefits or harm.
The wording of each recommendation was discussed until all members were satisfied they would score at least 80 on a scale of 1 (no agreement) to 100 (complete agreement) and then 20/23 members with full voting rights scored each recommendation on the same scale and the average was calculated to generate a strength of agreement (SOA) score. Two patient representatives and a data-analyst expressed concern that they did not have sufficient medical knowledge of all drugs reviewed to score all recommendations, so while they fully agreed with each, they did not wish to score each one and did not contribute to the final SOA score.
The search was conducted in December 2020. Limits were placed for English language and filters as described above. The guideline will be updated in five years.
Drugs are considered in the following categories: pain management; NSAIDs and low-dose aspirin in the management of multisystem rheumatic disease; anticoagulants; bisphosphonates; anti-hypertensive medication in the management of multisystem rheumatic disease; and pulmonary vasodilators. The overall findings for maternal and foetal breastmilk exposures to each drug, including information and key references from the previous BSR guideline are summarised in the full-length guideline and updated recommendations listed below. Paternal exposures and recommendations are described separately after maternal data. An overall summary of compatibility of each drug pre-conception, during pregnancy, breastmilk exposure and paternal exposure is shown in . Generic recommendations were developed based on evidence as shown in , available at Rheumatology online. Generic recommendations on prescribing in rheumatic disease in pregnancy Pre-conception counselling should be addressed by all healthcare professionals, with referral to professionals with relevant experience as appropriate to optimise all therapy, including non-pharmacological options for chronic pain management during pregnancy (GRADE 1A, SOA 99.5). The risks and benefits of drug treatment to mother and foetus should be discussed and clearly documented by all healthcare professionals involved in the patient’s care (GRADE 1A, SOA 99). The cause of pain and other symptoms should be assessed and managed appropriately (GRADE 1B, SOA 98.5). The requirement for analgesia should be assessed and minimum effective dose should be prescribed and titrated according to response (GRADE 1B, SOA 100). Tricyclic antidepressants are preferred over other antidepressant medications to manage chronic pain (GRADE 1B, SOA 98.1). Cessation of anti-depressant therapy that is being used as chronic pain medication in the post-natal period is not recommended, due to the risk of adverse impact on mood (GRADE 1C, SOA 96). LDA (≤150 mg/day) is recommended in all patients at high risk for pre-eclampsia (GRADE 1A, SOA 99.5). Low molecular weight heparin is the preferred anticoagulant (GRADE 1A, SOA 100). Nifedipine is the preferred vasodilator (GRADE 1B, SOA 98.5). Paternal drug exposure may reduce male fertility but has not been associated with adverse foetal development or pregnancy outcome. Although evidence is weak, we recommend that men are reassured about the safety of fathering a pregnancy whilst taking medicines to manage comorbidities as described in this guideline (GRADE 1C, SOA 98). Pain management: conventional analgesics Paracetamol Recommendations were based on two systematic reviews , UKTIS , Royal College of Obstetricians (RCOG) guidance and LactMed , as shown in , available at Rheumatology online. Recommendations for paracetamol in pregnancy and breastfeeding Paracetamol is the analgesic of choice and compatible peri-conception and throughout pregnancy (GRADE 1B, SOA 99). LactMed describes paracetamol as a good choice for analgesia and fever reduction in breastfeeding mothers (GRADE 2C, SOA 99.5). Codeine Recommendations were based on a systematic review (including a large Norwegian population-based cohort study , a large case-control study and a case-control study ), UKTIS , breastfeeding exposures from references and LactMed , as shown in , available at Rheumatology online. Recommendations for codeine in pregnancy and breastfeeding Codeine is compatible peri-conception and throughout pregnancy, although long-term use should be avoided. There is no consistent evidence to recommend a dose reduction pre-delivery but neonatologists should be aware of maternal use (GRADE 1B, SOA 97.8). Caution is advised with use of codeine in breastfeeding, due to the risk of CNS depression resulting from unpredictable metabolism of codeine to morphine (GRADE 1C, SOA 98). Tramadol Recommendations were based on a systematic review (including a large prospective cohort study of over 1.6 million women ), UKTIS (including case reports ), LactMed (including ) and RCOG guidance , as shown in , available at Rheumatology online. Recommendations for tramadol in pregnancy and breastfeeding Avoid tramadol peri-conception and in first trimester and only consider in second/third trimester if no alternative analgesia (GRADE 2B, SOA 97.8). Based on limited data, tramadol may be compatible with short-term use in breastfeeding (GRADE 2C, SOA 94.8). Other treatments for chronic pain Amitriptyline Recommendations were based on NICE guidance , a systematic review , UKTIS and LactMed , as shown in , available at Rheumatology online. Recommendations for amitriptyline in pregnancy and breastfeeding Amitriptyline is compatible with pregnancy. There is no evidence of adverse effect on IQ or developmental outcomes (GRADE 1C, SOA 100). Because very little amitriptyline is found in breastmilk with antidepressant doses and it is used at lower doses for chronic pain, it is unlikely to cause adverse effects in breastfed infants (GRADE 1C, SOA 100). Gabapentin and pregabalin Recommendations were based on [ , , , ] and , as shown in , available at Rheumatology online. Recommendations for gabapentin and pregabalin in pregnancy and breastfeeding Gabapentin at lowest effective dose may be considered in pregnancy with folic acid supplementation if no alternative analgesic suitable (GRADE 1B, SOA 95). Gabapentin may be considered in breastfeeding if no alternative analgesic is suitable (GRADE 2C, SOA 96). Pregabalin may be considered in pregnancy (with folic acid supplementation) and during breastfeeding (GRADE 2C, SOA 95.3). Serotonin–norepinephrine reuptake inhibitors (SNRIs) Recommendations were based on NICE guidance , two systematic reviews , UKTIS and LactMed , as shown in , available at Rheumatology online. Recommendations for SNRIs in pregnancy and breastfeeding Venlafaxine is compatible at conception and throughout pregnancy. There may be an increased risk of neonatal abstinence syndrome/short-term behavioural effects, but larger studies are needed to evaluate this finding (GRADE 2C, SOA 95.8). Duloxetine may be considered in pregnancy and breastfeeding but there are fewer data than for venlafaxine (GRADE 2C, SOA 95.3). Venlafaxine and duloxetine may be considered in breastfeeding if there is no alternative chronic pain medication (GRADE 2C, SOA 95.8). Selective serotonin reuptake inhibitors (SSRIs) Recommendations were based on [ , , , ] and , as shown in , available at Rheumatology online. Recommendations for SSRIs in pregnancy and breastfeeding Fluoxetine, paroxetine and sertraline are compatible with pregnancy (GRADE 1B, SOA 98.8). Based on limited evidence, SSRIs are compatible with breastfeeding (GRADE 2C, SOA 98.3). NSAIDs and anti-platelet drugs Recommendations were based on NICE guidance , four systematic reviews , a case report/review , UKTIS , LactMed and United States Food and Drug Administration (FDA) guidance , as shown in , available at Rheumatology online. Recommendations for NSAIDs and COX-2 inhibitors in pregnancy and breastfeeding Discordant findings from retrospective, large studies with population controls on the use of NSAIDs in the first trimester of pregnancy raise the possibility of a low risk of miscarriage and malformation. Therefore, these drugs should only be used intermittently in the first trimester of pregnancy (GRADE 1B, SOA 97.3). Intermittent rather than regular use of all non-selective NSAIDs except LDA is recommended throughout pregnancy and weaned from end of second trimester (26 weeks) to stop by gestational week 30 to avoid premature closure of the ductus arteriosus (GRADE 1B, SOA 98). At present there are limited data on selective COX-2 inhibitors; they should therefore be avoided during pregnancy (GRADE 2C, SOA 98.5). Non-selective NSAIDs (especially ibuprofen) are compatible with breastfeeding (GRADE 1C, SOA 98.8). Recommendations for low-dose aspirin and clopidogrel in pregnancy and breastfeeding LDA of ≤150 mg/day may be continued throughout pregnancy and NICE guidelines (2019) for hypertension in pregnancy advise treatment with LDA (for prophylaxis of pre-eclampsia) until delivery (GRADE 1B, SOA 99.0). LDA is compatible with breastfeeding (GRADE 2C, SOA 99.8). There are limited data on clopidogrel but it may be considered where alternative drugs are not suitable in pregnancy and breastfeeding (GRADE 2C, SOA 96.3). Colchicine and dapsone Recommendations were based on and , as shown in , available at Rheumatology online. Recommendations for colchicine and dapsone in pregnancy and breastfeeding Colchicine therapy may be considered during pregnancy (GRADE 1B, SOA 99.5). Dapsone may be used in pregnancy (GRADE 2C, SOA 95.0). Colchicine may be used in breastfeeding (GRADE 2C, SOA 98.3). Dapsone may be used in breastfeeding and due to the risk of haemolytic anaemia it is advised to monitor the infant for signs of haemolysis, especially in newborn or premature breastfed infants (GRADE 2C, SOA 90.7). Anticoagulants in rheumatic disease Recommendations were based on [ , , ] and , as shown in , available at Rheumatology online. Recommendations for anticoagulants in pregnancy and breastfeeding Low molecular weight heparin (LMWH) is compatible throughout pregnancy (GRADE 1A, SOA 100). LMWH is compatible with breastfeeding (GRADE 1C, SOA 100). The use of warfarin in pregnancy is associated with increased foetal risk throughout pregnancy and has limited indications, therefore should only be considered in exceptional circumstances (GRADE 1B, SOA 98.8). Warfarin is compatible with breastfeeding (GRADE 1A, SOA 100). Direct oral anticoagulants (DOACs) cannot be recommended in pregnancy (GRADE 1C, SOA 97.9) Rivaroxaban may be considered in breastfeeding (GRADE 2C, SOA 95.3) Other DOACs are not recommended in breastfeeding due to lack to human data and concerns from animal studies (GRADE 1C, SOA 97.4) Fondaparinux may be considered in pregnancy and breastfeeding if there is an allergy or adverse response to LMWH (GRADE 2C, SOA 95.5) Bisphosphonates Recommendations were based on [ , , ] and , as shown in , available at Rheumatology online. Recommendations for bisphosphonates in pregnancy and breastfeeding There is insufficient data upon which to recommend bisphosphonates in pregnancy or to advise a specific time for them to be stopped pre-conception. Given their biological half-life in bone of up to 10 years and no evidence of harm from limited reports of their use in pregnancy, a pragmatic recommendation is that they should be stopped 3 months in advance of pregnancy (GRADE 2C, SOA 96.8). There are no data on which to base a recommendation for the use of bisphosphonates during breastfeeding (GRADE 2C, SOA 98.5). Antihypertensive medication in rheumatic disease Recommendations were based on [ , , ] and , as shown in , available at Rheumatology online. Recommendations for angiotensin-converting enzyme inhibitors (ACEis)/angiotensin receptor blockers (ARBs) in pregnancy and breastfeeding ACEi and ARBs should be stopped as soon as possible when pregnancy is confirmed in the first trimester and if necessary an alternative antihypertensive compatible with pregnancy given (GRADE 1A, SOA 100). ACEis/ARBs should be avoided in the second and third trimester but may be considered under specialist advice in certain circumstances (GRADE 1C, SOA 98.5). Based on limited evidence, enalapril is compatible with breastfeeding (GRADE 2C, SOA 98.5). Calcium channel blockers Recommendations were based on [ , , ], as shown in , available at Rheumatology online. Recommendations for calcium channel blockers in pregnancy and breastfeeding Nifedipine is compatible with pregnancy with no direct evidence of harm at doses up to 90 mg/day (GRADE 1A, SOA 99.0). Nifedipine is compatible with breastfeeding (GRADE 1B, SOA 100). Amlodipine can be considered in pregnancy and breastfeeding as there is no evidence of harm (GRADE 1C, SOA 97.9). Pulmonary vasodilators Recommendations were based on [ , , , ] and , as shown in , available at Rheumatology online. Recommendations for pulmonary vasodilators in pregnancy and breastfeeding Established moderate-to-severe pulmonary hypertension (PHT) remains a contraindication to pregnancy. If pregnancy occurs, the use of these pulmonary vasodilator drugs in pregnancy should be considered only as part of a multidisciplinary team assessment (GRADE 1C, SOA 99.5). Limited evidence supports the use of prostacyclines to treat PHT during pregnancy (GRADE 2C, SOA 98.0). Limited evidence supports the use of sildenafil to treat PHT during pregnancy (GRADE 2C, SOA 98.0). Bosentan is teratogenic in animals and although there is no evidence of harm from human pregnancy, the evidence is insufficient to recommend in pregnancy (GRADE 1C, SOA 98.8). There are no data relating to breastfeeding exposure to pulmonary vasodilators on which to base a recommendation (GRADE 2C, SOA 98.8). Paternal exposures Recommendations were based on and , as shown in , available at Rheumatology online. Recommendations for paternal exposure Paracetamol is compatible with paternal exposure (GRADE 1B, SOA 98.5). Amitriptyline, SNRIs and SSRIs are compatible with paternal exposure (GRADE 1B, SOA 98.5). Non-selective NSAIDs are compatible with paternal exposure (GRADE 1C, SOA 98.4). Based on limited or no data and no association with adverse foetal development or pregnancy outcome, paternal exposure to all other drugs described in this guideline is unlikely to be harmful (GRADE 2C, SOA 97.3).
Pre-conception counselling should be addressed by all healthcare professionals, with referral to professionals with relevant experience as appropriate to optimise all therapy, including non-pharmacological options for chronic pain management during pregnancy (GRADE 1A, SOA 99.5). The risks and benefits of drug treatment to mother and foetus should be discussed and clearly documented by all healthcare professionals involved in the patient’s care (GRADE 1A, SOA 99). The cause of pain and other symptoms should be assessed and managed appropriately (GRADE 1B, SOA 98.5). The requirement for analgesia should be assessed and minimum effective dose should be prescribed and titrated according to response (GRADE 1B, SOA 100). Tricyclic antidepressants are preferred over other antidepressant medications to manage chronic pain (GRADE 1B, SOA 98.1). Cessation of anti-depressant therapy that is being used as chronic pain medication in the post-natal period is not recommended, due to the risk of adverse impact on mood (GRADE 1C, SOA 96). LDA (≤150 mg/day) is recommended in all patients at high risk for pre-eclampsia (GRADE 1A, SOA 99.5). Low molecular weight heparin is the preferred anticoagulant (GRADE 1A, SOA 100). Nifedipine is the preferred vasodilator (GRADE 1B, SOA 98.5). Paternal drug exposure may reduce male fertility but has not been associated with adverse foetal development or pregnancy outcome. Although evidence is weak, we recommend that men are reassured about the safety of fathering a pregnancy whilst taking medicines to manage comorbidities as described in this guideline (GRADE 1C, SOA 98).
Paracetamol Recommendations were based on two systematic reviews , UKTIS , Royal College of Obstetricians (RCOG) guidance and LactMed , as shown in , available at Rheumatology online. Recommendations for paracetamol in pregnancy and breastfeeding Paracetamol is the analgesic of choice and compatible peri-conception and throughout pregnancy (GRADE 1B, SOA 99). LactMed describes paracetamol as a good choice for analgesia and fever reduction in breastfeeding mothers (GRADE 2C, SOA 99.5).
Recommendations were based on two systematic reviews , UKTIS , Royal College of Obstetricians (RCOG) guidance and LactMed , as shown in , available at Rheumatology online. Recommendations for paracetamol in pregnancy and breastfeeding Paracetamol is the analgesic of choice and compatible peri-conception and throughout pregnancy (GRADE 1B, SOA 99). LactMed describes paracetamol as a good choice for analgesia and fever reduction in breastfeeding mothers (GRADE 2C, SOA 99.5).
Paracetamol is the analgesic of choice and compatible peri-conception and throughout pregnancy (GRADE 1B, SOA 99). LactMed describes paracetamol as a good choice for analgesia and fever reduction in breastfeeding mothers (GRADE 2C, SOA 99.5).
Recommendations were based on a systematic review (including a large Norwegian population-based cohort study , a large case-control study and a case-control study ), UKTIS , breastfeeding exposures from references and LactMed , as shown in , available at Rheumatology online. Recommendations for codeine in pregnancy and breastfeeding Codeine is compatible peri-conception and throughout pregnancy, although long-term use should be avoided. There is no consistent evidence to recommend a dose reduction pre-delivery but neonatologists should be aware of maternal use (GRADE 1B, SOA 97.8). Caution is advised with use of codeine in breastfeeding, due to the risk of CNS depression resulting from unpredictable metabolism of codeine to morphine (GRADE 1C, SOA 98).
Codeine is compatible peri-conception and throughout pregnancy, although long-term use should be avoided. There is no consistent evidence to recommend a dose reduction pre-delivery but neonatologists should be aware of maternal use (GRADE 1B, SOA 97.8). Caution is advised with use of codeine in breastfeeding, due to the risk of CNS depression resulting from unpredictable metabolism of codeine to morphine (GRADE 1C, SOA 98).
Recommendations were based on a systematic review (including a large prospective cohort study of over 1.6 million women ), UKTIS (including case reports ), LactMed (including ) and RCOG guidance , as shown in , available at Rheumatology online. Recommendations for tramadol in pregnancy and breastfeeding Avoid tramadol peri-conception and in first trimester and only consider in second/third trimester if no alternative analgesia (GRADE 2B, SOA 97.8). Based on limited data, tramadol may be compatible with short-term use in breastfeeding (GRADE 2C, SOA 94.8).
Avoid tramadol peri-conception and in first trimester and only consider in second/third trimester if no alternative analgesia (GRADE 2B, SOA 97.8). Based on limited data, tramadol may be compatible with short-term use in breastfeeding (GRADE 2C, SOA 94.8).
Amitriptyline Recommendations were based on NICE guidance , a systematic review , UKTIS and LactMed , as shown in , available at Rheumatology online. Recommendations for amitriptyline in pregnancy and breastfeeding Amitriptyline is compatible with pregnancy. There is no evidence of adverse effect on IQ or developmental outcomes (GRADE 1C, SOA 100). Because very little amitriptyline is found in breastmilk with antidepressant doses and it is used at lower doses for chronic pain, it is unlikely to cause adverse effects in breastfed infants (GRADE 1C, SOA 100).
Recommendations were based on NICE guidance , a systematic review , UKTIS and LactMed , as shown in , available at Rheumatology online. Recommendations for amitriptyline in pregnancy and breastfeeding Amitriptyline is compatible with pregnancy. There is no evidence of adverse effect on IQ or developmental outcomes (GRADE 1C, SOA 100). Because very little amitriptyline is found in breastmilk with antidepressant doses and it is used at lower doses for chronic pain, it is unlikely to cause adverse effects in breastfed infants (GRADE 1C, SOA 100).
Amitriptyline is compatible with pregnancy. There is no evidence of adverse effect on IQ or developmental outcomes (GRADE 1C, SOA 100). Because very little amitriptyline is found in breastmilk with antidepressant doses and it is used at lower doses for chronic pain, it is unlikely to cause adverse effects in breastfed infants (GRADE 1C, SOA 100).
Recommendations were based on [ , , , ] and , as shown in , available at Rheumatology online. Recommendations for gabapentin and pregabalin in pregnancy and breastfeeding Gabapentin at lowest effective dose may be considered in pregnancy with folic acid supplementation if no alternative analgesic suitable (GRADE 1B, SOA 95). Gabapentin may be considered in breastfeeding if no alternative analgesic is suitable (GRADE 2C, SOA 96). Pregabalin may be considered in pregnancy (with folic acid supplementation) and during breastfeeding (GRADE 2C, SOA 95.3).
Gabapentin at lowest effective dose may be considered in pregnancy with folic acid supplementation if no alternative analgesic suitable (GRADE 1B, SOA 95). Gabapentin may be considered in breastfeeding if no alternative analgesic is suitable (GRADE 2C, SOA 96). Pregabalin may be considered in pregnancy (with folic acid supplementation) and during breastfeeding (GRADE 2C, SOA 95.3).
Recommendations were based on NICE guidance , two systematic reviews , UKTIS and LactMed , as shown in , available at Rheumatology online. Recommendations for SNRIs in pregnancy and breastfeeding Venlafaxine is compatible at conception and throughout pregnancy. There may be an increased risk of neonatal abstinence syndrome/short-term behavioural effects, but larger studies are needed to evaluate this finding (GRADE 2C, SOA 95.8). Duloxetine may be considered in pregnancy and breastfeeding but there are fewer data than for venlafaxine (GRADE 2C, SOA 95.3). Venlafaxine and duloxetine may be considered in breastfeeding if there is no alternative chronic pain medication (GRADE 2C, SOA 95.8).
Venlafaxine is compatible at conception and throughout pregnancy. There may be an increased risk of neonatal abstinence syndrome/short-term behavioural effects, but larger studies are needed to evaluate this finding (GRADE 2C, SOA 95.8). Duloxetine may be considered in pregnancy and breastfeeding but there are fewer data than for venlafaxine (GRADE 2C, SOA 95.3). Venlafaxine and duloxetine may be considered in breastfeeding if there is no alternative chronic pain medication (GRADE 2C, SOA 95.8).
Recommendations were based on [ , , , ] and , as shown in , available at Rheumatology online. Recommendations for SSRIs in pregnancy and breastfeeding Fluoxetine, paroxetine and sertraline are compatible with pregnancy (GRADE 1B, SOA 98.8). Based on limited evidence, SSRIs are compatible with breastfeeding (GRADE 2C, SOA 98.3).
Fluoxetine, paroxetine and sertraline are compatible with pregnancy (GRADE 1B, SOA 98.8). Based on limited evidence, SSRIs are compatible with breastfeeding (GRADE 2C, SOA 98.3).
Recommendations were based on NICE guidance , four systematic reviews , a case report/review , UKTIS , LactMed and United States Food and Drug Administration (FDA) guidance , as shown in , available at Rheumatology online. Recommendations for NSAIDs and COX-2 inhibitors in pregnancy and breastfeeding Discordant findings from retrospective, large studies with population controls on the use of NSAIDs in the first trimester of pregnancy raise the possibility of a low risk of miscarriage and malformation. Therefore, these drugs should only be used intermittently in the first trimester of pregnancy (GRADE 1B, SOA 97.3). Intermittent rather than regular use of all non-selective NSAIDs except LDA is recommended throughout pregnancy and weaned from end of second trimester (26 weeks) to stop by gestational week 30 to avoid premature closure of the ductus arteriosus (GRADE 1B, SOA 98). At present there are limited data on selective COX-2 inhibitors; they should therefore be avoided during pregnancy (GRADE 2C, SOA 98.5). Non-selective NSAIDs (especially ibuprofen) are compatible with breastfeeding (GRADE 1C, SOA 98.8). Recommendations for low-dose aspirin and clopidogrel in pregnancy and breastfeeding LDA of ≤150 mg/day may be continued throughout pregnancy and NICE guidelines (2019) for hypertension in pregnancy advise treatment with LDA (for prophylaxis of pre-eclampsia) until delivery (GRADE 1B, SOA 99.0). LDA is compatible with breastfeeding (GRADE 2C, SOA 99.8). There are limited data on clopidogrel but it may be considered where alternative drugs are not suitable in pregnancy and breastfeeding (GRADE 2C, SOA 96.3).
Discordant findings from retrospective, large studies with population controls on the use of NSAIDs in the first trimester of pregnancy raise the possibility of a low risk of miscarriage and malformation. Therefore, these drugs should only be used intermittently in the first trimester of pregnancy (GRADE 1B, SOA 97.3). Intermittent rather than regular use of all non-selective NSAIDs except LDA is recommended throughout pregnancy and weaned from end of second trimester (26 weeks) to stop by gestational week 30 to avoid premature closure of the ductus arteriosus (GRADE 1B, SOA 98). At present there are limited data on selective COX-2 inhibitors; they should therefore be avoided during pregnancy (GRADE 2C, SOA 98.5). Non-selective NSAIDs (especially ibuprofen) are compatible with breastfeeding (GRADE 1C, SOA 98.8).
LDA of ≤150 mg/day may be continued throughout pregnancy and NICE guidelines (2019) for hypertension in pregnancy advise treatment with LDA (for prophylaxis of pre-eclampsia) until delivery (GRADE 1B, SOA 99.0). LDA is compatible with breastfeeding (GRADE 2C, SOA 99.8). There are limited data on clopidogrel but it may be considered where alternative drugs are not suitable in pregnancy and breastfeeding (GRADE 2C, SOA 96.3).
Recommendations were based on and , as shown in , available at Rheumatology online. Recommendations for colchicine and dapsone in pregnancy and breastfeeding Colchicine therapy may be considered during pregnancy (GRADE 1B, SOA 99.5). Dapsone may be used in pregnancy (GRADE 2C, SOA 95.0). Colchicine may be used in breastfeeding (GRADE 2C, SOA 98.3). Dapsone may be used in breastfeeding and due to the risk of haemolytic anaemia it is advised to monitor the infant for signs of haemolysis, especially in newborn or premature breastfed infants (GRADE 2C, SOA 90.7).
Colchicine therapy may be considered during pregnancy (GRADE 1B, SOA 99.5). Dapsone may be used in pregnancy (GRADE 2C, SOA 95.0). Colchicine may be used in breastfeeding (GRADE 2C, SOA 98.3). Dapsone may be used in breastfeeding and due to the risk of haemolytic anaemia it is advised to monitor the infant for signs of haemolysis, especially in newborn or premature breastfed infants (GRADE 2C, SOA 90.7).
Recommendations were based on [ , , ] and , as shown in , available at Rheumatology online. Recommendations for anticoagulants in pregnancy and breastfeeding Low molecular weight heparin (LMWH) is compatible throughout pregnancy (GRADE 1A, SOA 100). LMWH is compatible with breastfeeding (GRADE 1C, SOA 100). The use of warfarin in pregnancy is associated with increased foetal risk throughout pregnancy and has limited indications, therefore should only be considered in exceptional circumstances (GRADE 1B, SOA 98.8). Warfarin is compatible with breastfeeding (GRADE 1A, SOA 100). Direct oral anticoagulants (DOACs) cannot be recommended in pregnancy (GRADE 1C, SOA 97.9) Rivaroxaban may be considered in breastfeeding (GRADE 2C, SOA 95.3) Other DOACs are not recommended in breastfeeding due to lack to human data and concerns from animal studies (GRADE 1C, SOA 97.4) Fondaparinux may be considered in pregnancy and breastfeeding if there is an allergy or adverse response to LMWH (GRADE 2C, SOA 95.5)
Low molecular weight heparin (LMWH) is compatible throughout pregnancy (GRADE 1A, SOA 100). LMWH is compatible with breastfeeding (GRADE 1C, SOA 100). The use of warfarin in pregnancy is associated with increased foetal risk throughout pregnancy and has limited indications, therefore should only be considered in exceptional circumstances (GRADE 1B, SOA 98.8). Warfarin is compatible with breastfeeding (GRADE 1A, SOA 100). Direct oral anticoagulants (DOACs) cannot be recommended in pregnancy (GRADE 1C, SOA 97.9) Rivaroxaban may be considered in breastfeeding (GRADE 2C, SOA 95.3) Other DOACs are not recommended in breastfeeding due to lack to human data and concerns from animal studies (GRADE 1C, SOA 97.4) Fondaparinux may be considered in pregnancy and breastfeeding if there is an allergy or adverse response to LMWH (GRADE 2C, SOA 95.5)
Recommendations were based on [ , , ] and , as shown in , available at Rheumatology online. Recommendations for bisphosphonates in pregnancy and breastfeeding There is insufficient data upon which to recommend bisphosphonates in pregnancy or to advise a specific time for them to be stopped pre-conception. Given their biological half-life in bone of up to 10 years and no evidence of harm from limited reports of their use in pregnancy, a pragmatic recommendation is that they should be stopped 3 months in advance of pregnancy (GRADE 2C, SOA 96.8). There are no data on which to base a recommendation for the use of bisphosphonates during breastfeeding (GRADE 2C, SOA 98.5).
There is insufficient data upon which to recommend bisphosphonates in pregnancy or to advise a specific time for them to be stopped pre-conception. Given their biological half-life in bone of up to 10 years and no evidence of harm from limited reports of their use in pregnancy, a pragmatic recommendation is that they should be stopped 3 months in advance of pregnancy (GRADE 2C, SOA 96.8). There are no data on which to base a recommendation for the use of bisphosphonates during breastfeeding (GRADE 2C, SOA 98.5).
Recommendations were based on [ , , ] and , as shown in , available at Rheumatology online. Recommendations for angiotensin-converting enzyme inhibitors (ACEis)/angiotensin receptor blockers (ARBs) in pregnancy and breastfeeding ACEi and ARBs should be stopped as soon as possible when pregnancy is confirmed in the first trimester and if necessary an alternative antihypertensive compatible with pregnancy given (GRADE 1A, SOA 100). ACEis/ARBs should be avoided in the second and third trimester but may be considered under specialist advice in certain circumstances (GRADE 1C, SOA 98.5). Based on limited evidence, enalapril is compatible with breastfeeding (GRADE 2C, SOA 98.5).
ACEi and ARBs should be stopped as soon as possible when pregnancy is confirmed in the first trimester and if necessary an alternative antihypertensive compatible with pregnancy given (GRADE 1A, SOA 100). ACEis/ARBs should be avoided in the second and third trimester but may be considered under specialist advice in certain circumstances (GRADE 1C, SOA 98.5). Based on limited evidence, enalapril is compatible with breastfeeding (GRADE 2C, SOA 98.5).
Recommendations were based on [ , , ], as shown in , available at Rheumatology online. Recommendations for calcium channel blockers in pregnancy and breastfeeding Nifedipine is compatible with pregnancy with no direct evidence of harm at doses up to 90 mg/day (GRADE 1A, SOA 99.0). Nifedipine is compatible with breastfeeding (GRADE 1B, SOA 100). Amlodipine can be considered in pregnancy and breastfeeding as there is no evidence of harm (GRADE 1C, SOA 97.9).
Nifedipine is compatible with pregnancy with no direct evidence of harm at doses up to 90 mg/day (GRADE 1A, SOA 99.0). Nifedipine is compatible with breastfeeding (GRADE 1B, SOA 100). Amlodipine can be considered in pregnancy and breastfeeding as there is no evidence of harm (GRADE 1C, SOA 97.9).
Recommendations were based on [ , , , ] and , as shown in , available at Rheumatology online. Recommendations for pulmonary vasodilators in pregnancy and breastfeeding Established moderate-to-severe pulmonary hypertension (PHT) remains a contraindication to pregnancy. If pregnancy occurs, the use of these pulmonary vasodilator drugs in pregnancy should be considered only as part of a multidisciplinary team assessment (GRADE 1C, SOA 99.5). Limited evidence supports the use of prostacyclines to treat PHT during pregnancy (GRADE 2C, SOA 98.0). Limited evidence supports the use of sildenafil to treat PHT during pregnancy (GRADE 2C, SOA 98.0). Bosentan is teratogenic in animals and although there is no evidence of harm from human pregnancy, the evidence is insufficient to recommend in pregnancy (GRADE 1C, SOA 98.8). There are no data relating to breastfeeding exposure to pulmonary vasodilators on which to base a recommendation (GRADE 2C, SOA 98.8).
Established moderate-to-severe pulmonary hypertension (PHT) remains a contraindication to pregnancy. If pregnancy occurs, the use of these pulmonary vasodilator drugs in pregnancy should be considered only as part of a multidisciplinary team assessment (GRADE 1C, SOA 99.5). Limited evidence supports the use of prostacyclines to treat PHT during pregnancy (GRADE 2C, SOA 98.0). Limited evidence supports the use of sildenafil to treat PHT during pregnancy (GRADE 2C, SOA 98.0). Bosentan is teratogenic in animals and although there is no evidence of harm from human pregnancy, the evidence is insufficient to recommend in pregnancy (GRADE 1C, SOA 98.8). There are no data relating to breastfeeding exposure to pulmonary vasodilators on which to base a recommendation (GRADE 2C, SOA 98.8).
Recommendations were based on and , as shown in , available at Rheumatology online. Recommendations for paternal exposure Paracetamol is compatible with paternal exposure (GRADE 1B, SOA 98.5). Amitriptyline, SNRIs and SSRIs are compatible with paternal exposure (GRADE 1B, SOA 98.5). Non-selective NSAIDs are compatible with paternal exposure (GRADE 1C, SOA 98.4). Based on limited or no data and no association with adverse foetal development or pregnancy outcome, paternal exposure to all other drugs described in this guideline is unlikely to be harmful (GRADE 2C, SOA 97.3).
Paracetamol is compatible with paternal exposure (GRADE 1B, SOA 98.5). Amitriptyline, SNRIs and SSRIs are compatible with paternal exposure (GRADE 1B, SOA 98.5). Non-selective NSAIDs are compatible with paternal exposure (GRADE 1C, SOA 98.4). Based on limited or no data and no association with adverse foetal development or pregnancy outcome, paternal exposure to all other drugs described in this guideline is unlikely to be harmful (GRADE 2C, SOA 97.3).
Implementation Awareness of these guidelines will aid clinical practitioners and patients in decision making and will be raised through presentation at local, regional and national meetings. No barriers to implementation of these guidelines are anticipated. Key standards of care Patients with rheumatic disease should receive tailored pre-pregnancy counselling and then be reviewed during pregnancy and the 4-month post-partum period by clinical practitioners with expertise in the management of rheumatic disease in pregnancy, in addition to their routine obstetric care. They should have access to written information on relevant medications in pregnancy and breastfeeding that is accurate and allows them to make informed decisions regarding compatibility of certain drugs in pregnancy. Future research agenda The limitation of current evidence highlights the need for a national pregnancy registry for patients with rheumatic disease, as currently exists for women with epilepsy. All women with rheumatic disease who become pregnant would be eligible to register, whether or not they are on anti-rheumatic treatment. The prospective pregnancy outcome data would then be published to display information on outcomes such as miscarriage and congenital anomalies in patients treated with anti-rheumatic and other drug therapy. These data would also be used to answer specific questions where data is currently lacking. Data relating to the impact of paternal exposure to these drugs (both fertility and male-mediated teratogenicity), as well as breastfeeding exposure is particularly limited, and further research in these areas is urgently required. Mechanism for audit of the guideline An audit pro forma to assess compliance with these guidelines is shown in , available at Rheumatology online. The full guideline is available at Rheumatology online.
Awareness of these guidelines will aid clinical practitioners and patients in decision making and will be raised through presentation at local, regional and national meetings. No barriers to implementation of these guidelines are anticipated.
Patients with rheumatic disease should receive tailored pre-pregnancy counselling and then be reviewed during pregnancy and the 4-month post-partum period by clinical practitioners with expertise in the management of rheumatic disease in pregnancy, in addition to their routine obstetric care. They should have access to written information on relevant medications in pregnancy and breastfeeding that is accurate and allows them to make informed decisions regarding compatibility of certain drugs in pregnancy.
The limitation of current evidence highlights the need for a national pregnancy registry for patients with rheumatic disease, as currently exists for women with epilepsy. All women with rheumatic disease who become pregnant would be eligible to register, whether or not they are on anti-rheumatic treatment. The prospective pregnancy outcome data would then be published to display information on outcomes such as miscarriage and congenital anomalies in patients treated with anti-rheumatic and other drug therapy. These data would also be used to answer specific questions where data is currently lacking. Data relating to the impact of paternal exposure to these drugs (both fertility and male-mediated teratogenicity), as well as breastfeeding exposure is particularly limited, and further research in these areas is urgently required.
An audit pro forma to assess compliance with these guidelines is shown in , available at Rheumatology online. The full guideline is available at Rheumatology online.
are available at Rheumatology online.
keac559_Supplementary_Data Click here for additional data file.
|
British Society for Rheumatology guideline on prescribing drugs in pregnancy and breastfeeding: comorbidity medications used in rheumatology practice | d348ae8e-f0ab-4d9d-86ff-5ca296e8323e | 10070063 | Internal Medicine[mh] | Background The rationale behind this update on the 2016 British Society for Rheumatology (BSR) guidelines on prescribing anti-rheumatic drugs in pregnancy and breastfeeding was described in detail in the guideline scope . In brief, despite the existence of additional evidence-based guidelines on prescribing/managing rheumatic disease in pregnancy the information contained within them requires continual review to include emerging information on the safety of new and existing drugs in pregnancy. Chronic disease adversely affects pregnancy. Data from Mothers and Babies: Reducing Risk through Audits and Confidential Enquiries across the UK (MBRRACE-UK), reports regularly from a national programme of work conducting surveillance and investigating the causes of maternal deaths, stillbirths and infant deaths . Data from 2017–19 found that 8.8 women per 100 000 died during pregnancy or up to six weeks after childbirth or the end of pregnancy, and most women who died had multiple health problems or other vulnerabilities . In all decisions regarding medication choices and changes, it is also important to consider the potential for deterioration in the mother's wellbeing through side effects or reduced disease control (and its adverse impact on the baby). Therefore, the exposure of the foetus to different drugs when switches are made must be balanced against possible foetal gains and understanding the potential impact of reduced control of the medical disorder on a pregnancy is vital . Need for guideline Patients with inflammatory rheumatic disease (IRD) should be counselled to achieve and then maintain remission or low disease activity before/during pregnancy to reduce the risk of adverse pregnancy outcomes . This goal is primarily achieved through adjustment of therapy to ensure disease control with disease modifying anti-rheumatic drugs (DMARDs) and/or immunosuppressive drugs that are compatible with pregnancy. These medications are reviewed in the BSR guideline on prescribing drugs in pregnancy and breastfeeding: immunomodulatory anti-rheumatic drugs and corticosteroids . Many patients with IRD, however, have an additional burden of pain and comorbid illness that require treatment with other medications. The compatibility of various comorbidity medications relevant to rheumatic disease will be covered in this update. This updated information will provide advice for healthcare professionals and patients to ensure more confident prescribing in these scenarios and will highlight any medications that should be stopped and/or avoided in the reproductive age group unless highly effective contraception is used, in line with guidance issued by the Medicines and Healthcare Products Regulatory Agency (MHRA) and Faculty of Sexual and Reproductive Healthcare . Objectives of guideline To update the previous BSR guidelines on prescribing in pregnancy in rheumatic disease for the following drug categories: pain management; NSAIDs and low-dose aspirin (LDA); anticoagulants; colchicine; dapsone; bisphosphonates; anti-hypertensives; and pulmonary vasodilators. This revised guideline was produced by consensus review of current evidence to answer specific questions in relation to each drug as follows. Should it be stopped pre-conception? Is it compatible with pregnancy? Is it compatible with breastmilk exposure? Where possible, recommendations are made regarding compatibility with paternal exposure. Target audience The primary audience consists of health professionals in the UK directly involved in managing patients with rheumatic disease who are (or are planning to become) pregnant and/or breastfeeding, men planning to conceive, and patients who have unintentionally conceived while taking these medications. This audience includes rheumatologists, rheumatology nurses/allied health professionals, rheumatology speciality trainees and pharmacists, as well as the patients themselves. The guideline will also be useful to obstetricians, obstetric physicians, renal physicians, dermatologists and general practitioners who may prescribe these medications to patients in pregnancy. This guideline uses the terms ‘woman’, ‘maternal’ or ‘mother’ throughout. These should be taken to include people who do not identify as women but are pregnant or have given birth . Where the term ‘breastfeeding’ is used in this guideline it also refers to infant breastmilk exposure via other methods (e.g. expressed breastmilk, administered via a bottle). The areas the guideline does not cover This guideline does not cover the management of infertility or acute pain relief during labour, hence morphine was excluded. Other drug categories: antimalarials; corticosteroids; disease modifying anti-rheumatic and immunosuppressive therapies; and biologic drugs are considered in another guideline . All recommendations in this guideline were formulated by the working group on the basis of published evidence at the time of the systematic literature search, and do not necessarily refer to licensing information or Summary of Product Characteristics for individual medications. Stakeholder involvement This guideline was commissioned by the BSR Standards, Audit and Guidelines Working Group. A Guideline Working group (GWG) was created, consisting of a chair (I.G.), alongside representatives from relevant stakeholders ( ). In accordance with BSR policy, all members of the GWG made declarations of interest, available on the BSR website. Involvement and affiliations of stakeholder groups involved in guideline development The GWG consisted of rheumatologists from a range of clinical backgrounds, various allied health professionals, other specialists in women’s health, lay members and representatives from the United Kingdom Tetralogy Information Service (UKTIS). All members of the working group contributed to the process for agreeing key questions, guideline content, recommendations and strength of agreement.
The rationale behind this update on the 2016 British Society for Rheumatology (BSR) guidelines on prescribing anti-rheumatic drugs in pregnancy and breastfeeding was described in detail in the guideline scope . In brief, despite the existence of additional evidence-based guidelines on prescribing/managing rheumatic disease in pregnancy the information contained within them requires continual review to include emerging information on the safety of new and existing drugs in pregnancy. Chronic disease adversely affects pregnancy. Data from Mothers and Babies: Reducing Risk through Audits and Confidential Enquiries across the UK (MBRRACE-UK), reports regularly from a national programme of work conducting surveillance and investigating the causes of maternal deaths, stillbirths and infant deaths . Data from 2017–19 found that 8.8 women per 100 000 died during pregnancy or up to six weeks after childbirth or the end of pregnancy, and most women who died had multiple health problems or other vulnerabilities . In all decisions regarding medication choices and changes, it is also important to consider the potential for deterioration in the mother's wellbeing through side effects or reduced disease control (and its adverse impact on the baby). Therefore, the exposure of the foetus to different drugs when switches are made must be balanced against possible foetal gains and understanding the potential impact of reduced control of the medical disorder on a pregnancy is vital .
Patients with inflammatory rheumatic disease (IRD) should be counselled to achieve and then maintain remission or low disease activity before/during pregnancy to reduce the risk of adverse pregnancy outcomes . This goal is primarily achieved through adjustment of therapy to ensure disease control with disease modifying anti-rheumatic drugs (DMARDs) and/or immunosuppressive drugs that are compatible with pregnancy. These medications are reviewed in the BSR guideline on prescribing drugs in pregnancy and breastfeeding: immunomodulatory anti-rheumatic drugs and corticosteroids . Many patients with IRD, however, have an additional burden of pain and comorbid illness that require treatment with other medications. The compatibility of various comorbidity medications relevant to rheumatic disease will be covered in this update. This updated information will provide advice for healthcare professionals and patients to ensure more confident prescribing in these scenarios and will highlight any medications that should be stopped and/or avoided in the reproductive age group unless highly effective contraception is used, in line with guidance issued by the Medicines and Healthcare Products Regulatory Agency (MHRA) and Faculty of Sexual and Reproductive Healthcare .
To update the previous BSR guidelines on prescribing in pregnancy in rheumatic disease for the following drug categories: pain management; NSAIDs and low-dose aspirin (LDA); anticoagulants; colchicine; dapsone; bisphosphonates; anti-hypertensives; and pulmonary vasodilators. This revised guideline was produced by consensus review of current evidence to answer specific questions in relation to each drug as follows. Should it be stopped pre-conception? Is it compatible with pregnancy? Is it compatible with breastmilk exposure? Where possible, recommendations are made regarding compatibility with paternal exposure.
The primary audience consists of health professionals in the UK directly involved in managing patients with rheumatic disease who are (or are planning to become) pregnant and/or breastfeeding, men planning to conceive, and patients who have unintentionally conceived while taking these medications. This audience includes rheumatologists, rheumatology nurses/allied health professionals, rheumatology speciality trainees and pharmacists, as well as the patients themselves. The guideline will also be useful to obstetricians, obstetric physicians, renal physicians, dermatologists and general practitioners who may prescribe these medications to patients in pregnancy. This guideline uses the terms ‘woman’, ‘maternal’ or ‘mother’ throughout. These should be taken to include people who do not identify as women but are pregnant or have given birth . Where the term ‘breastfeeding’ is used in this guideline it also refers to infant breastmilk exposure via other methods (e.g. expressed breastmilk, administered via a bottle).
This guideline does not cover the management of infertility or acute pain relief during labour, hence morphine was excluded. Other drug categories: antimalarials; corticosteroids; disease modifying anti-rheumatic and immunosuppressive therapies; and biologic drugs are considered in another guideline . All recommendations in this guideline were formulated by the working group on the basis of published evidence at the time of the systematic literature search, and do not necessarily refer to licensing information or Summary of Product Characteristics for individual medications.
This guideline was commissioned by the BSR Standards, Audit and Guidelines Working Group. A Guideline Working group (GWG) was created, consisting of a chair (I.G.), alongside representatives from relevant stakeholders ( ). In accordance with BSR policy, all members of the GWG made declarations of interest, available on the BSR website.
The GWG consisted of rheumatologists from a range of clinical backgrounds, various allied health professionals, other specialists in women’s health, lay members and representatives from the United Kingdom Tetralogy Information Service (UKTIS). All members of the working group contributed to the process for agreeing key questions, guideline content, recommendations and strength of agreement.
Statement of scope of literature search and strategy employed Most medications covered in this guideline have been comprehensively and systematically reviewed in multiple other documents, since the first BSR guideline on this topic. Therefore, a consensus-based approach was taken to compile and assess most significant evidence published since 2013 to December 2020 through a comprehensive search of MEDLINE, PubMed and EMBASE databases with specific search terms ( , available at Rheumatology online). Filters were applied to capture National Institute for Health and Care Excellence (NICE) guidance, international guidelines, systematic reviews, cohort studies or case series. Information was preferentially selected from NICE guidance and/or largest/most recent systematic reviews and where lacking was extracted from largest cohort, case series or abstract. Findings were cross-referenced with the previous BSR guideline , as well as the Cochrane, Lactmed (a National Library of Medicine database on drugs and lactation) and UKTIS databases. Two independent reviewers screened the title and abstract of 2997 articles, identified 130 and selected the most recent/largest systematic reviews or largest cohort study or case series as well as any NICE guidance and international guidelines. Thirty-six studies ( ) met the inclusion criteria and relevant information was extracted into data-extraction tables. Statement of methods used to formulate the recommendations (levels of evidence) The working group met regularly to formalise search strategy, review evidence, resolve disagreements and finally to determine recommendations. This guideline was developed in line with BSR’s Guidelines Protocol using Grading of Recommendations, Assessment, Development and Evaluations (GRADE) methodology to determine quality of evidence and strength of recommendation. Accompanying each recommendation in this guideline, in brackets, is the strength of recommendation, quality of evidence and strength of agreement (SOA). Strength of recommendation Using GRADE, recommendations were categorized as either strong (denoted by 1) or weak (denoted by 2), according to the balance between benefits and risks. A strong recommendation was made when the benefits clearly outweigh the risks (or vice versa). A weak recommendation denotes that the benefits are more closely balanced with the risk, or more uncertain. Quality of evidence Using the GRADE approach, the quality of evidence was determined as either high (A), moderate (B) or low/very low (C) reflecting the confidence in the estimates of benefits or harm. Strength of agreement The wording of each recommendation was discussed until all members were satisfied they would score at least 80 on a scale of 1 (no agreement) to 100 (complete agreement) and then 20/23 members with full voting rights scored each recommendation on the same scale and the average was calculated to generate a strength of agreement (SOA) score. Two patient representatives and a data-analyst expressed concern that they did not have sufficient medical knowledge of all drugs reviewed to score all recommendations, so while they fully agreed with each, they did not wish to score each one and did not contribute to the final SOA score. Statement of any limits of search and when guideline will be updated The search was conducted in December 2020. Limits were placed for English language and filters as described above. The guideline will be updated in five years.
Most medications covered in this guideline have been comprehensively and systematically reviewed in multiple other documents, since the first BSR guideline on this topic. Therefore, a consensus-based approach was taken to compile and assess most significant evidence published since 2013 to December 2020 through a comprehensive search of MEDLINE, PubMed and EMBASE databases with specific search terms ( , available at Rheumatology online). Filters were applied to capture National Institute for Health and Care Excellence (NICE) guidance, international guidelines, systematic reviews, cohort studies or case series. Information was preferentially selected from NICE guidance and/or largest/most recent systematic reviews and where lacking was extracted from largest cohort, case series or abstract. Findings were cross-referenced with the previous BSR guideline , as well as the Cochrane, Lactmed (a National Library of Medicine database on drugs and lactation) and UKTIS databases. Two independent reviewers screened the title and abstract of 2997 articles, identified 130 and selected the most recent/largest systematic reviews or largest cohort study or case series as well as any NICE guidance and international guidelines. Thirty-six studies ( ) met the inclusion criteria and relevant information was extracted into data-extraction tables.
The working group met regularly to formalise search strategy, review evidence, resolve disagreements and finally to determine recommendations. This guideline was developed in line with BSR’s Guidelines Protocol using Grading of Recommendations, Assessment, Development and Evaluations (GRADE) methodology to determine quality of evidence and strength of recommendation. Accompanying each recommendation in this guideline, in brackets, is the strength of recommendation, quality of evidence and strength of agreement (SOA).
Using GRADE, recommendations were categorized as either strong (denoted by 1) or weak (denoted by 2), according to the balance between benefits and risks. A strong recommendation was made when the benefits clearly outweigh the risks (or vice versa). A weak recommendation denotes that the benefits are more closely balanced with the risk, or more uncertain.
Using the GRADE approach, the quality of evidence was determined as either high (A), moderate (B) or low/very low (C) reflecting the confidence in the estimates of benefits or harm.
The wording of each recommendation was discussed until all members were satisfied they would score at least 80 on a scale of 1 (no agreement) to 100 (complete agreement) and then 20/23 members with full voting rights scored each recommendation on the same scale and the average was calculated to generate a strength of agreement (SOA) score. Two patient representatives and a data-analyst expressed concern that they did not have sufficient medical knowledge of all drugs reviewed to score all recommendations, so while they fully agreed with each, they did not wish to score each one and did not contribute to the final SOA score.
The search was conducted in December 2020. Limits were placed for English language and filters as described above. The guideline will be updated in five years.
Drugs are considered in the following categories: pain management; NSAIDs and LDA in the management of multisystem rheumatic disease; anticoagulants; bisphosphonates; anti-hypertensive medication in the management of multisystem rheumatic disease; and pulmonary vasodilators. The overall findings for maternal and foetal breastmilk exposures to each drug, including information and key references from the previous BSR guideline are summarised and recommendations updated accordingly. Paternal exposures and recommendations are described separately after maternal data. An overall summary of compatibility of each drug pre-conception, during pregnancy, breastmilk exposure and paternal exposure is shown in . Generic recommendations were developed based on evidence as shown in , available at Rheumatology online. Generic recommendations on prescribing in rheumatic disease in pregnancy Pre-conception counselling should be addressed by all healthcare professionals, with referral to professionals with relevant experience as appropriate to optimize all therapy, including non-pharmacological options for chronic pain management during pregnancy (GRADE 1A, SOA 99.5). The risks and benefits of drug treatment to mother and foetus should be discussed and clearly documented by all healthcare professionals involved in the patient’s care (GRADE 1A, SOA 99). The cause of pain and other symptoms should be assessed and managed appropriately (GRADE 1B, SOA 98.5). The requirement for analgesia should be assessed and minimum effective dose should be prescribed and titrated according to response (GRADE 1B, SOA 100). Tricyclic antidepressants are preferred over other antidepressant medications to manage chronic pain (GRADE 1B, SOA 98.1). Cessation of anti-depressant therapy that is being used as chronic pain medication in the post-natal period is not recommended, due to the risk of adverse impact on mood (GRADE 1C, SOA 96). LDA (≤150 mg/day) is recommended in all patients at high risk for pre-eclampsia (GRADE 1A, SOA 99.5). Low molecular weight heparin (LMWH) is the preferred anticoagulant (GRADE 1A, SOA 100). Nifedipine is the preferred vasodilator (GRADE 1B, SOA 98.5). Paternal drug exposure may reduce male fertility but has not been associated with adverse foetal development or pregnancy outcome. Although evidence is weak, we recommend that men are reassured about the safety of fathering a pregnancy while taking medicines to manage comorbidities as described in this guideline (GRADE 1C, SOA 98). Pain management: conventional analgesics Paracetamol Two systematic reviews were selected. Overall, paracetamol was considered to have a favourable safety profile in pregnancy. Potential links between paracetamol use in pregnancy and pre-term birth and adverse neurodevelopmental outcomes (principally autism and attention deficit hyperactivity disorder) were confounded by maternal disease and selection bias . UKTIS states that ‘In most studies, risk of abnormal neurodevelopment correlated with duration of paracetamol exposure. However, significant methodological limitations of these studies limit the conclusions that can be drawn, and a causal association remains unproven’ . There were no convincing associations with congenital malformations. UKTIS notes that findings of a possible increased risk of cryptorchidism in male offspring following paracetamol use during pregnancy are conflicting . Current Royal College of Obstetricians (RCOG) guidance is that paracetamol remains safe for use during pregnancy and breastfeeding, and its use in any trimester does not appear to increase the risk of major birth defects . There were conflicting results on the risk of developing wheeze in infants exposed to paracetamol during pregnancy. A recent meta-analysis of 13 articles and 1 043 109 individuals identified a statistically significant association between prenatal paracetamol exposure and increased risk of child asthma [odds ratio (OR) 1.19; 95% CI, 1.12, 1.27; P < 0.00001] in a random-effect model . This significant association was observed for first trimester as well as second and third trimester exposure. There were no studies identified which specifically examined neonatal outcomes after drug exposure in breastmilk. LactMed reports low amounts of this drug in breastmilk at levels much less than doses usually given to infants with few reports of adverse events. The following recommendations for paracetamol were developed based on evidence as shown in , available at Rheumatology online. Recommendations for paracetamol in pregnancy and breastfeeding Paracetamol is the analgesic of choice and compatible peri-conception and throughout pregnancy (GRADE 1B, SOA 99). LactMed describes paracetamol as a good choice for analgesia and fever reduction in breastfeeding mothers (GRADE 2C, SOA 99.5). Codeine One systematic review considered outcomes from opioid-exposed pregnancies, including three studies of congenital malformation in >20 000 codeine-exposed pregnancies with conflicting results. Of these three studies, a large Norwegian population-based cohort study found no statistically significant associations between neonatal codeine exposure and congenital malformations. In contrast, a large case-control study found that first trimester exposure to opioids (codeine and hydrocodone in 69% of cases) was significantly associated with congenital heart defects, spina bifida and gastroschisis , although there was a high risk for recall bias in this study. A case-control study found a significant association between spina bifida in infants and first trimester maternal opioid use, although it was not with codeine use specifically and potential confounders such as illicit drug use/dependence and sociodemographic data exist in this study . UKTIS notes that the methodological limitations of studies make it difficult to draw any firm conclusions and more robust research is needed before firm conclusions regarding the risk of congenital anomalies with codeine use can be provided . No convincing associations between first trimester opioid use and miscarriage have been found. A large population study found increased rates of premature delivery, reduced birthweight and pre-eclampsia in codeine-exposed patients compared with controls that was considered to be due to confounding factors such as maternal disease and lack of adjustment for gestational age . This study also found a significant increase in post-partum haemorrhage (18.3% vs 14.5%; OR 1.3; 95% CI 1.1, 1.5) in 2666 mothers exposed to codeine and hypothesized it was due to an opioid effect weakening myometrial contraction, but the precise mechanism is unknown. UKTIS describes theoretical concerns that maternal use of codeine near term may be associated with respiratory depression in the neonate and notes that the only study that has investigated the risk of neonatal respiratory depression found no increased risk . The potential risk of neonatal abstinence syndrome (NAS) is lower with short courses of short-acting opioids and must be weighed against the benefits for treating acute pain that often outweigh risks. Prolonged opioid use, however, should be avoided and if used in the last trimester, neonatologists should be advised due to the risks of NAS . Controversy remains over whether codeine is safe in breastfeeding. Central nervous system (CNS) depression was reported by mothers in 16.7% (35/210) babies exposed to codeine, compared with 0.5% exposed to paracetamol . In the same study there was one neonatal death and high morphine levels were found at post-mortem. The mother had received high doses of codeine (>2 mg/kg/day) and was subsequently found to be an ultra-rapid CYP2D6 metaboliser. Another study demonstrated dose-dependent CNS depression in 24% (17/72) of infants exposed to codeine through breastmilk . A large study of 7804 infants reported conflicting results, but specifically, there was no difference in poor Apgar scores, postnatal complications, admission to special care baby units, readmission to hospital, resuscitation or death in infants exposed and not exposed to codeine . LactMed notes that numerous professional organizations and regulatory agencies recommend that other agents are preferred over codeine during breastfeeding but that other opioid alternatives have been less studied and may not be safer . It is acknowledged that due to its unpredictable metabolism, administration of codeine results in delivery of an unknown quantity of morphine. Therefore, despite its widespread use and probable safety in most cases, we would advise caution with prolonged use of codeine in breastfeeding and appropriate advice to the mother to seek medical attention if she has any concerns regarding lethargy or drowsiness in her child. The following recommendations for codeine were based on evidence as shown in , available at Rheumatology online. Recommendations for codeine in pregnancy and breastfeeding Codeine is compatible peri-conception and throughout pregnancy, although long-term use should be avoided. There is no consistent evidence to recommend a dose reduction pre-delivery but neonatologists should be aware of maternal use (GRADE 1B, SOA 97.8). Caution is advised with use of codeine in breastfeeding, due to the risk of CNS depression resulting from unpredictable metabolism of codeine to morphine (GRADE 1C, SOA 98). Tramadol A systematic review considered outcomes from tramadol-exposed pregnancies. It included a large prospective cohort study of over 1.6 million women that reported a statistically significant association between tramadol use in pregnancy and major congenital malformations (OR 1.33; 95% CI 1.05, 1.70) and cardiovascular defects (OR 1.56; 95% CI 1.04, 2.29), although it did not adjust for indication, duration or dose of medication. UKTIS notes that a single study identified an increased risk of miscarriage among women who used tramadol in early pregnancy and recommended further studies considering the impact of confounding factors to clarify this finding. A single cohort study found no increased risk of preterm delivery within 146 pregnancies exposed to tramadol in at least the first trimester . A small number of case reports have reported NAS with long-term intrauterine exposure to tramadol , although none have compared the relative rate of NAS with tramadol compared with other opioid analgesics. LactMed states that the excretion of tramadol into milk is low and even lower amounts of the active metabolite, O-desmethyltramadol, are excreted and a study of breastfeeding in newborn infants found no adverse effects attributable to tramadol compared with controls . There has been one death, however, in the 8-month-old breastfed infant of a woman addicted to tramadol, although the death was not definitely attributable to tramadol exposure in breastmilk . Current RCOG advice states that tramadol can continue to be used (with caution) during breastfeeding and the lowest effective dose should be used for the shortest time possible . The following recommendations for tramadol were based on evidence as shown in , available at Rheumatology online. Recommendations for tramadol in pregnancy and breastfeeding Avoid tramadol peri-conception and in first trimester and only consider in second/third trimester if no alternative analgesia (GRADE 2B, SOA 97.8). Based on limited data, tramadol may be compatible with short-term use in breastfeeding (GRADE 2C, SOA 94.8). Other treatments for chronic pain Amitriptyline NICE guidance and a systematic review described amitriptyline use in pregnancy. Notably, the evidence base underlying amitriptyline use in pregnancy comes from its use as a tricyclic antidepressant (TCA) to treat depression at doses of 150–300 mg and evidence of its use in pregnancy to treat chronic widespread pain at 75 mg per day or less is lacking. In addition, most studies report on TCAs as a drug class and although they include (some) women exposed to amitriptyline, they lack a separate assessment of their pregnancy outcomes so do not provide information about specific amitriptyline exposure . Overall, no increased risk of congenital malformations has been found for TCAs . UKTIS reports conflicting findings, with possible associations with spontaneous abortion, preterm delivery, and autism spectrum disorder identified in some (but not all) studies . There is an association between discontinuation of antidepressants and a high risk for relapse of mood disorders that can adversely impact on pregnancy. It is unknown, however, whether a similar phenomenon of rebound pain exists when antidepressants used for analgesia are discontinued in pregnancy, particularly for amitriptyline that is used at 2–4-fold higher doses to treat depression compared with chronic pain. If a decision is made to stop amitriptyline in pregnancy, the dose should be tapered gradually where possible . Low levels of amitriptyline and its metabolites are reported in breastmilk with no adverse effects described with limited follow-up, summarised in LactMed . The following recommendations were based on evidence as shown in , available at Rheumatology online. Recommendations for amitriptyline in pregnancy and breastfeeding Amitriptyline is compatible with pregnancy. There is no evidence of adverse effect on IQ or developmental outcomes (GRADE 1C, SOA 100). Because very little amitriptyline is found in breastmilk with antidepressant doses and it is used at lower doses for chronic pain, it is unlikely to cause adverse effects in breastfed infants (GRADE 1C, SOA 100). Gabapentin and pregabalin Data on gabapentin use in pregnancy comes mostly from studies of the treatment of maternal epilepsy at doses generally higher that those used to treat chronic pain, while information on pregabalin use in pregnancy comes from studies of its use to treat neuropathic pain . A systematic review of gabapentin use in pregnancy to treat epilepsy did not find any evidence of an increased rate of major malformations or other adverse outcomes attributable to gabapentin exposure. No long-term outcomes were reported. Furthermore, a systematic review and meta-analysis published in abstract form only , of eight cohort studies (four prospective and four retrospective), reporting 5 072 286 unexposed and 949 exposed pregnancies found that first trimester exposure to gabapentin was not statistically significantly associated with major congenital malformations (OR 0.83; 95% CI 0.45, 1.53). A prospective cohort study using Teratology Information Services data from seven European countries found that first trimester use of pregabalin was associated with significantly higher rates of major congenital malformations when compared with unexposed pregnancies (OR 3.0; 95% CI 1.2, 7.9). This study, however, was limited by a small sample size (164 exposed pregnancies) and lack of adjustment for potential confounding by use of concomitant medications . A systematic review of pregabalin use, mostly to treat neuropathic pain in pregnancy, including data from this study and two others identified 651 pregnancy exposures and concluded that pregabalin exposure during pregnancy is not devoid of structural teratogenicity potential . UKTIS reviewed data from six controlled studies, with outcomes of >3000 exposed pregnancies of patients with chronic pain or other non-epilepsy indications. They conclude that this data does not indicate that maternal pregabalin use in pregnancy is associated with increased risks of malformation, miscarriage or adversely affects foetal growth . Interestingly, a systematic review and meta-analysis published in abstract form only , of six studies reporting 2319 exposed pregnancies and 4 982 778 unexposed pregnancies found that first trimester exposure to pregabalin was not significantly associated with an increased risk of major congenital malformations (OR 1.20; 95% CI 0.92, 1.57). Following completion of our guidelines, the MHRA issued a safety warning on the use of pregabalin in pregnancy based on new data available online . This new data has been considered in the latest UKTIS update in April 2022 that states that the available data do not provide conclusive evidence that maternal pregabalin use in the first trimester, or at any stage of pregnancy, is associated with increased risks of either overall malformation or any specific malformations. Therefore, our recommendations have not changed. UKTIS found no controlled studies of the risk of neonatal complication following prenatal gabapentin or pregabalin exposure, although one study has described a small number of affected infants, including one case of neonatal withdrawal following in-utero pregabalin exposure . Use of any centrally acting drug throughout pregnancy or near delivery may be associated with withdrawal symptoms in the neonate and/or NAS. Although evidence is lacking and it is uncertain if pregabalin or gabapentin impact upon maternal folate status, UK guidelines state that women who take any anti-epileptic medication should be prescribed high-dose folic acid (5 mg/day) preconceptually and in the first trimester . There remains limited data on use of these drugs in breastfeeding. According to LactMed, low levels of gabapentin and pregabalin have been found in breastmilk with no adverse effects on infants reported in limited case reports/series ( n < 10) for gabapentin. Therefore, both drugs may be considered if required by the breastfeeding mother . The following recommendations were based on evidence as shown in , available at Rheumatology online. Recommendations for gabapentin and pregabalin in pregnancy and breastfeeding Gabapentin at lowest effective dose may be considered in pregnancy with folic acid supplementation if no alternative analgesic is suitable (GRADE 1B, SOA 95). Gabapentin may be considered in breastfeeding if no alternative analgesic is suitable (GRADE 2C, SOA 96). Pregabalin may be considered in pregnancy (with folic acid supplementation) and during breastfeeding (GRADE 2C, SOA 95.3). Serotonin–norepinephrine reuptake inhibitors (SNRIs) NICE guidance and two systematic reviews described SNRI use in pregnancy. There was no association between first trimester exposure to venlafaxine and an increased risk for major congenital malformations . UKTIS notes that although there are some reports of an increased risk of miscarriage following gestational exposure to venlafaxine, the data are inconsistent and likely confounded by indication and other factors . Some studies have found a possible association with an increased risk for some perinatal complications, including a withdrawal syndrome with venlafaxine use in the third trimester . A more limited data set for duloxetine does not suggest a clinically important increased risk for major malformations but has identified prenatal antidepressant exposure syndrome in two of five case reports and conflicting reports of increased rates of gestational hypertension and spontaneous abortion . LactMed reports that infants receive venlafaxine and its active metabolite in breastmilk, and the metabolite of the drug can be found in the plasma of most breastfed infants, but no proven drug-related side effects have been reported in small case series. Little published information is available on the use of duloxetine during breastfeeding; however, the dose in milk is low and serum levels were low in two breastfed infants . The following recommendations were based on evidence as shown in , available at Rheumatology online. Recommendations for SNRIs in pregnancy and breastfeeding Venlafaxine is compatible at conception and throughout pregnancy. There may be an increased risk of neonatal abstinence syndrome/short-term behavioural effects, but larger studies are needed to evaluate this finding (GRADE 2C, SOA 95.8). Duloxetine may be considered in pregnancy and breastfeeding but there are fewer data than for venlafaxine (GRADE 2C, SOA 95.3). Venlafaxine and duloxetine may be considered in breastfeeding if there is no alternative chronic pain medication (GRADE 2C, SOA 95.8). Selective serotonin reuptake inhibitors (SSRIs) NICE guidance and a systematic review described SSRI use to treat depression in pregnancy. The SSRIs are used to treat chronic pain at similar doses used to treat depression. A systematic review considered data from >50 000 infants exposed to SSRIs in utero , which did not show an overall increased risk for congenital malformations . It describes ongoing debate about the risk for cardiovascular malformations with first-trimester use of SSRIs with risk found from some but not all studies . Overall, any increase in absolute risk was thought unlikely to be clinically significant and may be associated with particular SSRIs, principally fluoxetine and paroxetine . The UKTIS summary of findings to date from studies on SSRIs states that a causal association between use of SSRIs in pregnancy and any type of congenital malformation has not been confirmed. It also describes conflicting results from other outcomes and concludes that available data do not suggest that SSRI use in pregnancy increases the risk of stillbirth and that possible associations with neurodevelopmental impairment in infants requires further study. An increased risk, however, of persistent pulmonary hypertension (PPHN) of the newborn has also been reported following exposure to SSRIs as a class beyond 20 weeks of gestation and, although it remains an uncommon event (0.2–1.2% vs 0.1–0.2% in the background population), it represents a potentially serious neonatal complication . Because there is no robust evidence of a superior safety profile for any one drug, switching between drugs is not recommended if depression is stable on treatment. NICE recommendations on use of TCAs, SSRIs or SNRIs include consideration of: the uncertainty about whether any increased risk to the foetus and other problems for the woman or baby can be attributed directly to these drugs or may be caused by other factors; and the risk of discontinuation symptoms in the woman and neonatal adaptation syndrome in the baby with most TCAs, SSRIs and (S)NRIs, in particular paroxetine and venlafaxine . There is limited information on the use of these drugs in breastfeeding. One small study showed temporarily reduced growth during exposure to fluoxetine in breastmilk. There have been no studies specifically investigating compatibility of paroxetine and sertraline with breastfeeding, but sertraline is reported as having one of the lowest rates of transmission to breastmilk . The following recommendations were based on evidence as shown in , available at Rheumatology online. Recommendations for SSRIs in pregnancy and breastfeeding Fluoxetine, paroxetine and sertraline are compatible with pregnancy (GRADE 1B, SOA 98.8). Based on limited evidence, SSRIs are compatible with breastfeeding (GRADE 2C, SOA 98.3). NSAIDs and anti-platelet drugs Non-selective cyclooxygenase (COX) inhibitors have different indications in pregnancy. NSAIDs have analgesic and anti-inflammatory actions mediated through peripheral inhibition and differential selectivity of COX enzymes. In contrast, LDA at doses ≤150 mg/day is used to prevent thrombosis and pre-eclampsia in high-risk groups throughout pregnancy in patients with rheumatic diseases. Clopidogrel is an anti-platelet agent that may sometimes be used with or instead of LDA to reduce cardiovascular risk. NICE guidance , four systematic reviews and a case report/review evaluated LDA, NSAID and clopidogrel use in pregnancy. NSAIDs and COX-2 inhibitors Overall, there is no consistent evidence for an increased risk of teratogenic effects with NSAID use in pregnancy. There are mixed findings regarding a potential increased risk for miscarriage, with findings limited by methodology and larger associations reported for indomethacin and diclofenac use in the periconceptional period . There was no information on safety of COX2 inhibitors in pregnancy. NSAIDs are reported to increase the incidence of luteinized unruptured follicle (LUF) syndrome, whereby an anovulatory cycle results due to failure of normal follicular wall rupture despite normal ovarian follicular development and elevation of serum progesterone. COX-2 is active in the ovaries during follicular development; thus, inhibition via COX-2 inhibitors is thought to result in LUF. Although similar findings have been reported for both COX-1 and COX-2 NSAIDs, the risks have been found to be greater in patients with inactive disease and in those taking a COX-2 inhibitor (etoricoxib) rather than non-selective NSAIDs and it is reversible, following drug withdrawal . A systematic review of the effects of various drugs on foetal cardiac function evaluated by ultrasound found that all NSAIDs (including COX2 inhibitors) increased constriction in the ductus arteriosus, within 4–30 h of exposure and resolved by 72 h of discontinuation . In this study the critical gestational age (measured for indomethacin) increased from 5–10% of foetuses at weeks 26–27 to 50% at week 32. In 2020 the United States Food and Drug Administration (FDA) recommended that all NSAIDs be avoided from gestational week 20 rather than the previously advised 30 weeks. This advice was based on their updated review of published data to 2016 and 35 cases reported to FDA identifying an increased risk of oligohydramnios and renal impairment that began at 20 weeks of gestation and were mostly reversible on stopping NSAID. They clarify their advice for healthcare professionals, stating that the use of NSAIDs between 20 and 30 weeks of pregnancy should be limited to the lowest effective dose for the shortest duration . Low-dose aspirin LDA has been extensively studied and shown to improve outcomes in high-risk pregnancies. A Cochrane review of 77 trials, involving 40 249 women and their babies, found high quality evidence that antiplatelet agents (mostly LDA up to 150 mg/day) reduced pre-eclampsia and its complications . A systematic review of 22 RCTS of LDA plus heparin compared with other treatments in patients with APS, including 1515 treatment and 1531 control subjects, found that adverse pregnancy outcomes were significantly improved with LDA and heparin . The use of LDA in the third trimester of pregnancy is not associated with premature closure of ductus arteriosus and NICE guidelines for management of hypertension in pregnancy advises treatment with LDA until delivery . Clopidogrel There was no demonstrable maternofoetal toxicity in 13 (mostly second and third trimester only) pregnancy exposures to clopidogrel . UKTIS does not report on clopidogrel. LactMed considers various non-selective NSAIDs to be acceptable during breastfeeding and prefers ibuprofen because of its extremely low levels in breastmilk, short half-life and safe use in infants in doses much higher than those excreted in breastmilk, as an analgesic or anti-inflammatory agent in breastfeeding mothers. There was no information on COX2 inhibitors. Aspirin doses up to 325 mg daily are not excreted into breastmilk so LDA may be considered as an antiplatelet drug for use in breastfeeding women and is preferred to clopidogrel as no information is available on this drug . Recommendations for these drugs were based on evidence as shown in , available at Rheumatology online. Recommendations for NSAIDs and COX-2 inhibitors in pregnancy and breastfeeding Discordant findings from retrospective, large studies with population controls on the use of non-selective NSAIDs in the first trimester of pregnancy raise the possibility of a low risk of miscarriage and malformation. Therefore, these drugs should only be used intermittently in the first trimester of pregnancy (GRADE 1B, SOA 97.3). Intermittent rather than regular use of all non-selective NSAIDs except LDA is recommended throughout pregnancy and weaned from end of second trimester (26 weeks) to stop by gestational week 30 to avoid premature closure of the ductus arteriosus (GRADE 1B, SOA 98). At present there are limited data on selective cyclooxygenase-2 inhibitors; they should therefore be avoided during pregnancy (GRADE 2C, SOA 98.5). Non-selective NSAIDs (especially ibuprofen) are compatible with breastfeeding (GRADE 1C, SOA 98.8). Recommendations for LDA and clopidogrel in pregnancy and breastfeeding LDA may be continued throughout pregnancy and NICE guidelines (2019) for hypertension in pregnancy advise treatment with LDA (for prophylaxis of pre-eclampsia) until delivery (GRADE 1B, SOA 99.0). LDA is compatible with breastfeeding (GRADE 2C, SOA 99.8). There are limited data on clopidogrel but it may be considered where alternative drugs are not suitable in pregnancy and breastfeeding (GRADE 2C, SOA 96.3). Colchicine and dapsone These drugs were not considered in the previous BSR guideline and are now included because they may be used to treat certain inflammatory rheumatic diseases. UKTIS does not report on colchicine or dapsone. A systematic review and meta-analysis of colchicine use in 550 pregnancies of women with mostly familial Mediterranean fever (FMF) at doses of 1–2 mg per day, compared with 1263 non-exposed pregnancies found this drug did not significantly increase the incidence of foetal malformations or miscarriage when taken during pregnancy . However, the National Amyloidosis Centre recommends to continue the prescribed dose as there are no established safety concerns at colchicine doses >2 mg daily during pregnancy . No systematic review data was identified for dapsone. A review of 924 pregnancies exposed to dapsone to treat malaria was precluded from meaningful risk-benefit analysis due to limited reporting of outcomes . They concluded that the use of dapsone may be considered when no suitable alternative is available and the threat of malaria is the greater risk. It is a safe option in pregnant patients without glucose-6-phosphate dehydrogenase (G6PDH) deficiency and can be used during lactation while monitoring the baby for haemolysis and G6PDH deficiency . A review of the treatment of rheumatic and autoimmune skin disease in women during pregnancy concluded that dapsone may be safely and cautiously used during pregnancy . LactMed reports that long-term prophylactic maternal doses of colchicine up to 1.5 mg daily produce levels in milk that result in the infant receiving <10% of the maternal weight-adjusted dosage and no adverse effects have been reported from limited studies. It also states that the highest milk levels occur 2–4 h after a dose, so avoiding breastfeeding during this time can minimize the infant dose, or simply taking the drug after nursing. LactMed also states that dapsone can be used during breastfeeding; however, haemolytic anaemia might occur, especially in newborn infants and in those with G6PDH deficiency . Recommendations for these drugs were based on evidence as shown in , available at Rheumatology online. Recommendations for colchicine and dapsone in pregnancy and breastfeeding Colchicine therapy may be considered during pregnancy (GRADE 1B, SOA 99.5). Dapsone may be used in pregnancy (GRADE 2C, SOA 95.0). Colchicine may be used in breastfeeding (GRADE 2C, SOA 98.3). Dapsone may be used in breastfeeding and due to the risk of haemolytic anaemia it is advised to monitor the infant for signs of haemolysis, especially in newborn or premature breastfed infants (GRADE 2C, SOA 90.7). Anticoagulants in rheumatic disease The deleterious effects of warfarin and compatibility of heparin in pregnancy are well described and evidence-based guidelines for the management of venous thromboembolism (VTE) and thrombophilia in pregnancy exist . Heparin A systematic review and systematically produced guidelines describe the utility of heparin in the management of VTE and pregnancy morbidity in pregnant patients with antiphospholipid syndrome (APS). Heparin/LMWH does not cross the placenta . Heparins are compatible with breastfeeding. There were no additional studies identified, but LactMed states that no particular caution is required as the molecular weight of heparin is such that it is unlikely to be appreciably excreted into breastmilk. Warfarin Warfarin has the ability to cross the placenta and is associated with an increased risk of congenital abnormalities including a characteristic warfarin embryopathy (hypoplasia of the nasal bridge, congenital heart defects, ventriculomegaly, agenesis of the corpus callosum, stippled epiphyses) in ∼5% of foetuses exposed between 6 and 12 weeks of gestation. Warfarin should therefore be avoided between 6 and 12 weeks . While heparin/LMWH remains the anticoagulant of choice in pregnancy for the majority of patients considered to be at increased thrombotic risk in pregnancy, warfarin may be considered in pregnancy for women with mechanical heart valves (MHVs). The LMWH regimen could be used for those who refuse warfarin and other vitamin K antagonists (VKAs) . In management of atrial fibrillation, VKAs may be used after the first trimester . LactMed is reassuring both with regard to low drug levels in breastmilk and infant serum, and no reported adverse effects. New anticoagulants A systematic review identified 236 cases of direct oral anticoagulants (DOAC) use in pregnancy of rivaroxaban ( n = 178), dabigatran ( n = 27), apixaban ( n = 21) and edoxaban ( n = 10). DOACs were mostly used for prophylaxis or treatment of venous thromboembolism ( n = 91). DOACs were discontinued within the first 2 months of pregnancy in 84%, and the maximum reported duration of use was 26 weeks. Pregnancy outcome data were available for 140 pregnancies. Thirty-nine pregnancies were electively terminated. In the remaining 101 pregnancies, total miscarriage rate was 31% ( n = 31) and live birth rate was 68% ( n = 69, one missing). Foetal and neonatal abnormalities were reported in eight pregnancies, of which at least half were suspected to be related to rivaroxaban use during the first trimester of pregnancy. In only 18% of cases ( n = 42), the presence or absence of thrombotic and bleeding complications was reported. This limited evidence raises concern regarding embryo-foetal safety, with a high incidence of miscarriages and a 4% rate of congenital anomalies with the use of rivaroxaban. Overall, not enough data are available to judge safety and efficacy of the use of DOACs during pregnancy and rivaroxaban and dabigatran have both been shown to cause adverse obstetric outcomes in animal studies, described in summary of product characteristics . UKTIS does not report on DOACs. LactMed describes several case reports that consistently indicate that maternal doses of rivaroxaban of 15–30 mg daily produce low levels in milk that are considerably below doses required for anticoagulation in infants . Therefore, breastfeeding is not contraindicated if rivaroxaban is required by the mother. There are no data on the excretion of dabigatran into human milk. Fondaparinux Fondaparinux may be considered if there is an allergy or adverse response to LMWH, although solid data are lacking and minor transplacental passage has been demonstrated, without any adverse materno-foetal effects in five pregnancies . LactMed considers use of fondaparinux to be acceptable during breastfeeding . Recommendations on all anticoagulants were based on evidence as shown in , available at Rheumatology online. Recommendations for anticoagulants in pregnancy and breastfeeding LMWH heparin is compatible throughout pregnancy (GRADE 1A, SOA 100). LMWH is compatible with breastfeeding (GRADE 1C, SOA 100). The use of warfarin in pregnancy is associated with increased foetal risk throughout pregnancy and has limited indications, therefore should only be considered in exceptional circumstances (GRADE 1B, SOA 98.8). Warfarin is compatible with breastfeeding (GRADE 1A, SOA 100). Direct oral anticoagulants (DOACs) cannot be recommended in pregnancy (GRADE 1C, SOA 97.9). Rivaroxaban may be considered in breastfeeding (GRADE 2C, SOA 95.3). Other DOACs are not recommended in breastfeeding due to lack of human data and concerns from animal studies (GRADE 1C, SOA 97.4). Fondaparinux may be considered in pregnancy and breastfeeding if there is an allergy or adverse response to LMWH (GRADE 2C, SOA 95.5). Bisphosphonates Bisphosphonates are not ideal in women planning pregnancy because the absolute risk of fracture is small in this age group and the skeletal half-life of these drugs is very long. The number of human pregnancy exposures remains limited and a detailed literature review identified 40 pregnancies , while a systematic review published in abstract form described outcomes from 120 bisphosphonate-exposed pregnancies . Overall, rates of congenital malformation and miscarriages were comparable in bisphosphonate and controls, although limiting factors included: few comparator groups; heterogeneous maternal disease; concomitant medication; and small sample size. Controlled studies have demonstrated possible associations between preconceptual/gestational bisphosphonate exposure and increased risk of spontaneous abortion, decreased infant birth weight, and lower gestational age at delivery. These findings, however, may reflect data limitations and/or uncontrolled confounding (UKTIS). Therefore, further controlled studies are required to fully establish the safety of bisphosphonates in pregnancy and they are not drugs of choice in women planning pregnancy. LactMed states that limited evidence indicates that breastfeeding after cessation of long-term bisphosphonate treatment appears to have no adverse effects on the infant. There is no information on the use of alendronate or risedronate during breastfeeding. Limited information indicates that maternal doses of pamidronate of 30 mg intravenously produce very low levels in milk and because pamidronate has a serum half-life of ∼3 h, is highly bound to calcium and poorly absorbed orally, absorption of pamidronate by a breastfed infant is unlikely. Therefore, withholding breastfeeding for 12–24 h after a dose should ensure that the breastfed infant is exposed to little or no pamidronate. If the mother receives a bisphosphonate during pregnancy or nursing, some experts recommend monitoring the infant’s serum calcium during the first 2 months postpartum . These recommendations were based on evidence as shown in , available at Rheumatology online. Recommendations for bisphosphonates in pregnancy and breastfeeding There is insufficient data upon which to recommend bisphosphonates in pregnancy or to advise a specific time for them to be stopped pre-conception. Given their biological half-life in bone of up to 10 years and no evidence of harm from limited reports of their use in pregnancy, a pragmatic recommendation is that they should be stopped 3 months in advance of pregnancy (GRADE 2C, SOA 96.8). There are no data on which to base a recommendation for the use of bisphosphonates during breastfeeding (GRADE 2C, SOA 98.5). Antihypertensive medication in rheumatic disease Patients with ARD, particularly renal SLE and systemic sclerosis (SSc) frequently require anti-hypertensive treatment for sometimes life-threatening disease, such as scleroderma renal crisis, that may require specialist use of certain anti-hypertensive drugs. The introduction of ACEis for the treatment of SSc renal crisis has significantly reduced mortality rates of up to 50% in the 1970s to a mortality of up to 20% at 6 months after introducing ACEis. The use of ACEis may therefore be indicated (also in pregnancy) in exceptional circumstances . The management of pre-existing and new onset hypertension in pregnancy has been comprehensively reviewed and updated in the 2019 NICE guideline, Hypertension in Pregnancy: Diagnosis and Management . Angiotensin blockade Disruption of the renin-angiotensin system (RAS) in pregnancy by maternal treatment with ACEis or angiotensin receptor blockers (ARBs) in the second/third trimester leads to abnormal foetal renal development, known as foetal RAS blockade syndrome . There are conflicting results, however, on the risk of this fetopathy occurring after first trimester exposure. A systematic review and meta-analysis of 19 articles involving 4 163 753 pregnant women found a significant association between overall congenital malformations and first trimester‐only exposure to ACEis/ARBs (odds ratio 1.94; 95% CI 1.71, 2.21; P <0.0001). This review also found a significant association between cardiovascular malformations, miscarriage and stillbirth and ACEi/ARB exposure. A similar risk was observed in a cohort of 1 333 624 pregnancies , including 4107 (0.31%) following first trimester ACEi exposure that found an increased risk of overall malformations in the ACEi-exposed pregnancies [unadjusted relative risk (RR), 1.82; 95% CI 1.61, 2.06] and of cardiac malformations (RR 2.95; 95% CI 2.50, 3.47). On further analysis, however, restricting the cohort to pregnancies complicated by chronic hypertension (both exposed and unexposed) and accounting for other confounding factors, there was no significant increase in the risk of any of the outcomes assessed. NICE guidelines state that women taking ACEis/ARBs should be advised of the increased risk of congenital anomalies if these drugs are taken during pregnancy and to discuss alternative antihypertensive drugs with their clinician responsible for managing their condition . If they become pregnant on ACEis/ARBs they should be stopped (preferably within 2 working days of notification) and other antihypertensive treatments offered. UKTIS recommends that where prolonged first trimester exposure has occurred, a 20-week anomaly scan should focus on cardiovascular, renal and neurological development, in addition to the routine anatomical checks. UKTIS states that ACEi fetopathy following exposure to ACEis in the second and third trimesters of pregnancy is well-described and may include oligohydramnios, renal tubular dysgenesis, neonatal anuria, hypocalvaria, pulmonary hypoplasia, persistent patent ductus arteriosus, mild-to-severe intrauterine growth restriction, and foetal or neonatal death. It is proposed that these effects occur as a result of a direct effect on the foetal RAS which begins to function from ∼26 weeks gestation. A small prospective case series has suggested that the risk period for ACEi fetopathy is with exposure beyond 20 weeks gestation. Due to data limitations, the absolute risk of ACEi fetopathy is unclear. Due to the risk of ACEi fetopathy, use of ACEis in the second and third trimesters is generally contraindicated and should only be reserved for cases of severe maternal illness that cannot be managed using alternative drugs . Negligible amounts of enalapril and captopril are transferred into breastmilk with no adverse effects reported on the breastfed infants of mothers treated with short-acting ACEis . NICE states that enalapril may be offered to treat hypertension in breastfeeding mothers with appropriate monitoring of maternal renal function and serum potassium . The following recommendations were based on evidence as shown in , available at Rheumatology online. Recommendations for ACEis/ARBs in pregnancy and breastfeeding ACEis and ARBs should be stopped as soon as possible when pregnancy is confirmed in the first trimester and if necessary an alternative antihypertensive compatible with pregnancy given (GRADE 1A, SOA 100). ACEis/ARBs should be avoided in the second and third trimester but may be considered under specialist advice in certain circumstances (GRADE 1C, SOA 98.5). Based on limited evidence, enalapril is compatible with breastfeeding (GRADE 2C, SOA 98.5). Calcium channel blockers Calcium channel blockers (CCBs) including amlodipine, diltiazem, felodipine, lacidipine, lercanidipine, nicardipine, nifedipine, nimodipine and verapamil are mainly used for the treatment and prophylaxis of angina, and the treatment of hypertension where an ACEi/ARB is unsuitable. In patients with rheumatic disease, nifedipine or amlodipine are also used to treat Raynaud’s phenomenon. UKTIS has not identified an increased risk of congenital malformations or other adverse pregnancy outcomes with CCBs, although data remains too limited to draw firm conclusions on many outcomes. Data on rates of preterm delivery, foetal growth and neurodevelopmental outcomes are too limited and/or confounded to permit an accurate risk assessment, but where a CCB is required to treat maternal hypertension or as a tocolytic, foetal benefits of use are likely to outweigh any unspecified risk and treatment should not be withheld on this basis . Data is more limited for amlodipine so it is not included in alternatives to treat hypertension in pregnancy that include labetalol, nifedipine or methyldopa in order of preference . LactMed describes low levels of nifedipine and amlodipine in breastmilk, without any adverse effects being reported among exposed infants . NICE guidance of other antihypertensive drugs that may be offered in breastfeeding mothers include nifedipine, amlodipine, atenolol or labetolol . Recommendations were based on evidence as shown in , available at Rheumatology online. Recommendations for CCBs in pregnancy and breastfeeding Nifedipine is compatible with pregnancy with no direct evidence of harm at doses up to 90 mg/day (GRADE 1A, SOA 99.0). Nifedipine is compatible with breastfeeding (GRADE 1B, SOA 100). Amlodipine can be considered in pregnancy and breastfeeding as there is no evidence of harm (GRADE 1C, SOA 97.9). Pulmonary vasodilators Moderate-to-severe pulmonary hypertension (PHT) is a rare complication of certain ARDs and remains a contraindication to planned pregnancy with high mortality. Unintentional pregnancy and/or patient choice, however, means that treatment of this condition with specific pulmonary vasodilators may be required in pregnancy. Limited information on use of these drugs in human pregnancy was identified in the previous BSR guideline . No studies were identified examining pregnancy outcomes after paternal exposure to any of these pulmonary vasodilators. Sildenafil Sildenafil has been studied in the context of trying to improve utero-placental circulation in pregnancies affected by severe foetal growth restriction. A systematic review and meta-analysis were identified, examining the utility of sildenafil being used for treatment or prevention of obstetric diseases compared with placebo. They analysed 598 pregnant women from seven clinical trials with pre-eclampsia ( n = 139), intrauterine growth restriction ( n = 275) and oligohydramnios ( n = 184) and found a significant improvement in birthweight following sildenafil treatment during pregnancy, with no difference in other outcomes . However, in 2018 a randomised controlled study looking at using sildenafil to treat pregnant women in whom there was significant foetal growth restriction was halted early due to a number of neonates having persistent pulmonary hypertension of the newborn (PPHN). This multi-centered study enrolled women with a singleton pregnancy between 18 and 30 weeks with severe foetal growth restriction of likely placental origin, where the likelihood of perinatal death/severe morbidity was estimated to be significant. Participants were randomised into sildenafil or placebo arms. One of three study sites reported that PPHN appeared to be more prevalent in infants exposed in utero to sildenafil compared with placebo-exposed infants ( n = 17/93 vs n = 3/90), that death among infants with PPHN was more common following sildenafil exposure ( n = 11/17 vs n = 0/3), and that when the overall neonatal death rate was considered, there was a non-statistically significant trend towards an increased risk following sildenafil exposure ( n = 19/93 vs n = 9/90). As a result, the trial was halted early and sildenafil is no longer recommended to improve placental function in severely growth-restricted babies . However, it should be noted that two further study sites for this trial did not detect an increased risk of PPHN and overall, this trial has not identified a clear beneficial effect of sildenafil on foetal outcome. PPHN is more prevalent in premature and growth-restricted foetuses and is a relatively rare complication in healthy babies delivered at term. UKTIS recommends that general use of sildenafil in pregnancy should be avoided where possible . However, in the context of maternal pulmonary hypertension there is clear benefit in the use of sildenafil to reduce the effects of pulmonary hypertension, which often gets worse during pregnancy. Case studies/case series data suggest that sildenafil exposure was not associated with miscarriage or congenital anomaly; however, the data is extremely limited (around 18 reported pregnancies). The risks and benefits of continuing sildenafil should be discussed with the patients, but most will likely require ongoing treatment. LactMed describes limited data showing that sildenafil and its active metabolite are poorly excreted into breastmilk and amounts ingested by the infant are small and would not be expected to cause any adverse effects in breastfed infants . Bosentan Animal data have revealed teratogenicity due to bosentan, including malformations of the head, mouth, face and large blood vessels in addition to an increased number of stillbirths and increased mortality . Previously, we identified data from 12 pregnancies of women with PHT treated with bosentan in pregnancy plus multiple other medications, including sildenafil and iloprost, with reduced pregnancy duration of 37 weeks in one and reduced birth weight in two cases but no other maternal complications or foetal loss . We identified one further case report of a patient with Eisenmenger syndrome exposed to long-term bosentan before and during pregnancy that was delivered by caesarean section at 27 weeks due to severe maternal PHT without any evidence of teratogenic effects of bosentan . UKTIS does not report on bosentan. LactMed states there is little published experience with bosentan during breastfeeding and an alternate drug may be preferred, especially while nursing a newborn or preterm infant . Prostacyclines Previously we identified data on 23 pregnancies of patients with PHT (three with SLE) treated with iloprost ( n = 5 pregnancies) or epoprostenol ( n = 15 pregnancies) and three other prostacyclines (unspecified type) in patients who were taking multiple other medications, including immunosuppressants, sildenafil and bosentan . Findings of premature deliveries and reduced birthweight were confounded by maternal disease. Furthermore, maternal complications were attributable to PHT. We did not identify any new evidence and UKTIS and LactMed do not report on iloprost or epoprostenol. The recommendations were based on evidence as shown in , available at Rheumatology online. Recommendations for pulmonary vasodilators in pregnancy and breastfeeding Established moderate to severe PHT remains a contraindication to pregnancy. If pregnancy occurs, the use of these pulmonary vasodilator drugs in pregnancy should be considered only as part of a multidisciplinary team assessment (GRADE 1C, SOA 99.5). Limited evidence supports the use of prostacyclines to treat PHT during pregnancy (GRADE 2C, SOA 98.0). Limited evidence supports the use of sildenafil to treat PHT during pregnancy (GRADE 2C, SOA 98.0). Bosentan is teratogenic in animals and although there is no evidence of harm from human pregnancy, the evidence is insufficient to recommend in pregnancy (GRADE 1C, SOA 98.8). There are no data relating to breastfeeding exposure to pulmonary vasodilators on which to base a recommendation (GRADE 2C, SOA 98.8). Paternal exposures There remains limited data on paternal exposure to drugs used to treat rheumatic disease and reports of teratogenic effects linked with paternal exposure are lacking for all drugs considered in this guideline. Our updated search revealed the following additional information. Links with adverse neurodevelopmental outcomes but not asthma have been reported following paternal exposure to paracetamol from the largest cohort study of paracetamol in pregnancy. A direct causal effect, however, remains unproven. An observational prospective cohort study from Sweden of 3983 children born to fathers receiving antidepressant treatment around conception, included fathers treated with: SSRIs, n = 2865; SNRIs, n = 470; and TCAs, n = 240. This study found paternal intake of all antidepressants studied to be safe with respect to the risk of preterm birth, malformation, autism or intellectual disability. A systematic review of the effect of paternal exposure to immunosuppressive drugs on sexual function, reproductive hormones, fertility, pregnancy and offspring outcomes found very weak evidence of reduced sperm parameters with codeine, tramadol and CCBs but not NSAIDs, LDA or ACEi and no abnormalities in offspring were reported for any drug. Interestingly, low-dose lisinopril (2.5 mg/day) has been shown to increase total sperm count and motility in a randomised, controlled, crossover pilot of study of normotensive men with idiopathic oligospermia leading to an unassisted pregnancy rate of 48.5% . Small studies of chronic use of NSAIDs in men with rheumatic disease have not indicated any impairment of spermatogenesis and no evidence for harmful effects of NSAIDs on offspring . A systematic review of eight studies including 166 cases of paternal exposure found inconsistent reports of adverse effects of colchicine on sperm quality and only one study ( n = 53) of paternal exposure did not find an adverse effect on offspring . The following recommendations were based on evidence as shown in , available at Rheumatology online. Recommendations for paternal exposure Paracetamol is compatible with paternal exposure (GRADE 1B, SOA 98.5). Amitriptyline, SNRIs and SSRIs are compatible with paternal exposure (GRADE 1B, SOA 98.5). Non-selective NSAIDs are compatible with paternal exposure (GRADE 1C, SOA 98.4). Based on limited or no data and no association with adverse foetal development or pregnancy outcome, paternal exposure to all other drugs described in this guideline are unlikely to be harmful (GRADE 2C, SOA 97.3).
Pre-conception counselling should be addressed by all healthcare professionals, with referral to professionals with relevant experience as appropriate to optimize all therapy, including non-pharmacological options for chronic pain management during pregnancy (GRADE 1A, SOA 99.5). The risks and benefits of drug treatment to mother and foetus should be discussed and clearly documented by all healthcare professionals involved in the patient’s care (GRADE 1A, SOA 99). The cause of pain and other symptoms should be assessed and managed appropriately (GRADE 1B, SOA 98.5). The requirement for analgesia should be assessed and minimum effective dose should be prescribed and titrated according to response (GRADE 1B, SOA 100). Tricyclic antidepressants are preferred over other antidepressant medications to manage chronic pain (GRADE 1B, SOA 98.1). Cessation of anti-depressant therapy that is being used as chronic pain medication in the post-natal period is not recommended, due to the risk of adverse impact on mood (GRADE 1C, SOA 96). LDA (≤150 mg/day) is recommended in all patients at high risk for pre-eclampsia (GRADE 1A, SOA 99.5). Low molecular weight heparin (LMWH) is the preferred anticoagulant (GRADE 1A, SOA 100). Nifedipine is the preferred vasodilator (GRADE 1B, SOA 98.5). Paternal drug exposure may reduce male fertility but has not been associated with adverse foetal development or pregnancy outcome. Although evidence is weak, we recommend that men are reassured about the safety of fathering a pregnancy while taking medicines to manage comorbidities as described in this guideline (GRADE 1C, SOA 98).
Paracetamol Two systematic reviews were selected. Overall, paracetamol was considered to have a favourable safety profile in pregnancy. Potential links between paracetamol use in pregnancy and pre-term birth and adverse neurodevelopmental outcomes (principally autism and attention deficit hyperactivity disorder) were confounded by maternal disease and selection bias . UKTIS states that ‘In most studies, risk of abnormal neurodevelopment correlated with duration of paracetamol exposure. However, significant methodological limitations of these studies limit the conclusions that can be drawn, and a causal association remains unproven’ . There were no convincing associations with congenital malformations. UKTIS notes that findings of a possible increased risk of cryptorchidism in male offspring following paracetamol use during pregnancy are conflicting . Current Royal College of Obstetricians (RCOG) guidance is that paracetamol remains safe for use during pregnancy and breastfeeding, and its use in any trimester does not appear to increase the risk of major birth defects . There were conflicting results on the risk of developing wheeze in infants exposed to paracetamol during pregnancy. A recent meta-analysis of 13 articles and 1 043 109 individuals identified a statistically significant association between prenatal paracetamol exposure and increased risk of child asthma [odds ratio (OR) 1.19; 95% CI, 1.12, 1.27; P < 0.00001] in a random-effect model . This significant association was observed for first trimester as well as second and third trimester exposure. There were no studies identified which specifically examined neonatal outcomes after drug exposure in breastmilk. LactMed reports low amounts of this drug in breastmilk at levels much less than doses usually given to infants with few reports of adverse events. The following recommendations for paracetamol were developed based on evidence as shown in , available at Rheumatology online. Recommendations for paracetamol in pregnancy and breastfeeding Paracetamol is the analgesic of choice and compatible peri-conception and throughout pregnancy (GRADE 1B, SOA 99). LactMed describes paracetamol as a good choice for analgesia and fever reduction in breastfeeding mothers (GRADE 2C, SOA 99.5).
Two systematic reviews were selected. Overall, paracetamol was considered to have a favourable safety profile in pregnancy. Potential links between paracetamol use in pregnancy and pre-term birth and adverse neurodevelopmental outcomes (principally autism and attention deficit hyperactivity disorder) were confounded by maternal disease and selection bias . UKTIS states that ‘In most studies, risk of abnormal neurodevelopment correlated with duration of paracetamol exposure. However, significant methodological limitations of these studies limit the conclusions that can be drawn, and a causal association remains unproven’ . There were no convincing associations with congenital malformations. UKTIS notes that findings of a possible increased risk of cryptorchidism in male offspring following paracetamol use during pregnancy are conflicting . Current Royal College of Obstetricians (RCOG) guidance is that paracetamol remains safe for use during pregnancy and breastfeeding, and its use in any trimester does not appear to increase the risk of major birth defects . There were conflicting results on the risk of developing wheeze in infants exposed to paracetamol during pregnancy. A recent meta-analysis of 13 articles and 1 043 109 individuals identified a statistically significant association between prenatal paracetamol exposure and increased risk of child asthma [odds ratio (OR) 1.19; 95% CI, 1.12, 1.27; P < 0.00001] in a random-effect model . This significant association was observed for first trimester as well as second and third trimester exposure. There were no studies identified which specifically examined neonatal outcomes after drug exposure in breastmilk. LactMed reports low amounts of this drug in breastmilk at levels much less than doses usually given to infants with few reports of adverse events. The following recommendations for paracetamol were developed based on evidence as shown in , available at Rheumatology online. Recommendations for paracetamol in pregnancy and breastfeeding Paracetamol is the analgesic of choice and compatible peri-conception and throughout pregnancy (GRADE 1B, SOA 99). LactMed describes paracetamol as a good choice for analgesia and fever reduction in breastfeeding mothers (GRADE 2C, SOA 99.5).
Paracetamol is the analgesic of choice and compatible peri-conception and throughout pregnancy (GRADE 1B, SOA 99). LactMed describes paracetamol as a good choice for analgesia and fever reduction in breastfeeding mothers (GRADE 2C, SOA 99.5).
One systematic review considered outcomes from opioid-exposed pregnancies, including three studies of congenital malformation in >20 000 codeine-exposed pregnancies with conflicting results. Of these three studies, a large Norwegian population-based cohort study found no statistically significant associations between neonatal codeine exposure and congenital malformations. In contrast, a large case-control study found that first trimester exposure to opioids (codeine and hydrocodone in 69% of cases) was significantly associated with congenital heart defects, spina bifida and gastroschisis , although there was a high risk for recall bias in this study. A case-control study found a significant association between spina bifida in infants and first trimester maternal opioid use, although it was not with codeine use specifically and potential confounders such as illicit drug use/dependence and sociodemographic data exist in this study . UKTIS notes that the methodological limitations of studies make it difficult to draw any firm conclusions and more robust research is needed before firm conclusions regarding the risk of congenital anomalies with codeine use can be provided . No convincing associations between first trimester opioid use and miscarriage have been found. A large population study found increased rates of premature delivery, reduced birthweight and pre-eclampsia in codeine-exposed patients compared with controls that was considered to be due to confounding factors such as maternal disease and lack of adjustment for gestational age . This study also found a significant increase in post-partum haemorrhage (18.3% vs 14.5%; OR 1.3; 95% CI 1.1, 1.5) in 2666 mothers exposed to codeine and hypothesized it was due to an opioid effect weakening myometrial contraction, but the precise mechanism is unknown. UKTIS describes theoretical concerns that maternal use of codeine near term may be associated with respiratory depression in the neonate and notes that the only study that has investigated the risk of neonatal respiratory depression found no increased risk . The potential risk of neonatal abstinence syndrome (NAS) is lower with short courses of short-acting opioids and must be weighed against the benefits for treating acute pain that often outweigh risks. Prolonged opioid use, however, should be avoided and if used in the last trimester, neonatologists should be advised due to the risks of NAS . Controversy remains over whether codeine is safe in breastfeeding. Central nervous system (CNS) depression was reported by mothers in 16.7% (35/210) babies exposed to codeine, compared with 0.5% exposed to paracetamol . In the same study there was one neonatal death and high morphine levels were found at post-mortem. The mother had received high doses of codeine (>2 mg/kg/day) and was subsequently found to be an ultra-rapid CYP2D6 metaboliser. Another study demonstrated dose-dependent CNS depression in 24% (17/72) of infants exposed to codeine through breastmilk . A large study of 7804 infants reported conflicting results, but specifically, there was no difference in poor Apgar scores, postnatal complications, admission to special care baby units, readmission to hospital, resuscitation or death in infants exposed and not exposed to codeine . LactMed notes that numerous professional organizations and regulatory agencies recommend that other agents are preferred over codeine during breastfeeding but that other opioid alternatives have been less studied and may not be safer . It is acknowledged that due to its unpredictable metabolism, administration of codeine results in delivery of an unknown quantity of morphine. Therefore, despite its widespread use and probable safety in most cases, we would advise caution with prolonged use of codeine in breastfeeding and appropriate advice to the mother to seek medical attention if she has any concerns regarding lethargy or drowsiness in her child. The following recommendations for codeine were based on evidence as shown in , available at Rheumatology online. Recommendations for codeine in pregnancy and breastfeeding Codeine is compatible peri-conception and throughout pregnancy, although long-term use should be avoided. There is no consistent evidence to recommend a dose reduction pre-delivery but neonatologists should be aware of maternal use (GRADE 1B, SOA 97.8). Caution is advised with use of codeine in breastfeeding, due to the risk of CNS depression resulting from unpredictable metabolism of codeine to morphine (GRADE 1C, SOA 98).
Codeine is compatible peri-conception and throughout pregnancy, although long-term use should be avoided. There is no consistent evidence to recommend a dose reduction pre-delivery but neonatologists should be aware of maternal use (GRADE 1B, SOA 97.8). Caution is advised with use of codeine in breastfeeding, due to the risk of CNS depression resulting from unpredictable metabolism of codeine to morphine (GRADE 1C, SOA 98).
A systematic review considered outcomes from tramadol-exposed pregnancies. It included a large prospective cohort study of over 1.6 million women that reported a statistically significant association between tramadol use in pregnancy and major congenital malformations (OR 1.33; 95% CI 1.05, 1.70) and cardiovascular defects (OR 1.56; 95% CI 1.04, 2.29), although it did not adjust for indication, duration or dose of medication. UKTIS notes that a single study identified an increased risk of miscarriage among women who used tramadol in early pregnancy and recommended further studies considering the impact of confounding factors to clarify this finding. A single cohort study found no increased risk of preterm delivery within 146 pregnancies exposed to tramadol in at least the first trimester . A small number of case reports have reported NAS with long-term intrauterine exposure to tramadol , although none have compared the relative rate of NAS with tramadol compared with other opioid analgesics. LactMed states that the excretion of tramadol into milk is low and even lower amounts of the active metabolite, O-desmethyltramadol, are excreted and a study of breastfeeding in newborn infants found no adverse effects attributable to tramadol compared with controls . There has been one death, however, in the 8-month-old breastfed infant of a woman addicted to tramadol, although the death was not definitely attributable to tramadol exposure in breastmilk . Current RCOG advice states that tramadol can continue to be used (with caution) during breastfeeding and the lowest effective dose should be used for the shortest time possible . The following recommendations for tramadol were based on evidence as shown in , available at Rheumatology online. Recommendations for tramadol in pregnancy and breastfeeding Avoid tramadol peri-conception and in first trimester and only consider in second/third trimester if no alternative analgesia (GRADE 2B, SOA 97.8). Based on limited data, tramadol may be compatible with short-term use in breastfeeding (GRADE 2C, SOA 94.8).
Avoid tramadol peri-conception and in first trimester and only consider in second/third trimester if no alternative analgesia (GRADE 2B, SOA 97.8). Based on limited data, tramadol may be compatible with short-term use in breastfeeding (GRADE 2C, SOA 94.8).
Amitriptyline NICE guidance and a systematic review described amitriptyline use in pregnancy. Notably, the evidence base underlying amitriptyline use in pregnancy comes from its use as a tricyclic antidepressant (TCA) to treat depression at doses of 150–300 mg and evidence of its use in pregnancy to treat chronic widespread pain at 75 mg per day or less is lacking. In addition, most studies report on TCAs as a drug class and although they include (some) women exposed to amitriptyline, they lack a separate assessment of their pregnancy outcomes so do not provide information about specific amitriptyline exposure . Overall, no increased risk of congenital malformations has been found for TCAs . UKTIS reports conflicting findings, with possible associations with spontaneous abortion, preterm delivery, and autism spectrum disorder identified in some (but not all) studies . There is an association between discontinuation of antidepressants and a high risk for relapse of mood disorders that can adversely impact on pregnancy. It is unknown, however, whether a similar phenomenon of rebound pain exists when antidepressants used for analgesia are discontinued in pregnancy, particularly for amitriptyline that is used at 2–4-fold higher doses to treat depression compared with chronic pain. If a decision is made to stop amitriptyline in pregnancy, the dose should be tapered gradually where possible . Low levels of amitriptyline and its metabolites are reported in breastmilk with no adverse effects described with limited follow-up, summarised in LactMed . The following recommendations were based on evidence as shown in , available at Rheumatology online. Recommendations for amitriptyline in pregnancy and breastfeeding Amitriptyline is compatible with pregnancy. There is no evidence of adverse effect on IQ or developmental outcomes (GRADE 1C, SOA 100). Because very little amitriptyline is found in breastmilk with antidepressant doses and it is used at lower doses for chronic pain, it is unlikely to cause adverse effects in breastfed infants (GRADE 1C, SOA 100).
NICE guidance and a systematic review described amitriptyline use in pregnancy. Notably, the evidence base underlying amitriptyline use in pregnancy comes from its use as a tricyclic antidepressant (TCA) to treat depression at doses of 150–300 mg and evidence of its use in pregnancy to treat chronic widespread pain at 75 mg per day or less is lacking. In addition, most studies report on TCAs as a drug class and although they include (some) women exposed to amitriptyline, they lack a separate assessment of their pregnancy outcomes so do not provide information about specific amitriptyline exposure . Overall, no increased risk of congenital malformations has been found for TCAs . UKTIS reports conflicting findings, with possible associations with spontaneous abortion, preterm delivery, and autism spectrum disorder identified in some (but not all) studies . There is an association between discontinuation of antidepressants and a high risk for relapse of mood disorders that can adversely impact on pregnancy. It is unknown, however, whether a similar phenomenon of rebound pain exists when antidepressants used for analgesia are discontinued in pregnancy, particularly for amitriptyline that is used at 2–4-fold higher doses to treat depression compared with chronic pain. If a decision is made to stop amitriptyline in pregnancy, the dose should be tapered gradually where possible . Low levels of amitriptyline and its metabolites are reported in breastmilk with no adverse effects described with limited follow-up, summarised in LactMed . The following recommendations were based on evidence as shown in , available at Rheumatology online. Recommendations for amitriptyline in pregnancy and breastfeeding Amitriptyline is compatible with pregnancy. There is no evidence of adverse effect on IQ or developmental outcomes (GRADE 1C, SOA 100). Because very little amitriptyline is found in breastmilk with antidepressant doses and it is used at lower doses for chronic pain, it is unlikely to cause adverse effects in breastfed infants (GRADE 1C, SOA 100).
Amitriptyline is compatible with pregnancy. There is no evidence of adverse effect on IQ or developmental outcomes (GRADE 1C, SOA 100). Because very little amitriptyline is found in breastmilk with antidepressant doses and it is used at lower doses for chronic pain, it is unlikely to cause adverse effects in breastfed infants (GRADE 1C, SOA 100).
Data on gabapentin use in pregnancy comes mostly from studies of the treatment of maternal epilepsy at doses generally higher that those used to treat chronic pain, while information on pregabalin use in pregnancy comes from studies of its use to treat neuropathic pain . A systematic review of gabapentin use in pregnancy to treat epilepsy did not find any evidence of an increased rate of major malformations or other adverse outcomes attributable to gabapentin exposure. No long-term outcomes were reported. Furthermore, a systematic review and meta-analysis published in abstract form only , of eight cohort studies (four prospective and four retrospective), reporting 5 072 286 unexposed and 949 exposed pregnancies found that first trimester exposure to gabapentin was not statistically significantly associated with major congenital malformations (OR 0.83; 95% CI 0.45, 1.53). A prospective cohort study using Teratology Information Services data from seven European countries found that first trimester use of pregabalin was associated with significantly higher rates of major congenital malformations when compared with unexposed pregnancies (OR 3.0; 95% CI 1.2, 7.9). This study, however, was limited by a small sample size (164 exposed pregnancies) and lack of adjustment for potential confounding by use of concomitant medications . A systematic review of pregabalin use, mostly to treat neuropathic pain in pregnancy, including data from this study and two others identified 651 pregnancy exposures and concluded that pregabalin exposure during pregnancy is not devoid of structural teratogenicity potential . UKTIS reviewed data from six controlled studies, with outcomes of >3000 exposed pregnancies of patients with chronic pain or other non-epilepsy indications. They conclude that this data does not indicate that maternal pregabalin use in pregnancy is associated with increased risks of malformation, miscarriage or adversely affects foetal growth . Interestingly, a systematic review and meta-analysis published in abstract form only , of six studies reporting 2319 exposed pregnancies and 4 982 778 unexposed pregnancies found that first trimester exposure to pregabalin was not significantly associated with an increased risk of major congenital malformations (OR 1.20; 95% CI 0.92, 1.57). Following completion of our guidelines, the MHRA issued a safety warning on the use of pregabalin in pregnancy based on new data available online . This new data has been considered in the latest UKTIS update in April 2022 that states that the available data do not provide conclusive evidence that maternal pregabalin use in the first trimester, or at any stage of pregnancy, is associated with increased risks of either overall malformation or any specific malformations. Therefore, our recommendations have not changed. UKTIS found no controlled studies of the risk of neonatal complication following prenatal gabapentin or pregabalin exposure, although one study has described a small number of affected infants, including one case of neonatal withdrawal following in-utero pregabalin exposure . Use of any centrally acting drug throughout pregnancy or near delivery may be associated with withdrawal symptoms in the neonate and/or NAS. Although evidence is lacking and it is uncertain if pregabalin or gabapentin impact upon maternal folate status, UK guidelines state that women who take any anti-epileptic medication should be prescribed high-dose folic acid (5 mg/day) preconceptually and in the first trimester . There remains limited data on use of these drugs in breastfeeding. According to LactMed, low levels of gabapentin and pregabalin have been found in breastmilk with no adverse effects on infants reported in limited case reports/series ( n < 10) for gabapentin. Therefore, both drugs may be considered if required by the breastfeeding mother . The following recommendations were based on evidence as shown in , available at Rheumatology online. Recommendations for gabapentin and pregabalin in pregnancy and breastfeeding Gabapentin at lowest effective dose may be considered in pregnancy with folic acid supplementation if no alternative analgesic is suitable (GRADE 1B, SOA 95). Gabapentin may be considered in breastfeeding if no alternative analgesic is suitable (GRADE 2C, SOA 96). Pregabalin may be considered in pregnancy (with folic acid supplementation) and during breastfeeding (GRADE 2C, SOA 95.3).
Gabapentin at lowest effective dose may be considered in pregnancy with folic acid supplementation if no alternative analgesic is suitable (GRADE 1B, SOA 95). Gabapentin may be considered in breastfeeding if no alternative analgesic is suitable (GRADE 2C, SOA 96). Pregabalin may be considered in pregnancy (with folic acid supplementation) and during breastfeeding (GRADE 2C, SOA 95.3).
NICE guidance and two systematic reviews described SNRI use in pregnancy. There was no association between first trimester exposure to venlafaxine and an increased risk for major congenital malformations . UKTIS notes that although there are some reports of an increased risk of miscarriage following gestational exposure to venlafaxine, the data are inconsistent and likely confounded by indication and other factors . Some studies have found a possible association with an increased risk for some perinatal complications, including a withdrawal syndrome with venlafaxine use in the third trimester . A more limited data set for duloxetine does not suggest a clinically important increased risk for major malformations but has identified prenatal antidepressant exposure syndrome in two of five case reports and conflicting reports of increased rates of gestational hypertension and spontaneous abortion . LactMed reports that infants receive venlafaxine and its active metabolite in breastmilk, and the metabolite of the drug can be found in the plasma of most breastfed infants, but no proven drug-related side effects have been reported in small case series. Little published information is available on the use of duloxetine during breastfeeding; however, the dose in milk is low and serum levels were low in two breastfed infants . The following recommendations were based on evidence as shown in , available at Rheumatology online. Recommendations for SNRIs in pregnancy and breastfeeding Venlafaxine is compatible at conception and throughout pregnancy. There may be an increased risk of neonatal abstinence syndrome/short-term behavioural effects, but larger studies are needed to evaluate this finding (GRADE 2C, SOA 95.8). Duloxetine may be considered in pregnancy and breastfeeding but there are fewer data than for venlafaxine (GRADE 2C, SOA 95.3). Venlafaxine and duloxetine may be considered in breastfeeding if there is no alternative chronic pain medication (GRADE 2C, SOA 95.8).
Venlafaxine is compatible at conception and throughout pregnancy. There may be an increased risk of neonatal abstinence syndrome/short-term behavioural effects, but larger studies are needed to evaluate this finding (GRADE 2C, SOA 95.8). Duloxetine may be considered in pregnancy and breastfeeding but there are fewer data than for venlafaxine (GRADE 2C, SOA 95.3). Venlafaxine and duloxetine may be considered in breastfeeding if there is no alternative chronic pain medication (GRADE 2C, SOA 95.8).
NICE guidance and a systematic review described SSRI use to treat depression in pregnancy. The SSRIs are used to treat chronic pain at similar doses used to treat depression. A systematic review considered data from >50 000 infants exposed to SSRIs in utero , which did not show an overall increased risk for congenital malformations . It describes ongoing debate about the risk for cardiovascular malformations with first-trimester use of SSRIs with risk found from some but not all studies . Overall, any increase in absolute risk was thought unlikely to be clinically significant and may be associated with particular SSRIs, principally fluoxetine and paroxetine . The UKTIS summary of findings to date from studies on SSRIs states that a causal association between use of SSRIs in pregnancy and any type of congenital malformation has not been confirmed. It also describes conflicting results from other outcomes and concludes that available data do not suggest that SSRI use in pregnancy increases the risk of stillbirth and that possible associations with neurodevelopmental impairment in infants requires further study. An increased risk, however, of persistent pulmonary hypertension (PPHN) of the newborn has also been reported following exposure to SSRIs as a class beyond 20 weeks of gestation and, although it remains an uncommon event (0.2–1.2% vs 0.1–0.2% in the background population), it represents a potentially serious neonatal complication . Because there is no robust evidence of a superior safety profile for any one drug, switching between drugs is not recommended if depression is stable on treatment. NICE recommendations on use of TCAs, SSRIs or SNRIs include consideration of: the uncertainty about whether any increased risk to the foetus and other problems for the woman or baby can be attributed directly to these drugs or may be caused by other factors; and the risk of discontinuation symptoms in the woman and neonatal adaptation syndrome in the baby with most TCAs, SSRIs and (S)NRIs, in particular paroxetine and venlafaxine . There is limited information on the use of these drugs in breastfeeding. One small study showed temporarily reduced growth during exposure to fluoxetine in breastmilk. There have been no studies specifically investigating compatibility of paroxetine and sertraline with breastfeeding, but sertraline is reported as having one of the lowest rates of transmission to breastmilk . The following recommendations were based on evidence as shown in , available at Rheumatology online. Recommendations for SSRIs in pregnancy and breastfeeding Fluoxetine, paroxetine and sertraline are compatible with pregnancy (GRADE 1B, SOA 98.8). Based on limited evidence, SSRIs are compatible with breastfeeding (GRADE 2C, SOA 98.3).
Fluoxetine, paroxetine and sertraline are compatible with pregnancy (GRADE 1B, SOA 98.8). Based on limited evidence, SSRIs are compatible with breastfeeding (GRADE 2C, SOA 98.3).
Non-selective cyclooxygenase (COX) inhibitors have different indications in pregnancy. NSAIDs have analgesic and anti-inflammatory actions mediated through peripheral inhibition and differential selectivity of COX enzymes. In contrast, LDA at doses ≤150 mg/day is used to prevent thrombosis and pre-eclampsia in high-risk groups throughout pregnancy in patients with rheumatic diseases. Clopidogrel is an anti-platelet agent that may sometimes be used with or instead of LDA to reduce cardiovascular risk. NICE guidance , four systematic reviews and a case report/review evaluated LDA, NSAID and clopidogrel use in pregnancy.
Overall, there is no consistent evidence for an increased risk of teratogenic effects with NSAID use in pregnancy. There are mixed findings regarding a potential increased risk for miscarriage, with findings limited by methodology and larger associations reported for indomethacin and diclofenac use in the periconceptional period . There was no information on safety of COX2 inhibitors in pregnancy. NSAIDs are reported to increase the incidence of luteinized unruptured follicle (LUF) syndrome, whereby an anovulatory cycle results due to failure of normal follicular wall rupture despite normal ovarian follicular development and elevation of serum progesterone. COX-2 is active in the ovaries during follicular development; thus, inhibition via COX-2 inhibitors is thought to result in LUF. Although similar findings have been reported for both COX-1 and COX-2 NSAIDs, the risks have been found to be greater in patients with inactive disease and in those taking a COX-2 inhibitor (etoricoxib) rather than non-selective NSAIDs and it is reversible, following drug withdrawal . A systematic review of the effects of various drugs on foetal cardiac function evaluated by ultrasound found that all NSAIDs (including COX2 inhibitors) increased constriction in the ductus arteriosus, within 4–30 h of exposure and resolved by 72 h of discontinuation . In this study the critical gestational age (measured for indomethacin) increased from 5–10% of foetuses at weeks 26–27 to 50% at week 32. In 2020 the United States Food and Drug Administration (FDA) recommended that all NSAIDs be avoided from gestational week 20 rather than the previously advised 30 weeks. This advice was based on their updated review of published data to 2016 and 35 cases reported to FDA identifying an increased risk of oligohydramnios and renal impairment that began at 20 weeks of gestation and were mostly reversible on stopping NSAID. They clarify their advice for healthcare professionals, stating that the use of NSAIDs between 20 and 30 weeks of pregnancy should be limited to the lowest effective dose for the shortest duration .
LDA has been extensively studied and shown to improve outcomes in high-risk pregnancies. A Cochrane review of 77 trials, involving 40 249 women and their babies, found high quality evidence that antiplatelet agents (mostly LDA up to 150 mg/day) reduced pre-eclampsia and its complications . A systematic review of 22 RCTS of LDA plus heparin compared with other treatments in patients with APS, including 1515 treatment and 1531 control subjects, found that adverse pregnancy outcomes were significantly improved with LDA and heparin . The use of LDA in the third trimester of pregnancy is not associated with premature closure of ductus arteriosus and NICE guidelines for management of hypertension in pregnancy advises treatment with LDA until delivery .
There was no demonstrable maternofoetal toxicity in 13 (mostly second and third trimester only) pregnancy exposures to clopidogrel . UKTIS does not report on clopidogrel. LactMed considers various non-selective NSAIDs to be acceptable during breastfeeding and prefers ibuprofen because of its extremely low levels in breastmilk, short half-life and safe use in infants in doses much higher than those excreted in breastmilk, as an analgesic or anti-inflammatory agent in breastfeeding mothers. There was no information on COX2 inhibitors. Aspirin doses up to 325 mg daily are not excreted into breastmilk so LDA may be considered as an antiplatelet drug for use in breastfeeding women and is preferred to clopidogrel as no information is available on this drug . Recommendations for these drugs were based on evidence as shown in , available at Rheumatology online. Recommendations for NSAIDs and COX-2 inhibitors in pregnancy and breastfeeding Discordant findings from retrospective, large studies with population controls on the use of non-selective NSAIDs in the first trimester of pregnancy raise the possibility of a low risk of miscarriage and malformation. Therefore, these drugs should only be used intermittently in the first trimester of pregnancy (GRADE 1B, SOA 97.3). Intermittent rather than regular use of all non-selective NSAIDs except LDA is recommended throughout pregnancy and weaned from end of second trimester (26 weeks) to stop by gestational week 30 to avoid premature closure of the ductus arteriosus (GRADE 1B, SOA 98). At present there are limited data on selective cyclooxygenase-2 inhibitors; they should therefore be avoided during pregnancy (GRADE 2C, SOA 98.5). Non-selective NSAIDs (especially ibuprofen) are compatible with breastfeeding (GRADE 1C, SOA 98.8). Recommendations for LDA and clopidogrel in pregnancy and breastfeeding LDA may be continued throughout pregnancy and NICE guidelines (2019) for hypertension in pregnancy advise treatment with LDA (for prophylaxis of pre-eclampsia) until delivery (GRADE 1B, SOA 99.0). LDA is compatible with breastfeeding (GRADE 2C, SOA 99.8). There are limited data on clopidogrel but it may be considered where alternative drugs are not suitable in pregnancy and breastfeeding (GRADE 2C, SOA 96.3).
Discordant findings from retrospective, large studies with population controls on the use of non-selective NSAIDs in the first trimester of pregnancy raise the possibility of a low risk of miscarriage and malformation. Therefore, these drugs should only be used intermittently in the first trimester of pregnancy (GRADE 1B, SOA 97.3). Intermittent rather than regular use of all non-selective NSAIDs except LDA is recommended throughout pregnancy and weaned from end of second trimester (26 weeks) to stop by gestational week 30 to avoid premature closure of the ductus arteriosus (GRADE 1B, SOA 98). At present there are limited data on selective cyclooxygenase-2 inhibitors; they should therefore be avoided during pregnancy (GRADE 2C, SOA 98.5). Non-selective NSAIDs (especially ibuprofen) are compatible with breastfeeding (GRADE 1C, SOA 98.8).
LDA may be continued throughout pregnancy and NICE guidelines (2019) for hypertension in pregnancy advise treatment with LDA (for prophylaxis of pre-eclampsia) until delivery (GRADE 1B, SOA 99.0). LDA is compatible with breastfeeding (GRADE 2C, SOA 99.8). There are limited data on clopidogrel but it may be considered where alternative drugs are not suitable in pregnancy and breastfeeding (GRADE 2C, SOA 96.3).
These drugs were not considered in the previous BSR guideline and are now included because they may be used to treat certain inflammatory rheumatic diseases. UKTIS does not report on colchicine or dapsone. A systematic review and meta-analysis of colchicine use in 550 pregnancies of women with mostly familial Mediterranean fever (FMF) at doses of 1–2 mg per day, compared with 1263 non-exposed pregnancies found this drug did not significantly increase the incidence of foetal malformations or miscarriage when taken during pregnancy . However, the National Amyloidosis Centre recommends to continue the prescribed dose as there are no established safety concerns at colchicine doses >2 mg daily during pregnancy . No systematic review data was identified for dapsone. A review of 924 pregnancies exposed to dapsone to treat malaria was precluded from meaningful risk-benefit analysis due to limited reporting of outcomes . They concluded that the use of dapsone may be considered when no suitable alternative is available and the threat of malaria is the greater risk. It is a safe option in pregnant patients without glucose-6-phosphate dehydrogenase (G6PDH) deficiency and can be used during lactation while monitoring the baby for haemolysis and G6PDH deficiency . A review of the treatment of rheumatic and autoimmune skin disease in women during pregnancy concluded that dapsone may be safely and cautiously used during pregnancy . LactMed reports that long-term prophylactic maternal doses of colchicine up to 1.5 mg daily produce levels in milk that result in the infant receiving <10% of the maternal weight-adjusted dosage and no adverse effects have been reported from limited studies. It also states that the highest milk levels occur 2–4 h after a dose, so avoiding breastfeeding during this time can minimize the infant dose, or simply taking the drug after nursing. LactMed also states that dapsone can be used during breastfeeding; however, haemolytic anaemia might occur, especially in newborn infants and in those with G6PDH deficiency . Recommendations for these drugs were based on evidence as shown in , available at Rheumatology online. Recommendations for colchicine and dapsone in pregnancy and breastfeeding Colchicine therapy may be considered during pregnancy (GRADE 1B, SOA 99.5). Dapsone may be used in pregnancy (GRADE 2C, SOA 95.0). Colchicine may be used in breastfeeding (GRADE 2C, SOA 98.3). Dapsone may be used in breastfeeding and due to the risk of haemolytic anaemia it is advised to monitor the infant for signs of haemolysis, especially in newborn or premature breastfed infants (GRADE 2C, SOA 90.7).
Colchicine therapy may be considered during pregnancy (GRADE 1B, SOA 99.5). Dapsone may be used in pregnancy (GRADE 2C, SOA 95.0). Colchicine may be used in breastfeeding (GRADE 2C, SOA 98.3). Dapsone may be used in breastfeeding and due to the risk of haemolytic anaemia it is advised to monitor the infant for signs of haemolysis, especially in newborn or premature breastfed infants (GRADE 2C, SOA 90.7).
The deleterious effects of warfarin and compatibility of heparin in pregnancy are well described and evidence-based guidelines for the management of venous thromboembolism (VTE) and thrombophilia in pregnancy exist .
A systematic review and systematically produced guidelines describe the utility of heparin in the management of VTE and pregnancy morbidity in pregnant patients with antiphospholipid syndrome (APS). Heparin/LMWH does not cross the placenta . Heparins are compatible with breastfeeding. There were no additional studies identified, but LactMed states that no particular caution is required as the molecular weight of heparin is such that it is unlikely to be appreciably excreted into breastmilk.
Warfarin has the ability to cross the placenta and is associated with an increased risk of congenital abnormalities including a characteristic warfarin embryopathy (hypoplasia of the nasal bridge, congenital heart defects, ventriculomegaly, agenesis of the corpus callosum, stippled epiphyses) in ∼5% of foetuses exposed between 6 and 12 weeks of gestation. Warfarin should therefore be avoided between 6 and 12 weeks . While heparin/LMWH remains the anticoagulant of choice in pregnancy for the majority of patients considered to be at increased thrombotic risk in pregnancy, warfarin may be considered in pregnancy for women with mechanical heart valves (MHVs). The LMWH regimen could be used for those who refuse warfarin and other vitamin K antagonists (VKAs) . In management of atrial fibrillation, VKAs may be used after the first trimester . LactMed is reassuring both with regard to low drug levels in breastmilk and infant serum, and no reported adverse effects.
A systematic review identified 236 cases of direct oral anticoagulants (DOAC) use in pregnancy of rivaroxaban ( n = 178), dabigatran ( n = 27), apixaban ( n = 21) and edoxaban ( n = 10). DOACs were mostly used for prophylaxis or treatment of venous thromboembolism ( n = 91). DOACs were discontinued within the first 2 months of pregnancy in 84%, and the maximum reported duration of use was 26 weeks. Pregnancy outcome data were available for 140 pregnancies. Thirty-nine pregnancies were electively terminated. In the remaining 101 pregnancies, total miscarriage rate was 31% ( n = 31) and live birth rate was 68% ( n = 69, one missing). Foetal and neonatal abnormalities were reported in eight pregnancies, of which at least half were suspected to be related to rivaroxaban use during the first trimester of pregnancy. In only 18% of cases ( n = 42), the presence or absence of thrombotic and bleeding complications was reported. This limited evidence raises concern regarding embryo-foetal safety, with a high incidence of miscarriages and a 4% rate of congenital anomalies with the use of rivaroxaban. Overall, not enough data are available to judge safety and efficacy of the use of DOACs during pregnancy and rivaroxaban and dabigatran have both been shown to cause adverse obstetric outcomes in animal studies, described in summary of product characteristics . UKTIS does not report on DOACs. LactMed describes several case reports that consistently indicate that maternal doses of rivaroxaban of 15–30 mg daily produce low levels in milk that are considerably below doses required for anticoagulation in infants . Therefore, breastfeeding is not contraindicated if rivaroxaban is required by the mother. There are no data on the excretion of dabigatran into human milk.
Fondaparinux may be considered if there is an allergy or adverse response to LMWH, although solid data are lacking and minor transplacental passage has been demonstrated, without any adverse materno-foetal effects in five pregnancies . LactMed considers use of fondaparinux to be acceptable during breastfeeding . Recommendations on all anticoagulants were based on evidence as shown in , available at Rheumatology online. Recommendations for anticoagulants in pregnancy and breastfeeding LMWH heparin is compatible throughout pregnancy (GRADE 1A, SOA 100). LMWH is compatible with breastfeeding (GRADE 1C, SOA 100). The use of warfarin in pregnancy is associated with increased foetal risk throughout pregnancy and has limited indications, therefore should only be considered in exceptional circumstances (GRADE 1B, SOA 98.8). Warfarin is compatible with breastfeeding (GRADE 1A, SOA 100). Direct oral anticoagulants (DOACs) cannot be recommended in pregnancy (GRADE 1C, SOA 97.9). Rivaroxaban may be considered in breastfeeding (GRADE 2C, SOA 95.3). Other DOACs are not recommended in breastfeeding due to lack of human data and concerns from animal studies (GRADE 1C, SOA 97.4). Fondaparinux may be considered in pregnancy and breastfeeding if there is an allergy or adverse response to LMWH (GRADE 2C, SOA 95.5).
LMWH heparin is compatible throughout pregnancy (GRADE 1A, SOA 100). LMWH is compatible with breastfeeding (GRADE 1C, SOA 100). The use of warfarin in pregnancy is associated with increased foetal risk throughout pregnancy and has limited indications, therefore should only be considered in exceptional circumstances (GRADE 1B, SOA 98.8). Warfarin is compatible with breastfeeding (GRADE 1A, SOA 100). Direct oral anticoagulants (DOACs) cannot be recommended in pregnancy (GRADE 1C, SOA 97.9). Rivaroxaban may be considered in breastfeeding (GRADE 2C, SOA 95.3). Other DOACs are not recommended in breastfeeding due to lack of human data and concerns from animal studies (GRADE 1C, SOA 97.4). Fondaparinux may be considered in pregnancy and breastfeeding if there is an allergy or adverse response to LMWH (GRADE 2C, SOA 95.5).
Bisphosphonates are not ideal in women planning pregnancy because the absolute risk of fracture is small in this age group and the skeletal half-life of these drugs is very long. The number of human pregnancy exposures remains limited and a detailed literature review identified 40 pregnancies , while a systematic review published in abstract form described outcomes from 120 bisphosphonate-exposed pregnancies . Overall, rates of congenital malformation and miscarriages were comparable in bisphosphonate and controls, although limiting factors included: few comparator groups; heterogeneous maternal disease; concomitant medication; and small sample size. Controlled studies have demonstrated possible associations between preconceptual/gestational bisphosphonate exposure and increased risk of spontaneous abortion, decreased infant birth weight, and lower gestational age at delivery. These findings, however, may reflect data limitations and/or uncontrolled confounding (UKTIS). Therefore, further controlled studies are required to fully establish the safety of bisphosphonates in pregnancy and they are not drugs of choice in women planning pregnancy. LactMed states that limited evidence indicates that breastfeeding after cessation of long-term bisphosphonate treatment appears to have no adverse effects on the infant. There is no information on the use of alendronate or risedronate during breastfeeding. Limited information indicates that maternal doses of pamidronate of 30 mg intravenously produce very low levels in milk and because pamidronate has a serum half-life of ∼3 h, is highly bound to calcium and poorly absorbed orally, absorption of pamidronate by a breastfed infant is unlikely. Therefore, withholding breastfeeding for 12–24 h after a dose should ensure that the breastfed infant is exposed to little or no pamidronate. If the mother receives a bisphosphonate during pregnancy or nursing, some experts recommend monitoring the infant’s serum calcium during the first 2 months postpartum . These recommendations were based on evidence as shown in , available at Rheumatology online. Recommendations for bisphosphonates in pregnancy and breastfeeding There is insufficient data upon which to recommend bisphosphonates in pregnancy or to advise a specific time for them to be stopped pre-conception. Given their biological half-life in bone of up to 10 years and no evidence of harm from limited reports of their use in pregnancy, a pragmatic recommendation is that they should be stopped 3 months in advance of pregnancy (GRADE 2C, SOA 96.8). There are no data on which to base a recommendation for the use of bisphosphonates during breastfeeding (GRADE 2C, SOA 98.5).
There is insufficient data upon which to recommend bisphosphonates in pregnancy or to advise a specific time for them to be stopped pre-conception. Given their biological half-life in bone of up to 10 years and no evidence of harm from limited reports of their use in pregnancy, a pragmatic recommendation is that they should be stopped 3 months in advance of pregnancy (GRADE 2C, SOA 96.8). There are no data on which to base a recommendation for the use of bisphosphonates during breastfeeding (GRADE 2C, SOA 98.5).
Patients with ARD, particularly renal SLE and systemic sclerosis (SSc) frequently require anti-hypertensive treatment for sometimes life-threatening disease, such as scleroderma renal crisis, that may require specialist use of certain anti-hypertensive drugs. The introduction of ACEis for the treatment of SSc renal crisis has significantly reduced mortality rates of up to 50% in the 1970s to a mortality of up to 20% at 6 months after introducing ACEis. The use of ACEis may therefore be indicated (also in pregnancy) in exceptional circumstances . The management of pre-existing and new onset hypertension in pregnancy has been comprehensively reviewed and updated in the 2019 NICE guideline, Hypertension in Pregnancy: Diagnosis and Management .
Disruption of the renin-angiotensin system (RAS) in pregnancy by maternal treatment with ACEis or angiotensin receptor blockers (ARBs) in the second/third trimester leads to abnormal foetal renal development, known as foetal RAS blockade syndrome . There are conflicting results, however, on the risk of this fetopathy occurring after first trimester exposure. A systematic review and meta-analysis of 19 articles involving 4 163 753 pregnant women found a significant association between overall congenital malformations and first trimester‐only exposure to ACEis/ARBs (odds ratio 1.94; 95% CI 1.71, 2.21; P <0.0001). This review also found a significant association between cardiovascular malformations, miscarriage and stillbirth and ACEi/ARB exposure. A similar risk was observed in a cohort of 1 333 624 pregnancies , including 4107 (0.31%) following first trimester ACEi exposure that found an increased risk of overall malformations in the ACEi-exposed pregnancies [unadjusted relative risk (RR), 1.82; 95% CI 1.61, 2.06] and of cardiac malformations (RR 2.95; 95% CI 2.50, 3.47). On further analysis, however, restricting the cohort to pregnancies complicated by chronic hypertension (both exposed and unexposed) and accounting for other confounding factors, there was no significant increase in the risk of any of the outcomes assessed. NICE guidelines state that women taking ACEis/ARBs should be advised of the increased risk of congenital anomalies if these drugs are taken during pregnancy and to discuss alternative antihypertensive drugs with their clinician responsible for managing their condition . If they become pregnant on ACEis/ARBs they should be stopped (preferably within 2 working days of notification) and other antihypertensive treatments offered. UKTIS recommends that where prolonged first trimester exposure has occurred, a 20-week anomaly scan should focus on cardiovascular, renal and neurological development, in addition to the routine anatomical checks. UKTIS states that ACEi fetopathy following exposure to ACEis in the second and third trimesters of pregnancy is well-described and may include oligohydramnios, renal tubular dysgenesis, neonatal anuria, hypocalvaria, pulmonary hypoplasia, persistent patent ductus arteriosus, mild-to-severe intrauterine growth restriction, and foetal or neonatal death. It is proposed that these effects occur as a result of a direct effect on the foetal RAS which begins to function from ∼26 weeks gestation. A small prospective case series has suggested that the risk period for ACEi fetopathy is with exposure beyond 20 weeks gestation. Due to data limitations, the absolute risk of ACEi fetopathy is unclear. Due to the risk of ACEi fetopathy, use of ACEis in the second and third trimesters is generally contraindicated and should only be reserved for cases of severe maternal illness that cannot be managed using alternative drugs . Negligible amounts of enalapril and captopril are transferred into breastmilk with no adverse effects reported on the breastfed infants of mothers treated with short-acting ACEis . NICE states that enalapril may be offered to treat hypertension in breastfeeding mothers with appropriate monitoring of maternal renal function and serum potassium . The following recommendations were based on evidence as shown in , available at Rheumatology online. Recommendations for ACEis/ARBs in pregnancy and breastfeeding ACEis and ARBs should be stopped as soon as possible when pregnancy is confirmed in the first trimester and if necessary an alternative antihypertensive compatible with pregnancy given (GRADE 1A, SOA 100). ACEis/ARBs should be avoided in the second and third trimester but may be considered under specialist advice in certain circumstances (GRADE 1C, SOA 98.5). Based on limited evidence, enalapril is compatible with breastfeeding (GRADE 2C, SOA 98.5).
ACEis and ARBs should be stopped as soon as possible when pregnancy is confirmed in the first trimester and if necessary an alternative antihypertensive compatible with pregnancy given (GRADE 1A, SOA 100). ACEis/ARBs should be avoided in the second and third trimester but may be considered under specialist advice in certain circumstances (GRADE 1C, SOA 98.5). Based on limited evidence, enalapril is compatible with breastfeeding (GRADE 2C, SOA 98.5).
Calcium channel blockers (CCBs) including amlodipine, diltiazem, felodipine, lacidipine, lercanidipine, nicardipine, nifedipine, nimodipine and verapamil are mainly used for the treatment and prophylaxis of angina, and the treatment of hypertension where an ACEi/ARB is unsuitable. In patients with rheumatic disease, nifedipine or amlodipine are also used to treat Raynaud’s phenomenon. UKTIS has not identified an increased risk of congenital malformations or other adverse pregnancy outcomes with CCBs, although data remains too limited to draw firm conclusions on many outcomes. Data on rates of preterm delivery, foetal growth and neurodevelopmental outcomes are too limited and/or confounded to permit an accurate risk assessment, but where a CCB is required to treat maternal hypertension or as a tocolytic, foetal benefits of use are likely to outweigh any unspecified risk and treatment should not be withheld on this basis . Data is more limited for amlodipine so it is not included in alternatives to treat hypertension in pregnancy that include labetalol, nifedipine or methyldopa in order of preference . LactMed describes low levels of nifedipine and amlodipine in breastmilk, without any adverse effects being reported among exposed infants . NICE guidance of other antihypertensive drugs that may be offered in breastfeeding mothers include nifedipine, amlodipine, atenolol or labetolol . Recommendations were based on evidence as shown in , available at Rheumatology online. Recommendations for CCBs in pregnancy and breastfeeding Nifedipine is compatible with pregnancy with no direct evidence of harm at doses up to 90 mg/day (GRADE 1A, SOA 99.0). Nifedipine is compatible with breastfeeding (GRADE 1B, SOA 100). Amlodipine can be considered in pregnancy and breastfeeding as there is no evidence of harm (GRADE 1C, SOA 97.9).
Nifedipine is compatible with pregnancy with no direct evidence of harm at doses up to 90 mg/day (GRADE 1A, SOA 99.0). Nifedipine is compatible with breastfeeding (GRADE 1B, SOA 100). Amlodipine can be considered in pregnancy and breastfeeding as there is no evidence of harm (GRADE 1C, SOA 97.9).
Moderate-to-severe pulmonary hypertension (PHT) is a rare complication of certain ARDs and remains a contraindication to planned pregnancy with high mortality. Unintentional pregnancy and/or patient choice, however, means that treatment of this condition with specific pulmonary vasodilators may be required in pregnancy. Limited information on use of these drugs in human pregnancy was identified in the previous BSR guideline . No studies were identified examining pregnancy outcomes after paternal exposure to any of these pulmonary vasodilators.
Sildenafil has been studied in the context of trying to improve utero-placental circulation in pregnancies affected by severe foetal growth restriction. A systematic review and meta-analysis were identified, examining the utility of sildenafil being used for treatment or prevention of obstetric diseases compared with placebo. They analysed 598 pregnant women from seven clinical trials with pre-eclampsia ( n = 139), intrauterine growth restriction ( n = 275) and oligohydramnios ( n = 184) and found a significant improvement in birthweight following sildenafil treatment during pregnancy, with no difference in other outcomes . However, in 2018 a randomised controlled study looking at using sildenafil to treat pregnant women in whom there was significant foetal growth restriction was halted early due to a number of neonates having persistent pulmonary hypertension of the newborn (PPHN). This multi-centered study enrolled women with a singleton pregnancy between 18 and 30 weeks with severe foetal growth restriction of likely placental origin, where the likelihood of perinatal death/severe morbidity was estimated to be significant. Participants were randomised into sildenafil or placebo arms. One of three study sites reported that PPHN appeared to be more prevalent in infants exposed in utero to sildenafil compared with placebo-exposed infants ( n = 17/93 vs n = 3/90), that death among infants with PPHN was more common following sildenafil exposure ( n = 11/17 vs n = 0/3), and that when the overall neonatal death rate was considered, there was a non-statistically significant trend towards an increased risk following sildenafil exposure ( n = 19/93 vs n = 9/90). As a result, the trial was halted early and sildenafil is no longer recommended to improve placental function in severely growth-restricted babies . However, it should be noted that two further study sites for this trial did not detect an increased risk of PPHN and overall, this trial has not identified a clear beneficial effect of sildenafil on foetal outcome. PPHN is more prevalent in premature and growth-restricted foetuses and is a relatively rare complication in healthy babies delivered at term. UKTIS recommends that general use of sildenafil in pregnancy should be avoided where possible . However, in the context of maternal pulmonary hypertension there is clear benefit in the use of sildenafil to reduce the effects of pulmonary hypertension, which often gets worse during pregnancy. Case studies/case series data suggest that sildenafil exposure was not associated with miscarriage or congenital anomaly; however, the data is extremely limited (around 18 reported pregnancies). The risks and benefits of continuing sildenafil should be discussed with the patients, but most will likely require ongoing treatment. LactMed describes limited data showing that sildenafil and its active metabolite are poorly excreted into breastmilk and amounts ingested by the infant are small and would not be expected to cause any adverse effects in breastfed infants .
Animal data have revealed teratogenicity due to bosentan, including malformations of the head, mouth, face and large blood vessels in addition to an increased number of stillbirths and increased mortality . Previously, we identified data from 12 pregnancies of women with PHT treated with bosentan in pregnancy plus multiple other medications, including sildenafil and iloprost, with reduced pregnancy duration of 37 weeks in one and reduced birth weight in two cases but no other maternal complications or foetal loss . We identified one further case report of a patient with Eisenmenger syndrome exposed to long-term bosentan before and during pregnancy that was delivered by caesarean section at 27 weeks due to severe maternal PHT without any evidence of teratogenic effects of bosentan . UKTIS does not report on bosentan. LactMed states there is little published experience with bosentan during breastfeeding and an alternate drug may be preferred, especially while nursing a newborn or preterm infant .
Previously we identified data on 23 pregnancies of patients with PHT (three with SLE) treated with iloprost ( n = 5 pregnancies) or epoprostenol ( n = 15 pregnancies) and three other prostacyclines (unspecified type) in patients who were taking multiple other medications, including immunosuppressants, sildenafil and bosentan . Findings of premature deliveries and reduced birthweight were confounded by maternal disease. Furthermore, maternal complications were attributable to PHT. We did not identify any new evidence and UKTIS and LactMed do not report on iloprost or epoprostenol. The recommendations were based on evidence as shown in , available at Rheumatology online. Recommendations for pulmonary vasodilators in pregnancy and breastfeeding Established moderate to severe PHT remains a contraindication to pregnancy. If pregnancy occurs, the use of these pulmonary vasodilator drugs in pregnancy should be considered only as part of a multidisciplinary team assessment (GRADE 1C, SOA 99.5). Limited evidence supports the use of prostacyclines to treat PHT during pregnancy (GRADE 2C, SOA 98.0). Limited evidence supports the use of sildenafil to treat PHT during pregnancy (GRADE 2C, SOA 98.0). Bosentan is teratogenic in animals and although there is no evidence of harm from human pregnancy, the evidence is insufficient to recommend in pregnancy (GRADE 1C, SOA 98.8). There are no data relating to breastfeeding exposure to pulmonary vasodilators on which to base a recommendation (GRADE 2C, SOA 98.8).
Established moderate to severe PHT remains a contraindication to pregnancy. If pregnancy occurs, the use of these pulmonary vasodilator drugs in pregnancy should be considered only as part of a multidisciplinary team assessment (GRADE 1C, SOA 99.5). Limited evidence supports the use of prostacyclines to treat PHT during pregnancy (GRADE 2C, SOA 98.0). Limited evidence supports the use of sildenafil to treat PHT during pregnancy (GRADE 2C, SOA 98.0). Bosentan is teratogenic in animals and although there is no evidence of harm from human pregnancy, the evidence is insufficient to recommend in pregnancy (GRADE 1C, SOA 98.8). There are no data relating to breastfeeding exposure to pulmonary vasodilators on which to base a recommendation (GRADE 2C, SOA 98.8).
There remains limited data on paternal exposure to drugs used to treat rheumatic disease and reports of teratogenic effects linked with paternal exposure are lacking for all drugs considered in this guideline. Our updated search revealed the following additional information. Links with adverse neurodevelopmental outcomes but not asthma have been reported following paternal exposure to paracetamol from the largest cohort study of paracetamol in pregnancy. A direct causal effect, however, remains unproven. An observational prospective cohort study from Sweden of 3983 children born to fathers receiving antidepressant treatment around conception, included fathers treated with: SSRIs, n = 2865; SNRIs, n = 470; and TCAs, n = 240. This study found paternal intake of all antidepressants studied to be safe with respect to the risk of preterm birth, malformation, autism or intellectual disability. A systematic review of the effect of paternal exposure to immunosuppressive drugs on sexual function, reproductive hormones, fertility, pregnancy and offspring outcomes found very weak evidence of reduced sperm parameters with codeine, tramadol and CCBs but not NSAIDs, LDA or ACEi and no abnormalities in offspring were reported for any drug. Interestingly, low-dose lisinopril (2.5 mg/day) has been shown to increase total sperm count and motility in a randomised, controlled, crossover pilot of study of normotensive men with idiopathic oligospermia leading to an unassisted pregnancy rate of 48.5% . Small studies of chronic use of NSAIDs in men with rheumatic disease have not indicated any impairment of spermatogenesis and no evidence for harmful effects of NSAIDs on offspring . A systematic review of eight studies including 166 cases of paternal exposure found inconsistent reports of adverse effects of colchicine on sperm quality and only one study ( n = 53) of paternal exposure did not find an adverse effect on offspring . The following recommendations were based on evidence as shown in , available at Rheumatology online. Recommendations for paternal exposure Paracetamol is compatible with paternal exposure (GRADE 1B, SOA 98.5). Amitriptyline, SNRIs and SSRIs are compatible with paternal exposure (GRADE 1B, SOA 98.5). Non-selective NSAIDs are compatible with paternal exposure (GRADE 1C, SOA 98.4). Based on limited or no data and no association with adverse foetal development or pregnancy outcome, paternal exposure to all other drugs described in this guideline are unlikely to be harmful (GRADE 2C, SOA 97.3).
Paracetamol is compatible with paternal exposure (GRADE 1B, SOA 98.5). Amitriptyline, SNRIs and SSRIs are compatible with paternal exposure (GRADE 1B, SOA 98.5). Non-selective NSAIDs are compatible with paternal exposure (GRADE 1C, SOA 98.4). Based on limited or no data and no association with adverse foetal development or pregnancy outcome, paternal exposure to all other drugs described in this guideline are unlikely to be harmful (GRADE 2C, SOA 97.3).
Implementation Awareness of these guidelines will aid clinical practitioners and patients in decision making and will be raised through presentation at local, regional and national meetings. No barriers to implementation of these guidelines are anticipated. Key standards of care Patients with rheumatic disease should receive tailored pre-pregnancy counselling and then be reviewed during pregnancy and the four month post-partum period by clinical practitioners with expertise in the management of rheumatic disease in pregnancy, in addition to their routine obstetric care. They should have access to written information on relevant medications in pregnancy and breastfeeding that is accurate and allows them to make informed decisions regarding compatibility of certain drugs in pregnancy. Future research agenda The limitation of current evidence highlights the need for a national pregnancy registry for patients with rheumatic disease, as currently exists for women with epilepsy. All women with rheumatic disease who become pregnant would be eligible to register, whether or not they are on anti-rheumatic treatment. The prospective pregnancy outcome data would then be published to display information on outcomes such as miscarriage and congenital anomalies in patients treated with anti-rheumatic and other drug therapy. These data would also be used to answer specific questions where data is currently lacking. Data relating to the impact of paternal exposure to these drugs (both fertility and male-mediated teratogenicity), as well as breastfeeding exposure is particularly limited, and further research in these areas is urgently required. Mechanism for audit of the guideline An audit pro forma to assess compliance with these guidelines is shown in , available at Rheumatology online.
Awareness of these guidelines will aid clinical practitioners and patients in decision making and will be raised through presentation at local, regional and national meetings. No barriers to implementation of these guidelines are anticipated.
Patients with rheumatic disease should receive tailored pre-pregnancy counselling and then be reviewed during pregnancy and the four month post-partum period by clinical practitioners with expertise in the management of rheumatic disease in pregnancy, in addition to their routine obstetric care. They should have access to written information on relevant medications in pregnancy and breastfeeding that is accurate and allows them to make informed decisions regarding compatibility of certain drugs in pregnancy.
The limitation of current evidence highlights the need for a national pregnancy registry for patients with rheumatic disease, as currently exists for women with epilepsy. All women with rheumatic disease who become pregnant would be eligible to register, whether or not they are on anti-rheumatic treatment. The prospective pregnancy outcome data would then be published to display information on outcomes such as miscarriage and congenital anomalies in patients treated with anti-rheumatic and other drug therapy. These data would also be used to answer specific questions where data is currently lacking. Data relating to the impact of paternal exposure to these drugs (both fertility and male-mediated teratogenicity), as well as breastfeeding exposure is particularly limited, and further research in these areas is urgently required.
An audit pro forma to assess compliance with these guidelines is shown in , available at Rheumatology online.
are available at Rheumatology online.
keac552_Supplementary_Data Click here for additional data file.
|
Executive Summary: British Society for Rheumatology guideline on prescribing drugs in pregnancy and breastfeeding: immunomodulatory anti-rheumatic drugs and corticosteroids | 9fec4aaf-d26d-4ee2-92bc-685f1942bf2f | 10070067 | Internal Medicine[mh] | Need for guideline The rationale behind this update of the 2016 British Society for Rheumatology (BSR) guidelines on prescribing anti-rheumatic drugs in pregnancy and breastfeeding was previously described . Additional evidence-based guidelines on managing rheumatic disease in pregnancy now exist . However, emerging information on pregnancy exposures to biologic DMARDs, biosimilars and targeted synthetic DMARDs requires regular review to assess their safety in pregnancy. Uncertainty around use of these drugs in pregnancy may lead to withdrawal of treatment from women with inflammatory rheumatic diseases (IRDs) unnecessarily , which can increase the risk of poor disease control during pregnancy . Data from Mothers and Babies: Reducing Risk through Audits and Confidential Enquiries across the UK (MBRRACE-UK) between 2017 and 2019 found that 8.8 women per 100 000 died during pregnancy or up to six weeks after childbirth or the end of pregnancy, and most women who died had multiple health problems or other vulnerabilities . In all decisions regarding medication choices and changes, it is important to consider the potential for deterioration in the mother's wellbeing through side effects or reduced disease control (and its adverse impact on the baby) . As such, the potential benefit to the foetus from any drug changes in the mother must be balanced against the possible risks to the foetus from loss of disease control in the mother . Objectives of guideline To update the previous BSR guidelines on prescribing in pregnancy in rheumatic disease of corticosteroids and immunomodulatory drugs ( , available at Rheumatology online) by systematically reviewing all evidence published since the previous guideline. This guideline will highlight medications that should be stopped and/or avoided in the reproductive age group unless highly effective contraception is used, in line with guidance issued by the Medicines and Healthcare Products Regulatory Agency and the Faculty of Sexual and Reproductive Healthcare . Target audience This guideline is directed at healthcare professionals in the UK involved in managing patients with rheumatic disease who are (or are planning to become) pregnant and/or breastfeeding, men with rheumatic disease who are planning to conceive, and patients with rheumatic disease who have unintentionally conceived while taking these medications. This audience includes rheumatologists, rheumatology nurses/allied health professionals, rheumatology speciality trainees and pharmacists, as well as the patients themselves. The guideline will also be useful to obstetricians, obstetric physicians, midwives, renal physicians, dermatologists, gastroenterologists, respiratory physicians and general practitioners who prescribe these medications in pregnancy. This guideline uses the terms ‘woman’, ‘maternal’ or ‘mother’ throughout. These should be taken to include people who do not identify as women but are pregnant or have given birth . Where the term ‘breastfeeding’ is used in this guideline it also refers to infant breastmilk exposure via other methods (e.g. expressed breastmilk, administered via a bottle). The areas the guideline does not cover This guideline does not cover the management of infertility or the indications for these drugs in rheumatic diseases in pregnancy. Other drug categories (pain management; non-steroidal anti-inflammatory drugs and low dose aspirin; anticoagulants; bisphosphonates; anti-hypertensives; and pulmonary vasodilators) are considered in the BSR guideline on prescribing drugs in pregnancy and breastfeeding: comorbidity medications used in rheumatology practice . All recommendations in this guideline were formulated by the working group on the basis of published evidence at the time of the systematic literature search, and do not necessarily refer to licencing information or Summary of Product Characteristics for individual medications. Stakeholder involvement A guideline working group (GWG), chaired by I.G., consisted of rheumatologists from a range of clinical backgrounds, allied health professionals, other specialists in women’s health, lay members, and representatives from the United Kingdom Teratology Information Service (UKTIS), this is shown in , available at Rheumatology online.
The rationale behind this update of the 2016 British Society for Rheumatology (BSR) guidelines on prescribing anti-rheumatic drugs in pregnancy and breastfeeding was previously described . Additional evidence-based guidelines on managing rheumatic disease in pregnancy now exist . However, emerging information on pregnancy exposures to biologic DMARDs, biosimilars and targeted synthetic DMARDs requires regular review to assess their safety in pregnancy. Uncertainty around use of these drugs in pregnancy may lead to withdrawal of treatment from women with inflammatory rheumatic diseases (IRDs) unnecessarily , which can increase the risk of poor disease control during pregnancy . Data from Mothers and Babies: Reducing Risk through Audits and Confidential Enquiries across the UK (MBRRACE-UK) between 2017 and 2019 found that 8.8 women per 100 000 died during pregnancy or up to six weeks after childbirth or the end of pregnancy, and most women who died had multiple health problems or other vulnerabilities . In all decisions regarding medication choices and changes, it is important to consider the potential for deterioration in the mother's wellbeing through side effects or reduced disease control (and its adverse impact on the baby) . As such, the potential benefit to the foetus from any drug changes in the mother must be balanced against the possible risks to the foetus from loss of disease control in the mother .
To update the previous BSR guidelines on prescribing in pregnancy in rheumatic disease of corticosteroids and immunomodulatory drugs ( , available at Rheumatology online) by systematically reviewing all evidence published since the previous guideline. This guideline will highlight medications that should be stopped and/or avoided in the reproductive age group unless highly effective contraception is used, in line with guidance issued by the Medicines and Healthcare Products Regulatory Agency and the Faculty of Sexual and Reproductive Healthcare .
This guideline is directed at healthcare professionals in the UK involved in managing patients with rheumatic disease who are (or are planning to become) pregnant and/or breastfeeding, men with rheumatic disease who are planning to conceive, and patients with rheumatic disease who have unintentionally conceived while taking these medications. This audience includes rheumatologists, rheumatology nurses/allied health professionals, rheumatology speciality trainees and pharmacists, as well as the patients themselves. The guideline will also be useful to obstetricians, obstetric physicians, midwives, renal physicians, dermatologists, gastroenterologists, respiratory physicians and general practitioners who prescribe these medications in pregnancy. This guideline uses the terms ‘woman’, ‘maternal’ or ‘mother’ throughout. These should be taken to include people who do not identify as women but are pregnant or have given birth . Where the term ‘breastfeeding’ is used in this guideline it also refers to infant breastmilk exposure via other methods (e.g. expressed breastmilk, administered via a bottle).
This guideline does not cover the management of infertility or the indications for these drugs in rheumatic diseases in pregnancy. Other drug categories (pain management; non-steroidal anti-inflammatory drugs and low dose aspirin; anticoagulants; bisphosphonates; anti-hypertensives; and pulmonary vasodilators) are considered in the BSR guideline on prescribing drugs in pregnancy and breastfeeding: comorbidity medications used in rheumatology practice . All recommendations in this guideline were formulated by the working group on the basis of published evidence at the time of the systematic literature search, and do not necessarily refer to licencing information or Summary of Product Characteristics for individual medications.
A guideline working group (GWG), chaired by I.G., consisted of rheumatologists from a range of clinical backgrounds, allied health professionals, other specialists in women’s health, lay members, and representatives from the United Kingdom Teratology Information Service (UKTIS), this is shown in , available at Rheumatology online.
Statement of scope of literature search and strategy employed Evidence used to develop these guidelines was compiled from a systematic literature search conducted according to guidelines of Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) . Studies were identified by searching MEDLINE and Embase databases from 1 January 2014 to 31 December 2020, with additional studies identified through the Cochrane, LactMed and UKTIS databases ( , available at Rheumatology online). The searches were conducted in January 2021. Limits were placed for English language studies. The searches were not limited by disease indication; studies in non-rheumatic diseases were considered, if relevant. Evidence from the current searches (referred to as ‘recent studies’ in the text) was combined with data from the previous guideline (referred to as ‘previous studies’) , to inform recommendations. Relevant pharmaceutical companies were also contacted for additional data. Two independent reviewers screened the titles and abstracts of articles, reviewed the full texts of studies that met inclusion criteria ( , available at Rheumatology online), and performed data extraction. Any disagreements were resolved by group discussion. Statement of methods used to formulate the recommendations (levels of evidence) This guideline was developed in line with BSR’s Guidelines Protocol using Grading of Recommendations, Assessment, Development and Evaluations (GRADE) methodology, to determine quality of evidence and strength of recommendation. Accompanying each recommendation, in brackets, is the strength of recommendation, quality of evidence and strength of agreement (SOA). Recommendations were categorized as strong (1), when the benefits clearly outweigh the risks (or vice versa), or weak (2), when the benefits were more closely balanced with the risks or more uncertain. The quality of evidence was determined as high (A), moderate (B) or low/very low (C), reflecting the confidence in the estimates of benefits or harm. The wording of each recommendation was agreed by all members and subjected to a vote for SOA on a scale of 1–100 (no to complete agreement). The guideline will be updated in five years.
Evidence used to develop these guidelines was compiled from a systematic literature search conducted according to guidelines of Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) . Studies were identified by searching MEDLINE and Embase databases from 1 January 2014 to 31 December 2020, with additional studies identified through the Cochrane, LactMed and UKTIS databases ( , available at Rheumatology online). The searches were conducted in January 2021. Limits were placed for English language studies. The searches were not limited by disease indication; studies in non-rheumatic diseases were considered, if relevant. Evidence from the current searches (referred to as ‘recent studies’ in the text) was combined with data from the previous guideline (referred to as ‘previous studies’) , to inform recommendations. Relevant pharmaceutical companies were also contacted for additional data. Two independent reviewers screened the titles and abstracts of articles, reviewed the full texts of studies that met inclusion criteria ( , available at Rheumatology online), and performed data extraction. Any disagreements were resolved by group discussion.
This guideline was developed in line with BSR’s Guidelines Protocol using Grading of Recommendations, Assessment, Development and Evaluations (GRADE) methodology, to determine quality of evidence and strength of recommendation. Accompanying each recommendation, in brackets, is the strength of recommendation, quality of evidence and strength of agreement (SOA). Recommendations were categorized as strong (1), when the benefits clearly outweigh the risks (or vice versa), or weak (2), when the benefits were more closely balanced with the risks or more uncertain. The quality of evidence was determined as high (A), moderate (B) or low/very low (C), reflecting the confidence in the estimates of benefits or harm. The wording of each recommendation was agreed by all members and subjected to a vote for SOA on a scale of 1–100 (no to complete agreement). The guideline will be updated in five years.
A flow diagram of study selection is shown in . The findings for all drug exposures, including information from the previous BSR guideline , are summarized in the full-length guideline. An overall summary of the compatibility of each drug pre-conception, during pregnancy, with breastmilk exposure, and with paternal exposure is shown in . For each drug, maternal information is summarized in and , while paternal information is summarized in . The data synthesis strategy for is shown in , available at Rheumatology online. The list of references included within the evidence tables is shown in , also available at Rheumatology online. Generic recommendations on prescribing immunomodulatory drugs and/or corticosteroids in rheumatic disease in pregnancy Pre-conception counselling should be addressed by all healthcare professionals, with referral to professionals with relevant expertise as appropriate, to optimize disease control before pregnancy; with advice on the timing of pregnancy, and drug therapy before, during and after pregnancy, including contraception (GRADE 1A, SOA 99.5%). If a woman is planning pregnancy, avoid pregnancy-incompatible drugs (GRADE 1A, SOA 100%). The risks and benefits to the mother and foetus of drug treatment to control maternal disease should be discussed and clearly documented by all healthcare professionals involved in the patient’s care (GRADE 1A, SOA 99.5%). Immunomodulatory drugs that are contraindicated in pregnancy should be switched to a pregnancy-compatible alternative in advance of conception to ensure maintenance of disease control on the new medication (GRADE 1A, SOA 100%). When no pregnancy-compatible drugs are suitable, control of severe/life-threatening maternal disease should take priority over concerns for potential foetal outcomes (GRADE 1B, SOA 99.0%). All biologic DMARDs may be continued throughout pregnancy if required to control active/severe maternal disease (GRADE 1B, SOA 98.5%). Immunisation schedules in infants after in-utero exposure to biologic DMARDs will depend on timing of exposure, bioavailability and persistence of the drug, mechanism of action of the drug, and live vaccines (GRADE 1C, SOA 99.5%). Where possible, the minimum effective dose of immunomodulatory drug or corticosteroid should be used to maintain maternal disease suppression, and stopping the drug during pregnancy may be considered in women at low risk of disease flare on withdrawal of therapy (GRADE 1B, SOA 100%). Some drugs may reduce male fertility, but paternal drug exposure in humans has not convincingly been associated with adverse foetal development or pregnancy outcome. Although the evidence is weak, men who take rheumatological medicines should be reassured about the safety of conceiving (GRADE 2C, SOA 98.4%). Antimalarials Recommendations were based on 23 studies of HCQ identified in the current literature search and 23 studies identified in the previous guideline’s literature search, in addition to four studies of chloroquine and no studies of mepacrine. Three recent and one previous study of breastmilk exposure to HCQ were identified. Recommendations for hydroxychloroquine in pregnancy and breastmilk exposure HCQ remains the antimalarial of choice in women planning a pregnancy with rheumatic disease in need of treatment, and should be continued during pregnancy at dose of ≤400 mg/day (GRADE 1B, SOA 100%). HCQ is compatible with breastmilk exposure (GRADE 1B, SOA 99.5%). Corticosteroids Recommendations were based on 11 recent studies of prednisolone, one study of methylprednisolone, and previous studies: 47 on prednisolone; 31 on dexamethasone; 27 on betamethasone; and 10 on general corticosteroid use. Recommendations for corticosteroids in pregnancy and breastmilk exposure Prednisolone is compatible with pregnancy and is the preferred corticosteroid in the treatment of maternal rheumatological disease in pregnancy and requires shared care with obstetric teams to monitor maternal blood pressure and blood glucose (GRADE 1B, SOA 100%). Where possible, the dose of prednisolone should be <20 mg/day and tapered to the minimum effective dose to control maternal disease, in conjunction with steroid-sparing drugs compatible with pregnancy (GRADE 1C, SOA 99.5%). Prednisolone is compatible with breastmilk exposure (GRADE 1B, SOA 100%). Methylprednisolone has similar rates of placental transfer to prednisolone and would therefore be expected to be compatible with pregnancy and breastmilk exposure (GRADE 2C, SOA 99%). Conventional synthetic DMARDs Methotrexate Recommendations were based on 12 recent and 10 previous studies of MTX. Recommendations for methotrexate in pregnancy and breastmilk exposure MTX at any dose should be avoided in pregnancy and stopped at least one month in advance of planned conception, when it should be switched to another pregnancy-compatible drug to ensure maintenance of maternal disease suppression (GRADE 1A, SOA 98%). In women treated with low-dose (≤25 mg/week) MTX within one month prior to conception, folic acid supplementation (5 mg/day) should be continued up to 12 weeks of pregnancy (GRADE 1B, SOA 99.5%). In unintended pregnancy on low-dose (≤25 mg/week) MTX, there is minimal risk to the foetus; the drug should be stopped immediately, folic acid supplementation (5 mg/day) continued, and a careful evaluation of foetal risk with early referral to a foetal medicine department considered (GRADE 1C, SOA 100%). Although only minute amounts of MTX are excreted into breastmilk, MTX cannot be recommended in breastfeeding because of theoretical risks and insufficient data on outcomes (GRADE 2C, SOA 99%). Sulfasalazine Recommendations were based on six studies of SSZ identified in the previous guideline. Recommendations for sulfasalazine in pregnancy and breastmilk exposure SSZ is compatible throughout pregnancy, with folic acid 5 mg/day recommended in the periconception period and during the first trimester (GRADE 1B, SOA 100%). SSZ is compatible with breastmilk exposure in healthy, full-term infants (GRADE 1C, SOA 99.5%). Leflunomide Recommendations were based on three recent and seven previous studies of LEF. Recommendations for leflunomide in pregnancy and breastmilk exposure LEF may not be a human teratogen but there remains insufficient evidence to support use at the time of conception or during pregnancy (GRADE 1B, SOA 98%). Women on LEF considering pregnancy should stop and undergo a standard cholestyramine washout procedure, and switch to alternative medication compatible with pregnancy (GRADE 1B, SOA 98.8%). If unintended conception occurs on LEF, the drug should be stopped immediately and a standard cholestyramine washout procedure given, with early referral to a foetal medicine department considered (GRADE 1B, SOA 99%). LEF is not recommended while breastfeeding (GRADE 1C, SOA 99.5%). Azathioprine Recommendations were based on nine recent and 28 previous studies of AZA. Recommendations for azathioprine in pregnancy and breastmilk exposure AZA is compatible throughout pregnancy (GRADE 1B, SOA 100%). AZA is compatible with breastmilk exposure (GRADE 2C, SOA 99.5%). Ciclosporin Recommendations were based on five recent and 13 previous studies of ciclosporin (CsA). Recommendations for ciclosporin in pregnancy and breastmilk exposure CsA is compatible throughout pregnancy with monitoring of maternal blood pressure, renal function, blood glucose and drug levels (GRADE 1B, SOA 100%). CsA is compatible with breastmilk exposure (GRADE 2C, SOA 99.7%). Tacrolimus Recommendations were based on eight recent and seven previous studies of tacrolimus. Recommendations for tacrolimus in pregnancy and breastmilk exposure Tacrolimus is compatible throughout pregnancy with monitoring of maternal blood pressure, renal function, blood glucose and drug levels (GRADE 2B, SOA 100%). Tacrolimus is compatible with breastmilk exposure (GRADE 2C, SOA 99.8%). Cyclophosphamide Recommendations were based on one recent and five previous studies of CYC. Recommendations for cyclophosphamide in pregnancy and breastmilk exposure CYC is a known teratogen and gonadotoxic, and therefore should only be considered in pregnancy in cases of severe life/organ-threatening maternal disease when there is appreciable risk of maternal and foetal morbidity and mortality without this therapy (GRADE 1B, SOA 99.5%). CYC is not recommended while breastfeeding (GRADE 2C, SOA 100%). Mycophenolate mofetil Recommendations were based on eight recent and 16 previous studies of MMF. Recommendations for mycophenolate mofetil in pregnancy and breastmilk exposure MMF remains contraindicated during pregnancy, and should be avoided in women planning pregnancy or switched to a pregnancy-compatible alternative at least 6 weeks before attempting to conceive (GRADE 1B, SOA 100%). In cases of unintended conception, switch MMF to a pregnancy-compatible alternative and refer to local experts for further advice and risk assessment (GRADE 1B, SOA 100%). MMF is not recommended while breastfeeding (GRADE 2C, SOA 99.7%). Intravenous immunoglobulin Two recent and 16 previous studies of IVIG were identified. Recommendations for intravenous immunoglobulin in pregnancy and breastmilk exposure IVIG is compatible with pregnancy (GRADE 1B, SOA 99.5%). IVIG is compatible with breastmilk exposure (GRADE 2C, SOA 100%). Biologic DMARDs Anti-TNFα drugs Recommendations were based on 50 recent studies, two breastmilk transfer studies, and 29 previous studies of tumour necrosis factor-α inhibitors (TNFi). Recommendations for anti-TNFα medications in pregnancy and breastmilk exposure Women with no/low disease activity established on a TNFi with known placental transfer [infliximab (INF), adalimumab (ADA), golimumab (GOL)] do not need to be switched to an alternative TNFi with established minimal placental transfer [certolizumab pegol (CZP)] either before or during pregnancy (GRADE 1B, SOA 100%). CZP is compatible with all three trimesters of pregnancy, has no to minimal placental transfer compared with other TNFi, and does not require any alteration to the infant vaccination schedule (GRADE 1B, SOA 100%). Women considered to have low risk of disease flare on withdrawal of TNFi in pregnancy could stop INF at 20 weeks, ADA and GOL at 28 weeks, and ETA at 32 weeks so that a full-term infant can have a normal vaccination schedule, with rotavirus vaccination at 8 weeks as per the UK schedule (GRADE 1B, SOA 99.5%). INF, ADA, ETA or GOL may be continued throughout pregnancy to maintain maternal disease control; in these circumstances, live vaccines should be avoided in infants until they are 6 months of age (GRADE 1B, SOA 100%). If a TNFi is stopped in pregnancy, it can be restarted as soon as practical post-partum in the absence of infections or surgical complications, regardless of breastfeeding status, to ensure control of maternal disease (GRADE 1C, SOA 100%). TNFi are compatible with breastmilk exposure (GRADE 1C, SOA 100%). Rituximab Recommendations were based on five recent studies, a breastmilk transfer study, and eight previous studies of rituximab (RTX). Recommendations for rituximab in pregnancy and breastmilk exposure Limited evidence has not shown RTX to be teratogenic; however, there remains insufficient evidence to be confident that it is compatible with pregnancy. Consider stopping the drug at conception (GRADE 2C, SOA 99.3%). RTX may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 99.7%). If RTX is used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 98.7%). Based on limited evidence, maternal treatment with RTX is compatible with breastmilk exposure (GRADE 2C, SOA 99.5%). Interleukin-6 inhibitors We identified three recent studies and one previous study of tocilizumab (TCZ). No studies of sarilumab were identified. Recommendations for IL-6 inhibitors in pregnancy and breastmilk exposure Limited evidence has not shown IL-6 inhibitors (IL-6i) to be teratogenic; however, there remains insufficient evidence to be confident that they are compatible with pregnancy. Consider stopping the drug at conception. Any exposure, however, during pregnancy is unlikely to be harmful (GRADE 2C, SOA 99.7%). IL-6i may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 100%). If IL-6i are used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 99.3%). Based on limited evidence, maternal treatment with IL-6i is compatible with breastmilk exposure (GRADE 2C, SOA 100%). Interleukin-1 inhibitors We identified four recent and three previous studies of anakinra and one study of canakinumab. Recommendations for IL-1 inhibitors in pregnancy and breastmilk exposure Limited evidence has not shown IL-1 inhibitors (IL-1i) to be teratogenic; however, there remains insufficient evidence to be confident that they are compatible with pregnancy. Consider stopping the drug at conception. Any exposure, however, during pregnancy is unlikely to be harmful (GRADE 2C, SOA 99.8%). IL-1i may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 100%). If IL-1i are used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 99.3%). Based on limited evidence, maternal treatment with IL-1i is compatible with breastmilk exposure (GRADE 2C, SOA 100%). Abatacept We identified two recent and three previous studies of abatacept (ABA). Recommendations for abatacept in pregnancy and breastmilk exposure Limited evidence has not shown ABA to be teratogenic; however, there remains insufficient evidence to be confident that it is compatible with pregnancy. Consider stopping the drug at conception. Any exposure, however, during pregnancy is unlikely to be harmful (GRADE 2C, SOA 99.3%). ABA may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 99.3%). If ABA is used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 99.3%). Based on limited evidence, maternal treatment with ABA is compatible with breastmilk exposure (GRADE 2C, SOA 99.5%). Belimumab We identified one recent and two previous studies of belimumab (BEL). Recommendations for belimumab in pregnancy and breastmilk exposure Limited evidence has not shown BEL to be teratogenic; however, there remains insufficient evidence to be confident that it is compatible with pregnancy. Consider stopping the drug at conception. Any exposure, however, during pregnancy is unlikely to be harmful (GRADE 2C, SOA 99.3%). BEL may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 99.5%). If BEL is used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 98.8%). Based on limited evidence, maternal treatment with BEL is compatible with breastmilk exposure (GRADE 2C, SOA 99.5%). Interleukin-17 inhibitors We identified three studies of secukinumab and ixekizumab. Recommendations for interleukin-17 inhibitors in pregnancy and breastmilk exposure Limited evidence has not shown interleukin-17 inhibitors (IL-17i) to be teratogenic; however, there remains insufficient evidence to be confident that they are compatible with pregnancy. Consider stopping the drug at conception. Any exposure, however, during pregnancy is unlikely to be harmful (GRADE 2C, SOA 99.3%). IL-17i may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 99%). If IL-17i are used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 99.3%). Based on limited evidence, maternal treatment with IL-17i is compatible with breastmilk exposure (GRADE 2C, SOA 99.5%). Interleukin-12/23 inhibitors We identified three studies of ustekinumab (UST). Recommendations for interleukin-12/23 inhibitors in pregnancy and breastmilk exposure Limited evidence has not shown UST to be teratogenic; however, there remains insufficient evidence to be confident that it is compatible with pregnancy. Consider stopping the drug at conception. Any exposure, however, during pregnancy is unlikely to be harmful (GRADE 2C, SOA 99.3%). UST may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 98.8%). If UST is used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 99.3%). Based on limited evidence, maternal treatment with UST is compatible with breastmilk exposure (GRADE 2C, SOA 99.5%). Anifrolumab Anifrolumab was not NICE approved at the time of our literature search, and so was not included in the search. Targeted synthetic DMARDs Apremilast No studies relating to apremilast use in pregnancy were found. JAK inhibitors We identified three studies of tofacitinib (TOF) and no studies of baricitinib or upadacitinib. Filgotinib was NICE approved after our search, and so was not included. Recommendations for JAK inhibitors in pregnancy and breastmilk exposure There are insufficient data to make a recommendation on Janus kinase inhibitor (JAKi) use during pregnancy and they should be stopped at least two weeks before planned conception (GRADE 2C, SOA 99.5%). There are insufficient data to recommend JAKi in breastfeeding and, given they are likely to transfer into breastmilk, they should be avoided (GRADE 2C, SOA 99.5%). Paternal exposure Recommendations for paternal exposure to immunomodulatory drugs Due to the adverse effect of CYC on male fertility, semen cryopreservation is recommended for men prior to paternal exposure (GRADE 1C, SOA 99.5%). Men who take SSZ may have reduced fertility. There is little evidence to suggest that SSZ should be stopped pre-conception, unless conception is delayed by more than 12 months when stopping SSZ should be considered along with other causes of infertility (GRADE 1C, SOA 99.0%). Paternal exposure to the following anti-rheumatic medication is compatible with pregnancy: prednisolone, low-dose (≤25 mg/week) methotrexate, azathioprine (GRADE 1B); TNFi, cyclosporin (GRADE 1C); hydroxychloroquine, leflunomide, tacrolimus, mycophenolate mofetil, intravenous immunoglobulin, rituximab, interleukin-6 inhibitors, interleukin-1 inhibitors, abatacept, belimumab, interleukin-17 inhibitors, ustekinumab and JAKi (GRADE 2C, SOA 99.3%).
Pre-conception counselling should be addressed by all healthcare professionals, with referral to professionals with relevant expertise as appropriate, to optimize disease control before pregnancy; with advice on the timing of pregnancy, and drug therapy before, during and after pregnancy, including contraception (GRADE 1A, SOA 99.5%). If a woman is planning pregnancy, avoid pregnancy-incompatible drugs (GRADE 1A, SOA 100%). The risks and benefits to the mother and foetus of drug treatment to control maternal disease should be discussed and clearly documented by all healthcare professionals involved in the patient’s care (GRADE 1A, SOA 99.5%). Immunomodulatory drugs that are contraindicated in pregnancy should be switched to a pregnancy-compatible alternative in advance of conception to ensure maintenance of disease control on the new medication (GRADE 1A, SOA 100%). When no pregnancy-compatible drugs are suitable, control of severe/life-threatening maternal disease should take priority over concerns for potential foetal outcomes (GRADE 1B, SOA 99.0%). All biologic DMARDs may be continued throughout pregnancy if required to control active/severe maternal disease (GRADE 1B, SOA 98.5%). Immunisation schedules in infants after in-utero exposure to biologic DMARDs will depend on timing of exposure, bioavailability and persistence of the drug, mechanism of action of the drug, and live vaccines (GRADE 1C, SOA 99.5%). Where possible, the minimum effective dose of immunomodulatory drug or corticosteroid should be used to maintain maternal disease suppression, and stopping the drug during pregnancy may be considered in women at low risk of disease flare on withdrawal of therapy (GRADE 1B, SOA 100%). Some drugs may reduce male fertility, but paternal drug exposure in humans has not convincingly been associated with adverse foetal development or pregnancy outcome. Although the evidence is weak, men who take rheumatological medicines should be reassured about the safety of conceiving (GRADE 2C, SOA 98.4%).
Recommendations were based on 23 studies of HCQ identified in the current literature search and 23 studies identified in the previous guideline’s literature search, in addition to four studies of chloroquine and no studies of mepacrine. Three recent and one previous study of breastmilk exposure to HCQ were identified. Recommendations for hydroxychloroquine in pregnancy and breastmilk exposure HCQ remains the antimalarial of choice in women planning a pregnancy with rheumatic disease in need of treatment, and should be continued during pregnancy at dose of ≤400 mg/day (GRADE 1B, SOA 100%). HCQ is compatible with breastmilk exposure (GRADE 1B, SOA 99.5%).
HCQ remains the antimalarial of choice in women planning a pregnancy with rheumatic disease in need of treatment, and should be continued during pregnancy at dose of ≤400 mg/day (GRADE 1B, SOA 100%). HCQ is compatible with breastmilk exposure (GRADE 1B, SOA 99.5%).
Recommendations were based on 11 recent studies of prednisolone, one study of methylprednisolone, and previous studies: 47 on prednisolone; 31 on dexamethasone; 27 on betamethasone; and 10 on general corticosteroid use. Recommendations for corticosteroids in pregnancy and breastmilk exposure Prednisolone is compatible with pregnancy and is the preferred corticosteroid in the treatment of maternal rheumatological disease in pregnancy and requires shared care with obstetric teams to monitor maternal blood pressure and blood glucose (GRADE 1B, SOA 100%). Where possible, the dose of prednisolone should be <20 mg/day and tapered to the minimum effective dose to control maternal disease, in conjunction with steroid-sparing drugs compatible with pregnancy (GRADE 1C, SOA 99.5%). Prednisolone is compatible with breastmilk exposure (GRADE 1B, SOA 100%). Methylprednisolone has similar rates of placental transfer to prednisolone and would therefore be expected to be compatible with pregnancy and breastmilk exposure (GRADE 2C, SOA 99%). Conventional synthetic DMARDs Methotrexate Recommendations were based on 12 recent and 10 previous studies of MTX. Recommendations for methotrexate in pregnancy and breastmilk exposure MTX at any dose should be avoided in pregnancy and stopped at least one month in advance of planned conception, when it should be switched to another pregnancy-compatible drug to ensure maintenance of maternal disease suppression (GRADE 1A, SOA 98%). In women treated with low-dose (≤25 mg/week) MTX within one month prior to conception, folic acid supplementation (5 mg/day) should be continued up to 12 weeks of pregnancy (GRADE 1B, SOA 99.5%). In unintended pregnancy on low-dose (≤25 mg/week) MTX, there is minimal risk to the foetus; the drug should be stopped immediately, folic acid supplementation (5 mg/day) continued, and a careful evaluation of foetal risk with early referral to a foetal medicine department considered (GRADE 1C, SOA 100%). Although only minute amounts of MTX are excreted into breastmilk, MTX cannot be recommended in breastfeeding because of theoretical risks and insufficient data on outcomes (GRADE 2C, SOA 99%). Sulfasalazine Recommendations were based on six studies of SSZ identified in the previous guideline. Recommendations for sulfasalazine in pregnancy and breastmilk exposure SSZ is compatible throughout pregnancy, with folic acid 5 mg/day recommended in the periconception period and during the first trimester (GRADE 1B, SOA 100%). SSZ is compatible with breastmilk exposure in healthy, full-term infants (GRADE 1C, SOA 99.5%). Leflunomide Recommendations were based on three recent and seven previous studies of LEF. Recommendations for leflunomide in pregnancy and breastmilk exposure LEF may not be a human teratogen but there remains insufficient evidence to support use at the time of conception or during pregnancy (GRADE 1B, SOA 98%). Women on LEF considering pregnancy should stop and undergo a standard cholestyramine washout procedure, and switch to alternative medication compatible with pregnancy (GRADE 1B, SOA 98.8%). If unintended conception occurs on LEF, the drug should be stopped immediately and a standard cholestyramine washout procedure given, with early referral to a foetal medicine department considered (GRADE 1B, SOA 99%). LEF is not recommended while breastfeeding (GRADE 1C, SOA 99.5%). Azathioprine Recommendations were based on nine recent and 28 previous studies of AZA. Recommendations for azathioprine in pregnancy and breastmilk exposure AZA is compatible throughout pregnancy (GRADE 1B, SOA 100%). AZA is compatible with breastmilk exposure (GRADE 2C, SOA 99.5%). Ciclosporin Recommendations were based on five recent and 13 previous studies of ciclosporin (CsA). Recommendations for ciclosporin in pregnancy and breastmilk exposure CsA is compatible throughout pregnancy with monitoring of maternal blood pressure, renal function, blood glucose and drug levels (GRADE 1B, SOA 100%). CsA is compatible with breastmilk exposure (GRADE 2C, SOA 99.7%). Tacrolimus Recommendations were based on eight recent and seven previous studies of tacrolimus. Recommendations for tacrolimus in pregnancy and breastmilk exposure Tacrolimus is compatible throughout pregnancy with monitoring of maternal blood pressure, renal function, blood glucose and drug levels (GRADE 2B, SOA 100%). Tacrolimus is compatible with breastmilk exposure (GRADE 2C, SOA 99.8%). Cyclophosphamide Recommendations were based on one recent and five previous studies of CYC. Recommendations for cyclophosphamide in pregnancy and breastmilk exposure CYC is a known teratogen and gonadotoxic, and therefore should only be considered in pregnancy in cases of severe life/organ-threatening maternal disease when there is appreciable risk of maternal and foetal morbidity and mortality without this therapy (GRADE 1B, SOA 99.5%). CYC is not recommended while breastfeeding (GRADE 2C, SOA 100%). Mycophenolate mofetil Recommendations were based on eight recent and 16 previous studies of MMF. Recommendations for mycophenolate mofetil in pregnancy and breastmilk exposure MMF remains contraindicated during pregnancy, and should be avoided in women planning pregnancy or switched to a pregnancy-compatible alternative at least 6 weeks before attempting to conceive (GRADE 1B, SOA 100%). In cases of unintended conception, switch MMF to a pregnancy-compatible alternative and refer to local experts for further advice and risk assessment (GRADE 1B, SOA 100%). MMF is not recommended while breastfeeding (GRADE 2C, SOA 99.7%). Intravenous immunoglobulin Two recent and 16 previous studies of IVIG were identified. Recommendations for intravenous immunoglobulin in pregnancy and breastmilk exposure IVIG is compatible with pregnancy (GRADE 1B, SOA 99.5%). IVIG is compatible with breastmilk exposure (GRADE 2C, SOA 100%).
Prednisolone is compatible with pregnancy and is the preferred corticosteroid in the treatment of maternal rheumatological disease in pregnancy and requires shared care with obstetric teams to monitor maternal blood pressure and blood glucose (GRADE 1B, SOA 100%). Where possible, the dose of prednisolone should be <20 mg/day and tapered to the minimum effective dose to control maternal disease, in conjunction with steroid-sparing drugs compatible with pregnancy (GRADE 1C, SOA 99.5%). Prednisolone is compatible with breastmilk exposure (GRADE 1B, SOA 100%). Methylprednisolone has similar rates of placental transfer to prednisolone and would therefore be expected to be compatible with pregnancy and breastmilk exposure (GRADE 2C, SOA 99%).
Methotrexate Recommendations were based on 12 recent and 10 previous studies of MTX. Recommendations for methotrexate in pregnancy and breastmilk exposure MTX at any dose should be avoided in pregnancy and stopped at least one month in advance of planned conception, when it should be switched to another pregnancy-compatible drug to ensure maintenance of maternal disease suppression (GRADE 1A, SOA 98%). In women treated with low-dose (≤25 mg/week) MTX within one month prior to conception, folic acid supplementation (5 mg/day) should be continued up to 12 weeks of pregnancy (GRADE 1B, SOA 99.5%). In unintended pregnancy on low-dose (≤25 mg/week) MTX, there is minimal risk to the foetus; the drug should be stopped immediately, folic acid supplementation (5 mg/day) continued, and a careful evaluation of foetal risk with early referral to a foetal medicine department considered (GRADE 1C, SOA 100%). Although only minute amounts of MTX are excreted into breastmilk, MTX cannot be recommended in breastfeeding because of theoretical risks and insufficient data on outcomes (GRADE 2C, SOA 99%).
Recommendations were based on 12 recent and 10 previous studies of MTX. Recommendations for methotrexate in pregnancy and breastmilk exposure MTX at any dose should be avoided in pregnancy and stopped at least one month in advance of planned conception, when it should be switched to another pregnancy-compatible drug to ensure maintenance of maternal disease suppression (GRADE 1A, SOA 98%). In women treated with low-dose (≤25 mg/week) MTX within one month prior to conception, folic acid supplementation (5 mg/day) should be continued up to 12 weeks of pregnancy (GRADE 1B, SOA 99.5%). In unintended pregnancy on low-dose (≤25 mg/week) MTX, there is minimal risk to the foetus; the drug should be stopped immediately, folic acid supplementation (5 mg/day) continued, and a careful evaluation of foetal risk with early referral to a foetal medicine department considered (GRADE 1C, SOA 100%). Although only minute amounts of MTX are excreted into breastmilk, MTX cannot be recommended in breastfeeding because of theoretical risks and insufficient data on outcomes (GRADE 2C, SOA 99%).
MTX at any dose should be avoided in pregnancy and stopped at least one month in advance of planned conception, when it should be switched to another pregnancy-compatible drug to ensure maintenance of maternal disease suppression (GRADE 1A, SOA 98%). In women treated with low-dose (≤25 mg/week) MTX within one month prior to conception, folic acid supplementation (5 mg/day) should be continued up to 12 weeks of pregnancy (GRADE 1B, SOA 99.5%). In unintended pregnancy on low-dose (≤25 mg/week) MTX, there is minimal risk to the foetus; the drug should be stopped immediately, folic acid supplementation (5 mg/day) continued, and a careful evaluation of foetal risk with early referral to a foetal medicine department considered (GRADE 1C, SOA 100%). Although only minute amounts of MTX are excreted into breastmilk, MTX cannot be recommended in breastfeeding because of theoretical risks and insufficient data on outcomes (GRADE 2C, SOA 99%).
Recommendations were based on six studies of SSZ identified in the previous guideline. Recommendations for sulfasalazine in pregnancy and breastmilk exposure SSZ is compatible throughout pregnancy, with folic acid 5 mg/day recommended in the periconception period and during the first trimester (GRADE 1B, SOA 100%). SSZ is compatible with breastmilk exposure in healthy, full-term infants (GRADE 1C, SOA 99.5%).
SSZ is compatible throughout pregnancy, with folic acid 5 mg/day recommended in the periconception period and during the first trimester (GRADE 1B, SOA 100%). SSZ is compatible with breastmilk exposure in healthy, full-term infants (GRADE 1C, SOA 99.5%).
Recommendations were based on three recent and seven previous studies of LEF. Recommendations for leflunomide in pregnancy and breastmilk exposure LEF may not be a human teratogen but there remains insufficient evidence to support use at the time of conception or during pregnancy (GRADE 1B, SOA 98%). Women on LEF considering pregnancy should stop and undergo a standard cholestyramine washout procedure, and switch to alternative medication compatible with pregnancy (GRADE 1B, SOA 98.8%). If unintended conception occurs on LEF, the drug should be stopped immediately and a standard cholestyramine washout procedure given, with early referral to a foetal medicine department considered (GRADE 1B, SOA 99%). LEF is not recommended while breastfeeding (GRADE 1C, SOA 99.5%).
LEF may not be a human teratogen but there remains insufficient evidence to support use at the time of conception or during pregnancy (GRADE 1B, SOA 98%). Women on LEF considering pregnancy should stop and undergo a standard cholestyramine washout procedure, and switch to alternative medication compatible with pregnancy (GRADE 1B, SOA 98.8%). If unintended conception occurs on LEF, the drug should be stopped immediately and a standard cholestyramine washout procedure given, with early referral to a foetal medicine department considered (GRADE 1B, SOA 99%). LEF is not recommended while breastfeeding (GRADE 1C, SOA 99.5%).
Recommendations were based on nine recent and 28 previous studies of AZA. Recommendations for azathioprine in pregnancy and breastmilk exposure AZA is compatible throughout pregnancy (GRADE 1B, SOA 100%). AZA is compatible with breastmilk exposure (GRADE 2C, SOA 99.5%).
AZA is compatible throughout pregnancy (GRADE 1B, SOA 100%). AZA is compatible with breastmilk exposure (GRADE 2C, SOA 99.5%).
Recommendations were based on five recent and 13 previous studies of ciclosporin (CsA). Recommendations for ciclosporin in pregnancy and breastmilk exposure CsA is compatible throughout pregnancy with monitoring of maternal blood pressure, renal function, blood glucose and drug levels (GRADE 1B, SOA 100%). CsA is compatible with breastmilk exposure (GRADE 2C, SOA 99.7%).
CsA is compatible throughout pregnancy with monitoring of maternal blood pressure, renal function, blood glucose and drug levels (GRADE 1B, SOA 100%). CsA is compatible with breastmilk exposure (GRADE 2C, SOA 99.7%).
Recommendations were based on eight recent and seven previous studies of tacrolimus. Recommendations for tacrolimus in pregnancy and breastmilk exposure Tacrolimus is compatible throughout pregnancy with monitoring of maternal blood pressure, renal function, blood glucose and drug levels (GRADE 2B, SOA 100%). Tacrolimus is compatible with breastmilk exposure (GRADE 2C, SOA 99.8%).
Tacrolimus is compatible throughout pregnancy with monitoring of maternal blood pressure, renal function, blood glucose and drug levels (GRADE 2B, SOA 100%). Tacrolimus is compatible with breastmilk exposure (GRADE 2C, SOA 99.8%).
Recommendations were based on one recent and five previous studies of CYC. Recommendations for cyclophosphamide in pregnancy and breastmilk exposure CYC is a known teratogen and gonadotoxic, and therefore should only be considered in pregnancy in cases of severe life/organ-threatening maternal disease when there is appreciable risk of maternal and foetal morbidity and mortality without this therapy (GRADE 1B, SOA 99.5%). CYC is not recommended while breastfeeding (GRADE 2C, SOA 100%).
CYC is a known teratogen and gonadotoxic, and therefore should only be considered in pregnancy in cases of severe life/organ-threatening maternal disease when there is appreciable risk of maternal and foetal morbidity and mortality without this therapy (GRADE 1B, SOA 99.5%). CYC is not recommended while breastfeeding (GRADE 2C, SOA 100%).
Recommendations were based on eight recent and 16 previous studies of MMF. Recommendations for mycophenolate mofetil in pregnancy and breastmilk exposure MMF remains contraindicated during pregnancy, and should be avoided in women planning pregnancy or switched to a pregnancy-compatible alternative at least 6 weeks before attempting to conceive (GRADE 1B, SOA 100%). In cases of unintended conception, switch MMF to a pregnancy-compatible alternative and refer to local experts for further advice and risk assessment (GRADE 1B, SOA 100%). MMF is not recommended while breastfeeding (GRADE 2C, SOA 99.7%).
MMF remains contraindicated during pregnancy, and should be avoided in women planning pregnancy or switched to a pregnancy-compatible alternative at least 6 weeks before attempting to conceive (GRADE 1B, SOA 100%). In cases of unintended conception, switch MMF to a pregnancy-compatible alternative and refer to local experts for further advice and risk assessment (GRADE 1B, SOA 100%). MMF is not recommended while breastfeeding (GRADE 2C, SOA 99.7%).
Two recent and 16 previous studies of IVIG were identified. Recommendations for intravenous immunoglobulin in pregnancy and breastmilk exposure IVIG is compatible with pregnancy (GRADE 1B, SOA 99.5%). IVIG is compatible with breastmilk exposure (GRADE 2C, SOA 100%).
IVIG is compatible with pregnancy (GRADE 1B, SOA 99.5%). IVIG is compatible with breastmilk exposure (GRADE 2C, SOA 100%).
Anti-TNFα drugs Recommendations were based on 50 recent studies, two breastmilk transfer studies, and 29 previous studies of tumour necrosis factor-α inhibitors (TNFi). Recommendations for anti-TNFα medications in pregnancy and breastmilk exposure Women with no/low disease activity established on a TNFi with known placental transfer [infliximab (INF), adalimumab (ADA), golimumab (GOL)] do not need to be switched to an alternative TNFi with established minimal placental transfer [certolizumab pegol (CZP)] either before or during pregnancy (GRADE 1B, SOA 100%). CZP is compatible with all three trimesters of pregnancy, has no to minimal placental transfer compared with other TNFi, and does not require any alteration to the infant vaccination schedule (GRADE 1B, SOA 100%). Women considered to have low risk of disease flare on withdrawal of TNFi in pregnancy could stop INF at 20 weeks, ADA and GOL at 28 weeks, and ETA at 32 weeks so that a full-term infant can have a normal vaccination schedule, with rotavirus vaccination at 8 weeks as per the UK schedule (GRADE 1B, SOA 99.5%). INF, ADA, ETA or GOL may be continued throughout pregnancy to maintain maternal disease control; in these circumstances, live vaccines should be avoided in infants until they are 6 months of age (GRADE 1B, SOA 100%). If a TNFi is stopped in pregnancy, it can be restarted as soon as practical post-partum in the absence of infections or surgical complications, regardless of breastfeeding status, to ensure control of maternal disease (GRADE 1C, SOA 100%). TNFi are compatible with breastmilk exposure (GRADE 1C, SOA 100%). Rituximab Recommendations were based on five recent studies, a breastmilk transfer study, and eight previous studies of rituximab (RTX). Recommendations for rituximab in pregnancy and breastmilk exposure Limited evidence has not shown RTX to be teratogenic; however, there remains insufficient evidence to be confident that it is compatible with pregnancy. Consider stopping the drug at conception (GRADE 2C, SOA 99.3%). RTX may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 99.7%). If RTX is used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 98.7%). Based on limited evidence, maternal treatment with RTX is compatible with breastmilk exposure (GRADE 2C, SOA 99.5%). Interleukin-6 inhibitors We identified three recent studies and one previous study of tocilizumab (TCZ). No studies of sarilumab were identified. Recommendations for IL-6 inhibitors in pregnancy and breastmilk exposure Limited evidence has not shown IL-6 inhibitors (IL-6i) to be teratogenic; however, there remains insufficient evidence to be confident that they are compatible with pregnancy. Consider stopping the drug at conception. Any exposure, however, during pregnancy is unlikely to be harmful (GRADE 2C, SOA 99.7%). IL-6i may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 100%). If IL-6i are used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 99.3%). Based on limited evidence, maternal treatment with IL-6i is compatible with breastmilk exposure (GRADE 2C, SOA 100%). Interleukin-1 inhibitors We identified four recent and three previous studies of anakinra and one study of canakinumab. Recommendations for IL-1 inhibitors in pregnancy and breastmilk exposure Limited evidence has not shown IL-1 inhibitors (IL-1i) to be teratogenic; however, there remains insufficient evidence to be confident that they are compatible with pregnancy. Consider stopping the drug at conception. Any exposure, however, during pregnancy is unlikely to be harmful (GRADE 2C, SOA 99.8%). IL-1i may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 100%). If IL-1i are used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 99.3%). Based on limited evidence, maternal treatment with IL-1i is compatible with breastmilk exposure (GRADE 2C, SOA 100%). Abatacept We identified two recent and three previous studies of abatacept (ABA). Recommendations for abatacept in pregnancy and breastmilk exposure Limited evidence has not shown ABA to be teratogenic; however, there remains insufficient evidence to be confident that it is compatible with pregnancy. Consider stopping the drug at conception. Any exposure, however, during pregnancy is unlikely to be harmful (GRADE 2C, SOA 99.3%). ABA may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 99.3%). If ABA is used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 99.3%). Based on limited evidence, maternal treatment with ABA is compatible with breastmilk exposure (GRADE 2C, SOA 99.5%). Belimumab We identified one recent and two previous studies of belimumab (BEL). Recommendations for belimumab in pregnancy and breastmilk exposure Limited evidence has not shown BEL to be teratogenic; however, there remains insufficient evidence to be confident that it is compatible with pregnancy. Consider stopping the drug at conception. Any exposure, however, during pregnancy is unlikely to be harmful (GRADE 2C, SOA 99.3%). BEL may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 99.5%). If BEL is used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 98.8%). Based on limited evidence, maternal treatment with BEL is compatible with breastmilk exposure (GRADE 2C, SOA 99.5%). Interleukin-17 inhibitors We identified three studies of secukinumab and ixekizumab. Recommendations for interleukin-17 inhibitors in pregnancy and breastmilk exposure Limited evidence has not shown interleukin-17 inhibitors (IL-17i) to be teratogenic; however, there remains insufficient evidence to be confident that they are compatible with pregnancy. Consider stopping the drug at conception. Any exposure, however, during pregnancy is unlikely to be harmful (GRADE 2C, SOA 99.3%). IL-17i may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 99%). If IL-17i are used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 99.3%). Based on limited evidence, maternal treatment with IL-17i is compatible with breastmilk exposure (GRADE 2C, SOA 99.5%). Interleukin-12/23 inhibitors We identified three studies of ustekinumab (UST). Recommendations for interleukin-12/23 inhibitors in pregnancy and breastmilk exposure Limited evidence has not shown UST to be teratogenic; however, there remains insufficient evidence to be confident that it is compatible with pregnancy. Consider stopping the drug at conception. Any exposure, however, during pregnancy is unlikely to be harmful (GRADE 2C, SOA 99.3%). UST may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 98.8%). If UST is used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 99.3%). Based on limited evidence, maternal treatment with UST is compatible with breastmilk exposure (GRADE 2C, SOA 99.5%). Anifrolumab Anifrolumab was not NICE approved at the time of our literature search, and so was not included in the search.
Recommendations were based on 50 recent studies, two breastmilk transfer studies, and 29 previous studies of tumour necrosis factor-α inhibitors (TNFi). Recommendations for anti-TNFα medications in pregnancy and breastmilk exposure Women with no/low disease activity established on a TNFi with known placental transfer [infliximab (INF), adalimumab (ADA), golimumab (GOL)] do not need to be switched to an alternative TNFi with established minimal placental transfer [certolizumab pegol (CZP)] either before or during pregnancy (GRADE 1B, SOA 100%). CZP is compatible with all three trimesters of pregnancy, has no to minimal placental transfer compared with other TNFi, and does not require any alteration to the infant vaccination schedule (GRADE 1B, SOA 100%). Women considered to have low risk of disease flare on withdrawal of TNFi in pregnancy could stop INF at 20 weeks, ADA and GOL at 28 weeks, and ETA at 32 weeks so that a full-term infant can have a normal vaccination schedule, with rotavirus vaccination at 8 weeks as per the UK schedule (GRADE 1B, SOA 99.5%). INF, ADA, ETA or GOL may be continued throughout pregnancy to maintain maternal disease control; in these circumstances, live vaccines should be avoided in infants until they are 6 months of age (GRADE 1B, SOA 100%). If a TNFi is stopped in pregnancy, it can be restarted as soon as practical post-partum in the absence of infections or surgical complications, regardless of breastfeeding status, to ensure control of maternal disease (GRADE 1C, SOA 100%). TNFi are compatible with breastmilk exposure (GRADE 1C, SOA 100%).
Women with no/low disease activity established on a TNFi with known placental transfer [infliximab (INF), adalimumab (ADA), golimumab (GOL)] do not need to be switched to an alternative TNFi with established minimal placental transfer [certolizumab pegol (CZP)] either before or during pregnancy (GRADE 1B, SOA 100%). CZP is compatible with all three trimesters of pregnancy, has no to minimal placental transfer compared with other TNFi, and does not require any alteration to the infant vaccination schedule (GRADE 1B, SOA 100%). Women considered to have low risk of disease flare on withdrawal of TNFi in pregnancy could stop INF at 20 weeks, ADA and GOL at 28 weeks, and ETA at 32 weeks so that a full-term infant can have a normal vaccination schedule, with rotavirus vaccination at 8 weeks as per the UK schedule (GRADE 1B, SOA 99.5%). INF, ADA, ETA or GOL may be continued throughout pregnancy to maintain maternal disease control; in these circumstances, live vaccines should be avoided in infants until they are 6 months of age (GRADE 1B, SOA 100%). If a TNFi is stopped in pregnancy, it can be restarted as soon as practical post-partum in the absence of infections or surgical complications, regardless of breastfeeding status, to ensure control of maternal disease (GRADE 1C, SOA 100%). TNFi are compatible with breastmilk exposure (GRADE 1C, SOA 100%).
Recommendations were based on five recent studies, a breastmilk transfer study, and eight previous studies of rituximab (RTX). Recommendations for rituximab in pregnancy and breastmilk exposure Limited evidence has not shown RTX to be teratogenic; however, there remains insufficient evidence to be confident that it is compatible with pregnancy. Consider stopping the drug at conception (GRADE 2C, SOA 99.3%). RTX may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 99.7%). If RTX is used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 98.7%). Based on limited evidence, maternal treatment with RTX is compatible with breastmilk exposure (GRADE 2C, SOA 99.5%).
Limited evidence has not shown RTX to be teratogenic; however, there remains insufficient evidence to be confident that it is compatible with pregnancy. Consider stopping the drug at conception (GRADE 2C, SOA 99.3%). RTX may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 99.7%). If RTX is used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 98.7%). Based on limited evidence, maternal treatment with RTX is compatible with breastmilk exposure (GRADE 2C, SOA 99.5%).
We identified three recent studies and one previous study of tocilizumab (TCZ). No studies of sarilumab were identified.
Limited evidence has not shown IL-6 inhibitors (IL-6i) to be teratogenic; however, there remains insufficient evidence to be confident that they are compatible with pregnancy. Consider stopping the drug at conception. Any exposure, however, during pregnancy is unlikely to be harmful (GRADE 2C, SOA 99.7%). IL-6i may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 100%). If IL-6i are used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 99.3%). Based on limited evidence, maternal treatment with IL-6i is compatible with breastmilk exposure (GRADE 2C, SOA 100%).
Limited evidence has not shown IL-6 inhibitors (IL-6i) to be teratogenic; however, there remains insufficient evidence to be confident that they are compatible with pregnancy. Consider stopping the drug at conception. Any exposure, however, during pregnancy is unlikely to be harmful (GRADE 2C, SOA 99.7%). IL-6i may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 100%). If IL-6i are used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 99.3%). Based on limited evidence, maternal treatment with IL-6i is compatible with breastmilk exposure (GRADE 2C, SOA 100%).
We identified four recent and three previous studies of anakinra and one study of canakinumab. Recommendations for IL-1 inhibitors in pregnancy and breastmilk exposure Limited evidence has not shown IL-1 inhibitors (IL-1i) to be teratogenic; however, there remains insufficient evidence to be confident that they are compatible with pregnancy. Consider stopping the drug at conception. Any exposure, however, during pregnancy is unlikely to be harmful (GRADE 2C, SOA 99.8%). IL-1i may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 100%). If IL-1i are used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 99.3%). Based on limited evidence, maternal treatment with IL-1i is compatible with breastmilk exposure (GRADE 2C, SOA 100%).
Limited evidence has not shown IL-1 inhibitors (IL-1i) to be teratogenic; however, there remains insufficient evidence to be confident that they are compatible with pregnancy. Consider stopping the drug at conception. Any exposure, however, during pregnancy is unlikely to be harmful (GRADE 2C, SOA 99.8%). IL-1i may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 100%). If IL-1i are used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 99.3%). Based on limited evidence, maternal treatment with IL-1i is compatible with breastmilk exposure (GRADE 2C, SOA 100%).
Limited evidence has not shown IL-1 inhibitors (IL-1i) to be teratogenic; however, there remains insufficient evidence to be confident that they are compatible with pregnancy. Consider stopping the drug at conception. Any exposure, however, during pregnancy is unlikely to be harmful (GRADE 2C, SOA 99.8%). IL-1i may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 100%). If IL-1i are used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 99.3%). Based on limited evidence, maternal treatment with IL-1i is compatible with breastmilk exposure (GRADE 2C, SOA 100%).
We identified two recent and three previous studies of abatacept (ABA). Recommendations for abatacept in pregnancy and breastmilk exposure Limited evidence has not shown ABA to be teratogenic; however, there remains insufficient evidence to be confident that it is compatible with pregnancy. Consider stopping the drug at conception. Any exposure, however, during pregnancy is unlikely to be harmful (GRADE 2C, SOA 99.3%). ABA may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 99.3%). If ABA is used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 99.3%). Based on limited evidence, maternal treatment with ABA is compatible with breastmilk exposure (GRADE 2C, SOA 99.5%).
Limited evidence has not shown ABA to be teratogenic; however, there remains insufficient evidence to be confident that it is compatible with pregnancy. Consider stopping the drug at conception. Any exposure, however, during pregnancy is unlikely to be harmful (GRADE 2C, SOA 99.3%). ABA may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 99.3%). If ABA is used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 99.3%). Based on limited evidence, maternal treatment with ABA is compatible with breastmilk exposure (GRADE 2C, SOA 99.5%).
We identified one recent and two previous studies of belimumab (BEL). Recommendations for belimumab in pregnancy and breastmilk exposure Limited evidence has not shown BEL to be teratogenic; however, there remains insufficient evidence to be confident that it is compatible with pregnancy. Consider stopping the drug at conception. Any exposure, however, during pregnancy is unlikely to be harmful (GRADE 2C, SOA 99.3%). BEL may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 99.5%). If BEL is used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 98.8%). Based on limited evidence, maternal treatment with BEL is compatible with breastmilk exposure (GRADE 2C, SOA 99.5%).
Limited evidence has not shown BEL to be teratogenic; however, there remains insufficient evidence to be confident that it is compatible with pregnancy. Consider stopping the drug at conception. Any exposure, however, during pregnancy is unlikely to be harmful (GRADE 2C, SOA 99.3%). BEL may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 99.5%). If BEL is used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 98.8%). Based on limited evidence, maternal treatment with BEL is compatible with breastmilk exposure (GRADE 2C, SOA 99.5%).
We identified three studies of secukinumab and ixekizumab. Recommendations for interleukin-17 inhibitors in pregnancy and breastmilk exposure Limited evidence has not shown interleukin-17 inhibitors (IL-17i) to be teratogenic; however, there remains insufficient evidence to be confident that they are compatible with pregnancy. Consider stopping the drug at conception. Any exposure, however, during pregnancy is unlikely to be harmful (GRADE 2C, SOA 99.3%). IL-17i may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 99%). If IL-17i are used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 99.3%). Based on limited evidence, maternal treatment with IL-17i is compatible with breastmilk exposure (GRADE 2C, SOA 99.5%).
Limited evidence has not shown interleukin-17 inhibitors (IL-17i) to be teratogenic; however, there remains insufficient evidence to be confident that they are compatible with pregnancy. Consider stopping the drug at conception. Any exposure, however, during pregnancy is unlikely to be harmful (GRADE 2C, SOA 99.3%). IL-17i may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 99%). If IL-17i are used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 99.3%). Based on limited evidence, maternal treatment with IL-17i is compatible with breastmilk exposure (GRADE 2C, SOA 99.5%).
We identified three studies of ustekinumab (UST). Recommendations for interleukin-12/23 inhibitors in pregnancy and breastmilk exposure Limited evidence has not shown UST to be teratogenic; however, there remains insufficient evidence to be confident that it is compatible with pregnancy. Consider stopping the drug at conception. Any exposure, however, during pregnancy is unlikely to be harmful (GRADE 2C, SOA 99.3%). UST may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 98.8%). If UST is used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 99.3%). Based on limited evidence, maternal treatment with UST is compatible with breastmilk exposure (GRADE 2C, SOA 99.5%).
Limited evidence has not shown UST to be teratogenic; however, there remains insufficient evidence to be confident that it is compatible with pregnancy. Consider stopping the drug at conception. Any exposure, however, during pregnancy is unlikely to be harmful (GRADE 2C, SOA 99.3%). UST may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 98.8%). If UST is used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 99.3%). Based on limited evidence, maternal treatment with UST is compatible with breastmilk exposure (GRADE 2C, SOA 99.5%).
Anifrolumab was not NICE approved at the time of our literature search, and so was not included in the search.
Apremilast No studies relating to apremilast use in pregnancy were found. JAK inhibitors We identified three studies of tofacitinib (TOF) and no studies of baricitinib or upadacitinib. Filgotinib was NICE approved after our search, and so was not included. Recommendations for JAK inhibitors in pregnancy and breastmilk exposure There are insufficient data to make a recommendation on Janus kinase inhibitor (JAKi) use during pregnancy and they should be stopped at least two weeks before planned conception (GRADE 2C, SOA 99.5%). There are insufficient data to recommend JAKi in breastfeeding and, given they are likely to transfer into breastmilk, they should be avoided (GRADE 2C, SOA 99.5%).
No studies relating to apremilast use in pregnancy were found.
We identified three studies of tofacitinib (TOF) and no studies of baricitinib or upadacitinib. Filgotinib was NICE approved after our search, and so was not included. Recommendations for JAK inhibitors in pregnancy and breastmilk exposure There are insufficient data to make a recommendation on Janus kinase inhibitor (JAKi) use during pregnancy and they should be stopped at least two weeks before planned conception (GRADE 2C, SOA 99.5%). There are insufficient data to recommend JAKi in breastfeeding and, given they are likely to transfer into breastmilk, they should be avoided (GRADE 2C, SOA 99.5%).
There are insufficient data to make a recommendation on Janus kinase inhibitor (JAKi) use during pregnancy and they should be stopped at least two weeks before planned conception (GRADE 2C, SOA 99.5%). There are insufficient data to recommend JAKi in breastfeeding and, given they are likely to transfer into breastmilk, they should be avoided (GRADE 2C, SOA 99.5%).
Recommendations for paternal exposure to immunomodulatory drugs Due to the adverse effect of CYC on male fertility, semen cryopreservation is recommended for men prior to paternal exposure (GRADE 1C, SOA 99.5%). Men who take SSZ may have reduced fertility. There is little evidence to suggest that SSZ should be stopped pre-conception, unless conception is delayed by more than 12 months when stopping SSZ should be considered along with other causes of infertility (GRADE 1C, SOA 99.0%). Paternal exposure to the following anti-rheumatic medication is compatible with pregnancy: prednisolone, low-dose (≤25 mg/week) methotrexate, azathioprine (GRADE 1B); TNFi, cyclosporin (GRADE 1C); hydroxychloroquine, leflunomide, tacrolimus, mycophenolate mofetil, intravenous immunoglobulin, rituximab, interleukin-6 inhibitors, interleukin-1 inhibitors, abatacept, belimumab, interleukin-17 inhibitors, ustekinumab and JAKi (GRADE 2C, SOA 99.3%).
Due to the adverse effect of CYC on male fertility, semen cryopreservation is recommended for men prior to paternal exposure (GRADE 1C, SOA 99.5%). Men who take SSZ may have reduced fertility. There is little evidence to suggest that SSZ should be stopped pre-conception, unless conception is delayed by more than 12 months when stopping SSZ should be considered along with other causes of infertility (GRADE 1C, SOA 99.0%). Paternal exposure to the following anti-rheumatic medication is compatible with pregnancy: prednisolone, low-dose (≤25 mg/week) methotrexate, azathioprine (GRADE 1B); TNFi, cyclosporin (GRADE 1C); hydroxychloroquine, leflunomide, tacrolimus, mycophenolate mofetil, intravenous immunoglobulin, rituximab, interleukin-6 inhibitors, interleukin-1 inhibitors, abatacept, belimumab, interleukin-17 inhibitors, ustekinumab and JAKi (GRADE 2C, SOA 99.3%).
Implementation Awareness of these guidelines will aid clinical practitioners and patients in decision making. No barriers to implementation of these guidelines are anticipated. Key standards of care Patients with rheumatic disease should receive pre-pregnancy counselling and regular review during pregnancy and for 6 months post-partum by clinical practitioners with expertise in the management of rheumatic disease in pregnancy. They should have access to information on relevant medications in pregnancy and breastfeeding, enabling them to make informed decisions on drug use in pregnancy. Future research agenda Research questions include: should biologic DMARDs with known placental transfer be stopped or switched before/during pregnancy; are targeted synthetic DMARDs compatible with pregnancy; is it safe to give certain live vaccines to infants ≤6 months of age after third trimester exposure to biologic DMARDs with high placental transfer? Mechanism for audit of the guideline An audit tool to assess compliance with these guidelines is shown in , available at Rheumatology online.
Awareness of these guidelines will aid clinical practitioners and patients in decision making. No barriers to implementation of these guidelines are anticipated.
Patients with rheumatic disease should receive pre-pregnancy counselling and regular review during pregnancy and for 6 months post-partum by clinical practitioners with expertise in the management of rheumatic disease in pregnancy. They should have access to information on relevant medications in pregnancy and breastfeeding, enabling them to make informed decisions on drug use in pregnancy.
Research questions include: should biologic DMARDs with known placental transfer be stopped or switched before/during pregnancy; are targeted synthetic DMARDs compatible with pregnancy; is it safe to give certain live vaccines to infants ≤6 months of age after third trimester exposure to biologic DMARDs with high placental transfer?
An audit tool to assess compliance with these guidelines is shown in , available at Rheumatology online.
are available at Rheumatology online.
keac558_Supplementary_Data Click here for additional data file.
|
British Society for Rheumatology guideline on prescribing drugs in pregnancy and breastfeeding: immunomodulatory anti-rheumatic drugs and corticosteroids | fe6fc434-fdc0-4968-98fe-d47dd97970d0 | 10070073 | Internal Medicine[mh] | Background The rationale behind this update of the 2016 British Society for Rheumatology (BSR) guidelines on prescribing anti-rheumatic drugs in pregnancy and breastfeeding was described in detail in the guideline scope . In brief, despite the existence of additional evidence-based guidelines on prescribing/managing rheumatic disease in pregnancy , the information contained within them requires continual review to include emerging information on the safety of new and existing drugs in pregnancy. Chronic disease adversely affects pregnancy. Data from Mothers and Babies: Reducing Risk through Audits and Confidential Enquiries across the UK (MBRRACE-UK) reports regularly from a national programme of work conducting surveillance and investigating the causes of maternal deaths, stillbirths and infant deaths . Data from 2017–19 found that 8.8 women per 100 000 died during pregnancy or up to six weeks after childbirth or the end of pregnancy, and most women who died had multiple health problems or other vulnerabilities . In all decisions regarding medication choices and changes, it is important to consider the potential for deterioration in the mother's wellbeing through side effects or reduced disease control (and its adverse impact on the baby). As such, the potential benefit to the foetus from any drug changes in the mother must be balanced against the possible risks to the foetus from loss of disease control in the mother . Need for guideline There has been an appreciable increase in the number of published pregnancy exposures to biologic DMARDs (bDMARDs), and two of these drugs are now licensed for use in pregnancy. In addition, therapeutic advances in management of various inflammatory rheumatic diseases (IRDs) have led to an expansion of bDMARDs and biosimilars with different modes of action, as well as a new class of targeted synthetic DMARDs (tsDMARDs). The continuing expansion of existing and novel DMARDs means that uncertainty remains around the use of many of these drugs in pregnancy. This uncertainty may still lead to withdrawal of treatment from pregnant women unnecessarily . Discontinuation of treatment in preparation for or during early pregnancy can increase the risk of disease activity and flares during pregnancy, and are reported following discontinuation of biologics in patients with IRDs . The compatibility of various immunosuppressive and disease-modifying medications relevant to rheumatic disease will be covered in this update. This updated information will provide advice for healthcare professionals and patients, to ensure more confident prescribing in these scenarios, and will highlight any medications that should be stopped and/or avoided in the reproductive age group unless highly effective contraception is used, in line with guidance issued by the Medicines and Healthcare Products Regulatory Agency (MHRA) and the Faculty of Sexual and Reproductive Healthcare . Objectives of guideline To update the previous BSR guidelines on prescribing in pregnancy in rheumatic disease of the following drug categories: antimalarials; corticosteroids; conventional synthetic (cs)DMARDs and immunosuppressive therapies; bDMARDs; and tsDMARDs. The full list of medications is shown in , available at Rheumatology online. This revised guideline was produced by systematically reviewing all evidence published since the previous guideline, to answer specific questions in relation to each drug, as follows: Should it be stopped pre-conception? Is it compatible with pregnancy? Is it compatible with breastmilk exposure? Where possible, recommendations are made regarding compatibility with paternal exposure. Target audience The primary audience consists of health professionals in the UK directly involved in managing patients with rheumatic disease who are (or are planning to become) pregnant and/or breastfeeding, men with rheumatic disease who are planning to conceive, and patients with rheumatic disease who have unintentionally conceived while taking these medications. This audience includes rheumatologists, rheumatology nurses/allied health professionals, rheumatology speciality trainees and pharmacists, as well as the patients themselves. The guideline will also be useful to obstetricians, obstetric physicians, midwives, renal physicians, dermatologists, gastroenterologists, respiratory physicians and general practitioners who prescribe these medications in pregnancy. This guideline uses the terms ‘woman’, ‘maternal’ or ‘mother’ throughout. These should be taken to include people who do not identify as women but are pregnant or have given birth . Where the term ‘breastfeeding’ is used in this guideline it also refers to infant breastmilk exposure via other methods (e.g. expressed breastmilk, administered via a bottle). The areas the guideline does not cover This guideline does not cover the management of infertility or the indications for these drugs in specific rheumatic diseases in pregnancy. Other drug categories (pain management; NSAIDs and low dose aspirin; anticoagulants; bisphosphonates; anti-hypertensives; and pulmonary vasodilators) are considered in the BSR guideline on prescribing drugs in pregnancy and breastfeeding: comorbidity medications used in rheumatology practice ( https://doi.org/10.1093/rheumatology/keac552 ). All recommendations in this guideline were formulated by the working group on the basis of published evidence at the time of the systematic literature search, and do not necessarily refer to licensing information or Summary of Product Characteristics for individual medications. Stakeholder involvement This guideline was commissioned by the BSR Standards, Guidelines and Audit Working Group. A Guideline Working group (GWG) was created, consisting of a chair (I.G.), alongside representatives from relevant stakeholders shown in , available at Rheumatology online. In accordance with BSR policy, all members of the GWG made declarations of interest, available on the BSR website. Involvement and affiliations of stakeholder groups involved in guideline development The GWG consisted of rheumatologists from a range of clinical backgrounds, various allied health professionals, other specialists in women’s health, lay members and representatives from the United Kingdom Teratology Information Service (UKTIS). All members of the working group contributed to the process for agreeing key questions, guideline content, recommendations and strength of agreement.
The rationale behind this update of the 2016 British Society for Rheumatology (BSR) guidelines on prescribing anti-rheumatic drugs in pregnancy and breastfeeding was described in detail in the guideline scope . In brief, despite the existence of additional evidence-based guidelines on prescribing/managing rheumatic disease in pregnancy , the information contained within them requires continual review to include emerging information on the safety of new and existing drugs in pregnancy. Chronic disease adversely affects pregnancy. Data from Mothers and Babies: Reducing Risk through Audits and Confidential Enquiries across the UK (MBRRACE-UK) reports regularly from a national programme of work conducting surveillance and investigating the causes of maternal deaths, stillbirths and infant deaths . Data from 2017–19 found that 8.8 women per 100 000 died during pregnancy or up to six weeks after childbirth or the end of pregnancy, and most women who died had multiple health problems or other vulnerabilities . In all decisions regarding medication choices and changes, it is important to consider the potential for deterioration in the mother's wellbeing through side effects or reduced disease control (and its adverse impact on the baby). As such, the potential benefit to the foetus from any drug changes in the mother must be balanced against the possible risks to the foetus from loss of disease control in the mother .
There has been an appreciable increase in the number of published pregnancy exposures to biologic DMARDs (bDMARDs), and two of these drugs are now licensed for use in pregnancy. In addition, therapeutic advances in management of various inflammatory rheumatic diseases (IRDs) have led to an expansion of bDMARDs and biosimilars with different modes of action, as well as a new class of targeted synthetic DMARDs (tsDMARDs). The continuing expansion of existing and novel DMARDs means that uncertainty remains around the use of many of these drugs in pregnancy. This uncertainty may still lead to withdrawal of treatment from pregnant women unnecessarily . Discontinuation of treatment in preparation for or during early pregnancy can increase the risk of disease activity and flares during pregnancy, and are reported following discontinuation of biologics in patients with IRDs . The compatibility of various immunosuppressive and disease-modifying medications relevant to rheumatic disease will be covered in this update. This updated information will provide advice for healthcare professionals and patients, to ensure more confident prescribing in these scenarios, and will highlight any medications that should be stopped and/or avoided in the reproductive age group unless highly effective contraception is used, in line with guidance issued by the Medicines and Healthcare Products Regulatory Agency (MHRA) and the Faculty of Sexual and Reproductive Healthcare .
To update the previous BSR guidelines on prescribing in pregnancy in rheumatic disease of the following drug categories: antimalarials; corticosteroids; conventional synthetic (cs)DMARDs and immunosuppressive therapies; bDMARDs; and tsDMARDs. The full list of medications is shown in , available at Rheumatology online. This revised guideline was produced by systematically reviewing all evidence published since the previous guideline, to answer specific questions in relation to each drug, as follows: Should it be stopped pre-conception? Is it compatible with pregnancy? Is it compatible with breastmilk exposure? Where possible, recommendations are made regarding compatibility with paternal exposure.
The primary audience consists of health professionals in the UK directly involved in managing patients with rheumatic disease who are (or are planning to become) pregnant and/or breastfeeding, men with rheumatic disease who are planning to conceive, and patients with rheumatic disease who have unintentionally conceived while taking these medications. This audience includes rheumatologists, rheumatology nurses/allied health professionals, rheumatology speciality trainees and pharmacists, as well as the patients themselves. The guideline will also be useful to obstetricians, obstetric physicians, midwives, renal physicians, dermatologists, gastroenterologists, respiratory physicians and general practitioners who prescribe these medications in pregnancy. This guideline uses the terms ‘woman’, ‘maternal’ or ‘mother’ throughout. These should be taken to include people who do not identify as women but are pregnant or have given birth . Where the term ‘breastfeeding’ is used in this guideline it also refers to infant breastmilk exposure via other methods (e.g. expressed breastmilk, administered via a bottle).
This guideline does not cover the management of infertility or the indications for these drugs in specific rheumatic diseases in pregnancy. Other drug categories (pain management; NSAIDs and low dose aspirin; anticoagulants; bisphosphonates; anti-hypertensives; and pulmonary vasodilators) are considered in the BSR guideline on prescribing drugs in pregnancy and breastfeeding: comorbidity medications used in rheumatology practice ( https://doi.org/10.1093/rheumatology/keac552 ). All recommendations in this guideline were formulated by the working group on the basis of published evidence at the time of the systematic literature search, and do not necessarily refer to licensing information or Summary of Product Characteristics for individual medications.
This guideline was commissioned by the BSR Standards, Guidelines and Audit Working Group. A Guideline Working group (GWG) was created, consisting of a chair (I.G.), alongside representatives from relevant stakeholders shown in , available at Rheumatology online. In accordance with BSR policy, all members of the GWG made declarations of interest, available on the BSR website.
The GWG consisted of rheumatologists from a range of clinical backgrounds, various allied health professionals, other specialists in women’s health, lay members and representatives from the United Kingdom Teratology Information Service (UKTIS). All members of the working group contributed to the process for agreeing key questions, guideline content, recommendations and strength of agreement.
Statement of scope of literature search and strategy employed The evidence used to develop these guidelines was compiled from a systematic literature search conducted according to guidelines of Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) . Studies were identified by searching MEDLINE and Embase databases from 1 January 2014 to 31 December 2020 using combinations of the key MESH and free terms: pregnancy; lactation; breastfeeding; paternal exposure; and the name of each drug. The full electronic search strategies for the MEDLINE and Embase databases are shown in , available at Rheumatology online. Searches were not limited by disease indication; in addition to IRDs, studies in non-rheumatic diseases, such as psoriasis, inflammatory bowel disease (IBD) and organ transplantation were considered, if relevant. Additional published studies were identified through the Cochrane, LactMed (a National Library of Medicine database on drugs and lactation) and UKTIS databases (weblinks shown in , available at Rheumatology online), and checking of reference lists from recently published national and international guidelines and systematic literature reviews. Due to the paucity of data pertaining to the use of non-TNFi biologic drugs and tsDMARDs in pregnancy and breastmilk exposure, relevant pharmaceutical companies were contacted between July and November 2021, and asked for any further available data. Two independent reviewers screened the titles and abstracts of articles from the searches then reviewed the full texts of relevant studies, selecting articles that met inclusion criteria of: randomized and non-randomized controlled trials; cohort studies; case-control studies; and case series with more than ten participants. For medications with data on fewer than 300 pregnancy exposures, case series with more than five participants were eligible for inclusion. Conference abstracts were eligible for inclusion if they contained sufficient relevant data and there was no corresponding published manuscript. Case reports, and case series with fewer than five participants, were excluded, as were animal studies. Data extraction was performed by two reviewers. Disagreements arising during screening and extraction were resolved by group discussion, with involvement of a third reviewer where necessary. Statement of methods used to formulate the recommendations (levels of evidence) The working group met regularly to formalize the search strategy, review evidence, resolve disagreements and, finally, to determine recommendations. This guideline was developed in line with BSR’s Guidelines Protocol using Grading of Recommendations, Assessment, Development and Evaluations (GRADE) methodology to determine quality of evidence and strength of recommendation. Accompanying each recommendation in this guideline, in brackets, is the strength of recommendation, quality of evidence and strength of agreement (SOA). Strength of recommendation Using GRADE, recommendations were categorized as either strong (denoted by 1) or weak (denoted by 2), according to the balance between benefits and risks. A strong recommendation was made when the benefits clearly outweigh the risks (or vice versa). A weak recommendation denotes that the benefits are more closely balanced with the risk or more uncertain. Quality of evidence Using the GRADE approach, the quality of evidence was determined as either high (A), moderate (B) or low/very low (C), reflecting the confidence in the estimates of benefits or harm. Strength of agreement The wording of each recommendation was revised until all members were satisfied that they would score at least 80 on a scale of 1 (no agreement) to 100 (complete agreement). The 20/24 working group members with full voting rights then scored each recommendation on the same scale, and the average was calculated to generate a strength of agreement (SOA) score. Two patient representatives and data analysts expressed concern that they did not have sufficient medical knowledge of all drugs reviewed to score all recommendations; so while they fully agreed with each recommendation, they did not wish to score each one, and did not contribute to the final SOA score. Statement of any limits of search and when the guideline will be updated The search was conducted in January 2021. Limits were placed for English language and filters as described above. The guideline will be updated in five years.
The evidence used to develop these guidelines was compiled from a systematic literature search conducted according to guidelines of Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) . Studies were identified by searching MEDLINE and Embase databases from 1 January 2014 to 31 December 2020 using combinations of the key MESH and free terms: pregnancy; lactation; breastfeeding; paternal exposure; and the name of each drug. The full electronic search strategies for the MEDLINE and Embase databases are shown in , available at Rheumatology online. Searches were not limited by disease indication; in addition to IRDs, studies in non-rheumatic diseases, such as psoriasis, inflammatory bowel disease (IBD) and organ transplantation were considered, if relevant. Additional published studies were identified through the Cochrane, LactMed (a National Library of Medicine database on drugs and lactation) and UKTIS databases (weblinks shown in , available at Rheumatology online), and checking of reference lists from recently published national and international guidelines and systematic literature reviews. Due to the paucity of data pertaining to the use of non-TNFi biologic drugs and tsDMARDs in pregnancy and breastmilk exposure, relevant pharmaceutical companies were contacted between July and November 2021, and asked for any further available data. Two independent reviewers screened the titles and abstracts of articles from the searches then reviewed the full texts of relevant studies, selecting articles that met inclusion criteria of: randomized and non-randomized controlled trials; cohort studies; case-control studies; and case series with more than ten participants. For medications with data on fewer than 300 pregnancy exposures, case series with more than five participants were eligible for inclusion. Conference abstracts were eligible for inclusion if they contained sufficient relevant data and there was no corresponding published manuscript. Case reports, and case series with fewer than five participants, were excluded, as were animal studies. Data extraction was performed by two reviewers. Disagreements arising during screening and extraction were resolved by group discussion, with involvement of a third reviewer where necessary.
The working group met regularly to formalize the search strategy, review evidence, resolve disagreements and, finally, to determine recommendations. This guideline was developed in line with BSR’s Guidelines Protocol using Grading of Recommendations, Assessment, Development and Evaluations (GRADE) methodology to determine quality of evidence and strength of recommendation. Accompanying each recommendation in this guideline, in brackets, is the strength of recommendation, quality of evidence and strength of agreement (SOA).
Using GRADE, recommendations were categorized as either strong (denoted by 1) or weak (denoted by 2), according to the balance between benefits and risks. A strong recommendation was made when the benefits clearly outweigh the risks (or vice versa). A weak recommendation denotes that the benefits are more closely balanced with the risk or more uncertain.
Using the GRADE approach, the quality of evidence was determined as either high (A), moderate (B) or low/very low (C), reflecting the confidence in the estimates of benefits or harm.
The wording of each recommendation was revised until all members were satisfied that they would score at least 80 on a scale of 1 (no agreement) to 100 (complete agreement). The 20/24 working group members with full voting rights then scored each recommendation on the same scale, and the average was calculated to generate a strength of agreement (SOA) score. Two patient representatives and data analysts expressed concern that they did not have sufficient medical knowledge of all drugs reviewed to score all recommendations; so while they fully agreed with each recommendation, they did not wish to score each one, and did not contribute to the final SOA score.
The search was conducted in January 2021. Limits were placed for English language and filters as described above. The guideline will be updated in five years.
A flow diagram of study selection is shown in , displaying the initial number of articles screened ( n = 27 908), the number of articles selected for full-length review ( n = 706), and the number included in the final analysis from this updated search ( n = 154). This information was then merged with the results of the previous guideline’s systematic review to give the total exposure data for each drug. The following data were extracted where possible for each medication: number of studies and study type; number of pregnancy exposures; number of live births; pregnancy duration; birth weight; maternal complications; miscarriages; number and type of congenital anomalies (where possible, congenital anomalies described in original publications were classified as major/minor according to European surveillance of congenital anomalies (EUROCAT) definitions ); breastmilk exposure; long-term follow-up; and paternal exposure. An overall summary of compatibility of each drug pre-conception, during pregnancy, with breastmilk exposure, and with paternal exposure is shown in . For each drug, maternal information is summarized in the text and in and , while paternal exposures and recommendations are described separately and shown in . The data synthesis strategy for is shown in , available at Rheumatology online. Other relevant papers identified in our search that did not meet the inclusion criteria are discussed in the main text. Generic recommendations on prescribing immunomodulatory drugs and/or corticosteroids in rheumatic disease in pregnancy Pre-conception counselling should be addressed by all healthcare professionals, with referral to professionals with relevant expertise as appropriate, to optimize disease control before pregnancy; with advice on the timing of pregnancy, and drug therapy before, during and after pregnancy, including contraception (GRADE 1A, SOA 99.5%). If a woman is planning pregnancy, avoid pregnancy-incompatible drugs (GRADE 1A, SOA 100%). The risks and benefits to the mother and foetus of drug treatment to control maternal disease should be discussed and clearly documented by all healthcare professionals involved in the patient’s care (GRADE 1A, SOA 99.5%). Immunomodulatory drugs that are contraindicated in pregnancy should be switched to a pregnancy-compatible alternative in advance of conception to ensure maintenance of disease control on the new medication (GRADE 1A, SOA 100%). When no pregnancy-compatible drugs are suitable, control of severe/life-threatening maternal disease should take priority over concerns for potential foetal outcomes (GRADE 1B, SOA 99.0%). All biologic DMARDs may be continued throughout pregnancy if required to control active/severe maternal disease (GRADE 1B, SOA 98.5%). Immunization schedules in infants after in-utero exposure to biologic DMARDs will depend on timing of exposure, bioavailability and persistence of the drug, mechanism of action of the drug, and live vaccines (GRADE 1C, SOA 99.5%). Where possible, the minimum effective dose of immunomodulatory drug or corticosteroid should be used to maintain maternal disease suppression, and stopping the drug during pregnancy may be considered in women at low risk of disease flare on withdrawal of therapy (GRADE 1B, SOA 100%). Some drugs may reduce male fertility, but paternal drug exposure in humans has not convincingly been associated with adverse foetal development or pregnancy outcome. Although the evidence is weak, men who take rheumatological medicines should be reassured about the safety of conceiving (GRADE 2C, SOA 98.4%). Antimalarials HCQ is the antimalarial drug most used to treat rheumatic disease and has been extensively studied in pregnancy. We identified an additional 23 studies that, combined with the previous 23 studies , reported on ( n = 4701) pregnancy exposures to HCQ, with very limited information on other antimalarials [ , , , ]. Many of these studies were confounded by primarily reporting pregnancy outcomes in patients with SLE treated with other immunosuppressive agents, including MMF and corticosteroids, and use in anti-Ro/La positive patients in the prevention of congenital heart block (CHB). Despite these limitations, there were no appreciable adverse effects of HCQ on pregnancy duration or birth weight in the largest studies. In fact, several studies comparing HCQ-treated and untreated cohorts with rheumatic disease (mostly SLE) either found no significant difference between cohorts [ , , , , , , , ], or significantly longer pregnancy durations and/or higher birth weight in the HCQ-treated pregnancies [ , , , , , , , , , ]. The weighted mean for gestation across 15 studies reporting pregnancy duration in HCQ-exposed vs HCQ-unexposed pregnancies was 36.4 weeks and 34.7 weeks, respectively [ , , , , , , , , , ]. The weighted mean for birth weight for HCQ-exposed vs HCQ-unexposed pregnancies was 2847 and 2733 g, respectively, in 10 studies reporting these outcomes [ , , , , , , , , , ]. A total of 60 first trimester miscarriages were reported from 524 HCQ-exposed pregnancies (11.5%) in 10 studies, compared with 117 first trimester miscarriages in 718 HCQ-unexposed pregnancies (16.3%) [ , , , , , ]. No specific pattern of congenital malformations was observed in association with HCQ exposure. No increased risk of adverse foetal outcomes was reported in >3229 chloroquine-exposed pregnancies in four studies [ , , , ], including two studies where chloroquine was used as malaria prophylaxis during pregnancy; although, in these two studies, higher rates of maternal adverse events were reported, relative to the comparator (sulfadoxine-pyrimethamine). No information was found on mepacrine. The findings for HCQ were consistent across all studies apart from a large population-based cohort study comparing HCQ-exposed ( n = 2045) and HCQ-unexposed ( n = 21 679) pregnancies in patients with rheumatic disease, which did not control fully for disease, comorbidity-related pregnancy risk factors, dose of corticosteroids and combination with specific immunosuppressive drugs . This study found a small increase in the risk of congenital malformations associated with first trimester HCQ use, mainly oral clefts, respiratory anomalies and urinary defects, with wide confidence intervals for specific malformations. A statistically significant increase in risk, however, was only found with daily doses of ≥400 mg of HCQ. This study concluded that for most patients with autoimmune rheumatic disorders, the benefits of treatment during pregnancy will likely outweigh this risk. Importantly, a more recent study (published after our search date) of pregnant women prospectively enrolled into MotherToBaby/Organisation of Teratology Information Specialists (OTIS) pregnancy studies, compared outcomes for HCQ-exposed pregnancies ( n = 279) with disease-matched ( n = 279) and healthy comparator ( n = 279) HCQ-unexposed groups . Reassuringly, this study found no evidence of an increased risk for structural defects or other adverse outcomes with HCQ at any dose (average 325 mg/day; range 100–800 mg/day), except for an isolated finding of reduced head circumference at birth with HCQ exposure, which was not thought to be of any clinical significance. Therefore, advice on HCQ dosage in pregnancy relates to general guidance for reducing ophthalmic risk outside of pregnancy to a maximum of 400 mg/day, as pharmacokinetic changes in pregnancy reduce the reliability of weight-based dosing . Ultimately, it is important to maintain HCQ during pregnancy, as discontinuation of this drug in pregnancy may increase risk of disease flares and foetal loss . Disease flares would increase the need for alternative medications with more potential risks for mother or baby in pregnancy. Previous studies of breastmilk exposure to HCQ were mostly limited to case reports, showing that <1% of the maternal dose of HCQ was found in breastmilk . Three more recent studies of HCQ use ( n = 195) confirmed very low concentrations of HCQ in breastmilk and no adverse effects on breastfed infants [ , , ]. There remain limited studies of long-term outcomes in children, but no adverse immunological or clinical findings have been reported . Recommendations for hydroxychloroquine in pregnancy and breastmilk exposure HCQ remains the antimalarial of choice in women planning a pregnancy with rheumatic disease in need of treatment, and should be continued during pregnancy at a dose of ≤400 mg/day (GRADE 1B, SOA 100%). HCQ is compatible with breastmilk exposure (GRADE 1B, SOA 99.5%). Corticosteroids Corticosteroids used to treat rheumatic disease (prednisolone, prednisone and methylprednisolone) are metabolized in the placenta, and so 10% or less of the active drug reaches the foetus. Previously, we identified 47 studies on prednisolone and found it to be compatible with pregnancy and breastmilk exposure . Studies of corticosteroid use in pregnancy were confounded by multiple concomitant medications and use in high-risk pregnancies; particularly the fluorinated steroids, which are used to prevent or treat preterm labour and complications such as foetal lung immaturity. Therefore, we searched for further evidence on corticosteroids used to treat rheumatic disease and identified additional studies: 11 on prednisolone with ( n = 1218) pregnancies and one on methylprednisolone with ( n = 12) pregnancies . This evidence was combined with the previous studies: 47 on prednisolone ( n = 1503) [ , , , , , , , ]; 31 on dexamethasone ( n = 11 214) [ , , , , ]; 27 on betamethasone ( n = 27 746) [ , , , , , , , , ]; and 10 on general corticosteroid use ( n = 785) [ , , , , ]. Studies on the use of methylprednisolone in pregnancy were not specifically sought in the previous guideline because it is generally used as rescue therapy for severe disease. Compared with prednisolone, parenteral administration of methylprednisolone has a prolonged duration of action with similar rates of placental transfer to prednisolone . Previously, we found that following prednisolone (or unspecified corticosteroid) exposure, average pregnancy duration in the majority of randomized controlled trial (RCT), case-control, cohort and case-series studies (where reported) was usually term, at ≥37 weeks [ , , , , , , , , , , , ]. Other studies reporting ≤37-week delivery were confounded by factors such as maternal disease and concomitant medications [ , , , , , , , , , , , ]. Birth weights followed a similar pattern and were affected by preterm deliveries and confounding factors, as described above. For instance, prednisolone exposure in those RCTs, cohorts, case-control studies and case series which reported average gestations of ≥37 weeks, average birth weights ranged from 2.6–3.4 kg [ , , , , , , , , , , ]. Overall, prednisolone itself was not felt to have contributed to low birth weight (LBW) in any study . High rates of maternal complications compatible with underlying disease were previously reported for prednisolone and dexamethasone, but none were specifically attributed to these medications . The major congenital malformations observed with prednisolone were frequently confounded by concomitant teratogenic drug exposure, such as MMF , and the overall incidence was not significantly higher than in drug-free controls. Studies reporting major malformations with fluorinated steroid exposure [e.g. patent ductus arteriosus (PDA), blindness and deafness ] did not attribute them to steroid therapy. Furthermore, in the majority of cases, the steroids were used for treatment of underlying conditions such as preterm delivery , where steroids were found to be beneficial in improving outcomes, or treatment of maternal autoantibody-mediated cardiomyopathy . A large study analysing 832 636 live births did not show an increased risk of orofacial cleft palate with the use of corticosteroids in pregnancy . foetal loss in studies of prednisolone and fluorinated steroids was attributed to underlying disease rather than steroid therapy, such as in APS and complete atrio-ventricular block . Most (8/11) of the additional studies on maternal prednisolone exposure that we found in our updated search did not identify any adverse effects of prednisolone use on pregnancy outcomes [ , , ]. In contrast, a population-based study from Norway exploring the associations between disease activity and medications with offspring birth weight, pre-eclampsia and preterm birth in SLE found prednisolone use to be significantly associated with lower birth weight, increased risk of pre-eclampsia, and a 3-fold increase in preterm birth . A conference abstract reported that continuation of high-dose glucocorticoids during 164 pregnancies increased the risks of preterm birth, low birth weight and preterm premature rupture of membranes (PPROM) at prednisolone cut-off doses of 7.5 mg, 6.7 mg, 5.0 mg per day, respectively . In contrast, another conference abstract of 143 SLE pregnancies found that foetal complications were associated with prednisone >25 mg, and that low (10 mg/day) to moderate (10–24 mg/day) doses of prednisone during pregnancy were not associated with adverse foetal outcomes . Similarly, the largest prospective study of SLE pregnancy outcomes did not identify prednisolone ≤20 mg/day as a risk factor for adverse pregnancy outcomes . UKTIS notes that many of the studies reporting pregnancy outcomes following gestational exposure to systemic corticosteroids are limited by a lack of stratification to account for differing doses, treatment duration and steroid potencies, as well as confounding by maternal disease . It concludes that preterm delivery may be associated with gestational exposure to systemic corticosteroids, and further well-controlled studies are required to address this question. Therefore, an increased risk of adverse foetal effects following use of high-dose/potency corticosteroids, or use for extended periods, cannot be ruled out. Based on limited evidence, prednisone, prednisolone and methylprednisolone are considered compatible with breastmilk exposure . There remain few breastmilk exposure studies. One study, comprising 19 pregnancy and breastmilk exposures, found that prednisone and prednisolone exhibit dose- and concentration-dependent pharmacokinetics during pregnancy, and infant exposure to these agents via breastmilk is minimal . Another study of 12 patients with multiple sclerosis found the transfer of methylprednisolone into breastmilk to be low even when maternal serum concentration levels were highest at the end of an infusion, and although these levels were not considered to pose a threat to the infant, they state that mothers may choose to wait two to four h to further limit an infant’s exposure . Previously, long-term follow-up studies had not reported any adverse events after prednisolone exposure in pregnancy . Two additional studies did not report any adverse events from 9–12 months of post-partum follow-up of 227 non-rheumatic disease pregnancies exposed to prednisolone . Recommendations for corticosteroids in pregnancy and breastmilk exposure Prednisolone is compatible with pregnancy and is the preferred corticosteroid in the treatment of maternal rheumatological disease in pregnancy and requires shared care with obstetric teams to monitor maternal blood pressure and blood glucose (GRADE 1B, SOA 100%). Where possible, the dose of prednisolone should be <20 mg/day and tapered to the minimum effective dose to control maternal disease, in conjunction with steroid-sparing drugs compatible with pregnancy (GRADE 1C, SOA 99.5%). Prednisolone is compatible with breastmilk exposure (GRADE 1B, SOA 100%). Methylprednisolone has similar rates of placental transfer to prednisolone and would therefore be expected to be compatible with pregnancy and breastmilk exposure (GRADE 2C, SOA 99%).
Pre-conception counselling should be addressed by all healthcare professionals, with referral to professionals with relevant expertise as appropriate, to optimize disease control before pregnancy; with advice on the timing of pregnancy, and drug therapy before, during and after pregnancy, including contraception (GRADE 1A, SOA 99.5%). If a woman is planning pregnancy, avoid pregnancy-incompatible drugs (GRADE 1A, SOA 100%). The risks and benefits to the mother and foetus of drug treatment to control maternal disease should be discussed and clearly documented by all healthcare professionals involved in the patient’s care (GRADE 1A, SOA 99.5%). Immunomodulatory drugs that are contraindicated in pregnancy should be switched to a pregnancy-compatible alternative in advance of conception to ensure maintenance of disease control on the new medication (GRADE 1A, SOA 100%). When no pregnancy-compatible drugs are suitable, control of severe/life-threatening maternal disease should take priority over concerns for potential foetal outcomes (GRADE 1B, SOA 99.0%). All biologic DMARDs may be continued throughout pregnancy if required to control active/severe maternal disease (GRADE 1B, SOA 98.5%). Immunization schedules in infants after in-utero exposure to biologic DMARDs will depend on timing of exposure, bioavailability and persistence of the drug, mechanism of action of the drug, and live vaccines (GRADE 1C, SOA 99.5%). Where possible, the minimum effective dose of immunomodulatory drug or corticosteroid should be used to maintain maternal disease suppression, and stopping the drug during pregnancy may be considered in women at low risk of disease flare on withdrawal of therapy (GRADE 1B, SOA 100%). Some drugs may reduce male fertility, but paternal drug exposure in humans has not convincingly been associated with adverse foetal development or pregnancy outcome. Although the evidence is weak, men who take rheumatological medicines should be reassured about the safety of conceiving (GRADE 2C, SOA 98.4%).
HCQ is the antimalarial drug most used to treat rheumatic disease and has been extensively studied in pregnancy. We identified an additional 23 studies that, combined with the previous 23 studies , reported on ( n = 4701) pregnancy exposures to HCQ, with very limited information on other antimalarials [ , , , ]. Many of these studies were confounded by primarily reporting pregnancy outcomes in patients with SLE treated with other immunosuppressive agents, including MMF and corticosteroids, and use in anti-Ro/La positive patients in the prevention of congenital heart block (CHB). Despite these limitations, there were no appreciable adverse effects of HCQ on pregnancy duration or birth weight in the largest studies. In fact, several studies comparing HCQ-treated and untreated cohorts with rheumatic disease (mostly SLE) either found no significant difference between cohorts [ , , , , , , , ], or significantly longer pregnancy durations and/or higher birth weight in the HCQ-treated pregnancies [ , , , , , , , , , ]. The weighted mean for gestation across 15 studies reporting pregnancy duration in HCQ-exposed vs HCQ-unexposed pregnancies was 36.4 weeks and 34.7 weeks, respectively [ , , , , , , , , , ]. The weighted mean for birth weight for HCQ-exposed vs HCQ-unexposed pregnancies was 2847 and 2733 g, respectively, in 10 studies reporting these outcomes [ , , , , , , , , , ]. A total of 60 first trimester miscarriages were reported from 524 HCQ-exposed pregnancies (11.5%) in 10 studies, compared with 117 first trimester miscarriages in 718 HCQ-unexposed pregnancies (16.3%) [ , , , , , ]. No specific pattern of congenital malformations was observed in association with HCQ exposure. No increased risk of adverse foetal outcomes was reported in >3229 chloroquine-exposed pregnancies in four studies [ , , , ], including two studies where chloroquine was used as malaria prophylaxis during pregnancy; although, in these two studies, higher rates of maternal adverse events were reported, relative to the comparator (sulfadoxine-pyrimethamine). No information was found on mepacrine. The findings for HCQ were consistent across all studies apart from a large population-based cohort study comparing HCQ-exposed ( n = 2045) and HCQ-unexposed ( n = 21 679) pregnancies in patients with rheumatic disease, which did not control fully for disease, comorbidity-related pregnancy risk factors, dose of corticosteroids and combination with specific immunosuppressive drugs . This study found a small increase in the risk of congenital malformations associated with first trimester HCQ use, mainly oral clefts, respiratory anomalies and urinary defects, with wide confidence intervals for specific malformations. A statistically significant increase in risk, however, was only found with daily doses of ≥400 mg of HCQ. This study concluded that for most patients with autoimmune rheumatic disorders, the benefits of treatment during pregnancy will likely outweigh this risk. Importantly, a more recent study (published after our search date) of pregnant women prospectively enrolled into MotherToBaby/Organisation of Teratology Information Specialists (OTIS) pregnancy studies, compared outcomes for HCQ-exposed pregnancies ( n = 279) with disease-matched ( n = 279) and healthy comparator ( n = 279) HCQ-unexposed groups . Reassuringly, this study found no evidence of an increased risk for structural defects or other adverse outcomes with HCQ at any dose (average 325 mg/day; range 100–800 mg/day), except for an isolated finding of reduced head circumference at birth with HCQ exposure, which was not thought to be of any clinical significance. Therefore, advice on HCQ dosage in pregnancy relates to general guidance for reducing ophthalmic risk outside of pregnancy to a maximum of 400 mg/day, as pharmacokinetic changes in pregnancy reduce the reliability of weight-based dosing . Ultimately, it is important to maintain HCQ during pregnancy, as discontinuation of this drug in pregnancy may increase risk of disease flares and foetal loss . Disease flares would increase the need for alternative medications with more potential risks for mother or baby in pregnancy. Previous studies of breastmilk exposure to HCQ were mostly limited to case reports, showing that <1% of the maternal dose of HCQ was found in breastmilk . Three more recent studies of HCQ use ( n = 195) confirmed very low concentrations of HCQ in breastmilk and no adverse effects on breastfed infants [ , , ]. There remain limited studies of long-term outcomes in children, but no adverse immunological or clinical findings have been reported . Recommendations for hydroxychloroquine in pregnancy and breastmilk exposure HCQ remains the antimalarial of choice in women planning a pregnancy with rheumatic disease in need of treatment, and should be continued during pregnancy at a dose of ≤400 mg/day (GRADE 1B, SOA 100%). HCQ is compatible with breastmilk exposure (GRADE 1B, SOA 99.5%).
HCQ remains the antimalarial of choice in women planning a pregnancy with rheumatic disease in need of treatment, and should be continued during pregnancy at a dose of ≤400 mg/day (GRADE 1B, SOA 100%). HCQ is compatible with breastmilk exposure (GRADE 1B, SOA 99.5%).
Corticosteroids used to treat rheumatic disease (prednisolone, prednisone and methylprednisolone) are metabolized in the placenta, and so 10% or less of the active drug reaches the foetus. Previously, we identified 47 studies on prednisolone and found it to be compatible with pregnancy and breastmilk exposure . Studies of corticosteroid use in pregnancy were confounded by multiple concomitant medications and use in high-risk pregnancies; particularly the fluorinated steroids, which are used to prevent or treat preterm labour and complications such as foetal lung immaturity. Therefore, we searched for further evidence on corticosteroids used to treat rheumatic disease and identified additional studies: 11 on prednisolone with ( n = 1218) pregnancies and one on methylprednisolone with ( n = 12) pregnancies . This evidence was combined with the previous studies: 47 on prednisolone ( n = 1503) [ , , , , , , , ]; 31 on dexamethasone ( n = 11 214) [ , , , , ]; 27 on betamethasone ( n = 27 746) [ , , , , , , , , ]; and 10 on general corticosteroid use ( n = 785) [ , , , , ]. Studies on the use of methylprednisolone in pregnancy were not specifically sought in the previous guideline because it is generally used as rescue therapy for severe disease. Compared with prednisolone, parenteral administration of methylprednisolone has a prolonged duration of action with similar rates of placental transfer to prednisolone . Previously, we found that following prednisolone (or unspecified corticosteroid) exposure, average pregnancy duration in the majority of randomized controlled trial (RCT), case-control, cohort and case-series studies (where reported) was usually term, at ≥37 weeks [ , , , , , , , , , , , ]. Other studies reporting ≤37-week delivery were confounded by factors such as maternal disease and concomitant medications [ , , , , , , , , , , , ]. Birth weights followed a similar pattern and were affected by preterm deliveries and confounding factors, as described above. For instance, prednisolone exposure in those RCTs, cohorts, case-control studies and case series which reported average gestations of ≥37 weeks, average birth weights ranged from 2.6–3.4 kg [ , , , , , , , , , , ]. Overall, prednisolone itself was not felt to have contributed to low birth weight (LBW) in any study . High rates of maternal complications compatible with underlying disease were previously reported for prednisolone and dexamethasone, but none were specifically attributed to these medications . The major congenital malformations observed with prednisolone were frequently confounded by concomitant teratogenic drug exposure, such as MMF , and the overall incidence was not significantly higher than in drug-free controls. Studies reporting major malformations with fluorinated steroid exposure [e.g. patent ductus arteriosus (PDA), blindness and deafness ] did not attribute them to steroid therapy. Furthermore, in the majority of cases, the steroids were used for treatment of underlying conditions such as preterm delivery , where steroids were found to be beneficial in improving outcomes, or treatment of maternal autoantibody-mediated cardiomyopathy . A large study analysing 832 636 live births did not show an increased risk of orofacial cleft palate with the use of corticosteroids in pregnancy . foetal loss in studies of prednisolone and fluorinated steroids was attributed to underlying disease rather than steroid therapy, such as in APS and complete atrio-ventricular block . Most (8/11) of the additional studies on maternal prednisolone exposure that we found in our updated search did not identify any adverse effects of prednisolone use on pregnancy outcomes [ , , ]. In contrast, a population-based study from Norway exploring the associations between disease activity and medications with offspring birth weight, pre-eclampsia and preterm birth in SLE found prednisolone use to be significantly associated with lower birth weight, increased risk of pre-eclampsia, and a 3-fold increase in preterm birth . A conference abstract reported that continuation of high-dose glucocorticoids during 164 pregnancies increased the risks of preterm birth, low birth weight and preterm premature rupture of membranes (PPROM) at prednisolone cut-off doses of 7.5 mg, 6.7 mg, 5.0 mg per day, respectively . In contrast, another conference abstract of 143 SLE pregnancies found that foetal complications were associated with prednisone >25 mg, and that low (10 mg/day) to moderate (10–24 mg/day) doses of prednisone during pregnancy were not associated with adverse foetal outcomes . Similarly, the largest prospective study of SLE pregnancy outcomes did not identify prednisolone ≤20 mg/day as a risk factor for adverse pregnancy outcomes . UKTIS notes that many of the studies reporting pregnancy outcomes following gestational exposure to systemic corticosteroids are limited by a lack of stratification to account for differing doses, treatment duration and steroid potencies, as well as confounding by maternal disease . It concludes that preterm delivery may be associated with gestational exposure to systemic corticosteroids, and further well-controlled studies are required to address this question. Therefore, an increased risk of adverse foetal effects following use of high-dose/potency corticosteroids, or use for extended periods, cannot be ruled out. Based on limited evidence, prednisone, prednisolone and methylprednisolone are considered compatible with breastmilk exposure . There remain few breastmilk exposure studies. One study, comprising 19 pregnancy and breastmilk exposures, found that prednisone and prednisolone exhibit dose- and concentration-dependent pharmacokinetics during pregnancy, and infant exposure to these agents via breastmilk is minimal . Another study of 12 patients with multiple sclerosis found the transfer of methylprednisolone into breastmilk to be low even when maternal serum concentration levels were highest at the end of an infusion, and although these levels were not considered to pose a threat to the infant, they state that mothers may choose to wait two to four h to further limit an infant’s exposure . Previously, long-term follow-up studies had not reported any adverse events after prednisolone exposure in pregnancy . Two additional studies did not report any adverse events from 9–12 months of post-partum follow-up of 227 non-rheumatic disease pregnancies exposed to prednisolone . Recommendations for corticosteroids in pregnancy and breastmilk exposure Prednisolone is compatible with pregnancy and is the preferred corticosteroid in the treatment of maternal rheumatological disease in pregnancy and requires shared care with obstetric teams to monitor maternal blood pressure and blood glucose (GRADE 1B, SOA 100%). Where possible, the dose of prednisolone should be <20 mg/day and tapered to the minimum effective dose to control maternal disease, in conjunction with steroid-sparing drugs compatible with pregnancy (GRADE 1C, SOA 99.5%). Prednisolone is compatible with breastmilk exposure (GRADE 1B, SOA 100%). Methylprednisolone has similar rates of placental transfer to prednisolone and would therefore be expected to be compatible with pregnancy and breastmilk exposure (GRADE 2C, SOA 99%).
Prednisolone is compatible with pregnancy and is the preferred corticosteroid in the treatment of maternal rheumatological disease in pregnancy and requires shared care with obstetric teams to monitor maternal blood pressure and blood glucose (GRADE 1B, SOA 100%). Where possible, the dose of prednisolone should be <20 mg/day and tapered to the minimum effective dose to control maternal disease, in conjunction with steroid-sparing drugs compatible with pregnancy (GRADE 1C, SOA 99.5%). Prednisolone is compatible with breastmilk exposure (GRADE 1B, SOA 100%). Methylprednisolone has similar rates of placental transfer to prednisolone and would therefore be expected to be compatible with pregnancy and breastmilk exposure (GRADE 2C, SOA 99%).
Methotrexate MTX is contraindicated in pregnancy and was previously recommended to be stopped at least three months in advance of conception . UKTIS considers MTX risk in pregnancy to be dependent on its use at high (>25 mg per week) or low (≤25 mg per week) dosages . Rheumatology usage of MTX to treat inflammatory arthritis falls into the low-dose category and is far removed from the high doses used as a chemotherapeutic agent in the treatment of various cancers (e.g. >500 mg/m 2 ) or as an abortifacient at 50 mg/m 2 . UKTIS concludes that exposure to high-dose MTX in early pregnancy confers a risk of severe embryopathy (including craniofacial defects, malformations of the digits and defects of the spine and ribs) in the foetus, and the option of termination of pregnancy should be discussed with the patient. In contrast, for exposure to lower doses of MTX prior to conception, additional foetal monitoring is advised, as well as counselling of women and their partners about the lack of available data to facilitate quantification of risk of adverse pregnancy outcomes. Previously, we identified a high proportion of major anomalies following MTX (and other DMARD) exposure, predominantly during the first trimester of pregnancy, in 27 pregnancies from 10 studies [ , , , ]. An additional 12 studies of MTX exposure in 2765 pregnancies were identified: six maternal studies and six paternal studies . Several studies reported on the risks of pre-conceptual and pregnancy exposure to MTX. Data from the National Birth Defects Prevention Study, a US case-control study of major birth defects, reported that 4/10 113 (0.04%) mothers of foetus/infants without major birth defects (controls) had been exposed to MTX, compared with 16/27 623 (0.06%) mothers of live-born infants with a major birth defect (cases) who had been exposed to MTX . The dose of MTX was not reported, but indications included a neoplasm of endocrine glands and so was presumably of high dose in at least one case. Of the 16 cases with major birth defects, 15 were exposed from three months pre-conception to the end of the first trimester. A cohort study of 240 SLE pregnancies in whom 36.8% were exposed to MTX before and during the first trimester reported an increased risk of foetal complications . A large prospective observational multicentre cohort study of 324 pregnancies exposed to MTX found an increase in the cumulative incidence of spontaneous miscarriage (42.5%) and major congenital anomalies (6.6%) among pregnancies ( n = 188) exposed to a median dose of 10 mg/week of MTX after a median of 4.3 weeks post-conception . This difference reached statistical significance when compared with a cohort of women without autoimmune diseases, but not when compared with a disease-matched cohort. No increased risk of miscarriage or major congenital anomaly was found in pregnancies ( n = 136) exposed to a median dose of 15 mg/week of MTX that was stopped three months pre-conception. Not all studies reported increased risks with MTX exposure. A study of pre-conception use of MTX on miscarriage rates in 114 RA pregnancies, compared with 48 MTX-unexposed RA pregnancies, did not find a statistically significant association between miscarriage and MTX use . An analysis of 18 pregnancies exposed to MTX (≤20 mg/week) from up to one year pre-conception and in the first trimester found a high percentage of live-born children with no malformations . Analysis of 23 first trimester exposures to low-dose MTX, identified from three United States health plan databases, did not reveal a significant increase in the risk of congenital malformations, foetal death or neonatal complications in women with chronic autoimmune disease, compared with those who received MTX before, but not during, pregnancy . These studies provide some evidence that a 3-month MTX-free interval prior to conception might not be required. Therefore, unintentional exposure to low-dose MTX during the peri-conceptional period confers minimal risk in unintended pregnancy exposures, and so termination of pregnancy is not routinely recommended for MTX exposure unless it is maternally requested due to unplanned pregnancy [ , , ]. Studies of MTX in breastfeeding remain very limited. Although they did not meet our inclusion criteria, we identified two case reports that found low levels of MTX in breastmilk and no adverse effects on the breastfed infants . LactMed describes low levels of MTX in breastmilk and conflicting expert opinion on whether it can safely be used during breastfeeding . It states that withholding breastfeeding for 24 h after a weekly low-dose of MTX may decrease the infant's dose by 40%, and that if breastfeeding is undertaken during long-term, low-dose MTX use, monitoring of the infant's complete blood count and differential could be considered. Post-partum follow-up of up to 14 months after first trimester MTX exposure was reported in three infants with long-term complications of foetal MTX syndrome, including semi lobar holoprosencephaly, cardiac abnormalities, tracheostomy and requirement for antiepileptic therapy . Recommendations for methotrexate in pregnancy and breastmilk exposure MTX at any dose should be avoided in pregnancy and stopped at least one month in advance of planned conception, when it should be switched to another pregnancy-compatible drug to ensure maintenance of maternal disease suppression (GRADE 1A, SOA 98%). In women treated with low-dose (≤25 mg/week) MTX within one month prior to conception, folic acid supplementation (5 mg/day) should be continued up to 12 weeks of pregnancy (GRADE 1B, SOA 99.5%). In unintended pregnancy on low-dose (≤25 mg/week) MTX, there is minimal risk to the foetus; the drug should be stopped immediately, folic acid supplementation (5 mg/day) continued, and a careful evaluation of foetal risk with early referral to a foetal medicine department considered (GRADE 1C, SOA 100%). Although only minute amounts of MTX are excreted into breastmilk, MTX cannot be recommended in breastfeeding because of theoretical risks and insufficient data on outcomes (GRADE 2C, SOA 99%). Sulfasalazine Previously, we recommended that SSZ is compatible with pregnancy and breastmilk exposure and can be continued with adequate folic acid supplementation (5 mg/day) . This recommendation was based on six publications reporting SSZ exposure in 178 pregnancies in patients with RA, osteoporosis and ankylosing spondylitis (AS) [ , , , , , ]. These studies contained limited information relating to miscarriage rate, pregnancy duration, birth weight or malformation rate; overall, however, there were no significant adverse effects highlighted that were considered to be directly attributable to SSZ. We did not identify any additional studies on the use of SSZ in pregnancy, breastmilk exposure or paternal exposure. UKTIS does not identify any specific risks with SSZ exposure. It comments that although high-dose folic acid (5 mg/day) is generally recommended, no studies have investigated whether there is increased benefit of this higher dose of folic acid compared with a standard dose of 400 micrograms/day. Minimal amounts of SSZ are expressed in breastmilk, and it can be used during breastfeeding if the infant is full term and healthy, although it should be avoided in ill, stressed or premature infants, and in infants with hyperbilirubinaemia or glucose-6-phosphate dehydrogenase deficiency . Recommendations for sulfasalazine in pregnancy and breastmilk exposure SSZ is compatible throughout pregnancy, with folic acid 5 mg/day recommended in the periconception period and during the first trimester (GRADE 1B, SOA 100%). SSZ is compatible with breastmilk exposure in healthy, full-term infants (GRADE 1C, SOA 99.5%). Leflunomide Based upon limited evidence, we previously found that LEF may not be a human teratogen, but there was insufficient evidence to support its compatibility in human pregnancy, so our recommendation was that LEF is not the DMARD of choice in women planning pregnancy . This recommendation was based on data from seven studies [ , , ] reporting on 111 pregnancies exposed to LEF (discontinued in almost all cases in the first trimester, and frequently followed by a cholestyramine washout). Overall, the findings were largely reassuring, with no direct evidence of human teratogenicity. We identified three additional studies of 703 pregnancies exposed to LEF at various stages of pregnancy, with varying exposure to washout and/or plasma testing of LEF metabolites, which did not find an increased risk of adverse pregnancy outcomes compared with the general population . Although it was not included within our systematic review, pregnancy outcomes have been reported for teriflunomide—the principal active metabolite responsible for leflunomide’s activity in vivo —which, at recommended doses, results in a similar range of plasma concentrations to leflunomide . The known outcomes from 222 pregnancy exposures to teriflunomide for relapsing forms of multiple sclerosis also found outcomes consistent with the general population . Overall, these findings do not indicate a teratogenic risk of LEF in human pregnancies. The practicality of previous recommendations regarding the testing of plasma levels of teriflunomide has been questioned , and testing is not currently routinely available in the UK. We did not identify any data on breastmilk exposure to LEF, and no information is available in LactMed . Recommendations for leflunomide in pregnancy and breastmilk exposure LEF may not be a human teratogen but there remains insufficient evidence to support use at the time of conception or during pregnancy (GRADE 1B, SOA 98%). Women on LEF considering pregnancy should stop and undergo a standard cholestyramine washout procedure, and switch to alternative medication compatible with pregnancy (GRADE 1B, SOA 98.8%). If unintended conception occurs on LEF, the drug should be stopped immediately and a standard cholestyramine washout procedure given, with early referral to a foetal medicine department considered (GRADE 1B, SOA 99%). LEF is not recommended while breastfeeding (GRADE 1C, SOA 99.5%). Azathioprine Previously, we recommended that AZA is compatible with pregnancy at doses ≤2 mg/kg, with breastmilk exposure and with paternal exposure . These recommendations were based on 28 studies [ , , , , , , , , , , , , , , , , , ] in 738 AZA-exposed pregnancies, which included a wide range of diagnoses and concomitant medications, compared with 1121 disease-matched and 667 healthy controls. These data did not demonstrate an increased risk of miscarriage, preterm birth, low birth weight or congenital malformation due to AZA exposure in pregnancy. We identified an additional nine studies of 3699 pregnancy exposures to AZA: six maternal studies [ , , ] and three paternal studies [ , , ]. Overall, the findings from maternal exposures ( n = 1019) to AZA did not identify any adverse pregnancy outcomes. One study, reporting on AZA metabolism in 30 IBD pregnancies, measured active metabolites and found only 6-thioguanine nucleotide (6-TGN) but not 6-mercaptopurine (6-MP) in umbilical cord blood at delivery; no major teratogenicity was observed, although 60% of the infants had anaemia, which was suspected to be due to maternal thiopurine use . Two of these studies extended follow-up to 3 months and nearly 10 years, without any adverse effects being reported. The majority of studies did not specify the mean/median dose of AZA utilized in the study populations, and there is no clear evidence regarding a dose limit. Use of AZA at an effective dose should be supported by monitoring of blood tests, following local guidelines. Based on our previous evidence from 26 infants breastfed by mothers on AZA or 6-MP, minimal amounts of AZA were detected in breastmilk, and no adverse effects were identified . We did not identify any new studies of breastmilk exposure to AZA. LactMed states that avoiding breastfeeding for 4 h after maternal ingestion of AZA should markedly reduce the dose received by the infant in breastmilk . In routine clinical practice, there is no concern in the management of solid organ transplant patients who breastfeed on this drug . Recommendations for azathioprine in pregnancy and breastmilk exposure AZA is compatible throughout pregnancy (GRADE 1B, SOA 100%). AZA is compatible with breastmilk exposure (GRADE 2C, SOA 99.5%). Ciclosporin An earlier consensus document reviewed evidence from >800 human pregnancies exposed to ciclosporin (CsA) . Our previous search from 2005 onwards identified a further 13 studies/reports [ , , , , , , , , , , , , ] of 98 pregnancies in patients with a variety of diseases and multiple concomitant medications who had been exposed to CsA at 2–6 mg/kg during pregnancy. Reports of increased rates of preterm delivery and low birth weight were confounded by maternal disease and concomitant medications, and there was no evidence of an increased malformation risk . Comorbidities, such as hypertension, pre-eclampsia and gestational diabetes mellitus, were reported at higher incidences than the general population. Based upon this evidence, CsA was considered compatible with pregnancy at the lowest effective dose, with monitoring of blood pressure, blood glucose and renal function . UKTIS draws a similar conclusion . We identified an additional five studies of 550 pregnancy exposures [ , , ]. Three studies reported on maternal exposure [ , , ]. A cohort study of 240 SLE pregnancies, in whom 50% were exposed to CsA before and during the first trimester, increased the risk of pancytopenia and/or pre-eclampsia in maternal outcomes . A single-centre experience of outcomes of pregnancy ( n = 117) following liver transplantation did not find any difference between those on CsA ( n = 34) compared with tacrolimus ( n = 81), and so did not attribute these outcomes to medication . A study of the efficacy and safety of CsA in 29 pregnancies of patients with systemic autoimmune diseases did not find an increased risk of maternal–foetal complications, and stated that it should be continued in patients who benefit from therapy . Previously identified studies described small amounts of CsA in breastmilk and almost universally undetectable blood levels in infants , without any adverse effects reported during breastmilk exposure. We found a further study that reported low transfer of CsA and its metabolites into the breastmilk of seven post-transplant mothers in the first two days post-partum, although this study was not designed to make a corresponding assessment of drug safety . LactMed recommends that breastfed infants should be monitored if CsA is used during lactation, possibly with measurement of serum levels if there is a concern for toxicity . In routine clinical practice, there is no concern in the management of solid organ transplant patients who breastfeed on this drug . No additional studies of long-term follow-up were identified to those found previously on 10 infants exposed to CsA in utero , which reported no complications at 11–14 months [ , , ]. Recommendations for ciclosporin in pregnancy and breastmilk exposure CsA is compatible throughout pregnancy with monitoring of maternal blood pressure, renal function, blood glucose and drug levels (GRADE 1B, SOA 100%). CsA is compatible with breastmilk exposure (GRADE 2C, SOA 99.7%). Tacrolimus Based upon previous consensus and our previous review of six studies [ , , , , , ] of 26 pregnancies exposed to tacrolimus and two breastmilk exposure studies , tacrolimus was considered compatible with pregnancy and breastmilk exposure. There were complex confounding issues in many of these studies and, overall, no adverse outcomes were considered to be directly attributable to tacrolimus . We found additional evidence from eight studies of 489 pregnancy exposures to tacrolimus . Studies of maternal outcomes, mostly from solid organ transplant recipients, reported varying incidences of adverse maternal-foetal outcomes, but these outcomes were confounded by transplant-associated comorbidities and concomitant immunosuppression, particularly MMF [ , , ]. UKTIS concludes that the available data do not suggest an association between spontaneous miscarriage, congenital malformation or intrauterine death and exposure to tacrolimus during pregnancy, but data are limited and potentially confounded; therefore, an increased risk of these outcomes cannot be excluded . Previously, we found studies reporting low levels of tacrolimus in umbilical cord blood and breastmilk in small numbers of breastfed infants without any adverse effects . These findings were confirmed in an additional study of 13 breastfed infants of mothers with SLE . This study found concentrations of tacrolimus in the umbilical cord blood were lower than those in the maternal blood; the relative infant dose in breastfed infants of tacrolimus was <1%, and the level of tacrolimus in infant blood was below detectable limits. LactMed suggests that exclusively breastfed infants should be monitored . In routine clinical practice, there is no concern in the management of solid organ transplant patients who breastfeed on this drug . Recommendations for tacrolimus in pregnancy and breastmilk exposure Tacrolimus is compatible throughout pregnancy with monitoring of maternal blood pressure, renal function, blood glucose and drug levels (GRADE 2B, SOA 100%). Tacrolimus is compatible with breastmilk exposure (GRADE 2C, SOA 99.8%). Cyclophosphamide CYC is a known human teratogen and is gonadotoxic in men and women . Our previous findings from reports of predominantly first trimester use of CYC in nine pregnancies, revealed multiple adverse outcomes in mothers with severe maternal disease and multiple concomitant medications [ , , , , ]. No maternal complications of CYC were reported. The nine pregnancies ended in two first trimester miscarriages, six healthy infants and one major congenital anomaly (Klippel–Feil syndrome). Follow-up to 87–90 months in four live births reported normal development in three children and the single case of Klippel–Feil syndrome . We found an additional study of pregnancies ( n = 11) in women with multiple sclerosis who had been exposed to CYC prior to conception ; 10 women had a successful delivery [five preterm delivery and one small for gestational age (SGA)], while one underwent elective termination. It should be noted, however, that the time between the last dose of CYC and conception in this study was an average of 3.7 ± 1.5 years (range 0.33–5.9 years). Although it did not meet our inclusion criteria, one case report analysed breastmilk levels of CYC in a women with multiple sclerosis . CYC levels in breastmilk samples were measured after IV CYC at a dose of 2.8 g, with relatively low levels identified in the milk. The authors reported an average relative infant dose for a period of four days that varied from 4.7% at day 1 to 0.9% at day 4. Recommendations for cyclophosphamide in pregnancy and breastmilk exposure CYC is a known teratogen and gonadotoxic, and therefore should only be considered in pregnancy in cases of severe life/organ-threatening maternal disease when there is appreciable risk of maternal and foetal morbidity and mortality without this therapy (GRADE 1B, SOA 99.5%). CYC is not recommended while breastfeeding (GRADE 2C, SOA 100%). Mycophenolate mofetil MMF is a known teratogen and is recommended to be stopped at least 6 weeks before a planned pregnancy . It is rapidly absorbed following oral administration and hydrolysed to form the active ingredient, mycophenolic acid (MPA). This active metabolite has a mean apparent half-life of 17 h after a 1 g oral dose of MMF, and undergoes enterohepatic circulation, with a secondary plasma peak at 6–12 h after an oral or intravenous dose . We previously reviewed data from 16 studies/reports [ , , , , , , , , , ] of 90 pregnancies exposed to MMF, mostly from renal transplant patients in whom there was concomitant exposure to prednisolone and tacrolimus. Increased rates of premature delivery, low birth weight and major congenital malformations were reported, including malformations typical for the previously described MMF embryopathy (including cleft lip and/or palate, microtia with aural atresia, micrognathia and ocular anomalies) . Our updated search found eight further studies of 934 pregnancy exposures to MMF: five maternal exposure studies and three paternal exposure studies [ , , ]. The five studies of maternal exposure in pregnancies ( n = 714) all reported increased risks of miscarriage and birth defects, with 351 foetal losses, eight stillbirths and 38 cases of congenital malformation . These studies were mostly of first trimester exposure, with three including second and third trimester exposures. One study, however, found that following discontinuation of MMF within 6 weeks of conception, outcomes including the rates of birth defects and miscarriages were similar to pregnancies not exposed to MMF . UKTIS describes the increased risks of first trimester pregnancy loss, as well as major congenital anomalies, and states that women of childbearing potential who are prescribed MMF or MPA should be informed of the associated risks to the foetus and, therefore, the importance of adequate contraception. It notes that European Medicines Agency guidelines for male and female patients, published in October 2015 following a periodic safety update review, recommend additional measures to prevent foetal exposure to MMF and should be read prior to prescribing MMF . As in our previous search, we did not identify any data on breastmilk exposure. Similarly, LactMed reports that no information is available on the excretion of MMF into breastmilk, and that a few infants have reportedly been breastfed during MMF therapy with no adverse effects reported . We did not identify any additional long-term follow-up data to that previously found of one case of ‘small for age’ with otherwise normal development , and another study reporting on 3/6 exposed children (one with normal development, one who required hearing aids, and one who had motor and speech delay) . Recommendations for mycophenolate mofetil in pregnancy and breastmilk exposure MMF remains contraindicated during pregnancy, and should be avoided in women planning pregnancy or switched to a pregnancy-compatible alternative at least 6 weeks before attempting to conceive (GRADE 1B, SOA 100%). In cases of unintended conception, switch MMF to a pregnancy-compatible alternative and refer to local experts for further advice and risk assessment (GRADE 1B, SOA 100%). MMF is not recommended while breastfeeding (GRADE 2C, SOA 99.7%). Intravenous immunoglobulin IVIG is considered to be compatible with pregnancy and breastmilk exposure . Our previous review of data found 16 studies/reports [ , , , , , , , , , ] of 336 pregnancies in which IVIG was used, mostly in APS or in the prevention of CHB in anti-Ro/La positive mothers. The studies identified were focused on therapeutic efficacy rather than the safety of IVIG; hence, all outcomes were confounded by use in patients with high-risk pregnancies and multiple concomitant medications. Overall, the number and type of maternal and foetal complications observed were compatible with known effects of the underlying maternal disease on pregnancy, rather than being specific to IVIG. The studies reviewed did not raise any new concerns to question the accepted safety of IVIG in pregnancy . We identified two further studies of 67 exposures to IVIG to treat immune thrombocytopenia (ITP) in pregnancy . One study found that glucocorticoids increased the risk of maternal hypertension, while the addition of IVIG to corticosteroid regimes did not adversely affect pregnancy outcomes . The other study found comparable benefits of IVIG compared with corticosteroids in treating ITP in pregnancy, compared with no treatment, and similar neonatal outcomes between the treatment groups . UKTIS does not report on IVIG. None of the studies we previously or recently identified addressed the use of IVIG in breastmilk exposure or with paternal exposure. LactMed states that immunoglobulin is a normal component of breastmilk, and data from two mothers indicate that IgG concentrations in milk are normal or higher, and IgM levels in milk are normal or lower, during IVIG therapy . Recommendations for intravenous immunoglobulin in pregnancy and breastmilk exposure IVIG is compatible with pregnancy (GRADE 1B, SOA 99.5%). IVIG is compatible with breastmilk exposure (GRADE 2C, SOA 100%).
MTX is contraindicated in pregnancy and was previously recommended to be stopped at least three months in advance of conception . UKTIS considers MTX risk in pregnancy to be dependent on its use at high (>25 mg per week) or low (≤25 mg per week) dosages . Rheumatology usage of MTX to treat inflammatory arthritis falls into the low-dose category and is far removed from the high doses used as a chemotherapeutic agent in the treatment of various cancers (e.g. >500 mg/m 2 ) or as an abortifacient at 50 mg/m 2 . UKTIS concludes that exposure to high-dose MTX in early pregnancy confers a risk of severe embryopathy (including craniofacial defects, malformations of the digits and defects of the spine and ribs) in the foetus, and the option of termination of pregnancy should be discussed with the patient. In contrast, for exposure to lower doses of MTX prior to conception, additional foetal monitoring is advised, as well as counselling of women and their partners about the lack of available data to facilitate quantification of risk of adverse pregnancy outcomes. Previously, we identified a high proportion of major anomalies following MTX (and other DMARD) exposure, predominantly during the first trimester of pregnancy, in 27 pregnancies from 10 studies [ , , , ]. An additional 12 studies of MTX exposure in 2765 pregnancies were identified: six maternal studies and six paternal studies . Several studies reported on the risks of pre-conceptual and pregnancy exposure to MTX. Data from the National Birth Defects Prevention Study, a US case-control study of major birth defects, reported that 4/10 113 (0.04%) mothers of foetus/infants without major birth defects (controls) had been exposed to MTX, compared with 16/27 623 (0.06%) mothers of live-born infants with a major birth defect (cases) who had been exposed to MTX . The dose of MTX was not reported, but indications included a neoplasm of endocrine glands and so was presumably of high dose in at least one case. Of the 16 cases with major birth defects, 15 were exposed from three months pre-conception to the end of the first trimester. A cohort study of 240 SLE pregnancies in whom 36.8% were exposed to MTX before and during the first trimester reported an increased risk of foetal complications . A large prospective observational multicentre cohort study of 324 pregnancies exposed to MTX found an increase in the cumulative incidence of spontaneous miscarriage (42.5%) and major congenital anomalies (6.6%) among pregnancies ( n = 188) exposed to a median dose of 10 mg/week of MTX after a median of 4.3 weeks post-conception . This difference reached statistical significance when compared with a cohort of women without autoimmune diseases, but not when compared with a disease-matched cohort. No increased risk of miscarriage or major congenital anomaly was found in pregnancies ( n = 136) exposed to a median dose of 15 mg/week of MTX that was stopped three months pre-conception. Not all studies reported increased risks with MTX exposure. A study of pre-conception use of MTX on miscarriage rates in 114 RA pregnancies, compared with 48 MTX-unexposed RA pregnancies, did not find a statistically significant association between miscarriage and MTX use . An analysis of 18 pregnancies exposed to MTX (≤20 mg/week) from up to one year pre-conception and in the first trimester found a high percentage of live-born children with no malformations . Analysis of 23 first trimester exposures to low-dose MTX, identified from three United States health plan databases, did not reveal a significant increase in the risk of congenital malformations, foetal death or neonatal complications in women with chronic autoimmune disease, compared with those who received MTX before, but not during, pregnancy . These studies provide some evidence that a 3-month MTX-free interval prior to conception might not be required. Therefore, unintentional exposure to low-dose MTX during the peri-conceptional period confers minimal risk in unintended pregnancy exposures, and so termination of pregnancy is not routinely recommended for MTX exposure unless it is maternally requested due to unplanned pregnancy [ , , ]. Studies of MTX in breastfeeding remain very limited. Although they did not meet our inclusion criteria, we identified two case reports that found low levels of MTX in breastmilk and no adverse effects on the breastfed infants . LactMed describes low levels of MTX in breastmilk and conflicting expert opinion on whether it can safely be used during breastfeeding . It states that withholding breastfeeding for 24 h after a weekly low-dose of MTX may decrease the infant's dose by 40%, and that if breastfeeding is undertaken during long-term, low-dose MTX use, monitoring of the infant's complete blood count and differential could be considered. Post-partum follow-up of up to 14 months after first trimester MTX exposure was reported in three infants with long-term complications of foetal MTX syndrome, including semi lobar holoprosencephaly, cardiac abnormalities, tracheostomy and requirement for antiepileptic therapy . Recommendations for methotrexate in pregnancy and breastmilk exposure MTX at any dose should be avoided in pregnancy and stopped at least one month in advance of planned conception, when it should be switched to another pregnancy-compatible drug to ensure maintenance of maternal disease suppression (GRADE 1A, SOA 98%). In women treated with low-dose (≤25 mg/week) MTX within one month prior to conception, folic acid supplementation (5 mg/day) should be continued up to 12 weeks of pregnancy (GRADE 1B, SOA 99.5%). In unintended pregnancy on low-dose (≤25 mg/week) MTX, there is minimal risk to the foetus; the drug should be stopped immediately, folic acid supplementation (5 mg/day) continued, and a careful evaluation of foetal risk with early referral to a foetal medicine department considered (GRADE 1C, SOA 100%). Although only minute amounts of MTX are excreted into breastmilk, MTX cannot be recommended in breastfeeding because of theoretical risks and insufficient data on outcomes (GRADE 2C, SOA 99%).
MTX at any dose should be avoided in pregnancy and stopped at least one month in advance of planned conception, when it should be switched to another pregnancy-compatible drug to ensure maintenance of maternal disease suppression (GRADE 1A, SOA 98%). In women treated with low-dose (≤25 mg/week) MTX within one month prior to conception, folic acid supplementation (5 mg/day) should be continued up to 12 weeks of pregnancy (GRADE 1B, SOA 99.5%). In unintended pregnancy on low-dose (≤25 mg/week) MTX, there is minimal risk to the foetus; the drug should be stopped immediately, folic acid supplementation (5 mg/day) continued, and a careful evaluation of foetal risk with early referral to a foetal medicine department considered (GRADE 1C, SOA 100%). Although only minute amounts of MTX are excreted into breastmilk, MTX cannot be recommended in breastfeeding because of theoretical risks and insufficient data on outcomes (GRADE 2C, SOA 99%).
Previously, we recommended that SSZ is compatible with pregnancy and breastmilk exposure and can be continued with adequate folic acid supplementation (5 mg/day) . This recommendation was based on six publications reporting SSZ exposure in 178 pregnancies in patients with RA, osteoporosis and ankylosing spondylitis (AS) [ , , , , , ]. These studies contained limited information relating to miscarriage rate, pregnancy duration, birth weight or malformation rate; overall, however, there were no significant adverse effects highlighted that were considered to be directly attributable to SSZ. We did not identify any additional studies on the use of SSZ in pregnancy, breastmilk exposure or paternal exposure. UKTIS does not identify any specific risks with SSZ exposure. It comments that although high-dose folic acid (5 mg/day) is generally recommended, no studies have investigated whether there is increased benefit of this higher dose of folic acid compared with a standard dose of 400 micrograms/day. Minimal amounts of SSZ are expressed in breastmilk, and it can be used during breastfeeding if the infant is full term and healthy, although it should be avoided in ill, stressed or premature infants, and in infants with hyperbilirubinaemia or glucose-6-phosphate dehydrogenase deficiency . Recommendations for sulfasalazine in pregnancy and breastmilk exposure SSZ is compatible throughout pregnancy, with folic acid 5 mg/day recommended in the periconception period and during the first trimester (GRADE 1B, SOA 100%). SSZ is compatible with breastmilk exposure in healthy, full-term infants (GRADE 1C, SOA 99.5%).
SSZ is compatible throughout pregnancy, with folic acid 5 mg/day recommended in the periconception period and during the first trimester (GRADE 1B, SOA 100%). SSZ is compatible with breastmilk exposure in healthy, full-term infants (GRADE 1C, SOA 99.5%).
Based upon limited evidence, we previously found that LEF may not be a human teratogen, but there was insufficient evidence to support its compatibility in human pregnancy, so our recommendation was that LEF is not the DMARD of choice in women planning pregnancy . This recommendation was based on data from seven studies [ , , ] reporting on 111 pregnancies exposed to LEF (discontinued in almost all cases in the first trimester, and frequently followed by a cholestyramine washout). Overall, the findings were largely reassuring, with no direct evidence of human teratogenicity. We identified three additional studies of 703 pregnancies exposed to LEF at various stages of pregnancy, with varying exposure to washout and/or plasma testing of LEF metabolites, which did not find an increased risk of adverse pregnancy outcomes compared with the general population . Although it was not included within our systematic review, pregnancy outcomes have been reported for teriflunomide—the principal active metabolite responsible for leflunomide’s activity in vivo —which, at recommended doses, results in a similar range of plasma concentrations to leflunomide . The known outcomes from 222 pregnancy exposures to teriflunomide for relapsing forms of multiple sclerosis also found outcomes consistent with the general population . Overall, these findings do not indicate a teratogenic risk of LEF in human pregnancies. The practicality of previous recommendations regarding the testing of plasma levels of teriflunomide has been questioned , and testing is not currently routinely available in the UK. We did not identify any data on breastmilk exposure to LEF, and no information is available in LactMed . Recommendations for leflunomide in pregnancy and breastmilk exposure LEF may not be a human teratogen but there remains insufficient evidence to support use at the time of conception or during pregnancy (GRADE 1B, SOA 98%). Women on LEF considering pregnancy should stop and undergo a standard cholestyramine washout procedure, and switch to alternative medication compatible with pregnancy (GRADE 1B, SOA 98.8%). If unintended conception occurs on LEF, the drug should be stopped immediately and a standard cholestyramine washout procedure given, with early referral to a foetal medicine department considered (GRADE 1B, SOA 99%). LEF is not recommended while breastfeeding (GRADE 1C, SOA 99.5%).
LEF may not be a human teratogen but there remains insufficient evidence to support use at the time of conception or during pregnancy (GRADE 1B, SOA 98%). Women on LEF considering pregnancy should stop and undergo a standard cholestyramine washout procedure, and switch to alternative medication compatible with pregnancy (GRADE 1B, SOA 98.8%). If unintended conception occurs on LEF, the drug should be stopped immediately and a standard cholestyramine washout procedure given, with early referral to a foetal medicine department considered (GRADE 1B, SOA 99%). LEF is not recommended while breastfeeding (GRADE 1C, SOA 99.5%).
Previously, we recommended that AZA is compatible with pregnancy at doses ≤2 mg/kg, with breastmilk exposure and with paternal exposure . These recommendations were based on 28 studies [ , , , , , , , , , , , , , , , , , ] in 738 AZA-exposed pregnancies, which included a wide range of diagnoses and concomitant medications, compared with 1121 disease-matched and 667 healthy controls. These data did not demonstrate an increased risk of miscarriage, preterm birth, low birth weight or congenital malformation due to AZA exposure in pregnancy. We identified an additional nine studies of 3699 pregnancy exposures to AZA: six maternal studies [ , , ] and three paternal studies [ , , ]. Overall, the findings from maternal exposures ( n = 1019) to AZA did not identify any adverse pregnancy outcomes. One study, reporting on AZA metabolism in 30 IBD pregnancies, measured active metabolites and found only 6-thioguanine nucleotide (6-TGN) but not 6-mercaptopurine (6-MP) in umbilical cord blood at delivery; no major teratogenicity was observed, although 60% of the infants had anaemia, which was suspected to be due to maternal thiopurine use . Two of these studies extended follow-up to 3 months and nearly 10 years, without any adverse effects being reported. The majority of studies did not specify the mean/median dose of AZA utilized in the study populations, and there is no clear evidence regarding a dose limit. Use of AZA at an effective dose should be supported by monitoring of blood tests, following local guidelines. Based on our previous evidence from 26 infants breastfed by mothers on AZA or 6-MP, minimal amounts of AZA were detected in breastmilk, and no adverse effects were identified . We did not identify any new studies of breastmilk exposure to AZA. LactMed states that avoiding breastfeeding for 4 h after maternal ingestion of AZA should markedly reduce the dose received by the infant in breastmilk . In routine clinical practice, there is no concern in the management of solid organ transplant patients who breastfeed on this drug . Recommendations for azathioprine in pregnancy and breastmilk exposure AZA is compatible throughout pregnancy (GRADE 1B, SOA 100%). AZA is compatible with breastmilk exposure (GRADE 2C, SOA 99.5%).
AZA is compatible throughout pregnancy (GRADE 1B, SOA 100%). AZA is compatible with breastmilk exposure (GRADE 2C, SOA 99.5%).
An earlier consensus document reviewed evidence from >800 human pregnancies exposed to ciclosporin (CsA) . Our previous search from 2005 onwards identified a further 13 studies/reports [ , , , , , , , , , , , , ] of 98 pregnancies in patients with a variety of diseases and multiple concomitant medications who had been exposed to CsA at 2–6 mg/kg during pregnancy. Reports of increased rates of preterm delivery and low birth weight were confounded by maternal disease and concomitant medications, and there was no evidence of an increased malformation risk . Comorbidities, such as hypertension, pre-eclampsia and gestational diabetes mellitus, were reported at higher incidences than the general population. Based upon this evidence, CsA was considered compatible with pregnancy at the lowest effective dose, with monitoring of blood pressure, blood glucose and renal function . UKTIS draws a similar conclusion . We identified an additional five studies of 550 pregnancy exposures [ , , ]. Three studies reported on maternal exposure [ , , ]. A cohort study of 240 SLE pregnancies, in whom 50% were exposed to CsA before and during the first trimester, increased the risk of pancytopenia and/or pre-eclampsia in maternal outcomes . A single-centre experience of outcomes of pregnancy ( n = 117) following liver transplantation did not find any difference between those on CsA ( n = 34) compared with tacrolimus ( n = 81), and so did not attribute these outcomes to medication . A study of the efficacy and safety of CsA in 29 pregnancies of patients with systemic autoimmune diseases did not find an increased risk of maternal–foetal complications, and stated that it should be continued in patients who benefit from therapy . Previously identified studies described small amounts of CsA in breastmilk and almost universally undetectable blood levels in infants , without any adverse effects reported during breastmilk exposure. We found a further study that reported low transfer of CsA and its metabolites into the breastmilk of seven post-transplant mothers in the first two days post-partum, although this study was not designed to make a corresponding assessment of drug safety . LactMed recommends that breastfed infants should be monitored if CsA is used during lactation, possibly with measurement of serum levels if there is a concern for toxicity . In routine clinical practice, there is no concern in the management of solid organ transplant patients who breastfeed on this drug . No additional studies of long-term follow-up were identified to those found previously on 10 infants exposed to CsA in utero , which reported no complications at 11–14 months [ , , ]. Recommendations for ciclosporin in pregnancy and breastmilk exposure CsA is compatible throughout pregnancy with monitoring of maternal blood pressure, renal function, blood glucose and drug levels (GRADE 1B, SOA 100%). CsA is compatible with breastmilk exposure (GRADE 2C, SOA 99.7%).
CsA is compatible throughout pregnancy with monitoring of maternal blood pressure, renal function, blood glucose and drug levels (GRADE 1B, SOA 100%). CsA is compatible with breastmilk exposure (GRADE 2C, SOA 99.7%).
Based upon previous consensus and our previous review of six studies [ , , , , , ] of 26 pregnancies exposed to tacrolimus and two breastmilk exposure studies , tacrolimus was considered compatible with pregnancy and breastmilk exposure. There were complex confounding issues in many of these studies and, overall, no adverse outcomes were considered to be directly attributable to tacrolimus . We found additional evidence from eight studies of 489 pregnancy exposures to tacrolimus . Studies of maternal outcomes, mostly from solid organ transplant recipients, reported varying incidences of adverse maternal-foetal outcomes, but these outcomes were confounded by transplant-associated comorbidities and concomitant immunosuppression, particularly MMF [ , , ]. UKTIS concludes that the available data do not suggest an association between spontaneous miscarriage, congenital malformation or intrauterine death and exposure to tacrolimus during pregnancy, but data are limited and potentially confounded; therefore, an increased risk of these outcomes cannot be excluded . Previously, we found studies reporting low levels of tacrolimus in umbilical cord blood and breastmilk in small numbers of breastfed infants without any adverse effects . These findings were confirmed in an additional study of 13 breastfed infants of mothers with SLE . This study found concentrations of tacrolimus in the umbilical cord blood were lower than those in the maternal blood; the relative infant dose in breastfed infants of tacrolimus was <1%, and the level of tacrolimus in infant blood was below detectable limits. LactMed suggests that exclusively breastfed infants should be monitored . In routine clinical practice, there is no concern in the management of solid organ transplant patients who breastfeed on this drug . Recommendations for tacrolimus in pregnancy and breastmilk exposure Tacrolimus is compatible throughout pregnancy with monitoring of maternal blood pressure, renal function, blood glucose and drug levels (GRADE 2B, SOA 100%). Tacrolimus is compatible with breastmilk exposure (GRADE 2C, SOA 99.8%).
Tacrolimus is compatible throughout pregnancy with monitoring of maternal blood pressure, renal function, blood glucose and drug levels (GRADE 2B, SOA 100%). Tacrolimus is compatible with breastmilk exposure (GRADE 2C, SOA 99.8%).
CYC is a known human teratogen and is gonadotoxic in men and women . Our previous findings from reports of predominantly first trimester use of CYC in nine pregnancies, revealed multiple adverse outcomes in mothers with severe maternal disease and multiple concomitant medications [ , , , , ]. No maternal complications of CYC were reported. The nine pregnancies ended in two first trimester miscarriages, six healthy infants and one major congenital anomaly (Klippel–Feil syndrome). Follow-up to 87–90 months in four live births reported normal development in three children and the single case of Klippel–Feil syndrome . We found an additional study of pregnancies ( n = 11) in women with multiple sclerosis who had been exposed to CYC prior to conception ; 10 women had a successful delivery [five preterm delivery and one small for gestational age (SGA)], while one underwent elective termination. It should be noted, however, that the time between the last dose of CYC and conception in this study was an average of 3.7 ± 1.5 years (range 0.33–5.9 years). Although it did not meet our inclusion criteria, one case report analysed breastmilk levels of CYC in a women with multiple sclerosis . CYC levels in breastmilk samples were measured after IV CYC at a dose of 2.8 g, with relatively low levels identified in the milk. The authors reported an average relative infant dose for a period of four days that varied from 4.7% at day 1 to 0.9% at day 4. Recommendations for cyclophosphamide in pregnancy and breastmilk exposure CYC is a known teratogen and gonadotoxic, and therefore should only be considered in pregnancy in cases of severe life/organ-threatening maternal disease when there is appreciable risk of maternal and foetal morbidity and mortality without this therapy (GRADE 1B, SOA 99.5%). CYC is not recommended while breastfeeding (GRADE 2C, SOA 100%).
CYC is a known teratogen and gonadotoxic, and therefore should only be considered in pregnancy in cases of severe life/organ-threatening maternal disease when there is appreciable risk of maternal and foetal morbidity and mortality without this therapy (GRADE 1B, SOA 99.5%). CYC is not recommended while breastfeeding (GRADE 2C, SOA 100%).
MMF is a known teratogen and is recommended to be stopped at least 6 weeks before a planned pregnancy . It is rapidly absorbed following oral administration and hydrolysed to form the active ingredient, mycophenolic acid (MPA). This active metabolite has a mean apparent half-life of 17 h after a 1 g oral dose of MMF, and undergoes enterohepatic circulation, with a secondary plasma peak at 6–12 h after an oral or intravenous dose . We previously reviewed data from 16 studies/reports [ , , , , , , , , , ] of 90 pregnancies exposed to MMF, mostly from renal transplant patients in whom there was concomitant exposure to prednisolone and tacrolimus. Increased rates of premature delivery, low birth weight and major congenital malformations were reported, including malformations typical for the previously described MMF embryopathy (including cleft lip and/or palate, microtia with aural atresia, micrognathia and ocular anomalies) . Our updated search found eight further studies of 934 pregnancy exposures to MMF: five maternal exposure studies and three paternal exposure studies [ , , ]. The five studies of maternal exposure in pregnancies ( n = 714) all reported increased risks of miscarriage and birth defects, with 351 foetal losses, eight stillbirths and 38 cases of congenital malformation . These studies were mostly of first trimester exposure, with three including second and third trimester exposures. One study, however, found that following discontinuation of MMF within 6 weeks of conception, outcomes including the rates of birth defects and miscarriages were similar to pregnancies not exposed to MMF . UKTIS describes the increased risks of first trimester pregnancy loss, as well as major congenital anomalies, and states that women of childbearing potential who are prescribed MMF or MPA should be informed of the associated risks to the foetus and, therefore, the importance of adequate contraception. It notes that European Medicines Agency guidelines for male and female patients, published in October 2015 following a periodic safety update review, recommend additional measures to prevent foetal exposure to MMF and should be read prior to prescribing MMF . As in our previous search, we did not identify any data on breastmilk exposure. Similarly, LactMed reports that no information is available on the excretion of MMF into breastmilk, and that a few infants have reportedly been breastfed during MMF therapy with no adverse effects reported . We did not identify any additional long-term follow-up data to that previously found of one case of ‘small for age’ with otherwise normal development , and another study reporting on 3/6 exposed children (one with normal development, one who required hearing aids, and one who had motor and speech delay) . Recommendations for mycophenolate mofetil in pregnancy and breastmilk exposure MMF remains contraindicated during pregnancy, and should be avoided in women planning pregnancy or switched to a pregnancy-compatible alternative at least 6 weeks before attempting to conceive (GRADE 1B, SOA 100%). In cases of unintended conception, switch MMF to a pregnancy-compatible alternative and refer to local experts for further advice and risk assessment (GRADE 1B, SOA 100%). MMF is not recommended while breastfeeding (GRADE 2C, SOA 99.7%).
MMF remains contraindicated during pregnancy, and should be avoided in women planning pregnancy or switched to a pregnancy-compatible alternative at least 6 weeks before attempting to conceive (GRADE 1B, SOA 100%). In cases of unintended conception, switch MMF to a pregnancy-compatible alternative and refer to local experts for further advice and risk assessment (GRADE 1B, SOA 100%). MMF is not recommended while breastfeeding (GRADE 2C, SOA 99.7%).
IVIG is considered to be compatible with pregnancy and breastmilk exposure . Our previous review of data found 16 studies/reports [ , , , , , , , , , ] of 336 pregnancies in which IVIG was used, mostly in APS or in the prevention of CHB in anti-Ro/La positive mothers. The studies identified were focused on therapeutic efficacy rather than the safety of IVIG; hence, all outcomes were confounded by use in patients with high-risk pregnancies and multiple concomitant medications. Overall, the number and type of maternal and foetal complications observed were compatible with known effects of the underlying maternal disease on pregnancy, rather than being specific to IVIG. The studies reviewed did not raise any new concerns to question the accepted safety of IVIG in pregnancy . We identified two further studies of 67 exposures to IVIG to treat immune thrombocytopenia (ITP) in pregnancy . One study found that glucocorticoids increased the risk of maternal hypertension, while the addition of IVIG to corticosteroid regimes did not adversely affect pregnancy outcomes . The other study found comparable benefits of IVIG compared with corticosteroids in treating ITP in pregnancy, compared with no treatment, and similar neonatal outcomes between the treatment groups . UKTIS does not report on IVIG. None of the studies we previously or recently identified addressed the use of IVIG in breastmilk exposure or with paternal exposure. LactMed states that immunoglobulin is a normal component of breastmilk, and data from two mothers indicate that IgG concentrations in milk are normal or higher, and IgM levels in milk are normal or lower, during IVIG therapy . Recommendations for intravenous immunoglobulin in pregnancy and breastmilk exposure IVIG is compatible with pregnancy (GRADE 1B, SOA 99.5%). IVIG is compatible with breastmilk exposure (GRADE 2C, SOA 100%).
IVIG is compatible with pregnancy (GRADE 1B, SOA 99.5%). IVIG is compatible with breastmilk exposure (GRADE 2C, SOA 100%).
Biological therapies are commonly used as second-line agents to treat various forms of IRDs. They are recombinant proteins, most commonly monoclonal IgG1 antibodies directed against specific targets, or fusion proteins containing the Fc portion of IgG1 joined to receptor-blocking proteins. The presence of the Fc region of IgG1 in most of these biologic drugs is required for their active placental transfer, which accelerates by active transport from the second trimester onwards. Biologic drugs are often given alongside other DMARDs, and decisions regarding continuation of treatment should be taken for each drug independently. Although the evidence base for biosimilar use in pregnancy and breastmilk exposure is more limited than for originator biologics, they would be expected to have comparable effects. Therefore, for pragmatic reasons, our recommendations are applicable to equivalent licensed biosimilars. Five studies were identified that assessed the impact of biologic drugs as a whole on pregnancy outcomes . These studies included a total of 379 pregnancies in women with autoimmune disease on predominantly anti-TNFα drugs, but also rituximab (RTX), abatacept (ABA), tocilizumab (TCZ), ustekinumab (UST) and anakinra. Separate birth outcomes for specific medications were not reported. Overall, the authors of these studies found no increased risk of miscarriage, stillbirth or congenital anomalies in biologic-exposed patients. One study that reported outcomes for 120 pregnancies in women with autoimmune diseases (predominantly RA and IBD) found a slightly increased risk of prematurity and a trend towards low birth weight in the biologic-exposed group compared with those not exposed . However, once statistical modelling had been performed to correct for confounding by indication and proxies of unmeasured confounders, no association was found between biologic exposure and birth weight or gestational age. In addition, the same authors found that biologic use was not associated with an increased risk of serious infections in mothers, during post-partum, or in infants during the first year of life . Anti-TNFα drugs Five biologic agents that inhibit TNFα (TNFi) are currently licensed to treat IRDs: etanercept (ETA), infliximab (INF), adalimumab (ADA), golimumab (GOL) and certolizumab pegol (CZP). Three of these drugs (INF, ADA and GOL) are monoclonal IgG1 directed against TNFα, one (ETA) is a fusion protein of the TNF receptor joined to the Fc region of IgG1, while CZP is an antigen-binding fragment (Fab′) of a monoclonal anti-TNFα antibody which lacks the Fc region of IgG1 and has been conjugated with polyethylene glycol (PEG). These drugs have different half-lives, bioavailability and rates of placental transfer, which are relevant when considering their potential use in pregnancy. Initial 2006/8 consensus recommendations advised avoidance of ETA, INF and ADA in pregnancy and breastfeeding due to a lack of evidence rather than evidence of harm . Previously, we reviewed outcome data from TNFi-exposed pregnancies ( n = 706) of patients with predominantly IBD but also rheumatic disease and non-autoimmune-mediated recurrent spontaneous miscarriage, compared with ( n = 399) disease and ( n = 170) healthy control pregnancies [ , , , , , , , , , ]. There were multiple confounders of concomitant therapies (including MTX, LEF and MMF) and active inflammatory disease. Overall, these studies did not describe an increased incidence of adverse effects upon miscarriage rates, pregnancy duration, birth weight, foetal death or congenital malformation that was attributable to ETA, INF, ADA or CZP. At that time, there was limited information on placental transfer, and no published studies of GOL in human pregnancy or breastmilk exposure. TNFi exposure in pregnancy and with breastmilk exposure has been extensively studied since our last search. We identified an additional 50 studies, reporting 12 491 pregnancy exposures to TNFi, including INF ( n ≥ 5377), ADA ( n ≥ 2797), ETA ( n ≥ 2210), CZP ( n ≥ 776) and GOL ( n ≥ 196) [ , , ]. Many studies reported combined outcomes for exposure to different TNFi agents. The majority of studies of maternal exposure did not report an increased risk of preterm birth, miscarriage, low birth weight or congenital malformations [ , , , , , , , , , , , ]. Different adverse outcomes were reported in some studies, however. A study of ETA found that the proportion of infants with major birth defects was higher (9.4% vs 3.5%, respectively) in ETA-exposed pregnancies ( n = 370) than in pregnancies of disease-matched, non-exposed women ( n = 164) ; however, the lack of a specific pattern of birth defects and the expected minimal placental transfer of ETA in early pregnancy did not support the biologic plausibility of a drug-related effect. A study reporting a lower live birth rate in INF-exposed pregnancies ( n = 99) in women with Crohn’s disease considered their findings to be confounded by more severe disease in those patients exposed to INF and increased exposure to other immunosuppressive agents . A population-based study of TNFi-exposed pregnancies ( n = 1027) found increased risks of preterm birth, caesarean section and SGA babies in comparison with TNFi-unexposed pregnancies ( n = 9399) ; however, the authors noted that these associations may have been related to underlying disease activity rather than agent-specific effects, due to diverse findings across disease groups. A retrospective cohort study of TNFi-exposed pregnancies ( n = 1457) in women with IBD found TNFi exposure to be an independent risk factor for maternal complications and infections when compared with TNFi-unexposed pregnancies ( n = 9818) . In this study, TNFi exposure did not associate with congenital malformations or an increased risk of infection in children during the first year of life. Furthermore, there was no difference in the risk of complications in women exposed to TNFi during the third trimester, relative to cessation before week 24, although disease relapses were more common in those stopping TNFi prior to the third trimester. A study of 4961 pregnant women with autoimmune inflammatory conditions found similar risks of serious infections in women taking steroids, csDMARDs or TNFi during pregnancy, but found that higher doses of steroids were an independent risk factor for serious infections in pregnancy . A registry-based study from Denmark and Sweden reported a non-statistically significant higher risk of having children with birth defects in women with RA, AS, psoriatic arthritis (PsA), IBD or psoriasis who had received TNFi during pregnancy ( n = 683), relative to women with chronic inflammatory disease but without TNFi exposure ( n = 21 549) ; however, the heterogeneity of observed birth defects went against a common aetiology. A prospective cohort study of TNFi pregnancy exposures ( n = 495) in women with chronic inflammatory disease (RA, AS, PsA, psoriasis and IBD) found prenatal TNFi exposure to be associated with an increased risk of birth defects without a distinct pattern of malformations, when compared with non-disease-matched, TNFi-unexposed controls ( n = 1532) . An increased risk of preterm births and reduced birth weight, but not spontaneous miscarriage, was also noted. The authors concluded that, although TNFi may carry a risk of adverse pregnancy outcomes of moderate clinical relevance, they may remain a treatment option, considering the impact of inadequately controlled disease on the mother and unborn child. A small number of studies specifically compared risks between TNFi agents. In one study, INF was found to be associated with a greater risk for preterm births relative to ETA, and a higher prevalence of severely SGA babies relative to ETA and ADA, in pregnant women with RA, AS, PsA or psoriasis. In IBD, however, the risk of preterm births and SGA babies did not differ between INF and ADA . In a study of individual safety reports in pregnant IBD patients exposed to TNFi ( n = 783), the odds for maternal or foetal adverse events were found to be lower for CZP monotherapy, but not for INF or ADA monotherapy, when compared with an aminosalicylate monotherapy comparator in multi-level regression models . In another study, the risk of birth defects did not differ significantly between ADA, INF or ETA-exposed women with chronic inflammatory diseases . Several studies specifically compared outcomes for pregnancies exposed to TNFi during late vs early trimesters [ , , , , , , , , , , , ], the majority of which reported no significant concerns with late trimester exposure. In a study of INF-exposed pregnancies ( n = 1850) in women with IBD, RA, AS, PsA and psoriasis, frequencies of congenital abnormalities and other adverse birth and infant outcomes (including neonatal infections) were similar when comparing first and third trimester exposure . In a study comparing early discontinuation of INF (>90 days before delivery; n = 68) to late discontinuation (<90 days before delivery; n = 318) in pregnancies of women with IBD, early discontinuation was associated with increased disease flares, more steroid usage and more preterm births than late discontinuation . Rates of other adverse outcomes, including congenital malformations and infant respiratory infections, were similar between these groups. A further study of TNFi-exposed pregnancies ( n = 153) in women with IBD reported that continuation of TNFi after gestational week 30, relative to cessation before week 30, was independently associated with modestly lower birth weights in multivariate regression models after adjustment for disease activity, but not other adverse infant outcomes . Different rates of placental transfer of TNFi and timing of drug exposure in the second and/or third trimester of pregnancy influenced previous advice regarding avoidance of live vaccines in the first 7 months of life. Previously, a small cohort study found CZP (a PEGylated Fab’ fragment, lacking the Fc region) to have minimal rates of placental transfer compared with INF and ADA , and two case reports demonstrated very low rates of placental transfer of ETA administered throughout pregnancy . Therefore, previous recommendations described discontinuation of ADA and ETA at the end of the second trimester to ensure negligible or no drug is detectable in cord blood at delivery. For INF, due to its prolonged bioavailability and higher rate of placental transfer, it was recommended to be stopped earlier in pregnancy (at 16 weeks) for it to be undetectable in cord blood at delivery. Since the last guideline, we found increased data demonstrating different rates of placental transfer of TNFi [ , , , , ]. In a study of pregnant women with IBD exposed to INF ( n = 44) or ADA ( n = 36), the median time to drug clearance was 4 months for ADA and 7 months for INF . In this study, continuation of TNFi in the third trimester did not increase the risk of childhood infection, relative to discontinuation before the third trimester. In a prospective study of ADA ( n = 58) and INF-exposed pregnancies ( n = 73) in women with IBD, cord blood samples showed significantly higher levels of INF than ADA at birth . In this study, placental transfer of INF increased exponentially over the third trimester, while ADA transportation was limited and increased in a linear fashion. Maternal and birth outcomes were comparable between these groups, as were one-year infant health outcomes, including infection and adverse reactions to vaccinations. A study of infants ( n = 14) with third trimester maternal exposure to CZP found minimal rates of placental transfer, supporting continuation of this treatment during pregnancy . Recent data, published after our literature search and highlighted by UCB Pharma during public consultation, reported no signal for adverse pregnancy outcomes following maternal CZP exposure in a large ( n = 1425) cohort . UKTIS reports that studies that investigate the use of TNFi (ADA, CZP, ETA and INF only) during pregnancy have not found an overall increased risk of congenital malformation for these therapies as a class; there is also no compelling evidence of an increased risk for spontaneous miscarriage, intrauterine death or adverse neurodevelopmental outcomes; however, data are currently too limited to exclude adverse effects on the foetus. LBW and preterm birth have been associated with in-utero TNFi exposure in some studies but are confounded by maternal disease. UKTIS also states that there are theoretical concerns that the use of immunosuppressant antibodies, which actively cross the placenta, may result in neonatal or infant immunosuppression and increase the risk of infection; therefore, a delay is advised in administration of live vaccines to infants of: 5 months after last dose of ADA; 16 weeks after last dose of ETA; and until 6 months of age after in-utero INF. In contrast, as CZP is minimally transferred across the placenta, it is unlikely that infants born to women who used CZP in pregnancy would experience sufficient levels of TNFα inhibition to significantly inhibit their immune response . Previously, we found case reports and case series reporting detection of ADA in breastmilk but not in infant serum , and detection of ETA and INF in breastmilk in some [ , , , ] but not all studies , with no adverse effects detectable in any of these breastfed infants. We found nine additional studies of TNFi reporting on 133 breastmilk exposures [ , , , , , , , , ]. Overall, these studies did not find any adverse effects of TNFi. A study of mothers ( n = 17) breastfeeding while taking CZP found minimal transfer of this drug into breastmilk . Another study reported low or undetectable concentrations of INF, ADA, CZP, GOL and UST in 72 breastmilk samples . In this study, breastfed infants on biologics ( n = 243), thiopurines ( n = 102) or combination therapy ( n = 67) were found to have similar risks of infection and rates of milestone achievement compared with infants unexposed to these drugs via breastmilk or not breastfed. LactMed states that: CZP is excreted into breastmilk in some, but not all, women in small amounts; INF is usually either not detectable in breastmilk or detectable at very low levels; ETA, ADA and GOL are minimally excreted into breastmilk, with all TNFi being predicted to be poorly absorbed by the infant due to large molecular weight of each drug . While some evidence suggests that IgG antibodies may not be digested by the gut in the early neonatal period , other studies demonstrate marked digestion of IgG by the infant gut . Previously, we found long-term follow-up data in children exposed in utero to ETA [ , , , ], ADA and CZP . We found an additional 16 studies reporting long-term follow-up data after exposure to TNFi [ , , , , , , ]. In one study, 196 children with intrauterine exposure to TNFi (ADA, n = 81; INF, n = 115) for maternal IBD were followed up for 5 years, finding no association with long-term adverse health outcomes (including childhood infections and vaccination adverse reactions) when compared with TNFi-unexposed controls . This study included women continuing TNFi during the third trimester, where no increased risk of infection was noted in their offspring. In a retrospective cohort study of children ( n = 388) exposed to TNFi (ADA, n = 164; INF, n = 223; CZP, n = 1) in utero for maternal IBD, the incidence of severe infections was similar to TNFi-unexposed children of IBD mothers after median follow-up of 5 years . Similarly, in another study, the risk of serious or opportunistic infections during the first year of life in live-born infants ( n = 229) exposed to ADA during pregnancy in women with RA or IBD was not significantly different to disease-matched, non-exposed controls and healthy controls . In this study, the risk of infection remained similar when restricting to infants exposed to ADA during the third trimester. Two additional studies, which were not included in our final analysis because they did not report primarily on pregnancy or breastmilk exposure outcomes, described infections in the first 3 years of life in children after in-utero exposure to TNFi (predominantly ETA, ADA and INF) and/or csDMARDs for IRDs, psoriasis and IBD. One study of 493 children exposed in utero to TNFi reported an increased risk of some site-specific infections but not other adverse outcomes within the first year of life only . The other study of 1027 children demonstrated a slightly increased risk of paediatric infections associated with both TNFi and csDMARDs in the first and second year after birth; however, the authors noted that this association was present regardless of third trimester exposure and could also have been confounded by disease severity . Studies of vaccine safety and efficacy were not specifically sought through our systematic literature search but are relevant to consider, because most guidance recommends avoidance of live vaccines up to 6–12 months post-partum in infants exposed to bDMARDs in the second/third trimester . This advice is heavily influenced by the finding that placental transfer of bDMARDs can lead to persistence of drug levels by up to 12 months following in-utero exposure, with a median clearance time of 6 months and longest clearance times for INF. In addition, a fatal case of disseminated TB-like disease had been reported in a 4-month-old infant who had received Bacille Calmette et Guerin (BCG) vaccination following in-utero exposure to INF . This guidance impacts on rotavirus vaccination and, if indicated, the BCG vaccine, while the measles, mumps and rubella (MMR) vaccine (typically given at 12 months) is not affected. These restrictions significantly impact on rotavirus only, since the BCG vaccine may easily be deferred to be given later in life, while the rotavirus course of vaccination must be completed by 24 weeks of age due to risk of intussusception . Currently, the restrictions may be avoided by discontinuing TNFi in the second or early third trimester, several half-lives prior to delivery. Rotavirus continues to be a major cause of acute gastroenteritis in young children and has been estimated to result in >500 000 deaths and 2.4 million hospital admissions worldwide . The rotavirus vaccine is over 85% effective at protecting against severe rotavirus gastroenteritis in the first two years of life . Although current UK guidance recommends avoidance of rotavirus vaccination in infants of mothers exposed to biologics during pregnancy , it refers readers to the green book . This text states that, although the vaccine is a live attenuated virus, with the exception of severe combined immunodeficiency (SCID), the benefit from vaccination may exceed risk in other forms of immunosuppression . There are increasing reports of the safe use of rotavirus vaccination in infants following perinatal exposure to bDMARDs, including INF. A systematic review described cohort studies and case reports of infants ( n = 54) of mothers who received antenatal bDMARDs (mostly TNFi, including INF) who then received rotavirus vaccine without significant adverse effects . The authors of that review recommended that otherwise healthy newborns with a history of perinatal exposure to bDMARDS should receive rotavirus vaccinations as per the recommended schedule, while the BCG vaccine should be withheld in the first year of life. Not all consensus review articles have reached the same conclusion, however, with some recommending avoidance of both rotavirus and BCG vaccinations in infants exposed to bDMARDs in utero in the first 6 months of life unless levels of biological drugs are undetectable , or avoidance of live vaccines until 6–12 months of age in infants exposed to biologics that may cross the placenta at clinically significant levels . A systematic review published in abstract form after our search date evaluated vaccine safety in infants exposed to bDMARDs or tsDMARDs in pregnancy . It identified in-utero exposures to ADA ( n = 326), CZP ( n = 18), ETA ( n = 1), INF ( n = 408), GOL ( n = 1), RTX ( n = 1), TCZ ( n = 3), UST ( n = 1) and no tsDMARD exposures in mostly IBD ( n = 849) pregnancies. Infant vaccination included: BCG ( n = 111) and/or rotavirus ( n = 48) in the first year of life (many <6 months); and MMR at 12 months ( n = 590), 6–9 months ( n = 12) and at 1, 2 or 4 months ( n = 3). Adverse events with BCG vaccination included one death, two large local skin reactions, and one infant with axillary lymphadenopathy. A freedom of information request to the MHRA revealed four further suspected fatal BCG infections in infants exposed to TNFi in utero (INF, n = 2; ADA, n = 1; and unspecified TNFi, n = 1). Adverse effects noted in infants given rotavirus vaccination were mild and at similar frequency to those in biologic-unexposed infants. No complications were reported with MMR vaccination. Overall, the most evidence of clinically harmful effects was found after administration of BCG to infants <3 months of age and after in-utero exposure to INF. In contrast, outcomes following rotavirus (mostly <6 months) and MMR (mostly at a year) vaccinations were reassuring. Notably, disseminated rotavirus infection has not been reported. Other systematic reviews have also evaluated vaccine efficacy, and report adequate vaccination response (measured by antibody levels) following non-live vaccination in infants exposed to TNFi , although there are conflicting reports, with low antibody responses to the Haemophilus influenzae type-B vaccine reported in some infants exposed to INF or ADA . Although our literature search did not assess evidence relating to the peri-operative use of TNFi or other bDMARDs (for example, in the context of caesarean sections), relevant guidance can be found in other BSR guidelines . Recommendations for anti-TNFα medications in pregnancy and breastmilk exposure Women with no/low disease activity established on a TNFi with known placental transfer (INF, ADA, GOL) do not need to be switched to an alternative TNFi with established minimal placental transfer (CZP) either before or during pregnancy (GRADE 1B, SOA 100%). CZP is compatible with all three trimesters of pregnancy, has no to minimal placental transfer compared with other TNFi, and does not require any alteration to the infant vaccination schedule (GRADE 1B, SOA 100%). Women considered to have low risk of disease flare on withdrawal of TNFi in pregnancy could stop INF at 20 weeks, ADA and GOL at 28 weeks, and ETA at 32 weeks so that a full-term infant can have a normal vaccination schedule, with rotavirus vaccination at 8 weeks as per the UK schedule (GRADE 1B, SOA 99.5%). INF, ADA, ETA or GOL may be continued throughout pregnancy to maintain maternal disease control; in these circumstances, live vaccines should be avoided in infants until they are 6 months of age (GRADE 1B, SOA 100%). If a TNFi is stopped in pregnancy, it can be restarted as soon as practical post-partum in the absence of infections or surgical complications, regardless of breastfeeding status, to ensure control of maternal disease (GRADE 1C, SOA 100%). TNFi are compatible with breastmilk exposure (GRADE 1C, SOA 100%).
Five biologic agents that inhibit TNFα (TNFi) are currently licensed to treat IRDs: etanercept (ETA), infliximab (INF), adalimumab (ADA), golimumab (GOL) and certolizumab pegol (CZP). Three of these drugs (INF, ADA and GOL) are monoclonal IgG1 directed against TNFα, one (ETA) is a fusion protein of the TNF receptor joined to the Fc region of IgG1, while CZP is an antigen-binding fragment (Fab′) of a monoclonal anti-TNFα antibody which lacks the Fc region of IgG1 and has been conjugated with polyethylene glycol (PEG). These drugs have different half-lives, bioavailability and rates of placental transfer, which are relevant when considering their potential use in pregnancy. Initial 2006/8 consensus recommendations advised avoidance of ETA, INF and ADA in pregnancy and breastfeeding due to a lack of evidence rather than evidence of harm . Previously, we reviewed outcome data from TNFi-exposed pregnancies ( n = 706) of patients with predominantly IBD but also rheumatic disease and non-autoimmune-mediated recurrent spontaneous miscarriage, compared with ( n = 399) disease and ( n = 170) healthy control pregnancies [ , , , , , , , , , ]. There were multiple confounders of concomitant therapies (including MTX, LEF and MMF) and active inflammatory disease. Overall, these studies did not describe an increased incidence of adverse effects upon miscarriage rates, pregnancy duration, birth weight, foetal death or congenital malformation that was attributable to ETA, INF, ADA or CZP. At that time, there was limited information on placental transfer, and no published studies of GOL in human pregnancy or breastmilk exposure. TNFi exposure in pregnancy and with breastmilk exposure has been extensively studied since our last search. We identified an additional 50 studies, reporting 12 491 pregnancy exposures to TNFi, including INF ( n ≥ 5377), ADA ( n ≥ 2797), ETA ( n ≥ 2210), CZP ( n ≥ 776) and GOL ( n ≥ 196) [ , , ]. Many studies reported combined outcomes for exposure to different TNFi agents. The majority of studies of maternal exposure did not report an increased risk of preterm birth, miscarriage, low birth weight or congenital malformations [ , , , , , , , , , , , ]. Different adverse outcomes were reported in some studies, however. A study of ETA found that the proportion of infants with major birth defects was higher (9.4% vs 3.5%, respectively) in ETA-exposed pregnancies ( n = 370) than in pregnancies of disease-matched, non-exposed women ( n = 164) ; however, the lack of a specific pattern of birth defects and the expected minimal placental transfer of ETA in early pregnancy did not support the biologic plausibility of a drug-related effect. A study reporting a lower live birth rate in INF-exposed pregnancies ( n = 99) in women with Crohn’s disease considered their findings to be confounded by more severe disease in those patients exposed to INF and increased exposure to other immunosuppressive agents . A population-based study of TNFi-exposed pregnancies ( n = 1027) found increased risks of preterm birth, caesarean section and SGA babies in comparison with TNFi-unexposed pregnancies ( n = 9399) ; however, the authors noted that these associations may have been related to underlying disease activity rather than agent-specific effects, due to diverse findings across disease groups. A retrospective cohort study of TNFi-exposed pregnancies ( n = 1457) in women with IBD found TNFi exposure to be an independent risk factor for maternal complications and infections when compared with TNFi-unexposed pregnancies ( n = 9818) . In this study, TNFi exposure did not associate with congenital malformations or an increased risk of infection in children during the first year of life. Furthermore, there was no difference in the risk of complications in women exposed to TNFi during the third trimester, relative to cessation before week 24, although disease relapses were more common in those stopping TNFi prior to the third trimester. A study of 4961 pregnant women with autoimmune inflammatory conditions found similar risks of serious infections in women taking steroids, csDMARDs or TNFi during pregnancy, but found that higher doses of steroids were an independent risk factor for serious infections in pregnancy . A registry-based study from Denmark and Sweden reported a non-statistically significant higher risk of having children with birth defects in women with RA, AS, psoriatic arthritis (PsA), IBD or psoriasis who had received TNFi during pregnancy ( n = 683), relative to women with chronic inflammatory disease but without TNFi exposure ( n = 21 549) ; however, the heterogeneity of observed birth defects went against a common aetiology. A prospective cohort study of TNFi pregnancy exposures ( n = 495) in women with chronic inflammatory disease (RA, AS, PsA, psoriasis and IBD) found prenatal TNFi exposure to be associated with an increased risk of birth defects without a distinct pattern of malformations, when compared with non-disease-matched, TNFi-unexposed controls ( n = 1532) . An increased risk of preterm births and reduced birth weight, but not spontaneous miscarriage, was also noted. The authors concluded that, although TNFi may carry a risk of adverse pregnancy outcomes of moderate clinical relevance, they may remain a treatment option, considering the impact of inadequately controlled disease on the mother and unborn child. A small number of studies specifically compared risks between TNFi agents. In one study, INF was found to be associated with a greater risk for preterm births relative to ETA, and a higher prevalence of severely SGA babies relative to ETA and ADA, in pregnant women with RA, AS, PsA or psoriasis. In IBD, however, the risk of preterm births and SGA babies did not differ between INF and ADA . In a study of individual safety reports in pregnant IBD patients exposed to TNFi ( n = 783), the odds for maternal or foetal adverse events were found to be lower for CZP monotherapy, but not for INF or ADA monotherapy, when compared with an aminosalicylate monotherapy comparator in multi-level regression models . In another study, the risk of birth defects did not differ significantly between ADA, INF or ETA-exposed women with chronic inflammatory diseases . Several studies specifically compared outcomes for pregnancies exposed to TNFi during late vs early trimesters [ , , , , , , , , , , , ], the majority of which reported no significant concerns with late trimester exposure. In a study of INF-exposed pregnancies ( n = 1850) in women with IBD, RA, AS, PsA and psoriasis, frequencies of congenital abnormalities and other adverse birth and infant outcomes (including neonatal infections) were similar when comparing first and third trimester exposure . In a study comparing early discontinuation of INF (>90 days before delivery; n = 68) to late discontinuation (<90 days before delivery; n = 318) in pregnancies of women with IBD, early discontinuation was associated with increased disease flares, more steroid usage and more preterm births than late discontinuation . Rates of other adverse outcomes, including congenital malformations and infant respiratory infections, were similar between these groups. A further study of TNFi-exposed pregnancies ( n = 153) in women with IBD reported that continuation of TNFi after gestational week 30, relative to cessation before week 30, was independently associated with modestly lower birth weights in multivariate regression models after adjustment for disease activity, but not other adverse infant outcomes . Different rates of placental transfer of TNFi and timing of drug exposure in the second and/or third trimester of pregnancy influenced previous advice regarding avoidance of live vaccines in the first 7 months of life. Previously, a small cohort study found CZP (a PEGylated Fab’ fragment, lacking the Fc region) to have minimal rates of placental transfer compared with INF and ADA , and two case reports demonstrated very low rates of placental transfer of ETA administered throughout pregnancy . Therefore, previous recommendations described discontinuation of ADA and ETA at the end of the second trimester to ensure negligible or no drug is detectable in cord blood at delivery. For INF, due to its prolonged bioavailability and higher rate of placental transfer, it was recommended to be stopped earlier in pregnancy (at 16 weeks) for it to be undetectable in cord blood at delivery. Since the last guideline, we found increased data demonstrating different rates of placental transfer of TNFi [ , , , , ]. In a study of pregnant women with IBD exposed to INF ( n = 44) or ADA ( n = 36), the median time to drug clearance was 4 months for ADA and 7 months for INF . In this study, continuation of TNFi in the third trimester did not increase the risk of childhood infection, relative to discontinuation before the third trimester. In a prospective study of ADA ( n = 58) and INF-exposed pregnancies ( n = 73) in women with IBD, cord blood samples showed significantly higher levels of INF than ADA at birth . In this study, placental transfer of INF increased exponentially over the third trimester, while ADA transportation was limited and increased in a linear fashion. Maternal and birth outcomes were comparable between these groups, as were one-year infant health outcomes, including infection and adverse reactions to vaccinations. A study of infants ( n = 14) with third trimester maternal exposure to CZP found minimal rates of placental transfer, supporting continuation of this treatment during pregnancy . Recent data, published after our literature search and highlighted by UCB Pharma during public consultation, reported no signal for adverse pregnancy outcomes following maternal CZP exposure in a large ( n = 1425) cohort . UKTIS reports that studies that investigate the use of TNFi (ADA, CZP, ETA and INF only) during pregnancy have not found an overall increased risk of congenital malformation for these therapies as a class; there is also no compelling evidence of an increased risk for spontaneous miscarriage, intrauterine death or adverse neurodevelopmental outcomes; however, data are currently too limited to exclude adverse effects on the foetus. LBW and preterm birth have been associated with in-utero TNFi exposure in some studies but are confounded by maternal disease. UKTIS also states that there are theoretical concerns that the use of immunosuppressant antibodies, which actively cross the placenta, may result in neonatal or infant immunosuppression and increase the risk of infection; therefore, a delay is advised in administration of live vaccines to infants of: 5 months after last dose of ADA; 16 weeks after last dose of ETA; and until 6 months of age after in-utero INF. In contrast, as CZP is minimally transferred across the placenta, it is unlikely that infants born to women who used CZP in pregnancy would experience sufficient levels of TNFα inhibition to significantly inhibit their immune response . Previously, we found case reports and case series reporting detection of ADA in breastmilk but not in infant serum , and detection of ETA and INF in breastmilk in some [ , , , ] but not all studies , with no adverse effects detectable in any of these breastfed infants. We found nine additional studies of TNFi reporting on 133 breastmilk exposures [ , , , , , , , , ]. Overall, these studies did not find any adverse effects of TNFi. A study of mothers ( n = 17) breastfeeding while taking CZP found minimal transfer of this drug into breastmilk . Another study reported low or undetectable concentrations of INF, ADA, CZP, GOL and UST in 72 breastmilk samples . In this study, breastfed infants on biologics ( n = 243), thiopurines ( n = 102) or combination therapy ( n = 67) were found to have similar risks of infection and rates of milestone achievement compared with infants unexposed to these drugs via breastmilk or not breastfed. LactMed states that: CZP is excreted into breastmilk in some, but not all, women in small amounts; INF is usually either not detectable in breastmilk or detectable at very low levels; ETA, ADA and GOL are minimally excreted into breastmilk, with all TNFi being predicted to be poorly absorbed by the infant due to large molecular weight of each drug . While some evidence suggests that IgG antibodies may not be digested by the gut in the early neonatal period , other studies demonstrate marked digestion of IgG by the infant gut . Previously, we found long-term follow-up data in children exposed in utero to ETA [ , , , ], ADA and CZP . We found an additional 16 studies reporting long-term follow-up data after exposure to TNFi [ , , , , , , ]. In one study, 196 children with intrauterine exposure to TNFi (ADA, n = 81; INF, n = 115) for maternal IBD were followed up for 5 years, finding no association with long-term adverse health outcomes (including childhood infections and vaccination adverse reactions) when compared with TNFi-unexposed controls . This study included women continuing TNFi during the third trimester, where no increased risk of infection was noted in their offspring. In a retrospective cohort study of children ( n = 388) exposed to TNFi (ADA, n = 164; INF, n = 223; CZP, n = 1) in utero for maternal IBD, the incidence of severe infections was similar to TNFi-unexposed children of IBD mothers after median follow-up of 5 years . Similarly, in another study, the risk of serious or opportunistic infections during the first year of life in live-born infants ( n = 229) exposed to ADA during pregnancy in women with RA or IBD was not significantly different to disease-matched, non-exposed controls and healthy controls . In this study, the risk of infection remained similar when restricting to infants exposed to ADA during the third trimester. Two additional studies, which were not included in our final analysis because they did not report primarily on pregnancy or breastmilk exposure outcomes, described infections in the first 3 years of life in children after in-utero exposure to TNFi (predominantly ETA, ADA and INF) and/or csDMARDs for IRDs, psoriasis and IBD. One study of 493 children exposed in utero to TNFi reported an increased risk of some site-specific infections but not other adverse outcomes within the first year of life only . The other study of 1027 children demonstrated a slightly increased risk of paediatric infections associated with both TNFi and csDMARDs in the first and second year after birth; however, the authors noted that this association was present regardless of third trimester exposure and could also have been confounded by disease severity . Studies of vaccine safety and efficacy were not specifically sought through our systematic literature search but are relevant to consider, because most guidance recommends avoidance of live vaccines up to 6–12 months post-partum in infants exposed to bDMARDs in the second/third trimester . This advice is heavily influenced by the finding that placental transfer of bDMARDs can lead to persistence of drug levels by up to 12 months following in-utero exposure, with a median clearance time of 6 months and longest clearance times for INF. In addition, a fatal case of disseminated TB-like disease had been reported in a 4-month-old infant who had received Bacille Calmette et Guerin (BCG) vaccination following in-utero exposure to INF . This guidance impacts on rotavirus vaccination and, if indicated, the BCG vaccine, while the measles, mumps and rubella (MMR) vaccine (typically given at 12 months) is not affected. These restrictions significantly impact on rotavirus only, since the BCG vaccine may easily be deferred to be given later in life, while the rotavirus course of vaccination must be completed by 24 weeks of age due to risk of intussusception . Currently, the restrictions may be avoided by discontinuing TNFi in the second or early third trimester, several half-lives prior to delivery. Rotavirus continues to be a major cause of acute gastroenteritis in young children and has been estimated to result in >500 000 deaths and 2.4 million hospital admissions worldwide . The rotavirus vaccine is over 85% effective at protecting against severe rotavirus gastroenteritis in the first two years of life . Although current UK guidance recommends avoidance of rotavirus vaccination in infants of mothers exposed to biologics during pregnancy , it refers readers to the green book . This text states that, although the vaccine is a live attenuated virus, with the exception of severe combined immunodeficiency (SCID), the benefit from vaccination may exceed risk in other forms of immunosuppression . There are increasing reports of the safe use of rotavirus vaccination in infants following perinatal exposure to bDMARDs, including INF. A systematic review described cohort studies and case reports of infants ( n = 54) of mothers who received antenatal bDMARDs (mostly TNFi, including INF) who then received rotavirus vaccine without significant adverse effects . The authors of that review recommended that otherwise healthy newborns with a history of perinatal exposure to bDMARDS should receive rotavirus vaccinations as per the recommended schedule, while the BCG vaccine should be withheld in the first year of life. Not all consensus review articles have reached the same conclusion, however, with some recommending avoidance of both rotavirus and BCG vaccinations in infants exposed to bDMARDs in utero in the first 6 months of life unless levels of biological drugs are undetectable , or avoidance of live vaccines until 6–12 months of age in infants exposed to biologics that may cross the placenta at clinically significant levels . A systematic review published in abstract form after our search date evaluated vaccine safety in infants exposed to bDMARDs or tsDMARDs in pregnancy . It identified in-utero exposures to ADA ( n = 326), CZP ( n = 18), ETA ( n = 1), INF ( n = 408), GOL ( n = 1), RTX ( n = 1), TCZ ( n = 3), UST ( n = 1) and no tsDMARD exposures in mostly IBD ( n = 849) pregnancies. Infant vaccination included: BCG ( n = 111) and/or rotavirus ( n = 48) in the first year of life (many <6 months); and MMR at 12 months ( n = 590), 6–9 months ( n = 12) and at 1, 2 or 4 months ( n = 3). Adverse events with BCG vaccination included one death, two large local skin reactions, and one infant with axillary lymphadenopathy. A freedom of information request to the MHRA revealed four further suspected fatal BCG infections in infants exposed to TNFi in utero (INF, n = 2; ADA, n = 1; and unspecified TNFi, n = 1). Adverse effects noted in infants given rotavirus vaccination were mild and at similar frequency to those in biologic-unexposed infants. No complications were reported with MMR vaccination. Overall, the most evidence of clinically harmful effects was found after administration of BCG to infants <3 months of age and after in-utero exposure to INF. In contrast, outcomes following rotavirus (mostly <6 months) and MMR (mostly at a year) vaccinations were reassuring. Notably, disseminated rotavirus infection has not been reported. Other systematic reviews have also evaluated vaccine efficacy, and report adequate vaccination response (measured by antibody levels) following non-live vaccination in infants exposed to TNFi , although there are conflicting reports, with low antibody responses to the Haemophilus influenzae type-B vaccine reported in some infants exposed to INF or ADA . Although our literature search did not assess evidence relating to the peri-operative use of TNFi or other bDMARDs (for example, in the context of caesarean sections), relevant guidance can be found in other BSR guidelines . Recommendations for anti-TNFα medications in pregnancy and breastmilk exposure Women with no/low disease activity established on a TNFi with known placental transfer (INF, ADA, GOL) do not need to be switched to an alternative TNFi with established minimal placental transfer (CZP) either before or during pregnancy (GRADE 1B, SOA 100%). CZP is compatible with all three trimesters of pregnancy, has no to minimal placental transfer compared with other TNFi, and does not require any alteration to the infant vaccination schedule (GRADE 1B, SOA 100%). Women considered to have low risk of disease flare on withdrawal of TNFi in pregnancy could stop INF at 20 weeks, ADA and GOL at 28 weeks, and ETA at 32 weeks so that a full-term infant can have a normal vaccination schedule, with rotavirus vaccination at 8 weeks as per the UK schedule (GRADE 1B, SOA 99.5%). INF, ADA, ETA or GOL may be continued throughout pregnancy to maintain maternal disease control; in these circumstances, live vaccines should be avoided in infants until they are 6 months of age (GRADE 1B, SOA 100%). If a TNFi is stopped in pregnancy, it can be restarted as soon as practical post-partum in the absence of infections or surgical complications, regardless of breastfeeding status, to ensure control of maternal disease (GRADE 1C, SOA 100%). TNFi are compatible with breastmilk exposure (GRADE 1C, SOA 100%).
Women with no/low disease activity established on a TNFi with known placental transfer (INF, ADA, GOL) do not need to be switched to an alternative TNFi with established minimal placental transfer (CZP) either before or during pregnancy (GRADE 1B, SOA 100%). CZP is compatible with all three trimesters of pregnancy, has no to minimal placental transfer compared with other TNFi, and does not require any alteration to the infant vaccination schedule (GRADE 1B, SOA 100%). Women considered to have low risk of disease flare on withdrawal of TNFi in pregnancy could stop INF at 20 weeks, ADA and GOL at 28 weeks, and ETA at 32 weeks so that a full-term infant can have a normal vaccination schedule, with rotavirus vaccination at 8 weeks as per the UK schedule (GRADE 1B, SOA 99.5%). INF, ADA, ETA or GOL may be continued throughout pregnancy to maintain maternal disease control; in these circumstances, live vaccines should be avoided in infants until they are 6 months of age (GRADE 1B, SOA 100%). If a TNFi is stopped in pregnancy, it can be restarted as soon as practical post-partum in the absence of infections or surgical complications, regardless of breastfeeding status, to ensure control of maternal disease (GRADE 1C, SOA 100%). TNFi are compatible with breastmilk exposure (GRADE 1C, SOA 100%).
Data relating to non-TNFi biologic use in pregnancy remain scarce, and in this guideline updated case reports and small case series were excluded for consistency. A recent systematic review summarized all available data (including case reports) on non-TNFi bDMARDs and tsDMARDs and did not identify any adverse safety signals . Rituximab Previously, we found insufficient evidence to be confident that RTX is compatible with pregnancy and recommended it should be stopped 6 months before conception, although there were no direct reports of teratogenicity and only second/third trimester exposure was associated with neonatal B-cell depletion. RTX is a monoclonal IgG1 that actively crosses the placenta from 16 weeks of pregnancy onwards. Previously, we found eight studies on 173 RTX-exposed pregnancies that met our inclusion criteria. Overall, these studies reported reassuring pregnancy outcomes, but found low B cells at birth in infants exposed to RTX in the second/third trimester, no human breastmilk exposure studies were identified . We identified five further studies of 143 RTX pregnancy exposures in addition to eight exposures in studies that reported on combined outcomes of multiple biologic-exposed pregnancies (see ‘Biologic DMARDs’ above). Four studies ( n = 135) of mostly pre-conception or first trimester maternal exposure did not report an increased risk of preterm birth, miscarriage, LBW or congenital anomaly . An additional cohort study of eight mothers treated with one to four cycles of RTX during pregnancy ( n = 6 for diffuse large B-cell lymphoma; n = 2 for SLE) found a high rate of preterm birth and intrauterine infections, but noted the potential for confounding by the underlying disease and concomitant medications . UKTIS reports that there is insufficient evidence to assess whether the risk for spontaneous miscarriage, congenital malformation, birth weight, intrauterine death or adverse neurodevelopmental outcomes is increased following exposure to RTX in utero . Due to a lack of data on the effects of RTX on the neonatal/infant immune system, it recommends delaying administration of live vaccines following in-utero exposure to RTX, although no specific timescale is given . There were three studies of breastmilk exposure ( n = 23) to RTX [ , , ]. One study detected minimal transfer of RTX into the breastmilk of nine breastfeeding mothers, with a relative infant dose of rituximab well below theoretically acceptable levels (<0.4%) . No adverse effects attributable to RTX exposure were reported in the breastfed infants, and the drug was considered to have an acceptable benefit-to-risk ratio, supporting both maternal treatment and breastmilk exposure. LactMed states that the amount of RTX in breastmilk is very low and absorption is unlikely because it is a protein with a high molecular weight that is likely to be partially destroyed in the infant’s gastrointestinal tract; thus, absorption by the infant is probably minimal . Vaccine efficacy after prenatal exposure to RTX is much less studied, and systematic reviews have identified adequate non-live vaccination outcomes from 5/6 RTX-exposed infants, with low immunity to diphtheria detected in one 11-month-old infant exposed to RTX at conception . Recommendations for rituximab in pregnancy and breastmilk exposure Limited evidence has not shown RTX to be teratogenic; however, there remains insufficient evidence to be confident that it is compatible with pregnancy. Consider stopping the drug at conception (GRADE 2C, SOA 99.3%). RTX may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 99.7%). If RTX is used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 98.7%). Based on limited evidence, maternal treatment with RTX is compatible with breastmilk exposure (GRADE 2C, SOA 99.5%). Interleukin-6 inhibitors Tocilizumab Due to insufficient data, we previously recommended that TCZ be stopped at least three months pre-conception, but that unintentional exposure early in the first trimester is unlikely to be harmful due to its IgG1 structure. This limited data came from outcomes of 33 pregnancies in 32 patients, published in abstract form . We identified three further studies of 365 pregnancy exposures to TCZ in addition to four exposures in the papers described above which reported combined outcomes of multiple biologic-exposed pregnancies (see ‘DMARDs’ above). These studies of mainly first trimester exposure did not find increased rates of congenital abnormalities in patients with rheumatic disease. Two studies commented on a higher than background rate of spontaneous miscarriage , and could not exclude an effect on birth weight and risk of prematurity, while acknowledging potential confounding factors. For example, stopping effective treatment early in pregnancy could destabilize the rheumatic disease, with adverse consequences for the pregnancy. UKTIS notes a small number of studies lacking comparator groups that are confounded by maternal disease, and states that there is currently no compelling evidence that TCZ is teratogenic or fetotoxic. Due to lack of data on effects on the neonatal/infant immune system, it recommends that live vaccines are avoided until the infant is 6 months of age . One study of n = 2 breastmilk exposure to TCZ did not report any adverse effects attributable to the drug . Saito et al. have described a total of four cases of babies exposed to TCZ via breastmilk with no complications in the infants (three publications which did not meet our inclusion criteria) . Concentrations of TCZ were measured in the breastmilk in these studies and two additional cases reports, with no clinical adverse effects reported . Levels in the breastmilk were found to peak on day three following the infusion , but were significantly lower than the corresponding maternal serum concentrations in all of these cases, ranging from 11% in colostrum , down to as low as 1:2000 . Sarilumab No papers were found that met our inclusion criteria. Sanofi provided data on 13 patients who became pregnant in the sarilumab and DMARD long-term safety population, of whom seven had a miscarriage . In addition, two male patients fathered two healthy children. Recommendations for IL-6 inhibitors in pregnancy and breastmilk exposure Limited evidence has not shown IL-6 inhibitors (IL-6i) to be teratogenic; however, there remains insufficient evidence to be confident that they are compatible with pregnancy. Consider stopping the drug at conception. Any exposure, however, during pregnancy is unlikely to be harmful (GRADE 2C, SOA 99.7%). IL-6i may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 100%). If IL-6i are used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 99.3%). Based on limited evidence, maternal treatment with IL-6i is compatible with breastmilk exposure (GRADE 2C, SOA 100%). Interleukin-1 inhibitors Anakinra Anakinra is a recombinant form of human interleukin-1 receptor antagonist (IL-1Ra) with a high molecular weight which was not found to cross ex-vivo full-term human placentae . Previously, there was insufficient information evidence on which to base a recommendation for anakinra in pregnancy, but unintentional exposure in the first trimester was considered unlikely to be harmful . This limited evidence came from reports on five pregnancies in three studies/reports, with no evidence of harm [ , , ]. Subsequently, we identified four studies of 43 pregnancy exposures to anakinra , in addition to one exposure in the papers described above which reported combined outcomes of multiple biologic-exposed pregnancies (see ‘Biologic DMARDs’ above). These exposures were mostly in patients with periodic fevers and severe maternal disease. Overall, outcomes were reassuring, although two congenital renal anomalies and two cases of oligohydramnios (which can be linked to foetal renal anomalies) were reported [ , , ]. It was unclear, however, whether the renal abnormalities were associated with antenatal anakinra use or maternal hyperthermia or both. Given the significant beneficial effects of anakinra in suppressing maternal disease with limited pregnancy-compatible options, it was considered a safe alternative in managing disease in women with periodic fever in pregnancy. UKTIS does not report on anakinra. Two studies of n = 12 breastmilk exposures to anakinra did not report any adverse effects attributable to the drug . LactMed states that IL-1Ra is a normal component of human milk, possibly as an anti-inflammatory agent, and that several infants have been breastfed during maternal anakinra therapy with no obvious adverse effects. If anakinra is required by the mother, it is not a reason to discontinue breastfeeding . Canakinumab Canakinumab is a human monoclonal antibody to IL-1. There is, therefore, the potential for active transport across the placenta from the second trimester onwards. One paper reporting canakinumab use from pre-conception in eight pregnancies (stopped during first trimester in 5/8 pregnancies) met our inclusion criteria . There were seven live births, all of whom were healthy, full term and normal birth weight. One mother with refractory Cogan syndrome had an early miscarriage at 6 weeks (after a miscarriage the previous year while on anakinra). In addition, Novartis supplied information from analyses of the Novartis Global Safety Database, which included 76 maternal and nine paternal exposures . There were 47 known pregnancy outcomes, with 27 healthy newborns, one preterm birth (with meconium staining not thought likely to be related to canakinumab exposure), nine spontaneous miscarriages (including two miscarriages related to paternal exposure in one father) and three elective terminations. Seven other adverse events were reported: one case of congenital pyelocaliectasis following paternal exposure to canakinumab; one case of congenital musculoskeletal abnormality following maternal exposure, with insufficient data on other contributory factors; one case of inherited genetic disease with neonatal RSV infection and pyrexia following infant vaccinations; and four adverse events very unlikely to relate to canakinumab exposure (one newborn non-serious hypotension and three cases of likely inherited disease: Muckle–Wells syndrome and CAPS). One cohort study reported breastmilk exposure in four infants whose mothers were prescribed regular canakinumab, with no reported serious infections or developmental abnormalities . Recommendations for IL-1 inhibitors in pregnancy and breastmilk exposure Limited evidence has not shown IL-1 inhibitors (IL-1i) to be teratogenic; however, there remains insufficient evidence to be confident that they are compatible with pregnancy. Consider stopping the drug at conception. Any exposure, however, during pregnancy is unlikely to be harmful (GRADE 2C, SOA 99.8%). IL-1i may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 100%). If IL-1i are used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 99.3%). Based on limited evidence, maternal treatment with IL-1i is compatible with breastmilk exposure (GRADE 2C, SOA 100%). Abatacept ABA is a fusion protein containing the Fc region of IgG1 fused to the extracellular domain of CTLA-4; therefore, it is able to cross the placental barrier from approximately week 16. Previously, there was insufficient data to recommend ABA in pregnancy, but unintentional exposure early in the first trimester was considered unlikely to be harmful . This conclusion was based on reports from 11 pregnancy exposures (and a further eight exposures reported in abstract form only), many of which were confounded by concomitant MTX [ , , ]. We identified two further studies of at least 196 pregnancy exposures to ABA (with some overlap of data) , in addition to three exposures in the papers described above which reported combined outcomes of multiple biologic-exposed pregnancies (see ‘Biologic DMARDs’ above). The data from these studies did not suggest an increased risk of adverse pregnancy outcomes with ABA, and many of the congenital abnormalities were considered by the authors to be associated with concomitant use of other teratogenic DMARDs. The apparently high rate of spontaneous miscarriage (total ≥48/184) is difficult to interpret due to confounding by indication and other medications such as MTX, as well as detection bias of early miscarriages during the close monitoring of clinical trials . The authors concluded that ABA should only be used during pregnancy if the benefit to the mother justifies potential risk to the foetus. UKTIS does not report on ABA. Although it did not meet our inclusion criteria, we identified one case report of breastmilk exposure to ABA, with no adverse effects, in which ABA was secreted into breastmilk at levels 1/200–1/300 of those in serum . LactMed describes one case report that reported concentrations of ABA in milk were very low and did not appear to affect the breastfed infant, concluding that if ABA is required by the mother, it is not a reason to discontinue breastfeeding, although alternative drugs may be preferred . Recommendations for abatacept in pregnancy and breastmilk exposure Limited evidence has not shown ABA to be teratogenic; however, there remains insufficient evidence to be confident that it is compatible with pregnancy. Consider stopping the drug at conception. Any exposure, however, during pregnancy is unlikely to be harmful (GRADE 2C, SOA 99.3%). ABA may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 99.3%). If ABA is used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 99.3%). Based on limited evidence, maternal treatment with ABA is compatible with breastmilk exposure (GRADE 2C, SOA 99.5%). Belimumab Belimumab (BEL) is a fully humanized monoclonal IgG1 that inhibits B-cell activating factor. Previously, our systematic search did not identify any publications that met our search criteria on use of BEL in human pregnancy, although we found reference to pregnancy outcomes from placebo-controlled, phase 2 and 3 studies in 83 SLE pregnancies , and BEL pregnancy registry data describing known outcomes from 118 SLE pregnancies . Overall, these studies did not identify any pattern of adverse effects in pregnancy directly attributable to BEL. We subsequently identified one study of 66 BEL pregnancy exposures . Total foetal losses in BEL-treated subjects were similar to background estimates in SLE patients (∼25%), although data remain limited. Another six studies, which did not meet our search criteria , reported an additional 124 pregnancies exposed to BEL. There was only one reported congenital abnormality: a case report of mild Ebstein’s anomaly in a baby following successful control of SLE (with previous lupus nephritis and APS, previously only controlled on MMF) . Two case series (26 pregnancy exposures) reported preterm delivery in 12/22 live births, with six babies being SGA . However, these cohorts included patients with complex rheumatic disease (SLE and APS), including some patients with previous lupus nephritis and active disease at conception , and a high average maternal age and rate of previous pregnancy losses . Both authors concluded that, while careful consideration and further research is required, BEL could be a reasonable treatment option for patients with SLE requiring treatment in pregnancy. UKTIS does not report on BEL. Although they did not meet our inclusion criteria, we identified two cases of breastmilk exposure while on BEL . No neonatal outcomes, adverse or otherwise, were reported in one case . In the other case, there were no adverse effects, and BEL was secreted into breastmilk at levels 1/200–1/500 of those in serum . LactMed states that, based on preliminary information that BEL levels in breastmilk are very low and infant absorption is probably minimal, if BEL is required by the mother, it is not a reason to discontinue breastfeeding, although caution is required, especially while nursing a newborn or preterm infant . Recommendations for belimumab in pregnancy and breastmilk exposure Limited evidence has not shown BEL to be teratogenic; however, there remains insufficient evidence to be confident that it is compatible with pregnancy. Consider stopping the drug at conception. Any exposure, however, during pregnancy is unlikely to be harmful (GRADE 2C, SOA 99.3%). BEL may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 99.5%). If BEL is used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 98.8%). Based on limited evidence, maternal treatment with BEL is compatible with breastmilk exposure (GRADE 2C, SOA 99.5%). Interleukin-17 inhibitors Interleukin-17 inhibitors (IL-17i) were not in general use and so not considered in our previous search. Secukinumab has an IgG1 structure, which may theoretically suggest increased transplacental transfer compared with the IgG4 structure of ixekizumab. We found three studies of mostly first trimester pregnancy exposures to secukinumab ( n = 244) and ixekizumab ( n = 18) . The authors did not report any increased incidence of adverse outcomes directly attributable to the drugs, although information, particularly for ixekizumab, remains very limited. In addition, Novartis provided data from a search of their Global Safety Database, relating to 298 reports of maternal pregnancy exposures and 90 paternal exposures to secukinumab, mostly during the first trimester . No outcome data were provided, but analysis within Novartis did not reveal any new safety information. UKTIS does not report on these drugs. There were no data on breastmilk exposure to IL-17i identified in our search or in LactMed. Data provided from the Novartis Global Safety Database identified six breastmilk exposures to secukinumab, including one report of newborn pyrexia during breastfeeding . The data in all cases were too limited to draw conclusions, but no new safety concerns were inferred. Recommendations for interleukin-17 inhibitors in pregnancy and breastmilk exposure Limited evidence has not shown IL-17i to be teratogenic; however, there remains insufficient evidence to be confident that they are compatible with pregnancy. Consider stopping the drug at conception. Any exposure, however, during pregnancy is unlikely to be harmful (GRADE 2C, SOA 99.3%). IL-17i may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 99%). If IL-17i are used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 99.3%). Based on limited evidence, maternal treatment with IL-17i is compatible with breastmilk exposure (GRADE 2C, SOA 99.5%). Interleukin-12/23 inhibitors UST is an interleukin-12/23 inhibitor (IL-12/23i) with an IgG1 structure. UST was not in general use and not considered in our previous search. We found three studies of mostly first and second trimester pregnancy exposures to UST ( n = 517) for maternal psoriasis, PsA and IBD . Overall, these data showed that the rates of live births, spontaneous miscarriages and congenital anomalies were consistent with the general population and TNFi-exposed pregnancies. Similarly, UKTIS did not identify any specific drug-related risk but was unable to provide a reliable evidence-based evaluation of risk due to limited information . Although they did not meet our inclusion criteria, we identified two reports of exposure to UST during breastmilk exposure . In one case report, the trough level of UST in breastmilk after restarting this drug post-partum in a Crohn’s disease patient was initially in the same range as the corresponding serum trough level, and then decreased during maintenance therapy . In another study assessing breastfeeding mothers taking a range of biologic medication for IBD, UST was detected at low levels in 4/6 mothers taking this medication . Overall, in this study, breastfed infants of mothers on biologics were found to have similar risks of infection and rates of milestone achievement compared with non-breastfed infants or infants unexposed to these drugs. LactMed states that UST is either not detectable or found at very low levels in breastmilk and infant absorption is probably minimal; as such, if UST is required by the mother, it is not a reason to discontinue breastfeeding, although caution is required especially while nursing a newborn or preterm infant . Recommendations for interleukin-12/23 inhibitors in pregnancy and breastmilk exposure Limited evidence has not shown UST to be teratogenic; however, there remains insufficient evidence to be confident that it is compatible with pregnancy. Consider stopping the drug at conception. Any exposure, however, during pregnancy is unlikely to be harmful (GRADE 2C, SOA 99.3%). UST may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 98.8%). If UST is used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 99.3%). Based on limited evidence, maternal treatment with UST is compatible with breastmilk exposure (GRADE 2C, SOA 99.5%). Anifrolumab Anifrolumab is a fully human, IgG1κ monoclonal antibody to type I interferon receptor subunit 1. Although it is not currently licensed for use in the UK, anifrolumab is approved for use in other countries, including FDA approval to treat moderate-to-severe SLE in the USA . Although there is no published evidence on the use of this drug in pregnancy, breastmilk or paternal exposure, further information was provided by AstraZeneca on data from completed trials. Despite the mandatory contraception use in anifrolumab clinical trials, pregnancies have been reported in 20 SLE patients receiving anifrolumab in studies completed by December 2021. Patients who became pregnant during studies had to immediately discontinue the investigational product. No anifrolumab-associated congenital abnormalities or anifrolumab-associated adverse events have been observed in the clinical trials. Current data on anifrolumab pregnancy exposure are insufficient to inform about potential drug-related risks and therefore post-authorisation pregnancy studies are planned. UKTIS does not report on anifrolumab. Although there is no information regarding breastmilk exposure to anifrolumab, LactMed states that due to its large molecular weight, the amount in milk is likely to be very low, and it is also likely to be partially destroyed in the infant's gastrointestinal tract with minimal absorption by the infant . Based on this limited evidence, we have not made any recommendations on this drug.
Previously, we found insufficient evidence to be confident that RTX is compatible with pregnancy and recommended it should be stopped 6 months before conception, although there were no direct reports of teratogenicity and only second/third trimester exposure was associated with neonatal B-cell depletion. RTX is a monoclonal IgG1 that actively crosses the placenta from 16 weeks of pregnancy onwards. Previously, we found eight studies on 173 RTX-exposed pregnancies that met our inclusion criteria. Overall, these studies reported reassuring pregnancy outcomes, but found low B cells at birth in infants exposed to RTX in the second/third trimester, no human breastmilk exposure studies were identified . We identified five further studies of 143 RTX pregnancy exposures in addition to eight exposures in studies that reported on combined outcomes of multiple biologic-exposed pregnancies (see ‘Biologic DMARDs’ above). Four studies ( n = 135) of mostly pre-conception or first trimester maternal exposure did not report an increased risk of preterm birth, miscarriage, LBW or congenital anomaly . An additional cohort study of eight mothers treated with one to four cycles of RTX during pregnancy ( n = 6 for diffuse large B-cell lymphoma; n = 2 for SLE) found a high rate of preterm birth and intrauterine infections, but noted the potential for confounding by the underlying disease and concomitant medications . UKTIS reports that there is insufficient evidence to assess whether the risk for spontaneous miscarriage, congenital malformation, birth weight, intrauterine death or adverse neurodevelopmental outcomes is increased following exposure to RTX in utero . Due to a lack of data on the effects of RTX on the neonatal/infant immune system, it recommends delaying administration of live vaccines following in-utero exposure to RTX, although no specific timescale is given . There were three studies of breastmilk exposure ( n = 23) to RTX [ , , ]. One study detected minimal transfer of RTX into the breastmilk of nine breastfeeding mothers, with a relative infant dose of rituximab well below theoretically acceptable levels (<0.4%) . No adverse effects attributable to RTX exposure were reported in the breastfed infants, and the drug was considered to have an acceptable benefit-to-risk ratio, supporting both maternal treatment and breastmilk exposure. LactMed states that the amount of RTX in breastmilk is very low and absorption is unlikely because it is a protein with a high molecular weight that is likely to be partially destroyed in the infant’s gastrointestinal tract; thus, absorption by the infant is probably minimal . Vaccine efficacy after prenatal exposure to RTX is much less studied, and systematic reviews have identified adequate non-live vaccination outcomes from 5/6 RTX-exposed infants, with low immunity to diphtheria detected in one 11-month-old infant exposed to RTX at conception . Recommendations for rituximab in pregnancy and breastmilk exposure Limited evidence has not shown RTX to be teratogenic; however, there remains insufficient evidence to be confident that it is compatible with pregnancy. Consider stopping the drug at conception (GRADE 2C, SOA 99.3%). RTX may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 99.7%). If RTX is used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 98.7%). Based on limited evidence, maternal treatment with RTX is compatible with breastmilk exposure (GRADE 2C, SOA 99.5%).
Limited evidence has not shown RTX to be teratogenic; however, there remains insufficient evidence to be confident that it is compatible with pregnancy. Consider stopping the drug at conception (GRADE 2C, SOA 99.3%). RTX may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 99.7%). If RTX is used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 98.7%). Based on limited evidence, maternal treatment with RTX is compatible with breastmilk exposure (GRADE 2C, SOA 99.5%).
Tocilizumab Due to insufficient data, we previously recommended that TCZ be stopped at least three months pre-conception, but that unintentional exposure early in the first trimester is unlikely to be harmful due to its IgG1 structure. This limited data came from outcomes of 33 pregnancies in 32 patients, published in abstract form . We identified three further studies of 365 pregnancy exposures to TCZ in addition to four exposures in the papers described above which reported combined outcomes of multiple biologic-exposed pregnancies (see ‘DMARDs’ above). These studies of mainly first trimester exposure did not find increased rates of congenital abnormalities in patients with rheumatic disease. Two studies commented on a higher than background rate of spontaneous miscarriage , and could not exclude an effect on birth weight and risk of prematurity, while acknowledging potential confounding factors. For example, stopping effective treatment early in pregnancy could destabilize the rheumatic disease, with adverse consequences for the pregnancy. UKTIS notes a small number of studies lacking comparator groups that are confounded by maternal disease, and states that there is currently no compelling evidence that TCZ is teratogenic or fetotoxic. Due to lack of data on effects on the neonatal/infant immune system, it recommends that live vaccines are avoided until the infant is 6 months of age . One study of n = 2 breastmilk exposure to TCZ did not report any adverse effects attributable to the drug . Saito et al. have described a total of four cases of babies exposed to TCZ via breastmilk with no complications in the infants (three publications which did not meet our inclusion criteria) . Concentrations of TCZ were measured in the breastmilk in these studies and two additional cases reports, with no clinical adverse effects reported . Levels in the breastmilk were found to peak on day three following the infusion , but were significantly lower than the corresponding maternal serum concentrations in all of these cases, ranging from 11% in colostrum , down to as low as 1:2000 . Sarilumab No papers were found that met our inclusion criteria. Sanofi provided data on 13 patients who became pregnant in the sarilumab and DMARD long-term safety population, of whom seven had a miscarriage . In addition, two male patients fathered two healthy children. Recommendations for IL-6 inhibitors in pregnancy and breastmilk exposure Limited evidence has not shown IL-6 inhibitors (IL-6i) to be teratogenic; however, there remains insufficient evidence to be confident that they are compatible with pregnancy. Consider stopping the drug at conception. Any exposure, however, during pregnancy is unlikely to be harmful (GRADE 2C, SOA 99.7%). IL-6i may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 100%). If IL-6i are used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 99.3%). Based on limited evidence, maternal treatment with IL-6i is compatible with breastmilk exposure (GRADE 2C, SOA 100%).
Due to insufficient data, we previously recommended that TCZ be stopped at least three months pre-conception, but that unintentional exposure early in the first trimester is unlikely to be harmful due to its IgG1 structure. This limited data came from outcomes of 33 pregnancies in 32 patients, published in abstract form . We identified three further studies of 365 pregnancy exposures to TCZ in addition to four exposures in the papers described above which reported combined outcomes of multiple biologic-exposed pregnancies (see ‘DMARDs’ above). These studies of mainly first trimester exposure did not find increased rates of congenital abnormalities in patients with rheumatic disease. Two studies commented on a higher than background rate of spontaneous miscarriage , and could not exclude an effect on birth weight and risk of prematurity, while acknowledging potential confounding factors. For example, stopping effective treatment early in pregnancy could destabilize the rheumatic disease, with adverse consequences for the pregnancy. UKTIS notes a small number of studies lacking comparator groups that are confounded by maternal disease, and states that there is currently no compelling evidence that TCZ is teratogenic or fetotoxic. Due to lack of data on effects on the neonatal/infant immune system, it recommends that live vaccines are avoided until the infant is 6 months of age . One study of n = 2 breastmilk exposure to TCZ did not report any adverse effects attributable to the drug . Saito et al. have described a total of four cases of babies exposed to TCZ via breastmilk with no complications in the infants (three publications which did not meet our inclusion criteria) . Concentrations of TCZ were measured in the breastmilk in these studies and two additional cases reports, with no clinical adverse effects reported . Levels in the breastmilk were found to peak on day three following the infusion , but were significantly lower than the corresponding maternal serum concentrations in all of these cases, ranging from 11% in colostrum , down to as low as 1:2000 .
No papers were found that met our inclusion criteria. Sanofi provided data on 13 patients who became pregnant in the sarilumab and DMARD long-term safety population, of whom seven had a miscarriage . In addition, two male patients fathered two healthy children.
Limited evidence has not shown IL-6 inhibitors (IL-6i) to be teratogenic; however, there remains insufficient evidence to be confident that they are compatible with pregnancy. Consider stopping the drug at conception. Any exposure, however, during pregnancy is unlikely to be harmful (GRADE 2C, SOA 99.7%). IL-6i may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 100%). If IL-6i are used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 99.3%). Based on limited evidence, maternal treatment with IL-6i is compatible with breastmilk exposure (GRADE 2C, SOA 100%).
Anakinra Anakinra is a recombinant form of human interleukin-1 receptor antagonist (IL-1Ra) with a high molecular weight which was not found to cross ex-vivo full-term human placentae . Previously, there was insufficient information evidence on which to base a recommendation for anakinra in pregnancy, but unintentional exposure in the first trimester was considered unlikely to be harmful . This limited evidence came from reports on five pregnancies in three studies/reports, with no evidence of harm [ , , ]. Subsequently, we identified four studies of 43 pregnancy exposures to anakinra , in addition to one exposure in the papers described above which reported combined outcomes of multiple biologic-exposed pregnancies (see ‘Biologic DMARDs’ above). These exposures were mostly in patients with periodic fevers and severe maternal disease. Overall, outcomes were reassuring, although two congenital renal anomalies and two cases of oligohydramnios (which can be linked to foetal renal anomalies) were reported [ , , ]. It was unclear, however, whether the renal abnormalities were associated with antenatal anakinra use or maternal hyperthermia or both. Given the significant beneficial effects of anakinra in suppressing maternal disease with limited pregnancy-compatible options, it was considered a safe alternative in managing disease in women with periodic fever in pregnancy. UKTIS does not report on anakinra. Two studies of n = 12 breastmilk exposures to anakinra did not report any adverse effects attributable to the drug . LactMed states that IL-1Ra is a normal component of human milk, possibly as an anti-inflammatory agent, and that several infants have been breastfed during maternal anakinra therapy with no obvious adverse effects. If anakinra is required by the mother, it is not a reason to discontinue breastfeeding . Canakinumab Canakinumab is a human monoclonal antibody to IL-1. There is, therefore, the potential for active transport across the placenta from the second trimester onwards. One paper reporting canakinumab use from pre-conception in eight pregnancies (stopped during first trimester in 5/8 pregnancies) met our inclusion criteria . There were seven live births, all of whom were healthy, full term and normal birth weight. One mother with refractory Cogan syndrome had an early miscarriage at 6 weeks (after a miscarriage the previous year while on anakinra). In addition, Novartis supplied information from analyses of the Novartis Global Safety Database, which included 76 maternal and nine paternal exposures . There were 47 known pregnancy outcomes, with 27 healthy newborns, one preterm birth (with meconium staining not thought likely to be related to canakinumab exposure), nine spontaneous miscarriages (including two miscarriages related to paternal exposure in one father) and three elective terminations. Seven other adverse events were reported: one case of congenital pyelocaliectasis following paternal exposure to canakinumab; one case of congenital musculoskeletal abnormality following maternal exposure, with insufficient data on other contributory factors; one case of inherited genetic disease with neonatal RSV infection and pyrexia following infant vaccinations; and four adverse events very unlikely to relate to canakinumab exposure (one newborn non-serious hypotension and three cases of likely inherited disease: Muckle–Wells syndrome and CAPS). One cohort study reported breastmilk exposure in four infants whose mothers were prescribed regular canakinumab, with no reported serious infections or developmental abnormalities . Recommendations for IL-1 inhibitors in pregnancy and breastmilk exposure Limited evidence has not shown IL-1 inhibitors (IL-1i) to be teratogenic; however, there remains insufficient evidence to be confident that they are compatible with pregnancy. Consider stopping the drug at conception. Any exposure, however, during pregnancy is unlikely to be harmful (GRADE 2C, SOA 99.8%). IL-1i may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 100%). If IL-1i are used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 99.3%). Based on limited evidence, maternal treatment with IL-1i is compatible with breastmilk exposure (GRADE 2C, SOA 100%).
Anakinra is a recombinant form of human interleukin-1 receptor antagonist (IL-1Ra) with a high molecular weight which was not found to cross ex-vivo full-term human placentae . Previously, there was insufficient information evidence on which to base a recommendation for anakinra in pregnancy, but unintentional exposure in the first trimester was considered unlikely to be harmful . This limited evidence came from reports on five pregnancies in three studies/reports, with no evidence of harm [ , , ]. Subsequently, we identified four studies of 43 pregnancy exposures to anakinra , in addition to one exposure in the papers described above which reported combined outcomes of multiple biologic-exposed pregnancies (see ‘Biologic DMARDs’ above). These exposures were mostly in patients with periodic fevers and severe maternal disease. Overall, outcomes were reassuring, although two congenital renal anomalies and two cases of oligohydramnios (which can be linked to foetal renal anomalies) were reported [ , , ]. It was unclear, however, whether the renal abnormalities were associated with antenatal anakinra use or maternal hyperthermia or both. Given the significant beneficial effects of anakinra in suppressing maternal disease with limited pregnancy-compatible options, it was considered a safe alternative in managing disease in women with periodic fever in pregnancy. UKTIS does not report on anakinra. Two studies of n = 12 breastmilk exposures to anakinra did not report any adverse effects attributable to the drug . LactMed states that IL-1Ra is a normal component of human milk, possibly as an anti-inflammatory agent, and that several infants have been breastfed during maternal anakinra therapy with no obvious adverse effects. If anakinra is required by the mother, it is not a reason to discontinue breastfeeding .
Canakinumab is a human monoclonal antibody to IL-1. There is, therefore, the potential for active transport across the placenta from the second trimester onwards. One paper reporting canakinumab use from pre-conception in eight pregnancies (stopped during first trimester in 5/8 pregnancies) met our inclusion criteria . There were seven live births, all of whom were healthy, full term and normal birth weight. One mother with refractory Cogan syndrome had an early miscarriage at 6 weeks (after a miscarriage the previous year while on anakinra). In addition, Novartis supplied information from analyses of the Novartis Global Safety Database, which included 76 maternal and nine paternal exposures . There were 47 known pregnancy outcomes, with 27 healthy newborns, one preterm birth (with meconium staining not thought likely to be related to canakinumab exposure), nine spontaneous miscarriages (including two miscarriages related to paternal exposure in one father) and three elective terminations. Seven other adverse events were reported: one case of congenital pyelocaliectasis following paternal exposure to canakinumab; one case of congenital musculoskeletal abnormality following maternal exposure, with insufficient data on other contributory factors; one case of inherited genetic disease with neonatal RSV infection and pyrexia following infant vaccinations; and four adverse events very unlikely to relate to canakinumab exposure (one newborn non-serious hypotension and three cases of likely inherited disease: Muckle–Wells syndrome and CAPS). One cohort study reported breastmilk exposure in four infants whose mothers were prescribed regular canakinumab, with no reported serious infections or developmental abnormalities .
Limited evidence has not shown IL-1 inhibitors (IL-1i) to be teratogenic; however, there remains insufficient evidence to be confident that they are compatible with pregnancy. Consider stopping the drug at conception. Any exposure, however, during pregnancy is unlikely to be harmful (GRADE 2C, SOA 99.8%). IL-1i may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 100%). If IL-1i are used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 99.3%). Based on limited evidence, maternal treatment with IL-1i is compatible with breastmilk exposure (GRADE 2C, SOA 100%).
ABA is a fusion protein containing the Fc region of IgG1 fused to the extracellular domain of CTLA-4; therefore, it is able to cross the placental barrier from approximately week 16. Previously, there was insufficient data to recommend ABA in pregnancy, but unintentional exposure early in the first trimester was considered unlikely to be harmful . This conclusion was based on reports from 11 pregnancy exposures (and a further eight exposures reported in abstract form only), many of which were confounded by concomitant MTX [ , , ]. We identified two further studies of at least 196 pregnancy exposures to ABA (with some overlap of data) , in addition to three exposures in the papers described above which reported combined outcomes of multiple biologic-exposed pregnancies (see ‘Biologic DMARDs’ above). The data from these studies did not suggest an increased risk of adverse pregnancy outcomes with ABA, and many of the congenital abnormalities were considered by the authors to be associated with concomitant use of other teratogenic DMARDs. The apparently high rate of spontaneous miscarriage (total ≥48/184) is difficult to interpret due to confounding by indication and other medications such as MTX, as well as detection bias of early miscarriages during the close monitoring of clinical trials . The authors concluded that ABA should only be used during pregnancy if the benefit to the mother justifies potential risk to the foetus. UKTIS does not report on ABA. Although it did not meet our inclusion criteria, we identified one case report of breastmilk exposure to ABA, with no adverse effects, in which ABA was secreted into breastmilk at levels 1/200–1/300 of those in serum . LactMed describes one case report that reported concentrations of ABA in milk were very low and did not appear to affect the breastfed infant, concluding that if ABA is required by the mother, it is not a reason to discontinue breastfeeding, although alternative drugs may be preferred . Recommendations for abatacept in pregnancy and breastmilk exposure Limited evidence has not shown ABA to be teratogenic; however, there remains insufficient evidence to be confident that it is compatible with pregnancy. Consider stopping the drug at conception. Any exposure, however, during pregnancy is unlikely to be harmful (GRADE 2C, SOA 99.3%). ABA may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 99.3%). If ABA is used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 99.3%). Based on limited evidence, maternal treatment with ABA is compatible with breastmilk exposure (GRADE 2C, SOA 99.5%).
Limited evidence has not shown ABA to be teratogenic; however, there remains insufficient evidence to be confident that it is compatible with pregnancy. Consider stopping the drug at conception. Any exposure, however, during pregnancy is unlikely to be harmful (GRADE 2C, SOA 99.3%). ABA may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 99.3%). If ABA is used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 99.3%). Based on limited evidence, maternal treatment with ABA is compatible with breastmilk exposure (GRADE 2C, SOA 99.5%).
Belimumab (BEL) is a fully humanized monoclonal IgG1 that inhibits B-cell activating factor. Previously, our systematic search did not identify any publications that met our search criteria on use of BEL in human pregnancy, although we found reference to pregnancy outcomes from placebo-controlled, phase 2 and 3 studies in 83 SLE pregnancies , and BEL pregnancy registry data describing known outcomes from 118 SLE pregnancies . Overall, these studies did not identify any pattern of adverse effects in pregnancy directly attributable to BEL. We subsequently identified one study of 66 BEL pregnancy exposures . Total foetal losses in BEL-treated subjects were similar to background estimates in SLE patients (∼25%), although data remain limited. Another six studies, which did not meet our search criteria , reported an additional 124 pregnancies exposed to BEL. There was only one reported congenital abnormality: a case report of mild Ebstein’s anomaly in a baby following successful control of SLE (with previous lupus nephritis and APS, previously only controlled on MMF) . Two case series (26 pregnancy exposures) reported preterm delivery in 12/22 live births, with six babies being SGA . However, these cohorts included patients with complex rheumatic disease (SLE and APS), including some patients with previous lupus nephritis and active disease at conception , and a high average maternal age and rate of previous pregnancy losses . Both authors concluded that, while careful consideration and further research is required, BEL could be a reasonable treatment option for patients with SLE requiring treatment in pregnancy. UKTIS does not report on BEL. Although they did not meet our inclusion criteria, we identified two cases of breastmilk exposure while on BEL . No neonatal outcomes, adverse or otherwise, were reported in one case . In the other case, there were no adverse effects, and BEL was secreted into breastmilk at levels 1/200–1/500 of those in serum . LactMed states that, based on preliminary information that BEL levels in breastmilk are very low and infant absorption is probably minimal, if BEL is required by the mother, it is not a reason to discontinue breastfeeding, although caution is required, especially while nursing a newborn or preterm infant . Recommendations for belimumab in pregnancy and breastmilk exposure Limited evidence has not shown BEL to be teratogenic; however, there remains insufficient evidence to be confident that it is compatible with pregnancy. Consider stopping the drug at conception. Any exposure, however, during pregnancy is unlikely to be harmful (GRADE 2C, SOA 99.3%). BEL may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 99.5%). If BEL is used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 98.8%). Based on limited evidence, maternal treatment with BEL is compatible with breastmilk exposure (GRADE 2C, SOA 99.5%).
Limited evidence has not shown BEL to be teratogenic; however, there remains insufficient evidence to be confident that it is compatible with pregnancy. Consider stopping the drug at conception. Any exposure, however, during pregnancy is unlikely to be harmful (GRADE 2C, SOA 99.3%). BEL may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 99.5%). If BEL is used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 98.8%). Based on limited evidence, maternal treatment with BEL is compatible with breastmilk exposure (GRADE 2C, SOA 99.5%).
Interleukin-17 inhibitors (IL-17i) were not in general use and so not considered in our previous search. Secukinumab has an IgG1 structure, which may theoretically suggest increased transplacental transfer compared with the IgG4 structure of ixekizumab. We found three studies of mostly first trimester pregnancy exposures to secukinumab ( n = 244) and ixekizumab ( n = 18) . The authors did not report any increased incidence of adverse outcomes directly attributable to the drugs, although information, particularly for ixekizumab, remains very limited. In addition, Novartis provided data from a search of their Global Safety Database, relating to 298 reports of maternal pregnancy exposures and 90 paternal exposures to secukinumab, mostly during the first trimester . No outcome data were provided, but analysis within Novartis did not reveal any new safety information. UKTIS does not report on these drugs. There were no data on breastmilk exposure to IL-17i identified in our search or in LactMed. Data provided from the Novartis Global Safety Database identified six breastmilk exposures to secukinumab, including one report of newborn pyrexia during breastfeeding . The data in all cases were too limited to draw conclusions, but no new safety concerns were inferred. Recommendations for interleukin-17 inhibitors in pregnancy and breastmilk exposure Limited evidence has not shown IL-17i to be teratogenic; however, there remains insufficient evidence to be confident that they are compatible with pregnancy. Consider stopping the drug at conception. Any exposure, however, during pregnancy is unlikely to be harmful (GRADE 2C, SOA 99.3%). IL-17i may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 99%). If IL-17i are used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 99.3%). Based on limited evidence, maternal treatment with IL-17i is compatible with breastmilk exposure (GRADE 2C, SOA 99.5%).
Limited evidence has not shown IL-17i to be teratogenic; however, there remains insufficient evidence to be confident that they are compatible with pregnancy. Consider stopping the drug at conception. Any exposure, however, during pregnancy is unlikely to be harmful (GRADE 2C, SOA 99.3%). IL-17i may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 99%). If IL-17i are used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 99.3%). Based on limited evidence, maternal treatment with IL-17i is compatible with breastmilk exposure (GRADE 2C, SOA 99.5%).
UST is an interleukin-12/23 inhibitor (IL-12/23i) with an IgG1 structure. UST was not in general use and not considered in our previous search. We found three studies of mostly first and second trimester pregnancy exposures to UST ( n = 517) for maternal psoriasis, PsA and IBD . Overall, these data showed that the rates of live births, spontaneous miscarriages and congenital anomalies were consistent with the general population and TNFi-exposed pregnancies. Similarly, UKTIS did not identify any specific drug-related risk but was unable to provide a reliable evidence-based evaluation of risk due to limited information . Although they did not meet our inclusion criteria, we identified two reports of exposure to UST during breastmilk exposure . In one case report, the trough level of UST in breastmilk after restarting this drug post-partum in a Crohn’s disease patient was initially in the same range as the corresponding serum trough level, and then decreased during maintenance therapy . In another study assessing breastfeeding mothers taking a range of biologic medication for IBD, UST was detected at low levels in 4/6 mothers taking this medication . Overall, in this study, breastfed infants of mothers on biologics were found to have similar risks of infection and rates of milestone achievement compared with non-breastfed infants or infants unexposed to these drugs. LactMed states that UST is either not detectable or found at very low levels in breastmilk and infant absorption is probably minimal; as such, if UST is required by the mother, it is not a reason to discontinue breastfeeding, although caution is required especially while nursing a newborn or preterm infant . Recommendations for interleukin-12/23 inhibitors in pregnancy and breastmilk exposure Limited evidence has not shown UST to be teratogenic; however, there remains insufficient evidence to be confident that it is compatible with pregnancy. Consider stopping the drug at conception. Any exposure, however, during pregnancy is unlikely to be harmful (GRADE 2C, SOA 99.3%). UST may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 98.8%). If UST is used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 99.3%). Based on limited evidence, maternal treatment with UST is compatible with breastmilk exposure (GRADE 2C, SOA 99.5%).
Limited evidence has not shown UST to be teratogenic; however, there remains insufficient evidence to be confident that it is compatible with pregnancy. Consider stopping the drug at conception. Any exposure, however, during pregnancy is unlikely to be harmful (GRADE 2C, SOA 99.3%). UST may be considered to manage severe maternal disease in pregnancy if no other pregnancy-compatible drugs are suitable (GRADE 2C, SOA 98.8%). If UST is used to treat severe maternal disease in the third trimester, it is currently recommended to avoid all live vaccines in the infant vaccination schedule until 6 months of age (GRADE 2C, SOA 99.3%). Based on limited evidence, maternal treatment with UST is compatible with breastmilk exposure (GRADE 2C, SOA 99.5%).
Anifrolumab is a fully human, IgG1κ monoclonal antibody to type I interferon receptor subunit 1. Although it is not currently licensed for use in the UK, anifrolumab is approved for use in other countries, including FDA approval to treat moderate-to-severe SLE in the USA . Although there is no published evidence on the use of this drug in pregnancy, breastmilk or paternal exposure, further information was provided by AstraZeneca on data from completed trials. Despite the mandatory contraception use in anifrolumab clinical trials, pregnancies have been reported in 20 SLE patients receiving anifrolumab in studies completed by December 2021. Patients who became pregnant during studies had to immediately discontinue the investigational product. No anifrolumab-associated congenital abnormalities or anifrolumab-associated adverse events have been observed in the clinical trials. Current data on anifrolumab pregnancy exposure are insufficient to inform about potential drug-related risks and therefore post-authorisation pregnancy studies are planned. UKTIS does not report on anifrolumab. Although there is no information regarding breastmilk exposure to anifrolumab, LactMed states that due to its large molecular weight, the amount in milk is likely to be very low, and it is also likely to be partially destroyed in the infant's gastrointestinal tract with minimal absorption by the infant . Based on this limited evidence, we have not made any recommendations on this drug.
Data relating to tsDMARD use in pregnancy remain scarce, and in this guideline, updated case reports and small case series were excluded for consistency. A recent systematic review summarized all available data including case reports and raised no additional safety concerns . No studies were found relating to apremilast use in pregnancy, and therefore we were unable to make any recommendations. JAK inhibitors Tofacitinib (TOF), baricitinib (BAR) and upadacitinib (UPA) are oral Janus kinase inhibitors (JAKi) which were not in general use and so not considered in our previous search. They are small molecule inhibitors of low molecular weight which could theoretically cross the placenta. JAKi have a short half-life (∼3 h for TOF), although biological changes can persist for longer (for example, dose-dependent reductions in NK cells and CRP do not appear to reverse before 2 weeks after discontinuation) . We identified three published reports of pregnancies during RCTs and post-marketing surveillance reports for TOF, but due to probable overlap, only the later 2018 paper was included (further updates were published in abstract form in 2020) . This included 116 pregnancy exposures to TOF, with the known reported outcomes including 15/72 first trimester miscarriages, and major malformations in 2/44 live births (one pulmonary valve stenosis and one ventricular septal defect), which was in line with background risks in the general population . Pfizer medical information supplied further details on pregnancies reported during the TOF clinical trials programmes . The RA and psoriasis trials included a total of 60 maternal exposures during the first trimester with 51 known outcomes, with concomitant MTX use either before or during pregnancy in at least 19 of these cases . There were 28 healthy newborns, two premature infants, and one congenital malformation where the mother was also taking losartan. In these trials, there were nine spontaneous miscarriages, nine medical terminations, and one elective termination. In the ulcerative colitis (UC) clinical trial programmes, there were 15 maternal exposures with 13 known outcomes (nine healthy newborns, two spontaneous miscarriages and two medical terminations) . In the PsA programmes, there were a further four maternal and three paternal exposures with six known outcomes (three healthy newborns, two spontaneous miscarriages and one medical termination) . Post-marketing reports up to 2017 were included in the analysis by Mahadevan et al. with 42 (predominantly first trimester) maternal exposures and three paternal exposures, with only 12 known outcomes (seven healthy newborns, one congenital malformation, three spontaneous miscarriages and one medical termination) . Although not eligible for inclusion, we found one case report of an RA pregnancy exposed to BAR from pre-conception up to 17 weeks gestation, with a healthy infant delivered at 38 weeks gestation by caesarean section . In addition, Eli Lilly provided data on 36 maternal pregnancies during the BAR clinical trials programme, with 25 known outcomes (trimester and duration unspecified) . Many were taking concomitant medications, including MTX ( n = 14). Outcomes included 13 healthy newborns (including three born preterm), seven spontaneous miscarriages (with MTX exposure in six of these), and five elective terminations. In addition, there have been 22 post-marketing reports of pregnancy exposures to BAR, with four known outcomes as of January 2021, including three healthy newborns (including one preterm), and one spontaneous miscarriage at 13 weeks in a 36-year-old mother (early first trimester exposure to BAR with concomitant medications for RA including HCQ, GOL, prednisolone, MTX, folic acid and ibuprofen) . UKTIS does not reported on JAKi. We found no evidence relating to breastmilk exposure to JAKi. Given that they are small molecules and likely to transfer into breastmilk, they should be avoided. LactMed states that no information is available on the use of TOF or BAR during breastmilk exposure, and alternate drugs are preferred, especially while nursing a newborn or preterm infant . There were no studies identified that met our inclusion criteria for UPA. Data provided by AbbVie included 54 maternal pregnancies inadvertently exposed to UPA in the month prior to conception or during the first trimester of pregnancy during the clinical trials programme . The 41 known outcomes included 17 healthy live births (including two premature deliveries), 14 spontaneous miscarriages (10 of whom were on concomitant MTX), nine elective terminations (all with no reported foetal defects) and one ectopic pregnancy. Filgotinib was NICE-approved after our search window, and so was not included in our search. Recommendations for JAK inhibitors in pregnancy and breastmilk exposure There are insufficient data to make a recommendation on JAKi use during pregnancy and they should be stopped at least two weeks before planned conception (GRADE 2C, SOA 99.5%). There are insufficient data to recommend JAKi in breastfeeding and, given they are likely to transfer into breastmilk, they should be avoided (GRADE 2C, SOA 99.5%).
Tofacitinib (TOF), baricitinib (BAR) and upadacitinib (UPA) are oral Janus kinase inhibitors (JAKi) which were not in general use and so not considered in our previous search. They are small molecule inhibitors of low molecular weight which could theoretically cross the placenta. JAKi have a short half-life (∼3 h for TOF), although biological changes can persist for longer (for example, dose-dependent reductions in NK cells and CRP do not appear to reverse before 2 weeks after discontinuation) . We identified three published reports of pregnancies during RCTs and post-marketing surveillance reports for TOF, but due to probable overlap, only the later 2018 paper was included (further updates were published in abstract form in 2020) . This included 116 pregnancy exposures to TOF, with the known reported outcomes including 15/72 first trimester miscarriages, and major malformations in 2/44 live births (one pulmonary valve stenosis and one ventricular septal defect), which was in line with background risks in the general population . Pfizer medical information supplied further details on pregnancies reported during the TOF clinical trials programmes . The RA and psoriasis trials included a total of 60 maternal exposures during the first trimester with 51 known outcomes, with concomitant MTX use either before or during pregnancy in at least 19 of these cases . There were 28 healthy newborns, two premature infants, and one congenital malformation where the mother was also taking losartan. In these trials, there were nine spontaneous miscarriages, nine medical terminations, and one elective termination. In the ulcerative colitis (UC) clinical trial programmes, there were 15 maternal exposures with 13 known outcomes (nine healthy newborns, two spontaneous miscarriages and two medical terminations) . In the PsA programmes, there were a further four maternal and three paternal exposures with six known outcomes (three healthy newborns, two spontaneous miscarriages and one medical termination) . Post-marketing reports up to 2017 were included in the analysis by Mahadevan et al. with 42 (predominantly first trimester) maternal exposures and three paternal exposures, with only 12 known outcomes (seven healthy newborns, one congenital malformation, three spontaneous miscarriages and one medical termination) . Although not eligible for inclusion, we found one case report of an RA pregnancy exposed to BAR from pre-conception up to 17 weeks gestation, with a healthy infant delivered at 38 weeks gestation by caesarean section . In addition, Eli Lilly provided data on 36 maternal pregnancies during the BAR clinical trials programme, with 25 known outcomes (trimester and duration unspecified) . Many were taking concomitant medications, including MTX ( n = 14). Outcomes included 13 healthy newborns (including three born preterm), seven spontaneous miscarriages (with MTX exposure in six of these), and five elective terminations. In addition, there have been 22 post-marketing reports of pregnancy exposures to BAR, with four known outcomes as of January 2021, including three healthy newborns (including one preterm), and one spontaneous miscarriage at 13 weeks in a 36-year-old mother (early first trimester exposure to BAR with concomitant medications for RA including HCQ, GOL, prednisolone, MTX, folic acid and ibuprofen) .
]. There were no studies identified that met our inclusion criteria for UPA. Data provided by AbbVie included 54 maternal pregnancies inadvertently exposed to UPA in the month prior to conception or during the first trimester of pregnancy during the clinical trials programme . The 41 known outcomes included 17 healthy live births (including two premature deliveries), 14 spontaneous miscarriages (10 of whom were on concomitant MTX), nine elective terminations (all with no reported foetal defects) and one ectopic pregnancy. Filgotinib was NICE-approved after our search window, and so was not included in our search. Recommendations for JAK inhibitors in pregnancy and breastmilk exposure There are insufficient data to make a recommendation on JAKi use during pregnancy and they should be stopped at least two weeks before planned conception (GRADE 2C, SOA 99.5%). There are insufficient data to recommend JAKi in breastfeeding and, given they are likely to transfer into breastmilk, they should be avoided (GRADE 2C, SOA 99.5%).
There are insufficient data to make a recommendation on JAKi use during pregnancy and they should be stopped at least two weeks before planned conception (GRADE 2C, SOA 99.5%). There are insufficient data to recommend JAKi in breastfeeding and, given they are likely to transfer into breastmilk, they should be avoided (GRADE 2C, SOA 99.5%).
Hydroxychloroquine For HCQ, no additional paternal exposures were identified to the previously identified cohort study and case series of 13 pregnancies after paternal exposure to HCQ, which did not find any increased risk of adverse foetal outcomes. Corticosteroids Previously, four cohort studies and two case series reported on outcomes from ≥2127 pregnancies after paternal exposure to prednisolone, and a case-control study and a case series reported on outcomes from ( n = 4) pregnancies after paternal exposure to methylprednisolone. Overall, the quality of these studies was low, but reassuringly they did not identify an increased risk of adverse foetal outcomes. Since then, an additional study of 2380 paternal exposures to corticosteroids did not identify any statistically significant increase in adverse birth outcomes . Methotrexate Previous low-quality evidence from outcomes of pregnancies after paternal exposure ( n = 263) to predominantly low-dose MTX did not find any adverse effects . We identified an additional six studies of paternal exposures to MTX ( n = 2026) within three months of conception that similarly found no increased risk of adverse foetal outcomes when compared with MTX-unexposed controls ( n = 4 700 599) . An additional study examined foetal outcomes with paternal MTX (and other DMARD) use compared with abatacept; however, outcomes for individual drugs were not reported . Sulfasalazine Our previous review of three cohort [ , , ] and one case-control study reporting on 237 pregnancies after paternal exposure to SSZ did not find an increased risk of adverse foetal outcomes, although the quality of evidence was low. SSZ may also affect male fertility, with oligospermia, reduced sperm motility and increased proportions of abnormal sperm previously reported . No further studies of paternal SSZ exposure were identified in our search. Leflunomide Previously, we identified a cohort study and case report describing outcomes from ( n = 2) pregnancies after paternal exposure to LEF within three months of conception, and subsequent pregnancy exposure (with intercourse without a condom) in at least one case with no reported washout. No adverse foetal outcomes were observed. An additional study was identified examining foetal outcomes with paternal leflunomide (and other DMARD) use compared with abatacept; however, outcomes for individual drugs were not reported . Azathioprine In addition to the previous 602 paternal exposures [ , , , ], we identified three studies of n = 2680 pregnancies after paternal exposure to AZA [ , , ]. Overall, no increased risk of adverse foetal outcomes was observed. Ciclosporin In addition to previous studies [ , , , ] on outcomes from pregnancies ( n ≥ 254) after paternal exposure to CsA, we found a Danish population-based cohort study of birth outcomes in 247 children fathered by men treated with CsA before conception . Overall, these studies did not identify an increased risk of adverse foetal outcomes with paternal exposure to CsA. Tacrolimus No additional paternal studies were found to complement previous findings reporting outcomes from pregnancies ( n ≥ 120) after paternal exposure to tacrolimus, which did not identify an increased risk of adverse foetal outcomes [ , , ]. Cyclophosphamide No new studies of paternal exposure to CYC were identified. In addition to a potential long-term impact on spermatogenesis (and hence fertility) in men , there is evidence of an adverse impact on germ cell development and male-mediated teratogenicity from animal studies , although this has not been proven in humans . Mycophenolate mofetil In addition to the three previous studies [ , , ] of paternal exposures to MMF ( n ≥ 72), we found three additional studies of pregnancies ( n = 220) after paternal exposure [ , , ]. Overall, the quality of these studies was low, but they did not identify an increased risk of adverse foetal outcomes. Intravenous immunoglobulin There is no evidence relating to paternal exposure but based on maternal compatibility it is unlikely to be harmful. Anti-TNFα drugs Previously, we found five cohort studies [ , , , , ], two case series , two case reports and a case-control study that reported on outcomes from pregnancies ( n = 131) after paternal exposure to INF, ETA and ADA. Overall, the quality of these studies was deemed to be low, but they did not identify an increased risk of adverse foetal outcomes. We identified an additional eight cohort studies reporting outcomes from 620 pregnancies after paternal exposure to TNFi [ , , , ], with no significant findings to suggest adverse foetal outcomes were more likely after TNFi exposure. Rituximab We did not identify any further paternal exposures to the 11 found previously , which did not identify any problems in relation to paternal exposure to RTX. IL-6 Inhibitors Two studies of n = 15 paternal exposures to TCZ did not find any drug-related effects . IL-1 Inhibitors One study of n = 5 paternal exposures to anakinra did not find any drug-related effects . Youngstein et al. reported on five healthy pregnancies to three fathers on long-term canakinumab treatment . Abatacept In addition to the previous case of one healthy pregnancy following paternal ABA exposure , a study of clinical trial and post-marketing data submitted to the manufacturer (up to 2014) reported 10 paternal ABA exposures, resulting in nine healthy live births with one elective termination . IL-17 inhibitors Two studies of paternal exposure to secukinumab ( n = 54) and ixekizumab ( n = 34) did not report any adverse drug-related effects . JAK-inhibitors There were 87 paternal exposures to TOF in a study included in our search . Further details provided by Pfizer reported paternal exposures to TOF in the PsA clinical trials ( n = 3) and post-marketing reports ( n = 3), described above. During the RA and psoriasis clinical trials programmes, 66 men were exposed to TOF with 45 known outcomes, including 37 normal healthy newborns, two premature deliveries (with one subsequent neonatal death) and six spontaneous miscarriages . In the UC clinical trial programmes, there were 19 paternal exposures with 17 known outcomes (15 healthy newborns including one healthy preterm delivery at 34 weeks, and two spontaneous miscarriages) . Data provided by Eli Lilly included eight paternal exposures to BAR in the clinical trials programmes, and outcomes included six full-term healthy newborns and two spontaneous miscarriages . No data were found relating to paternal exposure to BEL, IL-12/23i or anifrolumab. Recommendations for paternal exposure to immunomodulatory drugs Due to the adverse effect of CYC on male fertility, semen cryopreservation is recommended for men prior to paternal exposure (GRADE 1C, SOA 99.5%). Men who take SSZ may have reduced fertility. There is little evidence to suggest that SSZ should be stopped pre-conception, unless conception is delayed by more than 12 months when stopping SSZ should be considered along with other causes of infertility (GRADE 1C, SOA 99.0%). Paternal exposure to the following anti-rheumatic medication is compatible with pregnancy: prednisolone, low-dose (≤25 mg/week) MTX, AZA (GRADE 1B); TNFi, cyclosporin (GRADE 1C); HCQ, LEF, tacrolimus, MMF, IVIG, RTX, IL-6i, IL-1i, ABA, BEL, IL-17i, UST and JAKi (GRADE 2C, SOA 99.3%).
For HCQ, no additional paternal exposures were identified to the previously identified cohort study and case series of 13 pregnancies after paternal exposure to HCQ, which did not find any increased risk of adverse foetal outcomes.
Previously, four cohort studies and two case series reported on outcomes from ≥2127 pregnancies after paternal exposure to prednisolone, and a case-control study and a case series reported on outcomes from ( n = 4) pregnancies after paternal exposure to methylprednisolone. Overall, the quality of these studies was low, but reassuringly they did not identify an increased risk of adverse foetal outcomes. Since then, an additional study of 2380 paternal exposures to corticosteroids did not identify any statistically significant increase in adverse birth outcomes .
Previous low-quality evidence from outcomes of pregnancies after paternal exposure ( n = 263) to predominantly low-dose MTX did not find any adverse effects . We identified an additional six studies of paternal exposures to MTX ( n = 2026) within three months of conception that similarly found no increased risk of adverse foetal outcomes when compared with MTX-unexposed controls ( n = 4 700 599) . An additional study examined foetal outcomes with paternal MTX (and other DMARD) use compared with abatacept; however, outcomes for individual drugs were not reported .
Our previous review of three cohort [ , , ] and one case-control study reporting on 237 pregnancies after paternal exposure to SSZ did not find an increased risk of adverse foetal outcomes, although the quality of evidence was low. SSZ may also affect male fertility, with oligospermia, reduced sperm motility and increased proportions of abnormal sperm previously reported . No further studies of paternal SSZ exposure were identified in our search.
Previously, we identified a cohort study and case report describing outcomes from ( n = 2) pregnancies after paternal exposure to LEF within three months of conception, and subsequent pregnancy exposure (with intercourse without a condom) in at least one case with no reported washout. No adverse foetal outcomes were observed. An additional study was identified examining foetal outcomes with paternal leflunomide (and other DMARD) use compared with abatacept; however, outcomes for individual drugs were not reported .
In addition to the previous 602 paternal exposures [ , , , ], we identified three studies of n = 2680 pregnancies after paternal exposure to AZA [ , , ]. Overall, no increased risk of adverse foetal outcomes was observed.
In addition to previous studies [ , , , ] on outcomes from pregnancies ( n ≥ 254) after paternal exposure to CsA, we found a Danish population-based cohort study of birth outcomes in 247 children fathered by men treated with CsA before conception . Overall, these studies did not identify an increased risk of adverse foetal outcomes with paternal exposure to CsA.
No additional paternal studies were found to complement previous findings reporting outcomes from pregnancies ( n ≥ 120) after paternal exposure to tacrolimus, which did not identify an increased risk of adverse foetal outcomes [ , , ].
No new studies of paternal exposure to CYC were identified. In addition to a potential long-term impact on spermatogenesis (and hence fertility) in men , there is evidence of an adverse impact on germ cell development and male-mediated teratogenicity from animal studies , although this has not been proven in humans .
In addition to the three previous studies [ , , ] of paternal exposures to MMF ( n ≥ 72), we found three additional studies of pregnancies ( n = 220) after paternal exposure [ , , ]. Overall, the quality of these studies was low, but they did not identify an increased risk of adverse foetal outcomes.
There is no evidence relating to paternal exposure but based on maternal compatibility it is unlikely to be harmful.
Previously, we found five cohort studies [ , , , , ], two case series , two case reports and a case-control study that reported on outcomes from pregnancies ( n = 131) after paternal exposure to INF, ETA and ADA. Overall, the quality of these studies was deemed to be low, but they did not identify an increased risk of adverse foetal outcomes. We identified an additional eight cohort studies reporting outcomes from 620 pregnancies after paternal exposure to TNFi [ , , , ], with no significant findings to suggest adverse foetal outcomes were more likely after TNFi exposure.
We did not identify any further paternal exposures to the 11 found previously , which did not identify any problems in relation to paternal exposure to RTX.
Two studies of n = 15 paternal exposures to TCZ did not find any drug-related effects .
One study of n = 5 paternal exposures to anakinra did not find any drug-related effects . Youngstein et al. reported on five healthy pregnancies to three fathers on long-term canakinumab treatment .
In addition to the previous case of one healthy pregnancy following paternal ABA exposure , a study of clinical trial and post-marketing data submitted to the manufacturer (up to 2014) reported 10 paternal ABA exposures, resulting in nine healthy live births with one elective termination .
Two studies of paternal exposure to secukinumab ( n = 54) and ixekizumab ( n = 34) did not report any adverse drug-related effects .
There were 87 paternal exposures to TOF in a study included in our search . Further details provided by Pfizer reported paternal exposures to TOF in the PsA clinical trials ( n = 3) and post-marketing reports ( n = 3), described above. During the RA and psoriasis clinical trials programmes, 66 men were exposed to TOF with 45 known outcomes, including 37 normal healthy newborns, two premature deliveries (with one subsequent neonatal death) and six spontaneous miscarriages . In the UC clinical trial programmes, there were 19 paternal exposures with 17 known outcomes (15 healthy newborns including one healthy preterm delivery at 34 weeks, and two spontaneous miscarriages) . Data provided by Eli Lilly included eight paternal exposures to BAR in the clinical trials programmes, and outcomes included six full-term healthy newborns and two spontaneous miscarriages . No data were found relating to paternal exposure to BEL, IL-12/23i or anifrolumab. Recommendations for paternal exposure to immunomodulatory drugs Due to the adverse effect of CYC on male fertility, semen cryopreservation is recommended for men prior to paternal exposure (GRADE 1C, SOA 99.5%). Men who take SSZ may have reduced fertility. There is little evidence to suggest that SSZ should be stopped pre-conception, unless conception is delayed by more than 12 months when stopping SSZ should be considered along with other causes of infertility (GRADE 1C, SOA 99.0%). Paternal exposure to the following anti-rheumatic medication is compatible with pregnancy: prednisolone, low-dose (≤25 mg/week) MTX, AZA (GRADE 1B); TNFi, cyclosporin (GRADE 1C); HCQ, LEF, tacrolimus, MMF, IVIG, RTX, IL-6i, IL-1i, ABA, BEL, IL-17i, UST and JAKi (GRADE 2C, SOA 99.3%).
Due to the adverse effect of CYC on male fertility, semen cryopreservation is recommended for men prior to paternal exposure (GRADE 1C, SOA 99.5%). Men who take SSZ may have reduced fertility. There is little evidence to suggest that SSZ should be stopped pre-conception, unless conception is delayed by more than 12 months when stopping SSZ should be considered along with other causes of infertility (GRADE 1C, SOA 99.0%). Paternal exposure to the following anti-rheumatic medication is compatible with pregnancy: prednisolone, low-dose (≤25 mg/week) MTX, AZA (GRADE 1B); TNFi, cyclosporin (GRADE 1C); HCQ, LEF, tacrolimus, MMF, IVIG, RTX, IL-6i, IL-1i, ABA, BEL, IL-17i, UST and JAKi (GRADE 2C, SOA 99.3%).
Implementation Awareness of these guidelines will aid clinical practitioners and patients in decision making and will be raised through presentation at local, regional and national meetings. No barriers to implementation of these guidelines are anticipated. Key standards of care Ideally, patients with rheumatic disease should receive tailored pre-pregnancy counselling and then be reviewed during pregnancy and the four-month post-partum period by clinical practitioners with expertise in the management of rheumatic disease in pregnancy, in addition to their routine obstetric care. They should have access to written information on relevant medications in pregnancy and breastfeeding that is accurate and allows them to make informed decisions regarding compatibility of certain drugs in pregnancy. Future research agenda The limitation of current evidence highlights the need for a national pregnancy registry for patients with rheumatic disease, as currently exists for women with epilepsy. All women with rheumatic disease who become pregnant would be eligible to register, whether or not they are on anti-rheumatic treatment. The prospective pregnancy outcome data would then be published to display information on outcomes such as miscarriage and congenital anomalies in patients treated with anti-rheumatic therapy. These data would also be used to answer specific questions such as the most suitable time to stop MTX pre-conception. Data relating to the impact of paternal exposure to these drugs (both fertility and male-mediated teratogenicity), as well as breastmilk exposure, are particularly limited, and further research in these areas is urgently required. Other research questions include: should bDMARDs with known placental transfer be stopped or switched before/during pregnancy; are tsDMARDs compatible with pregnancy; is it safe to give certain live vaccines to infants ≤6 months after in-utero exposure to bDMARDs with known placental transfer in the third trimester of pregnancy? Mechanism for audit of the guideline An audit pro forma to assess compliance with these guidelines is shown in the audit tool in , available at Rheumatology online.
Awareness of these guidelines will aid clinical practitioners and patients in decision making and will be raised through presentation at local, regional and national meetings. No barriers to implementation of these guidelines are anticipated.
Ideally, patients with rheumatic disease should receive tailored pre-pregnancy counselling and then be reviewed during pregnancy and the four-month post-partum period by clinical practitioners with expertise in the management of rheumatic disease in pregnancy, in addition to their routine obstetric care. They should have access to written information on relevant medications in pregnancy and breastfeeding that is accurate and allows them to make informed decisions regarding compatibility of certain drugs in pregnancy.
The limitation of current evidence highlights the need for a national pregnancy registry for patients with rheumatic disease, as currently exists for women with epilepsy. All women with rheumatic disease who become pregnant would be eligible to register, whether or not they are on anti-rheumatic treatment. The prospective pregnancy outcome data would then be published to display information on outcomes such as miscarriage and congenital anomalies in patients treated with anti-rheumatic therapy. These data would also be used to answer specific questions such as the most suitable time to stop MTX pre-conception. Data relating to the impact of paternal exposure to these drugs (both fertility and male-mediated teratogenicity), as well as breastmilk exposure, are particularly limited, and further research in these areas is urgently required. Other research questions include: should bDMARDs with known placental transfer be stopped or switched before/during pregnancy; are tsDMARDs compatible with pregnancy; is it safe to give certain live vaccines to infants ≤6 months after in-utero exposure to bDMARDs with known placental transfer in the third trimester of pregnancy?
An audit pro forma to assess compliance with these guidelines is shown in the audit tool in , available at Rheumatology online.
are available at Rheumatology online.
keac551_Supplementary_Data Click here for additional data file.
|
Moving Spatially Resolved Multiplexed Protein Profiling toward Clinical Oncology | 2e6752cd-ecde-430f-a01d-a8db3b92abe1 | 10070169 | Internal Medicine[mh] | Since the mid-19th century, the diagnosis and classification of cancer have relied on the description of abnormal cell morphology and tissue structure during review of hematoxylin and eosin (H&E)–stained tumor tissue sections under the light microscope . Ancillary molecular tests, in particular targeted DNA resequencing and IHC, have become more critical in recent years for the identification of specific tumor subtypes. IHC testing uses antibodies to determine the expression of a protein target in a formalin-fixed, paraffin-embedded (FFPE) tissue section. Numerous IHC stains are often required to establish an accurate diagnosis and document the expression of proteins that inform the patient's therapy, such as the estrogen receptor (ER) in breast cancer, V600E-mutant BRAF in melanoma, or programmed cell death ligand 1 (PD-L1) in non–small cell lung cancer (NSCLC). The demand for additional tissue sections often exceeds the amount of available tumor, particularly for small tumor biopsies. This pressure point is likely to intensify as proteolysis targeting chimera (PROTAC), antibody–drug conjugates, and targeted radionuclides are unlocking new protein targets for oncology drug development. “Multiplexed” antibody-based imaging technologies can measure the expression of multiple protein targets in a single tumor section and could significantly mitigate this problem . A second driving force for the development of multiplexed antibody-based imaging assays is the recent discovery in most human cancers of a multitude of coexistent tumor and immune cell functional states, also referred to as “cell phenotypes.” These phenotypes and their transition states were first identified through single-cell RNA sequencing of dissociated tumors and need to be confirmed within the spatial context of the intact tumor microenvironment (TME). It is hoped that a spatially resolved picture of protein expression, posttranslational protein modifications, cell phenotypes, and communities of cell phenotypes will not only advance our understanding of tumor evolution but also guide patient treatment response prediction modeling and the development of novel cancer therapeutics targeting the TME, such as bispecific T-cell engagers, immune-checkpoint inhibitors (ICI), and chimeric antigen receptor (CAR) T cells. The past 5 years have witnessed an explosion of multiplexed antibody-based imaging technologies that can provide simultaneous single-cell quantification of 2 to 60+ proteins while preserving spatial information, revealing features of cell type (e.g., CD8 + T cell), cell function (e.g., PD1 + CD8 + T cell), and cell localization (e.g., stromal PD1 + CD8 + T cell). These approaches can be applied using only one section of tissue from routinely collected FFPE and/or fresh frozen patient biopsies. This article provides a brief overview of multiplexed antibody-based imaging technologies, describes examples of their application in cancer research, and discusses challenges toward implementation of these technologies in clinical oncology. Of note, there are multiplexed technologies that are not imaging-based, capture RNA and protein (e.g., NanoString GeoMx Digital Spatial Profiling; ref. ) or RNA only (e.g., 10X Genomics Visium Spatial Transcriptomics; ref. ), and can provide complementary views. However, the focus of this review will be on multiplexed antibody-based imaging.
The shared feature between all techniques is the use of protein-specific antibodies to detect multiple proteins of interest in a single assay. The primary antibody bound to a protein epitope can be recognized after direct conjugation with a chromogen, fluorophore, or elemental isotope and subsequently detected using chromogenic, immunofluorescent, or mass spectrometry assays, respectively, or by using a secondary antibody to recognize the primary antibody detected in a similar manner. Staining of antibodies is performed either by adding all antibodies to the slide simultaneously or by using a cyclical and/or sequential stain and strip approach . Similarly, imaging of the stain is either performed in a one-step or multistep fashion. The maximal tumor area that can be analyzed is somewhat dependent on the selected method and can range from a single core within a tissue microarray (TMA; millimeters 2 , typically hundreds to thousands of cells) to the entire tissue section (centimeters 2 , typically thousands to millions of cells). Here, we will provide an overview of various multiplexed antibody-based imaging technologies (Supplementary Table S1). Among techniques that use chromogenic detection, multiplexed IHC consecutive staining on single slide (MICSSS) directly builds on IHC. MICSSS uses iterative rounds of staining and chromogen stripping with multistep whole slide imaging. Although MICSSS is a slower assay to complete, it has the advantage of capturing the entire tissue using a standard brightfield scanner without confounding tissue autofluorescence (AF). Additionally, given the widespread use of IHC in the clinical setting, MICSSS has the advantage of generating data most comparable with what pathologists examine in routine diagnostic practice. A growing number of technologies use immunofluorescence (IF) to quantify the expression of multiple proteins in a single tissue section. Cell DIVE, formerly known as MultiOmyx, uses iterative rounds of staining and fluorophore bleaching with multistep whole slide imaging to achieve up to 60 proteins on a single slide. Cell DIVE takes advantage of a background fluorescence image that is captured following each cycle of staining for AF removal, resulting in low AF images. However, this additional step introduces additional imaging time. PhenoImager HT, formerly known as Vectra Polaris, is a high-throughput yet lower dimensional technology, utilizing multistep staining with primary and secondary antibodies with fluorescent tyramide signal amplification to achieve up to nine proteins on a single slide with one-step whole slide imaging. The PhenoImager HT process requires spectral unmixing to deconvolute overlapping fluorescent spectra. CyCIF (cyclic immunofluorescence) is a method that uses cyclical staining/fluorophore bleaching/imaging (like Cell DIVE), uses open-source software that is compatible with many microscopes, and has been reported to measure up to 60 proteins on a single slide with whole slide imaging. Additional IF-based multiplexed imaging technologies have emerged more recently, including PhenoCycler-Fusion (formerly known as CODEX), iterative indirect immunofluorescence imaging (4i), Orion, COMET, iterative bleaching extends multiplexity (IBEX), and InSituPlex (ISP). PhenoCycler-Fusion uses one-step staining with multistep whole slide imaging to achieve up to 60 proteins. Orion uses not only one-step staining but also one-step whole slide imaging to achieve up to 21 proteins. 4i, IBEX, ISP, and COMET all use multistep staining and imaging to achieve up to 40, 65, 12, and 40 proteins, respectively. Although IBEX and ISP can capture the entire slide, 4i and COMET are currently limited to smaller regions. The number of protein markers that can be measured for many of these higher plex methods is limited by reagent availability or maintenance of tissue or epitope integrity throughout the process. The last group of multiplexed antibody-based imaging technologies utilizes mass spectrometry for the detection of multiple proteins in a single slide. These technologies produce high-dimensional data (currently up to 40 proteins) and have the advantage of a one-step staining and imaging protocol, resulting in less hands-on time and tissue damage. However, these detection techniques are limited by comparatively slower imaging times, and much of their application thus far has been limited to small tumor areas in the context of TMAs. The latter group of technologies includes imaging mass cytometry (IMC), multiplexed ion beam imaging (MIBI), and multiplexed ion beam imaging by time of flight (MIBI-TOF).
The ability to measure the expression of multiple proteins at single-cell resolution in situ , with the added advantage of protein colocalization, quantitative reporting, and spatial information, has enabled the identification of distinct functional states for many types of cells and their localization within anatomically defined tumor areas (e.g., the tumor/stroma barrier) or functionally defined “neighborhoods” or “cell communities.” Much of the initial work with multiplexed antibody-based imaging focused on breast cancer (Supplementary Table S2). Using a 32-plex antibody panel and IMC, one study described distinct cell populations and cell interactions and linked them to previously classified breast cancer subtypes (Giesen, Supplementary Table S2). Another study described integrated histologic and immunophenotypic features, including colocalization of multiple markers [e.g., ER, progesterone receptor (PR), Ki-67; Angelo, Supplementary Table S2]. In another study, eight distinct tumor cell phenotypes were identified, linked to established intrinsic breast cancer subtypes, and associated with in vivo tumor uptake of the radiotracer 18F-fluorodeoxyglucose (FDG) using antibodies against members of the glycolysis, hypoxia, and PI3K pathways (Sood, Supplementary Table S2). The expression of 36 proteins in triple-negative breast cancer was measured using MIBI-TOF, revealing archetypes of tumor–immune composition (cold, mixed, compartmentalized) and expression of immunoregulatory proteins in a cell-type and location-specific manner (Keren 2018, Supplementary Table S2). These findings were further extended in a follow-up study (Keren 2019, Supplementary Table S2). A panel of 35 antibodies and IMC were used to define 59 distinct tumor cell phenotypes and distinct microenvironment communities that contained tumor and stromal cell components (Jackson, Supplementary Table S2). In a follow-up study, the authors linked cell phenotypes (including epithelial, stromal, and immune cells) and neighborhoods with comprehensive genomic annotation of the same tumors (Raza Ali, Supplementary Table S2). The inclusion of antibodies against proteins involved in vascular and stromal heterogeneity allowed authors of one study to describe 10 multicellular TME structures that were differentially enriched in distinct breast cancer types and associated with clinical outcomes (Danenberg, Supplementary Table S2). Another study delineated complex cell phenotypes and TME states of normal breast, ductal carcinoma in situ , and invasive breast cancer and identified states that were associated with tumor recurrence (Risom, Supplementary Table S2). Studies in colorectal cancer build on the earlier finding that not only the type and density of T cells but also their intratumoral localization provides prognostic information (ref. ; Supplementary Table S2). One study examined 61 protein antigens in 747 colorectal tumor samples and described patterns of coexpression for multiple members of the mTOR and MAPK signaling pathways (Gerdes, Supplementary Table S2). CODEX was used to measure the expression of 56 proteins, and the authors defined nine conserved cellular neighborhoods that were associated with outcomes in patients with colorectal cancer (Schürch, Supplementary Table S2). In another study, PhenoImager HT was used to characterize the density and spatial distribution of natural killer cells, natural killer T-like cells, and other complex cellular phenotypes and their relationship to patient outcomes in colorectal cancer (Väyrynen, Supplementary Table S2). The ability of multiplexed antibody-based imaging to uncover new aspects of cancer biology has motivated similar studies in other cancer types (Supplementary Table S2). Spatial uniform manifold approximation and projection was developed to identify spatial signatures associated with prognosis in advanced melanoma (Giraldo 2021, Supplementary Table S2). Another study used a combination of IMC and RNA transcript analysis to measure the expression of chemokines and their relationship to immune cell phenotypes and tumor infiltration patterns in metastatic melanoma (Hoch, Supplementary Table S2). Authors of another study reported cell phenotypes and their spatial representation in localized prostate cancer (De Vargas Roditi, Supplementary Table S2).
Another important application of multiplexed antibody-based imaging is the generation of drug response biomarker hypotheses that could inform the clinical development and optimal use of novel cancer therapies. Early tissue-based studies, which did not yet leverage the full potential of these technologies, identified candidate predictive tissue-based biomarkers and documented TME changes in cancer patients treated with ICIs . These types of studies now increasingly use the latest multiplex antibody-based imaging technologies (Supplementary Table S2). Several studies have examined tumor tissue before or during the treatment of cancer patients with antibodies against programmed cell death 1 (PD-1), PD-L1, or cytotoxic T-lymphocyte associated protein 4 (CTLA-4). In one instance, pretreatment biopsies from melanoma patients treated with PD-1 blockade were profiled and the expression of 25 tumor- and immune-related protein markers were reported and associated with progression-free survival and overall survival (Martinez-Morilla, Supplementary Table S2). Another study examined the expression of six markers (PD-1, PD-L1, CD8, FOXP3, CD163, and SOX10/S100) in pretreatment tumor biopsies from patients with melanoma receiving anti–PD-1 therapy and reported cell phenotypes associated with treatment response (Berry, Supplementary Table S2). In a different setting, pretreatment tumor biopsies from melanoma patients receiving ICIs were examined for the expression of 35 protein markers. It was found that the presence of proliferating “antigen-experienced” cytotoxic T cells (CD8 + CD45RO + Ki-67 + ) in close proximity to tumor cells was predictive of response to ICIs (Moldoveanu, Supplementary Table S2). Another study examined the composition and spatial distribution of tumor-infiltrating subsets in baseline tumor biopsies from patients with NSCLC treated with PD-1 blockade (Lopez de Rodas, Supplementary Table S2). The authors reported that a higher density of stromal CD8 + T cells was significantly associated with improved progression-free and overall survival in patients with PD-L1 + tumors. An evaluation of pretreatment tumor samples from patients with Merkel cell carcinoma receiving anti–PD-1 checkpoint blockade revealed that the presence or absence of PD-L1 expression alone was not predictive of response, whereas the density of PD1 + and PD-L1 + cells and the frequency of PD1 + cells in close proximity to PD-L1 + cells correlated with clinical response (Giraldo 2018, Supplementary Table S2). Interestingly, a meta-analysis of studies that assessed the prediction of response to PD-1/PD-L1–directed therapy for cancer suggested that multiplexed antibody-based imaging and multimodal biomarker strategies are superior in predicting treatment response compared with PD-L1 IHC, tumor mutation burden, or gene expression profiling alone . Other studies have examined tumor tissue before or during treatment with other immune-directed therapies. For example, one study examined baseline and on-treatment tumor biopsies from patients with advanced melanoma receiving intratumoral injections of a modified herpes simplex virus type 1 (talimogene laherparepvec, TVEC) and systemic administration of the anti–PD-1 antibody pembrolizumab (Ribas, Supplementary Table S2). The authors reported broad changes in immune cell infiltration and increased expression of PD-L1 following injection of TVEC. In a study of matched untreated and IL2-injected “in-transit” melanoma metastases, changes in cell phenotypes following IL2 administration and predictive biomarkers of IL2 response were examined (Pourmaleki, Supplementary Table S2). Regressed lesions were characterized by nonproliferating CD8 + T cells lacking expression of PD-1, LAG3, and TIM3. In contrast, pretreatment lesions from patients who showed a complete IL2 response were characterized by proliferating CD8 + T cells with an exhausted phenotype (PD-1 + LAG3 + TIM3 + ), stromal B-cell aggregates, and membranous tumor cell expression of MHC-I.
Given the remarkable progress with spatially resolved multiplexed protein profiling technologies, the current pipeline of cancer therapeutics targeting novel proteins existing in complex spatial topologies, and the increasing reliance on molecular data for the accurate diagnosis of cancer, it seems timely to ask what it will take to bring these multiplexed antibody-based imaging technologies to clinical pathology and patients with cancer. First, further progress needs to be made toward harmonizing the reporting of multiplexed antibody-based imaging data . It is often difficult to compare findings from different studies, even studies reporting on the same cancer type using the same methodology. This is due to differences in reagents (i.e., antibody clones and antibody conjugates) and their validation, cell segmentation, managing signal and background, cell type and phenotype assignment, and spatial metrics in addition to the amount and type of publicly accessible source data and often limited clinical annotation of patient samples. Although data reporting must be quantitative, pathologists should be able to easily access the raw data (i.e., images), requiring the need for a standard digital image viewer. Second, there is a need to rigorously test the performance characteristic of each method (e.g., linearity of the assay, sensitivity, specificity, and reproducibility) in clinical samples. These efforts have begun for smaller multiplexed IF panels but must be expanded to larger panels and other technologies. To do so, we must address several questions. How do we distinguish findings that are statistically significant from findings that are not only statistically significant but also biologically meaningful and perhaps even clinically relevant? In planning a study, how much data (e.g., number of patient samples, number of cells, number of proteins) are typically needed to reach certain types of conclusions without overfitting data? And what type of tissue should be included as positive and negative staining controls? Greater priority and resources should be given to prospective Institutional Review Board–approved studies with well-defined biospecimen collection methods, validated antibody panels, and preplanned analysis goals and primary/secondary study endpoints. Third, early consideration should be given to the logistical aspects of translating cancer discovery science to hospital workflows. Most current studies have been conducted in individual research labs, replete with manual steps and individualized decision algorithms. In contrast, most hospital pathology departments heavily rely on automated staining and data analysis pipelines to allow rapid scaling to meet demand and the tight timelines required for clinical decision-making (days). Furthermore, pathologists rapidly scan through tissue sections from multiple tissue blocks to arrive at the correct diagnosis, whereas many multiplexed antibody-based imaging technologies are limited to the analysis of very small preselected areas within each cancer specimen. Technologies that allow for whole slide scanning, ideally from multiple different tumor blocks for each patient, would mirror current hospital practices for histopathologic assessment and may provide a more accurate representation of the intratumoral heterogeneity found in many human cancers. Lastly, it will be important to develop a financial framework that will consider how multiplexed antibody-based imaging diagnostics, perhaps assembled into disease-specific panels, could gradually replace time-honored but biospecimen-intensive and largely “semiquantitative” methods of protein detection. It is also important to consider how multiplexed antibody-based imaging could be efficiently integrated with the genomic annotation of cancer samples to achieve integrated diagnostics, accelerate clinical drug development, and ultimately promote cancer care that is more effective, more affordable, more accessible, and less toxic.
Supplementary Tables S1-S2 Supplementary Table S1: Overview of multiplexed antibody-based imaging technologies.Supplementary Table S2: Application of multiplexed antibody-based imaging in cancer research. Click here for additional data file.
|
Association of Surgical Timing with Outcomes in Early Stage Lung Cancer | bd8637a2-d46f-4409-9814-bb158720efed | 10070299 | Internal Medicine[mh] | The time point at which delays to treatment lead to worse patient outcomes in lung cancer is not entirely clear . Some have suggested worsened outcomes with various degrees of delays, though other evidence suggests that outcomes of patients with particularly short times to surgery may suffer due to rushed preoperative assessments . While prior studies have identified time cutoffs to surgery for lung cancer beyond which patients experience adverse outcomes, these studies examined either small patient cohorts or data gathered from large databases in which time from diagnosis to treatment was not consistently defined . For example, in many cases, date of diagnosis drawn from a national database is the same as the date of surgical resection (the time of tissue diagnosis). These studies also report only on overall survival and do not assess disease recurrence— a more specific finding related to lung cancer treatment. While the specific timing to definitive treatment may not be entirely clear, a systematic review suggests that expedient time to surgery does seem to convey favorable outcomes among patients with early stage lung cancer . In 2021, Heiden et al. published a Veterans Health Administration-based study in which they found worse survival and increased risk of recurrence among Veterans undergoing surgery more than 12 weeks after diagnosis of stage I non-small cell lung cancer (NSCLC). Notably, the authors attempted to standardize the measure of time to surgery (TTS) commencing from most recent suspicious abnormal computed tomography (CT) scan to date of surgery. Neither the Veterans Affairs hospital system nor its patients, however, are necessarily representative of those outside of that system. In this study, 96% of patients were male and average TTS was 10 weeks, which is likely longer than that at many centers . The optimal interval from diagnosis of early stage lung cancer to surgery, therefore, remains a topic of interest. In this study, we aimed to evaluate the association of TTS with outcomes among patients within a large integrated health system undergoing surgical resection for early stage lung cancer. We hypothesized that shorter TTS, measured as the interval from confirmatory imaging or biopsy prior to the operation scheduling date to surgical resection, would correlate with improved outcomes.
Setting In 2015, Kaiser Permanente Northern California (KPNC) implemented centers of excellence (COE) for lung cancer care, reducing the number of facilities that performed lung cancer surgery from 16 to 5 in an effort to promote standardization and specialization of care . Lung cancer diagnosis and treatment consist of a multidisciplinary review board that meets weekly across all medical centers between radiology, pulmonology, oncology, and thoracic surgery. After the diagnosis of stage I or II lung cancer is suspected upon imaging with or without biopsy, the patient and surgeon discuss treatment options. More than 98% of patients undergo pulmonary function testing. Once the surgeon and patient decide to proceed with surgery, designated surgery schedulers determine the earliest available time. All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. The study was approved by the institutional review board of KPNC (IRB: 1,610,759-1) and individual consent was waived because the risk to patients was deemed minimal. Study design and population This retrospective cohort study consists of patients aged 18–84 years who underwent elective surgery for stage I or II lung cancer between 1/1/2009 and 12/31/2019 who also had a CT or positron emission tomography (PET) scan of the lungs in the 6 months prior to surgery. Patients were identified using our regional cancer registry, which adheres to data standards of the Surveillance, Epidemiology, and End Results (SEER) program. We defined TTS as the time period from the confirmatory testing date (imaging or biopsy) prior to scheduling of surgery to the surgery date. We believe this is a more standardized definition given that some, but not all, patients with suspicious lesions undergo preoperative biopsy in our system. The last biopsy or scan prior to the surgery scheduling date was chosen as the start point for the TTS interval as this was deemed the point at which surgeons had the necessary staging and diagnostic information to confirm high likelihood of stage I or II lung cancer and therefore recommended surgery. Eligible scans of the lungs included CT with or without intravenous contrast, CT angiography, CT guidance for biopsies, or PET scans. Fleischner Society criteria for pulmonary nodules are used with weekly evaluation by a multidisciplinary panel of radiologists, pulmonologists, thoracic surgeons, and oncologists to define suspicious pulmonary nodules . Patients with less than one year of prior health plan membership were excluded to ensure capture of comorbidities, and patients with prior documented lung cancer were excluded. In accordance with National Comprehensive Cancer Network recommendations, our early stage lung cancer patients proceed directly to surgery . Hence, both stage I and II disease were included to be inclusive of all disease in which upfront surgery is performed. Patients downstaged to stage I or II disease with neoadjuvant chemotherapy in the year prior to surgery were excluded. We assessed three outcomes: unplanned return to care, all-cause mortality, and recurrence. Unplanned return to care included any emergency room visits, readmissions, or reoperations in the 30 days following lung cancer surgery. Death information was available using an internal database which aggregates data on vital status from KPNC hospitals, social security files, California state death registry, and the National Death Index. We followed patients from the date of surgery until the first of death (outcome), disenrollment from the health plan, the end of data availability on 12/31/2019, or the end of 5 years of follow-up. Recurrence information was collected from the KPNC cancer registry which began systematically collecting cancer recurrence data in 2016. Therefore, we only assessed recurrence as an outcome for our lung cancer cases that were diagnosed 2016–2019. The main exposure of interest was TTS, as defined above. Based on prior research and our own internal data, we assessed three separate cut-points for TTS, 2 weeks (≤ 4 days), 4 weeks (≤28 days), and 12 weeks (≤ 4 days) . The 2-week cut-point was chosen based off our institution’s internal quality benchmarks set prior to the study start date which target a time from surgical consultation to surgery of less than 2 weeks. This is also in line with the Cancer Care Ontario guidelines for cancer surgery . The 4-week cut-point was chosen based on our own institution’s initial feasibility data which is similar to but shorter than the 38 days suggested by Yang et al.. The 12-week cut-point was chosen based on the previously mentioned study by Heiden et al. . Other variables included in our models were patient age at surgery, sex, race/ethnicity, Charlson comorbidity score, diagnosis of non-lung cancer in the year prior to surgery, smoking history (ever vs never), cancer stage, histology, performance of mediastinoscopy, location, year, and hospital of surgery. Histology was grouped into adenocarcinoma, squamous cell, and other histologies. Year of surgery was separated into 2009–2014 and 2015–2019 based on the regionalization of COE for lung surgery occurring in 2015. Statistical analysis Logistic regression was used to model the association of TTS with any unplanned return to care within 30 days of surgery. Cox proportional hazards modeling was used to model associations with death and recurrence. The models for recurrence were restricted to only surgeries performed from 2016 to 2019. All models adjusted for all of the covariables defined above. Hospital of surgery was included in all models as a random intercept (cluster variable) to account for variations in practice by hospital. The proportional hazards assumption was assessed for all cut-point values (2, 4, 12 weeks) in models for death and recurrence.
In 2015, Kaiser Permanente Northern California (KPNC) implemented centers of excellence (COE) for lung cancer care, reducing the number of facilities that performed lung cancer surgery from 16 to 5 in an effort to promote standardization and specialization of care . Lung cancer diagnosis and treatment consist of a multidisciplinary review board that meets weekly across all medical centers between radiology, pulmonology, oncology, and thoracic surgery. After the diagnosis of stage I or II lung cancer is suspected upon imaging with or without biopsy, the patient and surgeon discuss treatment options. More than 98% of patients undergo pulmonary function testing. Once the surgeon and patient decide to proceed with surgery, designated surgery schedulers determine the earliest available time. All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. The study was approved by the institutional review board of KPNC (IRB: 1,610,759-1) and individual consent was waived because the risk to patients was deemed minimal.
This retrospective cohort study consists of patients aged 18–84 years who underwent elective surgery for stage I or II lung cancer between 1/1/2009 and 12/31/2019 who also had a CT or positron emission tomography (PET) scan of the lungs in the 6 months prior to surgery. Patients were identified using our regional cancer registry, which adheres to data standards of the Surveillance, Epidemiology, and End Results (SEER) program. We defined TTS as the time period from the confirmatory testing date (imaging or biopsy) prior to scheduling of surgery to the surgery date. We believe this is a more standardized definition given that some, but not all, patients with suspicious lesions undergo preoperative biopsy in our system. The last biopsy or scan prior to the surgery scheduling date was chosen as the start point for the TTS interval as this was deemed the point at which surgeons had the necessary staging and diagnostic information to confirm high likelihood of stage I or II lung cancer and therefore recommended surgery. Eligible scans of the lungs included CT with or without intravenous contrast, CT angiography, CT guidance for biopsies, or PET scans. Fleischner Society criteria for pulmonary nodules are used with weekly evaluation by a multidisciplinary panel of radiologists, pulmonologists, thoracic surgeons, and oncologists to define suspicious pulmonary nodules . Patients with less than one year of prior health plan membership were excluded to ensure capture of comorbidities, and patients with prior documented lung cancer were excluded. In accordance with National Comprehensive Cancer Network recommendations, our early stage lung cancer patients proceed directly to surgery . Hence, both stage I and II disease were included to be inclusive of all disease in which upfront surgery is performed. Patients downstaged to stage I or II disease with neoadjuvant chemotherapy in the year prior to surgery were excluded. We assessed three outcomes: unplanned return to care, all-cause mortality, and recurrence. Unplanned return to care included any emergency room visits, readmissions, or reoperations in the 30 days following lung cancer surgery. Death information was available using an internal database which aggregates data on vital status from KPNC hospitals, social security files, California state death registry, and the National Death Index. We followed patients from the date of surgery until the first of death (outcome), disenrollment from the health plan, the end of data availability on 12/31/2019, or the end of 5 years of follow-up. Recurrence information was collected from the KPNC cancer registry which began systematically collecting cancer recurrence data in 2016. Therefore, we only assessed recurrence as an outcome for our lung cancer cases that were diagnosed 2016–2019. The main exposure of interest was TTS, as defined above. Based on prior research and our own internal data, we assessed three separate cut-points for TTS, 2 weeks (≤ 4 days), 4 weeks (≤28 days), and 12 weeks (≤ 4 days) . The 2-week cut-point was chosen based off our institution’s internal quality benchmarks set prior to the study start date which target a time from surgical consultation to surgery of less than 2 weeks. This is also in line with the Cancer Care Ontario guidelines for cancer surgery . The 4-week cut-point was chosen based on our own institution’s initial feasibility data which is similar to but shorter than the 38 days suggested by Yang et al.. The 12-week cut-point was chosen based on the previously mentioned study by Heiden et al. . Other variables included in our models were patient age at surgery, sex, race/ethnicity, Charlson comorbidity score, diagnosis of non-lung cancer in the year prior to surgery, smoking history (ever vs never), cancer stage, histology, performance of mediastinoscopy, location, year, and hospital of surgery. Histology was grouped into adenocarcinoma, squamous cell, and other histologies. Year of surgery was separated into 2009–2014 and 2015–2019 based on the regionalization of COE for lung surgery occurring in 2015.
Logistic regression was used to model the association of TTS with any unplanned return to care within 30 days of surgery. Cox proportional hazards modeling was used to model associations with death and recurrence. The models for recurrence were restricted to only surgeries performed from 2016 to 2019. All models adjusted for all of the covariables defined above. Hospital of surgery was included in all models as a random intercept (cluster variable) to account for variations in practice by hospital. The proportional hazards assumption was assessed for all cut-point values (2, 4, 12 weeks) in models for death and recurrence.
A total of 2861 patients met inclusion criteria. Baseline characteristics can be found in Tables , . Of all patients, 52% were treated in the later time period, 61% were female, and 78% had stage 1 disease. Notably, while mediastinoscopy staging was performed in 10.9% of all patients, it was not associated with TTS when compared to patients without mediastinoscopy ( p = 0.48). Also, while minimally invasive adenocarcinoma and bronchioalveolar carcinoma made up 5.70% of cases, there was no association with these subtypes and time to surgery compared to cases with other histologic subtypes ( p = 0.17). In adjusted logistic regression models, patients with TTS greater than 12 weeks had decreased odds of unplanned return to care within 30 days compared to patients with TTS 12 weeks or shorter (odds ratio (OR) 0.69, 95% confidence interval (CI) 0.49–0.98) (Fig. , Table ). Models with cut-points at 2 (OR 1.09, 95% CI 0.82–1.44) or 4 weeks (OR 1.08, 95% CI 0.89–1.32) did not have statistically significant values for an association between TTS and unplanned return to care. Patients with TTS greater than 4 weeks had an increased rate of both death (HR 1.18, CI 1.00–1.39) and recurrence (HR 1.33, 1.10–1.62). Though not statistically significant, the results for 2- and 12-week cut-points did follow the same pattern with patients with longer TTS trending toward higher rates of death and recurrence (2-week cut-point for death model HR 1.23, CI 0.93–1.64; 12-week cut-point for death model HR 1.35, CI 0.97–1.88; 2-week cut-point for recurrence model HR 1.54, 0.85–2.80; 12-week cut-point for recurrence model HR 2.28, CI 0.80–6.46). Of all deaths, 73.7% were due to lung cancer. Average lengths of follow-up among patients in the 2009–2014 and 2015–2019 cohorts who died were 4.0 and 2.2 years, respectively. Patients had average follow-up for recurrence of 1.7 years with a maximum follow-up of 4 years and an interquartile range of 0.74–2.6 years. In addition to TTS, other variables were associated with our outcomes of interest . Black patients and patients with middle or lower-lobe cancer had reduced odds of having an unplanned return to care within 30 days, while patients with smoking history or squamous cell histology were more likely to have an unplanned return to care. Black patients were also less likely than White patients to die during follow-up, although Black patients had a slightly shorter average length of follow-up (2.8 years vs 3.1 years). Higher rates of recurrence were found in Asian (HR 2.13, CI 1.46–3.11) and Latinx (HR 1.26, CI 1.01–1.57) patients. Of the 2861 patients, 76.8% underwent lobectomy. Given this large majority, we performed a sub-group analysis of these patients at the 4-week time cut-point. We found similar results to our overall results with significantly increased hazard ratios for death and recurrence (1.17, CI 1.00–1.38 and 1.34, CI 1.09–1.65, respectively).
Our study suggests that TTS longer than 4 weeks is associated with increased rate of recurrence and death. The use of confirmatory testing prior to surgery scheduling as a starting point allows for a standardized definition for TTS. In 2008, Gould et al . used the concept of interval between imaging and treatment initiation to define time to treatment; however, they used the initial chest radiograph or CT scan in which a suspicious finding was later confirmed to be lung cancer. We believe this designated interval does not allow for the same degree of standardization for comparison. Suspicious findings may undergo various time intervals of lung nodule surveillance or workup prior to the decision to biopsy or treat, and evidence suggests that this time period of evaluation for unclear imaging findings may not have a bearing on patient outcome as long as the patient maintains continuity of care . Our study also avoids the potential systematic bias in which tissue diagnosis alone defines the beginning of the TTS interval. In these cases, if tissue diagnosis is performed at the time of surgery, TTS is zero. This aberrantly affects any comparison to TTS for patients with tissue diagnosis prior to surgery. Additionally, Tang et al. found that patients with tissue diagnosis at the time of surgery generally had smaller tumors and were better surgical candidates than those with prior tissue diagnoses emphasizing that these are not comparable groups. Our study shares similarities to that by Heiden et al . in that both examine a large patient population, define a standardized interval prior to surgery to define TTS, and find that a delay in TTS negatively impacts risk of recurrence and mortality. Our study, however, includes a population more generalizable to the population at large, includes more patients treated within 4 weeks, and, critically, finds that the TTS at which patients experience worse outcomes begins at 4 weeks rather than 12. Additionally, Heiden et al. do not account for patients who undergo preoperative biopsy, so despite their attempts at standardization, some patients may have undergone preoperative biopsy while others went directly to surgery. By using either biopsy or last imaging, our definition of TTS further standardizes this time interval and provides a more comparable starting point. The TTS within our system is affected by our internal standards process, prompt access to virtual preoperative anesthesia evaluation appointments, and ability to quickly book operative cases with a multitude of surgery schedulers in each surgical department. Also, our system allows for quick transfer of information via a standardized staff messaging system integrated for all of KPNC, thus minimizing delays for physicians and staff when accessing imaging, discussing patient information, and communicating with patients. Our internal standards emphasizing expedient surgical scheduling likely play a role in why mediastinal staging via mediastinoscopy is not associated with delays to surgery. The finding that TTS shorter than 2 weeks does not improve outcomes may occur for several reasons. It is possible that the evaluation and medical optimization process which may consist of further cardiopulmonary testing prior to surgery may have been rushed in a subset of these patients or omitted completely due to time constraints. It could also be that features on the imaging for some patients were more aggressive in appearance prompting the surgeon to opt for earlier surgical intervention for more aggressive disease. Also, given the range of the confidence intervals of this group’s outcomes, some patients may have benefited from the shorter TTS while others may not have. If this is the case, identifying these patients would provide an area for further improvement. Additionally, there was a much smaller number of patients with surgery within 2 weeks, so our findings may also be due to reduced statistical power for this group. The finding that a TTS shorter or longer than 12 weeks was not associated with survival or recurrence is interesting given the prior literature. Only 5% of our patients had TTS longer than 12 weeks and the variability in the confidence intervals of all three outcomes in this group was also relatively large. It is possible that some of these patients would have benefitted from earlier surgery while others would not have. Specific features on imaging were not collected, but it is possible that some of these patients had features that, while concerning enough to prompt surgery, were less aggressive in appearance. Interestingly, patients with TTS longer than 12 weeks were the only ones that were less likely to have an unplanned return to care within 30 days of surgery, perhaps due to increased time for medical optimization leading to less complicated post-operative recoveries. Additionally, characteristics that may appear less aggressive on imaging likely include smaller size, more peripheral location, and more ground glass component rather than solid. Such tumors may require less technically complicated dissections and, in some cases, less lung being resected with subsequent lower likelihood for post-operative issues that would lead to an unplanned return to care within 30 days. While preoperative workup and comorbidity status may influence TTS and death, the distribution of Charlson comorbidity scores was similar between patients with TTS ≤ and > 4 weeks. Also, the vast majority of deaths were related directly to lung cancer. Notably, our study included 61% females which contrasts with the 3.7% included in the study by Heiden et al . and broadens the generalizability of our results. Also, Black patients experienced reduced odds of unplanned return to care within 30 days and had reduced risk of death during follow-up compared to White patients. Asian and Latinx patients had increased risk of recurrence compared to White patients. These findings warrant further investigation. Of note, the group prior to regionalization experienced higher mortality, but, due to the study end-point at the end of 2019, the average follow-up duration was nearly twice as long in this group. Given that lobectomy is commonly used for definitive resection of early stage lung cancer, the sub-group analysis of only these patients further validates our results at the 4-week cut-point and allows for a more granular interpretation of our data. We believe this allows our results to be more easily translatable to clinical practice. There are a few additional limitations to our study. Given the retrospective design, we cannot definitively rule out the possibility of unmeasured confounders leading to the associations observed. Also, while this TTS definition allows for improved standardization and comparison among patient groups, it is not perfect. Comparison with this metric alone does not take into account more or less worrisome features on imaging that may affect how quickly a patient is scheduled for surgery such as nodule type, ground glass appearance, the presence of spiculations, location, and SUV. Such imaging characteristics along with histological characteristics such as tumor grade that may suggest aggressive lesions are likely treated with more urgency than those that appear more indolent. Additionally, our recurrence data were only available from 2016 to 2019, with a maximum follow-up for recurrence of four years. However, recurrence data are not available prior to 2016. Lastly, we do not have information on the exact reason initial imaging was performed for each case. We plan to further investigate this to distinguish whether imaging was incidental, conducted for screening, or prompted by symptoms. Stage I and II lung cancer patients with TTS of less than or equal to 4 weeks experienced lower rates of recurrence and a trend toward decreased mortality during follow-up than those with TTS greater than 4 weeks. Expeditious workup without compromising optimization of comorbidities may play a role in improving outcomes among patients with early stage lung cancer.
|
A novel Raman spectroscopic method for detecting traces of blood on an interfering substrate | 7ed33a8e-8814-496d-af26-1481b6aa108a | 10070500 | Forensic Medicine[mh] | Body fluid traces discovered at a crime scene play a significant role in reconstructing the event and are the primary source of DNA, RNA, etc. The majority of current methods for body fluid detection and identification are based on biochemical reactions . Several presumptive and confirmatory tests have been developed for bloodstains, which are often found at the scenes of violent crimes. Presumptive blood tests, which can be conducted at the scene, are mainly based on the peroxidase catalysis of hemoglobin (Hb) from red blood cells. These tests can potentially result in false positives caused by environmental oxidants , . Confirmatory tests for blood, including Teichmann and Takayama hemoglobin crystal tests, and immunological tests, such as ELISA and LDH assays, are labor intensive and costly and require a laboratory environment . Several emerging technologies have been recently developed for body fluid identification, including blood. Liquid chromatography–mass spectrometry and capillary electrophoresis can provide confirmatory identification of all main body fluids. However, these tests are time-consuming and require extensive sample preparation and a laboratory setting , . The analysis of mRNA expression has also been introduced in forensic science as a tool to identify body fluids and tissues due to its specificity and sensitivity by targeting RNA sequencing of upregulated biomarkers. These RNA assays have successfully expanded into the study of multiplex body fluid samples potentially found in sexual assault cases , . Spectroscopic methods such as IR, UV‒Vis absorption, and fluorescence have been shown to have great potential for detecting and identifying body fluid traces – . These techniques are nondestructive and could be applied at a crime scene since portable commercial instruments are available. Among these new methods, Raman spectroscopy appears very attractive as a universal, confirmatory method for the identification of all forensically relevant body fluids due to its specificity, ease of use, required minimal sample preparation, and possibility of being conducted at the scene of a crime , – . The benefits of Raman spectroscopy in forensics include the possibility to work with a small amount of material, as low as several picograms or femtoliters, high sensitivity to a sample’s chemical composition and structure, and a noncontacting and nondestructive method of analysis. Raman spectroscopy is already used by law enforcement agencies for confirmatory drug identification, trace evidence, paint and fiber analysis, etc. , . Chemometric analysis combined with Raman spectroscopy allows for the confirmatory identification of bloodstains , , determining the time since deposition , differentiating human and animal blood , and providing phenotypic information about the donor , . The specificity of body fluid trace detection at a crime scene can be affected by an underlying surface (substrate) such as floor tile, paper tissue, or contaminants, which can contribute to Raman scattering . The substrate's surface energy, the interaction between the body fluid and substrate, determines the wetting and affects the final morphology of the dried biofilm , . A substrate can produce Raman scattering that is stronger by orders of magnitude compared to a body fluid signal. To implement Raman spectroscopy in practical forensics, the interference signal from common substrates must be overcome . A popular experimental approach to avoid substrate interference is restoring an initial state of body fluid by a sample soluting in water . However, this is time-consuming and destructive because adding water to dried body fluid, accompanied by chemical reactions, can affect Raman spectra. Therefore, the vital problem of body fluid trace identification is the interference from a substrate. This problem can be solved in two ways: considering a substrate as an additional component in a combination of "sample & substrate" or extracting Raman spectra of a target body fluid sample from this combination without defining substrate characteristics. The former can be realized through methods similar to a multivariate curve resolution based on a bilinear model of a complex mixture spectrum in the form of a superposition of contributions of pure components – . In common, the problem is described by an equation set: 1 [12pt]{minimal}
$${}={}{{}}^{t},$$ W = C S t , where [12pt]{minimal}
$${}$$ S is the matrix of all component spectra in a composition, [12pt]{minimal}
$${}$$ C is the matrix of concentrations, and [12pt]{minimal}
$${}$$ W is the matrix of experimental spectra . Here, superscript character [12pt]{minimal}
$$t$$ t means matrix transposition. One of the main issues here is to have standard Raman spectra of a body fluid and a substrate separately. The latter can be solved easily using consequent measurements. The only way to acquire the standard spectrum of a body fluid is to measure it using a minimally interacting substrate. Boyd et al. compared Raman scattering from blood samples deposited on various substrates, including borosilicate glass, a silicon wafer, a polyethylene cup, and a microscope slide coated with commercial aluminum foil. Raman scattering peaks from all substrates, except aluminum foil, were detected. Therefore, the AI substrate is the most suitable for recording standard Raman spectra of targeted substances. This approach was applied to differentiate multicomponent Raman spectra and exclude interference from substrate contributions – . Sikirzhytskaya et al. successfully used alternating least squares statistics and multivariate curve resolution to decode blood signatures in the experimental Raman spectra of biological samples in the presence of contaminants. Gautam et al. used partial least squares discriminant analysis to distinguish the age of blood samples with high accuracy in the presence of polymer substrate interference. They used a rather strong assumption that the polymer is homogeneous and produces the same contribution to all spectra. The identification of a target body fluid on an interfering substrate without defining its characteristics (knowledge of [12pt]{minimal}
$${}$$ S is not complete) is more attractive. In this situation, Eq. ( ) can be solved for the case when we have experimental spectra for the compositions with varied concentrations of some components in a mixture during its evolution, for example, associated with a chemical process (Manne condition in a concentration space, see Fig. .) A Manne condition means that concentrations of two components in a mixture can be identified if intervals of evolution variable corresponding to their function [12pt]{minimal}
$$f(t)$$ f t nonzero values (the function carrier shown as a rectangle in Fig. ) do not overlap. In fact, this condition means that the concentration of a specific component can be restored if, during this mixture evolution, there is a situation when the concentration of the remaining components is zero. The latter is hardly implemented for the interfering substrate because it means that we should have a spatial point where substrate impact is absent. Of course, the opposite task of substrate characteristic identification can be easily solved by measuring at a spatial point on the substrate surface where a biofluid stain is absent. A weaker version of this condition can be fulfilled for a target component by combining multivariate curve resolution with the addition method (MCRAD) , . The latter can be implemented by varying the concentration of a target component by chemical manipulations or virtually (by computer simulations). The benefit of the MCRAD is that only the target component concentration has to be varied. Therefore, we do not need any information or special conditions for the interfering substrate. Another approach to extract a certain component concentration from an IR absorption spectrum of a complex gas mixture was developed by us , . The approach starts from degenerating Eq. ( ) in the following form: 2 [12pt]{minimal}
$${S}_{org}(k)={S}_{blank}(k)+C _{ref}(k),$$ S org k = S blank k + C · S ref k , where [12pt]{minimal}
$${S}_{org}$$ S org is an experimental spectrum, [12pt]{minimal}
$${S}_{ref}(k)$$ S ref k is a spectrum of a target component, [12pt]{minimal}
$$C$$ C is its concentration (or any other quantitative characteristic of this component volume fraction), which is a priori unknown, and [12pt]{minimal}
$${S}_{blank}$$ S blank is an unknown spectrum of other components in a mixture. [12pt]{minimal}
$$k$$ k is a wavenumber (Raman shift). This approach uses the concentration restoration criterion for a specific component based on reducing the spectrum complexity ( RSC ) when the spectral component is removed from the experimental spectrum (see Fig. ) . This criterion is associated with the minimization of the following functional: 3 [12pt]{minimal}
$$ f()= |_{org}- {S}_{ref})}{dk}|dk.$$ δ f C ~ = ∫ d ( S org - C ~ · S ref ) dk d k . Here, one needs to know the spectrum of the target component, and the latter has to have spectral peculiarities relative to other components. The latter is the same Manne condition but in a spectral space, which resembles the condition of applicability of DIAL (differential absorption LIDAR) or DOAS (differential optical absorption spectroscopy) approaches to study the molecular composition of the atmosphere using multifrequency absorption data. The MCRAD and RSC implement a "one-per-step" decomposition approach, which is more suitable for practical use. It should be noted that some variation of MCRAD has already been used for recovering a known Raman spectral component from a complex matrix , , while RSC has not been used yet for this purpose. This work investigated the capability, limitations, and benefits of the "one-per-step" decomposition model for the detection and correct identification of blood traces on interfering substrates using Raman spectroscopy. We applied MCRAD and RSC to Raman spectral data obtained for bloodstains on various common substrates, pure bloodstains, and pure substrates. The RSC method detected blood with a confidence probability close to 100%. The MCRAD method was shown to demonstrate a poor ability to detect bloodstains on blue polyester, denim, white polyester, and cotton fabric. The control studies aimed at apparent blood detection on pure substrates. Both methods demonstrated a good but not perfect ability to prove that bloodstains are absent on pure substrates. In our opinion, false positive errors are associated with a similarity between blood and substrate Raman spectra. We illustrated this conclusion using the Soergel distance between Raman spectra of blood and a substrate.
Bloodstain identification on interfering substrates To simulate realistic bloodstain evidence, which is typically recovered at the scene of a crime, droplets of whole blood of 10-μL volume were deposited on the surface of white cotton fabric, white polyester fabric, blue polyester fabric, and denim fabric using a micropipette. The bloodstains were left to dry overnight under ambient conditions. A bloodstain on aluminum foil was used as a standard sample on a noninterfering substrate . Automatic mapping was used to collect multiple Raman spectra from different spots of the sample to probe potential sample heterogeneity . Selected Raman spectra of bloodstains on various substrates as well as Raman spectra of the substrates are shown in Fig. . The Raman spectrum of blood on Al foil is consistent with the pure blood spectra reported previously . Spectra of bloodstains on various substrates show a significant contribution from substrates. The Raman spectrum of a bloodstain on denim is dominated by denim, which further illustrates the need for special data analytics to detect blood traces on such interfering substrates. The origin of specific blood Raman peaks is as follows. The pronounced peak at 1658 cm −1 corresponds to the amide I vibrations in a peptide chain. The peak at 1003 cm −1 and a doublet at 826 and 856 cm −1 corresponds to phenylalanine and tyrosine. The band at 754 cm −1 is associated with the pyrrole ring. The carbohydrates provide Raman peaks near 960, 1032, 1127 and 1208 cm −1 related to the stretching of C–O, C–C, C–O–H and C–O–C bonds. Peaks detected at 1449 and near 1340 cm −1 can be associated with lipoproteins but their content has individual variability. The Raman bands at 623 and 644 cm −1 refer to phenylalanine and tyrosine, respectively , , . The denim fabric most intense Raman peak at 1573 cm −1 is attributed to the indigo. Raman bands from 1030 to 1150 cm −1 and at 1380, 1340, 1090 and 460 cm −1 correspond to cotton fibers , . These bands are presented in white cotton Raman spectrum. For blue polyester, the Raman band at 1725 cm −1 corresponds to the stretching of the carbonyl group C=O, the band at 1612 cm −1 corresponds to C–C vibrations in the aromatic ring. The 702 cm −1 band also corresponds to the stretching of the C–C bonds in the ring. The Raman bands at 859 cm −1 , 998 cm −1 , 1096 cm −1 , 1179 cm −1 , 1291 cm −1 , 1416 cm −1 , 1463 cm −1 belong to a polyethylene terephthalate . For white polyester, the Raman bands at 1637cm −1 , 1440 cm −1 , 1080 cm −1 , 1280 cm −1 , 1300 cm −1 , 1128–1060 cm −1 , 1235 cm −1 are associated with nylon stripes . A Raman spectrum of a bloodstain on an interfering substrate is described by Eqs. ( ) or ( ), where [12pt]{minimal}
$$C$$ C is the volume fraction (VF) of the blood. The results of the application of MCRAD and RSC for the set of experimental Raman spectra of bloodstains on tested substrates are shown in Fig. . Calculations were conducted for a full Raman spectral dataset for a bloodstain on each common substrate and noninterfering Al foil. The latter was considered the blood spectral standard. The results of blood volume fraction restoration are presented in the form of the probability density function [12pt]{minimal}
$$f(C):$$ f C : [12pt]{minimal}
$$ f(C)dC=1,$$ ∫ f C d C = 1 , which characterizes the distribution of restored blood volume fraction values. The restored volume fractions are defined by all combinations of experimental Raman spectra of a bloodstain on a specific substrate and experimental Raman spectra of a bloodstain on an Al foil. Further data preprocessing included the calculation of a mean value and standard deviation for every value of restored volume fraction [12pt]{minimal}
$$C.$$ C . It was found that the MCRAD predicted mean values of [12pt]{minimal}
$$C$$ C close to zero, while the RSC predicted a mean value of approximately 0.1 for the bloodstain on the blue polyester, 0.4 for denim and white polyester, and 0.6 for cotton fabric. Notably, these results were obtained for samples containing bloodstains on the substrates. Therefore, MCRAD gave a quantitatively incorrect result (false negative). To further validate this conclusion using a statistical approach, we evaluated the hypothesis of the absence of blood on a substrate using the standard score criterion , : Z = (0 − μ)/σ, where μ is the mean value in a dataset and σ is the standard deviation. Here, the Z score shows how far the mean value of an experimental random parameter is from zero on a scale of the standard deviation. In other words, the larger |(0 − μ)|/σ is, the more confidently we can say that the estimated parameter is different from zero. The results of the Z score calculations and the confidence probability P of the blood absence in the sample are shown in Table for each of the distributions [12pt]{minimal}
$$f(C),$$ f C , which are presented in Fig. . These distributions were calculated using the MCRAD and RSC methods for all combinations of every Raman spectrum of a bloodstain on an Al foil with every Raman spectrum of a bloodstain on a corresponding substrate. After that, the mean value and standard deviation were calculated. If we choose the confidence level of 95%, it will correspond to the interval from − 1.96 to 1.96 in Table . The confidence probabilities of blood absence in a sample calculated according to Z scores are presented in Table . Therefore, the MCRAD method with a confidence probability of not less than 95% demonstrates the absence of blood for the bloodstains on blue polyester and denim. The same predictions are fulfilled for white polyester with a confidence probability of 59%. The MCRAD predicts blood presence on a cotton fabric with a confidence probability of 91%. The RSC method demonstrates the presence of blood for the same samples with a confidence probability close to 100%.
To simulate realistic bloodstain evidence, which is typically recovered at the scene of a crime, droplets of whole blood of 10-μL volume were deposited on the surface of white cotton fabric, white polyester fabric, blue polyester fabric, and denim fabric using a micropipette. The bloodstains were left to dry overnight under ambient conditions. A bloodstain on aluminum foil was used as a standard sample on a noninterfering substrate . Automatic mapping was used to collect multiple Raman spectra from different spots of the sample to probe potential sample heterogeneity . Selected Raman spectra of bloodstains on various substrates as well as Raman spectra of the substrates are shown in Fig. . The Raman spectrum of blood on Al foil is consistent with the pure blood spectra reported previously . Spectra of bloodstains on various substrates show a significant contribution from substrates. The Raman spectrum of a bloodstain on denim is dominated by denim, which further illustrates the need for special data analytics to detect blood traces on such interfering substrates. The origin of specific blood Raman peaks is as follows. The pronounced peak at 1658 cm −1 corresponds to the amide I vibrations in a peptide chain. The peak at 1003 cm −1 and a doublet at 826 and 856 cm −1 corresponds to phenylalanine and tyrosine. The band at 754 cm −1 is associated with the pyrrole ring. The carbohydrates provide Raman peaks near 960, 1032, 1127 and 1208 cm −1 related to the stretching of C–O, C–C, C–O–H and C–O–C bonds. Peaks detected at 1449 and near 1340 cm −1 can be associated with lipoproteins but their content has individual variability. The Raman bands at 623 and 644 cm −1 refer to phenylalanine and tyrosine, respectively , , . The denim fabric most intense Raman peak at 1573 cm −1 is attributed to the indigo. Raman bands from 1030 to 1150 cm −1 and at 1380, 1340, 1090 and 460 cm −1 correspond to cotton fibers , . These bands are presented in white cotton Raman spectrum. For blue polyester, the Raman band at 1725 cm −1 corresponds to the stretching of the carbonyl group C=O, the band at 1612 cm −1 corresponds to C–C vibrations in the aromatic ring. The 702 cm −1 band also corresponds to the stretching of the C–C bonds in the ring. The Raman bands at 859 cm −1 , 998 cm −1 , 1096 cm −1 , 1179 cm −1 , 1291 cm −1 , 1416 cm −1 , 1463 cm −1 belong to a polyethylene terephthalate . For white polyester, the Raman bands at 1637cm −1 , 1440 cm −1 , 1080 cm −1 , 1280 cm −1 , 1300 cm −1 , 1128–1060 cm −1 , 1235 cm −1 are associated with nylon stripes . A Raman spectrum of a bloodstain on an interfering substrate is described by Eqs. ( ) or ( ), where [12pt]{minimal}
$$C$$ C is the volume fraction (VF) of the blood. The results of the application of MCRAD and RSC for the set of experimental Raman spectra of bloodstains on tested substrates are shown in Fig. . Calculations were conducted for a full Raman spectral dataset for a bloodstain on each common substrate and noninterfering Al foil. The latter was considered the blood spectral standard. The results of blood volume fraction restoration are presented in the form of the probability density function [12pt]{minimal}
$$f(C):$$ f C : [12pt]{minimal}
$$ f(C)dC=1,$$ ∫ f C d C = 1 , which characterizes the distribution of restored blood volume fraction values. The restored volume fractions are defined by all combinations of experimental Raman spectra of a bloodstain on a specific substrate and experimental Raman spectra of a bloodstain on an Al foil. Further data preprocessing included the calculation of a mean value and standard deviation for every value of restored volume fraction [12pt]{minimal}
$$C.$$ C . It was found that the MCRAD predicted mean values of [12pt]{minimal}
$$C$$ C close to zero, while the RSC predicted a mean value of approximately 0.1 for the bloodstain on the blue polyester, 0.4 for denim and white polyester, and 0.6 for cotton fabric. Notably, these results were obtained for samples containing bloodstains on the substrates. Therefore, MCRAD gave a quantitatively incorrect result (false negative). To further validate this conclusion using a statistical approach, we evaluated the hypothesis of the absence of blood on a substrate using the standard score criterion , : Z = (0 − μ)/σ, where μ is the mean value in a dataset and σ is the standard deviation. Here, the Z score shows how far the mean value of an experimental random parameter is from zero on a scale of the standard deviation. In other words, the larger |(0 − μ)|/σ is, the more confidently we can say that the estimated parameter is different from zero. The results of the Z score calculations and the confidence probability P of the blood absence in the sample are shown in Table for each of the distributions [12pt]{minimal}
$$f(C),$$ f C , which are presented in Fig. . These distributions were calculated using the MCRAD and RSC methods for all combinations of every Raman spectrum of a bloodstain on an Al foil with every Raman spectrum of a bloodstain on a corresponding substrate. After that, the mean value and standard deviation were calculated. If we choose the confidence level of 95%, it will correspond to the interval from − 1.96 to 1.96 in Table . The confidence probabilities of blood absence in a sample calculated according to Z scores are presented in Table . Therefore, the MCRAD method with a confidence probability of not less than 95% demonstrates the absence of blood for the bloodstains on blue polyester and denim. The same predictions are fulfilled for white polyester with a confidence probability of 59%. The MCRAD predicts blood presence on a cotton fabric with a confidence probability of 91%. The RSC method demonstrates the presence of blood for the same samples with a confidence probability close to 100%.
It is of great importance for a new forensic method to determine the potential for false positives. The results of an attempt to detect blood on pure substrates using MCRAD and RSC are shown in Fig. . The blood volume fractions were estimated as follows. We used the MCRAD and RSC methods for all combinations of every Raman spectrum of a blood sample on an Al foil with the Raman spectrum of a sample of corresponding pure substrate. In total, RSC demonstrates an appropriate level of such error for more substrates compared to MCRAD. The issue is a denim substrate. Therefore, taking into account the results presented in Figs. and , RSC appears to be a more universal method in a case when we do not have a priori information about whether there is a biological sample on a substrate and which one. In our opinion, the bias in extracting a blood volume fraction from a pure substrate (see, for example, Fig. b) is associated with a similarity between blood and the substrate Raman spectra. To test this hypothesis, we used the Soergel distance to quantitatively estimate the similarity of two spectral curves: 4 [12pt]{minimal}
$$S=^{N}|{x}_{i}-{z}_{i}|}{_{i=1}^{N}({x}_{i},{z}_{i})},$$ S = ∑ i = 1 N | x i - z i | ∑ i = 1 N max ( x i , z i ) , where [12pt]{minimal}
$${x}_{i}$$ x i and [12pt]{minimal}
$${z}_{i}$$ z i are these curve abscissa values (Raman signal) for the same ordinate (Raman frequency shift). [12pt]{minimal}
$$N$$ N is the number of data points in the curves. Let us denote as [12pt]{minimal}
$${S}_{l}$$ S l the value of [12pt]{minimal}
$$S$$ S calculated according to (4) but for two spectra, which are preliminary averaged over a sliding spectral window including [12pt]{minimal}
$$l$$ l points (l< [12pt]{minimal}
$$N).$$ N ) . In other words, this means that Raman spectra are averaged in the sequential intervals including [12pt]{minimal}
$$l$$ l spectral points. Here, we used spectral preliminary normalization based on the area under a spectrum curve. In this case, [12pt]{minimal}
$$}{S}_{l}=0$$ lim l → N S l = 0 . Note that [12pt]{minimal}
$${S}_{l}=0$$ S l = 0 for identical curves for any [12pt]{minimal}
$$l$$ l . Therefore, [12pt]{minimal}
$${S}_{l}(l)$$ S l ( l ) may provide information about the similarity of the two spectra. The results of the [12pt]{minimal}
$${S}_{l}$$ S l calculation for a blood spectrum on an Al foil and a common substrate are presented in Fig. . Calculations were conducted for mean Raman spectra of blood on an Al foil and spatially averaged Raman spectra of a specific substrate. To calculate [12pt]{minimal}
$${S}_{l}(l),$$ S l l , we initially averaged Raman spectra for spectral subintervals with length [12pt]{minimal}
$$l$$ l . Then, we calculated [12pt]{minimal}
$${S}_{l}(l)$$ S l l for all combinations of every Raman spectrum of a blood sample on an Al foil with every Raman spectrum of a blood sample on a corresponding substrate. After that, the mean value and standard deviation were calculated. This procedure was repeated for [12pt]{minimal}
$$l$$ l varied in the interval [1, [12pt]{minimal}
$$N$$ N ]. We see that the dependence [12pt]{minimal}
$${S}_{l}$$ S l on [12pt]{minimal}
$$l$$ l for denim substrate has smaller values compared to other substrates, especially for [12pt]{minimal}
$$l>400$$ l > 400 . This substrate gives the largest errors in the estimation of a blood volume fraction on the pure substrates using the RSC (see Fig. b). In more detail, the Soergel distances calculated for individual sliding spectral windows for various Raman frequency shifts are presented in Fig. . These calculations were conducted in the same manner as the results presented in Fig. . We see that the difference between blood and denim Raman spectra is minimal compared to other substrates. This can be a reason for the largest error in the results presented in Fig. . Therefore, metrics such as the Soergel distance can estimate the spectral peculiarities of comparing spectra. Nevertheless, more work needs to be done to understand this interesting observation, although this is beyond the scope of this study.
In general, the accuracy of any analytical method of a mixture decomposition using Raman spectroscopy data is defined by; (i) a similarity of Raman spectra of pure components existing in a studied composition; (ii) the ration of the pure components volume fractions. The detection of a target component is complicated essentially in the case of its strong similarity and small volume fraction relatively other components in the studied composition. The RSC method is based on subtracting a target Raman spectrum with an unknown weight coefficient C from an experimental spectrum of a complex sample, achieving a minimum of the objective function (3). The possible reason for the greater stability and robustness of the RSC is as follows. RSC is based on the application of the L1 norm to a function [12pt]{minimal}
$$|_{org}-C {S}_{ref})}{dk}|$$ d ( S org - C · S ref ) dk through an estimation of an integral in Eq. ( ). The L1 norm is associated with function integration over an independent variable variation interval. Let us explicitly include random noise in the description. Both [12pt]{minimal}
$${S}_{org}$$ S org and [12pt]{minimal}
$${S}_{ref}$$ S ref can include an additive random nose ( [12pt]{minimal}
$${R}_{1}(k), {R}_{2}(k))$$ R 1 k , R 2 ( k ) ) : 5 [12pt]{minimal}
$${S}_{org}(k)={S}_{org}^{0}(k)+{R}_{1}(k),$$ S org k = S org 0 k + R 1 k , 6 [12pt]{minimal}
$${S}_{ref}(k)={S}_{ref}^{0}(k)+{R}_{2}(k),$$ S ref k = S ref 0 k + R 2 k , where [12pt]{minimal}
$${S}_{org}^{0}, {S}_{ref}^{0}$$ S org 0 , S ref 0 are the corresponding features without noise. Then, Eq. ( ) takes the form 7 [12pt]{minimal}
$$ f()= |_{blank}+{S}_{ref}^{0}(C- )+{R}_{1}-_{2})}{dk}|dk.$$ δ f C ~ = ∫ d S blank + S ref 0 C - C ~ + R 1 - C ~ R 2 dk d k . Let [12pt]{minimal}
$${R}_{1}(k), {R}_{2}(k)$$ R 1 k , R 2 ( k ) be stationary random functions with zero mean values: 8 [12pt]{minimal}
$$R_{1} ( k ) = {}( {} ), R_{2} ( k ) = {}( {} ),$$ R 1 k = ν · rand k , R 2 k = ν · rand k , where [12pt]{minimal}
$$$$ ν is the amplitude of the noise component presented in a relative fraction of a mean value of the set of Raman spectra of blood on an AI foil, and [12pt]{minimal}
$$()$$ Rand ( k ) is a set of random values varied in the interval [-0.5, 0.5]. If the function [12pt]{minimal}
$$|_{blank}+{S}_{ref}^{0}(C- )+{R}_{1}-_{2})}{dk}|$$ d S blank + S ref 0 C - C ~ + R 1 - C ~ R 2 dk is an ergodic random process, then integrating this function over evolution variable [12pt]{minimal}
$$k$$ k is equivalent to averaging over an ensemble of realizations. The latter causes noise reduction and influences the target component concentration (volume fraction) restoration results. To obtain arguments about this, we conducted numerical experiments with Eqs. ( ) and ( ), limited by a noise level up to 5% of the mean value of the Raman spectra set used, which exceeds the typical values of the noise component with a margin. We synthesized a set of 100 realizations of random functions [12pt]{minimal}
$${R}_{1}(k),$$ R 1 k , [12pt]{minimal}
$${R}_{2}(k)$$ R 2 ( k ) according to Eq. ( ) with [12pt]{minimal}
$$$$ ν varied in the interval [0.0, 0.05] and restored volume fraction [12pt]{minimal}
$$$$ C ^ using criterion (6). The calculation of the latter was conducted as follows. We took every Raman spectrum of a blood sample on an Al foil as a reference and used it to restore the volume fraction in the remaining Raman spectra of a blood sample on an Al foil. This procedure was repeated for all other Raman spectra of a blood sample on an Al foil. Then, the mean value and standard deviation were calculated. The results are presented in Fig. . One can see that the presence of such noise levels causes the target component (blood) volume fraction restoration relative error [12pt]{minimal}
$$ C$$ δ C up to 1%. Here, [12pt]{minimal}
$$ C=|C-|/C.$$ δ C = C - C ~ / C . Therefore, RSC is quite robust to random fluctuations of spectral data due to random experimental errors and intergroup variability. A possible reason for the weak stability and robustness of MCRAD is as follows. MCRAD is based on Eq. ( ) solution, with the matrix of concentrations [12pt]{minimal}
$${}$$ C containing a set of [12pt]{minimal}
$${}_{j}+$$ C ^ j + C ~ values, where [12pt]{minimal}
$$$$ C ~ is the unknown concentration (volume fraction) of a blood sample and [12pt]{minimal}
$${}_{j}$$ C ^ j are known additional volume fractions (VFs) according to the principle of standard addition. The evaluation of [12pt]{minimal}
$$$$ C ~ is conducted through an iterative solution of the set of equations , : 9 [12pt]{minimal}
$${}_{j}={S}_{org}+{(}_{j}+){S}_{ref}.$$ S ^ j = S org + ( C ^ j + C ~ ) S ref . The iterative procedure is based on the application of the L2 norm (the Euclidian norm) to a function similar to [12pt]{minimal}
$$({}-{}{{}}^{t})$$ ( W - C S t ) , where [12pt]{minimal}
$${}$$ S is the matrix of all component spectra in a composition, [12pt]{minimal}
$${}$$ C is the matrix of concentrations, and [12pt]{minimal}
$${}$$ W is the matrix of experimental spectra. Even in the case of a spectrum with additional random noise being an ergodic random process, the L2 norm cannot be averaged over an ensemble. We conducted the simulation using MCRAD with the same noise model (5), (7), and the same noise level as for the RSC. A Raman spectrum of a bloodstain on Al foil was used as a reference to restore the volume fraction in the rest of the Raman spectra of a blood sample on Al foil. This procedure was repeated for all other Raman spectra of blood on Al foil. After that, the mean value and standard deviation were calculated. The results of the concentration (volume fraction) [12pt]{minimal}
$$C$$ C restoration in 100 simulations are shown in Fig. . We see that the influence of noise on the concentration restoration accuracy is much stronger than that of the RSC. A possible reason is that the Euclidian norm does not allow the use of the benefits of ergodic random processes from the point of view of noise reduction. This can be a reason for the responsiveness of the MCRAD algorithm to random noise.
Potential errors (false positives and false negatives) of a new method could significantly reduce the interest of forensic practitioners. False negatives due to the low detection limit could result in missing valuable evidence. False positives could result in a significant waste of time and resources. To further reduce potential false positives for the method developed here, a second stage of the data analysis could be conducted as a part of a hierarchical approach. It is noteworthy here that running an additional analysis will not noticeably increase the total test time because of the fast spectral measurements and high speed/efficiency of modern computers. The second data analysis, which we propose, is the comparison of the obtained Raman spectra with the spectra of the corresponding pure substrate. The latter could already be in the spectral library of the software. If not, the mapping of the pure substrate could be conducted quickly at the crime scene or in the lab if the evidence sample on a piece of material is already collected and delivered to the lab. Of course, Raman spectrum of an analyzed real biofluid sample is not exactly the same compared to an etalon Raman spectrum and it is a source of bias. However, the variations in Raman spectra of all main body fluids did not prevent us from 100% accuracy in their identification when a high-quality Raman spectrum was measured for a “new” sample, which was not used for the training dataset , . In addition, blood is by far the most consistent body fluid (relative to other main body fluids including semen) from the viewpoint of biochemical composition. Therefore, we hypothesized that we can use a single reference Raman spectrum of dry blood in this study in contrast to a set of individual spectral components as we have done for semen traces in our earlier work . In this study, we created a reference Raman spectrum using several bloodstains on an aluminum substrate and then used this reference spectrum for the detection and identification of blood traces on interfering substrates. It is very important to emphasize here that the integrated bloodstains on interfering substrates were prepared from blood samples, which were not used for developing the reference Raman spectrum of dry blood (different donors). In any case, if there are doubts about the adequacy of the available reference spectrum of biological fluid to the sample under study, the RSC can be used in the reverse manner. We can measure the Raman spectrum of substrate in a spatial point without stain. After that, we can extract this component from the Raman spectrum measured in a spatial point with presence of a "biofluid stain & substrate" combination. The residue is a Raman spectrum of a specific biofluid stain sample. The latter can be identified by any suitable manner, for example, by comparing it with a library of biofluids’ Raman spectra. The decision about what biodluid is presented can be based, for example, on a fuzzy logic approach by comparing “distances’ of the residue with “standard’ Raman spectra of various biofluids from the library. The implementation of this approach is shown in Fig. . Here, the residuals [12pt]{minimal}
$${S}_{R}$$ S R between the experimental Raman spectra of bloodstain on a definite substrate and the Raman spectra of the same pure substrate are compared with the set [12pt]{minimal}
$${S}_{0}$$ S 0 of 100 Raman spectra of blood on Al foil and 100 Raman spectra of seminal fluid on an Al foil measured by us earlier . The proximity factor was calculated using formula 10 [12pt]{minimal}
$$=_{i}_{R,i}-{S}_{0,i}|}{|{S}_{R,i}+{S}_{0,i}|},$$ r = 1 2 ∑ i S R , i - S 0 , i S R , i + S 0 , i , where summation is conducted over all spectral points in the compared Raman spectra. For all cases, we can conclude that residual Raman spectrum corresponds to blood. Therefore, we can conclude that there are principal solutions of the issue about, strictly speaking, absence of absolute etalon Raman spectrum of a biofluid, which perfectly corresponds to a concrete experimental sample of biofluid analyzed “here and now”. The fundamental background of this positive for practical usefulness conclusion is as follows. Let the black line in Fig. correspond to the spatial positions of biofluid stain presence. Then, in a spatial area marked by a blue star, we can measure the Raman spectrum of the pure substrate because biofluid stain is absent. Therefore, this situation fully matches the Manne condition (see Fig. ) when evolutionary variable [12pt]{minimal}
$$t$$ t describes a spatial position on a substrate surface. A more deep analysis is not in the scope of current study and will be presented in the future papers. Another important issue about RSC robustness to false positive results. Once again, to validate this, we can use the Raman spectrum of substrate in a spatial point without stain. The following is an example of this approach implementation for bloodstain detection on various substrates. The goal is to confirm that the detected bloodstain is not a false positive. We suggest using additional experimental data from neighboring points on a substrate surface, which do not contain blood traces (pure substrate). We used the simplest unsupervised classification method, principal component analysis (PCA), to test whether Raman spectra from apparent bloodstains and a pure substrate could be differentiated. Figure shows a PCA score plot obtained for Raman spectra collected from a bloodstain on a common substrate and those collected from the same pure substrate. These two classes of Raman spectra could be differentiated with high confidence in the case of blue and white polyester and cotton. However, there is some overlap on the score plot for Raman spectra collected from a bloodstain on denim substrate and those collected from pure denim. We believe that this is because denim has a strong Raman signal and overwhelms the signal from blood. As evident in Fig. , Raman spectra of a bloodstain on denim substrate are very similar to the spectra of pure denim with no noticeable contribution from blood. Nevertheless, despite some overlap, there is a significant number of points on the score plot, which are well separated. Therefore, we believe that the proposed approach should allow for testing for false positives even in the case of denim if Raman spectra are collected from multiple points on the bloodstain and compared with those collected from a pure denim. Obviously, more work needs to be done to optimize this process, including using more robust statistical methods of supervised statistics. We plan to work on this in the near future.
Samples Blood samples were purchased from BioIVT, LLC (Westbury, NY), from five anonymous donors. Donors were negative for HbsAg, HCV, HIV-1&2, syphilis, and HIV-1 antigen. All samples were deposited onto one of the following substrates: aluminum tape, white cotton fabric, white polyester fabric, blue polyester fabric, and denim fabric, pipetting 10 μL on the surface and letting it dry overnight.
Blood samples were purchased from BioIVT, LLC (Westbury, NY), from five anonymous donors. Donors were negative for HbsAg, HCV, HIV-1&2, syphilis, and HIV-1 antigen. All samples were deposited onto one of the following substrates: aluminum tape, white cotton fabric, white polyester fabric, blue polyester fabric, and denim fabric, pipetting 10 μL on the surface and letting it dry overnight.
A Renishaw InVia confocal Raman spectrograph equipped with a research-grade Leica microscope, a long-range 50 × objective, and a Renishaw PRIOR stage for automatic mapping were used to collect the Raman spectra over a range of 400 –1800 cm −1 . A 785-nm laser light was utilized for excitation. The maximal laser power was about 80 mW. It was reduced from to ten percent capacity with a spectrum accumulation time of 10 s to avoid photodegradation. The spot size of the excitation beam on the sample was approximately 2 µm using standard confocal mode and a 50-µm slit. Multiple spectra were collected from different spots of each bloodstain using automatic mapping, and each spectrum was an average of ten accumulations. Peak accuracy was assured by verifying instrument calibration before each analysis using a silicon standard. All spectrum measurements were first treated using WiRE 3.4 software to remove any cosmic ray interference. The processing of the received data was performed with MATLAB software. Outliers were removed using the random forest method . The preprocessing of the experimental Raman spectra was conducted in 3 steps: background subtraction (a standard procedure for minimizing the fluorescence contribution), a random noise filtration, and normalization by the area under the curve. The background subtraction was implemented by shape-preserving piecewise cubic interpolation of a Raman spectrum at neighboring grid points in a gliding spectral window with a width of 200 spectral points (182.2 cm −1 ), the quantile value is set to 10%. The noise reduction was implemented using Savitsky-Goley filter with the following parameters’ value: the order of the polynomial was equal to 1, the gliding spectral window width was equal to 45 spectral points (41 cm −1 ). The finding this filter optimal parameters was estimated by the following way. The random nature of noise allows us to consider the average Raman spectrum [12pt]{minimal}
$${ }_{samp}$$ S ¯ samp of an experimental sample set of [12pt]{minimal}
$${S}_{samp,i}$$ S s a m p , i spectra: [12pt]{minimal}
$${ }_{samp}=_{i=1}^{N}{S}_{samp,i}$$ S ¯ samp = 1 N ∑ i = 1 N S s a m p , i as an approximation to the actual spectrum without noise. Here, [12pt]{minimal}
$$N$$ N is the volume of the experimental set. Let’s denote the Raman spectrum [12pt]{minimal}
$${S}_{samp,i}(k)$$ S s a m p , i ( k ) processed by Savitsky-Goley filter as [12pt]{minimal}
$${S}_{SG,i}(k),$$ S S G , i k , where [12pt]{minimal}
$$k=$$ k = 1 , K - is Raman shift. Optimal filter parameters are corresponded to minimum of the following functional r (see Eq. ( )): [12pt]{minimal}
$$=_{j,k}_{SG,j}(k)-{ }_{samp}(k)|}{|{S}_{SG,j}(k)+{ }_{samp}(k)|}$$ r = 1 2 N K ∑ j , k S S G , j ( k ) - S ¯ samp ( k ) S S G , j ( k ) + S ¯ samp ( k ) The dependence of [12pt]{minimal}
$$$$ r on the gliding spectral window width is presented in Fig. . Here, we used the first-order polynomial in Savitsky-Goley filter. In common, the choice of the gliding window width about 40 cm −1 is quite reasonable. The using polynomial of more high orders reduces the quality of filtration (see Fig. ) because less [12pt]{minimal}
$$$$ r value corresponds more close shape of a processed by Savitsky-Goley filter Raman spectrum to an average Raman spectrum of the respective experimental sample set. The typical Raman spectra signal-to-noise value near their maxima was about 62 dB, the mean value of this parameter was about 7 dB that is caused by presence of many small peaks.
The results of a comparative study of two methods for the detection of body fluid traces on common substrates are presented. The first method, referred to as MCRAD, is a standard in Raman spectroscopy that combines multivariate curve resolution with the addition method. Another method referred to as RSC is new in Raman spectroscopy applications, which is based on reducing spectrum complexity when we remove a spectral component from a mixture entirely. Both methods implement the "one-per-step" approach for complex sample Raman spectrum decomposition. An example of bloodstain detection on blue polyester, denim, cotton fabric, and white polyester substrates was considered. Both RSC and MCRAD were shown to allow for restoring a target component (body fluid) volume fraction from an experimental spectrum of this body fluid dried on an interfering substrate with only a priori knowledge of this body fluid etalon spectrum. RSC shows more reliable performance than MCRAD for the detection and identification of a bloodstain on interfering substrates. In complicated cases, the probability of obtaining a wrong result is essentially higher for MCRAD than for RSC. This conclusion is confirmed by the results of the decomposition of Raman spectra of bloodstains on various substrates (see Fig. and Table ). In our opinion, the better robustness of RSC compared to MCRAD is due to the implementation of ergodic theory in the target minimization function in the RSC method. When a substrate has a rather unique Raman spectrum compared to a body fluid, both RSC and MCRAD are quite efficient in avoiding false positive results. In complicated cases of a substrate in which the Raman spectrum has a low specificity (in other words, high similarity) compared to the tested body fluid, the level of these errors can be up to 0.2 in volume fraction instead of zero value. Such situations were met more often for MCRAD compared to RSC (see Fig. ). To test this hypothesis, we used the Soergel distance to quantitatively estimate the similarity of two spectra, which are preliminarily averaged over a sliding spectral window. The calculated dependencies of the Soergel distance between two Raman spectra on the size of the spectral window have evident peculiarities for the denim case, which was the largest bias in the decomposition results. For practical application of the developed method, we proposed a simple additional test (hierarchical approach) for a potential false positive using Raman spectra of a pure substrate and statistically compared them with Raman spectra obtained for the apparent bloodstain. Therefore, this work offers a novel approach in Raman spectroscopy named RSC for solving one of the most challenging problems for the identification of bloodstains for forensic purposes using Raman spectroscopy, which is the interference of common substrates. Some comments for future studies are as follows. The origin of established peculiarities in dependencies of the Soergel distance between two Raman spectra on the size of the spectral window should be studied in detail. We do not report on the sensitivity of the RSC method. The detection limit of the method needs to be investigated, compared with the current methods used by the law enforcement agencies and the needs of the practical application (for example, for DNA profiling).
|
A molecular reappraisal of matrix-producing breast metaplastic carcinoma highlighted by PLAG1 and MYC rearrangements | c2c51b79-03c1-4b3f-8c31-41b6c6fedc30 | 10070549 | Anatomy[mh] | Metaplastic carcinoma of the breast is a rare “special type” of tumor including different subtypes, all of which are extremely aggressive and with a poor prognosis. The most common is metaplastic carcinoma with squamous differentiation, with components showing mesenchymal differentiation (chondroid, muscular or fibrous). With lower frequency the mesenchymal overgrowth can simulate a true sarcomatous appearance. One type of metaplastic carcinoma is composed of a chondromyxoid background resembling the pleomorphic adenoma of the parotid gland (PA) and/or carcinoma ex pleomorphic adenoma (Ca-ex-PA) or a myxoid lipoblastoma in children. The authors discuss the role of PLAG1 and MYC involvement in development and aggressiveness of this tumor.
Between 2015 and 2017, we identified and followed-up 12 cases of matrix producing carcinoma (MPC). Fluorescence in-situ Hybridization (FISH) was performed using a dual-color break-apart (BA) PLAG1 probe from Empire Genomics (Buffalo, NY, USA). The PLAG1 probe is a dual-color break-apart probe consisting of a 265-kb 3′ centromeric side labeled in green and a 289-kb 5 ′ telomeric side labeled in orange. Formalin-fixed paraffin-embedded sections (FFPE) 4 μm thick, were air-dried and oven baked for 30 minutes at 60°C. FISH was performed in accordance with the manufacturer’s technical instructions; 10 μl of PLAG1 mix probe was added, and the slides were cover slipped and sealed. Slides were then denatured for three minutes at 83°C and hybridized overnight at 37°C on a Thermo-Brite (DAKO Hybridizer). Following hybridization, coverslips were removed by placing the slides in a room temperature bath of 2XSSC, moved and heated to 0.4xSSC 0,3% IGEPAL 72°C for two minutes and then rinsed in 2xSSC 0,1% IGEPAL at room temperature. A two-fusion signal pattern indicates no rearrangement involving the PLAG1 gene, whereas a distant separation between orange and green signals indicates PLAG1 involvement. A positive FISH score, for PLAG1 , was reported The cases with high Gene Copy Gain (GCG) for PLAG1 , were also analyzed using the LSI MYC Dual Color Break-Apart Rearrangement Probe (Abbott Molecular Inc., Des Plaines, IL, USA). This kit contains a mixture of LSI-MYC Spectrum-Orange probe and LSI-MYC Spectrum-Green probe. Briefly, 4-μm–thick, FFPE sections were air-dried and oven baked for 30 minutes at 60°C and then processed following the manufacturer’s recommendations on the Abbott Vysis. After the slides were processed, 10 μl of MYC break-apart probe (Abbott Molecular, Inc.) were added, and the slides were cover slipped and sealed. Slides were then denatured for 10 minutes at 80°C and hybridized overnight at 37°C on a Thermo-Brite. Following hybridization, coverslips were removed by placing the slides in a room temperature bath of 2 min 2X SSC and then moved to a second bath heated to 73°C for three minutes. They were then rinsed in Wash Buffer. DAPI (4’,6-Diamidino-2-Phenylindole, Dihydrochloride) was then applied to each slide, and a coverslip was applied. For amplification of MYC the threshold limit of MYC/CEP8 ratio was ⩾2 was used following the international guideline. Orange staining was used to evaluate the MYC locus at 8q24 while green staining was used to evaluate the centromeric 8 chromosome (8-CEP). , All cases underwent surgical procedures as mastectomy or quadrantectomy following our Institutional Guidelines and an adjuvant or neoadjuvant CT was performed in almost all cases.
This type of carcinoma showed a frequency up to 5% in our series and rarely expressed hormonal receptors (estrogen, progesterone and Her2/Neu). , The patients ranged from 40 to 78 years old with a mean of 64.7 years, and the tumor size ranged from 1.1 to 5 cm (mean 2 cm). Only one case showed an ER and PgR receptor expression (both in 90% of cells) with negative Her-2/Neu expression by IHC ( ). From a morphological point of view, the neoplasm was composed of epithelial cells with anaplastic features, aggregated in poorly cohesive nests or strands intermingled with a chondromyxoid matrix, and we also detected foci of coagulative necrosis ( ). All tumors had a high mitotic rate and a high proliferative index evaluated with Ki-67 labeling index (average 54.7%) ( ). We observed the classic rearrangement as BA (Break-Apart) of PLAG1 in eight cases (BA From 2% to 29%), four cases did not show a BA ( ). We observed an increase of GCG for PLAG1 from 1.7 up to 6.1 signal for cells. At times there was an increase of red 5’-signal (average gain 3.7 signal for cell) and a high level of green 3’-signal (average gain 22.4 signal for cells) ( , and ). Cases with a PLAG1 rearrangement showed a GCG (34.7%) and an increased green signal (68%). This unexpected observation suggested more involvement of 8-CHR. To ascertain whether or not this represented a true polysomy, we evaluated MYC status. Since MYC is present in 8-CHR, MYC evaluation is more accurate than PLAG1 for ascertaining polysomy. A true polysomy for MYC (threshold value >3) was found in a high percentage of cases (33.3%). An amplification of MYC , was found in three cases with MYC/CEP-8 ratio ⩾ 2. only one case showed a borderline value of 1.9 (case no.11) ( ). In two cases, a neoadjuvant therapy (NeoA-Therapy) was used. In one case with EC (case no.3), and the other with a taxane-based drug following an IEO trial Protocol (data not shown) (case no.2). Due to disease progression, the second case was excluded from the protocol and an Adjuvant therapy (Adj-Therapy) was carried out with EC. At the last follow-up, only two cases died with disease (DWD; case no. 2 and 7) ( ) and the others were NED (not evidence of disease) ( ; ). The cases with MYC amplification did not show PLAG1 -BA, while all the cases polysomic for MYC showed PLAG1-BA. Among Wild Type (WT) cases of MYC (four cases), one was negative for PLAG1 -BA rearrangement while others were rearranged with PLAG1 -BA ( ).
Pleomorphic Adenoma Gene-1 PLAG1 , located in CHR 8q12.1, encodes a nuclear zinc finger protein that is rearranged in many tumors while MYC is a transcriptional protein factor (bHLH transcription factor) linking many other genes and acquiring a pro-oncogene activity. This gene observed in CHR 8q24.21 plays a pivotal role in regulation, progression, cellular transformation and apoptosis in great number of tumors, hematological and solid ones. PLAG1 were observed rearranged in many cases of PA and/or Ca-ex-PA (frequency of 80% of cases), in myoepithelial tumor of salivary gland, in myxoid-lipoblastoma in children and in some mesenchymal tumors of the uterus. MPC is a rare breast tumor that shares a similar morphology with those tumors. Following these morphological oddities, our hypothesized that MPC could share the same PLAG1 alteration. , To the best of our knowledge, no association has been published in the biomedical literature between PLAG1 and MYC rearrangements and MPC. We observed a rearrangement of PLAG1 , involving 8-CHR, not only as a BA signal, but also as Gene Copy Gain (GCG). These alterations were observed and reported in salivary tumors and myxoid lipoblastoma. , MYC involves the same chromosome of PLAG1 and we tried to understand that PLAG1 -GCG was a true polysomy or not. We observed that the cases with PLAG1 -BA showed a MYC polysomy, while the cases without PLAG1 -BA showed MYC amplified, moreover, cases with high value PLAG1 -GCG showed a MYC -polysomy, while the cases with low value of PLAG1 -GCG showed an amplification of MYC ( ). These observations suggest that PLAG1 -GCG, in MPC, is due to a polysomy. PLAG1 -GCG was observed in salivary gland tumor together with PLAG1 deletion. The PLAG1 -GCG without a BA signal suggest a chromosomic rearrangement of CHR 8q12 followed by PLAG1 upregulation in a same manner used in lymphomas and due to an oncogenic mutation of promoted region of MYC . This is a cause of polysomy of 8-CHR in non rearranged PLAG1 cases. - The association between PLAG1 -GCG and MYC amplified-polysomic was observed in salivary glands. In cases reported in literature the amplification of MYC give an aggressive behavior in a subgroup of salivary tumors, this can also happen in MPC. , , MYC were amplified even in a subgroup of inflammatory breast carcinomas with poor prognosis that showed a high-expression of hormonal receptor (Estrogen and Progesteron) and absence of Her-2/Neu. In our cases it was difficult to identify this association because so few cases were studied. Among the two cases DWD one showed a MYC amplification without PLAG1 -BA and the others one was MYC -WT with PLAG1 -BA in 6% of neoplastic cells (cases n.2 and n.7; ). We suppose that MYC amplification gives an aggressive behavior together with PLAG1 -BA in the cases of MPC discussed here. - Rearrangement of PLAG1 in salivary gland tumors does not change the prognosis, in fact, PLAG1 was observed in benign and malignant salivary tumor with the same frequency. Only MYC changes the prognosis. Among MYC -WT cases we observed a PLAG1 -BA rearrangement, suggesting a difference between salivary tumors and MPC with involvement of PLAG1 -BA in the prognosis and aggressiveness of neoplasms. , Despite morphological similarities between MPC and other salivary and soft tissue myxoid tumors, its behavior is different. , Because of the few cases analyzed we cannot show a statistical correlation between prognosis and the PLAG1 and MYC alterations. In literature this is the first report that shows an involvement of PLAG1 and MYC in prognosis and behavior of breast MPC. So, this is should be a bases for a future studies.
|
SOCS3 as a potential driver of lung metastasis in colon cancer patients | e1cf26d1-a7dc-4940-b8a9-ec9bcbbbf5e0 | 10070831 | Anatomy[mh] | The JAK-STAT pathway is the center of signal transduction pathways to activate multiple cytokines, and its aberrant alterations involve the occurrence and development process of multiple malignant tumors and autoimmune diseases . Suppressors of cytokine signaling (SOCS) proteins are the negative physiologic regulator and compete with STATs by binding to the cytokine receptor to interfere with the signal transduction process . The SOCS family contains 8 intracellular protein members, including CIS and SOCS1-7 . These proteins all are centered on an SH2 domain, surrounded by a changeable length and sequence amino-terminal domain and a carboxyl-terminal module with 40-amino-acid, called the SOCS box . Among these regions, the SH2 domain is the major functional region and harbors different effects in different SOCS proteins. SOCS box is made of three α-helices and combined with an E3 ubiquitin ligase complex. This structure, together with the E1 ubiquitin-activating enzyme and the E2 ubiquitin-conjugating enzyme, mediates ubiquitination and proteasome degradation of the SOCS binding conjugate . Besides, SOCS1 and SOCS3 also possess a special kinase inhibition region (KIR) at the n-terminal, which directly inhibits JAK tyrosine kinase activity . The SH2 region in SOCS1 is directly combined with the activation loop of JAK to inhibit the JAK tyrosine kinase activity and the tyrosine phosphorylation of STAT1α . After recruiting the SH2 domain to the receptor cytoplasmic domain, SOCS3 exerts inhibition of JAK kinase activity and degradation of the proteasome through the interaction of the SH2 domain and KIR . Currently, SOCS3 has turned out to involve in the process of embryonic development, inflammatory response, immunoregulation, and tumor progression. SOCS3 selectively regulated the signal transduction process of multiple cytokines, including leukemia inhibitory factor (LIF), IL-6, IL-11, G-CSF, and leptin . Besides, the loss of SOCS3 in macrophages promoted the activation of STAT and the conversion to the M1 phenotype, improving the phagocytic capacity of macrophages and promoting the differentiation of Th1 and Th17 cells to regulate immune effects . Meanwhile, SOCS3 played different roles in different tumor types, being pro-oncogenic or tumor suppressive. In breast cancer, SOCS3 inhibited tumor metastasis and tumor cell proliferation regulated by multiple non-coding RNA . In lung cancer, SOCS3 inhibited tumor cell proliferation and angiogenesis in small-cell lung cancer cells via inhibiting HIF-1α and promoted apoptosis and cell proliferation in non-small-cell lung cancer through its methylation . Mouse model research illustrated that SOCS3 mediated the proliferation and hyperplasia of the crypt and the transformation of inflammation into cancer in the colon . Altered SOCS3 combined with chemotherapy or targeted therapy has been demonstrated to sensitize tumors and inhibit tumor progression . Besides, SOCS3 mimetic have been shown to reduce tumor growth and inhibit lung metastasis in triple negative breast cancer (TNBC) by suppressing the production of inflammatory cytokines . Therefore, based on the performance of SOCS3 in regulating cytokine signaling and immune response, it is of clinical application value to investigate the probable mechanism of SOCS3 in immune-related diseases and tumors and to develop mimetic compounds. However, current research on SOCS3 is mostly focused on immune regulation. The mechanism of SOCS3 regulating tumorigenesis and its involved signaling pathway is relatively complicated. Therefore, it is necessary to summarize the mechanism of SOCS3 in different tumors and conduct further mechanistic exploration. In this study, we explored the expression pattern and prognostic performance of SOCS3 in pan-cancer and its relationship with immune cell infiltration. Besides, we performed immunohistochemistry (IHC) staining in colon cancer patients with lung metastasis to explore the relationship between SOCS3 status and macrophage infiltration and tumor metastasis.
Pan-cancer expression and immune infiltration analysis The SOCS3 expression and corresponding clinical follow-up information from different tumors were collected from The Cancer Genome Atlas (TCGA, https://portal.gdc.cancer.gov/ ). Normal tissue data from the GTEx database ( http://gtexportal.org ) was the normal control. The SOCS3 expression difference between tumor tissues and normal tissues was displayed using a boxplot (R package, ggplot2, v 3.4.0). Clinical follow-up information consisted of overall survival (OS). The relationship between SOCS3 and the prognosis was examined using the Cox proportional hazards regression model (R package, survival, coxph, v 3.2-7) and Kaplan–Meier analysis (R package, survival, survfit, v 3.2-7). Median levels of SOCS3 expression divided patients into low and high expression groups in each tumor and performed differentially expressed genes (DEGs) analysis (R package, TCGAbiolinks, v 2.24.3). Subsequently, we performed gene set enrichment analysis (GSEA) of DEGs for KEGG and Hallmark pathway in the pan-cancer, of which the top ten were shown (R package, clusterProfiler, GSEA, v 4.4.4). Circular barplot displayed the result of enrichment analysis (R package, ggplot2, v 3.4.0). Cancer types and corresponding abbreviations were available in the official documentation ( https://www.cancer.gov/types ). TIMER (Tumor Immune Estimation Resource) database ( https://cistrome.shinyapps.io/timer/ ) was used to evaluate the abundance of tumor infiltrating immune cells (TIICs) in different cancer types, including B cells, CD4+ T cells, CD8+ T cells, neutrophils, macrophages, and dendritic cells . The ratio of the immune-stromal component in the tumor microenvironment (TME), the tumor mutation burden (TMB), and Microsatellite instability (MSI) were calculated using the Sangerbox online platform ( http://www.sangerbox.com/tool ) . ImmuneScore and StromalScore are the scores of each sample’s immune and matrix components in the TME, respectively, while ESTIMATEScore is the sum of the two, representing the combined ratio of the two components in the TME, which means that the higher score matched the higher the corresponding component in the TME. These scores were also calculated using the Sangerbox online platform. The correlation analysis between EXO5 and immune-gene signatures was performed by the TIGER (Tumor Immunotherapy Gene Expression Resource) database ( http://tiger.canceromics.org/#/ ) . Patient population and GEO data population We collected colon cancer patients who had lung metastasis at the First Affiliated Hospital of Zhejiang University from 2010 to 2020. All patients were histopathologically diagnosed by two pathologists and the attending physician. Clinical information was gathered, including gender, age at diagnosis and stage. The Tumor-Node-Metastasis (TNM) -based staging of colon cancers was defined by the new 8th editions of the relevant Union for International Cancer Control (UICC) and American Joint Committee on Cancer (AJCC) publications. Patients without clinical information and pathological section were excluded. The gene expression data and corresponding clinical information of GSE68468, GSE41258, GSE22834, and GSE6988 data sets were collected from GEO data sets ( https://www.ncbi.nlm.nih.gov/geo/ ). Immunohistochemistry staining A total of 32 colon cancer patients who had lung metastasis were included in this study. Normal intestinal mucosal samples from normal adjacent tissue were used as control samples. The CD68, CD163, and SOCS3 status of all primary tumors and metastases and normal intestinal mucosal samples were conducted by IHC staining through standard staining procedures (SOCS3 antibody: ab16030, Abcam, Cambridge, UK; CD68 antibody: ZM0464; ZSGB-BIO, Beijing, China; CD163 antibody: ZM-0428; ZSGB-BIO, Beijing, China). Brownish-yellow cytoplasmic staining was determined to be positive on target tissue cells. Each sample was assessed with a minimum of 500 cells from 5 representative high-power fields (200 ×) and the proportion of positively colored cells was computed. Positive cell density was divided using a four-point method: 1 point for the proportion of positively stained cells < 25%, 2 points for those between 25%-50%, 3 points for those between 50%-75%, and 4 points for those > 75%. The color intensity of positively stained cells was divided according to the three-point method: 1 point for weakly positive cells, 2 points for moderately positive cells, and 3 points for strongly positive cells. The IHC staining score was calculated by multiplying the density value and color intensity of positive cells, and was determined by the four-point method: 0-1, negative; 2-4, weak ; 5-8, moderate ; and 9-12, strong . Two pathologists who were blind to the patient’s clinical characteristics underwent independent IHC staining assessment. All patients were divided into two groups based on IHC staining score: low expression (negative and +) and high expression (++ and +++). Enrichment analysis using online web tools DEGs between the low SOCS3 expression group and the high SOCS3 expression group in colon primary tumor and lung metastasis were compared with |log 2 FC| > 1 and P < 0.05, and the shared DEGs were displayed through the heat map (R package, pheatmap, v 1.0.12). STRING database ( https://string-db.org/ ) was used to conduct the relationship between DEGs, and the PPI network was displayed using the Cytoscape 3.7.1 software. DAVID ( https://david.ncifcrf.gov/tools.jsp ) was used to perform the Gene Ontology (GO) and the Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis for DEGs . The Venn diagram and enrichment factor bubble chart were performed using the OmicStudio tools at https://www.omicstudio.cn/tool . Statistical analysis Pearson’s chi-squared test and student’s t -test were used to evaluate the association of CD68, CD163, and SOCS3 status with gender, age at diagnosis, and stage. The student’s t -test was used to evaluate the differences in SOCS3 expression of the GEO data set. A P-value <0.05 was considered statistically significant. Data management and analysis were performed using R version 4.2, GraphPad Prism 8.0 Software (GraphPadSoftware, Inc., SanDiego, CA) and SPSS version 19.0 (SPSS Inc., Chicago, IL, USA). All methods were carried out by relevant guidelines.
The SOCS3 expression and corresponding clinical follow-up information from different tumors were collected from The Cancer Genome Atlas (TCGA, https://portal.gdc.cancer.gov/ ). Normal tissue data from the GTEx database ( http://gtexportal.org ) was the normal control. The SOCS3 expression difference between tumor tissues and normal tissues was displayed using a boxplot (R package, ggplot2, v 3.4.0). Clinical follow-up information consisted of overall survival (OS). The relationship between SOCS3 and the prognosis was examined using the Cox proportional hazards regression model (R package, survival, coxph, v 3.2-7) and Kaplan–Meier analysis (R package, survival, survfit, v 3.2-7). Median levels of SOCS3 expression divided patients into low and high expression groups in each tumor and performed differentially expressed genes (DEGs) analysis (R package, TCGAbiolinks, v 2.24.3). Subsequently, we performed gene set enrichment analysis (GSEA) of DEGs for KEGG and Hallmark pathway in the pan-cancer, of which the top ten were shown (R package, clusterProfiler, GSEA, v 4.4.4). Circular barplot displayed the result of enrichment analysis (R package, ggplot2, v 3.4.0). Cancer types and corresponding abbreviations were available in the official documentation ( https://www.cancer.gov/types ). TIMER (Tumor Immune Estimation Resource) database ( https://cistrome.shinyapps.io/timer/ ) was used to evaluate the abundance of tumor infiltrating immune cells (TIICs) in different cancer types, including B cells, CD4+ T cells, CD8+ T cells, neutrophils, macrophages, and dendritic cells . The ratio of the immune-stromal component in the tumor microenvironment (TME), the tumor mutation burden (TMB), and Microsatellite instability (MSI) were calculated using the Sangerbox online platform ( http://www.sangerbox.com/tool ) . ImmuneScore and StromalScore are the scores of each sample’s immune and matrix components in the TME, respectively, while ESTIMATEScore is the sum of the two, representing the combined ratio of the two components in the TME, which means that the higher score matched the higher the corresponding component in the TME. These scores were also calculated using the Sangerbox online platform. The correlation analysis between EXO5 and immune-gene signatures was performed by the TIGER (Tumor Immunotherapy Gene Expression Resource) database ( http://tiger.canceromics.org/#/ ) .
We collected colon cancer patients who had lung metastasis at the First Affiliated Hospital of Zhejiang University from 2010 to 2020. All patients were histopathologically diagnosed by two pathologists and the attending physician. Clinical information was gathered, including gender, age at diagnosis and stage. The Tumor-Node-Metastasis (TNM) -based staging of colon cancers was defined by the new 8th editions of the relevant Union for International Cancer Control (UICC) and American Joint Committee on Cancer (AJCC) publications. Patients without clinical information and pathological section were excluded. The gene expression data and corresponding clinical information of GSE68468, GSE41258, GSE22834, and GSE6988 data sets were collected from GEO data sets ( https://www.ncbi.nlm.nih.gov/geo/ ).
A total of 32 colon cancer patients who had lung metastasis were included in this study. Normal intestinal mucosal samples from normal adjacent tissue were used as control samples. The CD68, CD163, and SOCS3 status of all primary tumors and metastases and normal intestinal mucosal samples were conducted by IHC staining through standard staining procedures (SOCS3 antibody: ab16030, Abcam, Cambridge, UK; CD68 antibody: ZM0464; ZSGB-BIO, Beijing, China; CD163 antibody: ZM-0428; ZSGB-BIO, Beijing, China). Brownish-yellow cytoplasmic staining was determined to be positive on target tissue cells. Each sample was assessed with a minimum of 500 cells from 5 representative high-power fields (200 ×) and the proportion of positively colored cells was computed. Positive cell density was divided using a four-point method: 1 point for the proportion of positively stained cells < 25%, 2 points for those between 25%-50%, 3 points for those between 50%-75%, and 4 points for those > 75%. The color intensity of positively stained cells was divided according to the three-point method: 1 point for weakly positive cells, 2 points for moderately positive cells, and 3 points for strongly positive cells. The IHC staining score was calculated by multiplying the density value and color intensity of positive cells, and was determined by the four-point method: 0-1, negative; 2-4, weak ; 5-8, moderate ; and 9-12, strong . Two pathologists who were blind to the patient’s clinical characteristics underwent independent IHC staining assessment. All patients were divided into two groups based on IHC staining score: low expression (negative and +) and high expression (++ and +++).
DEGs between the low SOCS3 expression group and the high SOCS3 expression group in colon primary tumor and lung metastasis were compared with |log 2 FC| > 1 and P < 0.05, and the shared DEGs were displayed through the heat map (R package, pheatmap, v 1.0.12). STRING database ( https://string-db.org/ ) was used to conduct the relationship between DEGs, and the PPI network was displayed using the Cytoscape 3.7.1 software. DAVID ( https://david.ncifcrf.gov/tools.jsp ) was used to perform the Gene Ontology (GO) and the Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis for DEGs . The Venn diagram and enrichment factor bubble chart were performed using the OmicStudio tools at https://www.omicstudio.cn/tool .
Pearson’s chi-squared test and student’s t -test were used to evaluate the association of CD68, CD163, and SOCS3 status with gender, age at diagnosis, and stage. The student’s t -test was used to evaluate the differences in SOCS3 expression of the GEO data set. A P-value <0.05 was considered statistically significant. Data management and analysis were performed using R version 4.2, GraphPad Prism 8.0 Software (GraphPadSoftware, Inc., SanDiego, CA) and SPSS version 19.0 (SPSS Inc., Chicago, IL, USA). All methods were carried out by relevant guidelines.
Pan-cancer expression and prognostic performance of SOCS3 To figure out the difference in SOCS3 expression between normal tissue and tumor tissue in different cancer types, we explored the expression data in TCGA and GTEx databases. The results indicated that urologic neoplasms (including BLCA, KICH, KIRP) and respiratory tumors (including LUAD and LUSC), and ACC, LAML, and THCA were more inclined to low SOCS3 expression, whereas reproductive system neoplasms (including OV, TGCT, UCS) and gastrointestinal cancer (including ESCA, STAD, and PAAD) were more likely to high SOCS3 expression. In addition, the expression of SOCS3 was significantly downregulated in PRAD and BRCA, which belong to reproductive system neoplasms, and COAD and LIHC, members of gastrointestinal cancer . According to the Cox proportional hazards regression model, the SOCS3 expression was less obviously associated with prognosis and tumor progression. High SOCS3 expression patients harbored a poor prognosis in ESCA, GBM, HNSC, KIRC, LGG, STAD, and UVM, whereas harbored a better prognosis in BRCA . Subsequently, we divided patients into low expression group and high expression group by the median of SOCS3 expression, and performed the Kaplan–Meier analysis. The prognosis of the low SOCS3 expression group in LGG and KIRC was significantly better than that of the high SOCS3 expression group, whereas the opposite was in BRCA . However, over time, the prognosis of the two groups converged. To shed further light on the potential function of SOCS3 in pan-cancer, we divided patients into low and high expression groups in each tumor according to median levels of SOCS3 expression, and performed DEGs analysis. Subsequently, we performed GSEA through the GO functional annotation, KEGG pathway and the Hallmark pathway and summarized the pathways that were common to different tumors in the enrichment analysis results . For GO enrichment analysis, the high SOCS3 expression group was chiefly enriched in developmental regulation, regulation of cell differentiation, and immune response, whereas the low expression group was in organonitrogen compound biosynthetic process and oxidative phosphorylation. The results of the KEGG enrichment analysis revealed that patients with high SOCS3 expression mainly exhibited enrichment of the ECM receptor interaction, immune-related disease, and antigen processing and presentation. Those patients showed enrichment of epithelial mesenchymal transition, TNFα signaling via NFκb, and interferon response for the Hallmark pathway. Therefore, we speculated that there was a possible association between SOCS3 and the immune response and microenvironment in different tumors. Relationship between SOCS3 expression and immune response Based on the above discovery, we explored the relationship between SOCS3 expression and immune cell infiltration in different tumors through the TIMER database, including B cell, CD4+ T cell, CD8+ T cell, neutrophil, macrophage, and dendritic. It is observed that a large scale of tumors was positively related to immune cell infiltration, in which the four most significantly associated tumors were BRCA, LIHC, PRAD, and COAD . Through the ESTIMATE algorithm, immune scores were calculated in different tumors , for which the top three significantly associated tumors were BLCA (R= 0.54, P< 0.001), BRCA (R= 0.43, P< 0.001), and COAD (R= 0.69, P< 0.001). Besides, TMB and MSI, two biomarkers associated with the immunotherapy response, were analyzed concerning the SOCS3 expression in different tumors. The expression level of SOCS3 positively and significantly correlated with TMB in several tumors, including COAD, LAML, LGG, SARC, and THYM, whereas negatively and significantly correlated with that in UCEC, ESCA, LIHC, PAAD, PRAD, and THCA . COAD exhibited positive correlations between SOCS3 expression and MSI, and DLBC, HNSC, STAD, and UCEC exhibited negative correlations . Subsequently, we analyzed the relationship between SOCS3 expression and 11 published immune-gene signatures, including T cell-inflamed GEP, CAF, TAM M2, IFNG, CD8, CD274, TLS, TLS-melanoma, T cell Dysfunction, T cell exclusion, and MDSC. We found that the SOCS3 expression in the majority of tumors positively correlated with immune-gene signatures, except for TAM M2 . SOCS3 status in colon primary tumor and lung metastasis and its association with macrophage marker The function of SOCS3 is mainly focused on immune response, which can regulate macrophage and TH cell functions. Also macrophages in tumors can promote cancer development and metastasis through various mechanisms. Therefore, based on the previous research and the distinct relationship between SOCS3 status and the immune response in COAD, we collected tumor tissue and adjacent normal tissue from 32 colon cancer patients who had lung metastasis, and performed IHC staining with several markers, including CD68, CD163, and SOCS3. CD68 and CD163 are currently the most widely used macrophage marker and are predominantly located in cell membranes and lysosomal membranes. IHC staining indicated that there was a clear aggregation distribution effect on the leading edge of the colon primary tumors and lung metastasis focus for CD68 and CD163 markers, and this effect was more remarkable in lung metastasis than in colon primary tumor . Compared with adjacent normal intestinal mucosal tissue, SOCS3 status in colon primary tumor tissue was slightly weak, mainly with moderate positive staining in the cytoplasm . Nevertheless, SOCS3 and CD163 staining in lung metastasis showed more intense membrane and cytoplasm staining than in colon primary tumors . To search out the relationship between SOCS3 expression and clinical characteristics, we gathered clinical information from these patients during the first visit and calculated the IHC staining score of SOCS3. Patients who had lymph node metastasis and distant metastasis at the initial diagnosis preferred high SOCS3 expression in lung metastasis focus, whereas had no significant difference in colon primary tumors focus . Besides, the high IHC staining score of SOCS3 and CD163 mostly correlated with lung metastasis focus . Compared with the colon primary tumor, lung metastasis harbored higher CD163 and SOCS3 expression . Subsequently, we analyzed the correlation between SOCS3 expression and macrophage infiltration, and found that SOCS3 expression significantly correlated with macrophage markers and high SOCS3 expression was more inclined to high CD163 expression in lung metastasis . The results suggested that SOCS3 might be associated with macrophage infiltration and tumor invasion in lung metastasis. SOCS3 expression of the colon primary tumor and metastasis in the GEO database and its enrichment analysis Based on IHC staining results, we investigated transcription data in the GEO database to explore whether the transcript level and protein level have the same characteristics. The most common site of blood metastasis of colon cancer is the liver, followed by the lung, brain and bone. Transcript data sets consisting of the colon primary tumor and metastasis in the GEO database were used, including GSE68468 and GSE41258 (colon primary tumor and lung metastasis), GSE22834 and GSE6988 (colon primary tumor and liver metastasis). The CD68 and CD163 expression levels in metastasis were higher than in the colon primary tumor. The SOCS3 expression of lung metastasis was significantly higher than in the colon primary tumor, whereas that in liver metastasis was the opposite . These results revealed that SOCS3 shows different manifestations in different metastasis. To explore the molecular mechanisms of SOCS3 in lung metastasis, we divided the colon cancer and lung metastasis patients into two groups in the GSE68468 dataset respectively via the median of SOCS3. Subsequently, DEGs were compared between low SOCS3 expression and high SOCS3 expression in colon primary tumor and lung metastasis with |log 2 FC| > 1 and p < 0.05. 367 DEGs were screened out in colon primary tumors and 459 DEGs in lung metastasis, and 51 genes were shared by two sets of DEGs . There were 353 upregulated genes and 4 downregulated genes in the colon primary tumor, and 247 upregulated genes and 212 downregulated genes in lung metastasis . Interestingly, four genes that were originally highly expressed in the high SOCS3 expression group in colon cancer became low expressed after lung metastasis, including TTC40, PCDHA2, SP3P and ZNF471. Through STRING and Cytoscape, we conducted the interplay network of 51 shared DEGs and screened out two genes as hub genes, including CCL2 and PTGS2 . Furthermore, the enrichment analysis of GO and KEGG pathways was performed to explore the underlying interplay of the exceptional DEGs in lung metastasis. For GO analysis, 20 biological processes (BPs) were enriched, such as immune response, MHC class II protein related process, antigen processing, T cell activation, and inflammatory response . Regarding the KEGG pathway analysis, Rheumatoid arthritis, Th17 cell differentiation, and Graft−versus−host disease were enriched . These results showed that immune responses and regulations were significantly enriched in both GO and KEGG analysis.
To figure out the difference in SOCS3 expression between normal tissue and tumor tissue in different cancer types, we explored the expression data in TCGA and GTEx databases. The results indicated that urologic neoplasms (including BLCA, KICH, KIRP) and respiratory tumors (including LUAD and LUSC), and ACC, LAML, and THCA were more inclined to low SOCS3 expression, whereas reproductive system neoplasms (including OV, TGCT, UCS) and gastrointestinal cancer (including ESCA, STAD, and PAAD) were more likely to high SOCS3 expression. In addition, the expression of SOCS3 was significantly downregulated in PRAD and BRCA, which belong to reproductive system neoplasms, and COAD and LIHC, members of gastrointestinal cancer . According to the Cox proportional hazards regression model, the SOCS3 expression was less obviously associated with prognosis and tumor progression. High SOCS3 expression patients harbored a poor prognosis in ESCA, GBM, HNSC, KIRC, LGG, STAD, and UVM, whereas harbored a better prognosis in BRCA . Subsequently, we divided patients into low expression group and high expression group by the median of SOCS3 expression, and performed the Kaplan–Meier analysis. The prognosis of the low SOCS3 expression group in LGG and KIRC was significantly better than that of the high SOCS3 expression group, whereas the opposite was in BRCA . However, over time, the prognosis of the two groups converged. To shed further light on the potential function of SOCS3 in pan-cancer, we divided patients into low and high expression groups in each tumor according to median levels of SOCS3 expression, and performed DEGs analysis. Subsequently, we performed GSEA through the GO functional annotation, KEGG pathway and the Hallmark pathway and summarized the pathways that were common to different tumors in the enrichment analysis results . For GO enrichment analysis, the high SOCS3 expression group was chiefly enriched in developmental regulation, regulation of cell differentiation, and immune response, whereas the low expression group was in organonitrogen compound biosynthetic process and oxidative phosphorylation. The results of the KEGG enrichment analysis revealed that patients with high SOCS3 expression mainly exhibited enrichment of the ECM receptor interaction, immune-related disease, and antigen processing and presentation. Those patients showed enrichment of epithelial mesenchymal transition, TNFα signaling via NFκb, and interferon response for the Hallmark pathway. Therefore, we speculated that there was a possible association between SOCS3 and the immune response and microenvironment in different tumors.
Based on the above discovery, we explored the relationship between SOCS3 expression and immune cell infiltration in different tumors through the TIMER database, including B cell, CD4+ T cell, CD8+ T cell, neutrophil, macrophage, and dendritic. It is observed that a large scale of tumors was positively related to immune cell infiltration, in which the four most significantly associated tumors were BRCA, LIHC, PRAD, and COAD . Through the ESTIMATE algorithm, immune scores were calculated in different tumors , for which the top three significantly associated tumors were BLCA (R= 0.54, P< 0.001), BRCA (R= 0.43, P< 0.001), and COAD (R= 0.69, P< 0.001). Besides, TMB and MSI, two biomarkers associated with the immunotherapy response, were analyzed concerning the SOCS3 expression in different tumors. The expression level of SOCS3 positively and significantly correlated with TMB in several tumors, including COAD, LAML, LGG, SARC, and THYM, whereas negatively and significantly correlated with that in UCEC, ESCA, LIHC, PAAD, PRAD, and THCA . COAD exhibited positive correlations between SOCS3 expression and MSI, and DLBC, HNSC, STAD, and UCEC exhibited negative correlations . Subsequently, we analyzed the relationship between SOCS3 expression and 11 published immune-gene signatures, including T cell-inflamed GEP, CAF, TAM M2, IFNG, CD8, CD274, TLS, TLS-melanoma, T cell Dysfunction, T cell exclusion, and MDSC. We found that the SOCS3 expression in the majority of tumors positively correlated with immune-gene signatures, except for TAM M2 .
The function of SOCS3 is mainly focused on immune response, which can regulate macrophage and TH cell functions. Also macrophages in tumors can promote cancer development and metastasis through various mechanisms. Therefore, based on the previous research and the distinct relationship between SOCS3 status and the immune response in COAD, we collected tumor tissue and adjacent normal tissue from 32 colon cancer patients who had lung metastasis, and performed IHC staining with several markers, including CD68, CD163, and SOCS3. CD68 and CD163 are currently the most widely used macrophage marker and are predominantly located in cell membranes and lysosomal membranes. IHC staining indicated that there was a clear aggregation distribution effect on the leading edge of the colon primary tumors and lung metastasis focus for CD68 and CD163 markers, and this effect was more remarkable in lung metastasis than in colon primary tumor . Compared with adjacent normal intestinal mucosal tissue, SOCS3 status in colon primary tumor tissue was slightly weak, mainly with moderate positive staining in the cytoplasm . Nevertheless, SOCS3 and CD163 staining in lung metastasis showed more intense membrane and cytoplasm staining than in colon primary tumors . To search out the relationship between SOCS3 expression and clinical characteristics, we gathered clinical information from these patients during the first visit and calculated the IHC staining score of SOCS3. Patients who had lymph node metastasis and distant metastasis at the initial diagnosis preferred high SOCS3 expression in lung metastasis focus, whereas had no significant difference in colon primary tumors focus . Besides, the high IHC staining score of SOCS3 and CD163 mostly correlated with lung metastasis focus . Compared with the colon primary tumor, lung metastasis harbored higher CD163 and SOCS3 expression . Subsequently, we analyzed the correlation between SOCS3 expression and macrophage infiltration, and found that SOCS3 expression significantly correlated with macrophage markers and high SOCS3 expression was more inclined to high CD163 expression in lung metastasis . The results suggested that SOCS3 might be associated with macrophage infiltration and tumor invasion in lung metastasis.
Based on IHC staining results, we investigated transcription data in the GEO database to explore whether the transcript level and protein level have the same characteristics. The most common site of blood metastasis of colon cancer is the liver, followed by the lung, brain and bone. Transcript data sets consisting of the colon primary tumor and metastasis in the GEO database were used, including GSE68468 and GSE41258 (colon primary tumor and lung metastasis), GSE22834 and GSE6988 (colon primary tumor and liver metastasis). The CD68 and CD163 expression levels in metastasis were higher than in the colon primary tumor. The SOCS3 expression of lung metastasis was significantly higher than in the colon primary tumor, whereas that in liver metastasis was the opposite . These results revealed that SOCS3 shows different manifestations in different metastasis. To explore the molecular mechanisms of SOCS3 in lung metastasis, we divided the colon cancer and lung metastasis patients into two groups in the GSE68468 dataset respectively via the median of SOCS3. Subsequently, DEGs were compared between low SOCS3 expression and high SOCS3 expression in colon primary tumor and lung metastasis with |log 2 FC| > 1 and p < 0.05. 367 DEGs were screened out in colon primary tumors and 459 DEGs in lung metastasis, and 51 genes were shared by two sets of DEGs . There were 353 upregulated genes and 4 downregulated genes in the colon primary tumor, and 247 upregulated genes and 212 downregulated genes in lung metastasis . Interestingly, four genes that were originally highly expressed in the high SOCS3 expression group in colon cancer became low expressed after lung metastasis, including TTC40, PCDHA2, SP3P and ZNF471. Through STRING and Cytoscape, we conducted the interplay network of 51 shared DEGs and screened out two genes as hub genes, including CCL2 and PTGS2 . Furthermore, the enrichment analysis of GO and KEGG pathways was performed to explore the underlying interplay of the exceptional DEGs in lung metastasis. For GO analysis, 20 biological processes (BPs) were enriched, such as immune response, MHC class II protein related process, antigen processing, T cell activation, and inflammatory response . Regarding the KEGG pathway analysis, Rheumatoid arthritis, Th17 cell differentiation, and Graft−versus−host disease were enriched . These results showed that immune responses and regulations were significantly enriched in both GO and KEGG analysis.
Previous studies showed the abnormal expression of SOCS3 is widely involved in the pathogenesis of various tumors, including migration, cell growth, immune infiltration, inflammation response, and cell death related to various kinds of cytokines . Our study comprehensively investigated the expression and potential functions of SOCS3 in different tumors. We found that SOCS3 participated in the occurrence and development of a majority of tumors and presented different patterns in diverse tumors. SOCS3 referred to the prognosis and disease progression of a small part of tumors, and its high expression was more inclined to poor prognosis. Meanwhile, we also found that SOCS3 was positively correlated with main immune cell infiltration in almost each cancer type, especially in COAD. SOCS3 was a predictive marker for poor prognosis in central nervous system tumors and several gastrointestinal cancers, including ESCA and STAD. In COAD, LIHC and PRAD, the SOCS3 expression was significantly positively related to multiple immune cell infiltration. These results indicated that SOCS3 might be a potential target for immunotherapeutic intervention. However, patients with high SOCS3 expression in BRCA have a better prognosis and high immune cell infiltration. This result was also consistent with previous studies that SOCS3 inhibits breast cancer progression and metastasis . Macrophages played an important role in response to tumor immune through two different polarized statuses, including M1 macrophage, promoting tumor cell death via interleukin (IL) 12, and M2 macrophage, promoting tumor progression via IL10 . Suppression of SOCS3 in macrophages exerted anti-inflammatory and anti-tumor effects through the hyperactivation of STAT3 in the myeloid cell . Therefore, SOCS3 was recognized as an important feedback regulator mediating the macrophages’ activities and functions , and its peptidomimetic was a potential treatment to activate inflammatory cytokines in diseases . Based on existing studies of SOCS3 in immune response and its prominent association with immune response in COAD, we explored SOCS3 status and macrophage infiltration in colon cancer patients and the differences in the primary focus and lung metastases of colon cancer. Compared with the colon primary tumor, lung metastasis harbored higher CD163 and SOCS3 expression, and high SOCS3 expression was more likely to be associated with high CD163 expression in lung metastasis. These suggested that SOCS3 might be involved in the process of lung metastasis in colon cancer patients through its interaction with macrophages. Differential analysis and enrichment analysis of colon primary tumor and lung metastasis also suggested that immune response and Th cell activation were involved in this process. Interestingly, SOCS3 showed two completely different trends in liver metastases and lung metastases in colon cancer patients. Also COAD and READ were significantly different both in differences between tumor and adjacent normal tissue and prognosis, and in relationship with immune cells and mutation. This might suggest that SOCS3 had different functions in intestinal tumors with different anatomical locations and might influence the anatomical location of metastasis. Further diverse samples are required to validate these findings and relevant experiments were carried out to investigate the probable mechanism. However, there are several shortages in our study. Firstly, there was a data deviation due to a retrospective study based on the public database. Patients in the TCGA database were dominated by white people and massive missing data in clinical information might affect the analytical result. Besides, colon cancer samples in our study were relatively small and only performed IHC staining analysis, as well as lacking corresponding follow-up information. Furthermore, we need further a corresponding experiment to certify our findings. For the next step, we will collect more colon cancer and rectal cancer samples and corresponding metastatic tissues at different sites to perform transcriptomic and proteomic analysis and investigate the pattern of other immune cell infiltration. Meanwhile, we will explore the effect of SOCS3 alterations in tumor cells on surrounding macrophage polarization and its role in tumor invasion. Taken together, our study comprehensively investigated the expression pattern, prognosis performance and relationship with immune cell infiltration of SOCS3 in pan-cancer, and showed that SOCS3 possessed value as a prognostic marker and target for immunotherapeutic intervention in different tumors. Meanwhile, our research gave a new perspective on the relationship between SOCS3 status and macrophage infiltration in colon cancer patients with lung metastasis, which provided support for the possibility of SOCS3 as a potential target of tumor progression and tumor immunotherapy in colon cancer.
The original contributions presented in the study are included in the article/supplementary materials, further inquiries can be directed to the corresponding author/s.
The studies involving human participants were reviewed and approved by The Ethics Committee of the First Affiliated Hospital of Zhejiang University School of Medicine (the ethics approval number: IIT20210083B). Written informed consent for participation was not required for this study in accordance with the national legislation and the institutional requirements.
XS and GT designed the research study. XL, ZY, and BC performed the data analysis and interpretation. XL, ZY, LG and BC performed the IHC staining. XL, ZY, LG and BC performed the Statistical analysis. XL wrote the original draft. XL, ZY, LG and BC performed the writing review and editing. All authors contributed to the article and approved the submitted version.
|
Walter Marget and a brief history of paediatric infectious diseases in Munich, Germany | 2b8d3d1a-a8b2-4f52-be8d-b0b2ef4006d5 | 10071229 | Internal Medicine[mh] | |
IPNA–ESPN Junior Master Class—a decade of successful continuing education and training in pediatric nephrology | 95dba4e5-9696-4fdd-a896-c2a58d3538d6 | 10071458 | Internal Medicine[mh] | The European Society for Paediatric Nephrology (ESPN) together with the International Pediatric Nephrology Association (IPNA) has now conducted the broadest and most successful pediatric nephrology educational program in the world, training more than over 1000 participants from five continents, during the last 10 years. The IPNA–ESPN Junior Master Class program was launched in 2014 with the main objective of completing and harmonizing the pediatric nephrology activities and education in Europe. It was especially designed for fellows in pediatric nephrology, young pediatric nephrologists, residents in pediatrics, and pediatricians with special interest in pediatric nephrology. Pediatric kidney care maintains substantial disparities among European countries. Additionally, the requirements for qualification as a pediatric nephrologist vary from country to country . Furthermore, a high degree of perceived workforce inadequacy and difficulty in recruiting pediatric nephrology trainees exists in most regions of the world. Lack of interest, financial constraints, complexity of patients with kidney failure, and perceived pediatric nephrology workload are mentioned as dissuading factors by potential pediatric nephrology trainees . These factors support the organization of effective training and education programs, both at the undergraduate and postgraduate level, aiming at transfer of knowledge and experience from the most resourced centers, connecting physicians in developing countries with high standard pediatric nephrology training centers, and building novel collaboration between groups of young nephrologists working across Europe and around the globe . Following the pediatric nephrology syllabus , approved by the ESPN council and the European Academy of Paediatrics in 2019, the IPNA–ESPN Junior Master Class program comprises the whole pediatric nephrology curriculum over a seven-day course, distributed throughout annual educational sessions, to complete the curriculum in three years. Since 2014, three courses (nine sessions) have been delivered, educating a total of 1073 participants from 77 different countries, covering five continents, successfully connecting colleagues from middle and low-resourced countries through high-standard pediatric nephrology training centers across Europe and around the world.
The IPNA–ESPN Junior Master Class scientific program is organized tri-annually, covering the complete pediatric nephrology curriculum presented in 50 lectures. Pediatric nephrologists, nephrologists, urologists, radiologists, clinical geneticists, and pathologists, IPNA- and ESPN-certified teachers, are responsible for the lectures, covering all main clinical aspects from basic science to the newest updates in clinical pediatric nephrology, all based on published and systematic reviews and guidelines endorsed by ESPN, IPNA, and other recognized societies. ESPN holds responsibility for the scientific and local organization of this course, whenever possible located right before the annual ESPN conference, in the same venue, facilitating participation of students from different parts of Europe and allowing both participants and teachers to extend their working days together, and stimulating more effective and fruitful discussions and exchanges. Due to the COVID-19 pandemic, in 2020 and 2021, the courses were held online, and in 2022, a hybrid session was delivered. Every year, the IPNA–ESPN Junior Master Class program is evaluated by the European Accreditation Council for Continuing Medical Education (EACCME®) and the sessions are granted with a number of European CME credits (ECMEC®s) (Supplementary Table 1). A CME certificate is distributed to the attending participants after each session. Participants who complete the three-year course also receive a master diploma in pediatric nephrology recognized by ESPN and IPNA. Each IPNA–ESPN Junior Master Class program session runs two to three complete working days, according to the number of modules covered in that session. All lectures are scheduled for 30 min, followed by, at least, 10 min of complementary discussion on each topic. The global philosophy is to combine high-level education on all major topics (international academic speakers) and a session aiming at networking professors and participants from various nations (East–West/North–South connections). During the first session, fluids and electrolytes, radiology, embryology, antenatal hydronephrosis, CAKUT (congenital anomalies of the kidney and urinary tract), voiding dysfunction, vesicoureteral reflux, urinary tract infection, podocytopathies, Alport syndrome, renal tubular acidosis, mitochondrial kidney diseases, ciliopathies, ADPKD and ARPKD, and kidney involvement in metabolic disorders are reviewed. The second session is dedicated to nephrocalcinosis and urolithiasis, disorders of calcium and magnesium metabolism, post-infectious glomerulonephritis, IgA vasculitis and IgA nephropathy, nephrotic syndrome, haemolytic uremic syndrome, C3 glomerulopathies, kidney involvement in vasculitis, systemic lupus erythematosus nephritis, cystinosis, Bartter and Gitelman syndromes, and epidemiology of kidney diseases. The third session reviews the current concepts on diagnosis, management, treatment, and complications related to acute kidney injury, chronic kidney disease, dialysis, and kidney transplantation. An interactive session on kidney and transplant pathology is included in both the second and the third sessions.
The IPNA–ESPN Junior Master Class educational program is publicized on both IPNA and ESPN websites and newsletters. The participants are invited to submit their application by email, enclosing their brief curriculum vitae. This course attendance is exclusive and free to ESPN active members. Organizational and course expenses are totally covered by IPNA and ESPN, independent from pharmaceutical companies, following an application submitted every year to both societies.
Since 2014, the IPNA–ESPN Junior Master Class has educated 1073 participants from 77 different countries, a total of 522 unique trainees and young pediatric nephrologists during this time. A hundred and eighty-four participants (35%) from 43 countries have now concluded the 3-year course and accomplished the IPNA–ESPN Master in Pediatric Nephrology Program. Demographic characteristics of the participants are summarized in Table . Due to the COVID-19 pandemic, the third IPNA–ESPN Junior Master Class course was held online (2020 and 2021) and in hybrid format (2022), facilitating the access to this course for a higher number of participants from numerous nationalities, as shown in Fig. . The number of participants completing the IPNA–ESPN Junior Master Class program (three-year sessions) and receiving the IPNA–ESPN Master certificate in pediatric nephrology has also increased during this time (Fig. ). The distribution of participants in the Junior Master Class program is shown in Fig. . After each session, all students complete a multiple-choice written exam evaluating the course content. Since 2020, the IPNA–ESPN Junior Master participants who completed the three-year course are also invited to complete the ESPN board examination in pediatric nephrology, an online exam, consisting of 100 case-based multiple-choice questions covering all aspects of pediatric nephrology, to earn a new certification and a quality label of theoretical knowledge and adequate clinical thinking in pediatric nephrology, implemented by the ESPN educational committee .
In accordance with the European Accreditation Council for Continuing Medical Education (EACCME®) requirements, each course is appraised by the participants. The participants grade the quality of the speakers, the quality of the course contents, the organization, the impact of this educational activity on their current practice, and the new knowledge acquired. The results are presented on a 0–10 scale with 10 being the highest rating, in Fig. . The quality of the speakers, quality of the content, and organization of the course were all rated > 9 by the participants. Some original comments from various participants are presented in Supplementary Table 2.
The IPNA–ESPN Junior Master Class program was designed to provide junior pediatric nephrologists with a comprehensive and structured educational activity that would help to improve training activities in knowledge domains and meet the educational needs of junior pediatric nephrologists in Europe . Considering the importance of the interaction between participants, the first course, in 2014, was originally designed to host a total of 50 participants in each session. The high number of applications, with diverse origins and heterogeneous training backgrounds, encouraged us to increase the number of offered places, enriching the value of the lectures and the importance of the discussions. Moreover, this IPNA–ESPN educational activity has proved to be successful in connecting physicians in developing countries with high standard pediatric nephrology training centers across Europe and around the globe . Interventions to support the growth of the pediatric nephrology workforce include developing and sharing novel teaching methods, strengthening existing trainee relationships, and exposing potential trainees to nephrology early in their education . The IPNA–ESPN Junior Master Class program has been able to successfully train a high number of junior physicians, coming from diverse regions of the globe, to complete the three-year course, and achieve the IPNA–ESPN Master in Pediatric Nephrology certificate. ESPN prioritizes the harmonization of training curricula in Europe and the development of European recognition criteria for pediatric nephrologists . The European Accreditation Council for Continuing Medical Education (EACCME®) and the number of European CME credits (ECMEC®s) granted indicate the quality of the IPNA–ESPN Junior Master Class program, and underscores its standing as an effective training program, the program’s contribution to addressing global pediatric nephrology workforce needs and the continued high-level of education delivered by the program. The COVID-19 pandemic had a profound impact on the organization of medical seminars and congresses. Notwithstanding the importance of social and face-to-face interaction, the advent of online and hybrid meetings, with the possibility for live-streaming and recording lectures that are made available for enduring use, increased the number of attendees from different countries after 2020. The long-term impact of this educational approach still needs to be followed and further investigated.
The IPNA–ESPN Junior Master Class program is one of the most important educational activities sponsored by ESPN, continuously growing since 2014. At present, it is the main support for pediatric nephrology continuous education and training for the ESPN board examination . Despite the growing popularity and success, over the next years, IPNA and ESPN can expect various challenges. First, the balance between the continuation of online format and the interest by learners to return to face-to-face meetings needs to be considered. While in terms of costs and accessibility to participants from different regions, the online format is clearly advantageous; the importance of networking and the direct interaction between the participants and the teachers during face-to-face meetings cannot be neglected. Participants have expressed on the feedback questionnaire a preference for in-person instruction, although approximately one-third favor the online format. To best meet the needs and preferences of the participants, we can best serve them and deliver education to the greatest number of participants through the interactive hybrid format which will provide wider outreach and accessibility, decrease costs for many participants (and the sponsoring organizations), reduce travel time and expenses, and decrease the environmental footprint of air travel by so many. Second, considering the tremendous scientific developments during the last few years, we may have to expand the number of presentations included in this course, and also work to assure development of the competencies of the practicing pediatric nephrologist in a variety of settings can be aided through the course. Finally, additional funding methods may be required to provide this first in class education to those participants in need.
The IPNA–ESPN Junior Master Class educational program has significantly contributed to development of the pediatric nephrology workforce in Europe and throughout the world. This program has also improved collaborations between junior pediatric nephrologists around the globe and thereby stimulated additional clinical advances and research opportunities. The quality of care for children with kidney diseases is dependent on the number of well-trained health care providers, their dedication, and the health care systems they can organize. The IPNA–ESPN Junior Master Class program has provided a major improvement in pediatric nephrology training activities, enabling students to develop better knowledge, attitudes, and clinical skills than they likely would not have gained through their training programs alone. The evident friendly cooperation between groups of nephrologists from different countries, working together to deliver this education, provides a remarkable example to the young pediatric nephrologists who will be caring for patients and families for some time. Despite the number of challenges, ESPN has made a joint commitment with IPNA to continue this Junior Master Class program in the forthcoming years.
Below is the link to the electronic supplementary material. Supplementary file1 (DOCX 15 KB)
|
Patient perspectives on data sharing regarding implementing and using artificial intelligence in general practice – a qualitative study | 79d00ecf-10c1-4844-b8f1-ea01cd18deb3 | 10071604 | Family Medicine[mh] | Many healthcare systems are currently under pressure because of an increasing longevity of life leading to more elderly and a growing number of patients with complex multimorbidity . General practitioners (GPs) and other healthcare professionals in general practice must work faster to keep up with the increasing workload. At the same time, the requirements for documentation and administration are increasing and general practice in particular feels the pressure since the patients’ first point of contact with the healthcare system happens here, and because general practice works as gatekeeper to the secondary sector. The increasing pressure on general practice is resulting in less time for talking to and examining the patients . In recent years, the development and use of data-driven artificial intelligence (AI) has also increased dramatically. In 2019 the Danish government proposed a national strategy for implementing AI in the healthcare system. The strategy was proposed because of the possibility for decision support tools and streamlining the healthcare system through their use and the sharing of the large amount of data which is available in the healthcare system . However, existing literature shows that when it comes to sharing data, patients are concerned with aspects regarding confidentiality, security, privacy, and trust in the unit or organization conducting the research . Patients seem most willing to share data with the established healthcare system and least likely to share data with insurance companies, companies associated with biotechnology, and pharmaceutical products . With AI added to the equation, a new perspective appears. Some studies have shown that patients are more receptive to the use of AI if it is implemented and used in a way that preserves the patient-doctor relationship, partly due to fear of replacement of humans and partly because AI challenges the humanistic aspect of health care . It has been suggested that the possibility for continuity and longevity of contact with the GP, along with the broad range of health-related concerns that people seek care for in general practice makes the patient-GP relationship particularly important in general practice . It has been argued that to provide the patient with good care and effective encounters, there is a need for a healthy patient-GP relationship based on trust and good communication . Trust is needed to create an atmosphere in which honest and good communication can arise, and it has been suggested that the quality of the patient-GP relationship can be directly related to the trust between them . In relation to medical doctors there is an expected level of competence since doctors’ medical degree and their right to practice mark a passage to being a professional , and due to their level of education and competences doctors are awarded status, power and authority in society . In fact, a study has shown that doctors are one of the most trusted professions and have the highest status in society both among the population and other professions . However, beginning around the turn of the 21st century and continuing to present day, there have been changes in the way of thinking about trust in relation to doctors as a profession, including GPs themselves. Studies point towards reduced professional status as a driving force in and around medicine . This may partly be because of the introduction of new medical technologies that can limit doctors’ fields of expertise and make them more dependent on technology . In addition to this, online health information is now freely available, and the use of Google searching for health advice is widely used . This access to information has made it possible for people to educate themselves online about medical conditions or health-related issues without having to consult with their GP . Another important aspect regarding trust in the patient-GP relationship is the sharing of health data. Trust regarding sharing health data is important in order to implement and use AI in general practice in a way with which patients’ feel comfortable and currently, sharing and reuse of patient’s health data has shown to be a sensitive matter that has developed into public issues in some European countries . Furthermore, trust in AI is a complex matter. In an article on a theory of trust for AI in healthcare, it is pointed out that the achievement of trust in AI in a healthcare setting can be even more complex than trust in AI in general, which can partly be explained by the limited existence of public literature about AI in a general practice context . It is argued that trust in technologies should be complemented by trust in those producing the technology and that such trust can only be achieved if the companies producing the technology are transparent about their data use and policies. When data is used to promote AI, the people providing the data should be aware of how their data are handled, stored, and how it is being used . Although trust in AI is a complex matter, some patients view the use of innovative techniques such as AI for processing health data as an opportunity for GPs to benefit from the use of AI solutions in the medical process . Furthermore, research points out that some patients perceive AI as a diagnostic tool that can increase the diagnostic speed, which they view as beneficial and possibly lifesaving, an understanding that is based on AI’s ability to draw on more data or experience than humans . The above observations call for more research on patient perspectives on data sharing specifically regarding AI in general practice, in order to expand our understanding of the potential challenges surrounding this. Therefore, the aim of this paper is to uncover new patient perspectives on the importance of trust regarding the patient-GP relationship, data sharing and AI in general practice.
The authors have followed the Standards for Reporting Qualitative Research (SRQR) table guidelines for reporting qualitative studies when conducting the section . Researchers’ characteristics and reflexivity The first author is a Ph.D. student with a background in health promotion and psychology, and patient perspectives is therefore a fundamental focus point when investigating implementing and using technology in general practice. The second author is a Ph.D. student with a background in medicine with industrial specialization and is preoccupied with AIs place in the health care system, especially regarding decision support. The third author is a researcher with specialty in qualitative methods and is also focused on patient perspectives regarding implementing technology in general practice. The fourth author is a professor and GP and is preoccupied with using technology as a support tool for general practice. The final author is a professor and GP and is focused on the implementation of technology in general practice. All the authors have the assumption that technology can be helpful in general practice, if it is implemented and used in an ethical manner that comply with the law regarding data protection. Context The study took place in the North Denmark Region, and in Denmark the healthcare system is public and free for all citizens financed by taxes and almost all patients are listed with a GP. The GPs are paid by capitation and fee-for-service reimbursement . GPs are responsible for providing patients with palliative care and can refer patients to specialist treatment if needed. GPs are also responsible for patient care 24hours a day . Sampling strategy The strategy was to get a broad perspective with both male and female patients in different age groups and professions. Other than that, the inclusion criteria for patients were that they must be Danish citizens above 18 years of age and registered at a general practice clinic. We only included a sample of 10 interviewees in the study, since we included based on the concept information power, which relates to the study purpose, quality of dialogue and analysis strategy , and when the same perspectives kept coming up we felt a data saturation. Data collection methods and interviewees The study consists of 10 patient interviews carried out between October 2019 and January 2022. Nine interviews were carried out by the second author and one interview by the first author. All the patients (interviewees) were recruited from the North Denmark Region through different Facebook groups. A Facebook post explaining the project was made and movie tickets were offered as compensation. The post was published on different city groups in North Denmark Region as well as the “ AAU (Aalborg University) seek, find and become test subject ” Facebook group. Each interview lasted 30 to 45 minutes and the interviewees decided where the interviews were held. The interviewees ended up consisting of four men 25 to 74 years in age, and six women 27 to 46 years of age at the time of the interview. The ten interviewees’ educational background years ranged from short to medium to long and some were students. See Table regarding interviewee information. Data collection instruments and technologies We used an audio recorder for data collection, and all interviews were transcribed verbatim by the first author. The Nvivo13 software program was used to transcribe and code the data by marking and naming selections of text within all datasets. Vignette method The interviews were conducted with the vignette method, which is a methodological tool that entails a written, visual, or crafted incident and then presenting it to the interviewees in order to elicit their opinions and perspectives . Vignettes are designed to simulate events in a hypothetical way to investigate how people might react to such events . Vignettes seek the understanding of people’s attitudes, perceptions, and beliefs, especially concerning sensitive subjects such as healthcare . For this study, three vignettes were developed in written form and the patient referred to in the vignettes is fictional (Appendix 1). We found this method particularly relevant for this study since patients in Denmark do not have much experience with AI in general and do not know much about data sharing in relation to implementing and using AI in general practice. Even though we refer to data-driven AI, we were not explicit about this towards the interviewees, who were only presented with the term “artificial intelligence”, since they probably would not have gained any further knowledge from the information, since they did not know much about AI in general. Right before the interviews started, the interviewees were asked about their age, where they live, education, and profession. We used a semi-structured interview guide (Script, Appendix 2) to enable unforeseen perspectives and creativity to come forward. The script consisted of different themes with multiple questions for each theme. The interviews started with a vignette and from there on the questions began. The first line of questions regarded understanding of the first vignette, health data, and AI. After the second vignette, the questions concerned understanding of the vignette, when should data be shared/made available, sensitivity of data, and data security. After the third and final vignette, the questions involved understanding of the vignette and trust in the use of AI in general practice. Data processing and data analysis An inductive approach was applied to analyze the data and we used member checking as a technique to enhance trustworthiness and credibility. The first author carried out a thematic analysis using the six phases of analysis , with inputs and feedback from the rest of the authors. Transcribing the data is the first key phase in a thematic analysis when interpreting data, and while listening to the interviews and reading, ideas and notes were made. The second phase regarded generating the initial codes by using an open coding approach. In this case, the coding began by re-reading all the transcriptions and review notes, and then looking for patterns and perspectives that occurred multiple times and in several interviews. Then the third phase began, which involved searching for themes by analyzing the codes, and sorting them into themes, see Fig. below. In practice this meant exploring the codes’ relations to each other and seeing how they fit into the overall story about the data. The fourth phase was reviewing the themes, the dataset was re-read, and then the fifth phase started, which involved defining and naming the themes, the final themes were defined and named based on the codes and subcodes found in Fig. . The sixth and last phase regarded producing the report. The themes ended up being 1) Patient-GP relationship, 2) Willingness for data sharing, 3) Worries about data sharing, 4) Artificial intelligence’s possibilities in general practice, and 5) Worries about the use of artificial intelligence in general practice.
The first author is a Ph.D. student with a background in health promotion and psychology, and patient perspectives is therefore a fundamental focus point when investigating implementing and using technology in general practice. The second author is a Ph.D. student with a background in medicine with industrial specialization and is preoccupied with AIs place in the health care system, especially regarding decision support. The third author is a researcher with specialty in qualitative methods and is also focused on patient perspectives regarding implementing technology in general practice. The fourth author is a professor and GP and is preoccupied with using technology as a support tool for general practice. The final author is a professor and GP and is focused on the implementation of technology in general practice. All the authors have the assumption that technology can be helpful in general practice, if it is implemented and used in an ethical manner that comply with the law regarding data protection.
The study took place in the North Denmark Region, and in Denmark the healthcare system is public and free for all citizens financed by taxes and almost all patients are listed with a GP. The GPs are paid by capitation and fee-for-service reimbursement . GPs are responsible for providing patients with palliative care and can refer patients to specialist treatment if needed. GPs are also responsible for patient care 24hours a day .
The strategy was to get a broad perspective with both male and female patients in different age groups and professions. Other than that, the inclusion criteria for patients were that they must be Danish citizens above 18 years of age and registered at a general practice clinic. We only included a sample of 10 interviewees in the study, since we included based on the concept information power, which relates to the study purpose, quality of dialogue and analysis strategy , and when the same perspectives kept coming up we felt a data saturation.
The study consists of 10 patient interviews carried out between October 2019 and January 2022. Nine interviews were carried out by the second author and one interview by the first author. All the patients (interviewees) were recruited from the North Denmark Region through different Facebook groups. A Facebook post explaining the project was made and movie tickets were offered as compensation. The post was published on different city groups in North Denmark Region as well as the “ AAU (Aalborg University) seek, find and become test subject ” Facebook group. Each interview lasted 30 to 45 minutes and the interviewees decided where the interviews were held. The interviewees ended up consisting of four men 25 to 74 years in age, and six women 27 to 46 years of age at the time of the interview. The ten interviewees’ educational background years ranged from short to medium to long and some were students. See Table regarding interviewee information.
We used an audio recorder for data collection, and all interviews were transcribed verbatim by the first author. The Nvivo13 software program was used to transcribe and code the data by marking and naming selections of text within all datasets.
The interviews were conducted with the vignette method, which is a methodological tool that entails a written, visual, or crafted incident and then presenting it to the interviewees in order to elicit their opinions and perspectives . Vignettes are designed to simulate events in a hypothetical way to investigate how people might react to such events . Vignettes seek the understanding of people’s attitudes, perceptions, and beliefs, especially concerning sensitive subjects such as healthcare . For this study, three vignettes were developed in written form and the patient referred to in the vignettes is fictional (Appendix 1). We found this method particularly relevant for this study since patients in Denmark do not have much experience with AI in general and do not know much about data sharing in relation to implementing and using AI in general practice. Even though we refer to data-driven AI, we were not explicit about this towards the interviewees, who were only presented with the term “artificial intelligence”, since they probably would not have gained any further knowledge from the information, since they did not know much about AI in general. Right before the interviews started, the interviewees were asked about their age, where they live, education, and profession. We used a semi-structured interview guide (Script, Appendix 2) to enable unforeseen perspectives and creativity to come forward. The script consisted of different themes with multiple questions for each theme. The interviews started with a vignette and from there on the questions began. The first line of questions regarded understanding of the first vignette, health data, and AI. After the second vignette, the questions concerned understanding of the vignette, when should data be shared/made available, sensitivity of data, and data security. After the third and final vignette, the questions involved understanding of the vignette and trust in the use of AI in general practice.
An inductive approach was applied to analyze the data and we used member checking as a technique to enhance trustworthiness and credibility. The first author carried out a thematic analysis using the six phases of analysis , with inputs and feedback from the rest of the authors. Transcribing the data is the first key phase in a thematic analysis when interpreting data, and while listening to the interviews and reading, ideas and notes were made. The second phase regarded generating the initial codes by using an open coding approach. In this case, the coding began by re-reading all the transcriptions and review notes, and then looking for patterns and perspectives that occurred multiple times and in several interviews. Then the third phase began, which involved searching for themes by analyzing the codes, and sorting them into themes, see Fig. below. In practice this meant exploring the codes’ relations to each other and seeing how they fit into the overall story about the data. The fourth phase was reviewing the themes, the dataset was re-read, and then the fifth phase started, which involved defining and naming the themes, the final themes were defined and named based on the codes and subcodes found in Fig. . The sixth and last phase regarded producing the report. The themes ended up being 1) Patient-GP relationship, 2) Willingness for data sharing, 3) Worries about data sharing, 4) Artificial intelligence’s possibilities in general practice, and 5) Worries about the use of artificial intelligence in general practice.
Patient-GP relationship One of the main findings of the study was that the interviewees felt comfortable around their GP and had general trust in them. This was reflected in the interviewees’ willingness to share their data if their GP asked them to. The interviewees’ willingness to share in this context indicates that they trust their GP, and five of the interviewees, two men and three women, mention that they have a close relationship with their GP or believe there is some kind of trust bond between patients and their GP. When asked about their feelings towards sharing data with the GP, four of the interviewees, two men and two women, began wondering about the GPs authority and one of the interviewees stated: “ You have trust towards your GP, and they are kind of an authority person, they know best, and you have to trust that. It is also a form of power ” (I 9). The interviewees that thought of their GP as an authority felt that authority contributed to them feeling in safe hands regarding data sharing. However, one interviewee expressed concerns about feeling pressured to please the GP and the fact that patients are deeply dependent on their GP: “ In that situation one will feel pressured to please the GP and say, “oh yes of course ”” (I 6). A specific concern form one of the interviewees related to the reduction in people’s trust in authorities in general: “ There has probably been a movement in relation to how much people believe in authorities. I can feel that in my profession (as a policeman), so the level of how much a person believe in authority will probably decrease over the years I’m afraid ” (I 10). Willingness for data sharing The purpose behind data sharing was relevant for the interviewees, and they all found implementing and using AI in general practice to be a purpose for which they would be willing to share their data. One interviewee noted: “ Both developing AI to help the GP and cure illnesses are in the same category. They are both equally good causes that I would share my data for ” (I 3). Another perspective appeared when the interviewees were asked about the difference between sharing health data and personal data such as e-mail address, home address and social security number. Three of the interviewees, two men and one woman, felt they had nothing interesting in their health journal and nothing to hide, so they had no problem sharing their health data. However, they did not like the thought of sharing their personal data, since that was more sensitive to them, as one of the interviewees stated: “ My medical record is really boring, so my personal information would probably be more critical to share, although I share them the most already ” (I 3). When the interviewees were asked about what kind of health data are more sensitive than other data, early retirement and mental illnesses came up several times. Two of the interviewees, a man and a woman, referred to mental illnesses such as depression as taboo. One of the interviewees noted: “ I could imagine that the depression would be a sensitive subject for her [the woman in the vignette] because it is a mental illness, there is probably also many emotions involved ” (I 1). When asked about data sharing in relation to sensitive data, eight of the interviewees, three men and five women, expressed that having sensitive health information in their journal would make them more skeptical and influence their decision to share data in a negative way. However, a common opinion among six of the interviewees, three men and three women, was that the assurance of anonymization made it feel safer and easier for them to agree to sharing their data. However, two of the interviewees, two women, still had reservations towards the use of their data concerning access to the data and how the data key could be recreated after anonymization or pseudonymization. Therefore, the interviewees wanted to be more involved in the actual process of data sharing. The interviewees were also asked about the topic of data sharing without consent, and many different opinions arose. An opinion between three of the female interviewees was that data should not be shared without consent no matter what, but four of the interviewees, one man and three women, felt that there could be exceptions if something terrible was to happen, as one interviewee noted: “ It should only be in extreme cases… an epidemic that could hurt half of the Danish population ” (I 3). Another reservation that four of the interviewees, two men and two women, expressed regarded fear of sharing their data due to the possibility for monitoring, deprivation of liberty, general misuse, hacking or further data sharing than they had agreed to. The interviewees that were willing to share their data still expressed some reservations regarding security and who would be able to access their data. One interviewee noted: “ There has to be some sort of limit, so everyone cannot snoop around ” (I 4). Six of the interviewees, two men and four women, had concerns specifically regarding commercial or private companies, employers, the public, job centers, the state, and the medical industry. The concerns seemed to come from the interviewees fear of the above-mentioned misuse of their data. Four of the interviewees, two men and two women, pictured themselves as examples and got worried about what would happen to them if their data were shared. One interviewee expressed: “ I have a sleep-app, that measures when I go to bed and when I wake up, and sometimes you can go to bed late and I have a sleeping illness, and I think what if the job center discovered that… What can they use it for? It would probably not benefit me ” (I 7). Artificial intelligence’s possibilities for general practice All of the interviewees could see possibilities for using AI in general practice and the most commonly perceived benefit among all the interviewees was if the GP could use AI as a support tool. Half of the interviewees expressed that they would even feel safer if the GP used AI as a support tool when diagnosing. This opinion was partly explained by the fact that GPs are quite busy and thus the risk of the GPs overlooking something important regarding patients’ health was perceived as a possibility that AI could prevent. One interviewee stated: “ Maybe AI could help keeping track of patient records… AI could get an overview fast and see that this patient has now shown these symptoms for the fifth time, so maybe it is time to look into that instead of the GP missing it ” (I 2). Three female interviewees mentioned that humans, including GPs, make mistakes, which they viewed as another good reason to use AI as a support tool. For example, one interviewee stated: “ We are only humans, so is the GP. Making an extra check with AI would make me feel safer ” (I 5). However, the interviewees still felt that the GP should be the primary decision maker and AI should only be used as a support tool in general practice. Some of the positive perspectives among five of the interviewees, one man and four women, regarding AI in general practice revolved around AI’s ability to find patterns in data, develop overviews, work faster than the GP, and improve the quality of problem solving. However, when asked if and how AI would affect the GPs’ workflow, one interviewee noted: “ It is a radical change compared to how health is being practiced now ” (I 3). Four other interviewees, two men and two women, thought that with time patients would begin to trust AI, especially if AI proved to be continuously correct and AI received more exposure in society. A couple of the interviewees compared it to the fact that nowadays people are used to other technologies such as phones and tablets, and with time it could be the same with AI. A topic that generated many different perspectives was if AI could be used to detect future illnesses. The different perspectives could be explained by the fact that the interviewees had never thought about this scenario before. If AI detected an illness three of the interviewees, two men and a woman, would want to know about it instantly, while two other female interviewees did not. Those who wanted to know right away, wanted to know so they could change their ways and hopefully prevent the illness from progressing. One interviewee noted:” It would be amazing if the illness could be detected by AI and prevented earlier, instead of trying all sorts of things before knowing what it actually is ” (I 8). Two female interviewees perceived the possibility of AI detecting future illnesses as an ethical dilemma. The ones who did not want to know thought knowing could lead to anxiety, paranoia, not being able to enjoy the present moment, or thought it would get too expensive for the healthcare system due to over-diagnosis and increased focus on disease prevention. Worries about the use of AI in general practice A worry regarding AI among six of the interviewees, three men and three women, concerned AI taking over the GPs’ position and the patients losing their relationship with the GP. As one interviewee stated: “ A person does not have a trustful relationship with a machine ” (I 3). Related worries among the interviewees dealt with the fear that using AI in general practice could change the GPs’ role from health care provider to data collector or give AI too much influence and possibly outsmart the GPs. When talking about what AI could possibly be used for in general practice in the future, three of the interviewees, one man and two women, expressed concern about AI not being able to detect emotional states and show a rather one-sided perspective only based on non-emotional health data. Because of this worry, the interviewees did not think AI could ever provide a complete overview of the patients, and therefore they felt like the GP should be critical towards using it. Lack of trust in AI was also reflected in one of the interviewees’ statements: “ AI is not a living creature and depending on what you feed it with, it can learn different things, so it is important to be critical ” (I 2). The critical mindset towards AI was also expressed when the interviewees were asked about their trust in a potential diagnosis set by AI. Two female interviewees were directly against the possibility that AI could one day make diagnose by itself, while two others, a man and a woman, were concerned about trusting AI’s “opinion” and whether the AI dataset could be large enough to provide a precise diagnosis.
One of the main findings of the study was that the interviewees felt comfortable around their GP and had general trust in them. This was reflected in the interviewees’ willingness to share their data if their GP asked them to. The interviewees’ willingness to share in this context indicates that they trust their GP, and five of the interviewees, two men and three women, mention that they have a close relationship with their GP or believe there is some kind of trust bond between patients and their GP. When asked about their feelings towards sharing data with the GP, four of the interviewees, two men and two women, began wondering about the GPs authority and one of the interviewees stated: “ You have trust towards your GP, and they are kind of an authority person, they know best, and you have to trust that. It is also a form of power ” (I 9). The interviewees that thought of their GP as an authority felt that authority contributed to them feeling in safe hands regarding data sharing. However, one interviewee expressed concerns about feeling pressured to please the GP and the fact that patients are deeply dependent on their GP: “ In that situation one will feel pressured to please the GP and say, “oh yes of course ”” (I 6). A specific concern form one of the interviewees related to the reduction in people’s trust in authorities in general: “ There has probably been a movement in relation to how much people believe in authorities. I can feel that in my profession (as a policeman), so the level of how much a person believe in authority will probably decrease over the years I’m afraid ” (I 10).
The purpose behind data sharing was relevant for the interviewees, and they all found implementing and using AI in general practice to be a purpose for which they would be willing to share their data. One interviewee noted: “ Both developing AI to help the GP and cure illnesses are in the same category. They are both equally good causes that I would share my data for ” (I 3). Another perspective appeared when the interviewees were asked about the difference between sharing health data and personal data such as e-mail address, home address and social security number. Three of the interviewees, two men and one woman, felt they had nothing interesting in their health journal and nothing to hide, so they had no problem sharing their health data. However, they did not like the thought of sharing their personal data, since that was more sensitive to them, as one of the interviewees stated: “ My medical record is really boring, so my personal information would probably be more critical to share, although I share them the most already ” (I 3). When the interviewees were asked about what kind of health data are more sensitive than other data, early retirement and mental illnesses came up several times. Two of the interviewees, a man and a woman, referred to mental illnesses such as depression as taboo. One of the interviewees noted: “ I could imagine that the depression would be a sensitive subject for her [the woman in the vignette] because it is a mental illness, there is probably also many emotions involved ” (I 1). When asked about data sharing in relation to sensitive data, eight of the interviewees, three men and five women, expressed that having sensitive health information in their journal would make them more skeptical and influence their decision to share data in a negative way. However, a common opinion among six of the interviewees, three men and three women, was that the assurance of anonymization made it feel safer and easier for them to agree to sharing their data. However, two of the interviewees, two women, still had reservations towards the use of their data concerning access to the data and how the data key could be recreated after anonymization or pseudonymization. Therefore, the interviewees wanted to be more involved in the actual process of data sharing. The interviewees were also asked about the topic of data sharing without consent, and many different opinions arose. An opinion between three of the female interviewees was that data should not be shared without consent no matter what, but four of the interviewees, one man and three women, felt that there could be exceptions if something terrible was to happen, as one interviewee noted: “ It should only be in extreme cases… an epidemic that could hurt half of the Danish population ” (I 3). Another reservation that four of the interviewees, two men and two women, expressed regarded fear of sharing their data due to the possibility for monitoring, deprivation of liberty, general misuse, hacking or further data sharing than they had agreed to. The interviewees that were willing to share their data still expressed some reservations regarding security and who would be able to access their data. One interviewee noted: “ There has to be some sort of limit, so everyone cannot snoop around ” (I 4). Six of the interviewees, two men and four women, had concerns specifically regarding commercial or private companies, employers, the public, job centers, the state, and the medical industry. The concerns seemed to come from the interviewees fear of the above-mentioned misuse of their data. Four of the interviewees, two men and two women, pictured themselves as examples and got worried about what would happen to them if their data were shared. One interviewee expressed: “ I have a sleep-app, that measures when I go to bed and when I wake up, and sometimes you can go to bed late and I have a sleeping illness, and I think what if the job center discovered that… What can they use it for? It would probably not benefit me ” (I 7).
All of the interviewees could see possibilities for using AI in general practice and the most commonly perceived benefit among all the interviewees was if the GP could use AI as a support tool. Half of the interviewees expressed that they would even feel safer if the GP used AI as a support tool when diagnosing. This opinion was partly explained by the fact that GPs are quite busy and thus the risk of the GPs overlooking something important regarding patients’ health was perceived as a possibility that AI could prevent. One interviewee stated: “ Maybe AI could help keeping track of patient records… AI could get an overview fast and see that this patient has now shown these symptoms for the fifth time, so maybe it is time to look into that instead of the GP missing it ” (I 2). Three female interviewees mentioned that humans, including GPs, make mistakes, which they viewed as another good reason to use AI as a support tool. For example, one interviewee stated: “ We are only humans, so is the GP. Making an extra check with AI would make me feel safer ” (I 5). However, the interviewees still felt that the GP should be the primary decision maker and AI should only be used as a support tool in general practice. Some of the positive perspectives among five of the interviewees, one man and four women, regarding AI in general practice revolved around AI’s ability to find patterns in data, develop overviews, work faster than the GP, and improve the quality of problem solving. However, when asked if and how AI would affect the GPs’ workflow, one interviewee noted: “ It is a radical change compared to how health is being practiced now ” (I 3). Four other interviewees, two men and two women, thought that with time patients would begin to trust AI, especially if AI proved to be continuously correct and AI received more exposure in society. A couple of the interviewees compared it to the fact that nowadays people are used to other technologies such as phones and tablets, and with time it could be the same with AI. A topic that generated many different perspectives was if AI could be used to detect future illnesses. The different perspectives could be explained by the fact that the interviewees had never thought about this scenario before. If AI detected an illness three of the interviewees, two men and a woman, would want to know about it instantly, while two other female interviewees did not. Those who wanted to know right away, wanted to know so they could change their ways and hopefully prevent the illness from progressing. One interviewee noted:” It would be amazing if the illness could be detected by AI and prevented earlier, instead of trying all sorts of things before knowing what it actually is ” (I 8). Two female interviewees perceived the possibility of AI detecting future illnesses as an ethical dilemma. The ones who did not want to know thought knowing could lead to anxiety, paranoia, not being able to enjoy the present moment, or thought it would get too expensive for the healthcare system due to over-diagnosis and increased focus on disease prevention.
A worry regarding AI among six of the interviewees, three men and three women, concerned AI taking over the GPs’ position and the patients losing their relationship with the GP. As one interviewee stated: “ A person does not have a trustful relationship with a machine ” (I 3). Related worries among the interviewees dealt with the fear that using AI in general practice could change the GPs’ role from health care provider to data collector or give AI too much influence and possibly outsmart the GPs. When talking about what AI could possibly be used for in general practice in the future, three of the interviewees, one man and two women, expressed concern about AI not being able to detect emotional states and show a rather one-sided perspective only based on non-emotional health data. Because of this worry, the interviewees did not think AI could ever provide a complete overview of the patients, and therefore they felt like the GP should be critical towards using it. Lack of trust in AI was also reflected in one of the interviewees’ statements: “ AI is not a living creature and depending on what you feed it with, it can learn different things, so it is important to be critical ” (I 2). The critical mindset towards AI was also expressed when the interviewees were asked about their trust in a potential diagnosis set by AI. Two female interviewees were directly against the possibility that AI could one day make diagnose by itself, while two others, a man and a woman, were concerned about trusting AI’s “opinion” and whether the AI dataset could be large enough to provide a precise diagnosis.
This study’s aim was to uncover patient perspectives on trust regarding the patient-GP relationship, data sharing and AI in general practice and the results uncovers many different perspectives viewpoints. The study provides some insights into how AI and data sharing feed into the patient-GP relationship, which the interviewees viewed as important. The interviewees generally had a high level of trust towards their GP, but one of the interviewees expressed concern regarding a general reduction in people’s trust in authorities, which correlates with previously mentioned study findings, that points toward a reduced professional status among doctors . Opposite to this perspective, the ten interviewees felt comfortable enough to share their health data with the GP for a research project regarding using AI in general practice. However, the interviewees had concerns especially in relation to outsiders’ access to their health data and lack of privacy and AI not being able to detect emotional states. Furthermore, the interviewees were afraid that using AI in general practice would change the patient-GP relationship, and they did not trust AI on its own, although they believed AI would be a beneficial support tool for GPs and general practice. The patient-GP relationship was viewed as very important for the interviewees and they expressed fear that AI could influence the trust between them and the GP stating that people cannot trust a machine and that people must be critical towards AI. As mentioned previously the patient-GP relationship is emphasized as important in several other studies . However, an Australian study considers that AI could be perceived as more ethical than a human GP, since it can be more effective, unbiased, and not prone to human fallibility. Furthermore, AI will not get tired after a long shift like a human GP will . Some of the mentioned perspectives were also considered by the interviewees, but they still trusted the GP much more than AI regarding general practice care. However, some users may be happy to take an algorithms claims based on trust, but it has been suggested that a trustworthy algorithm should be able to show how it is working to those who want to understand how it came to its conclusions and interested parties should be able to assess the reliability of such claims . The interviewees in this study expressed different concerns regarding data sharing, and a study consisting of focus groups with patients from 16 different countries showed similar concerns about data sharing, especially regarding insurance companies and employers gaining access to the data . Other studies from the United Kingdom revealed concerns associated with lack of privacy and fear of private companies using the data for profit . Insurance and private companies are not the only ones that patients are concerned with regarding data sharing. A study investigating the social, ethical, and legal issues that impact on genetic information and testing in employment in Europe, shows that there has also been anxiety among the public regarding employers using personal genetic data to discriminate employees who are at risk of a certain disease or condition . This all suggests a general fear regarding the misuse of data on multiple levels, which could be problematic if AI should be implemented and used in general practice, since the possibility for effective decision support through AI is linked to sharing large amounts of health data . In this study, the interviewees considered AI to be a beneficial support tool that could increase the diagnostic speed. Half of the interviewees even expressed that they would feel safer if the GP used AI as a support tool when diagnosing. The interviewees’ positive attitudes towards using AI as a support tool correlates with the increasing use of new medical technologies that, as previously argued, can possibly make GPs more dependent on technology and the prospect of GPs being more dependent on technologies could lead to reduced professional status . However, the interviewees still wanted the GPs to be the primary decision makers. Studies investigating America’s, the United Kingdom’s and the Dutch’s population views on AI in healthcare settings show similar results. Here, the patients believed that AI should be used as a support tool rather than a primary decision maker, which suggests that people are not interested in pursuing decision making pathways without a human involved . The consensus that AI should not be the primary decision maker suggests that medical technologies are not perceived as being able to take the GPs’ place. Some of the interviewees in this study were sure that with time they would trust AI, especially if AI proved to be correct time and time again and received more exposure in society. The interviewees reasoned that people would get used to AI as they did with iPhones and tablets over a decade ago. This opens the possibility that if AI is implemented in general practice in a way that preserves the patient-GP relationship and is used primarily as a support tool for the GP, patients may trust AI eventually. Strengths and limitations This study investigates patient perspectives on data sharing regarding implementing and using AI in general practice, which opens an important perspective that has received little attention until now. Therefore, this study could serve as a steppingstone for the creation of guidelines on how to implement and use AI in general practice in a patient-friendly way. The interviewees did not have prior experience with or much knowledge on data sharing regarding the use of AI in general practice or AI in general, and therefore the vignette method was considered a good dialogue starter. However, the interviewees’ limited knowledge on the topics meant that many of the interviewees changed their mind or expressed different opinions on the same topic during the interviews. This may partly be explained by the fact that the interviewees were introduced to different examples and scenarios in the vignettes they read during the interviews. Using vignettes in this way can lead to the interviewees getting information overload . To prevent information overload, the vignettes only used short sentences. Methodical we found three aspects to discuss. The first one being that the study was conducted in Denmark among Danish patients and therefore the results are probably most relevant and transferable to countries with similar healthcare systems. However, countries with similar healthcare systems can be different regarding access and availability when it comes to patient’s possibilities for making appointments with a GP. The GPs’ role and status as primary physician and therefore a stable and relatable figure is widespread, and GPs are universally highly educated. However, in countries that face problems regarding access and availability to GPs, it is much harder for the patients to develop trust towards their GP, since the continuity and longevity of contact with the GP is a big part of creating the patient-GP relationship and the trust between them . Second, a greater number of patient perspectives may have identified additional insights. However, the 10 interviews we carried out were in depth, and offered the patients a chance to truly consider the different scenarios in the vignettes, which enriched the analysis with different and carefully considered perspectives. One limitation that must be pointed out, however, is that age span of the female interviewees was rather narrow, as the oldest female interviewee was only 46 years of age, it could have provided further insights if we had been able to include an older female, since older people have higher tendency of having complex multimorbidity . Unfortunately, we were not able to recruit an older female in the time period of the study. Thirdly, it is a limitation that most of the interviews were performed by the second author and the analysis was made primarily by the first author, where some thoughts and ideas may have been lost in the process. Conversely, the two authors working together on the transcriptions, analysis and presentation of the results led to thorough discussion on the themes and topics that emerged, which may have led to further insight and enhanced trustworthiness. This study is a step towards expanding our knowledge on patient perspectives on data sharing specifically regarding using AI in general practice, which is currently a very limited field. The perspectives shown in this pilot study can be used as focal points for future research on implementing and using AI in general practice and the mentioned concerns need to be taken seriously if patients are to trust the use of AI in general practice. Future research could advantageously carry out focus group interviews mixed with patients and GPs, so both patients and GPs could get a deeper understanding of each other’s viewpoints and in that way cooperate in finding ways of implementing and using AI in general practice that works for both parties. Furthermore, a larger study of GPs and patients in general practice could tease out further concerns identified in this study.
This study investigates patient perspectives on data sharing regarding implementing and using AI in general practice, which opens an important perspective that has received little attention until now. Therefore, this study could serve as a steppingstone for the creation of guidelines on how to implement and use AI in general practice in a patient-friendly way. The interviewees did not have prior experience with or much knowledge on data sharing regarding the use of AI in general practice or AI in general, and therefore the vignette method was considered a good dialogue starter. However, the interviewees’ limited knowledge on the topics meant that many of the interviewees changed their mind or expressed different opinions on the same topic during the interviews. This may partly be explained by the fact that the interviewees were introduced to different examples and scenarios in the vignettes they read during the interviews. Using vignettes in this way can lead to the interviewees getting information overload . To prevent information overload, the vignettes only used short sentences. Methodical we found three aspects to discuss. The first one being that the study was conducted in Denmark among Danish patients and therefore the results are probably most relevant and transferable to countries with similar healthcare systems. However, countries with similar healthcare systems can be different regarding access and availability when it comes to patient’s possibilities for making appointments with a GP. The GPs’ role and status as primary physician and therefore a stable and relatable figure is widespread, and GPs are universally highly educated. However, in countries that face problems regarding access and availability to GPs, it is much harder for the patients to develop trust towards their GP, since the continuity and longevity of contact with the GP is a big part of creating the patient-GP relationship and the trust between them . Second, a greater number of patient perspectives may have identified additional insights. However, the 10 interviews we carried out were in depth, and offered the patients a chance to truly consider the different scenarios in the vignettes, which enriched the analysis with different and carefully considered perspectives. One limitation that must be pointed out, however, is that age span of the female interviewees was rather narrow, as the oldest female interviewee was only 46 years of age, it could have provided further insights if we had been able to include an older female, since older people have higher tendency of having complex multimorbidity . Unfortunately, we were not able to recruit an older female in the time period of the study. Thirdly, it is a limitation that most of the interviews were performed by the second author and the analysis was made primarily by the first author, where some thoughts and ideas may have been lost in the process. Conversely, the two authors working together on the transcriptions, analysis and presentation of the results led to thorough discussion on the themes and topics that emerged, which may have led to further insight and enhanced trustworthiness. This study is a step towards expanding our knowledge on patient perspectives on data sharing specifically regarding using AI in general practice, which is currently a very limited field. The perspectives shown in this pilot study can be used as focal points for future research on implementing and using AI in general practice and the mentioned concerns need to be taken seriously if patients are to trust the use of AI in general practice. Future research could advantageously carry out focus group interviews mixed with patients and GPs, so both patients and GPs could get a deeper understanding of each other’s viewpoints and in that way cooperate in finding ways of implementing and using AI in general practice that works for both parties. Furthermore, a larger study of GPs and patients in general practice could tease out further concerns identified in this study.
The interviewees agreed that they would share their health data with their GP and that a research project on GPs using AI in general practice was a sufficient cause for which they would share their health data. All the interviewees found that AI could be an efficient and beneficial support tool for the GP in general practice. It was pointed out that patients cannot have a trustful relationship with a machine, which was mostly how the interviewees viewed AI, and therefore the interviewees insisted that the GP should continue playing a primary role in the decision making of the general practice workflow. It can therefore be argued that if AI is implemented in a way that preserves the patient-GP relationship and is used primarily as a support tool for the GP, patients may trust AI in general practice in the future.
Below is the link to the electronic supplementary material Supplementary Material 1 Supplementary Material 2
|
People at heightened risk of deterioration from COVID-19: living ‘with’ or living ‘despite’ public health communications? | 84399a7d-bf42-4d7a-8b33-349622a381c4 | 10071764 | Health Communication[mh] | The coronavirus pandemic has produced the most serious global public health emergency in a generation. Since the pandemic was declared in early 2020, governments worldwide have enacted and adjusted a range of public health measures to control transmission, protect health services, and lower morbidity and mortality at a population level. Against a backdrop of widespread community vaccination, Australia has recently shifted from the acute phase of the emergency towards a longer-term ‘living with COVID-19’ phase. Public health messaging now emphasises the role of informed personal choice with respect to COVID-19 exposure. Yet within this ‘living with COVID-19’ messaging, the needs of those most at risk of deterioration from the disease are almost entirely absent. There is neither recognition of the diverse range of people at heightened risk within the community nor acknowledgement of their need for tailored, evidence-based information and supports to enable them to manage their risk against a background of high community transmission. In this commentary, we explore this issue of informed decision-making related to COVID-19 and argue that there is urgent need to address this gap so that vulnerable groups can meaningfully apply public health advice to make informed decisions and to live safely and well with COVID-19. Early preventative public health responses focused on implementing physical distancing measures to counteract COVID-19 transmission in the absence of effective treatments or vaccines. , , The success of such measures in counteracting COVID-19 depended fundamentally on whole populations taking up and adhering to required measures (e.g., restricted gatherings, isolation, mask wearing). At different timepoints, variably across Australian states, such measures have been enforced or mandated by government. In the most effective cases, the required behavioural changes were supported by both extensive public health communications to convey the importance of adhering to public health measures, and concrete, actionable messaging about what people needed to do to keep themselves safe. The requirements of communicating complex public health advice in emergency situations are substantial and well-documented. Information must be clear and up-to-date, from trusted sources, accessible, consistent, and reach all members of the community equally (regardless of preferred language or access to different forms of media) to enable people to understand what is required and to act. , , , , , Many groups within the community are less likely to be reached by public health messaging or less able to take up certain measures. This includes people with lower health literacy levels; those who are socially, medically, or economically disadvantaged; those from culturally diverse backgrounds who may or may not prefer to communicate in a language other than English; and those from vulnerable groups (such as people with disability, mental health conditions, experiencing homelessness). , , , , , , , , , , A lack of targeted communication for some of these groups results in much higher risks of being influenced by circulating misinformation and disinformation and in failure to adhere to recommended preventive actions. , Successive pandemic waves, and population-wide vaccination rollout, have required governments and health authorities to proactively communicate and adjust messaging over many months, , sometimes on a daily basis, to reflect changes to the situation within Australia. There is now a large body of evidence to inform how this can be done well, and although it is arguable how well authorities within Australia and internationally have followed good practices, there are generally now high levels of knowledge across most populations about COVID-19 disease, symptoms and risk mitigation actions. Widespread vaccination has reduced the risk of high mortality rates and overburdened hospitals, resulting in public health advice moving broadly from population-wide preventative measures to a more specific model of personal decision-making. The availability of vaccination and booster shots, and more recently, antivirals, has led many people to feel relatively less alarmed about contracting COVID-19 despite high community levels of circulating virus and an emerging understanding of the likely prevalence and severity of long-COVID. Over time, lowered perceived risk coupled with growing pandemic fatigue has led many to a degree of complacency about contracting COVID-19 and the need to adhere to preventive risk mitigation measures. , To a large extent, this has enabled life to return to a more normal state for many, reflected by reopening of schools and workplaces and re-engagement with hospitality, entertainment and travel. This return to near-normal activity has also led to soaring rates of community transmission, particularly as new COVID-19 variants have emerged. Fortunately, high vaccination rates have prevented an equivalent rise in numbers hospitalised or of people dying from the virus. However, public health measures and accompanying messaging that characterised the earlier pandemic response have now largely been abandoned in favour of a focus on vaccination to mitigate the impact of COVID-19. Within the mix of public health messaging over the pandemic, there has long been recognition of the need to protect the health of those at heightened risk of severe disease or death. , , , , , This includes older people, those with chronic disease or who are immunocompromised (collectively termed ‘vulnerable’), the focus of this commentary. For such people, living with COVID-19 continues to be uncertain as there is a need to continue to enact protective measures even while the rest of the community returns to COVID-normal life. , , , The recent relaxation of public health measures both in Australia and overseas, while COVID-19 variants continue to circulate widely, has made this tension even more manifest during 2022. There is now growing disparity between those able to participate in a more normal life with high circulating COVID-19 levels and those who must navigate their own elevated risk against this backdrop. In this situation, people at heightened risk must take on personal responsibility for assessing and negotiating everyday activities to minimise their own risk of contracting disease. , , This decision-making can only happen in a meaningful way when individuals can access accurate, evidence-based information about what they can, and cannot, safely do. , , Despite clear guidance from peak bodies globally, , , , , and some examples of successfully tailored materials within Aboriginal and CALD communities within Australia, , , information to more broadly support vulnerable groups to live well through the pandemic is largely absent. How then can vulnerable people truly live with COVID-19, rather than live in isolation from the rest of the community, until the pandemic ends? The number of people at heightened risk of COVID-19 within Australia is not small. Estimates suggest that almost half of Australians have at least one chronic condition, and almost 20% have two or more. Not all of these people will be at heightened risk from COVID-19, but even if only a fraction are then these issues directly concern a substantial number of Australians. The pandemic is now in its fourth year, with little sign of abating, and clear public health advice to support those in vulnerable groups is urgently needed. Without this, people at heightened risk must continue to navigate life in parallel, relatively isolated from the rest of the community, with the now well-documented negative outcomes that this brings. , , Additionally, long-standing healthcare access issues are exacerbated by new risks posed by COVID-19, further widening inequality between the well and the vulnerable. , , Requiring vulnerable people to isolate indefinitely into the future, without access to accurate information to inform decisions about risk can only be viewed as a failure of public health policy. We contend that there are several clear ways this tension can be reduced. All rest on providing people at heightened COVID-19 risk with clear, accessible, tailored information and public health advice to support informed choices about their own risk in relation to COVID-19. There is also a need to recognise the continually changing context in which individuals are managing COVID-19 and that tailored advice must be timely and up-to-date. Additional supports for decision-making, and practical supports tailored to need, , , are also part of addressing this major gap in public health policy and practice. First, there should be clear and explicit recognition of the range of issues that exist in daily life for people who are managing chronic diseases or are at otherwise heightened risk against the backdrop of COVID-19. Recognition of this is important as this group cannot be considered (clinically or from a public health perspective) to be at the same risk as the rest of the community. Neither can they be considered a homogeneous group. The decisions required of an older person at heightened risk are, for instance, not the same as those of a working mother with school-aged children or a university student undertaking study and work. Public health messaging about those at heightened risk typically implies that such people are older, frailer and living quiet lives in relative isolation irrespective of the pandemic. This may stem from the identification of older adults as a particular at-risk group for COVID-19, , but there now needs to be clear recognition that the group of people at heightened risk encompasses a far more diverse range of people across the lifespan and across life circumstances. , , , , Public health messaging and information then needs to consider this diverse group as a specific audience and recipient of actionable messages. Tailored, evidence-based information and solutions must be provided to this audience, with recognition that this information will need to be continually updated. , , , , , Developing such tailored advice and support requires a range of activities to occur. 1. Engaging with people at heightened risk, and the organisations that service them, to determine directly what their main concerns or challenges are in living with COVID-19 and to learn how they are managing risk (devising and sharing solutions) in an environment of increased COVID-19 transmission. , , , 2. Collecting and making sense of better individualised risk data to inform communications and public health messaging. Such communications must be tailored to need and pre-tested with users drawn from diverse community groups. 3. Advising about public health measures to be continued (e.g., mask wearing and physical distancing in high(er) risk settings) and communicating with the entire community so that everyone understands why this is necessary to protect those at heightened risk. Effective communication needs to: o present clear, timely, consistent messages from trusted sources o be framed with awareness that messages will need continual updating for new evidence o be tailored to ensure accessibility and reach throughout the community and to different groups at heightened risk o provide actionable advice so that people can act to protect their health and that of others , , , , , , 4. Acknowledging where information is not yet available. Clearly and transparently communicating about uncertainties is key to maintaining community trust in public health measures. , 5. Developing decision support tools to support common decisions. The need for accurate, timely, up-to-date information to support public health measures to protect health has been highlighted since the very earliest days of the pandemic, as has the critical role for community engagement and the need for practical supports. , , , , Vulnerable groups, including those at heightened risk of deteriorating due to COVID-19, are disproportionately disadvantaged by the pandemic. It is therefore not surprising that consequent negative social, economic, health and other impacts have emerged and are notable. , , , Despite massive growth in research related to the pandemic and to public health controls across the world, very little squarely addresses the needs of those at heightened risk. Within Australia as we work to control and contain COVID-19 over the remaining course of the pandemic, we cannot consider to have successfully lived with COVID-19 while those at heightened risk remain on the isolated fringes of society. In the face of widespread apathy, there is an urgent need for leadership and continued public health action to reduce the impacts of the pandemic. Directly recognising and addressing these issues for those at heightened COVID-19 risk is a major, pressing priority and represents a huge unmet need within the Australian community. Rebecca Ryan receives support for her position as Coordinating Editor of the Cochrane Consumers and Communication Group under the NHMRC funding to Australian Cochrane Groups (2020-2023). The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. |
Highly efficient bioconversion of icariin to icaritin by whole-cell catalysis | 99c36135-4b64-44bf-bd30-ef0bdafacc7e | 10071772 | Microbiology[mh] | Herba Epimedii (Yinyanghuo in Chinese), the traditional Chinese medicine, has been used in China for more than 2,000 years . It has proven anti-osteoporotics , anti-cancer , anti-inflammatory activities and reproductive functions . The main active substances in Herba Epimedii are flavonoid glycosides, including icariside I, baohuoside I, epimedin A, B, C and icariin . They are formed by varying degrees of glycosylation of the C-3 and C-7 positions of icaritin (Fig. ) . Icaritin is an aglycone of flavonoids without any glycosylation. It has higher bioactivity than the other flavonoid glycosides in the treatment of some diseases . In addition, it enhanced stem cell proliferation, migration and osteogenic differentiation and had a powerful effect in treatment of multiple cancers . Clinical trials of icaritin indicated that it has good performance in treatment of hepatocellular carcinoma . However, the natural icaritin content of Herba Epimedii is very low. The icariin is one of considerable components in extracts of Herba Epimedii . Icariin has rhamnosidic linkage and β-glucosidic linkage at the C-3 and C-7 positions, respectively. Thus, α-L-rhamnosidases and β-glucosidases have been widely applied in icaritin production. However, the reaction conditions of α-L-rhamnosidases and β-glucosidases are incompatibility, and thus a two-step process is usually used [ – ], which restricts the production efficiency of icaritin. In addition, the icariin is difficult to dissolve in water, which affected the efficiency of enzymatic catalysts. Therefore, exploring one-step enzymatic catalysts and hydrolyzing high concentration of substrate have the potential to improve icaritin production. Immobilization of α-L-rhamnosidases and β-glucosidases has been reported in hydrolysis of Epimedii flavonoids. Immobilization of the thermostable enzyme 1000NH-DthRha and 1000NH-DthBgl3 was reported for hydrolysis of Epimedii flavonoids to icaritin , the molar conversion rate was 87.21%, but the substrate concentration was only 10 g/L. Recently, co-immobilization of α-L-rhamnosidase (Rha1) and β-glucosidase (Glu4) using cross-linked enzyme aggregates was reported to achieve one-pot production of icaritin by hydrolyzing a high concentration of epimedin C. However, the icaritin yield was only 77.45% at an epimedin C concentration of 100 g/L . Although immobilized enzymes have higher stability and they can be reused, the operation processes are very complicated, which is a significant drawback for industrial application. Thus, further exploring efficient method is still important for icaritin production. Whole-cell catalysis omits isolation and purification of enzymes and reduces costs. The cells can provide a natural, protective environment for the enzymes . However, few reports focus on hydrolysis of flavonoid glycosides by whole-cell catalysis. A strain of Aspergillus niger with α-L-rhamnosidase and β-glucosidase activities has been reported, it efficiently hydrolyzed flavonoid glycosides by whole-cell catalysis, including hesperidin, rutin, naringin, neohesperidin and naringin dihydrochalcone . Whole-cell catalysis using resting cells to completely hydrolyze epimedin C to icariin following expression of α-L-rhamnosidase in E. coli has also been reported . However, to our knowledge, whole-cell catalysis icariin into icaritin has not yet been reported. In this work, α-L-rhamnosidase SPRHA2 from Novosphingobium sp. GX9 and β-glucosidase PBGL from Paenibacillus cookii GX-4 were cloned and identified to hydrolyze icariin to produce icaritin in one-pot enzymatic method. Moreover, whole-cell catalysis for icariin hydrolysis was investigated. The results showed whole-cell catalytic system had better performance in hydrolysis of high purity icariin and crude icariin extracts. This study have potential for use in industrial production of icaritin.
Purification and basic analysis of SPRHA2 and PBGL Icariin has rhamnosidic linkage and β-glucosidic linkage at the C-3 and C-7 positions, respectively. Icaritin can thus be produced by hydrolysis of icariin by α-L-rhamnosidases and β-glucosidases. The hydrolysis of icariin was investigated by α-L-rhamnosidase SPRHA2 and β-glucosidase PBGL. The ORF of sprha2 was 3,498 bp. Analysis results from SMART revealed that SPRHA2 contains a glycoside hydrolase (GH) domain that belongs to GH family 106 and a GH 2 N domain (sugar binding domain). Prediction using SignalP-5.0 indicated that amino acids 1 to 31 of SPRHA2 form a signal peptide. The sequence coding signal peptide was removed by PCR for recombinant intracellular expression in E. coli . The ORF of pbgl was 2,280 bp. PBGL contains two GH family 3 domains (residues 107 to 465, and 504 to 744) and did not have a predicted signal peptide. The purified recombinant SPRHA2 and PBGL were analyzed by SDS-PAGE. The molecular weights of SPRHA2 and PBGL were ~ 120 kDa and ~ 84 kDa (Fig. ), respectively. K m and V max values of SPRHA2 and PBGL were determined using the substrates p NPR and p NPG, respectively: SPRHA2, K m : 0.63 ± 0.02 mM, V max : 267.60 ± 2.86 µmol min -1 mg -1 ; PBGL, K m : 0.17 ± 0.02 mM, V max : 149.00 ± 3.40 µmol min -1 mg -1 . Characterization of the enzymes in hydrolysis of flavonoid glycosides In this study, co-hydrolyzed icariin by α-L-rhamnosidase SPRHA2 and β-glucosidase PBGL to produce icaritin was assayed. First, the enzymes were applied individually to catalyze icariin. HPLC analysis results showed that the α-L-rhamnosidase SPRHA2 catalyzed icariin into icariside I (Fig. ). Surprisingly, β-glucosidase PBGL converted icariin into baohuoside I and icaritin, exhibiting α-L-rhamnosidase activity, but icariside I was not found. Next, hydrolysis of the intermediates icariside I and baohuoside I was studied to further determine the properties of SPRHA2 and PBGL. The results suggested that SPRHA2 catalyzed conversion of both icariside I and baohuoside I to icaritin, but it showed low activity toward icariside I. PBGL could catalyze conversion of icariside I and baohuoside I to icaritin (Additional file 1: Figure ). Thus, these data indicate that SPRHA2 and PBGL are bifunctional enzyme. Either SPRHA2 or PBGL has the ability to hydrolyze icariin to icaritin. In a previous report, an Epimedii flavonoid-glycosidase from Aspergillus sp. y848 had the same effect in hydrolysis of icariin as PBGL . The effect of combination of SPRHA2 and PBGL on the hydrolysis of icariin to produce icaritin was better than that of SPRHA2 or PBGL alone. Detailed transformation pathways are shown in Fig. . The hydrolysis of epimedin A, B and C by SPRHA2 and PBGL was also studied. The results showed that epimedin A was converted into sagittatoside A, a small amount of baohuoside I and icaritin by PBGL. PBGL could cleave the β-glucosidic linkages of epimedin B and C, to respectively form sagittatoside B and C, but could not break the rhamnosidic linkages at the C-3 position (Additional file 1: Figure ). Epimedin A and B were respectively converted into sagittatoside A and B by SPRHA2, but the activity was lower than that of PBGL. These data confirmed that SPRHA2 could hydrolyze the β-glucosidic bond at position C-7. In addition, SPRHA2 catalyzed epimedin C into sagittatoside C, icarisde I and icaritin, but icariin was not detected. To determine if icariin was produced in the reaction, the time course of epimedin C hydrolysis by SPRHA2 was studied. The results confirmed that icariin was not a product (Additional file 1: Figure ). We speculate that SPRHA2 simultaneously hydrolyzed the β-glucosidic linkage at C-7 and two α-L-rhamnosidic linkages at C-3 to produce sagittatoside C and icariside I, respectively. Assay of icariin co-hydrolysis by SPRHA2 and PBGL The influence of temperature and pH on co-hydrolysis of icariin by SPRHA2 combined with PBGL was investigated. The optimal temperature for the reaction was 55 °C (Fig. a), and the highest activity was observed in 200 mM borate saline buffer at pH 8.5 (Fig. b). This is the first report of hydrolysis of icariin in alkaline environment. It was worth mentioning that compared with one-pot catalytic method, the temperature and pH of the reaction after the first step should be adjusted to satisfy the needs of the second step in two-step catalytic methods; this increases energy consumption and processing costs. Here, we developed a one-pot process that did not require adjustment of the conditions during the reaction, which is advantageous for icaritin production. To increase the production efficiency of icaritin and lower the amount of residual intermediates, the optimum weight ratio of SPRHA2 and PBGL was assessed. Reactions were performed with 0.2 g/L icariin and different weight ratios of the enzymes (total protein concentration: 10 mg/L) at 55 °C, pH 8.5 for 10 min. The results showed that optimum weight ratio of SPRHA2 to PBGL was 4:6 (w:w) by HPLC analysis (Fig. c). Finally, in the optimum reaction conditions determined in this study, high concentrations of icariin were co-hydrolyzed by SPRHA2 and PBGL. Icariin is difficult to dissolve in water, and solubility of its hydrolysis products is even lower, which hinders the effective hydrolysis of icariin. Therefore, the reactions were shaken to improve the efficiency of icariin hydrolysis. The different concentrations of icariin were catalyzed by 40 mg/L SPRHA2 and 60 mg/L PBGL at 55 °C and pH 8.5 with shaking at 220 rpm for 5 h. The conversion rate of icariin decreased with the increase of substrate concentration (Fig. d). After 5 h, the icaritin yield was 72.43% when the substrate icariin concentration was 25 g/L; at higher substrate concentrations, the yields were below 50% (Fig. e). Thus, the co-hydrolysis system of icariin by purified SPRHA2 and PBGL does not have good performance for icaritin production. Whole-cell catalysis for icaritin production by co-hydrolysis To increase the production efficiency of icaritin and reduce costs, the whole-cell catalysis was examined in this study. The reaction properties were studied on co-hydrolysis of icariin by mixtures of E. coli cells harboring plasmids pET- sprha2 and pET- pbgl respectively. The optimal temperature and pH of whole-cell catalysis were determined to be 55 °C and 9.0 (Fig. a and b), respectively. The optimum pH for the whole-cell catalysis was thus more alkaline than that for the purified enzymes, and it exhibited high activity even at pH 9.5. The optimal temperature was not changed compared with that for the purified enzymes. The optimal weight ratio of wet cells containing SPRHA2 and PBGL, respectively, was investigated. The reaction system (200 µL) contained 1 g/L icariin and different ratios of wet cells (Total wet cell weight: 5 g/L); it was incubated at 55 °C, pH 9.0 with shaking at 220 rpm for 10 min. The results showed that optimal weight ratio was 1:1 (Fig. c). The time course of whole-cell hydrolysis of icariin under optimal conditions was studied to analyze catalytic process. Icariin was completely converted into icaritin in 2 h. Baohuoside I was the only intermediate. Icariside I was not detected (Additional file 1: Figure ). Due to icariin is difficult to dissolve in water, the enzymatic reaction at high concentration of substrate may be insufficient. Thus we tested multiple concentrations of substrate icariin in whole-cell catalysis. When reactions contained 20 g/L wet weight SPRHA2 cells and 20 g/L wet weight PBGL cells (incubated at 55 °C, pH 9.0 with shaking at 220 rpm for 5 h), 50 g/L icariin was completely hydrolyzed within 1 h, and the conversion rates of the icariin (100–200 g/L) were more than 90% by this time (Fig. d). The 200 g/L icariin was completely hydrolyzed after 4 h, producing 103.69 g/L icaritin with a yield of 95.23% (Fig. e). The product icaritin was confirmed by NMR analysis (Additional file 1: Table ). High-purity substrates are more expensive as raw materials, therefore crude icariin extracts were selected (purity ~ 10%, ~ 20%, ~ 50%, and ~ 70%) and we assessed their hydrolysis in whole-cell catalysis. The reaction system (500 µL) contained 200 g/L icariin crude extract, 20 g/L wet weight SPRHA2 cells and 20 g/L wet weight PBGL cells, and was incubated at 55 °C, pH 9.0 with shaking at 220 rpm for 4 h. The results indicated that all conversion rates of icariin were more than 98% (Fig. a), 4.27, 8.37, 36.96 and 49.31 g/L icaritin was respectively obtained by hydrolysis of crude icariin extracts (Fig. b). However, compared with hydrolysis of 98% pure icariin, the conversion rate of icariin was slightly lower. We speculate that crude icariin extracts contain some unknown impurities that hampered the reactions in whole-cell catalysis. Whole-cell catalyzed conversion of crude icariin extracts into icaritin by co-expression Whole-cell catalysts can execute multiple reactions by assembling enzymes . Next, we attempted to co-express SPRHA2 and PBGL in E. coli to further simplify operations. The plasmids pET- sprha2 - pbgl and pET- pbgl - sprha2 were constructed. The genes sprha2 and pbgl shared one T7 promoter and had separate ribosome-binding sites. The reaction system (500 µL) contained 200 g/L icariin crude extract and 40 g/L wet weight cells of the co-expression strain; the reaction mixture was incubated at 55 °C, pH 9.0 with shaking at 220 rpm for 4 h. The results showed that conversion rate of icariin by the strain harboring plasmid pET- pbgl - sprha2 was more than 99%, which was higher than for the strain harboring plasmid pET- sprha2 - pbgl (Fig. a), suggesting that co-expression plasmid pET- pbgl - sprha2 was more suitable for icariin hydrolysis than plasmid pET- sprha2 - pbgl . The protein expression levels of the two plasmid constructions were analyzed by SDS-PAGE. The results showed that the PBGL expression level in the pET- pbgl - sprha2 was significant higher than that in the pET- sprha2 - pbgl . The total protein content of these two enzymes was also higher under the plasmid pET- pbgl - sprha2 condition (Additional file 1: Fig. ). The product icaritin concentrations from strain harboring plasmid pET- pbgl - sprha2 were 4.16, 7.35, 31.99 and 42.22 g/L from crude icariin extracts, respectively (Fig. b), lower than when SPRHA2 and PBGL were expressed in separate strains. HPLC analysis showed that more residual intermediates were produced in the strains co-expressing both SPRHA2 and PBGL than expression of individual genes. The strain with plasmid pET- pbgl - sprha2 had good performance in hydrolysis of crude icariin extracts, its icaritin yield was more than 85% of the yield of mixture cells on hydrolysis of ~ 50% and ~ 70% pure crude icariin extracts. Expression of multiple genes may cause a greater burdens on E. coli than expression of individual genes , which may explain why the strains co-expressing both PBGL and SPRHA2 were not as effective as was mixing cells of the strains individually expressing these enzymes.
Icariin has rhamnosidic linkage and β-glucosidic linkage at the C-3 and C-7 positions, respectively. Icaritin can thus be produced by hydrolysis of icariin by α-L-rhamnosidases and β-glucosidases. The hydrolysis of icariin was investigated by α-L-rhamnosidase SPRHA2 and β-glucosidase PBGL. The ORF of sprha2 was 3,498 bp. Analysis results from SMART revealed that SPRHA2 contains a glycoside hydrolase (GH) domain that belongs to GH family 106 and a GH 2 N domain (sugar binding domain). Prediction using SignalP-5.0 indicated that amino acids 1 to 31 of SPRHA2 form a signal peptide. The sequence coding signal peptide was removed by PCR for recombinant intracellular expression in E. coli . The ORF of pbgl was 2,280 bp. PBGL contains two GH family 3 domains (residues 107 to 465, and 504 to 744) and did not have a predicted signal peptide. The purified recombinant SPRHA2 and PBGL were analyzed by SDS-PAGE. The molecular weights of SPRHA2 and PBGL were ~ 120 kDa and ~ 84 kDa (Fig. ), respectively. K m and V max values of SPRHA2 and PBGL were determined using the substrates p NPR and p NPG, respectively: SPRHA2, K m : 0.63 ± 0.02 mM, V max : 267.60 ± 2.86 µmol min -1 mg -1 ; PBGL, K m : 0.17 ± 0.02 mM, V max : 149.00 ± 3.40 µmol min -1 mg -1 .
In this study, co-hydrolyzed icariin by α-L-rhamnosidase SPRHA2 and β-glucosidase PBGL to produce icaritin was assayed. First, the enzymes were applied individually to catalyze icariin. HPLC analysis results showed that the α-L-rhamnosidase SPRHA2 catalyzed icariin into icariside I (Fig. ). Surprisingly, β-glucosidase PBGL converted icariin into baohuoside I and icaritin, exhibiting α-L-rhamnosidase activity, but icariside I was not found. Next, hydrolysis of the intermediates icariside I and baohuoside I was studied to further determine the properties of SPRHA2 and PBGL. The results suggested that SPRHA2 catalyzed conversion of both icariside I and baohuoside I to icaritin, but it showed low activity toward icariside I. PBGL could catalyze conversion of icariside I and baohuoside I to icaritin (Additional file 1: Figure ). Thus, these data indicate that SPRHA2 and PBGL are bifunctional enzyme. Either SPRHA2 or PBGL has the ability to hydrolyze icariin to icaritin. In a previous report, an Epimedii flavonoid-glycosidase from Aspergillus sp. y848 had the same effect in hydrolysis of icariin as PBGL . The effect of combination of SPRHA2 and PBGL on the hydrolysis of icariin to produce icaritin was better than that of SPRHA2 or PBGL alone. Detailed transformation pathways are shown in Fig. . The hydrolysis of epimedin A, B and C by SPRHA2 and PBGL was also studied. The results showed that epimedin A was converted into sagittatoside A, a small amount of baohuoside I and icaritin by PBGL. PBGL could cleave the β-glucosidic linkages of epimedin B and C, to respectively form sagittatoside B and C, but could not break the rhamnosidic linkages at the C-3 position (Additional file 1: Figure ). Epimedin A and B were respectively converted into sagittatoside A and B by SPRHA2, but the activity was lower than that of PBGL. These data confirmed that SPRHA2 could hydrolyze the β-glucosidic bond at position C-7. In addition, SPRHA2 catalyzed epimedin C into sagittatoside C, icarisde I and icaritin, but icariin was not detected. To determine if icariin was produced in the reaction, the time course of epimedin C hydrolysis by SPRHA2 was studied. The results confirmed that icariin was not a product (Additional file 1: Figure ). We speculate that SPRHA2 simultaneously hydrolyzed the β-glucosidic linkage at C-7 and two α-L-rhamnosidic linkages at C-3 to produce sagittatoside C and icariside I, respectively.
The influence of temperature and pH on co-hydrolysis of icariin by SPRHA2 combined with PBGL was investigated. The optimal temperature for the reaction was 55 °C (Fig. a), and the highest activity was observed in 200 mM borate saline buffer at pH 8.5 (Fig. b). This is the first report of hydrolysis of icariin in alkaline environment. It was worth mentioning that compared with one-pot catalytic method, the temperature and pH of the reaction after the first step should be adjusted to satisfy the needs of the second step in two-step catalytic methods; this increases energy consumption and processing costs. Here, we developed a one-pot process that did not require adjustment of the conditions during the reaction, which is advantageous for icaritin production. To increase the production efficiency of icaritin and lower the amount of residual intermediates, the optimum weight ratio of SPRHA2 and PBGL was assessed. Reactions were performed with 0.2 g/L icariin and different weight ratios of the enzymes (total protein concentration: 10 mg/L) at 55 °C, pH 8.5 for 10 min. The results showed that optimum weight ratio of SPRHA2 to PBGL was 4:6 (w:w) by HPLC analysis (Fig. c). Finally, in the optimum reaction conditions determined in this study, high concentrations of icariin were co-hydrolyzed by SPRHA2 and PBGL. Icariin is difficult to dissolve in water, and solubility of its hydrolysis products is even lower, which hinders the effective hydrolysis of icariin. Therefore, the reactions were shaken to improve the efficiency of icariin hydrolysis. The different concentrations of icariin were catalyzed by 40 mg/L SPRHA2 and 60 mg/L PBGL at 55 °C and pH 8.5 with shaking at 220 rpm for 5 h. The conversion rate of icariin decreased with the increase of substrate concentration (Fig. d). After 5 h, the icaritin yield was 72.43% when the substrate icariin concentration was 25 g/L; at higher substrate concentrations, the yields were below 50% (Fig. e). Thus, the co-hydrolysis system of icariin by purified SPRHA2 and PBGL does not have good performance for icaritin production.
To increase the production efficiency of icaritin and reduce costs, the whole-cell catalysis was examined in this study. The reaction properties were studied on co-hydrolysis of icariin by mixtures of E. coli cells harboring plasmids pET- sprha2 and pET- pbgl respectively. The optimal temperature and pH of whole-cell catalysis were determined to be 55 °C and 9.0 (Fig. a and b), respectively. The optimum pH for the whole-cell catalysis was thus more alkaline than that for the purified enzymes, and it exhibited high activity even at pH 9.5. The optimal temperature was not changed compared with that for the purified enzymes. The optimal weight ratio of wet cells containing SPRHA2 and PBGL, respectively, was investigated. The reaction system (200 µL) contained 1 g/L icariin and different ratios of wet cells (Total wet cell weight: 5 g/L); it was incubated at 55 °C, pH 9.0 with shaking at 220 rpm for 10 min. The results showed that optimal weight ratio was 1:1 (Fig. c). The time course of whole-cell hydrolysis of icariin under optimal conditions was studied to analyze catalytic process. Icariin was completely converted into icaritin in 2 h. Baohuoside I was the only intermediate. Icariside I was not detected (Additional file 1: Figure ). Due to icariin is difficult to dissolve in water, the enzymatic reaction at high concentration of substrate may be insufficient. Thus we tested multiple concentrations of substrate icariin in whole-cell catalysis. When reactions contained 20 g/L wet weight SPRHA2 cells and 20 g/L wet weight PBGL cells (incubated at 55 °C, pH 9.0 with shaking at 220 rpm for 5 h), 50 g/L icariin was completely hydrolyzed within 1 h, and the conversion rates of the icariin (100–200 g/L) were more than 90% by this time (Fig. d). The 200 g/L icariin was completely hydrolyzed after 4 h, producing 103.69 g/L icaritin with a yield of 95.23% (Fig. e). The product icaritin was confirmed by NMR analysis (Additional file 1: Table ). High-purity substrates are more expensive as raw materials, therefore crude icariin extracts were selected (purity ~ 10%, ~ 20%, ~ 50%, and ~ 70%) and we assessed their hydrolysis in whole-cell catalysis. The reaction system (500 µL) contained 200 g/L icariin crude extract, 20 g/L wet weight SPRHA2 cells and 20 g/L wet weight PBGL cells, and was incubated at 55 °C, pH 9.0 with shaking at 220 rpm for 4 h. The results indicated that all conversion rates of icariin were more than 98% (Fig. a), 4.27, 8.37, 36.96 and 49.31 g/L icaritin was respectively obtained by hydrolysis of crude icariin extracts (Fig. b). However, compared with hydrolysis of 98% pure icariin, the conversion rate of icariin was slightly lower. We speculate that crude icariin extracts contain some unknown impurities that hampered the reactions in whole-cell catalysis.
Whole-cell catalysts can execute multiple reactions by assembling enzymes . Next, we attempted to co-express SPRHA2 and PBGL in E. coli to further simplify operations. The plasmids pET- sprha2 - pbgl and pET- pbgl - sprha2 were constructed. The genes sprha2 and pbgl shared one T7 promoter and had separate ribosome-binding sites. The reaction system (500 µL) contained 200 g/L icariin crude extract and 40 g/L wet weight cells of the co-expression strain; the reaction mixture was incubated at 55 °C, pH 9.0 with shaking at 220 rpm for 4 h. The results showed that conversion rate of icariin by the strain harboring plasmid pET- pbgl - sprha2 was more than 99%, which was higher than for the strain harboring plasmid pET- sprha2 - pbgl (Fig. a), suggesting that co-expression plasmid pET- pbgl - sprha2 was more suitable for icariin hydrolysis than plasmid pET- sprha2 - pbgl . The protein expression levels of the two plasmid constructions were analyzed by SDS-PAGE. The results showed that the PBGL expression level in the pET- pbgl - sprha2 was significant higher than that in the pET- sprha2 - pbgl . The total protein content of these two enzymes was also higher under the plasmid pET- pbgl - sprha2 condition (Additional file 1: Fig. ). The product icaritin concentrations from strain harboring plasmid pET- pbgl - sprha2 were 4.16, 7.35, 31.99 and 42.22 g/L from crude icariin extracts, respectively (Fig. b), lower than when SPRHA2 and PBGL were expressed in separate strains. HPLC analysis showed that more residual intermediates were produced in the strains co-expressing both SPRHA2 and PBGL than expression of individual genes. The strain with plasmid pET- pbgl - sprha2 had good performance in hydrolysis of crude icariin extracts, its icaritin yield was more than 85% of the yield of mixture cells on hydrolysis of ~ 50% and ~ 70% pure crude icariin extracts. Expression of multiple genes may cause a greater burdens on E. coli than expression of individual genes , which may explain why the strains co-expressing both PBGL and SPRHA2 were not as effective as was mixing cells of the strains individually expressing these enzymes.
To date, many studies have been performed the flavonoid glycosides hydrolysis by enzymatic catalysis for icaritin production. The flavonoid glycosides mainly include epimedin C and icariin. Therefore, α-L-rhamnosidases and β-glucosidases were widely applied in icaritin production. However, the reaction conditions of α-L-rhamnosidases and β-glucosidases are incompatibility, thus a two-step catalysis process is usually used [ , , ]. In previous report, the α-L-Rhamnosidase Rhase-I from Talaromyces stollii CLY-6 was expressed in P. pastoris , then Rhase-I and β-glucosidase Bglsk from Sanguibacter keddieii DSM 10,542 hydrolyzed epimedin C in two-step, 20 g/L of epimedin C was converted into 8.83 g/L of icaritin with a yield of 98.66% . To simplify operations and reduce costs, it is important to developed one-step catalysis. At present, the one-step catalysis was achieved by the bifunctional enzymes or immobilized enzymes. A glycosidase from Aspergillus sp.y848 has α-L-rhamnosidase and β-glucosidase activities, and it directly hydrolyzed icariin to icaritin, the 5.04 g icaritin with 98% purity was obtained from 10 g icariin in 18–20 h . However, the reaction time was so long. The α-L-rhamnosidase Rha1 and β-glucosidase Glu4 were co-immobilized by cross-linked enzyme aggregates, the 34.24 g/L icaritin was obtained from 100 g/L epimedin C in one-pot catalysis within 8 h . But the immobilization of enzymes has complicated operations. In this study, one-step catalysis of icariin was achieved by SPRHA2 and PBGL without any modification. In addition, PBGL could directly hydrolyze icariin to produce icaritin. But SPRHA2 and PBGL co-hydrolyzed icariin with better performance than PBGL alone. Next, the reaction conditions of icariin co-hydrolysis were optimized. The 9.85 g/L icaritin was produced from 25 g/L icariin in one-step in 5 h, and the yield was 72.43%. However, this productivity of icaritin by co-hydrolysis of SPRHA2 and PBGL was not as high as two-step catalysis and immobilized enzymes hydrolysis (Table ). In addition to icaritin production by hydrolysis, engineered Saccharomyces cerevisiae and E. coli were built with complete icaritin biosynthesis pathways, 19.7 mg/L icaritin was obtained in co-culture . However, this productivity was lower than reported hydrolysis methods. To improve the production efficiency of icaritin and reduce costs, we explored the whole-cell catalysis for icariin hydrolysis. The combination of SPRHA2 cells and PBGL cells in icariin hydrolysis was investigated. The optimum pH for the whole-cell catalysis was 9.0, and it exhibited high activity even at pH 9.5. However, most reported enzymes hydrolyzed flavonoid glycosides with a pH below 6.0 [ , , ]. It was first report on hydrolysis of flavonoid glycosides in high alkaline. Moreover, a very high concentration of substrate was hydrolyzed. The 200 g/L icariin was completely co-hydrolyzed by SPRHA2 cells and PBGL cells in 4 h, producing 103.69 g/L icaritin with a yield of 95.23%. Obviously, with high concentrations of substrate conditions, whole-cell catalysis exhibited superior performance in icaritin production with the final concentration was 10.5 times higher than purified SPRHA2 and PBGL (Table ). The reason is that the enzymes may be more stable inside cells . In previous report, the highest concentration of substrate by hydrolysis was 150 g/L epimedin C. However, the yield of icaritin was below 80% . To our knowledge, this is the first report of flavonoid glycosides hydrolysis at such a high concentration (200 g/L) by whole-cell catalysis, and the highest concentration of icaritin produced. In order to further increase the production efficiency and simplify operations, we also constructed recombinant E. coli strains that co-expressed SPRHA2 and PBGL for whole-cell catalysis. And crude icariin extracts, as cheap raw materials, were also effectively hydrolyzed by the whole-cell catalytic system. Thus this study has great potential to meet the need of industrial icaritin production.
In this study, the bioconversion of icariin to icaritin using α-L-rhamnosidase SPRHA2 from Novosphingobium sp. GX9 and β-glucosidase PBGL from Paenibacillus cookii GX-4 was assayed. One-pot production of icaritin was achieved. Moreover, we developed a whole-cell catalytic method for icaritin production, which exhibited higher production efficiency. 200 g/L icariin was completely hydrolyzed by whole-cell catalysis in 4 h; icaritin yield was 95.23% (103.69 g/L). Recombinant strains were also constructed that co-expressed SPRHA2 and PBGL for whole-cell catalysis. Crude icariin extracts were efficiently transformed to icaritin in whole-cell catalysis. This study provides an efficient method for industrial icaritin production.
Materials, chemicals and strains Icariin (purity 98.0%) and baohuoside I (purity 98.0%) were purchased from Shanghai Aladdin Bio-chemical Technology Co., Ltd. (Shanghai, China). Icariside I (purity 98.0%), icaritin (purity 99.0%), epimedin A, B and C (purity 98.0%) were purchased from Shanghai Yuanye Bio-Technology Co., Ltd. (Shanghai, China). Crude icariin extracts (purity ~ 10%, ~ 20%, ~ 50% and ~ 70%) were purchased from Shanxi Sunrun Bio-tech. Co., Ltd. (Shanxi, China). All restriction enzymes, PrimeSTAR DNA polymerase, T4 DNA ligase and In-Fusion HD Cloning kit were purchased from Takara Co., Ltd. (Dalian, China). E. coli DH5α was used for plasmid propagation and construction. E. coli BL21 (DE3) was used for expression of heterologous genes. Plasmid construction The coding sequences of SPRHA2 (GenBank accession: QGA89207.1) and PBGL (GenBank accession: OP810493) were amplified by PCR from genomic DNA of Novosphingobium sp. GX9 and Paenibacillus cookii GX-4, respectively. The primers were as follows: sprha2-1: 5′-CAGGGATCCGAGCCGGCACCCGATGCGGCCGC-3′; sprha2-2: 5′-CAGAAGCTTTCAGCGGGTCGTGCCCAGCGTGAC-3′; pbgl-1: 5′-CGCGGATCCAGAAACCATACTTCAGACACGATCAA-3′; and pbgl-2: 5′-CGCAAGCTTTCAGCTTCTACGGTATTTCTTGGTTC-3′. The DNA fragments were digested by restriction enzymes Bam HI and Hin d III and ligated by T4 ligase into plasmid pET30a digested with the same restriction enzymes, resulting in plasmids pET- sprha2 and pET- pbgl , respectively. The recombinant plasmids were verified by DNA sequencing (Ruibo Tech, Guangzhou, China). The plasmids pET- sprha2 and pET- pbgl were respectively transformed into E. coli BL21 (DE3) to produce proteins. To construct co-expression plasmids, the DNA fragments of sprha2 and pbgl were inserted into plasmids pET- pbgl and pET- sprha2 respectively by In-fusion clone to produce plasmids pET- pbgl - sprha2 and pET- sprha2 - pbgl . The recombinant plasmids were verified by DNA sequencing. The plasmids pET- sprha2-pbgl and pET- pbgl-sprha2 were transformed into E. coli BL21 (DE3) for whole-cell catalysis. The primers were as follows: pet30-1: 5′-CCACTGAGATCCGGCTGCTAACAAAGCCCGAAAGG-3′; pet30-2: 5′-TGGTGGTGGTGCTCGAGTGCGGCCGCAAGCTT-3′; Inf-1: 5′-CTCGAGCACCACCACCAAAGGAGATATACATATGCACCATCA-3′; sprha2-3: 5′-CAGCCGGATCTCAGTGGTCAGCGGGTCGTGCCCAGCGTGAC-3′; and pbgl-3: 5′-CAGCCGGATCTCAGTGGTCAGCTTCTACGGTATTTCTTGGTTC-3′. Production of recombinant proteins For recombinant protein expression, a single colony was picked and inoculated into a flask containing 10 mL of Luria-Bertani (LB) medium for seed preparations, which included kanamycin (50 mg/L) and cultured overnight at 37 °C with constant shaking at 220 rpm. Then, 4 mL of seed culture was inoculated into 200 mL of LB medium containing kanamycin (50 mg/L) in a 500 mL flask, which was incubated at 37 °C with shaking at 220 rpm. When OD 600nm reached 0.4–0.6, isopropyl-β-D-thiogalactoside (IPTG) was added to a final concentration of 0.5 mM to induce recombinant protein expression. To reduce inclusion body formation and improve soluble protein expression, the cells were incubated at 20 °C with shaking at 180 rpm. After 20 h, cells were harvested by centrifugation at 6000×g for 10 min. The cell pellet was resuspended in 10 mL of lysis buffer (50 mM NaH 2 PO 4 , 300 mM NaCl, 10 mM imidazole, pH 8.0) and the cells were lysed by sonication. The lysate was centrifuged at 12,000×g for 30 min. The supernatant was then added to a 1 mL Ni-NTA agarose column (Qiagen, Hilden, Germany). The column was put on ice and gently shaken for 1 h. Then, the column was washed with 6 mL of wash buffer (50 mM NaH 2 PO 4 , 300 mM NaCl, 20 mM imidazole, pH 8.0), and the recombinant protein was eluted with 1 mL of elution buffer (50 mM NaH 2 PO 4 , 300 mM NaCl, 250 mM imidazole, pH 8.0). Purified recombinant protein was desalted using Sephadex™ G-25 (GE Healthcare, Freiburg, Germany), eluted with 1 mL of 100 mM sodium phosphate buffer (pH 7.0). The molecular weight and degree of purification were analyzed by 10% SDS-PAGE. Enzyme assays The activity of SPRHA2 and PBGL were respectively determined using p -nitrophenyl- α -L-rhamnopyranoside ( p NPR) and p NP-β-D-glucopyranosideat ( p NPG) at 50 °C, pH 7.5 and 50 °C, pH 8.0 for 20 min and 15 min. The reactions were quenched by 2 M Na 2 CO 3 . One unit of enzyme activity was defined as the amount of enzyme that released 1 µmol of p NP per minute. Substrates epimedin A, B and C, icariin, icariside I and baohuoside I were hydrolyzed by SPRHA2 and PBGL. The reactions contained 0.1 g/L substrate and dilute enzyme and were incubated in 100 mM McIlvaine buffer (pH 7.0) at 37 °C. The reactions were terminated by boiling for 5 min, and the products were analyzed by high-performance liquid chromatography (HPLC). Icaritin production from icariin by SPRHA2 and PBGL co-hydrolysis The enzymatic catalysis mixture (200 µL) contained 100 mM McIlvaine buffer (pH 7.0), 0.2 g/L icariin, 10 mg/L SPRHA2 and 10 mg/L PBGL. It was incubated at 45–70 °C for 10 min to determine the optimum reaction temperature. At the optimal temperature, the influence of pH was investigated in 100 mM McIlvaine buffer (pH 6.5-8.0), 200 mM borate saline buffer (pH 8.0–9.0) and 50 mM glycine-NaOH buffer (pH 9.0–10.0). The reactions were terminated by boiling for 5 min, and the products were analyzed by HPLC. To decrease the amount of residual reaction intermediates and increase production efficiency, the optimum weight ratio of SPRHA2 and PBGL was investigated. Icariin (0.2 g/L) was reacted by a mixture of two pure enzymes at different weight ratios (total concentration 10 mg/L) at 55 °C, pH 8.5 for 10 min. The reactions were terminated by boiling for 5 min, and the products were analyzed by HPLC. High concentrations of substrate hydrolysis by SPRHA2 and PBGL was studied under optimal reaction conditions. Because the solubility of icariin in water was low, reactions were shook to enhance hydrolysis. Icariin (25–100 g/L) was catalyzed by 40 mg/L SPRHA2 and 60 mg/L PBGL, reactions (500 µL system) were incubated at 55 °C, pH 8.5 with shaking at 220 rpm for 5 h. The reactions were terminated by boiling for 5 min. The outcomes of reactions were analyzed by HPLC. All assays were performed in triplicate. Whole-cell catalysis for icaritin production The E. coli strains containing plasmids pET- sprha2 and pET- pbgl were respectively induced as above, and harvested by centrifugation at 6000×g for 10 min. Then, harvested cells were washed with 0.9% NaCl solution and resuspended in 100 mM sodium phosphate buffer (pH 7.0). The concentration of wet cells was adjusted in an appropriate range for whole-cell catalysis. To determine the optimal temperature for reaction, the whole-cell reaction system (200 µL) included 1 g/L icariin, 2.5 g/L wet weight SPRHA2 cells, 2.5 g/L wet weight PBGL cells, and 100 mM McIlvaine buffer (pH 7.0). The mixture was incubated at 45–65 °C with shaking at 220 rpm for 10 min. At the optimal temperature, the influence of pH was investigated in 100 mM McIlvaine buffer (pH 6.5-8.0), 200 mM borate saline buffer (pH 8.0–9.0) and 50 mM glycine-NaOH buffer (pH 9.0–10.0). At the optimal temperature and pH, the weight ratio of wet cells of the two strains was investigated: the whole-cell reaction system (200 µL) contained 1 g/L icariin and different weight ratios of wet cells (total wet cell weight: 5 g/L); it was incubated at 55 °C, pH 9.0 with shaking at 220 rpm for 10 min. High concentration of substrate catalysis was investigated under the optimal reaction conditions. The reaction system (500 µL) included 40 g/L wet weight cells and 50–200 g/L icariin; it was incubated at 55 °C and pH 9.0 with shaking at 220 rpm for 5 h. All reactions were quenched by boiling for 5 min. All assays were performed in triplicate. The results of reactions were analyzed by HPLC. Analysis and identification of icaritin Samples were analyzed and quantified using HPLC (Agilent 1260, USA), equipped with a reversed-phase C18 column (250 × 4.6 mm, 5 μm, Alltech). The samples were diluted by 50% (v:v) acetonitrile, then injected and eluted with a linear gradient of solvent A (H 2 O) and solvent B (acetonitrile) as follows: 32% B (0–5 min), 32–80% B (5–12 min), 80% B (12–17 min), 80%-32% B (17–20 min), and 32% B (20–23 min) at a flow rate of 1 mL/min, and the UV absorption was measured at 270 nm. The column temperature was 35 °C. After whole-cell catalyzed conversion of 200 g/L icariin into icaritin, the reaction mixture was washed three times with water to remove glucose and rhamnose produced by the reaction. The purified icaritin was freeze-dried and subjected to NMR analysis by Avance III HD600 NMR spectrometer (Bruker, Swiss). DMSO- d 6 was used to dissolve the icaritin, which was then transferred to a 5 mm NMR tube for 13 C NMR analysis.
Icariin (purity 98.0%) and baohuoside I (purity 98.0%) were purchased from Shanghai Aladdin Bio-chemical Technology Co., Ltd. (Shanghai, China). Icariside I (purity 98.0%), icaritin (purity 99.0%), epimedin A, B and C (purity 98.0%) were purchased from Shanghai Yuanye Bio-Technology Co., Ltd. (Shanghai, China). Crude icariin extracts (purity ~ 10%, ~ 20%, ~ 50% and ~ 70%) were purchased from Shanxi Sunrun Bio-tech. Co., Ltd. (Shanxi, China). All restriction enzymes, PrimeSTAR DNA polymerase, T4 DNA ligase and In-Fusion HD Cloning kit were purchased from Takara Co., Ltd. (Dalian, China). E. coli DH5α was used for plasmid propagation and construction. E. coli BL21 (DE3) was used for expression of heterologous genes.
The coding sequences of SPRHA2 (GenBank accession: QGA89207.1) and PBGL (GenBank accession: OP810493) were amplified by PCR from genomic DNA of Novosphingobium sp. GX9 and Paenibacillus cookii GX-4, respectively. The primers were as follows: sprha2-1: 5′-CAGGGATCCGAGCCGGCACCCGATGCGGCCGC-3′; sprha2-2: 5′-CAGAAGCTTTCAGCGGGTCGTGCCCAGCGTGAC-3′; pbgl-1: 5′-CGCGGATCCAGAAACCATACTTCAGACACGATCAA-3′; and pbgl-2: 5′-CGCAAGCTTTCAGCTTCTACGGTATTTCTTGGTTC-3′. The DNA fragments were digested by restriction enzymes Bam HI and Hin d III and ligated by T4 ligase into plasmid pET30a digested with the same restriction enzymes, resulting in plasmids pET- sprha2 and pET- pbgl , respectively. The recombinant plasmids were verified by DNA sequencing (Ruibo Tech, Guangzhou, China). The plasmids pET- sprha2 and pET- pbgl were respectively transformed into E. coli BL21 (DE3) to produce proteins. To construct co-expression plasmids, the DNA fragments of sprha2 and pbgl were inserted into plasmids pET- pbgl and pET- sprha2 respectively by In-fusion clone to produce plasmids pET- pbgl - sprha2 and pET- sprha2 - pbgl . The recombinant plasmids were verified by DNA sequencing. The plasmids pET- sprha2-pbgl and pET- pbgl-sprha2 were transformed into E. coli BL21 (DE3) for whole-cell catalysis. The primers were as follows: pet30-1: 5′-CCACTGAGATCCGGCTGCTAACAAAGCCCGAAAGG-3′; pet30-2: 5′-TGGTGGTGGTGCTCGAGTGCGGCCGCAAGCTT-3′; Inf-1: 5′-CTCGAGCACCACCACCAAAGGAGATATACATATGCACCATCA-3′; sprha2-3: 5′-CAGCCGGATCTCAGTGGTCAGCGGGTCGTGCCCAGCGTGAC-3′; and pbgl-3: 5′-CAGCCGGATCTCAGTGGTCAGCTTCTACGGTATTTCTTGGTTC-3′.
For recombinant protein expression, a single colony was picked and inoculated into a flask containing 10 mL of Luria-Bertani (LB) medium for seed preparations, which included kanamycin (50 mg/L) and cultured overnight at 37 °C with constant shaking at 220 rpm. Then, 4 mL of seed culture was inoculated into 200 mL of LB medium containing kanamycin (50 mg/L) in a 500 mL flask, which was incubated at 37 °C with shaking at 220 rpm. When OD 600nm reached 0.4–0.6, isopropyl-β-D-thiogalactoside (IPTG) was added to a final concentration of 0.5 mM to induce recombinant protein expression. To reduce inclusion body formation and improve soluble protein expression, the cells were incubated at 20 °C with shaking at 180 rpm. After 20 h, cells were harvested by centrifugation at 6000×g for 10 min. The cell pellet was resuspended in 10 mL of lysis buffer (50 mM NaH 2 PO 4 , 300 mM NaCl, 10 mM imidazole, pH 8.0) and the cells were lysed by sonication. The lysate was centrifuged at 12,000×g for 30 min. The supernatant was then added to a 1 mL Ni-NTA agarose column (Qiagen, Hilden, Germany). The column was put on ice and gently shaken for 1 h. Then, the column was washed with 6 mL of wash buffer (50 mM NaH 2 PO 4 , 300 mM NaCl, 20 mM imidazole, pH 8.0), and the recombinant protein was eluted with 1 mL of elution buffer (50 mM NaH 2 PO 4 , 300 mM NaCl, 250 mM imidazole, pH 8.0). Purified recombinant protein was desalted using Sephadex™ G-25 (GE Healthcare, Freiburg, Germany), eluted with 1 mL of 100 mM sodium phosphate buffer (pH 7.0). The molecular weight and degree of purification were analyzed by 10% SDS-PAGE.
The activity of SPRHA2 and PBGL were respectively determined using p -nitrophenyl- α -L-rhamnopyranoside ( p NPR) and p NP-β-D-glucopyranosideat ( p NPG) at 50 °C, pH 7.5 and 50 °C, pH 8.0 for 20 min and 15 min. The reactions were quenched by 2 M Na 2 CO 3 . One unit of enzyme activity was defined as the amount of enzyme that released 1 µmol of p NP per minute. Substrates epimedin A, B and C, icariin, icariside I and baohuoside I were hydrolyzed by SPRHA2 and PBGL. The reactions contained 0.1 g/L substrate and dilute enzyme and were incubated in 100 mM McIlvaine buffer (pH 7.0) at 37 °C. The reactions were terminated by boiling for 5 min, and the products were analyzed by high-performance liquid chromatography (HPLC).
The enzymatic catalysis mixture (200 µL) contained 100 mM McIlvaine buffer (pH 7.0), 0.2 g/L icariin, 10 mg/L SPRHA2 and 10 mg/L PBGL. It was incubated at 45–70 °C for 10 min to determine the optimum reaction temperature. At the optimal temperature, the influence of pH was investigated in 100 mM McIlvaine buffer (pH 6.5-8.0), 200 mM borate saline buffer (pH 8.0–9.0) and 50 mM glycine-NaOH buffer (pH 9.0–10.0). The reactions were terminated by boiling for 5 min, and the products were analyzed by HPLC. To decrease the amount of residual reaction intermediates and increase production efficiency, the optimum weight ratio of SPRHA2 and PBGL was investigated. Icariin (0.2 g/L) was reacted by a mixture of two pure enzymes at different weight ratios (total concentration 10 mg/L) at 55 °C, pH 8.5 for 10 min. The reactions were terminated by boiling for 5 min, and the products were analyzed by HPLC. High concentrations of substrate hydrolysis by SPRHA2 and PBGL was studied under optimal reaction conditions. Because the solubility of icariin in water was low, reactions were shook to enhance hydrolysis. Icariin (25–100 g/L) was catalyzed by 40 mg/L SPRHA2 and 60 mg/L PBGL, reactions (500 µL system) were incubated at 55 °C, pH 8.5 with shaking at 220 rpm for 5 h. The reactions were terminated by boiling for 5 min. The outcomes of reactions were analyzed by HPLC. All assays were performed in triplicate.
The E. coli strains containing plasmids pET- sprha2 and pET- pbgl were respectively induced as above, and harvested by centrifugation at 6000×g for 10 min. Then, harvested cells were washed with 0.9% NaCl solution and resuspended in 100 mM sodium phosphate buffer (pH 7.0). The concentration of wet cells was adjusted in an appropriate range for whole-cell catalysis. To determine the optimal temperature for reaction, the whole-cell reaction system (200 µL) included 1 g/L icariin, 2.5 g/L wet weight SPRHA2 cells, 2.5 g/L wet weight PBGL cells, and 100 mM McIlvaine buffer (pH 7.0). The mixture was incubated at 45–65 °C with shaking at 220 rpm for 10 min. At the optimal temperature, the influence of pH was investigated in 100 mM McIlvaine buffer (pH 6.5-8.0), 200 mM borate saline buffer (pH 8.0–9.0) and 50 mM glycine-NaOH buffer (pH 9.0–10.0). At the optimal temperature and pH, the weight ratio of wet cells of the two strains was investigated: the whole-cell reaction system (200 µL) contained 1 g/L icariin and different weight ratios of wet cells (total wet cell weight: 5 g/L); it was incubated at 55 °C, pH 9.0 with shaking at 220 rpm for 10 min. High concentration of substrate catalysis was investigated under the optimal reaction conditions. The reaction system (500 µL) included 40 g/L wet weight cells and 50–200 g/L icariin; it was incubated at 55 °C and pH 9.0 with shaking at 220 rpm for 5 h. All reactions were quenched by boiling for 5 min. All assays were performed in triplicate. The results of reactions were analyzed by HPLC.
Samples were analyzed and quantified using HPLC (Agilent 1260, USA), equipped with a reversed-phase C18 column (250 × 4.6 mm, 5 μm, Alltech). The samples were diluted by 50% (v:v) acetonitrile, then injected and eluted with a linear gradient of solvent A (H 2 O) and solvent B (acetonitrile) as follows: 32% B (0–5 min), 32–80% B (5–12 min), 80% B (12–17 min), 80%-32% B (17–20 min), and 32% B (20–23 min) at a flow rate of 1 mL/min, and the UV absorption was measured at 270 nm. The column temperature was 35 °C. After whole-cell catalyzed conversion of 200 g/L icariin into icaritin, the reaction mixture was washed three times with water to remove glucose and rhamnose produced by the reaction. The purified icaritin was freeze-dried and subjected to NMR analysis by Avance III HD600 NMR spectrometer (Bruker, Swiss). DMSO- d 6 was used to dissolve the icaritin, which was then transferred to a 5 mm NMR tube for 13 C NMR analysis.
Below is the link to the electronic supplementary material. Additional file 1: Figure S1. HPLC analysis of icariside I and baohuoside I hydrolysis by SPRHA2 and PBGL; Figure S2. HPLC analysis of epimedin A, B, and C hydrolysis by SPRHA2 and PBGL; Figure S3. HPLC analysis of the time course of epimedin C hydrolysis by SPRHA2; Figure S4. Time course of hydrolysis of whole-cell hydrolysis of icariin; Figure S5. SDS-PAGE analysis of pET- pbgl - sprha2 and pET- sprha2 - pbgl in E. coli . Table S1. NMR data for standard icaritin and icaritin produced
|
DNA methylome and transcriptome profiling reveal key electrophysiology and immune dysregulation in hypertrophic cardiomyopathy | b5000816-d3aa-4231-8b7a-e04a3229e777 | 10072074 | Physiology[mh] | Hypertrophic cardiomyopathy (HCM) is a common hereditary cardiovascular disease with a global incidence of~0.2% . It is a major cause of sudden death and its main clinical manifestation is unexplained left ventricular hypertrophy (≥15 mm). Physiological characteristics of HCM include hypercontraction of the myo fasciculus, impaired diastole, increased sensitivity to calcium ions, and perturbed metabolic signalling . Changes in immune status may also occur during the tissue pathogenesis of HCM . Immunocytes account for 5.3% of the total number of ventricular cells . They maintain cardiac homoeostasis and normal physiological function but also mediate adverse post-injury inflammatory responses, myocardial remodelling, and cardiac electrophysiological abnormalities . Cardiac hypertrophy may cause immune dysregulation that negatively impacts cardiac function . HCM patients are at a relatively high risk of sudden death, and heart transplantation is the only efficacious treatment. However, transplantation is often impeded by long wait times because of the shortage of histocompatible donors and the risk of immune rejection. Hence, innovative therapeutic modalities are urgently required to delay the progression of HCM and mitigate the risk of sudden cardiac death associated with it. DNA methylation (DNAme) is a type of epigenetic modification that affects the entire human genome. It enables the same DNA sequence to be transcribed differently in various cell types. DNAme regulates tissue and organ growth and development. DNAme abnormalities are implicated in the pathogenesis and histopathological changes related to various diseases . CpG hypermethylation in promoter regions downregulates gene transcription. By contrast, CpG hypermethylation in the gene body indicates active gene transcription . DNA methyltransferases (DNMTs) and ten-eleven translocation cytosine dioxygenases (TETs) regulate DNAme and demethylation, respectively . Aberrant DNAme is involved in the growth and pathological adaptation of cardiomyocytes . DNMT3A knockdown by CRISPR/Cas9 creates aberrant DNAme in cardiomyocytes which, in turn, modulates the expression of genes associated with altered myocardial contractility, damages the mitochondria, and causes defects in lipid and glucose metabolism . Changes in the DNAme alter the genes encoding potassium channels which play critical roles in cardiac conduction and arrhythmia . Increasing evidence supports the involvement of DNAme in cardiomyopathy, heart failure and transplantation, immune effects and distant effects (e.g., myocardial remodelling and activation of fibrotic processes) . In heart failure patients, for instance, satellite (SAT) element hypermethylation is associated with significant upregulation of SAT transcription and may play a role in maintaining chromosomal integrity and myocardial stress response . DNAme predicts the risk of complications after transplantation, and FOXP3 promoter hypermethylation predicts immune rejection of heart transplantation . In late-stage HCM, some patients may have heart failure. Thus, studies investigating heart failure may detect DNAme alterations occurring in HCM. Nevertheless, dynamic changes in DNAme exist across distinct stages in disease progression , and the profile for HCM patients prior to heart failure remains unexplored. Furthermore, aberrations of DNAme may regulate innate and adaptive immunity . However, little is known about the mechanisms of immune infiltration and aberrant DNAme in HCM myocardium. We performed a microarray-based genome-wide DNAme analysis paired with transcriptome data to compare HCM against normal left ventricle myocardium. The combination of DNAme screening and transcriptome signals enabled us to investigate HCM mechanisms. It also permitted us to evaluate the effects of the previously hypothesized roles of immune cell function, cardiomyocyte development, and cardiac electrophysiology on HCM. The potential of the immune-related genes identified herein for the treatment of HCM could be explored in subsequent functional studies. Myocardial specimens All subjects provided written informed consent prior to participation. All procedures were approved by the Ethics Committee of Beijing Anzhen Hospital (No. 2021155×). Thirty-two myocardial tissue samples were collected from 24 patients with HCM and eight patients with normal myocardia at Beijing Anzhen Hospital. The HCM samples were obtained during septal myectomy and transported to the laboratory in ice-cold cardioplegia solution until cryopreservation (<4 h). HCM diagnosis was defined according to the 2020 AHA/ACC Guideline for the Diagnosis and Treatment of Patients with Hypertrophic Cardiomyopathy . All patients were diagnosed by two experienced clinicians. HCM was confirmed by 2D echocardiography showing unexplained left ventricular hypertrophy with diastolic interventricular septal thickness≥15 mm and septal wall: posterior wall thickness ratio≥1.3 in the absence of any other cardiac or systemic disease causing a similar magnitude of hypertrophy. Normal, healthy myocardial tissues were collected from donor hearts of patients who voluntarily donated their bodies for research. For all donors, clinical examination and medical history displayed no indications of cardiac history nor structural heart disease. All hearts were arrested in situ using ice-cold cardioplegia solution and transported to the lab on wet ice (always<4 hours). The left ventricles (LVs) of hearts were dissected and flash frozen in liquid nitrogen for this study. Tissue DNA was isolated with a QIAmp DNA Mini Kit (Qiagen, Hilden, Germany) and quantified by Qubit fluorometer (Thermo Fisher Scientific, Waltham, MA, USA). Total DNA>0.5 ug was selected and all samples were used for DNAme data collection. Total RNA was isolated from 300 mg tissue with TRIzolTM reagent (Invitrogen, USA) according to the manufacturer’s instruction. RNA concentration was measured using Qubit® RNA Assay Kit in Qubit® 2.0 Fluorometer (Life Technologies, CA, USA). RNA integrity was checked using the RNA Nano 6000 Assay Kit on the Agilent Bioanalyzer 2100 system (Agilent Technologies, CA, USA). The selection criteria were total RNA>0.4 μg, and RNA integrity number (RIN) > 5. Of these, 23 (16 from HCM patients and 7 from normal controls) were qualified for further sequencing. RNA-seq protocol Poly(A) RNA was purified from 0.4 μg total RNA using Dynabeads Oligo(dT)25–61005 (Thermo Fisher, USA). The poly(A) RNA was fragmented into small pieces using Magnesium RNA Fragmentation Module (NEB, USA). The cleaved RNA fragments were reverse-transcribed into cDNA by SuperScript II Reverse Transcriptase (Invitrogen), which was subsequently used to synthesize U-labelled second-stranded DNAs with E. coli DNA polymerase I (NEB), RNase H (NEB) and dUTP Solution (Thermo Fisher). Size selection was performed with AMPure XP beads (Beckman Coulter, USA). After the heat-labile UDG enzyme (NEB) treatment of the U-labelled second-stranded DNAs, the ligated products were amplified with PCR. The average insert size for the final cDNA library was 300 ± 100 bp. The 2 × 150 bp pair-end sequencing for mRNA were performed on Illumina Hiseq 2500 (Genechem Technology Co. Ltd., Shanghai, China). DNA methylation DNA was bisulphite-converted with the EZ DNAm kit (Zymo Research) according to the manufacturer’s instructions and then hybridized to the Infinium MethylationEPIC BeadChip (850 K, Illumina). These microarrays were scanned using the Illumina HiScan SQ scanner by Emei Tongde Ltd. Histology HCM and control LV tissue samples were successively perfused by saline and 4% paraformaldehyde. LV tissues were fixed for 24 h (4%paraformaldehyde), transferred to ethanol (70%) for subsequent dehydration and paraffin embedding. LV samples were continuously cut into sections, and stained with Masson’s trichrome and Hematoxylin‑eosin (HE) staining by following the protocols. Variant calling and annotation Read mapping and variant calling were performed according to the Broad Institute GATK best practices workflow for SNP and Indel calling on RNA-seq data ( https://software.broadinstitute.org/gatk/documentation/article.php?id=3891 ). Briefly, paired-end reads were mapped onto the human reference genome (b37) by the STAR two-pass alignment method . Picardtools ( http://broadinstitute.github.io/picard/ ) was used to add and sort read groups, mark duplicates, and create an index. The SplitNCigarReads function ( https://gatk.broadinstitute.org/hc/en-us/articles/9570487998491-SplitNCigarReads ) was used to split the reads into exon segments. BaseRecalibrator ( https://gatk.broadinstitute.org/hc/en-us/articles/360036898312-BaseRecalibrator ) and ApplyBQSR ( https://gatk.broadinstitute.org/hc/en-us/articles/360037055712-ApplyBQSR ) were used to correct for systematic bias affecting the assignment of base quality scores by the sequencer. Variant calling was performed with HaplotypeCaller ( https://gatk.broadinstitute.org/hc/en-us/articles/360037225632-HaplotypeCaller ) and annotated with wANNOVAR ( https://wannovar.wglab.org ) . Variants in the exons of the HCM pathogenic genes ( MYBPC3 , MYH7 , MYL2 , MYL3 , TNNI3 , TNNT2 , TPM1 , and ACTC1 , and other 3769 HCM-related genes listed in Genecards) were screened from the annotation results . DNAme data QC and differentially methylated probe (DMP) analysis The bisulphite treatment in the EZ DNA Methylation-Gold Kit was used to convert the DNA samples according to the manufacturer’s instructions (Zymo Research, Irvine, CA, USA). DNAme was detected with an Infinium Human Methylation EPIC BeadChip (850k) Microarray (Illumina, San Diego, CA, USA). Raw fluorescence data for the DNAme were stored in IDAT files. The ‘ChAMP’ package ( https://bioconductor.org/packages/release/bioc/html/ChAMP.html ) in R (R Core Team, Vienna, Austria) was used for QC and data analysis. Raw data were loaded with the ‘champ.load’ function and transformed into β-values. The latter represent the ratios of the fluorescence intensities of methylated probes to those of unmethylated probes and range from 0 to 1. The β-value increases with the degree of methylation. This function was also used for initial probe screening and QC. Probes meeting any of the following criteria were removed: (1) p > 0.01; (2) < 3 beads in≥5% of all samples; (3) non-CpG; (4) multihits; (5) underlying SNPs; and (6) location on X or Y chromosome. The filtered probe set was used in the subsequent analyses. Beta mixture quantile expansion (BMIQ) normalization and singular value decomposition (SVD) were used to eliminate batch effects caused by differences in experimental time points. A principal component analysis (PCA) was performed using the ‘prcomp’ function in the ‘stats’ package ( https://statisticsglobe.com/stats-r-package ) of R. PCA reduced the dimensionality of the high-dimensional methylation site matrix/gene expression matrix into a form with only a few principal components. This small number of principal components can then effectively represent the variation at all methylation sites/gene expression, and therefore describes the differences between samples. The ‘Bumphunter’ method ( https://bioconductor.org/packages/release/bioc/html/bumphunter.html ) was used to screen DMPs. The △β-value is the arithmetic difference between the mean β-value of the HCM group and that of the normal control for a single probe. Correction of P values (adjusted P) for multiple comparisons was performed by using the Benjamini-Hochberg procedure. The DMP thresholds were adjusted P < 0.05 and log|FC| > 0.1. The latter term is equivalent to the effect of |△β-value| > 0.1. HCM hyper DMPs had logFC>0.1 whereas HCM hypo DMPs had logFC<0.1. DMP gene annotations were performed using the default annotation of the ‘ChAMP’ package in R. The reference genome was hg19 ( https://www.ncbi.nlm.nih.gov/assembly/GCF_000001405.13/ ). An in-house Python script was used to evaluate probe overrepresentation in each chromosome. To determine DMP overrepresentation in a chromosome in all DMPs (DMPSet), it was necessary to determine the a priori distribution of the number of probes (N) from a chromosome in a random probe set (ProbeSet R ) equal in size (M) to the DMPSet. To this end, M probes were sampled 10,000 times with replacements from all probes in the array, and the cumulative probability of observing N > M within each chromosome in a ProbeSet R was determined. For HCM hyper DMPs and HCM hypo DMPs, 2,202 and 1,864 probes, respectively, were randomly sampled 10,000 times from 732,724 total probes in the array. The significance of the overrepresented chromosome was the sum of the probabilities (P-values) of finding N > M probes from this chromosome in the DMPSet. P value<0.05 indicated that the number of probes was significantly increased on the chromosome. RNA-seq data QC and differential expression analysis RNA was extracted and purified from 16 HCM and seven normal samples. Sequencing libraries with insert fragment length = 380 bp were sequenced on the Illumina HiSeq X ten platform (paired-end; 2 × 150 bp) and generated 6Gb raw data/sample. Raw reads were filtered with Cutadapt ( https://pypi.org/project/cutadapt/ ) to remove 3’ adapters with≥10 bp overlap (AGATCGGAAG) while allowing for a 20% base error rate. FastQC ( https://www.bioinformatics.babraham.ac.uk/projects/fastqc/ ) was used to filter the reads using the default QC parameters. HISAT2 ( http://daehwankimlab.github.io/hisat2/ ) was used to align the filtered reads to the GRCh38 reference genome ( https://www.ncbi.nlm.nih.gov/data-hub/genome/GCF_000001405.40/ ). The ‘Union’ scheme in HTSeq ( https://pypi.org/project/HTSeq/ ) was used for counting. Raw read counts were analysed with DESeq2 ( https://bioconductor.org/packages/release/bioc/html/DESeq2.html ). Those with adjusted P < 0.05 and log 2 |FC| > 1.0 were considered differentially expressed genes (DEGs). PCA was performed using the built-in function in DESeq2. Gene set enrichment analysis (GSEA) and gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analyses GSEA and GO and KEGG enrichment analyses were performed using the ‘clusterProfiler’ package ( https://bioconductor.org/packages/release/bioc/html/clusterProfiler.html ) in R . All DMPs were annotated with hg19 to obtain the DMP genes. The GO (adjusted P < 0.05) and KEGG( P value<0.05) enrichment analyses were performed on all DMP genes and DEGs. The significantly enriched GO items were grouped in either dataset into the functional network within ClueGO v. 2.5.9 ( https://apps.cytoscape.org/apps/cluego ) in Cytoscape v. 3.9.1 ( https://github.com/cytoscape/cytoscape/releases ). The default parameters were edge-weighted, force-directed, and BioLayout ( http://biolayout.org/download.html ) for CluePedia ( https://apps.cytoscape.org/apps/cluepedia ). Cytoscape was used to construct a network from the overlapping pathways in the results of the GO enrichment analyses of the DMP genes and the DEGs. The κ (kappa) score threshold was 0.3 and the sharing group percentage was 30.0%. GSEA was applied only to DEGs with adjusted P < 0.05. Protein-protein interaction (PPI) network analysis and pathogenic gene screening for overlapping genes of DMP genes and DEGs A PPI network of overlapping genes was constructed based on the STRING ( https://string-db.org/ ), an online protein network database . Genes with moderate confidence interaction scores (>0.4) were selected and isolated node genes were removed. The remaining genes were clustered into subnetworks by the Markov clustering (MCL) method. The inflation parameter was 1.5. All other parameters used in this analysis were set by default. The immune-related gene dataset that was confirmed in earlier studies to be implicated in the natural immune process was downloaded from the IMMPORT database ( https://www.immport.org/shared/ ). The GeneCards ( https://www.genecards.org/ ) database was used to screen and obtain pathogenic HCM genes. The keyword was ‘hypertrophic cardiomyopathy’ . xCell immune infiltration analysis The xCell analysis uses a deconvolution algorithm and integrates the advantages of GSEA. The xCell calculated single-sample gene set enrichment analysis (ssGSEA) scores for 489 gene signatures. Gain compensation was then corrected and new scores were used to assess the relative abundance of 64 cell types in each tissue sample. Cell types included multiple adaptive and innate immune, haematopoietic progenitor, epithelial, and extracellular stromal cells as well as 48 TME-associated cells. Gene expression values (fragments per kilobase per million mapped reads or FPKM) from the bulk RNA-seq data were uploaded to http://xCell.ucsf.edu/ to obtain 64 cell scores per sample. Immune cells with P value<0.05 (t -test) were selected for display. Public HCM and normal myocardium RNA-seq data The public GSE130036 dataset ( https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE130036 ) contains raw RNA-seq data for 28 HCM and nine normal myocardial samples. Trimmomatic ( https://github.com/usadellab/Trimmomatic ) was used to remove the adapter, and FastQC ( https://www.bioinformatics.babraham.ac.uk/projects/fastqc/ ) with its default parameters was used for quality control. HISAT2 ( http://daehwankimlab.github.io/hisat2/ ) compared the filtered reads against the GRCh38 reference genome. The ‘Union’ scheme in HTSeq ( https://pypi.org/project/HTSeq/ ) was used for counting. Data availability The raw DNAme data (No. PRJCA009134; https://ngdc.cncb.ac.cn/omix/preview/ocAxwKJt ) and the raw RNA-seq data (No. PRJCA009145; https://ngdc.cncb.ac.cn/gsa-human/s/mRwWy2L3 ) reported herein were deposited to the Genome Sequence Archive of the China National Center for Bioinformation/Beijing Institute of Genomics of the Chinese Academy of Sciences and is publicly accessible. All subjects provided written informed consent prior to participation. All procedures were approved by the Ethics Committee of Beijing Anzhen Hospital (No. 2021155×). Thirty-two myocardial tissue samples were collected from 24 patients with HCM and eight patients with normal myocardia at Beijing Anzhen Hospital. The HCM samples were obtained during septal myectomy and transported to the laboratory in ice-cold cardioplegia solution until cryopreservation (<4 h). HCM diagnosis was defined according to the 2020 AHA/ACC Guideline for the Diagnosis and Treatment of Patients with Hypertrophic Cardiomyopathy . All patients were diagnosed by two experienced clinicians. HCM was confirmed by 2D echocardiography showing unexplained left ventricular hypertrophy with diastolic interventricular septal thickness≥15 mm and septal wall: posterior wall thickness ratio≥1.3 in the absence of any other cardiac or systemic disease causing a similar magnitude of hypertrophy. Normal, healthy myocardial tissues were collected from donor hearts of patients who voluntarily donated their bodies for research. For all donors, clinical examination and medical history displayed no indications of cardiac history nor structural heart disease. All hearts were arrested in situ using ice-cold cardioplegia solution and transported to the lab on wet ice (always<4 hours). The left ventricles (LVs) of hearts were dissected and flash frozen in liquid nitrogen for this study. Tissue DNA was isolated with a QIAmp DNA Mini Kit (Qiagen, Hilden, Germany) and quantified by Qubit fluorometer (Thermo Fisher Scientific, Waltham, MA, USA). Total DNA>0.5 ug was selected and all samples were used for DNAme data collection. Total RNA was isolated from 300 mg tissue with TRIzolTM reagent (Invitrogen, USA) according to the manufacturer’s instruction. RNA concentration was measured using Qubit® RNA Assay Kit in Qubit® 2.0 Fluorometer (Life Technologies, CA, USA). RNA integrity was checked using the RNA Nano 6000 Assay Kit on the Agilent Bioanalyzer 2100 system (Agilent Technologies, CA, USA). The selection criteria were total RNA>0.4 μg, and RNA integrity number (RIN) > 5. Of these, 23 (16 from HCM patients and 7 from normal controls) were qualified for further sequencing. RNA-seq protocol Poly(A) RNA was purified from 0.4 μg total RNA using Dynabeads Oligo(dT)25–61005 (Thermo Fisher, USA). The poly(A) RNA was fragmented into small pieces using Magnesium RNA Fragmentation Module (NEB, USA). The cleaved RNA fragments were reverse-transcribed into cDNA by SuperScript II Reverse Transcriptase (Invitrogen), which was subsequently used to synthesize U-labelled second-stranded DNAs with E. coli DNA polymerase I (NEB), RNase H (NEB) and dUTP Solution (Thermo Fisher). Size selection was performed with AMPure XP beads (Beckman Coulter, USA). After the heat-labile UDG enzyme (NEB) treatment of the U-labelled second-stranded DNAs, the ligated products were amplified with PCR. The average insert size for the final cDNA library was 300 ± 100 bp. The 2 × 150 bp pair-end sequencing for mRNA were performed on Illumina Hiseq 2500 (Genechem Technology Co. Ltd., Shanghai, China). DNA methylation DNA was bisulphite-converted with the EZ DNAm kit (Zymo Research) according to the manufacturer’s instructions and then hybridized to the Infinium MethylationEPIC BeadChip (850 K, Illumina). These microarrays were scanned using the Illumina HiScan SQ scanner by Emei Tongde Ltd. Histology HCM and control LV tissue samples were successively perfused by saline and 4% paraformaldehyde. LV tissues were fixed for 24 h (4%paraformaldehyde), transferred to ethanol (70%) for subsequent dehydration and paraffin embedding. LV samples were continuously cut into sections, and stained with Masson’s trichrome and Hematoxylin‑eosin (HE) staining by following the protocols. Poly(A) RNA was purified from 0.4 μg total RNA using Dynabeads Oligo(dT)25–61005 (Thermo Fisher, USA). The poly(A) RNA was fragmented into small pieces using Magnesium RNA Fragmentation Module (NEB, USA). The cleaved RNA fragments were reverse-transcribed into cDNA by SuperScript II Reverse Transcriptase (Invitrogen), which was subsequently used to synthesize U-labelled second-stranded DNAs with E. coli DNA polymerase I (NEB), RNase H (NEB) and dUTP Solution (Thermo Fisher). Size selection was performed with AMPure XP beads (Beckman Coulter, USA). After the heat-labile UDG enzyme (NEB) treatment of the U-labelled second-stranded DNAs, the ligated products were amplified with PCR. The average insert size for the final cDNA library was 300 ± 100 bp. The 2 × 150 bp pair-end sequencing for mRNA were performed on Illumina Hiseq 2500 (Genechem Technology Co. Ltd., Shanghai, China). DNA was bisulphite-converted with the EZ DNAm kit (Zymo Research) according to the manufacturer’s instructions and then hybridized to the Infinium MethylationEPIC BeadChip (850 K, Illumina). These microarrays were scanned using the Illumina HiScan SQ scanner by Emei Tongde Ltd. HCM and control LV tissue samples were successively perfused by saline and 4% paraformaldehyde. LV tissues were fixed for 24 h (4%paraformaldehyde), transferred to ethanol (70%) for subsequent dehydration and paraffin embedding. LV samples were continuously cut into sections, and stained with Masson’s trichrome and Hematoxylin‑eosin (HE) staining by following the protocols. Read mapping and variant calling were performed according to the Broad Institute GATK best practices workflow for SNP and Indel calling on RNA-seq data ( https://software.broadinstitute.org/gatk/documentation/article.php?id=3891 ). Briefly, paired-end reads were mapped onto the human reference genome (b37) by the STAR two-pass alignment method . Picardtools ( http://broadinstitute.github.io/picard/ ) was used to add and sort read groups, mark duplicates, and create an index. The SplitNCigarReads function ( https://gatk.broadinstitute.org/hc/en-us/articles/9570487998491-SplitNCigarReads ) was used to split the reads into exon segments. BaseRecalibrator ( https://gatk.broadinstitute.org/hc/en-us/articles/360036898312-BaseRecalibrator ) and ApplyBQSR ( https://gatk.broadinstitute.org/hc/en-us/articles/360037055712-ApplyBQSR ) were used to correct for systematic bias affecting the assignment of base quality scores by the sequencer. Variant calling was performed with HaplotypeCaller ( https://gatk.broadinstitute.org/hc/en-us/articles/360037225632-HaplotypeCaller ) and annotated with wANNOVAR ( https://wannovar.wglab.org ) . Variants in the exons of the HCM pathogenic genes ( MYBPC3 , MYH7 , MYL2 , MYL3 , TNNI3 , TNNT2 , TPM1 , and ACTC1 , and other 3769 HCM-related genes listed in Genecards) were screened from the annotation results . The bisulphite treatment in the EZ DNA Methylation-Gold Kit was used to convert the DNA samples according to the manufacturer’s instructions (Zymo Research, Irvine, CA, USA). DNAme was detected with an Infinium Human Methylation EPIC BeadChip (850k) Microarray (Illumina, San Diego, CA, USA). Raw fluorescence data for the DNAme were stored in IDAT files. The ‘ChAMP’ package ( https://bioconductor.org/packages/release/bioc/html/ChAMP.html ) in R (R Core Team, Vienna, Austria) was used for QC and data analysis. Raw data were loaded with the ‘champ.load’ function and transformed into β-values. The latter represent the ratios of the fluorescence intensities of methylated probes to those of unmethylated probes and range from 0 to 1. The β-value increases with the degree of methylation. This function was also used for initial probe screening and QC. Probes meeting any of the following criteria were removed: (1) p > 0.01; (2) < 3 beads in≥5% of all samples; (3) non-CpG; (4) multihits; (5) underlying SNPs; and (6) location on X or Y chromosome. The filtered probe set was used in the subsequent analyses. Beta mixture quantile expansion (BMIQ) normalization and singular value decomposition (SVD) were used to eliminate batch effects caused by differences in experimental time points. A principal component analysis (PCA) was performed using the ‘prcomp’ function in the ‘stats’ package ( https://statisticsglobe.com/stats-r-package ) of R. PCA reduced the dimensionality of the high-dimensional methylation site matrix/gene expression matrix into a form with only a few principal components. This small number of principal components can then effectively represent the variation at all methylation sites/gene expression, and therefore describes the differences between samples. The ‘Bumphunter’ method ( https://bioconductor.org/packages/release/bioc/html/bumphunter.html ) was used to screen DMPs. The △β-value is the arithmetic difference between the mean β-value of the HCM group and that of the normal control for a single probe. Correction of P values (adjusted P) for multiple comparisons was performed by using the Benjamini-Hochberg procedure. The DMP thresholds were adjusted P < 0.05 and log|FC| > 0.1. The latter term is equivalent to the effect of |△β-value| > 0.1. HCM hyper DMPs had logFC>0.1 whereas HCM hypo DMPs had logFC<0.1. DMP gene annotations were performed using the default annotation of the ‘ChAMP’ package in R. The reference genome was hg19 ( https://www.ncbi.nlm.nih.gov/assembly/GCF_000001405.13/ ). An in-house Python script was used to evaluate probe overrepresentation in each chromosome. To determine DMP overrepresentation in a chromosome in all DMPs (DMPSet), it was necessary to determine the a priori distribution of the number of probes (N) from a chromosome in a random probe set (ProbeSet R ) equal in size (M) to the DMPSet. To this end, M probes were sampled 10,000 times with replacements from all probes in the array, and the cumulative probability of observing N > M within each chromosome in a ProbeSet R was determined. For HCM hyper DMPs and HCM hypo DMPs, 2,202 and 1,864 probes, respectively, were randomly sampled 10,000 times from 732,724 total probes in the array. The significance of the overrepresented chromosome was the sum of the probabilities (P-values) of finding N > M probes from this chromosome in the DMPSet. P value<0.05 indicated that the number of probes was significantly increased on the chromosome. RNA was extracted and purified from 16 HCM and seven normal samples. Sequencing libraries with insert fragment length = 380 bp were sequenced on the Illumina HiSeq X ten platform (paired-end; 2 × 150 bp) and generated 6Gb raw data/sample. Raw reads were filtered with Cutadapt ( https://pypi.org/project/cutadapt/ ) to remove 3’ adapters with≥10 bp overlap (AGATCGGAAG) while allowing for a 20% base error rate. FastQC ( https://www.bioinformatics.babraham.ac.uk/projects/fastqc/ ) was used to filter the reads using the default QC parameters. HISAT2 ( http://daehwankimlab.github.io/hisat2/ ) was used to align the filtered reads to the GRCh38 reference genome ( https://www.ncbi.nlm.nih.gov/data-hub/genome/GCF_000001405.40/ ). The ‘Union’ scheme in HTSeq ( https://pypi.org/project/HTSeq/ ) was used for counting. Raw read counts were analysed with DESeq2 ( https://bioconductor.org/packages/release/bioc/html/DESeq2.html ). Those with adjusted P < 0.05 and log 2 |FC| > 1.0 were considered differentially expressed genes (DEGs). PCA was performed using the built-in function in DESeq2. GSEA and GO and KEGG enrichment analyses were performed using the ‘clusterProfiler’ package ( https://bioconductor.org/packages/release/bioc/html/clusterProfiler.html ) in R . All DMPs were annotated with hg19 to obtain the DMP genes. The GO (adjusted P < 0.05) and KEGG( P value<0.05) enrichment analyses were performed on all DMP genes and DEGs. The significantly enriched GO items were grouped in either dataset into the functional network within ClueGO v. 2.5.9 ( https://apps.cytoscape.org/apps/cluego ) in Cytoscape v. 3.9.1 ( https://github.com/cytoscape/cytoscape/releases ). The default parameters were edge-weighted, force-directed, and BioLayout ( http://biolayout.org/download.html ) for CluePedia ( https://apps.cytoscape.org/apps/cluepedia ). Cytoscape was used to construct a network from the overlapping pathways in the results of the GO enrichment analyses of the DMP genes and the DEGs. The κ (kappa) score threshold was 0.3 and the sharing group percentage was 30.0%. GSEA was applied only to DEGs with adjusted P < 0.05. A PPI network of overlapping genes was constructed based on the STRING ( https://string-db.org/ ), an online protein network database . Genes with moderate confidence interaction scores (>0.4) were selected and isolated node genes were removed. The remaining genes were clustered into subnetworks by the Markov clustering (MCL) method. The inflation parameter was 1.5. All other parameters used in this analysis were set by default. The immune-related gene dataset that was confirmed in earlier studies to be implicated in the natural immune process was downloaded from the IMMPORT database ( https://www.immport.org/shared/ ). The GeneCards ( https://www.genecards.org/ ) database was used to screen and obtain pathogenic HCM genes. The keyword was ‘hypertrophic cardiomyopathy’ . The xCell analysis uses a deconvolution algorithm and integrates the advantages of GSEA. The xCell calculated single-sample gene set enrichment analysis (ssGSEA) scores for 489 gene signatures. Gain compensation was then corrected and new scores were used to assess the relative abundance of 64 cell types in each tissue sample. Cell types included multiple adaptive and innate immune, haematopoietic progenitor, epithelial, and extracellular stromal cells as well as 48 TME-associated cells. Gene expression values (fragments per kilobase per million mapped reads or FPKM) from the bulk RNA-seq data were uploaded to http://xCell.ucsf.edu/ to obtain 64 cell scores per sample. Immune cells with P value<0.05 (t -test) were selected for display. The public GSE130036 dataset ( https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE130036 ) contains raw RNA-seq data for 28 HCM and nine normal myocardial samples. Trimmomatic ( https://github.com/usadellab/Trimmomatic ) was used to remove the adapter, and FastQC ( https://www.bioinformatics.babraham.ac.uk/projects/fastqc/ ) with its default parameters was used for quality control. HISAT2 ( http://daehwankimlab.github.io/hisat2/ ) compared the filtered reads against the GRCh38 reference genome. The ‘Union’ scheme in HTSeq ( https://pypi.org/project/HTSeq/ ) was used for counting. The raw DNAme data (No. PRJCA009134; https://ngdc.cncb.ac.cn/omix/preview/ocAxwKJt ) and the raw RNA-seq data (No. PRJCA009145; https://ngdc.cncb.ac.cn/gsa-human/s/mRwWy2L3 ) reported herein were deposited to the Genome Sequence Archive of the China National Center for Bioinformation/Beijing Institute of Genomics of the Chinese Academy of Sciences and is publicly accessible. Clinical characteristics of study patients and donors Thirty-two cardiac ventricular septal tissue samples were collected from 24 HCM patients and eight donors (Methods; ). DNAme profiles were determined for all samples, and paired RNA-seq analyses were performed on 23 samples . The HCM group consisted of 14 males and 10 females with average age 56 ± 13 y and maximum left ventricular wall thickness (LVST) = 19.3 ± 4.4 cm. There was no significant difference between the sexes in terms of their maximum LVST ( P = 0.6854; t -test). HE staining results indicated myocyte hypertrophy, myocyte disarray, nuclear hyperchromatism, and vacuolar degeneration in HCM patients compared with normal controls in 40× and 200× views. Interstitial fibrosis and replacement fibrosis were detected in Masson-stained samples of HCM patients. The HCM patients harboured different variants in the exons of their pathogenic HCM genes MYBPC3 , MYH7 , MYL2 , MYL3 , TNNI3 , TNNT2 , TPM1 , and ACTC1 . No major pathogenic variants were responsible for patient stratification (Methods; Supplementary Table S1). We did not find significant differences in either methylome or transcriptome in terms of gender (male, or female), BMI (>24 or<24) and diabetes (with or without). DMPs: methylation level and chromosome distribution After quality control, 732,724 CpG probes were obtained and used to cluster 32 samples into two groups consistent with disease grouping (PCA; Methods; ). The HCM myocardium presented with both hypomethylation and hypermethylation changes and exhibited elevated double β-value distribution peaks (Methods; ). There were 4,066 DMPs (adjusted P < 0.05; log|FC| > 0.1; Methods; Supplementary Table S2) of which 54% were hypermethylated (HCM hyper DMPs; Methods). Fluctuations in the hypermethylation and hypomethylation alterations were indicated by the △β-value ranges (Methods; ). These were −0.29 to −0.1 for the HCM hypo DMPs and 0.1 to 0.34 for the HCM hyper DMPs. The DMPs were unevenly distributed across the chromosomes (Genomic annotation; Methods). The HCM hyper DMPs were enriched on chromosomes 4,5, and 12 while the HCM hypo DMPs were located on chromosomes 2, 3, 9, 10, 13, 18, and 21 ( P value<0.05, ). In terms of genomic distribution , 48% and 42% of the HCM hypo DMPs and HCM hyper DMPs, respectively, were located within the gene body while 84% and 66% of HCM hypo DMPs and HCM hyper DMPs, respectively, were located within the opensea. Overall, the methylation levels of the 4,066 DMPs significantly differed and could effectively discriminate the HCM myocardium from the normal tissues . Hypermethylation and hypomethylation disturb crucial cardiac cell functions in HCM pathogenesis and are associated with corresponding changes in transcriptional activity DMPs were annotated for 1,927 genes (DMP genes; Genomic annotation; Methods). Hyper-DMPs and hypo-DMPs influenced distinct cellular functions. The HCM hyper DMP genes were associated with cardiac growth and development processes such as heart morphogenesis (Gene ontology enrichment; Methods; adjusted P = 1.02 × 10 −5 ), cardiac chamber morphogenesis (adjusted P = 1.22 × 10 −5 ), cardiac chamber development (adjusted P = 1.22 × 10 −5 ), and outflow tract morphogenesis (adjusted P = 1.83 × 10 −5 , Supplementary Figure S1A). However, the HCM hypo DMP genes were correlated with muscle structure-related processes including actin binding (adjusted P = 4.19 × 10 −5 ), myofibrils (adjusted P = 3.95 × 10 −4 ), sarcomeres (adjusted P = 7.11 × 10 −4 ), contractile fibres (adjusted P = 7.34 × 10 −4 ), contractile actin filament bundles (adjusted P = 7.34 × 10 −4 ), and stress fibres (adjusted P = 7.34 × 10 −4 , Supplementary Figure S1B). A paired RNA-seq analysis of the 23 samples was performed to identify gene expression alterations related to aberrant DNAmes (Supplementary Table 3; PCA, Methods; ). We identified 905 DEGs in the HCM samples (adjusted P < 0.05, log 2 |FC| > 1.0, Methods; Supplementary Table 4; ). The DEGs were significantly enriched in the pathways related to the innate immune system (GSEA; enrichment score = −0.43; adjusted P = 1.24 × 10 −3 ; , left) and the adaptive immune system (GSEA enrichment score = −0.43; adjusted P = 1.79 × 10 −2 ; , right). Their transcriptional activity was decreased in the HCM myocardium. A GO enrichment of the DEGs and DMP genes revealed pathways with extensive DNAme and gene expression alterations that were functionally grouped into various networks (Methods; ). Two major immune subnetworks included most of the GO items. One was characterized by ‘regulation of leukocyte migration’ while the other was characterized by ‘muscle system process.’ Calcium signalling was the only significantly enriched KEGG pathway for both the DMP genes ( P value = 2.18 × 10 −4 ) and the DEGs ( P value = 6.04 × 10 −3 ) . The transcriptional activity of the DNMTs and TETs did not explain the observed changes in HCM DNAme. There were no significant differences in transcription between HCM myocardium and normal tissues for any DNMT family member genes ( DNMT1 , DNMT3A , and DNMT3B ; P value>0.05; t -test; ) or TET-related genes ( TET1 , TET2 , and TET3 ; P value>0.05; t -test; ). A public HCM dataset validated our observations ( P value>0.05; t -test; ). Of the methyl-CpG-binding domain (MBD) proteins or epigenome readers , EGR1 was downregulated and BAZ2B was upregulated, but there were no DNAme alterations (Supplementary Table S4). Changes in DNAme were detected in MBD4 , ZBTB38 , KLF4 , WT1 , and UHRF2 but without any obvious alteration of transcription (Supplementary Table S2). Compared with previous observations in human cardiomyocytes (hCMs) , DMP genes influenced all three constructed function groups: muscle contraction, cardiac transcriptional regulation, and heart development (Supplementary Table S5). The nodes of the PPI network of DEGs harbouring DMPs underscored the roles of immune regulation, cardiac development, and electrophysiology in HCM pathogenesis To identify DEGs caused by DNAme alterations, we extracted 85 genes common to both the DEGs and the DMP genes (Supplementary Table 6; , left). Twenty-four percent (20/85) of them were HCM genes in GeneCards (Supplementary Table 7). TTN-AS1 was the most relevant gene (score = 169) to HCM. Hence, these disease-causing genes may also have DNAme alterations and contribute to HCM occurrence. Nine of the 85 common genes, namely, ESR1 , GDF6 , IL20RA , PMP2 , SCG2 , STC2 , TGFA , TGFB2 , and VEGFC , are associated with cytokines, cytokine receptors, the TGFb family, interleukin receptors, and antimicrobials (IMMPORT database; Supplementary Table 8). PPI networks were resolved for 79 DEGs with aberrant DNAme (Methods; , right). Genes associated with disease causality, namely, ESR1 , RUNX2 , and CACNA1A , were crucial nodes with edges>3. ESR1 and RUNX2 belonged to the larger of the two clusters associated with tissue and organ growth and development. CACNA1A was the central node of the cluster involved in cardiac electrophysiology. Twenty-five DEGs could be explained by the regulatory roles of DNAme alterations (Methods; ). According to the enriched KEGG and GO pathways, the DEGs were involved in biological processes related to immune response, myocardial growth and development, cardiac conduction, and electrophysiology. The immune response-related genes included ESR1 , ITLN1 , AHNAK2 , FFAR2 , IL20RA , and VEGFC . The myocardial growth and development-related genes included ESR1 , ITLN1 , EYA2 , VEGFC , and DCHS2 . The myocardial conduction and electrophysiology-related genes included ESR1 , KCNIP1 , KCNA5 , and AHNAK2 . The HCM hyper DMPs in the gene body indicated transcriptional upregulation of ESR1 and VEGFC while those in the promoter might explain IL20RA downregulation . DCHS2 was the most significantly upregulated gene (log2FC = 4.6; adjusted P = 2.17 × 10 −3 ; ) and it encodes a calcium-dependent cell-adhesion protein. ITLN1 was the most significantly downregulated gene (log2FC = − 6.3, adjusted P = 3.15 × 10 −7 ; ). Dysregulated immune responses were observed in cardiac hypertrophy and remodelling The heart consists of heterogeneous cell subtypes intensively involved in cardiac hypertrophy and remodelling . We estimated the immunocyte composition (macrophages, T cells, and dendritic cells (DCs)) of the HCM myocardium using the novel gene signature-based xCell method (xCell; Methods). The ImmuneScore was significantly lower for the ventricle tissues of the HCM group than it was for those of the normal group. Macrophages comprise a wide range of functionally heterogeneous phenotypes. Understanding the roles of various macrophage phenotypes during cardiac remodelling could help develop a promising therapeutic strategy . In the present study, we found that the total macrophages ( P value<0.01; t -test) and M1 macrophages ( P value<0.01; t -test) were significantly decreased in HCM patients . Macrophages respond to different environmental signals and activate various polarization programs . Similarly, monocytes ( P value<0.05; t -test) and induced dendritic cells (iDCs; P value<0.05; t -test) were significantly decreased in HCM patients. Relative to the public dataset, our results showed that the active dendritic cells (aDCs) ( P value<0.01; t -test) and M2 macrophages ( P value<0.05; t -test) were significantly decreased in HCM myocardial tissue (Methods; ). The preceding results reflect end-stage HCM which might differ from early disease onset. The stromal cells were significantly increased in the public datasets (GSE130036) as well as our own compared with the normal group . The foregoing findings indicate that nonmyocytes in general and immune cells, in particular, were dysregulated in HCM patients. The immune response and regeneration capacity may be significantly impaired in HCM patients as their immune, stromal, and stem cells are all dysregulated. Nevertheless, the mechanisms of these dysregulations remain to be elucidated and merit further investigation. Thirty-two cardiac ventricular septal tissue samples were collected from 24 HCM patients and eight donors (Methods; ). DNAme profiles were determined for all samples, and paired RNA-seq analyses were performed on 23 samples . The HCM group consisted of 14 males and 10 females with average age 56 ± 13 y and maximum left ventricular wall thickness (LVST) = 19.3 ± 4.4 cm. There was no significant difference between the sexes in terms of their maximum LVST ( P = 0.6854; t -test). HE staining results indicated myocyte hypertrophy, myocyte disarray, nuclear hyperchromatism, and vacuolar degeneration in HCM patients compared with normal controls in 40× and 200× views. Interstitial fibrosis and replacement fibrosis were detected in Masson-stained samples of HCM patients. The HCM patients harboured different variants in the exons of their pathogenic HCM genes MYBPC3 , MYH7 , MYL2 , MYL3 , TNNI3 , TNNT2 , TPM1 , and ACTC1 . No major pathogenic variants were responsible for patient stratification (Methods; Supplementary Table S1). We did not find significant differences in either methylome or transcriptome in terms of gender (male, or female), BMI (>24 or<24) and diabetes (with or without). After quality control, 732,724 CpG probes were obtained and used to cluster 32 samples into two groups consistent with disease grouping (PCA; Methods; ). The HCM myocardium presented with both hypomethylation and hypermethylation changes and exhibited elevated double β-value distribution peaks (Methods; ). There were 4,066 DMPs (adjusted P < 0.05; log|FC| > 0.1; Methods; Supplementary Table S2) of which 54% were hypermethylated (HCM hyper DMPs; Methods). Fluctuations in the hypermethylation and hypomethylation alterations were indicated by the △β-value ranges (Methods; ). These were −0.29 to −0.1 for the HCM hypo DMPs and 0.1 to 0.34 for the HCM hyper DMPs. The DMPs were unevenly distributed across the chromosomes (Genomic annotation; Methods). The HCM hyper DMPs were enriched on chromosomes 4,5, and 12 while the HCM hypo DMPs were located on chromosomes 2, 3, 9, 10, 13, 18, and 21 ( P value<0.05, ). In terms of genomic distribution , 48% and 42% of the HCM hypo DMPs and HCM hyper DMPs, respectively, were located within the gene body while 84% and 66% of HCM hypo DMPs and HCM hyper DMPs, respectively, were located within the opensea. Overall, the methylation levels of the 4,066 DMPs significantly differed and could effectively discriminate the HCM myocardium from the normal tissues . DMPs were annotated for 1,927 genes (DMP genes; Genomic annotation; Methods). Hyper-DMPs and hypo-DMPs influenced distinct cellular functions. The HCM hyper DMP genes were associated with cardiac growth and development processes such as heart morphogenesis (Gene ontology enrichment; Methods; adjusted P = 1.02 × 10 −5 ), cardiac chamber morphogenesis (adjusted P = 1.22 × 10 −5 ), cardiac chamber development (adjusted P = 1.22 × 10 −5 ), and outflow tract morphogenesis (adjusted P = 1.83 × 10 −5 , Supplementary Figure S1A). However, the HCM hypo DMP genes were correlated with muscle structure-related processes including actin binding (adjusted P = 4.19 × 10 −5 ), myofibrils (adjusted P = 3.95 × 10 −4 ), sarcomeres (adjusted P = 7.11 × 10 −4 ), contractile fibres (adjusted P = 7.34 × 10 −4 ), contractile actin filament bundles (adjusted P = 7.34 × 10 −4 ), and stress fibres (adjusted P = 7.34 × 10 −4 , Supplementary Figure S1B). A paired RNA-seq analysis of the 23 samples was performed to identify gene expression alterations related to aberrant DNAmes (Supplementary Table 3; PCA, Methods; ). We identified 905 DEGs in the HCM samples (adjusted P < 0.05, log 2 |FC| > 1.0, Methods; Supplementary Table 4; ). The DEGs were significantly enriched in the pathways related to the innate immune system (GSEA; enrichment score = −0.43; adjusted P = 1.24 × 10 −3 ; , left) and the adaptive immune system (GSEA enrichment score = −0.43; adjusted P = 1.79 × 10 −2 ; , right). Their transcriptional activity was decreased in the HCM myocardium. A GO enrichment of the DEGs and DMP genes revealed pathways with extensive DNAme and gene expression alterations that were functionally grouped into various networks (Methods; ). Two major immune subnetworks included most of the GO items. One was characterized by ‘regulation of leukocyte migration’ while the other was characterized by ‘muscle system process.’ Calcium signalling was the only significantly enriched KEGG pathway for both the DMP genes ( P value = 2.18 × 10 −4 ) and the DEGs ( P value = 6.04 × 10 −3 ) . The transcriptional activity of the DNMTs and TETs did not explain the observed changes in HCM DNAme. There were no significant differences in transcription between HCM myocardium and normal tissues for any DNMT family member genes ( DNMT1 , DNMT3A , and DNMT3B ; P value>0.05; t -test; ) or TET-related genes ( TET1 , TET2 , and TET3 ; P value>0.05; t -test; ). A public HCM dataset validated our observations ( P value>0.05; t -test; ). Of the methyl-CpG-binding domain (MBD) proteins or epigenome readers , EGR1 was downregulated and BAZ2B was upregulated, but there were no DNAme alterations (Supplementary Table S4). Changes in DNAme were detected in MBD4 , ZBTB38 , KLF4 , WT1 , and UHRF2 but without any obvious alteration of transcription (Supplementary Table S2). Compared with previous observations in human cardiomyocytes (hCMs) , DMP genes influenced all three constructed function groups: muscle contraction, cardiac transcriptional regulation, and heart development (Supplementary Table S5). To identify DEGs caused by DNAme alterations, we extracted 85 genes common to both the DEGs and the DMP genes (Supplementary Table 6; , left). Twenty-four percent (20/85) of them were HCM genes in GeneCards (Supplementary Table 7). TTN-AS1 was the most relevant gene (score = 169) to HCM. Hence, these disease-causing genes may also have DNAme alterations and contribute to HCM occurrence. Nine of the 85 common genes, namely, ESR1 , GDF6 , IL20RA , PMP2 , SCG2 , STC2 , TGFA , TGFB2 , and VEGFC , are associated with cytokines, cytokine receptors, the TGFb family, interleukin receptors, and antimicrobials (IMMPORT database; Supplementary Table 8). PPI networks were resolved for 79 DEGs with aberrant DNAme (Methods; , right). Genes associated with disease causality, namely, ESR1 , RUNX2 , and CACNA1A , were crucial nodes with edges>3. ESR1 and RUNX2 belonged to the larger of the two clusters associated with tissue and organ growth and development. CACNA1A was the central node of the cluster involved in cardiac electrophysiology. Twenty-five DEGs could be explained by the regulatory roles of DNAme alterations (Methods; ). According to the enriched KEGG and GO pathways, the DEGs were involved in biological processes related to immune response, myocardial growth and development, cardiac conduction, and electrophysiology. The immune response-related genes included ESR1 , ITLN1 , AHNAK2 , FFAR2 , IL20RA , and VEGFC . The myocardial growth and development-related genes included ESR1 , ITLN1 , EYA2 , VEGFC , and DCHS2 . The myocardial conduction and electrophysiology-related genes included ESR1 , KCNIP1 , KCNA5 , and AHNAK2 . The HCM hyper DMPs in the gene body indicated transcriptional upregulation of ESR1 and VEGFC while those in the promoter might explain IL20RA downregulation . DCHS2 was the most significantly upregulated gene (log2FC = 4.6; adjusted P = 2.17 × 10 −3 ; ) and it encodes a calcium-dependent cell-adhesion protein. ITLN1 was the most significantly downregulated gene (log2FC = − 6.3, adjusted P = 3.15 × 10 −7 ; ). The heart consists of heterogeneous cell subtypes intensively involved in cardiac hypertrophy and remodelling . We estimated the immunocyte composition (macrophages, T cells, and dendritic cells (DCs)) of the HCM myocardium using the novel gene signature-based xCell method (xCell; Methods). The ImmuneScore was significantly lower for the ventricle tissues of the HCM group than it was for those of the normal group. Macrophages comprise a wide range of functionally heterogeneous phenotypes. Understanding the roles of various macrophage phenotypes during cardiac remodelling could help develop a promising therapeutic strategy . In the present study, we found that the total macrophages ( P value<0.01; t -test) and M1 macrophages ( P value<0.01; t -test) were significantly decreased in HCM patients . Macrophages respond to different environmental signals and activate various polarization programs . Similarly, monocytes ( P value<0.05; t -test) and induced dendritic cells (iDCs; P value<0.05; t -test) were significantly decreased in HCM patients. Relative to the public dataset, our results showed that the active dendritic cells (aDCs) ( P value<0.01; t -test) and M2 macrophages ( P value<0.05; t -test) were significantly decreased in HCM myocardial tissue (Methods; ). The preceding results reflect end-stage HCM which might differ from early disease onset. The stromal cells were significantly increased in the public datasets (GSE130036) as well as our own compared with the normal group . The foregoing findings indicate that nonmyocytes in general and immune cells, in particular, were dysregulated in HCM patients. The immune response and regeneration capacity may be significantly impaired in HCM patients as their immune, stromal, and stem cells are all dysregulated. Nevertheless, the mechanisms of these dysregulations remain to be elucidated and merit further investigation. DNAme alterations have been detected in various diseases and are promising as diagnostic and prognostic biomarkers . As an inherited disease, abnormal myocardial function due to pathogenic genomic alterations is anticipated to be an important aetiological factor of HCM, and we believe that altered DNA methylation may be a secondary alteration in myocardial function compensation. Recent progress has shown methylation changes may result from altered transcription activity in response to a stimulus in cardiomyocytes . The authors found that the acute hypoxia stress response continuously activated specific gene expression patterns and resulted in DNAme changes in regulatory regions of the corresponding genes. Therefore, transcriptional and DNAme changes help elucidate these molecular mechanisms of HCM pathogenesis. However, it requires considerable effort to determine DNAme profiles for human cardiac tissues. Animal models of heart failure and tissue samples from heart failure patients showed DNAme alterations in targeted genomic regions , here we further provided DNAme profiles from HCM patients in surgery prior to heart failure to explore changes in a relatively earlier stage in disease progression. Unlike tumour cells, HCM cardiomyocytes retain their basic morphology and functions. As transcriptional changes in methylation-related enzymes do not occur in HCM myocardium, there may be fewer changes in HCM DNAme than in the tumour genome. Abnormal gene expression in tumours may explain the extensive epigenetic reprogramming characteristic of genome-wide DNAme alterations. Future research should endeavour to investigate whether changes in DNMT and TET protein abundance occur in human HCM myocardium and identify their contributions to alterations in HCM DNAme. Besides, methylation quantitative trait loci (meQTL)/Expression quantitative trait loci (eQTL) analysis was used to correlate genetic variant loci with DNAme levels at specific CpG loci/expression of specific genes to explain genetic variation in the disease . With accumulation of this type of data for HCM, meQTL/eQTL analysis will provide a better explanation of genetic variation in familial HCM. Myocardial hypercontraction, dysregulation of calcium homoeostasis, and metabolic signalling disorders are pathological manifestations of HCM . To the best of our knowledge, the present study provided DNAme profiles for HCM tissues with paired transcriptome datasets and determine the alterations in DNAme underlying changes in gene expression. DNAme profiles facilitate the exploration of the connections underlying myocardial remodelling and cardiac electrophysiological abnormalities during pathogenesis. They may also disclose candidate targets for the development of novel therapeutic strategies for HCM. We observed decreases in innate and adaptive immune activity as well as mononuclear macrophage system-related components in the HCM myocardium. These changes are closely related to cardiac electrophysiological abnormalities and myocardial remodelling. Macrophage depletion leads to abnormalities in cardiac electrical signalling and especially atrioventricular block and predisposes the heart to progressive cardiomyopathy, reduced cardiac output, diastolic dysfunction, and impaired haemodynamics . The most significant change in HCM myocardium was a reduction in iDC content. These antigen-presenting cells have a strong migratory capacity, and various types of iDCs reside in the myocardium . In dilated cardiomyopathy, apoptosis and insufficient angiogenesis decrease DC diversity and increase the number of mature DCs . This mechanism may explain the reductions in iDCs observed in HCM myocardium. In comparison, DEGs in hCMs were enriched only in pathways associated with cell lineage and tissue differences rather than immune-related ones . Immune cell population changes in HCM are likely to result from myocardial remodelling or myocardial injury, but further efforts are required to dissect their impacts on cardiomyocytes. Future integration of high-resolution single-cell methylation sequencing and single-cell transcriptome sequencing may reveal methylation alterations in immune cell and cardiomyocyte interactions. We found that ESR1 encodes a cytokine receptor also as a transcription factor that has motifs for estrogen binding, DNA binding, and transcription activation. It had the most edges in the HCM PPI network. The two HCM hyper DMPs within its gene body may explain its upregulated transcription level. Overexpression of the ESR1 protein product ERα might be a protective factor in disease progression, prevent cardiac hypertrophy , reduce arrhythmias , improve vascular endothelial function , and modulate innate immune signalling in DCs and macrophages . There were no differences between the sexes in terms of their myocardial ESR1 expression levels . Nevertheless, males have relatively higher incidence and significantly lower average age at HCM diagnosis than females . The hormone receptor ESR1 might help maintain basic physiological functions in lesions and slow disease progression. Future investigations into the therapeutic mechanism of ESR1 might be warranted. ITLN1 underwent the most significant transcriptional changes and corresponding DNAme alterations in all HCM samples. In HCM myocardium, ITLN1 has a hypermethylated site within 1500 bp upstream of the transcriptional start site, which may lead to its transcriptional down-regulation. ITLN1 is a component of the innate immune system. It is an anti-inflammatory adipocytokine present in epicardial adipose tissue and it has cardioprotective efficacy . A recent study demonstrated altered transcriptional activity in failing ventricles . Our HCM patients displayed significantly downregulated ITLN1 and, therefore, limited macrophage differentiation into the anti-inflammatory M2 phenotype . We also observed decreased macrophage composition in the transcriptome dissection for our HCM samples. Hence, ITLN1 downregulation is a key factor in this phenomenon. Although the present work characterized HCM at the DNA and RNA levels, these bulk tissue data can not identify changes specific to a cell population, and future efforts are required to determine the DNAme alterations in cardiocytes and validate the composition changes of immune cells in HCM myocardium. In the present study, we profiled the changes in DNAme in HCM myocardium and evaluated their corresponding impact by jointly analysing DNAme and the transcriptome. We built upon traditional pathological alterations in HCM and identified possible links among immune dysregulation, cardiac electrophysiological abnormalities, and myocardial remodelling. Immune-related genes that also regulate DNAme might help identify and develop novel therapeutic strategies for HCM. Supplemental Material Click here for additional data file. |
Effect of Surface Decalcification With Hydrochloric Acid on the Determination of Estrogen Receptor, Progesterone Receptor, Ki67, and Human Epidermal Growth Factor Receptor 2 Expressions in Invasive Breast Carcinoma Based on Immunohistochemistry and Fluorescence In Situ Hybridization | 4466e957-22fc-41c6-a413-25d79fa55e84 | 10072208 | Anatomy[mh] | Materials After obtaining Institutional Review Board approval, tissue was collected from 44 invasive breast tumors and processed according to routine procedures at the Department of Pathology of The First Affiliated Hospital of Fujian Medical University, China. Original diagnoses were made between August 2021 and July 2022. The clinicopathological characteristics are shown in Table (test cohort). All tissue samples were fixed in 4% buffered formaldehyde for 24 to 48 hours according to ASCO/CAP guidelines. Samples with no histologically identifiable carcinoma were excluded from the study. After fixation, BC tissue collected for diagnostic purposes was processed according to a standard protocol and embedded in paraffin. For each formalin-fixed paraffin-embedded fragment, a 4 μm hematoxylin and eosin-stained slide were reviewed by a single pathologist to confirm the presence of malignancy. Tumor-negative fragments were excluded. Five 4 μm unstained slides were prepared (charged surface adhesion microscope slides, 20 minutes at 45 ºC) for IHC and FISH as controls. Tissue blocks were then decalcified using HCl (5% to 9% HCl by weight, Wexis Biotech Limited Corporation) for 1 hour (the common duration for SD). Exposing bone biopsy specimens to HCl for 1 hour produced ~15 to 25 SD sections. The tissue blocks were then washed in water, and another five 4 μm unstained slides were prepared for IHC and FISH. Immunohistochemical Method IHC staining for ER (mouse monoclonal, clone SP1, prediluted; Ventana Medical Systems), PR (mouse monoclonal, clone 1E2, prediluted; Ventana Medical Systems), HER2 (Her2/neu, rabbit monoclonal, clone 4B5, prediluted; Ventana Medical Systems), and Ki67 (rabbit monoclonal, clone SP6, prediluted; Ventana Medical Systems) was performed on 4 mm-thick tissue sections using a Ventana automated immune-stainer (Ventana Benchmark Ultra, Ventana Medical Systems). Pretreatment was performed with the onboard antigen retrieval method at high pH (8.4) using a cell conditioning 1 antigen retrieval buffer system at 95 °C (Ventana Medical Systems). The epitope retrieval time for antibodies was 32 minutes. The duration of incubation for primary antibodies was 24 minutes for ER, 8 minutes for PR and Her2/neu, and 10 minutes for Ki67. An indirect biotin-free system for the detection of various primary antibodies (Ultraview Universal Diaminobenzidine kit) was used for detection. The slides were then counterstained with hematoxylin (Harris). Immunohistochemical Scoring IHC scoring was performed for each slide. ER and PR were scored using a modified H -score method by multiplying the percentage of cells staining at different intensities (0 = negative, 1+ = weak, 2+ = moderate, and 3+ = strong) and summing these numbers to produce an H -score score ranging from zero (completely negative) to 300 (100% of tumor cells strongly 3+ positive). For this study, an H -score ≥1 was considered positive, an H -score of 200 to 300 was considered the strong expression, an H -score of 100 to 199 was considered the moderate expression, an H -score of 1 to 99 was considered the weak expression, and an H -score <1 was considered negative. HER2 was scored per the FDA-cleared Interpretation Guide for Ventana anti-HER2/neu (4B5) antibody clone. Ki67 was scored as a percentage of cells showing nuclear expression at any level of intensity. IHC-stained slides were evaluated independently by 2 investigators with extensive experience in breast pathology (Z.L. and H.S.). All discordant results were resolved by mutual agreement under a multihead microscope. Human Epidermal Growth Factor Receptor 2/neu Fluorescence In Situ Hybridization According to routine diagnostic procedures, cases with HER2 expressions of 2+ or 3+ were subjected to FISH using a HER2/CEP17 dual FISH probe (Abbott) with 4 μm slides. Analysis was performed on a Leica DM5500 B microscope system with Application Suite Advanced Fluorescence Software. Formalin-fixed, paraffin-embedded slides were deparaffinized and pretreated with citrate and protease enzyme solution. Next, they were dehydrated and hybridized with 10 μl of FISH probe at 37 °C overnight. The next day, the slides were washed in saline-sodium citrate buffer, counterstained with 4′,6-diamidino-2-phenylindole, dehydrated, and mounted using a mounting medium. One hundred tumor cell nuclei per tumor were assessed for HER2 gene and CEP17 probe signals at 100× magnification. The HER2/CEP17 ratio was calculated as well. A ratio <1.8 was defined as a normal copy number, a ratio of 1.8 to 2.2 was defined as an equivocal copy number, and a ratio >2.2 was defined as gene amplification, as prescribed in the ASCO and CAP guidelines. Validation of Surface Decalcification in Bone Metastases To validate our findings in real metastases, we retrospectively selected formalin-fixed paraffin-embedded bone metastasis material from 20 patients from our diagnostic pathology archives. The original diagnoses of bone metastases were made between September 2018 and May 2022. All obtained bone metastasis biopsies were taken from the vertebrae or pelvis; pathologic records indicated 10 cases with SD for 1 hour and another 10 cases treated with EDTA for time periods ranging from 6 hours to overnight according to routine procedures. These samples were subjected to IHC assessment as mentioned. Statistics Statistical analysis was performed in GraphPad Prism v 8.0. We applied paired-samples t -tests (PST) or Wilcoxon signed-rank test (WSR) for paired samples to determine the statistical significance of differences in mean expression between the studied groups. P <0.05 was considered to indicate statistical significance.
After obtaining Institutional Review Board approval, tissue was collected from 44 invasive breast tumors and processed according to routine procedures at the Department of Pathology of The First Affiliated Hospital of Fujian Medical University, China. Original diagnoses were made between August 2021 and July 2022. The clinicopathological characteristics are shown in Table (test cohort). All tissue samples were fixed in 4% buffered formaldehyde for 24 to 48 hours according to ASCO/CAP guidelines. Samples with no histologically identifiable carcinoma were excluded from the study. After fixation, BC tissue collected for diagnostic purposes was processed according to a standard protocol and embedded in paraffin. For each formalin-fixed paraffin-embedded fragment, a 4 μm hematoxylin and eosin-stained slide were reviewed by a single pathologist to confirm the presence of malignancy. Tumor-negative fragments were excluded. Five 4 μm unstained slides were prepared (charged surface adhesion microscope slides, 20 minutes at 45 ºC) for IHC and FISH as controls. Tissue blocks were then decalcified using HCl (5% to 9% HCl by weight, Wexis Biotech Limited Corporation) for 1 hour (the common duration for SD). Exposing bone biopsy specimens to HCl for 1 hour produced ~15 to 25 SD sections. The tissue blocks were then washed in water, and another five 4 μm unstained slides were prepared for IHC and FISH.
IHC staining for ER (mouse monoclonal, clone SP1, prediluted; Ventana Medical Systems), PR (mouse monoclonal, clone 1E2, prediluted; Ventana Medical Systems), HER2 (Her2/neu, rabbit monoclonal, clone 4B5, prediluted; Ventana Medical Systems), and Ki67 (rabbit monoclonal, clone SP6, prediluted; Ventana Medical Systems) was performed on 4 mm-thick tissue sections using a Ventana automated immune-stainer (Ventana Benchmark Ultra, Ventana Medical Systems). Pretreatment was performed with the onboard antigen retrieval method at high pH (8.4) using a cell conditioning 1 antigen retrieval buffer system at 95 °C (Ventana Medical Systems). The epitope retrieval time for antibodies was 32 minutes. The duration of incubation for primary antibodies was 24 minutes for ER, 8 minutes for PR and Her2/neu, and 10 minutes for Ki67. An indirect biotin-free system for the detection of various primary antibodies (Ultraview Universal Diaminobenzidine kit) was used for detection. The slides were then counterstained with hematoxylin (Harris).
IHC scoring was performed for each slide. ER and PR were scored using a modified H -score method by multiplying the percentage of cells staining at different intensities (0 = negative, 1+ = weak, 2+ = moderate, and 3+ = strong) and summing these numbers to produce an H -score score ranging from zero (completely negative) to 300 (100% of tumor cells strongly 3+ positive). For this study, an H -score ≥1 was considered positive, an H -score of 200 to 300 was considered the strong expression, an H -score of 100 to 199 was considered the moderate expression, an H -score of 1 to 99 was considered the weak expression, and an H -score <1 was considered negative. HER2 was scored per the FDA-cleared Interpretation Guide for Ventana anti-HER2/neu (4B5) antibody clone. Ki67 was scored as a percentage of cells showing nuclear expression at any level of intensity. IHC-stained slides were evaluated independently by 2 investigators with extensive experience in breast pathology (Z.L. and H.S.). All discordant results were resolved by mutual agreement under a multihead microscope.
According to routine diagnostic procedures, cases with HER2 expressions of 2+ or 3+ were subjected to FISH using a HER2/CEP17 dual FISH probe (Abbott) with 4 μm slides. Analysis was performed on a Leica DM5500 B microscope system with Application Suite Advanced Fluorescence Software. Formalin-fixed, paraffin-embedded slides were deparaffinized and pretreated with citrate and protease enzyme solution. Next, they were dehydrated and hybridized with 10 μl of FISH probe at 37 °C overnight. The next day, the slides were washed in saline-sodium citrate buffer, counterstained with 4′,6-diamidino-2-phenylindole, dehydrated, and mounted using a mounting medium. One hundred tumor cell nuclei per tumor were assessed for HER2 gene and CEP17 probe signals at 100× magnification. The HER2/CEP17 ratio was calculated as well. A ratio <1.8 was defined as a normal copy number, a ratio of 1.8 to 2.2 was defined as an equivocal copy number, and a ratio >2.2 was defined as gene amplification, as prescribed in the ASCO and CAP guidelines.
To validate our findings in real metastases, we retrospectively selected formalin-fixed paraffin-embedded bone metastasis material from 20 patients from our diagnostic pathology archives. The original diagnoses of bone metastases were made between September 2018 and May 2022. All obtained bone metastasis biopsies were taken from the vertebrae or pelvis; pathologic records indicated 10 cases with SD for 1 hour and another 10 cases treated with EDTA for time periods ranging from 6 hours to overnight according to routine procedures. These samples were subjected to IHC assessment as mentioned.
Statistical analysis was performed in GraphPad Prism v 8.0. We applied paired-samples t -tests (PST) or Wilcoxon signed-rank test (WSR) for paired samples to determine the statistical significance of differences in mean expression between the studied groups. P <0.05 was considered to indicate statistical significance.
Immunohistochemical Supplementary Table 1 (Supplemental Digital Content 1, http://links.lww.com/AIMM/A403 ) shows the IHC expressions of ER, PR, HER2, and Ki67 in the 44 breast tumors (control/SD). Of the 44 cases selected, 1 was excluded from Ki67 analysis, 1 was excluded from HER2 analysis, and 2 were excluded from ER analysis because of the failure of the stain. Figure shows the IHC expressions of ER, PR, Her2, and Ki67 in the control group (routine processing) and in the treatment group (SD in HCl for 1 hour) in the 44 blinded cases of BC. No statistically significant differences in expression were identified between the control and SD groups: ER (PST, P = 0.491; WSR, P = 0.466); PR (PST, P = 0.385; WSR, P = 0.418); and Her2 (PST, P = 0.491; WSR, P = 0.466). The only marker with a statistically significant difference in mean expression was Ki67 (PST, P = 0.039; WSR: P = 0.125); the average Ki67 expression decreased between 22% to 13% after SD in HCl 1 hour ( P < 0.05). In the control tissue, 31 cases were positive for ER and 26 cases were positive for PR. No statistically significant differences in mean expression were identified between the control and treatment groups for any of the markers tested. For ER expression, categorical changes (decreased expressions) were identified in 9/31 cases after SD. Seven of these categorical changes were changes from strong to moderate (Fig. A), and 2 cases were changes from moderate to weak (Fig. B). For PR expression, categorical changes (decreased expressions) were identified in 10/26 cases after SD. Two of these changes were from strong to moderate (Fig. A), 4 were changes from moderate to weak (Fig. B), and 4 were changes from positive H -scores (10, 10, 5, and 2) in the control tissue to negative scores after SD (Fig. C). No statistically significant difference in mean HER2 expression was identified between the control and treatment groups (PST, P = 0.491; WSR, P = 0.466). In the control tissue, 23 of the 44 cases were HER2-negative (IHC score 0 or 1+), whereas 12 and 8 cases were categorized as equivocal (IHC score 2+) and HER2-positive (IHC score 3+), respectively. After SD, the IHC category in 4 cases changed from equivocal (IHC score 2+) to negative (IHC score 1+). Among the known HER2-positive cases in the control tissue, all cases remained positive after SD (IHC score 3+; Fig. ). Human Epidermal Growth Factor Receptor 2/neu Fluorescence In Situ Hybridization Twenty control tissues showed IHC scores for HER2 of 2+ or 3+ and were subjected to FISH. Supplementary Table 2 (Supplemental Digital Content 2, http://links.lww.com/AIMM/A404 ) compares the HER2 expressions obtained by IHC and FISH in the control and decalcified tissues. For HER2 copy number and HER2/CEP17, PST ( P = 0.603 and 0.578, respectively) and WSR ( P = 0.317 and 0.588, respectively) indicated no statistically significant difference between the control and SD groups (Fig. ). One HER2-positive case based on IHC (case #16) was HER2-positive based on FISH in the control but HER2-negative based on FISH in the SD samples. No clinically relevant changes (ie, HER2-positive to HER2-negative or vice versa) were identified for HER2 copy number or HER2/CEP17 in the other cases (Fig. ). Validation of Surface Decalcification in Bone Metastases ER positivity, defined as nuclear staining in >1% of tumor cell nuclei, was present in 100% of the 10 samples in the SD group and 90% of the 10 samples in the EDTA group. The mean H -scores for ER in the SD and EDTA groups were 207 and 183.5, respectively. PR positivity was seen in 40% of the SD samples and 70% of the EDTA samples. The mean H -scores for PR in the SD and EDTA groups were 90 and 71.8, respectively. The averages for Ki67 in the SD and EDTA groups were 27.5% and 34.6%, respectively, whereas those for Her2/neu were 1.7 and 1.2, respectively. The nuclear, stromal, and epithelial morphologies and staining intensity did not differ between the SD and EDTA groups (Supplementary Fig. 1, http://links.lww.com/AIMM/A405 and Supplementary Fig. 2 http://links.lww.com/AIMM/A406 ).
Supplementary Table 1 (Supplemental Digital Content 1, http://links.lww.com/AIMM/A403 ) shows the IHC expressions of ER, PR, HER2, and Ki67 in the 44 breast tumors (control/SD). Of the 44 cases selected, 1 was excluded from Ki67 analysis, 1 was excluded from HER2 analysis, and 2 were excluded from ER analysis because of the failure of the stain. Figure shows the IHC expressions of ER, PR, Her2, and Ki67 in the control group (routine processing) and in the treatment group (SD in HCl for 1 hour) in the 44 blinded cases of BC. No statistically significant differences in expression were identified between the control and SD groups: ER (PST, P = 0.491; WSR, P = 0.466); PR (PST, P = 0.385; WSR, P = 0.418); and Her2 (PST, P = 0.491; WSR, P = 0.466). The only marker with a statistically significant difference in mean expression was Ki67 (PST, P = 0.039; WSR: P = 0.125); the average Ki67 expression decreased between 22% to 13% after SD in HCl 1 hour ( P < 0.05). In the control tissue, 31 cases were positive for ER and 26 cases were positive for PR. No statistically significant differences in mean expression were identified between the control and treatment groups for any of the markers tested. For ER expression, categorical changes (decreased expressions) were identified in 9/31 cases after SD. Seven of these categorical changes were changes from strong to moderate (Fig. A), and 2 cases were changes from moderate to weak (Fig. B). For PR expression, categorical changes (decreased expressions) were identified in 10/26 cases after SD. Two of these changes were from strong to moderate (Fig. A), 4 were changes from moderate to weak (Fig. B), and 4 were changes from positive H -scores (10, 10, 5, and 2) in the control tissue to negative scores after SD (Fig. C). No statistically significant difference in mean HER2 expression was identified between the control and treatment groups (PST, P = 0.491; WSR, P = 0.466). In the control tissue, 23 of the 44 cases were HER2-negative (IHC score 0 or 1+), whereas 12 and 8 cases were categorized as equivocal (IHC score 2+) and HER2-positive (IHC score 3+), respectively. After SD, the IHC category in 4 cases changed from equivocal (IHC score 2+) to negative (IHC score 1+). Among the known HER2-positive cases in the control tissue, all cases remained positive after SD (IHC score 3+; Fig. ).
Twenty control tissues showed IHC scores for HER2 of 2+ or 3+ and were subjected to FISH. Supplementary Table 2 (Supplemental Digital Content 2, http://links.lww.com/AIMM/A404 ) compares the HER2 expressions obtained by IHC and FISH in the control and decalcified tissues. For HER2 copy number and HER2/CEP17, PST ( P = 0.603 and 0.578, respectively) and WSR ( P = 0.317 and 0.588, respectively) indicated no statistically significant difference between the control and SD groups (Fig. ). One HER2-positive case based on IHC (case #16) was HER2-positive based on FISH in the control but HER2-negative based on FISH in the SD samples. No clinically relevant changes (ie, HER2-positive to HER2-negative or vice versa) were identified for HER2 copy number or HER2/CEP17 in the other cases (Fig. ).
ER positivity, defined as nuclear staining in >1% of tumor cell nuclei, was present in 100% of the 10 samples in the SD group and 90% of the 10 samples in the EDTA group. The mean H -scores for ER in the SD and EDTA groups were 207 and 183.5, respectively. PR positivity was seen in 40% of the SD samples and 70% of the EDTA samples. The mean H -scores for PR in the SD and EDTA groups were 90 and 71.8, respectively. The averages for Ki67 in the SD and EDTA groups were 27.5% and 34.6%, respectively, whereas those for Her2/neu were 1.7 and 1.2, respectively. The nuclear, stromal, and epithelial morphologies and staining intensity did not differ between the SD and EDTA groups (Supplementary Fig. 1, http://links.lww.com/AIMM/A405 and Supplementary Fig. 2 http://links.lww.com/AIMM/A406 ).
Approximately 1% to 4% of BC patients present with bone metastasis at the time of primary BC diagnosis. The ASCO/CAP guidelines were developed to improve the accuracy of HER2 and ER/PR testing in invasive BCs. Evaluation of the IHC profile in breast carcinoma (ie, ER, PR, and HER2) is critical for clinical management decisions, and there is often a need to repeat IHC staining when BC metastasizes to distant organs. Treatment with acid decreases the antigenicity of tumor cells, thereby altering the immunomarker status of the tumor at the metastatic site. , In the case of bone-only metastatic disease, the assessment of ER, PR, and HER2 may be hindered by decalcification, and the decalcification process is known to affect HER2 FISH analysis. , Decalcification is traditionally considered to be a crucial step in processing bone metastasis specimens before routine processing and embedding. Although no ideal decalcification method has been identified, EDTA decalcification has been reported to have no significant impact on biomarker evaluation. , However, small bone tissues such as bone marrow take ~24 to 48 hours for decalcification in the EDTA protocol, which is considered unacceptable in some institutions where the rapid processing of bone marrow trephine cores is a priority. SD, which is often used for small pieces of bone tissue such as core needle biopsies, is a step in processing bone metastasis specimens after routine processing and embedding. After the tissue is exposed, the block can be removed from the microtome and placed face down in an acidic decalcification solution for 15 to 60 minutes. This surface treatment allows the acidic decalcification solution to penetrate a small distance into the block and dissolve the calcium. The block can then be thoroughly rinsed in water to remove residual acid, chilled, and sectioned. However, it is unclear whether IHC assessments of ER, PR, and HER2 can be reliably performed on SD core needle biopsy samples from bone metastasis patients with BC. In this study, we evaluated the applicability of SD for bone metastases in BC. To our knowledge, this is the first study to assess ER, PR, Ki67, and HER2 in surface-decalcified bone metastasis samples from patients with invasive BC. The findings conclusively demonstrate that SD did not have statistically significant effects on the IHC analysis results of ER, PR, and HER2, whereas it did influence the IHC analysis of Ki67. However, we observed clinically relevant effects of SD; 4 weakly PR-positive cases became negative after SD and 4 HER2-equivocal cases (IHC score 2+) became HER2-negative (IHC score 1+) after SD. As ER and PR are used to confirm a BC diagnosis, the small decrease in signal after SD does not reduce the diagnostic utility. PR is an important marker in BC. Therefore, false-negative PR can lead to fewer falsely stratified patients for hormonal therapy. However, this decrease is only clinically relevant if it affects treatment decision-making; in 4 of these cases, ER remained positive after SD; thus SD did not influence treatment decision-making. Ki67 is occasionally used for prognostication and treatment guidance. However, the substantial loss of Ki67 staining limits the application of Ki67 in the assessment of cell proliferation in decalcified tissue; this is true not only for BC but also for other cancers. Fortunately, Ki67 is less critical in the metastatic setting compared with primary BC. Gertych et al studied the effects of decalcification solution on the expression of Ki67 and found that Ki67 expression decreased from an average of 35% to below the limit of detection after 1 hour due to the effects on HCl on antigenic epitopes. In general, more significant effects are identified with nuclear antigens. In our study, Ki67 decreased from an average of 22% to 13%. In clinical pathology practice, HER2 IHC is the primary test for HER2 status in invasive BC. A score of 0 or 1+ is interpreted as negative, and a score of 3+ is interpreted as positive without further testing. HER2 FISH was used for specimens with ambiguous HER2 IHC scores (2+). HER2 is overexpressed and/or amplified in ~15% to 20% of BCs. Overall, 18.1% (8 of 44) of the samples included in our prospective cohort showed HER2 amplification. After SD, the HER2 IHC score changed from equivocal (2+) to negative (1+) in 4 of the 12 cases. All the HER2-positive (IHC score 3+) cases in the control group remained HER2-positive after SD. Maclary and colleagues reported that heterogeneity in marker distribution progresses from diffuse to multifocal with aberrant loss of expression after decalcification. IHC markers (eg, Her2/neu) with strong expression (initial score of 3+ in nondecalcified tissue) retained their expressions under extended decalcification. We obtained the same results in the present study. The IHC results showed eight HER2 3+ scores, although only 1 FISH amplified case was not observed. The concordance rate of FISH in cases with 3+ IHC scores was 87.5%. This is typically achieved in formalin-fixed primary tumors. Discordance between IHC and FISH results has been described before but is relatively rare in 3+ samples. This discordance was seen in the SD samples but not in the control samples; the influence of the decalcification process could explain this discrepancy. Finally, we evaluated the above findings using samples from 10 patients with SD-treated bone metastases and 10 patients with EDTA-treated bone metastases. The IHC findings were comparable in these cases, validating the SD protocol in real samples. Overall, SD is an alternative decalcification method for bony metastases and allows the accurate assessment of ER, PR, and HER2 (both IHC and FISH) in MBC. SD is considered acceptable in some institutions where priority is given to the rapid processing of bone marrow trephine cores. In any case, adequate formalin fixation before decalcification and daily periodic control of the process is necessary.
SUPPLEMENTARY MATERIAL
|
Hilfreich in der Hausarztpraxis: Neues Faktenblatt zum Impfen | 293c8c23-cb64-4b83-b75d-5561d70af8d9 | 10072913 | Family Medicine[mh] | |
Ultra-high field imaging, plasma markers and autopsy data uncover a specific rostral locus coeruleus vulnerability to hyperphosphorylated tau | f3ee5f78-2e3e-4aff-a127-416b48cc66c1 | 10073793 | Forensic Medicine[mh] | Alzheimer’s disease (AD) pathogenesis starts two to three decades before the emergence of clinical symptoms . Recognition of this protracted evolvement of the disease along with the observation that clinical trials targeting patients in the prodromal or later stages of the disease have not successfully resulted in functional changes, underscore the importance of detecting and implementing interventions for AD at a much earlier asymptomatic stage . In the search for the earliest and meaningful markers of AD-related changes, the brainstem locus coeruleus (LC) has attained significant attention . Autopsy studies reported accumulation of hyperphosphorylated tau in the LC starting as early as age 20 . Detecting tau in the LC using current positron emission tomography (PET) ligands is arduous given the off-target binding to neuromelanin cells that also accumulate in the LC [ – ]. But, developments in MRI methods have enabled the localization of the LC in vivo , and work by our group indicated that the LC MRI-signal can convey information related to tau and risk of AD . We demonstrated that lower MRI-based LC integrity measures correlate with greater tau deposition in early cortical regions in clinically normal individuals starting approximately 54 years of age. In addition, at subthreshold levels of β-amyloid (Aβ)-PET, we observed a steeper decline in memory for individuals with lower LC integrity as compared to those with higher LC integrity, demonstrating a synergistic effect of LC integrity and early AD pathology on downstream clinical symptoms . Importantly, postmortem work indicated that the rostral and middle sections of the LC are possibly susceptible to volume loss at an earlier stage in the disease than the caudal section of the LC . Because the LC does not undergo significant neuronal loss before Braak stage IV , this suggests that these volumetric differences may be tau related . Accordingly, recent 3 T MRI studies also reported that lower integrity in the rostral section of the LC correlated with memory decline, lower cortical thickness and greater risk of mild cognitive impairment (MCI) [ – ]. However, there is inconsistent data on the relationship between tau-PET and regional LC integrity. A study combining data from healthy controls, MCI and AD patients, showed that lower middle-caudal LC integrity was associated with greater temporal lobe MK-6240 tau-PET burden . On the other hand, lower rostral-middle LC integrity in a group of autosomal dominant Alzheimer’s disease (ADAD) patients (symptomatic and asymptomatic) was associated with greater occipito-temporo-parietal FTP-PET burden . These inconsistencies may be due to the merging of distinct disease stages, a more accelerated disease progression in ADAD , or the use of different tau-tracers. Even though MK-6240 is presumed to signal earlier tau aggregations than the FTP tracer, both tracers are less sensitive to earlier non-fibrillar forms of tau . New advances in fluid biomarkers can act as proxies for soluble tau originating from the central nervous system, ushering a new era for the early detection of AD. Based on postmortem and the available imaging data, we hypothesized that specifically the rostral part of the LC would be vulnerable to early AD tau-related processes. This hypothesis can now be evaluated by taking advantage of exciting developments in fluid biomarkers using high sensitivity immunoassay technologies that resulted in several well-validated and robust plasma markers of neurodegeneration (neurofilament light (NfL), total tau) , β-amyloid (Aβ) and various hyperphosphorylated tau (ptau) species . Plasma ptau epitopes (ptau 181 , ptau 217 , ptau 231 ) levels are concordant with autopsy findings, have high sensitivity and specificity to detect tau-pathology on PET, and increase early in the preclinical stage of AD [ – ]. In particular, ptau 231 associated more robustly with tau-PET in asymptomatic individuals and was able to differentiate PET Braak stage 0 from Braak stage I/II and detect incipient amyloid pathology [ – ]. To examine regional vulnerability within the LC, it is important to consider that 3 T MRI methods acquire the LC in anisotropic voxels size (i.e. 0.3x0.3x3mm), which are often resampled to isotropic resolution. Such procedures can introduce noise and distortions in small regions such as the LC. Our efforts in 7 T MRI now provide a unique, detailed window in localization and anatomy of the LC at 0.4 mm near-isotropic voxel size . Here, we related LC integrity using 7 T MRI to different AD plasma markers in asymptomatic individuals across a wide age range, including individuals with preclinical AD. We then examined whether LC integrity and plasma markers are interactively or independently associated with cognition. To validate our in vivo findings, we also compared the proportion of neurons that were tau-positive in the rostral LC versus those in the caudal LC in 77 autopsy cases from the Rush Memory and Aging Project (MAP).
Participants The 7T dataset Cognitively unimpaired individuals (age range = 30–85, 52 females (52.50%)) were recruited from the general community in the most Southern region of the Netherlands. All participants were screened to exclude a history of major psychiatric or neurological disorders, having a history of brain injury of brain surgery, taking medications that may influence cognitive functioning, or being not eligible for ultra-high field MRI-scanning. To exclude individuals with depressive symptoms, we screened all participants using the Hamilton Rating Scale for Depression (all within normal range = 0–12; mean ± SD = 2.23 ± 2.53). All participants received monetary compensation and provided written informed consent. Approval of the experimental protocol was obtained from the local ethical committee of the Faculty of Health, Medicine and Life Sciences at Maastricht University, The Netherlands. The MAP dataset The dataset included 77 participants from the Rush Memory and Aging Project (MAP), an ongoing longitudinal clinical-pathologic studies that started in 1997 . Eligibility criteria included age>55 years, absence of a previous dementia diagnosis and consent to annual clinical evaluation and brain autopsy at death. Participants were recruited from retirement communities, social service agencies and subsidized housing facilities, and individual homes in the Chicago metropolitan region. This sample included individuals for whom detailed LC neuropathology data was available and consisted of individuals with normal cognition ( n = 29), mild cognitive impairment (MCI) ( n = 27) or AD ( n = 21) at their last clinical visit prior to autopsy. Diagnosis was done each year by a neuropsychologist and clinician, and final diagnosis was by a neurologist blinded to postmortem data, based on the National Institute of Neurological and Communicative Disorders and Stroke and the AD and Related Disorders Association (NINCDS/ADRDA) criteria [ – ].The average time interval between last visit and death for these participants was 0.77years (SD = 0.60). All data were de-identified and shared with a Data User Agreement. The study was approved by an institutional review board of Rush University Medical Center. All participants signed an informed consent, an Anatomical Gift Act, and a repository consent which allowed their data to be shared. Structural magnetic resonance imaging (7 T dataset) MRI data acquisition MR scans were performed in a 7 T Magnetom Siemens (Siemens Healthineers, Erlangen, Germany) with a 32-channel head coil (Nova Medical, Wilmington, MA, USA). First, we acquired a Magnetization Prepared 2 Rapid Acquisition Gradient Echoes (MP2RAGE) sequence for whole brain imaging (TR = 5000 ms, TE = 2.47 ms, flip angle = 5 o /3 o , voxel size = 0.7 × 0.7 × 0.7 mm 3 , number of slices = 240). An in-house developed magnetization transfer-weighted turbo flash (MT-TFL) sequence was performed to image the LC at high resolution. The sequence consisted of a multi-shot 3D-readout (TR = 538 ms, TE = 4.08, flip angle=8 o , voxel size=0.4 × 0.4 × 0.5 mm 3 , number of slices = 60) with center-out k-space sampling, preceded by 20 long off-resonant Gaussian sinc pulses (pulse length = 5.12 ms, bandwidth = 250 Hz, B 1 = 0.25 μT). For the MT-TFL sequence, the field-of-view was placed perpendicular to the pons and covered the area between the inferior colliculus and the inferior border of the pons. MRI data processing The MP2RAGE images were processed with FreeSurfer v6.0.0 ( https://surfer.nmr.mgh.harvard.edu ) using the software package’s automated reconstruction protocol as described previously, including expert options for 7 T images . Briefly, each T1-weighted image was subjected to an automated segmentation process involving intensity normalization, skull stripping, segregating left and right hemispheres, removing brainstem and cerebellum, correcting topology defects, defining the borders between gray/white matter and gray/cerebrospinal fluid, and parcellating cortical and subcortical areas. We visually inspected and, if necessary, edited each image. As a control measure for the LC, we extracted the bilateral hippocampal volume and adjusted it for intracranial volume (eTIV) using the following equation : [12pt]{minimal}
$${{{{{{{}}}}}}}\;{{{{{{{}}}}}}}\;{{{{{{{}}}}}}} = \, {{{{{{{}}}}}}}\;{{{{{{{}}}}}}}\;{{{{{{{}}}}}}}\\ - b( {{{{{{{{}}}}}}}\;{{{{{{{}}}}}}}} )$$ Adjusted hippocampal volume = raw hippocampal volume − b eTIV - Mean eTIV where b indicates the regression coefficient when raw hippocampal volume is regressed against eTIV. The MT-TFL images were intensity-normalized in a slice-specific manner using the subject-specific mean intensity of a 10 × 10voxel region-of-interest located in the pontine tegmentum (PT). From these images, a study-specific template of the LC scans was created with an iterative diffeomorphic warp estimate using the buildtemplateparallel.sh script of ANTS, as described previously . The LC was manually delineated on the common space twice by an expert (HJ, spatial correlation r = 0.90) and by another rater (spatial correlation between raters r = 0.83), guided by the voxel intensities and known LC anatomy. This segmented LC mask was applied to the high-resolution individual spatial and intensity-normalized LC images (Supplemental Fig. ). From the template we constructed a surface rendering for visualization purposes of our results. To show regional specificity, we also extracted intensity values from the entire LC. Using the inverse warp, we further obtained subject-specific LC segmentations, extracted the bilateral LC volume and adjusted it for eTIV for control analyses. Plasma markers (7T dataset) Fasted EDTA plasma samples were obtained through venipuncture from the antecubital vein and processed according to the SOP stipulated by the central biobank of Maastricht University Medical Center. Samples were centrifuged at 2000 × g , aliquoted in polypropylene tubes, and stored at −80 °C in our biobank within 60 min of collection. Plasma biomarkers were analyzed in randomized order using ultra-sensitive Single molecule array (Simoa) assays (Quanterix, Inc) for Aβ 42 and Aβ 40 (to create Aβ 42/40 ratio), total tau (Neurology 3-Plex A Advantage Kit), ptau 181 (pTau-181 V2 Advantage Kit), ptau 231 (University Gothenburg) and NfL (NF-light™ Advantage Kit) at the University of Gothenburg (Sweden). Analyses were performed in duplicates using a 1:4 automated dilution protocol for all markers, except for 1:2 dilution protocol for ptau 231 . Analysis of plasma ptau 217 was performed at Lund University (Sweden) using the Meso Scale Discovery (MSD) platform as previously described . The range of values measured is consistent with other studies examining similar cohorts (Table accompanying Supplemental Fig. ). Based on a previously defined cut-off (Aβ 42/40 ratio=0.080 ) in an asymptomatic cohort using the same assay in the same lab (University Gothenburg) we identified 20 Aβ + individuals. APOE genotyping was performed using polymerase chain reaction based on blood sample DNA extraction. Participants’ APOE status was defined as ‘ε4 carrier’ if they carry at least one ε4 allele. 48.5% ( n = 48) carried at least one ε4 allele and/or has elevated plasma Aβ 42/40 values, and hence can be considered at-risk of developing AD. Technicians handling the blood samples were blinded to the participant, cognitive and imaging data, and staff collecting cognitive or imaging data were blinded to blood results. Neuropathological measures (MAP dataset) Immediately after participants’ death, brains were extracted, weighed, and the brainstem and cerebellar hemispheres removed. Both hemispheres and the brainstem were sectioned into 1 cm-thick coronal slabs. One hemisphere was frozen as were select samples of the brainstem; the remained was fixed in 4% paraformaldehyde. The average postmortem time was 7.09 h (SD:3.89 h). Neuronal density (per mm 2 ) and paired helical filaments (PHF) tau tangle density of the LC were examined using immunohistochemistry with a monoclonal anti-tyrosine hydroxylase antibody and an anti-paired helical filaments tau antibody AT8, respectively, each bilaterally at two levels of the LC (rostral-to-middle (“rostral”) and main body or middle-to-caudal (“caudal”)) . To allow for unbiased comparisons, we divided tangle density by the neuronal density for each section of the LC. We selected participants who had neuropathologic data on both sections of the LC ( n = 77). Cortical Aβ load was quantified as percent area occupied by Aβ, labeled with a N-terminal directed monoclonal antibody, which identifies both the 1–40 and 1–42 length Aβ fragments, while cortical PHF tau tangles, were quantified as the density of paired helical filament tau tangles with an antibody specific for phosphorylated tau, AT8 (density per square millimeter) across 8 regions . Modified Bielschowsky silver quantification was used for Braak scoring of neurofibrillary pathology and Consortium to Establish a Registry for AD (CERAD) scoring of neuritic plaques. Using this information, the likelihood of AD pathology was identified according to the modified National Institute of Aging (NIA)-Reagan diagnosis of AD and grouped into not present (no or low likelihood) and present (intermediate and high likelihood). This evaluation is performed independent of clinical information, including the diagnosis . Neuropsychological assessment (7T dataset) The preclinical Alzheimer’s cognitive composite (PACC) was designed to be sensitive to cognitive change among individuals with preclinical AD, and consists of the average of z-transformed scores on the Mini-Mental-State Examination, Logical Memory Delayed recall test, Digit Symbol Substitution Test, Free and Cued Selective Reminding test (free and total recall) and was later expanded to also include the category fluency . We created a PACC-score based on the average of the z-scores of the performance on cognitive tests available in this cohort: Mini-Mental-State Examination, Digit Symbol Substitution Test, Rey-Auditory Verbal Learning Test (total and delayed free recall) and the category fluency test. Other collected behavioral measures were not analyzed for this study. Statistical analyses Statistical analyses were performed in R (version 4.1.2, http://www.r-project.org/ ). Group characteristics are represented in mean and standard deviation. Associations between the plasma markers, and age, sex, or APOE-status were examined with robust regression using the Huber M estimator. Robust regression is suited to handle outliers that are often observed in the skewed distributions of biomarker data. It effectively downweighs their influence on the coefficients and regression fit and therefore, allows including individuals with more extreme (or abnormal) values, who are less prevalent in healthy populations but reflect the normal population. Voxel-wise robust regressions between LC intensity and each plasma marker, or their interactive or additive effects with plasma Aβ 42/40 were corrected for age, sex and APOE- ε4 status and adjusted for multiple testing using the probabilistic Threshold Free Cluster Enhancement at two-sided p < 0.001 per analysis (not across analyses). From each plasma marker cluster on the LC surface, we extracted subject-specific LC intensity values for further analyses. These plasma-based analyses were repeated for the entire LC intensity, bilateral LC volume, and hippocampal volume to examine the specificity of our findings. To examine whether associations between the plasma markers and LC intensity in their respective clusters occurred within a specific age-range, we ran sliding window analyses with 20-year age bins with bootstrapped 95% confidence intervals (5,000 iterations) . Finally, using robust regression we examined whether the plasma markers that were significant from the previous analyses and the extracted LC intensity from the plasma-specific clusters were interactively or independently associated with PACC performance, with age, sex, education and APOE- ε4 status as covariates. We adjusted here for multiple testing using Bonferroni correction. Using the Johnson-Neyman approach we determined at which value the plasma marker modified the relationship between LC intensity and PACC. The threshold for statistical significance was set at two-sided p < 0.05, unless otherwise specified. For the MAP dataset, we performed a repeated measures ANOVA with proportion of tangles in each section of the LC as within-subject factor, and age, sex, APOE- ε4 status, and postmortem interval as covariates. To test if these effects are independent of AD pathology, we also added cortical Aβ or NIA-Reagan diagnosis of AD groups as covariate. In the second step, we included diagnosis (i.e., cognitively unimpaired, MCI or AD) as between-subject factor and interactions with LC sections, to determine whether the regional distribution of the proportion of LC tangles varied as a function of disease stage. Post-hoc adjustment for multiple comparisons was performed with Tukey’s HSD.
The 7T dataset Cognitively unimpaired individuals (age range = 30–85, 52 females (52.50%)) were recruited from the general community in the most Southern region of the Netherlands. All participants were screened to exclude a history of major psychiatric or neurological disorders, having a history of brain injury of brain surgery, taking medications that may influence cognitive functioning, or being not eligible for ultra-high field MRI-scanning. To exclude individuals with depressive symptoms, we screened all participants using the Hamilton Rating Scale for Depression (all within normal range = 0–12; mean ± SD = 2.23 ± 2.53). All participants received monetary compensation and provided written informed consent. Approval of the experimental protocol was obtained from the local ethical committee of the Faculty of Health, Medicine and Life Sciences at Maastricht University, The Netherlands. The MAP dataset The dataset included 77 participants from the Rush Memory and Aging Project (MAP), an ongoing longitudinal clinical-pathologic studies that started in 1997 . Eligibility criteria included age>55 years, absence of a previous dementia diagnosis and consent to annual clinical evaluation and brain autopsy at death. Participants were recruited from retirement communities, social service agencies and subsidized housing facilities, and individual homes in the Chicago metropolitan region. This sample included individuals for whom detailed LC neuropathology data was available and consisted of individuals with normal cognition ( n = 29), mild cognitive impairment (MCI) ( n = 27) or AD ( n = 21) at their last clinical visit prior to autopsy. Diagnosis was done each year by a neuropsychologist and clinician, and final diagnosis was by a neurologist blinded to postmortem data, based on the National Institute of Neurological and Communicative Disorders and Stroke and the AD and Related Disorders Association (NINCDS/ADRDA) criteria [ – ].The average time interval between last visit and death for these participants was 0.77years (SD = 0.60). All data were de-identified and shared with a Data User Agreement. The study was approved by an institutional review board of Rush University Medical Center. All participants signed an informed consent, an Anatomical Gift Act, and a repository consent which allowed their data to be shared.
Cognitively unimpaired individuals (age range = 30–85, 52 females (52.50%)) were recruited from the general community in the most Southern region of the Netherlands. All participants were screened to exclude a history of major psychiatric or neurological disorders, having a history of brain injury of brain surgery, taking medications that may influence cognitive functioning, or being not eligible for ultra-high field MRI-scanning. To exclude individuals with depressive symptoms, we screened all participants using the Hamilton Rating Scale for Depression (all within normal range = 0–12; mean ± SD = 2.23 ± 2.53). All participants received monetary compensation and provided written informed consent. Approval of the experimental protocol was obtained from the local ethical committee of the Faculty of Health, Medicine and Life Sciences at Maastricht University, The Netherlands.
The dataset included 77 participants from the Rush Memory and Aging Project (MAP), an ongoing longitudinal clinical-pathologic studies that started in 1997 . Eligibility criteria included age>55 years, absence of a previous dementia diagnosis and consent to annual clinical evaluation and brain autopsy at death. Participants were recruited from retirement communities, social service agencies and subsidized housing facilities, and individual homes in the Chicago metropolitan region. This sample included individuals for whom detailed LC neuropathology data was available and consisted of individuals with normal cognition ( n = 29), mild cognitive impairment (MCI) ( n = 27) or AD ( n = 21) at their last clinical visit prior to autopsy. Diagnosis was done each year by a neuropsychologist and clinician, and final diagnosis was by a neurologist blinded to postmortem data, based on the National Institute of Neurological and Communicative Disorders and Stroke and the AD and Related Disorders Association (NINCDS/ADRDA) criteria [ – ].The average time interval between last visit and death for these participants was 0.77years (SD = 0.60). All data were de-identified and shared with a Data User Agreement. The study was approved by an institutional review board of Rush University Medical Center. All participants signed an informed consent, an Anatomical Gift Act, and a repository consent which allowed their data to be shared.
MRI data acquisition MR scans were performed in a 7 T Magnetom Siemens (Siemens Healthineers, Erlangen, Germany) with a 32-channel head coil (Nova Medical, Wilmington, MA, USA). First, we acquired a Magnetization Prepared 2 Rapid Acquisition Gradient Echoes (MP2RAGE) sequence for whole brain imaging (TR = 5000 ms, TE = 2.47 ms, flip angle = 5 o /3 o , voxel size = 0.7 × 0.7 × 0.7 mm 3 , number of slices = 240). An in-house developed magnetization transfer-weighted turbo flash (MT-TFL) sequence was performed to image the LC at high resolution. The sequence consisted of a multi-shot 3D-readout (TR = 538 ms, TE = 4.08, flip angle=8 o , voxel size=0.4 × 0.4 × 0.5 mm 3 , number of slices = 60) with center-out k-space sampling, preceded by 20 long off-resonant Gaussian sinc pulses (pulse length = 5.12 ms, bandwidth = 250 Hz, B 1 = 0.25 μT). For the MT-TFL sequence, the field-of-view was placed perpendicular to the pons and covered the area between the inferior colliculus and the inferior border of the pons. MRI data processing The MP2RAGE images were processed with FreeSurfer v6.0.0 ( https://surfer.nmr.mgh.harvard.edu ) using the software package’s automated reconstruction protocol as described previously, including expert options for 7 T images . Briefly, each T1-weighted image was subjected to an automated segmentation process involving intensity normalization, skull stripping, segregating left and right hemispheres, removing brainstem and cerebellum, correcting topology defects, defining the borders between gray/white matter and gray/cerebrospinal fluid, and parcellating cortical and subcortical areas. We visually inspected and, if necessary, edited each image. As a control measure for the LC, we extracted the bilateral hippocampal volume and adjusted it for intracranial volume (eTIV) using the following equation : [12pt]{minimal}
$${{{{{{{}}}}}}}\;{{{{{{{}}}}}}}\;{{{{{{{}}}}}}} = \, {{{{{{{}}}}}}}\;{{{{{{{}}}}}}}\;{{{{{{{}}}}}}}\\ - b( {{{{{{{{}}}}}}}\;{{{{{{{}}}}}}}} )$$ Adjusted hippocampal volume = raw hippocampal volume − b eTIV - Mean eTIV where b indicates the regression coefficient when raw hippocampal volume is regressed against eTIV. The MT-TFL images were intensity-normalized in a slice-specific manner using the subject-specific mean intensity of a 10 × 10voxel region-of-interest located in the pontine tegmentum (PT). From these images, a study-specific template of the LC scans was created with an iterative diffeomorphic warp estimate using the buildtemplateparallel.sh script of ANTS, as described previously . The LC was manually delineated on the common space twice by an expert (HJ, spatial correlation r = 0.90) and by another rater (spatial correlation between raters r = 0.83), guided by the voxel intensities and known LC anatomy. This segmented LC mask was applied to the high-resolution individual spatial and intensity-normalized LC images (Supplemental Fig. ). From the template we constructed a surface rendering for visualization purposes of our results. To show regional specificity, we also extracted intensity values from the entire LC. Using the inverse warp, we further obtained subject-specific LC segmentations, extracted the bilateral LC volume and adjusted it for eTIV for control analyses.
MR scans were performed in a 7 T Magnetom Siemens (Siemens Healthineers, Erlangen, Germany) with a 32-channel head coil (Nova Medical, Wilmington, MA, USA). First, we acquired a Magnetization Prepared 2 Rapid Acquisition Gradient Echoes (MP2RAGE) sequence for whole brain imaging (TR = 5000 ms, TE = 2.47 ms, flip angle = 5 o /3 o , voxel size = 0.7 × 0.7 × 0.7 mm 3 , number of slices = 240). An in-house developed magnetization transfer-weighted turbo flash (MT-TFL) sequence was performed to image the LC at high resolution. The sequence consisted of a multi-shot 3D-readout (TR = 538 ms, TE = 4.08, flip angle=8 o , voxel size=0.4 × 0.4 × 0.5 mm 3 , number of slices = 60) with center-out k-space sampling, preceded by 20 long off-resonant Gaussian sinc pulses (pulse length = 5.12 ms, bandwidth = 250 Hz, B 1 = 0.25 μT). For the MT-TFL sequence, the field-of-view was placed perpendicular to the pons and covered the area between the inferior colliculus and the inferior border of the pons.
The MP2RAGE images were processed with FreeSurfer v6.0.0 ( https://surfer.nmr.mgh.harvard.edu ) using the software package’s automated reconstruction protocol as described previously, including expert options for 7 T images . Briefly, each T1-weighted image was subjected to an automated segmentation process involving intensity normalization, skull stripping, segregating left and right hemispheres, removing brainstem and cerebellum, correcting topology defects, defining the borders between gray/white matter and gray/cerebrospinal fluid, and parcellating cortical and subcortical areas. We visually inspected and, if necessary, edited each image. As a control measure for the LC, we extracted the bilateral hippocampal volume and adjusted it for intracranial volume (eTIV) using the following equation : [12pt]{minimal}
$${{{{{{{}}}}}}}\;{{{{{{{}}}}}}}\;{{{{{{{}}}}}}} = \, {{{{{{{}}}}}}}\;{{{{{{{}}}}}}}\;{{{{{{{}}}}}}}\\ - b( {{{{{{{{}}}}}}}\;{{{{{{{}}}}}}}} )$$ Adjusted hippocampal volume = raw hippocampal volume − b eTIV - Mean eTIV where b indicates the regression coefficient when raw hippocampal volume is regressed against eTIV. The MT-TFL images were intensity-normalized in a slice-specific manner using the subject-specific mean intensity of a 10 × 10voxel region-of-interest located in the pontine tegmentum (PT). From these images, a study-specific template of the LC scans was created with an iterative diffeomorphic warp estimate using the buildtemplateparallel.sh script of ANTS, as described previously . The LC was manually delineated on the common space twice by an expert (HJ, spatial correlation r = 0.90) and by another rater (spatial correlation between raters r = 0.83), guided by the voxel intensities and known LC anatomy. This segmented LC mask was applied to the high-resolution individual spatial and intensity-normalized LC images (Supplemental Fig. ). From the template we constructed a surface rendering for visualization purposes of our results. To show regional specificity, we also extracted intensity values from the entire LC. Using the inverse warp, we further obtained subject-specific LC segmentations, extracted the bilateral LC volume and adjusted it for eTIV for control analyses.
Fasted EDTA plasma samples were obtained through venipuncture from the antecubital vein and processed according to the SOP stipulated by the central biobank of Maastricht University Medical Center. Samples were centrifuged at 2000 × g , aliquoted in polypropylene tubes, and stored at −80 °C in our biobank within 60 min of collection. Plasma biomarkers were analyzed in randomized order using ultra-sensitive Single molecule array (Simoa) assays (Quanterix, Inc) for Aβ 42 and Aβ 40 (to create Aβ 42/40 ratio), total tau (Neurology 3-Plex A Advantage Kit), ptau 181 (pTau-181 V2 Advantage Kit), ptau 231 (University Gothenburg) and NfL (NF-light™ Advantage Kit) at the University of Gothenburg (Sweden). Analyses were performed in duplicates using a 1:4 automated dilution protocol for all markers, except for 1:2 dilution protocol for ptau 231 . Analysis of plasma ptau 217 was performed at Lund University (Sweden) using the Meso Scale Discovery (MSD) platform as previously described . The range of values measured is consistent with other studies examining similar cohorts (Table accompanying Supplemental Fig. ). Based on a previously defined cut-off (Aβ 42/40 ratio=0.080 ) in an asymptomatic cohort using the same assay in the same lab (University Gothenburg) we identified 20 Aβ + individuals. APOE genotyping was performed using polymerase chain reaction based on blood sample DNA extraction. Participants’ APOE status was defined as ‘ε4 carrier’ if they carry at least one ε4 allele. 48.5% ( n = 48) carried at least one ε4 allele and/or has elevated plasma Aβ 42/40 values, and hence can be considered at-risk of developing AD. Technicians handling the blood samples were blinded to the participant, cognitive and imaging data, and staff collecting cognitive or imaging data were blinded to blood results.
Immediately after participants’ death, brains were extracted, weighed, and the brainstem and cerebellar hemispheres removed. Both hemispheres and the brainstem were sectioned into 1 cm-thick coronal slabs. One hemisphere was frozen as were select samples of the brainstem; the remained was fixed in 4% paraformaldehyde. The average postmortem time was 7.09 h (SD:3.89 h). Neuronal density (per mm 2 ) and paired helical filaments (PHF) tau tangle density of the LC were examined using immunohistochemistry with a monoclonal anti-tyrosine hydroxylase antibody and an anti-paired helical filaments tau antibody AT8, respectively, each bilaterally at two levels of the LC (rostral-to-middle (“rostral”) and main body or middle-to-caudal (“caudal”)) . To allow for unbiased comparisons, we divided tangle density by the neuronal density for each section of the LC. We selected participants who had neuropathologic data on both sections of the LC ( n = 77). Cortical Aβ load was quantified as percent area occupied by Aβ, labeled with a N-terminal directed monoclonal antibody, which identifies both the 1–40 and 1–42 length Aβ fragments, while cortical PHF tau tangles, were quantified as the density of paired helical filament tau tangles with an antibody specific for phosphorylated tau, AT8 (density per square millimeter) across 8 regions . Modified Bielschowsky silver quantification was used for Braak scoring of neurofibrillary pathology and Consortium to Establish a Registry for AD (CERAD) scoring of neuritic plaques. Using this information, the likelihood of AD pathology was identified according to the modified National Institute of Aging (NIA)-Reagan diagnosis of AD and grouped into not present (no or low likelihood) and present (intermediate and high likelihood). This evaluation is performed independent of clinical information, including the diagnosis .
The preclinical Alzheimer’s cognitive composite (PACC) was designed to be sensitive to cognitive change among individuals with preclinical AD, and consists of the average of z-transformed scores on the Mini-Mental-State Examination, Logical Memory Delayed recall test, Digit Symbol Substitution Test, Free and Cued Selective Reminding test (free and total recall) and was later expanded to also include the category fluency . We created a PACC-score based on the average of the z-scores of the performance on cognitive tests available in this cohort: Mini-Mental-State Examination, Digit Symbol Substitution Test, Rey-Auditory Verbal Learning Test (total and delayed free recall) and the category fluency test. Other collected behavioral measures were not analyzed for this study.
Statistical analyses were performed in R (version 4.1.2, http://www.r-project.org/ ). Group characteristics are represented in mean and standard deviation. Associations between the plasma markers, and age, sex, or APOE-status were examined with robust regression using the Huber M estimator. Robust regression is suited to handle outliers that are often observed in the skewed distributions of biomarker data. It effectively downweighs their influence on the coefficients and regression fit and therefore, allows including individuals with more extreme (or abnormal) values, who are less prevalent in healthy populations but reflect the normal population. Voxel-wise robust regressions between LC intensity and each plasma marker, or their interactive or additive effects with plasma Aβ 42/40 were corrected for age, sex and APOE- ε4 status and adjusted for multiple testing using the probabilistic Threshold Free Cluster Enhancement at two-sided p < 0.001 per analysis (not across analyses). From each plasma marker cluster on the LC surface, we extracted subject-specific LC intensity values for further analyses. These plasma-based analyses were repeated for the entire LC intensity, bilateral LC volume, and hippocampal volume to examine the specificity of our findings. To examine whether associations between the plasma markers and LC intensity in their respective clusters occurred within a specific age-range, we ran sliding window analyses with 20-year age bins with bootstrapped 95% confidence intervals (5,000 iterations) . Finally, using robust regression we examined whether the plasma markers that were significant from the previous analyses and the extracted LC intensity from the plasma-specific clusters were interactively or independently associated with PACC performance, with age, sex, education and APOE- ε4 status as covariates. We adjusted here for multiple testing using Bonferroni correction. Using the Johnson-Neyman approach we determined at which value the plasma marker modified the relationship between LC intensity and PACC. The threshold for statistical significance was set at two-sided p < 0.05, unless otherwise specified. For the MAP dataset, we performed a repeated measures ANOVA with proportion of tangles in each section of the LC as within-subject factor, and age, sex, APOE- ε4 status, and postmortem interval as covariates. To test if these effects are independent of AD pathology, we also added cortical Aβ or NIA-Reagan diagnosis of AD groups as covariate. In the second step, we included diagnosis (i.e., cognitively unimpaired, MCI or AD) as between-subject factor and interactions with LC sections, to determine whether the regional distribution of the proportion of LC tangles varied as a function of disease stage. Post-hoc adjustment for multiple comparisons was performed with Tukey’s HSD.
Participant characteristics In the 7 T dataset, the mean age of the participants was 59.93 years (range 30–85 years), 52 were female (52.50%) and 37 (37%) carried at least one APOE- ε4 allele (Table ). All participants were cognitively healthy (mean MMSE-score:28.98). In the MAP sample, the mean age of the participants was 88.59 years, 56 were female (72.72%), 10 carried at least one APOE- ε4 allele (13%), and 47 individuals (61%) showed evidence of AD pathology at autopsy according to the NIA-Reagan AD criteria. Histograms depicting age distributions of each cohort are provided in Supplemental Fig. Correlations of plasma biomarkers with demographics First-order correlations between all the plasma markers, and age, sex or APOE are shown in Supplementary Tables , and Supplemental Fig. . In the following analyses, age analyses were adjusted for sex, and the sex analyses were adjusted for age. Older age was associated with lower Aβ 42/40 (t = −4.63, p < 0.001) and higher ptau 181 (t = 3.25, p = 0.002) and NfL (t = 9.67, p < 0.001). We detected no age-relationship for total tau (t = −1.73, p = 0.08), ptau 217 (t = 0.40, p = 0.69) or ptau 231 (t = 1.17, p = 0.25). Very weak or a lack of significant correlations between the ptau markers and Aβ 42/40 has been reported in similar cohorts . Females exhibited higher total tau (t = −2.23, p = 0.023), while males displayed higher ptau 181 (t = 3.21, p = 0.002). These sex differences remained significant when controlling for APOE- ε4 status and Aβ 42/40 (t = −2.22, p = 0.03 and t = 3.82, p = 0.003, respectively). There were no sex differences in Aβ 42 / 40 (t = 0.22, p = 0.83), NfL (t = 0.47, p = 0.64), ptau 217 (t = −0.81, p = 0.43) or ptau 231 (t = 1.47, p = 0.15). Those who carry at least one ε4 allele displayed lower Aβ 42/40 (t = −2.27, p = 0.025). We observed no differences between ε4-carriers and non-carriers for other biomarkers (total tau: p = 0.79, NfL: p = 0.95, ptau 181 : p = 0.18, ptau 217 : p = 0.15, ptau 231 : p = 0.66, Fig. ). Voxel-wise relationships between locus coeruleus intensity and plasma biomarkers Age, sex and APOE- ε4 status were included as covariates in the voxel-wise regression analyses, given their significant associations with the plasma biomarkers. Voxel-wise analyses revealed that higher ptau 231 was associated with lower LC intensity in bilateral dorso-rostral clusters (p TFCE <0.001). Ptau 231 had the largest cluster-size (160 voxels). Ptau 181 (44 voxels), ptau 217 (37 voxels) and t-tau (49 voxels) correlated negatively with LC intensity in right dorso-rostral clusters. Small clusters of negative associations between LC intensity and NfL (51 voxels) were distributed across the length of the LC, and more prominent in the middle-to-caudal part. LC intensity was not associated with Aβ 42/40 (Fig. ). Adjusting for Aβ 42/40 did not change these associations. We did not find interactions between Aβ 42/40 and any of the other plasma markers on LC intensity. Consistent with this, we observed similar negative slopes between the ptau markers, in particular the ptau 231 , and intensity in the LC clusters across Aβ + and Aβ- groups (Supplemental Fig. ). The control brain measures, hippocampal volume, bilateral LC volume, and average entire LC intensity, were not associated with any of the plasma markers (Supplemental Table ). Age-window of the relationship between locus coeruleus intensity and plasma biomarkers Bootstrapped sliding window analyses showed that the relationship between ptau 181 and LC intensity in its cluster was significant for individuals older than 60 (Fig. ), for ptau 217 starting from 60.5 years (Fig. ), and for ptau 231 the relationship with LC intensity was significant from 55.5 years and older (Fig. ). Of the individuals older than 55.5 years, 41% (25/61) was considered at increased risk for AD-related processes based on Aβ < 0.08 or presence of at least one APOE ε4 allele. For NfL and total tau, no robust age windows were detected. We confirmed these patterns in voxel-wise sliding window analyses: associations between ptau 181 and rostral dorsal LC intensity emerged in the early 60 s, while for ptau 217 , we detected clusters in the dorsal rostro-middle LC from the early-mid 60 s. The earliest age-associations with LC intensity were found for ptau 231 , revealing an anatomic pattern in the ventral and dorsal rostral parts of the LC starting from late 40 s and later associations (60’s) in the middle-caudal section of the LC (Fig. ). Independent and interactive relationships of plasma biomarkers and LC intensity on PACC Based on the previous results, we focused the cognitive analyses on the ptau biomarkers. The interaction between ptau 181 and its associated LC intensity cluster on PACC performance was significant at trend-level ( p = 0.06). There were no independent effects of ptau 181 and LC intensity on PACC. We found no interactive or independent relationships of ptau 217 with LC intensity in predicting PACC performance (Supplemental Table ). At higher levels of ptau 231 , in particular above 15.69 pg/ml, lower LC intensity was associated with worse PACC performance ( p = 0.016 or p BONF = 0.048, Fig. ). Regional specificity of tau in the LC in the autopsy dataset In the MAP dataset, normalized density of LC tangles in the rostral LC was significantly greater than those in the caudal LC (F (1,72) = 13.14, p < 0.001). This difference remained present when controlling for cortical Aβ (F (1,71) =13.39, p < 0.001) or NIA-Reagan diagnosis of AD groups (F (1,71) = 12.96, p < 0.001, Fig. ). Upon examining the interaction between LC section and the diagnostic groups, we found no significant difference in rostral versus caudal LC tangle density across the groups (F (1,70) = 0.07, p = 0.94, Supplemental Fig. ). Across all groups, rostral LC tangle density was higher than caudal LC tangle density (contrast of main effect LC section:β=0.50, t = 2.21, p = 0.031; mean rostro-caudal LC tangle density difference for cognitively normal = 0.05 ± 0.13, for MCI = 0.05 ± 0.09 and for AD = 0.05 ± 0.15).
In the 7 T dataset, the mean age of the participants was 59.93 years (range 30–85 years), 52 were female (52.50%) and 37 (37%) carried at least one APOE- ε4 allele (Table ). All participants were cognitively healthy (mean MMSE-score:28.98). In the MAP sample, the mean age of the participants was 88.59 years, 56 were female (72.72%), 10 carried at least one APOE- ε4 allele (13%), and 47 individuals (61%) showed evidence of AD pathology at autopsy according to the NIA-Reagan AD criteria. Histograms depicting age distributions of each cohort are provided in Supplemental Fig.
First-order correlations between all the plasma markers, and age, sex or APOE are shown in Supplementary Tables , and Supplemental Fig. . In the following analyses, age analyses were adjusted for sex, and the sex analyses were adjusted for age. Older age was associated with lower Aβ 42/40 (t = −4.63, p < 0.001) and higher ptau 181 (t = 3.25, p = 0.002) and NfL (t = 9.67, p < 0.001). We detected no age-relationship for total tau (t = −1.73, p = 0.08), ptau 217 (t = 0.40, p = 0.69) or ptau 231 (t = 1.17, p = 0.25). Very weak or a lack of significant correlations between the ptau markers and Aβ 42/40 has been reported in similar cohorts . Females exhibited higher total tau (t = −2.23, p = 0.023), while males displayed higher ptau 181 (t = 3.21, p = 0.002). These sex differences remained significant when controlling for APOE- ε4 status and Aβ 42/40 (t = −2.22, p = 0.03 and t = 3.82, p = 0.003, respectively). There were no sex differences in Aβ 42 / 40 (t = 0.22, p = 0.83), NfL (t = 0.47, p = 0.64), ptau 217 (t = −0.81, p = 0.43) or ptau 231 (t = 1.47, p = 0.15). Those who carry at least one ε4 allele displayed lower Aβ 42/40 (t = −2.27, p = 0.025). We observed no differences between ε4-carriers and non-carriers for other biomarkers (total tau: p = 0.79, NfL: p = 0.95, ptau 181 : p = 0.18, ptau 217 : p = 0.15, ptau 231 : p = 0.66, Fig. ).
Age, sex and APOE- ε4 status were included as covariates in the voxel-wise regression analyses, given their significant associations with the plasma biomarkers. Voxel-wise analyses revealed that higher ptau 231 was associated with lower LC intensity in bilateral dorso-rostral clusters (p TFCE <0.001). Ptau 231 had the largest cluster-size (160 voxels). Ptau 181 (44 voxels), ptau 217 (37 voxels) and t-tau (49 voxels) correlated negatively with LC intensity in right dorso-rostral clusters. Small clusters of negative associations between LC intensity and NfL (51 voxels) were distributed across the length of the LC, and more prominent in the middle-to-caudal part. LC intensity was not associated with Aβ 42/40 (Fig. ). Adjusting for Aβ 42/40 did not change these associations. We did not find interactions between Aβ 42/40 and any of the other plasma markers on LC intensity. Consistent with this, we observed similar negative slopes between the ptau markers, in particular the ptau 231 , and intensity in the LC clusters across Aβ + and Aβ- groups (Supplemental Fig. ). The control brain measures, hippocampal volume, bilateral LC volume, and average entire LC intensity, were not associated with any of the plasma markers (Supplemental Table ).
Bootstrapped sliding window analyses showed that the relationship between ptau 181 and LC intensity in its cluster was significant for individuals older than 60 (Fig. ), for ptau 217 starting from 60.5 years (Fig. ), and for ptau 231 the relationship with LC intensity was significant from 55.5 years and older (Fig. ). Of the individuals older than 55.5 years, 41% (25/61) was considered at increased risk for AD-related processes based on Aβ < 0.08 or presence of at least one APOE ε4 allele. For NfL and total tau, no robust age windows were detected. We confirmed these patterns in voxel-wise sliding window analyses: associations between ptau 181 and rostral dorsal LC intensity emerged in the early 60 s, while for ptau 217 , we detected clusters in the dorsal rostro-middle LC from the early-mid 60 s. The earliest age-associations with LC intensity were found for ptau 231 , revealing an anatomic pattern in the ventral and dorsal rostral parts of the LC starting from late 40 s and later associations (60’s) in the middle-caudal section of the LC (Fig. ).
Based on the previous results, we focused the cognitive analyses on the ptau biomarkers. The interaction between ptau 181 and its associated LC intensity cluster on PACC performance was significant at trend-level ( p = 0.06). There were no independent effects of ptau 181 and LC intensity on PACC. We found no interactive or independent relationships of ptau 217 with LC intensity in predicting PACC performance (Supplemental Table ). At higher levels of ptau 231 , in particular above 15.69 pg/ml, lower LC intensity was associated with worse PACC performance ( p = 0.016 or p BONF = 0.048, Fig. ).
In the MAP dataset, normalized density of LC tangles in the rostral LC was significantly greater than those in the caudal LC (F (1,72) = 13.14, p < 0.001). This difference remained present when controlling for cortical Aβ (F (1,71) =13.39, p < 0.001) or NIA-Reagan diagnosis of AD groups (F (1,71) = 12.96, p < 0.001, Fig. ). Upon examining the interaction between LC section and the diagnostic groups, we found no significant difference in rostral versus caudal LC tangle density across the groups (F (1,70) = 0.07, p = 0.94, Supplemental Fig. ). Across all groups, rostral LC tangle density was higher than caudal LC tangle density (contrast of main effect LC section:β=0.50, t = 2.21, p = 0.031; mean rostro-caudal LC tangle density difference for cognitively normal = 0.05 ± 0.13, for MCI = 0.05 ± 0.09 and for AD = 0.05 ± 0.15).
As clinical trials targeting Aβ provided marginal clinical effects, the AD field has oriented its focus on earlier time windows in the disease process . These earlier windows have compelled researchers to focus on tau pathology, which is more closely related to the emergence of clinical symptoms compared to Aβ . The LC is one of the earliest brain regions accumulating hyperphosphorylated tau and recent work emphasized its potential as early marker of future AD-related processes, including tau and cognitive decline [ , , ]. Here we aimed to examine whether different tau species and markers of neurodegenerations correlate to specific anatomic patterns of integrity within the LC across the adult lifespan. We found that lower integrity in bilateral dorso-rostral clusters of the LC was associated with greater ptau 231 concentrations, starting from midlife (~55 years). Higher ptau 217 and ptau 181 levels were associated with lower LC integrity in smaller right dorso-rostral clusters, starting from age 60. Similarly, the autopsy data revealed a higher tangle density in the rostral compared to the caudal part of the LC, independent of AD likelihood. Furthermore, lower PACC scores were associated with lower rostral LC integrity, particularly for individuals with higher plasma ptau 231 . By contrast, higher NfL was associated with predominantly lower middle-to-caudle LC integrity, independent of age. As tau phosphorylation at threonine 231 is one of the earliest events in the phosphorylation cascade hindering tubulin assembly , these findings illustrate that changes in rostral LC intensity can reflect processes related to very early tau aggregation, whereas integrity changes towards the caudal direction in the LC may reflect more nonspecific neurodegeneration. The fine-grained topography of correlations between LC integrity and ptau markers, demonstrating a predilection of AD-related processes for rostral regions of the LC, is consistent with autopsy reports . Work by Theofilas and colleagues reported 8.40% volume loss – not neuronal loss – per Braak stage, starting in Braak stage 0 for the rostral and middle LC . Consistent with our previous work and with the observation that the caudal LC contains fewer AD-related changes , we found that indicators of nonspecific neurodegeneration were associated with lower caudal LC integrity. Ultra-high field imaging of the LC in Parkinson’s disease reported lower integrity in the caudal part in patients compared to controls . We speculate that these observations emphasize a rosto-caudal gradient of vulnerabilities to specific pathologic events within the LC that may also be disease-specific. In accordance with this hypothesis, we found that rostral-middle parts of the LC are more vulnerable to hyperphosphorylated tau whereas the caudal part may be more affected by non-specific neurodegeneration. With respect to the plasma ptau epitopes, we found that ptau 231 was more robustly and earlier in life associated with LC integrity compared with the ptau 217 and ptau 181 markers. Neuropathology studies demonstrated that tau phosphorylation at threonine 231 may signal features of tau that precede the pre-neurofibrillary tangle and reflect the earliest events in the phosphorylation cascade hindering tubulin assembly . In the ALFA + cohort of asymptomatic individuals, ptau231 changes preceded changes in other markers and predicted increases in Aβ-PET burden among individuals with no PET-evidence of Aβ-pathology as well as in the younger group (45–60 years) . Given the concordance between these spatial patterns in the voxel-wise analyses and the higher rostral vulnerability in the autopsy data, our previous report on converging patterns of associations between LC integrity and tau-PET with autopsy data , and the similarities between our cohort and the ALFA + cohort, our findings here support the idea that LC integrity measures may capture very early aberrant tau-related processes [ , , ]. The finding that ptau 231 may be an early marker fits with recent work showing that ptau 231 was able to differentiate between PET-derived Braak stage 0 and Braak stage I-II, whereas ptau 181 was not . Furthermore, elevations in ptau 231 can be detected at lower amounts of AD-type pathology compared to ptau 181 . Importantly, similar observations have been made for cerebrospinal fluid ptau markers . Plasma ptau 181 correlates well with tau-PET, can detect elevated Aβ-PET and discriminate AD patients from non-AD patients . Previous work demonstrated that plasma ptau 217 detects AD-related processes slightly earlier than ptau 181 as it was better in predicting elevated entorhinal tau . We did not see a remarkable difference between the topography on the LC for ptau 217 and ptau 181 . Furthermore, our sliding window analyses revealed a similar age-range (>60 years) during which both ptau 217 and ptau 181 have their strongest relationship with LC integrity. Our plasma Aβ 42/40 marker did not correlate with the ptau markers, and this lack of correlations between the Aβ and ptau markers is consistent with observations in the ALFA + cohort . Within the Aβ + and Aβ- groups, we observed similar associations between the ptau markers and LC integrity, in particular for ptau 231 , and we posit that these ptau 231 -related LC changes may capture early processes on the AD pathophysiologic continuum. Importantly, the sensitivity of detecting early tau-related LC changes is higher when honing in on the rostral LC instead of the entire average LC or LC volume. Furthermore, we did not observe tau-related associations with hippocampal volume, further emphasizing the specificity of the rostral LC as a potential marker to detect very early AD-related processes. This rostro-caudal gradient of tau-related vulnerability within the LC also aligns with the topographic arrangement of axonal projections and the cytoarchitecture of the LC . Projections of the LC are not randomly organized, as neurons in the rostro-middle sections of the LC preferentially project to the higher-order cognitive regions of the cortex, while the majority of the neurons in the caudal section of the LC projects to the cerebellum and spinal cord . Furthermore, the rostral and caudal parts of the LC have distinct cell types and cell sizes organized in clusters, supporting functional heterogeneity . The dorso-rostral LC contains almost uniquely densely packed small fusiform neurons, which have long, thin dendrites projecting to the hippocampus and cortex, and are more heavily pigmented and more vulnerable to pathology than the larger cells. The caudal part of the LC contains mostly large multipolar neurons [ – ]. This distribution may also support the rostral LC’s critical contribution in modulating cognition, because fusiform neurons tend to have shorter action potentials with larger amplitudes compared to multipolar cells . These electrophysiologic properties enable fusiform cells to sustain high frequency firing critical to focused attention, learning and responding to salient stimuli . Thus, the specific vulnerability of rostral neurons to hyperphosphorylated tau aggregation may affect the functioning of these smaller cells and hence, affect its capacity to modulate cognition. Notably, of all three ptau markers, ptau 231 was also the only marker predicting PACC performance in interaction with rostral LC integrity, at a threshold value that was slightly higher than a recently published study reporting faster cognitive decline of unimpaired older individuals with elevated ptau 231 . Furthermore, in populations consisting of unimpaired and impaired individuals higher ptau 231 was associated with faster decline on the MMSE, Dementia Rating Scale and Clinical Dementia Rating-Sum of Boxes scale as compared to ptau 181 . These findings further confirm that lower rostral LC integrity measures can signal risk for early AD-related processes. This study has several limitations. First, while this is currently the largest cohort using state-of-the-art 7 T LC imaging, our sample size is moderately large. Due to the greater forces of the magnetic field, ultra-high field imaging has more strict inclusion criteria, resulting in lower enrollment and possibly a selection-bias to the healthier population. Second, due to the SARS-CoV-2 pandemic longitudinal assessments were delayed, limiting our current analyses to the cross-sectional data. Therefore, it also remains uncertain whether these individuals will exhibit progression consistent with AD trajectories or primary age-related tauopathy (PART) . As we do not have topographical information of tau and because we cannot preclude disease progression given that this sample is still cognitively unimpaired, making a distinction between these pathways is difficult. The fact that ptau 231 has shown to be a sensitive early marker of underlying AD pathology , predicts accumulation of Aβ , that almost half of the individuals >55.5 years in whom we found ptau-LC associations has either a genetic risk factor for AD or presence of elevated Aβ, and that these associations are also detected in the Aβ + individuals separately, makes it more likely that our observations reflect AD-related pathways. Nonetheless, follow-up of these individuals will be required to describe their phenotypic trajectory. Given that our neuropathologic observations confirm a rostral vulnerability independent of disease stage, similar associations can be expected in impaired individuals. Third, caution is needed when interpreting the caudal LC. As this region of the LC is more diffuse , it is also more challenging to image it and the data acquired may therefore not capture the full length of the LC. Therefore, volumetric LC measures should also be interpreted with caution. Finally, comparisons between the three ptau epitopes need to take into account that these markers were quantified using different platforms and antibodies . To conclude, the plasma ptau 231 marker is specifically associated with lower dorso-rostral LC integrity starting in midlife, and jointly, integrity of this LC section and plasma ptau 231 predict lower PACC performance. These findings link a plasma marker presumed to reflect a very early phase of hyperphosphorylation of tau specifically to the part of the LC exhibiting early vulnerability to tau deposition. Reductions in integrity of the dorso-rostral LC may thus signal very early AD risk, including early tau aggregation.
Supplemental Materials
|
eba3bf1a-8d6b-408d-9293-05c9f09c0f5c | 10073931 | Anatomy[mh] | CD47 is identified as an integrin‐associated protein on tumor cells and binds to signal regulatory protein alpha (SIRPα) on phagocytes, such as macrophages . The interaction between CD47 and SIRPα is known as a ‘do not eat me’ signal and functions to inhibit phagocytosis . Evidence suggests that CD47 is a dominant anti‐engulfment signal on tumor cells and is overexpressed in various cancers . Avoiding phagocytosis by tumor‐associated macrophages (TAMs) might promote growth and metastasis of malignant tumors . Thus, blocking of CD47‐SIRPα binding between tumor cells and innate immune cells can increase phagocytosis of tumor cells . Some reports have documented that high expression of CD47 is associated with aggressive breast cancer features and poor prognosis . Since CD47 also binds to thrombospondin‐1 (TSP‐1), the TSP1‐CD47 interaction might influence processes other than phagocytosis, such as proliferation, apoptosis, and migration, and could also play a role in angiogenesis and inflammation . In a previous study, we reported that tumor cell invasion into blood vessels strongly correlates with a basal‐like tumor phenotype and interval‐detected breast cancers . High levels of CD163 TAMs and tumor infiltrating lymphocyte (TIL) subsets were also related to blood vessel invasion (CD31 positive) . Here, we examined the combined tumor expression of CD47 and levels of CD68 TAMs, with particular focus on potential relations with TIL categories, stromal elastosis, lymphatic and blood vessel invasion by tumor cells, and tumor detection mode, examined in a population‐based retrospective series from the Norwegian Breast Cancer Screening Program.
Study population Patients were included from Vestfold County in Eastern Norway. Vestfold comprises around 5% of the Norwegian population with approximately 230,000 inhabitants. The Norwegian Breast Cancer Screening Program involves biennial mammography in the age‐group 50–69 years and was implemented in this county in 2004. The mean age of patients at the time of diagnosis was 60 years (range 49–70 years). A total of 37,977 women participated during the study period of March 2004–June 2009, with attendance rates of 71 and 76% in the initial and second screening rounds. The definition of an interval cancer was a tumor diagnosed between two screening sessions. During this period, 204 invasive screen‐detected cancers and 85 invasive interval cancers were diagnosed. Of these, seven patients were excluded: one screening cancer had no remaining tumor tissue; one screening case was re‐classified as a malignant phyllodes tumor; and one interval case had no biopsy or surgery performed due to multiple metastases. Further, four patients (two screening and two interval cancers) had simultaneous tumors in both breasts; the tumors with the worst prognostic profile (based on the Nottingham Prognostic Index) were selected for inclusion. Overall, 200 screening cancers and 82 interval cancers were included in this study. Clinical data, tumor stage, and survival information were retrieved from patient records. Immunohistochemical (IHC) surrogate markers for molecular subtypes of breast cancer were defined and applied according to the St. Gallen consensus from 2013 . The cutoff for estrogen receptor (ER) and progesterone receptor (PR) was 1% in the present study. Information on local tumor recurrence, distant metastases, or death was recorded for the 282 patients. The median follow‐up was 138 months (range 108–168; last clinical follow‐up was August 2018). The study was approved by the Regional Ethics Committee of Eastern Norway (reference #2018/1102). IHC staining Tissue microarray (TMA) slides were used in this study. Three tissue cores (diameter 1.0 mm) were extracted from paraffin‐embedded blocks with tumor tissue and inserted into TMA recipient blocks using a semi‐automated precision instrument (Minicore 3 Tissue Arrayer, Alphelys, France). In this study, a total of 226 of the 282 (80%) cases in the cohort were available for analyses on TMA sections. In 43 (15%) cases, the TMA cores had too limited tissue for evaluation, and whole sections were therefore used as substitutes. Regarding whole sections, three small areas from the tumor periphery, corresponding approximately to three TMA cores, were selected for evaluation of CD47 and CD68 expression in a similar way as used for TMA sections. In 13 additional cases (5%), tissue from the core needle biopsy (diameter 3–4 mm) was used, as in our previous studies . Dual IHC staining was performed on 4–5 μm standard tumor‐tissue sections. All sections were stained using the Ventana Discovery Ultra fully automated immuno‐stainer (Roche Diagnostics, Oslo, Norway) with standard protocol procedure. Pretreatment with Ventana Discovery CC1 (Roche Cat. No. 0641457001) for 64 min was performed. Anti‐CD47 antibody SP279 (Abcam ab226837; Abcam, Cambridge, UK) at dilution 1:100 was detected by anti‐Rabbit HQ (Roche Cat. No. 07017812001), followed by Discovery Anti‐HQ HRP (Roche Cat. No. 07017936001) and Discovery Purple (Roche Cat. No. 07053983001). Anti‐CD68 antibody PG‐M1 (Dako M0876, Dako, Oslo, Norway) at dilution 1:100 was detected by Discovery OmniMap anti‐Ms HRP (Roche 05269652001) followed by Discovery Teal HRP (Roche Cat. No. 08254338001). All slides were counterstained with hematoxylin. Anti‐CD47 and anti‐CD68 dual IHC staining generated purple membranous and cytoplasmic staining for CD47 and teal (turquoise) cytoplasmic staining for CD68. Primary antibodies were omitted for the negative controls. Tissues from breast cancer with known positive CD47 and CD68 expression were used as positive controls. Evaluation of CD47 and CD68 expression CD47 positive cases were evaluated for staining intensity in the tumor cell membrane or cytoplasm, as well as the staining area (fraction of positive tumor cells, %). A staining index (SI, values 0–9), obtained as a product of staining intensity (0–3) and proportion of immunopositive tumor cells (<10% = 1, 10–50% = 2, and >50% = 3), was calculated for the core with the strongest expression (‘hot core’), as previously published . Cases were dichotomized (based on frequency distribution analysis) as high or strong (SI 6–9; 33% of cases) versus low or weak (SI < 6; 67% of cases). CD68 positive staining was detected in the cytoplasm of TAMs and counted using an eye‐piece graticule (10 × 10 gridlines; 0.29 × 0.29 mm; total 0.084 mm 2 ) in the most active target areas (‘hot spot'), as previously published . High expression of combined CD47(tumor)–CD68(TAM) was defined as both high tumor cell expression of CD47 (SI 6–9) and high levels (upper quartile) of CD68 TAMs observed at the tumor periphery in the stroma or within the tumor epithelium. Some variability among the three TMA cores from the same tumor was observed. Notably, the TMA core with the most ‘active areas’ and strongest expression of CD47 and CD68 (‘hot spot’ or ‘hot core’) was selected to represent each case. For estimation of interobserver agreement, staining for CD47 and CD68 in 50 cases was evaluated by two pathologists (YC and TAK), showing good agreement with a kappa value of 0.81 for CD47 (supplementary material, Table ). CD68 showed a somewhat lower agreement with a kappa value of 0.66 (supplementary material, Table ). Information on IHC evaluation of CD3, CD4, CD8, CD45, FOXP3 TILs, CD163 TAMs, D2‐40 for lymphatic vessel invasion, and CD31 for blood vessel invasion was included from previous studies for comparison with CD47 tumor expression and CD68 TAM counts . In addition, for this study, counts derived from the CD45‐immunostained TMA sections were compared with TIL counts based on hematoxylin and eosin (H&E)‐stained whole slides (by two pathologists, YC and TAK) and following the criteria from the international TILs working group (50 random cases) . This comparison showed good agreement with a kappa value of 0.75. Baseline data on ER, PR, HER2, and Ki67 for this cohort have also been previously reported . Statistics All analyses were performed using the SPSS package, version 26.0 (IBM Corp, Armonk, NY, USA). Statistical significance (two‐sided) was considered as a P value less than 0.05. Continuous variables were categorized based on quartile limits. For statistical analysis, CD47 SI and CD68 TAM counts were dichotomized; high CD47 expression (SI 6–9) versus low CD47 expression (SI < 6), and high CD68 TAM counts (by upper quartile) versus low CD68 counts (others). Associations between categorical variables were assessed using Pearson's chi‐square test. The Cohen's kappa test for interobserver variability was used to examine the agreement between two observers. A total of 282 patients were accessible for survival analyses. The end point was disease‐specific survival (DSS), i.e. the time from diagnosis to death from breast cancer, or recurrence‐free survival (RFS), i.e. the time from diagnosis to disease recurrence for patients without metastases at the time of diagnosis. Univariate survival analyses were performed using the Kaplan–Meier method (log‐rank test for differences). Prognostic values of different variables were compared by Cox proportional hazards method for multivariate analyses, using the likelihood ratio test for differences. The variables were visually examined by log‐minus‐log plots to check for proportionality before incorporation into multivariate models. Hazard ratios (HRs) and their 95% confidence interval (CI) were estimated.
Patients were included from Vestfold County in Eastern Norway. Vestfold comprises around 5% of the Norwegian population with approximately 230,000 inhabitants. The Norwegian Breast Cancer Screening Program involves biennial mammography in the age‐group 50–69 years and was implemented in this county in 2004. The mean age of patients at the time of diagnosis was 60 years (range 49–70 years). A total of 37,977 women participated during the study period of March 2004–June 2009, with attendance rates of 71 and 76% in the initial and second screening rounds. The definition of an interval cancer was a tumor diagnosed between two screening sessions. During this period, 204 invasive screen‐detected cancers and 85 invasive interval cancers were diagnosed. Of these, seven patients were excluded: one screening cancer had no remaining tumor tissue; one screening case was re‐classified as a malignant phyllodes tumor; and one interval case had no biopsy or surgery performed due to multiple metastases. Further, four patients (two screening and two interval cancers) had simultaneous tumors in both breasts; the tumors with the worst prognostic profile (based on the Nottingham Prognostic Index) were selected for inclusion. Overall, 200 screening cancers and 82 interval cancers were included in this study. Clinical data, tumor stage, and survival information were retrieved from patient records. Immunohistochemical (IHC) surrogate markers for molecular subtypes of breast cancer were defined and applied according to the St. Gallen consensus from 2013 . The cutoff for estrogen receptor (ER) and progesterone receptor (PR) was 1% in the present study. Information on local tumor recurrence, distant metastases, or death was recorded for the 282 patients. The median follow‐up was 138 months (range 108–168; last clinical follow‐up was August 2018). The study was approved by the Regional Ethics Committee of Eastern Norway (reference #2018/1102).
Tissue microarray (TMA) slides were used in this study. Three tissue cores (diameter 1.0 mm) were extracted from paraffin‐embedded blocks with tumor tissue and inserted into TMA recipient blocks using a semi‐automated precision instrument (Minicore 3 Tissue Arrayer, Alphelys, France). In this study, a total of 226 of the 282 (80%) cases in the cohort were available for analyses on TMA sections. In 43 (15%) cases, the TMA cores had too limited tissue for evaluation, and whole sections were therefore used as substitutes. Regarding whole sections, three small areas from the tumor periphery, corresponding approximately to three TMA cores, were selected for evaluation of CD47 and CD68 expression in a similar way as used for TMA sections. In 13 additional cases (5%), tissue from the core needle biopsy (diameter 3–4 mm) was used, as in our previous studies . Dual IHC staining was performed on 4–5 μm standard tumor‐tissue sections. All sections were stained using the Ventana Discovery Ultra fully automated immuno‐stainer (Roche Diagnostics, Oslo, Norway) with standard protocol procedure. Pretreatment with Ventana Discovery CC1 (Roche Cat. No. 0641457001) for 64 min was performed. Anti‐CD47 antibody SP279 (Abcam ab226837; Abcam, Cambridge, UK) at dilution 1:100 was detected by anti‐Rabbit HQ (Roche Cat. No. 07017812001), followed by Discovery Anti‐HQ HRP (Roche Cat. No. 07017936001) and Discovery Purple (Roche Cat. No. 07053983001). Anti‐CD68 antibody PG‐M1 (Dako M0876, Dako, Oslo, Norway) at dilution 1:100 was detected by Discovery OmniMap anti‐Ms HRP (Roche 05269652001) followed by Discovery Teal HRP (Roche Cat. No. 08254338001). All slides were counterstained with hematoxylin. Anti‐CD47 and anti‐CD68 dual IHC staining generated purple membranous and cytoplasmic staining for CD47 and teal (turquoise) cytoplasmic staining for CD68. Primary antibodies were omitted for the negative controls. Tissues from breast cancer with known positive CD47 and CD68 expression were used as positive controls.
CD47 and CD68 expression CD47 positive cases were evaluated for staining intensity in the tumor cell membrane or cytoplasm, as well as the staining area (fraction of positive tumor cells, %). A staining index (SI, values 0–9), obtained as a product of staining intensity (0–3) and proportion of immunopositive tumor cells (<10% = 1, 10–50% = 2, and >50% = 3), was calculated for the core with the strongest expression (‘hot core’), as previously published . Cases were dichotomized (based on frequency distribution analysis) as high or strong (SI 6–9; 33% of cases) versus low or weak (SI < 6; 67% of cases). CD68 positive staining was detected in the cytoplasm of TAMs and counted using an eye‐piece graticule (10 × 10 gridlines; 0.29 × 0.29 mm; total 0.084 mm 2 ) in the most active target areas (‘hot spot'), as previously published . High expression of combined CD47(tumor)–CD68(TAM) was defined as both high tumor cell expression of CD47 (SI 6–9) and high levels (upper quartile) of CD68 TAMs observed at the tumor periphery in the stroma or within the tumor epithelium. Some variability among the three TMA cores from the same tumor was observed. Notably, the TMA core with the most ‘active areas’ and strongest expression of CD47 and CD68 (‘hot spot’ or ‘hot core’) was selected to represent each case. For estimation of interobserver agreement, staining for CD47 and CD68 in 50 cases was evaluated by two pathologists (YC and TAK), showing good agreement with a kappa value of 0.81 for CD47 (supplementary material, Table ). CD68 showed a somewhat lower agreement with a kappa value of 0.66 (supplementary material, Table ). Information on IHC evaluation of CD3, CD4, CD8, CD45, FOXP3 TILs, CD163 TAMs, D2‐40 for lymphatic vessel invasion, and CD31 for blood vessel invasion was included from previous studies for comparison with CD47 tumor expression and CD68 TAM counts . In addition, for this study, counts derived from the CD45‐immunostained TMA sections were compared with TIL counts based on hematoxylin and eosin (H&E)‐stained whole slides (by two pathologists, YC and TAK) and following the criteria from the international TILs working group (50 random cases) . This comparison showed good agreement with a kappa value of 0.75. Baseline data on ER, PR, HER2, and Ki67 for this cohort have also been previously reported .
All analyses were performed using the SPSS package, version 26.0 (IBM Corp, Armonk, NY, USA). Statistical significance (two‐sided) was considered as a P value less than 0.05. Continuous variables were categorized based on quartile limits. For statistical analysis, CD47 SI and CD68 TAM counts were dichotomized; high CD47 expression (SI 6–9) versus low CD47 expression (SI < 6), and high CD68 TAM counts (by upper quartile) versus low CD68 counts (others). Associations between categorical variables were assessed using Pearson's chi‐square test. The Cohen's kappa test for interobserver variability was used to examine the agreement between two observers. A total of 282 patients were accessible for survival analyses. The end point was disease‐specific survival (DSS), i.e. the time from diagnosis to death from breast cancer, or recurrence‐free survival (RFS), i.e. the time from diagnosis to disease recurrence for patients without metastases at the time of diagnosis. Univariate survival analyses were performed using the Kaplan–Meier method (log‐rank test for differences). Prognostic values of different variables were compared by Cox proportional hazards method for multivariate analyses, using the likelihood ratio test for differences. The variables were visually examined by log‐minus‐log plots to check for proportionality before incorporation into multivariate models. Hazard ratios (HRs) and their 95% confidence interval (CI) were estimated.
Expression of CD47 and CD68 in breast cancer Dual IHC staining for CD47 and CD68 in tumor tissues is shown in Figure . Of 282 cases, 93 (33%) showed high CD47 tumor cell expression, and high counts of CD68 TAMs were found in 68 cases (24%). CD47 and CD68 showed a significant positive association ( p < 0.001) (supplementary material, Table ). Associations between CD47 or CD68 , combined CD47–CD68 , and clinicopathological features, vascular invasion, and breast cancer molecular subgroups Separately, high expression of CD47, high levels of CD68‐TAMs, or combined high CD47–CD68 were significantly associated with negative ER, high Ki67, and high counts of all TIL categories (Table ). Further, high CD47 tumor expression or high combined CD47–CD68 were associated with histologic type and interval‐detected tumors. High levels of combined CD47–CD68 were associated with low stromal elastosis. High expression of CD47 or CD68 and combined CD47–CD68 were significantly associated with blood vessel invasion (CD31). CD68 and combined CD47–CD68 were also related to lymphatic vessel invasion (D2‐40) (Table ). High CD47 expression and high CD68 counts were strongly associated with combined lymphatic and blood vessel invasion (Table ). This series showed 51% luminal A, 32% luminal B (HER2−), 6% luminal B (HER2+), 4% HER2 type, and 7% triple negative cancers. High expression of CD47, CD68, and combined CD47–CD68 was more often found in triple negative and HER2 positive categories (Table ). Luminal A and luminal B (HER2−) tumors showed in general lower expression of CD47 or CD68 and combined CD47–CD68 (Table ). Association between CD47 or CD68 , combined CD47–CD68 , and prognosis Distant metastases were observed in 38 cases (13%), 6 patients had local recurrence only (2%), and 34 patients (12%) died of breast cancer during the follow‐up period. By univariate survival analyses of all cases, high expression of CD47 alone showed a trend of association with reduced RFS ( p = 0.051; Figure ). Among luminal A cases, high CD47 was associated with shorter RFS ( p = 0.025; Figure ). High levels of CD68 macrophages alone were not significantly related to RFS in the whole cohort (supplementary material, Figure A), whereas high CD68 TAM counts were associated with shorter RFS within the luminal A subset (supplementary material, Figure B). We found reduced RFS for combined CD47–CD68‐high cases in all tumors ( p = 0.018), and among luminal A cases ( p < 0.001) (Figure ), with shorter DSS in all luminal (HER2−) tumors ( p = 0.039). When analyzing each of the three other subgroups (of CD47–CD68 combinations) separately in comparison with the remaining cases, we found no significant differences (supplementary material, Figure ). Cases with a combination of high CD47–CD68 and blood vessel invasion (CD31 positive) showed significantly reduced RFS ( p < 0.001) and DSS ( p = 0.003) compared with other cases (Figure ). Other combinations of CD47–CD68 and blood vessel invasion (CD31 positive) were not significant when compared with the rest. For combined high CD47–CD68 and lymphatic vessel invasion (D2‐40) positive cases, no significant survival difference was present. Basic histopathological markers such as tumor diameter, histologic grade, lymph node status, and molecular subtype (including information on ER, HER2 status, and Ki67) together with CD47 or combined CD47–CD68 were included in multivariate analyses of RFS (all cases). High expression of CD47 alone, and combined high CD47–CD68, was significantly associated with shorter RFS (HR: 1.9, p = 0.042; HR: 2.1, p = 0.035) (Tables and ). High levels of CD68 macrophages alone were not significantly related to RFS by multivariate analysis (supplementary material, Table ). Interestingly, high expression of CD47 alone and combined CD47–CD68 indicated shorter RFS among luminal A tumors by multivariate assessment (HR: 3.9, p = 0.012; HR: 5.7, p = 0.004), adjusting for tumor diameter, histologic grade, and lymph node status (Tables and ) .
CD47 and CD68 in breast cancer Dual IHC staining for CD47 and CD68 in tumor tissues is shown in Figure . Of 282 cases, 93 (33%) showed high CD47 tumor cell expression, and high counts of CD68 TAMs were found in 68 cases (24%). CD47 and CD68 showed a significant positive association ( p < 0.001) (supplementary material, Table ).
CD47 or CD68 , combined CD47–CD68 , and clinicopathological features, vascular invasion, and breast cancer molecular subgroups Separately, high expression of CD47, high levels of CD68‐TAMs, or combined high CD47–CD68 were significantly associated with negative ER, high Ki67, and high counts of all TIL categories (Table ). Further, high CD47 tumor expression or high combined CD47–CD68 were associated with histologic type and interval‐detected tumors. High levels of combined CD47–CD68 were associated with low stromal elastosis. High expression of CD47 or CD68 and combined CD47–CD68 were significantly associated with blood vessel invasion (CD31). CD68 and combined CD47–CD68 were also related to lymphatic vessel invasion (D2‐40) (Table ). High CD47 expression and high CD68 counts were strongly associated with combined lymphatic and blood vessel invasion (Table ). This series showed 51% luminal A, 32% luminal B (HER2−), 6% luminal B (HER2+), 4% HER2 type, and 7% triple negative cancers. High expression of CD47, CD68, and combined CD47–CD68 was more often found in triple negative and HER2 positive categories (Table ). Luminal A and luminal B (HER2−) tumors showed in general lower expression of CD47 or CD68 and combined CD47–CD68 (Table ).
CD47 or CD68 , combined CD47–CD68 , and prognosis Distant metastases were observed in 38 cases (13%), 6 patients had local recurrence only (2%), and 34 patients (12%) died of breast cancer during the follow‐up period. By univariate survival analyses of all cases, high expression of CD47 alone showed a trend of association with reduced RFS ( p = 0.051; Figure ). Among luminal A cases, high CD47 was associated with shorter RFS ( p = 0.025; Figure ). High levels of CD68 macrophages alone were not significantly related to RFS in the whole cohort (supplementary material, Figure A), whereas high CD68 TAM counts were associated with shorter RFS within the luminal A subset (supplementary material, Figure B). We found reduced RFS for combined CD47–CD68‐high cases in all tumors ( p = 0.018), and among luminal A cases ( p < 0.001) (Figure ), with shorter DSS in all luminal (HER2−) tumors ( p = 0.039). When analyzing each of the three other subgroups (of CD47–CD68 combinations) separately in comparison with the remaining cases, we found no significant differences (supplementary material, Figure ). Cases with a combination of high CD47–CD68 and blood vessel invasion (CD31 positive) showed significantly reduced RFS ( p < 0.001) and DSS ( p = 0.003) compared with other cases (Figure ). Other combinations of CD47–CD68 and blood vessel invasion (CD31 positive) were not significant when compared with the rest. For combined high CD47–CD68 and lymphatic vessel invasion (D2‐40) positive cases, no significant survival difference was present. Basic histopathological markers such as tumor diameter, histologic grade, lymph node status, and molecular subtype (including information on ER, HER2 status, and Ki67) together with CD47 or combined CD47–CD68 were included in multivariate analyses of RFS (all cases). High expression of CD47 alone, and combined high CD47–CD68, was significantly associated with shorter RFS (HR: 1.9, p = 0.042; HR: 2.1, p = 0.035) (Tables and ). High levels of CD68 macrophages alone were not significantly related to RFS by multivariate analysis (supplementary material, Table ). Interestingly, high expression of CD47 alone and combined CD47–CD68 indicated shorter RFS among luminal A tumors by multivariate assessment (HR: 3.9, p = 0.012; HR: 5.7, p = 0.004), adjusting for tumor diameter, histologic grade, and lymph node status (Tables and ) .
The development and progress of breast cancer is an extremely complex process, involving multiple factors in tumor cells and the supporting microenvironment, for example inflammatory and immune cells, connective tissue cells, and features such as early tumor spread by vascular invasion . Several prior publications suggest that overexpression of CD47 by tumor cells inhibits phagocytosis by macrophages, allowing cancer cells to evade immune surveillance, and this is associated with tumor progression in several cancer types . However, the relationship between CD47 and other microenvironment features such as TIL subtypes, and how these factors promote early tumor spread by blood vessel invasion in breast cancer tissues are less investigated. To our knowledge, the present findings are the first to indicate that high expression of CD47 on tumor cells or combined high CD47–CD68 is significantly associated with blood vessel invasion, high levels of various TIL subsets, high CD163 TAMs, and interval tumor presentation in a population‐based mammography screening material. In line with other studies, we found that high CD47 expression and high combined CD47–CD68 are associated with features of aggressive tumors such as ER negativity, HER2 positivity, and high tumor cell proliferation by expression of Ki67. Thus, coordinated effects of CD47, TAMs, and TILs appear to enable lymphatic and blood vessel invasion and breast cancer progress. Notably, we suggest that tumors with combined high expression of CD47 and CD68 have increased both lymphatic and blood vessel invasion. Our findings are in line with a previous study on non‐small cell lung cancer (NSCLC), investigating both human tumor tissues and cancer cell lines, indicating that high expression of CD47 and CD68 promotes tumor invasion and metastasis . CD47 is known to play a role in angiogenesis, and studies demonstrate a crosstalk between CD47 and vascular endothelial growth factor receptor‐2 (VEGFR‐2) and TSP‐1, important angiogenic regulators . CD68 macrophages polarize into two phenotypes, M1 and M2. Whereas M1 macrophages function as promotors of inflammation, M2 macrophages are immunosuppressive and promote tumor progression. CD163 M2 macrophages and a small proportion of M1 macrophages are considered as TAMs . M1 macrophages can also switch to an M2‐like phenotype as the tumor begins to invade . Breast cancer is often infiltrated by TAMs, and an experimental model demonstrated multiple mechanisms of interaction between TAMs and tumor cells . TAMs are known to produce proangiogenic factors, such as VEGF, to increase the network of vessels, and promote migration and intravasation of tumor cells into blood vessels . Further, TAMs release anti‐inflammatory cytokines which decrease proinflammatory T‐cell immune response and promote tumor progression and metastasis . These experimental observations may support our findings that CD47 and CD68 combined are strongly related to both lymphatic and blood vessel invasion in breast cancer. Here, we found a strong association between overexpression of CD47 on tumor cells and high levels of all TIL subsets in breast cancers, supporting a link between CD47 and the innate immune system . The role of TILs as prognostic markers, especially CD8 and FOXP3 subtypes, has been evaluated in large studies, the majority focusing on ER−, HER2+, and triple negative breast cancers. In these subsets, the presence of high levels of TILs has been associated with better patient survival . However, some other studies and our previous study suggest that high TIL counts might be associated with poorer prognosis in ER+ patients . CD47 plays an important role in the immune evasion of tumor cells through direct or indirect interactions with different types of immune cells . However, the underlying mechanisms are not entirely clear, and increased number of various TIL subgroups does not present a simple reflection of immune function. CD47 has a complex and multifactorial role in anti‐cancer immunity and is involved in the regulation of different immune cell activities . CD47‐SIRPα participates in T‐cell recruitment at sites of inflammation in vivo and regulates T‐cell transendothelial migration . Interaction of CD47‐SIRPα results in inhibition of phagocytosis and stimulates T‐cell activation and neutrophil transepithelial migration . Thus, cancer cells may escape immune cells by upregulation of CD47 expression . Interestingly, our study indicates that FOXP3 has the strongest association with CD47, and infiltration of FOXP3 cells is correlated with poor prognosis in several cancers . Notably, anti‐CD47 treatment may enhance anti‐tumor T‐cell immunity , and a recent publication on pancreatic cancer indicated that CD47 targeting induces remodeling of tumor‐infiltrating immune cells of the tumor microenvironment . Several studies suggest that expression of CD47 is a prognostic marker in many different types of cancers, such as leukemia , lymphoma , hepatocellular carcinoma , bladder cancer , NSCLC , leiomyosarcoma , colorectal cancer , and pancreatic neuroendocrine tumors . Previous reports suggest that high expression of CD47‐SIRPα in the bone marrow and peripheral blood predicts recurrence in breast cancers , and increased CD47 in breast stem cells inhibits phagocytosis . Only a few studies report direct evidence of any correlation between the expression of CD47 and prognosis in breast cancer. A recent study showed that high CD47 expression was associated with epithelial–mesenchymal transition and poor prognosis in triple negative breast cancers . Another study indicated that high levels of combined CD47–CD68 represent an independent predictor of poor prognosis in breast cancer, especially in patients with hormone receptor‐negative tumors . Our findings are similar, except that we found reduced survival in the luminal A category. These results may indicate that when tumor CD47 expression is upregulated combined with increased number of macrophages, this may represent an adaptation to reduced phagocytic activity. As mentioned, we found that high CD47–CD68 was not only an independent prognostic factor among all breast cancers, but also within the hormone receptor‐positive luminal tumors. These mainly low‐grade breast cancers are frequently detected by mammography screening . It is still a challenge in clinical practice to predict who would benefit from one therapy approach over another in patients with luminal tumors, and it is therefore important to identify prognostic markers for this large subgroup, as a basis for improved patient stratification and more precise treatment. Baccelli et al reported that co‐expression of CD47 and MET in circulating tumor cells was associated with metastases and poor prognosis in luminal type breast cancers compared to expression of CD47 or MET alone . Possibly, CD47 and MET co‐expression could provide complementary information on these luminal tumors related to escape from macrophage‐mediated phagocytosis . Recent studies indicate that CD47‐SIRPα is not only a prognostic marker, but may also represent a potential target for treatment . Due to tumor heterogeneity, using TMA sections to evaluate CD47 tumor expression and CD68 TAM counts by IHC is a limitation of this study. To increase representation, three cores were extracted from different locations in the periphery of invasive tumor tissue. The relatively low number of patients in subgroup analyses provided weak statistical power for survival analysis. Studies in larger breast cancer cohorts are necessary to validate the potential of CD47 or CD47–CD68 as biomarkers in breast cancer . Furthermore, although semiquantitative and subjective scoring is a way of reporting tissue biomarker levels as numerical values for robust group comparisons, this method has its limitations. Validation studies need to be performed by using quantitative image‐based methods.
Our study indicates that CD47 and combined CD47–CD68 are significantly associated with high levels of several TIL subsets, blood vessel invasion (CD31 positive), and interval presentation of breast cancer in a population‐based mammography screening series. CD47 might represent an independent marker of reduced survival, including among hormone receptor‐positive luminal tumors. Our findings support that CD47 is a multifaceted actor in the tumor microenvironment and might represent a potentially relevant treatment target in breast cancer.
YC, TAK and LAA contributed to study design, data interpretation, literature search, and writing the manuscript. YC, TAK and HA participated in data collection. YC, TAK and EW performed data analysis and generation of figures. All authors approved the final version of the manuscript.
Figure S1. Estimated RFS according to high or low CD68 levels Figure S2. Estimated RFS and DSS according to different combinations of high or low CD47 or CD68 levels Table S1. Crosstabulation of interobserver analysis of CD47 expression by immunohistochemistry Table S2. Crosstabulation of interobserver analysis of CD68 count by immunohistochemistry Table S3. Cross‐correlations between CD47 and CD68 Table S4. Univariate and multivariate analysis of RFS (Cox proportional hazards method) of pathological variables and CD68 expression by IHC Click here for additional data file.
|
|
Virtual supervision in ophthalmology: a scoping review | 524ffb48-7bcd-4a87-9006-f0a29e6a6511 | 10074343 | Ophthalmology[mh] | The use of telemedicine (TM) in the USA increased dramatically in response to the COVID-19 public health emergency . One application of TM is virtual supervision (VS), where the supervisor and supervisee who are not in the same physical location interact for an episode of patient care via synchronous audio and/or video modalities .Virtual supervision can facilitate cost-effective mentoring by specialists from remote locations and is also useful as a teaching tool in graduate medical education (GME) . The information available in the literature on VS in ophthalmology is not well described. This scoping review aims to better understand the evidence and potential role for VS in ophthalmic practice and education. In consultation with a reference librarian, a literature search strategy was developed in accordance with Preferred Reporting Items for Systematic reviews and Meta-Analyses extension for Scoping Reviews (PRISMA-ScR) . We searched six databases (PubMed, Embase, Web of Science, CINAHL, PsycINFO, and ERIC) for articles published between January 1950 and August 2022. Our search included database-specific thesaurus terms such as medical subject headings (MeSH) and Emtree, as well as keywords related to VS in ophthalmology (Table ). The inclusion criteria were full-text articles published in an English-language peer-reviewed journal. We included studies that involved VS between physicians or between physicians and trainees in ophthalmology. We excluded studies in which the supervisor and supervisee were in the same physical location and studies in which clinical supervision was carried out retrospectively. We also excluded studies that included supervisors who were not ophthalmology physicians. Eligible studies were de-duplicated in EndNote (Clarivate Analytics, Philadelphia, PA) and imported into the systematic review software Covidence (Melbourne, Victoria, Australia) for screening, full-text review, and data extraction. The screening and selection process is displayed in a PRISMA flowchart (Fig. ). Two investigators (CK and CS) independently conducted title/abstract screening, full-text review, and data extraction in Covidence. Disagreements were resolved by the senior investigator (PBG). A data template was developed in Covidence to extract relevant information, including year of publication, location, study design, characteristics of study participants, sample size, and any type of outcome. Two investigators (CK and CS) independently appraised the methodological quality of the studies using the Mixed Methods Appraisal Tool (MMAT), and disagreements were resolved by the senior investigator (PBG). The Mixed Methods Appraisal Tool (MMAT) appraises studies based on five questions assessing the sampling strategy, outcome measurement, confounders, and statistical analysis of the study . All studies were scored on a scale of 1 to 5 based on the MMAT. Study characteristics The initial database search yielded a total of 2700 articles. Following duplicate removal, title and abstract screening, and full-text review, seven articles were included in our qualitative synthesis (Fig. ): four were case reports or case series (57%) , two were cohort studies (29%) , and one was a cross-sectional study (14%) (Table ) . The publication year of the articles ranged from 1998 to 2022. Most studies were conducted in the USA. Other studies were conducted in Finland , Israel , the UK , and the Philippines . Three studies were prospective in design (43%) , while other studies were retrospective (14%) or both (14%) . The objective of the studies ranged from determining the feasibility of a teleconference device (43%) , describing the use of VS in an emergency department (ED) or rural hospital (43%) , or determining whether fellows could safely operate independently with remote supervision (14%) . A total of 619 patients and 367 supervisees were included (Table ). Two studies included supervisees who were physicians, including an ophthalmic surgeon from a foreign country (14%) and a general practitioner in a rural hospital (14%) . Five studies included supervisees who were medical trainees, such as ophthalmology residents (43%) , vitreoretinal fellows (14%) , and emergency medicine (EM) residents (14%) . Study setting varied between the ED , operating room (OR) , rural hospital , and eye clinic . Virtual supervision was used for a clinical examination in three studies (43%) , surgical procedure in three studies (43%) , and in-office procedure in one study (14%) . For studies involving a surgical or in-office procedure, VS was used to teach surgeons endoscopic laser-assisted dacryocystorhinostomy (ELA-DCR) , perform orbital decompression on a patient with extensive facial trauma , remove a corneal foreign body and rust ring in a rural hospital , and supervise retina fellows repairing macula-on rhegmatogenous retinal detachments (RRD) during bank holidays and weekends . Devices used for VS included an optical coherence tomography (OCT)/fundus camera , slit lamp camera , network camera , and ceiling-mounted camera (Table ) . Synchronous communication between the supervisor and supervisee was facilitated through various modalities such as email , telephone , video conferencing software , and local area network (LAN) . Four studies reported successful transmission of real-time video of surgical or in-office procedures . In three studies on VS for clinical examinations, supervisees held synchronous communication with their supervisors to reach an agreement in diagnosis . Studies noted that VS was helpful for diagnosing ocular surface, anterior segment, and macular diseases but difficult to use when diagnosing vitreous and peripheral retinal conditions . One study noted that the slit lamp camera was sufficient to provide high enough real-time image quality for procedures , while another study using a network camera noted decreased transmission rate due to electronic traffic . Several studies also noted machine malfunction, image artifacts, insufficient image resolution, and lighting problems as factors that decreased the effectiveness of VS . One study reported patient outcomes related to VS and found no significant difference in 6-month retinal re-detachment rate when comparing remotely supervised to unsupervised fellows performing RRD repairs on bank holidays and weekends . Only one study reported educational outcomes. In this study, four out of eight residents stated that VS helped them diagnose patients more accurately, as they could synchronously share the images with the supervisor and talk through their thought processes . Critical appraisal of included studies The MMAT was used to critically appraise the quality of the studies. Limitations were noted in outcome measurement , statistical analysis , sampling strategy , and inclusion of confounding factors . Overall, two studies scored a 5/5 , one study scored a 4/5 , and four studies scored a 3/5 . The initial database search yielded a total of 2700 articles. Following duplicate removal, title and abstract screening, and full-text review, seven articles were included in our qualitative synthesis (Fig. ): four were case reports or case series (57%) , two were cohort studies (29%) , and one was a cross-sectional study (14%) (Table ) . The publication year of the articles ranged from 1998 to 2022. Most studies were conducted in the USA. Other studies were conducted in Finland , Israel , the UK , and the Philippines . Three studies were prospective in design (43%) , while other studies were retrospective (14%) or both (14%) . The objective of the studies ranged from determining the feasibility of a teleconference device (43%) , describing the use of VS in an emergency department (ED) or rural hospital (43%) , or determining whether fellows could safely operate independently with remote supervision (14%) . A total of 619 patients and 367 supervisees were included (Table ). Two studies included supervisees who were physicians, including an ophthalmic surgeon from a foreign country (14%) and a general practitioner in a rural hospital (14%) . Five studies included supervisees who were medical trainees, such as ophthalmology residents (43%) , vitreoretinal fellows (14%) , and emergency medicine (EM) residents (14%) . Study setting varied between the ED , operating room (OR) , rural hospital , and eye clinic . Virtual supervision was used for a clinical examination in three studies (43%) , surgical procedure in three studies (43%) , and in-office procedure in one study (14%) . For studies involving a surgical or in-office procedure, VS was used to teach surgeons endoscopic laser-assisted dacryocystorhinostomy (ELA-DCR) , perform orbital decompression on a patient with extensive facial trauma , remove a corneal foreign body and rust ring in a rural hospital , and supervise retina fellows repairing macula-on rhegmatogenous retinal detachments (RRD) during bank holidays and weekends . Devices used for VS included an optical coherence tomography (OCT)/fundus camera , slit lamp camera , network camera , and ceiling-mounted camera (Table ) . Synchronous communication between the supervisor and supervisee was facilitated through various modalities such as email , telephone , video conferencing software , and local area network (LAN) . Four studies reported successful transmission of real-time video of surgical or in-office procedures . In three studies on VS for clinical examinations, supervisees held synchronous communication with their supervisors to reach an agreement in diagnosis . Studies noted that VS was helpful for diagnosing ocular surface, anterior segment, and macular diseases but difficult to use when diagnosing vitreous and peripheral retinal conditions . One study noted that the slit lamp camera was sufficient to provide high enough real-time image quality for procedures , while another study using a network camera noted decreased transmission rate due to electronic traffic . Several studies also noted machine malfunction, image artifacts, insufficient image resolution, and lighting problems as factors that decreased the effectiveness of VS . One study reported patient outcomes related to VS and found no significant difference in 6-month retinal re-detachment rate when comparing remotely supervised to unsupervised fellows performing RRD repairs on bank holidays and weekends . Only one study reported educational outcomes. In this study, four out of eight residents stated that VS helped them diagnose patients more accurately, as they could synchronously share the images with the supervisor and talk through their thought processes . The MMAT was used to critically appraise the quality of the studies. Limitations were noted in outcome measurement , statistical analysis , sampling strategy , and inclusion of confounding factors . Overall, two studies scored a 5/5 , one study scored a 4/5 , and four studies scored a 3/5 . There is a limited evidence base on VS in ophthalmology. The few available studies suggest that VS is technologically feasible, may provide a positive experience for both supervisees and patients, and can permit synchronous communication and transmission of clinical data, which can be used to formulate diagnosis and management plans or learn new surgical skills. Virtual supervision can improve diagnostic accuracy by allowing fast, synchronous communication between supervisor and trainee. In one study, ophthalmology residents (training level unspecified) used VS to consult their supervisors regarding complicated cases during night shifts in the ED . The residents communicated with their supervisors via telephone and emailed images and videos captured by a miniature slit lamp camera. The authors found high agreement in diagnosis made during the night shift and the on-site examination by the supervisor the next day. In another study, first-year ophthalmology residents sent OCT/fundus photographs to the supervisor to synchronously discuss a differential diagnosis . The use of the device did not correlate with a longer duration of visit in the ED. Virtual supervision can also impact ophthalmic surgical GME. In one study, retina fellows who were deemed ready to operate independently by their supervisors performed surgery while supervisors provided synchronous feedback from a remote location during holidays and weekends . This readiness was assessed on a case-by-case basis by the supervisors after 2–3 months of supervised training based on a review of recorded surgeries and didactics. The study included only primary uncomplicated macula-on RRDs; patients with more complex retinal detachments were excluded. The study reported no statistical difference in the 6-month retinal re-detachment rate between off-hour cases utilizing VS and off-hour cases without any supervision. The authors did not mention if there were any adverse events, difficulties with VS, or a safety plan if a significant complication were to occur during the surgery. In addition, ophthalmologists can use VS to assist physicians in areas with limited access to ophthalmic care or continuing medical education (CME). In one study, a general practitioner in a remote area provided time-sensitive ophthalmic procedures with the assistance of synchronous communication with ophthalmologists . In another study, ophthalmologists in the USA were able to teach a new procedure to ophthalmologists in the Philippines who had the necessary equipment but not the surgical expertise . This review suggests that various methods can be used to ensure high image and video quality during VS, although challenges remain. In one study, patients were instructed to keep voluntary movements to a minimum and to fixate on a target, and a designated remote operator kept the slit lamp image focused . Despite these methods, there were technological glitches, image artifacts, and internet connection issues. Most studies reported that quality of images and videos were adequate for clinical diagnosis and procedure guidance, although in some cases, supervisees needed to spend a significant amount of time to find the optimal setup . Furthermore, the blue filter on the slit lamp with fluorescein was particularly difficult to visualize during VS due to degradation of the video quality . We identified several gaps in the literature on the use of VS in ophthalmology. First, most studies on ophthalmic GME lacked standardized curricula to implement the interventions and methodology to assess the outcomes. Surveys were often used to assess feasibility and usability of VS; only one study collected data from all supervisors, supervisees, and patients . Furthermore, no studies included control groups, which makes it difficult to compare the effectiveness of VS versus direct supervision in GME. Second, most studies focused on feasibility of technology for VS rather than on clinical or educational outcomes. One study that assessed educational outcomes relied on self-reported data without a comparison group . Third, there was significant heterogeneity in the types of supervisees such as residents, fellows, general practitioners, and physicians. Supervisees have varying needs based on their background and training experience, which makes it difficult to generalize the results . This review has several limitations. First, we did not include studies published in gray literature or in languages other than English. Second, the reproducibility of MMAT ratings is limited by the authors’ judgments about the quality of the study design. In conclusion, this scoping review highlights the potential utility of VS in ophthalmology and identifies areas for future research on this topic. Recent technological advancements allow supervisors to guide other physicians and trainees through VS, with potentially improved efficiency and collaboration between different institutions and countries. Future studies should employ larger sample sizes and more rigorous methodology to better define the safety and efficacy of VS in ophthalmic practice and education . |
Advancing Palliative Care Integration in Hematology: Building Upon Existing Evidence | a0b880f4-e343-4a51-9669-b9f482762de6 | 10074347 | Internal Medicine[mh] | Palliative care (PC) is defined by the World Health Organization as an “approach that improves the quality-of-life (QOL) of patients and their families facing the problems associated with life-threatening illness, through the prevention and relief of suffering by means of early identification and impeccable assessment and treatment of pain and other problems, physical, psychosocial, and spiritual” . PC specialists provide complex symptom management and family-centered biopsychosocial assessments with effective communication and focus on QOL . Though PC is often misperceived as end-of-life care, ideal PC is integrated early in the illness trajectory alongside life-prolonging or potentially curative therapies . While PC is sometimes provided by oncologists or other members of the cancer care team, referred to as “primary PC,” our use of the term refers to specialty PC services, or those provided by specialized clinicians on an interdisciplinary team.
Palliative care has many established benefits for patients with cancer. The 2009 ENABLE II randomized trial paved the way for numerous studies supporting PC integration in cancer care . In 2010, Temel demonstrated that early PC led to improved QOL, improved survival, and decreased the intensity of end-of-life care, among patients with metastatic non-small-cell lung cancer . In 2015, the ENABLE III randomized trial showed that patients receiving earlier, as compared to later, PC had improved one-year survival and reduced family and caregiver burden and depression . Multiple systematic reviews and meta-analyses also demonstrate benefit from PC interventions including improvements in patient QOL, symptom burden, caregiver outcomes, advance care planning, health care utilization, and, often, patient survival [ •, , ]. Resultantly, many organizations have called for the integration of PC into routine comprehensive oncology care .
Patients with HM suffer similar, or sometimes greater, symptom burdens to patients with metastatic solid tumors . Studies show that patients with blood cancers commonly present with distress and numerous physical and psychological symptoms including fatigue, insomnia, dry mouth, pain, and anxiety . Patients with HM also receive inadequate symptom management, psychological support, and engagement in advance care planning, which all contribute substantially to increased morbidity . While survival is improving, approximately 50,000 deaths annually are attributed to hematologic malignancies (HM) . As part of standard HM management, many patients receive intensive treatments, prolonged hospitalization, and oftentimes require life-long suppressive therapies . Patients with HM are more likely than those with solid tumors to receive intensive end-of-life care (e.g., chemotherapy or intensive care at end-of-life) and to die in the hospital . Patients with HM are also less likely than solid tumor patients to have documented care preferences such as advance care plans or be referred to hospice . Patients referred to hospice tend to experience shorter length of stays, which signals late referrals and limited benefit from hospice care . Patients with HM also experience significant barriers to hospice care including lack of access to blood transfusions for symptom support. Among survivors, symptoms often persist with long-term sequelae and quality-of-life implications. Survivors commonly experience fatigue, pain, neuropathy, cardiomyopathy, neurocognitive deficits, psychological distress, anticipatory grief, fear of recurrence, and post-traumatic stress .
While high-quality evidence supports the integration of PC in oncology, many barriers exist . Most randomized clinical trials of PC integration have excluded patients with HM. The integration of PC into standard HM care has therefore lagged behind that of solid tumor care . Hematologic malignancy specialists are less likely than solid tumor oncologists to request specialty PC consultation . Many also equate PC to end-of-life care and may not recognize the demonstrated benefits of early PC . Furthermore, hematologic malignancy specialists may wish to address the primary PC needs of their own patients. Surveys have shown, however, that hematologic malignancy specialists often express discomfort discussing death or hospice referral, as well as a sense of shame that this transition in treatment goals may indicate a personal failure . The treatment trajectories of blood cancers also contribute hurdles for service integration. The possibility of cure is rather unique to HM, especially when compared to most other advanced cancers, and drives aggressive clinical decision-making . Patients with HM often have a rapid and unpredictable decline at the end of life, which contributes to prognostic uncertainty and challenges both clinicians and patients; more than half of patients with HM have a different understanding of their prognosis than their hematologist . Prognostic uncertainties and misperceptions about treatment risks and benefits represent unmet PC needs for patients and families with HM .
Interest is growing for earlier integration of PC into hematologic malignancy care. Several recent studies demonstrate the benefits of PC integrated into HM care, yet only for some specific HM (ex. acute myeloid leukemia) and care settings (inpatient). HM are a heterogeneous group of diseases, with each patient with HM having unique needs, salient clinical features, treatment paradigms, and expected outcomes. Thus, each major disease group will likely have a different solution to the puzzle of PC-hemato-oncology integration. To follow, we summarize salient features and treatment paradigms for the various major HM disease sub-types. We then examine what is known about the PC needs specific to each disease, detail relevant studies of PC integration, and discuss the anticipated needs of each group, emphasizing areas warranting further study. A summary may be found in Table .
The three main categories of HM include (1) leukemias, (2) lymphomas, and (3) multiple myeloma (MM) . For clarity and completeness, we include further subcategories including acute versus chronic leukemias, myelodysplastic syndromes and myeloproliferative neoplasms, and cellular therapies, including CAR-T therapy and hematopoietic stem cell transplantation (HSCT) (which may be performed in several disease states and have specific PC considerations). While each disease is unique, HM as a group generally share at least one of the following: The need for intensive treatments to achieve remission or cure, which is associated with risk for early mortality and/or treatment toxicity Prognostic uncertainty and unpredictable illness courses, including wide variability in outcomes including possibility of cure Sometimes chronic, indolent, and/or a relapsing and remitting course, requiring indefinite and continuous oral suppressive therapies High patient and caregiving burden (physical, emotional, and/or spiritual), even after treatment completion
Treatment paradigms Leukemias are subdivided into two types: acute or chronic. Acute myeloid leukemia (AML) is the most common in adults followed by acute lymphoid leukemia (ALL). The abrupt presentation and rapidly progressive nature of acute leukemia lends itself to be more responsive to chemotherapy than solid tumors, which means a chance for cure. Cure, however, typically requires higher-dose intensity chemotherapy than typical solid tumor regimens and/or consolidative hematopoietic stem cell transplantation . This establishes a characteristic “high-risk, high-reward” treatment paradigm in acute leukemia care. Patient’s will receive intensive chemotherapy regimens, while other non-intensive treatments contributes to wide variability in treatment-associated morbidity and mortality. HSCT is considered for many patients and is discussed in a separate section below. Symptom burden and palliative care needs Acute leukemia may be among the most psychologically distressing of all cancers . The intensive treatments for acute leukemias necessitate weeks long and socially isolating hospitalizations . Treatments can carry high risk of death and treatment-related toxicities. Physical symptoms may be severe and include fevers, fatigue, mucositis, and other distressing gastrointestinal symptoms . Treatment effects may be long-lasting, contributing to emotional and psychological symptoms and worsened QOL . A secondary data analysis of 160 patients with AML found that a substantial proportion reported clinically significant post-traumatic stress symptoms one month after intensive chemotherapy . Approximately, a third of patients will report significant depressive or anxiety symptoms. Another one-third experience acute stress reactions from the shock of the diagnosis and unexpected urgent hospitalization . In addition to the high burden of physical and emotional symptoms, patients with acute leukemias have additional unmet PC-related needs. One study revealed that while 86% of AML patients were expected by their oncologists to have a poor prognosis, 74% of these patients reported at least a 50% chance of cure . Older adults with AML over-estimate their prognosis by threefold . This skew towards optimistic prognostication and aggressive care leads to increased healthcare utilization at the end of life. Patients with acute leukemia are more likely to choose aggressive therapies and die in the hospital, while accessing PC services less frequently than those patients with advanced solid tumors . A study of 168 deceased patients with acute leukemia revealed that 66.7% were hospitalized in the last week of life and over half received chemotherapy in the last 30 days of life . Another study of 200 leukemia patients reported the median time from last code status transition to death was only two days. Thirty-two percent (32%) of those code status conversations occurred at the time of clinical deterioration and 39.5% without the patient present or capable of making their own medical decisions . Thus, there are significant opportunities for improvement in advance care planning and symptom management in acute leukemia care. Palliative care integration In 2021, El-Jawahri published a multisite randomized clinical trial of 160 adults with AML undergoing intensive chemotherapy showing that integrated specialty PC significantly improves patient-reported QOL, depression, anxiety, and posttraumatic stress symptoms. Among the patients who died, those receiving integrated PC were more likely to have discussed their end-of-life preferences and less likely to receive chemotherapy at end-of-life [ ••, ]. Smaller studies reinforce the benefits of integrated PC in acute leukemia care, showing increased hospice use and fewer intensive care unit admissions . Next steps For patients with AML receiving intensive induction chemotherapy, the evidence of benefit of PC integration early in the disease course is clear. PC should be involved at the time of admission for induction chemotherapy or index hospitalization. PC services have much to offer including providing symptom management throughout the hospitalization, caregiver encouragement, support during potential HSCT, and, if needed, end-of-life care . Future work is needed in this area with at least one major clinical trial underway. SPRINT is an active multisite randomized controlled clinical trial examining collaborative palliative and leukemia care versus standard leukemia care alone in patients with AML and high risk MDS receiving non-intensive chemotherapy .
Leukemias are subdivided into two types: acute or chronic. Acute myeloid leukemia (AML) is the most common in adults followed by acute lymphoid leukemia (ALL). The abrupt presentation and rapidly progressive nature of acute leukemia lends itself to be more responsive to chemotherapy than solid tumors, which means a chance for cure. Cure, however, typically requires higher-dose intensity chemotherapy than typical solid tumor regimens and/or consolidative hematopoietic stem cell transplantation . This establishes a characteristic “high-risk, high-reward” treatment paradigm in acute leukemia care. Patient’s will receive intensive chemotherapy regimens, while other non-intensive treatments contributes to wide variability in treatment-associated morbidity and mortality. HSCT is considered for many patients and is discussed in a separate section below.
Acute leukemia may be among the most psychologically distressing of all cancers . The intensive treatments for acute leukemias necessitate weeks long and socially isolating hospitalizations . Treatments can carry high risk of death and treatment-related toxicities. Physical symptoms may be severe and include fevers, fatigue, mucositis, and other distressing gastrointestinal symptoms . Treatment effects may be long-lasting, contributing to emotional and psychological symptoms and worsened QOL . A secondary data analysis of 160 patients with AML found that a substantial proportion reported clinically significant post-traumatic stress symptoms one month after intensive chemotherapy . Approximately, a third of patients will report significant depressive or anxiety symptoms. Another one-third experience acute stress reactions from the shock of the diagnosis and unexpected urgent hospitalization . In addition to the high burden of physical and emotional symptoms, patients with acute leukemias have additional unmet PC-related needs. One study revealed that while 86% of AML patients were expected by their oncologists to have a poor prognosis, 74% of these patients reported at least a 50% chance of cure . Older adults with AML over-estimate their prognosis by threefold . This skew towards optimistic prognostication and aggressive care leads to increased healthcare utilization at the end of life. Patients with acute leukemia are more likely to choose aggressive therapies and die in the hospital, while accessing PC services less frequently than those patients with advanced solid tumors . A study of 168 deceased patients with acute leukemia revealed that 66.7% were hospitalized in the last week of life and over half received chemotherapy in the last 30 days of life . Another study of 200 leukemia patients reported the median time from last code status transition to death was only two days. Thirty-two percent (32%) of those code status conversations occurred at the time of clinical deterioration and 39.5% without the patient present or capable of making their own medical decisions . Thus, there are significant opportunities for improvement in advance care planning and symptom management in acute leukemia care.
In 2021, El-Jawahri published a multisite randomized clinical trial of 160 adults with AML undergoing intensive chemotherapy showing that integrated specialty PC significantly improves patient-reported QOL, depression, anxiety, and posttraumatic stress symptoms. Among the patients who died, those receiving integrated PC were more likely to have discussed their end-of-life preferences and less likely to receive chemotherapy at end-of-life [ ••, ]. Smaller studies reinforce the benefits of integrated PC in acute leukemia care, showing increased hospice use and fewer intensive care unit admissions .
For patients with AML receiving intensive induction chemotherapy, the evidence of benefit of PC integration early in the disease course is clear. PC should be involved at the time of admission for induction chemotherapy or index hospitalization. PC services have much to offer including providing symptom management throughout the hospitalization, caregiver encouragement, support during potential HSCT, and, if needed, end-of-life care . Future work is needed in this area with at least one major clinical trial underway. SPRINT is an active multisite randomized controlled clinical trial examining collaborative palliative and leukemia care versus standard leukemia care alone in patients with AML and high risk MDS receiving non-intensive chemotherapy .
Treatment paradigms In direct contrast with acute leukemias, chronic leukemias often present asymptomatically and have a slow disease course. Chronic myeloid leukemia (CML) and chronic lymphoid leukemia (CLL) are the most common types in adults. In general, prognosis is measured in years, sometimes even decades. Typical cases of CML are treated with oral tyrosine kinase inhibitors, which typically confer an excellent prognosis and often well-tolerated side effects. While prognosis is good, patients with CML often require indefinite and continuous oral targeted therapy, which has psychological, financial, and other implications. This ‘treatment-without-end’ paradigm is characteristic of CML care. Atypical cases of CML, such as those with resistant mutations or those that transform to AML, may have shortened survival or experience the effects described previously related to the transformed acute leukemia . Some patients face severe side effects or tolerability issues from their treatment. CLL is typified by older patient age and a slow-growing, indolent nature, with most patients presenting initially asymptomatic. The default management strategy for those with less aggressive variants is ‘active surveillance’ without treatment. Sometimes CLL can transform into an aggressive and life-limiting variant or aggressive lymphoma (“Richter’s transformation”), which indicates a poor prognosis. Regardless, patients with CLL can face frequent infectious and nosocomial complications, the need for hospitalization or intermittent treatments unpredictably, and, as with other HM, difficult prognostication . Some require indefinite oral therapies, at significant financial cost and sometimes with unfavorable side effect profiles . Symptom burden and palliative care needs There is marked clinical heterogeneity in how chronic leukemia variants may impacts a patient’s life. Patients with chronic leukemias may suffer severe toxicities and symptoms, often related to the chronicity of the illness and/or treatment. An international survey of 1482 patients with CLL found significantly worse emotional well-being in those with CLL than other patients with cancer . Life-long suppressive treatments, uncertainty related to the timing and severity of inevitable relapse, and nosocomial complications all contribute to detriments in physical, emotional, and financial well-being. Palliative care integration and next steps While it is recognized that patients with chronic leukemias experience significant symptom burden, often stemming from the cumulative toxicities of decades of continuous oral targeted therapies and recurring relapses, little has been studied regarding PC integration into standard chronic leukemia care. PC specialists could provide an extra layer of support for patients with chronic leukemias, particularly regarding the need for psycho-oncologic support interventions for patients with CLL and enhanced symptom management . Furthermore, there may be opportunity to provide support around the experience of living with a chronic illness, which may be experienced as a sword of Damocles or having the potential for transformation or progression in a manner that is difficult to predict. It is not clear, however, that every patient with a chronic leukemia needs or would benefit from specialist palliative care services. Perhaps those with CLL may do well with geriatrics or social work support with periodic PC consultation for those with specific and challenging PC needs. More research is needed to better understand the needs and the optimal involvement of PC in this population.
In direct contrast with acute leukemias, chronic leukemias often present asymptomatically and have a slow disease course. Chronic myeloid leukemia (CML) and chronic lymphoid leukemia (CLL) are the most common types in adults. In general, prognosis is measured in years, sometimes even decades. Typical cases of CML are treated with oral tyrosine kinase inhibitors, which typically confer an excellent prognosis and often well-tolerated side effects. While prognosis is good, patients with CML often require indefinite and continuous oral targeted therapy, which has psychological, financial, and other implications. This ‘treatment-without-end’ paradigm is characteristic of CML care. Atypical cases of CML, such as those with resistant mutations or those that transform to AML, may have shortened survival or experience the effects described previously related to the transformed acute leukemia . Some patients face severe side effects or tolerability issues from their treatment. CLL is typified by older patient age and a slow-growing, indolent nature, with most patients presenting initially asymptomatic. The default management strategy for those with less aggressive variants is ‘active surveillance’ without treatment. Sometimes CLL can transform into an aggressive and life-limiting variant or aggressive lymphoma (“Richter’s transformation”), which indicates a poor prognosis. Regardless, patients with CLL can face frequent infectious and nosocomial complications, the need for hospitalization or intermittent treatments unpredictably, and, as with other HM, difficult prognostication . Some require indefinite oral therapies, at significant financial cost and sometimes with unfavorable side effect profiles .
There is marked clinical heterogeneity in how chronic leukemia variants may impacts a patient’s life. Patients with chronic leukemias may suffer severe toxicities and symptoms, often related to the chronicity of the illness and/or treatment. An international survey of 1482 patients with CLL found significantly worse emotional well-being in those with CLL than other patients with cancer . Life-long suppressive treatments, uncertainty related to the timing and severity of inevitable relapse, and nosocomial complications all contribute to detriments in physical, emotional, and financial well-being.
While it is recognized that patients with chronic leukemias experience significant symptom burden, often stemming from the cumulative toxicities of decades of continuous oral targeted therapies and recurring relapses, little has been studied regarding PC integration into standard chronic leukemia care. PC specialists could provide an extra layer of support for patients with chronic leukemias, particularly regarding the need for psycho-oncologic support interventions for patients with CLL and enhanced symptom management . Furthermore, there may be opportunity to provide support around the experience of living with a chronic illness, which may be experienced as a sword of Damocles or having the potential for transformation or progression in a manner that is difficult to predict. It is not clear, however, that every patient with a chronic leukemia needs or would benefit from specialist palliative care services. Perhaps those with CLL may do well with geriatrics or social work support with periodic PC consultation for those with specific and challenging PC needs. More research is needed to better understand the needs and the optimal involvement of PC in this population.
Treatment paradigms MM is the second most common hematologic malignancy. While survival is improving with the advent of new therapies, MM remains generally incurable. The disease course is typified by periods of remission and relapse. The time in remission before relapse varies greatly depending on disease phenotype the clinical aggressiveness of the MM, though in ideal circumstances can last years. Patients often remain on maintenance therapies even when in remission. MM patients commonly are treated with five or more lines of therapy, including HSCT and multidrug regimens . Patients with ‘standard risk’ MM may be expected to live approximately 5–10 or more years . Symptom burden and palliative care needs MM patients receive indefinite therapy and must cope with the relapsing and remitting disease course, cumulative toxicities, and chronic survivorship. Symptoms, both physical and psychological, stem from the snowballing effects of treatments, the expectation and timing uncertainty of inevitable relapse, and the need for recurrent treatments . Patients with MM have been described as having worse physical function and global mental health than the general population, as well as diminished health-related QOL when compared to those with other HM . Physically, patients experience fatigue, pain, breathlessness, nausea, muscle weakness, and peripheral neuropathy . Psychologically, patients report the impact of social isolation, financial stress, relationship strain, anticipatory grief, and the toll of endless and unrelenting treatment. A recent cross-sectional, multisite study of 180 MM patients reported that nearly 25% of patients reported clinically significant depression, anxiety, and post-traumatic stress symptoms . There is growing evidence that the caregivers of patients with MM also struggle with psychological symptoms and could benefit from PC support. A cross-sectional, multisite study of 127 MM caregivers revealed that 44.1% have clinically significant anxiety, while another 24.4% reported post-traumatic stress symptoms. Caregivers reported higher rates of anxiety than the patients with MM themselves . Prognostic misunderstanding and patient–provider communication was thought to be a major contributor of stress for caregivers of patients with MM in this study. Palliative care integration and next steps Despite the high illness burden experienced by both MM patients and caregivers, there is a paucity of data on models for PC integration in the care of MM patients. PC integration research is needed in the MM population. During periods of disease progression, the possible benefits of PC involvement seem clear. During the prolonged periods of disease control, however, the ideal integration and involvement of palliative services is less apparent. We suspect that patients may still benefit from symptom assessments, psychosocial support, and assistance with coping during this period marked by survivorship and the anticipation of inevitable relapse. Assessing for unmet palliative-related needs and consideration of PC involvement is crucial throughout the trajectory of care.
MM is the second most common hematologic malignancy. While survival is improving with the advent of new therapies, MM remains generally incurable. The disease course is typified by periods of remission and relapse. The time in remission before relapse varies greatly depending on disease phenotype the clinical aggressiveness of the MM, though in ideal circumstances can last years. Patients often remain on maintenance therapies even when in remission. MM patients commonly are treated with five or more lines of therapy, including HSCT and multidrug regimens . Patients with ‘standard risk’ MM may be expected to live approximately 5–10 or more years .
MM patients receive indefinite therapy and must cope with the relapsing and remitting disease course, cumulative toxicities, and chronic survivorship. Symptoms, both physical and psychological, stem from the snowballing effects of treatments, the expectation and timing uncertainty of inevitable relapse, and the need for recurrent treatments . Patients with MM have been described as having worse physical function and global mental health than the general population, as well as diminished health-related QOL when compared to those with other HM . Physically, patients experience fatigue, pain, breathlessness, nausea, muscle weakness, and peripheral neuropathy . Psychologically, patients report the impact of social isolation, financial stress, relationship strain, anticipatory grief, and the toll of endless and unrelenting treatment. A recent cross-sectional, multisite study of 180 MM patients reported that nearly 25% of patients reported clinically significant depression, anxiety, and post-traumatic stress symptoms . There is growing evidence that the caregivers of patients with MM also struggle with psychological symptoms and could benefit from PC support. A cross-sectional, multisite study of 127 MM caregivers revealed that 44.1% have clinically significant anxiety, while another 24.4% reported post-traumatic stress symptoms. Caregivers reported higher rates of anxiety than the patients with MM themselves . Prognostic misunderstanding and patient–provider communication was thought to be a major contributor of stress for caregivers of patients with MM in this study.
Despite the high illness burden experienced by both MM patients and caregivers, there is a paucity of data on models for PC integration in the care of MM patients. PC integration research is needed in the MM population. During periods of disease progression, the possible benefits of PC involvement seem clear. During the prolonged periods of disease control, however, the ideal integration and involvement of palliative services is less apparent. We suspect that patients may still benefit from symptom assessments, psychosocial support, and assistance with coping during this period marked by survivorship and the anticipation of inevitable relapse. Assessing for unmet palliative-related needs and consideration of PC involvement is crucial throughout the trajectory of care.
Treatment paradigms Lymphomas are a heterogenous group of diseases. Lymphomas are divided into Hodgkin’s lymphoma (HL) and non-Hodgkin’s lymphoma (NHL). While HL have a chance of cure with intensive chemotherapy and a fair prognosis overall (approximately 90% 5-year survival), NHLs vary in severity, treatment responsiveness, and prognosis . There are over 3 dozen subtypes of NHL with presentations spanning the full range from indolent to aggressive. Aggressive NHL tends to respond favorably to chemotherapy and may be curable. Diffuse large B cell lymphoma (DLBCL), for example, carries an expected cure rate around 40–50% with multiagent chemotherapies. However, there are genetic and other risk factors, such as relapsed DLBCL, which may confer worse prognosis, especially if unresponsive to initial treatments . Indolent NHL, such as “follicular lymphoma,” may be incurable, but often with long expected survival rates. Indolent NHL may be experienced similarly to patients with some chronic leukemias in that the initial discovery of the disease may be met with “active surveillance” and without treatment. In general, the median survival is often greater than ten years. After treatment courses, patients usually experience periods of remission, sometimes lasting years, but disease relapse and progression are inevitable. Over time, multiple relapses and lines of treatment devolve into diminished treatment responsiveness, progressive decline, and steady disease progression. As with chronic leukemias, “transformation” to a phenotypically more aggressive variant with poor prognosis is possible . Symptom burden and palliative care needs Patients with lymphoma experience high symptom burden and PC-related needs. In HL patients, emotional and physical distress is common . NHL patients experience high rates of financial toxicity and physical symptoms. Fatigue, in particular, can be severe, debilitating, and persistent even in survivorship . More than 50% of NHL patients experience substantial treatment toxicities and high care utilization at the end of life . One study of 91 older NHL patients with aggressive disease demonstrated that in the last 30 days of life, 70% were hospitalized, one-third received systemic therapies, nearly one-quarter underwent admission to an intensive care unit, and more than half died in a healthcare facility. Fewer than half of these patients received PC consultation and even fewer were referred to hospice . As newer treatments bring improved prognosis for both types of lymphoma, patients are increasingly having to contend with issues of survivorship including persistent physical symptoms, post-traumatic stress, and financial toxicity . Palliative care integration and next steps Robust clinical trials examining PC integration in lymphoma care have not been conducted to date. Further research is needed to identify the optimal approach to PC integration in lymphoma care. We suspect that lymphoma patients may benefit from symptom screening, advance care planning, and an extra layer of support at numerous time points throughout the disease course .
Lymphomas are a heterogenous group of diseases. Lymphomas are divided into Hodgkin’s lymphoma (HL) and non-Hodgkin’s lymphoma (NHL). While HL have a chance of cure with intensive chemotherapy and a fair prognosis overall (approximately 90% 5-year survival), NHLs vary in severity, treatment responsiveness, and prognosis . There are over 3 dozen subtypes of NHL with presentations spanning the full range from indolent to aggressive. Aggressive NHL tends to respond favorably to chemotherapy and may be curable. Diffuse large B cell lymphoma (DLBCL), for example, carries an expected cure rate around 40–50% with multiagent chemotherapies. However, there are genetic and other risk factors, such as relapsed DLBCL, which may confer worse prognosis, especially if unresponsive to initial treatments . Indolent NHL, such as “follicular lymphoma,” may be incurable, but often with long expected survival rates. Indolent NHL may be experienced similarly to patients with some chronic leukemias in that the initial discovery of the disease may be met with “active surveillance” and without treatment. In general, the median survival is often greater than ten years. After treatment courses, patients usually experience periods of remission, sometimes lasting years, but disease relapse and progression are inevitable. Over time, multiple relapses and lines of treatment devolve into diminished treatment responsiveness, progressive decline, and steady disease progression. As with chronic leukemias, “transformation” to a phenotypically more aggressive variant with poor prognosis is possible .
Patients with lymphoma experience high symptom burden and PC-related needs. In HL patients, emotional and physical distress is common . NHL patients experience high rates of financial toxicity and physical symptoms. Fatigue, in particular, can be severe, debilitating, and persistent even in survivorship . More than 50% of NHL patients experience substantial treatment toxicities and high care utilization at the end of life . One study of 91 older NHL patients with aggressive disease demonstrated that in the last 30 days of life, 70% were hospitalized, one-third received systemic therapies, nearly one-quarter underwent admission to an intensive care unit, and more than half died in a healthcare facility. Fewer than half of these patients received PC consultation and even fewer were referred to hospice . As newer treatments bring improved prognosis for both types of lymphoma, patients are increasingly having to contend with issues of survivorship including persistent physical symptoms, post-traumatic stress, and financial toxicity .
Robust clinical trials examining PC integration in lymphoma care have not been conducted to date. Further research is needed to identify the optimal approach to PC integration in lymphoma care. We suspect that lymphoma patients may benefit from symptom screening, advance care planning, and an extra layer of support at numerous time points throughout the disease course .
Treatment paradigms Myelodysplastic syndromes (MDS) and myeloproliferative neoplasms (MPN) are another widely heterogeneous group of disorders. These disease processes result from mutations occurring in the stem cells of the bone marrow. It is possible, though not necessary, for one patient to have features of both MPN and MDS. MPN are a sub-type of various HM in which the bone marrow cancerously produces leukocytes, erythrocytes, or platelets, leading to (a) CML (as described previously), (b) polycythemia vera, (c) essential thrombocythemia, or (d) myelofibrosis, respectively. MDS occurs when the bone marrow fails to produce appropriate quantities of mature and functional blood cells and, instead, produces immature and dysplastic cells. There are several variants of MDS with substantial phenotypic variability. Higher-risk MDS confers a bleak prognosis with rapid disease progression, high risk of transformation to AML, and poor long-term survival. Presently, the only potentially curative therapy for MDS is HSCT, which carries its own risks and associated burden as described in a separate section below. Patients with MDS frequently require blood transfusions and other invasive treatments. While many new targeted therapies are available for other myeloid diseases like AML and CML, there are few approved treatment options for MDS and MPNs . Symptom burden and palliative care needs MPN and MDS patients experience significantly diminished health-related QOL. Fatigue and dyspnea are common and debilitating physical symptoms . Blood transfusions and HSCT have been found to be helpful in prolonging life and sometimes reducing symptoms of fatigue and dyspnea, though each also impart their own risks. Blood transfusion dependence is onerous—physically, emotionally, and financially. Patients may also experience unexpected urgent hospitalization for bleeding, complications, infections, or transformation to secondary AML. Psychological symptoms are thought to be common, though data are lacking. Patients with high-risk disease, functional impairment, and transfusion dependence carry higher risks of anxiety and depression. Caregivers may suffer similar to worse mental health outcomes when compared to MDS patients . Palliative care integration and next steps There have been no randomized trials to systematically study PC interventions in patients with MDS or MPN. We suspect that that patients with MDS and MPNs and their caregivers would benefit from the development of interventions aimed at promoting serious illness conversations, addressing symptom burden, and alleviating psychological distress. At the end of life, one major area of improvement could be addressing transfusion dependence as an exclusion criterion for hospice care, which impedes many MDS and MPN (as well as leukemias and other HM) patients from engaging with and benefiting from these services. Research is needed to implement supportive and psychosocial interventions for these patients and families. Symptom assessment and consideration for palliative care referral should be pursued throughout the continuum of MDS and MPN care.
Myelodysplastic syndromes (MDS) and myeloproliferative neoplasms (MPN) are another widely heterogeneous group of disorders. These disease processes result from mutations occurring in the stem cells of the bone marrow. It is possible, though not necessary, for one patient to have features of both MPN and MDS. MPN are a sub-type of various HM in which the bone marrow cancerously produces leukocytes, erythrocytes, or platelets, leading to (a) CML (as described previously), (b) polycythemia vera, (c) essential thrombocythemia, or (d) myelofibrosis, respectively. MDS occurs when the bone marrow fails to produce appropriate quantities of mature and functional blood cells and, instead, produces immature and dysplastic cells. There are several variants of MDS with substantial phenotypic variability. Higher-risk MDS confers a bleak prognosis with rapid disease progression, high risk of transformation to AML, and poor long-term survival. Presently, the only potentially curative therapy for MDS is HSCT, which carries its own risks and associated burden as described in a separate section below. Patients with MDS frequently require blood transfusions and other invasive treatments. While many new targeted therapies are available for other myeloid diseases like AML and CML, there are few approved treatment options for MDS and MPNs .
MPN and MDS patients experience significantly diminished health-related QOL. Fatigue and dyspnea are common and debilitating physical symptoms . Blood transfusions and HSCT have been found to be helpful in prolonging life and sometimes reducing symptoms of fatigue and dyspnea, though each also impart their own risks. Blood transfusion dependence is onerous—physically, emotionally, and financially. Patients may also experience unexpected urgent hospitalization for bleeding, complications, infections, or transformation to secondary AML. Psychological symptoms are thought to be common, though data are lacking. Patients with high-risk disease, functional impairment, and transfusion dependence carry higher risks of anxiety and depression. Caregivers may suffer similar to worse mental health outcomes when compared to MDS patients .
There have been no randomized trials to systematically study PC interventions in patients with MDS or MPN. We suspect that that patients with MDS and MPNs and their caregivers would benefit from the development of interventions aimed at promoting serious illness conversations, addressing symptom burden, and alleviating psychological distress. At the end of life, one major area of improvement could be addressing transfusion dependence as an exclusion criterion for hospice care, which impedes many MDS and MPN (as well as leukemias and other HM) patients from engaging with and benefiting from these services. Research is needed to implement supportive and psychosocial interventions for these patients and families. Symptom assessment and consideration for palliative care referral should be pursued throughout the continuum of MDS and MPN care.
Treatment paradigms Chimeric Antigen Receptor (CAR) T cell therapy is an exciting new treatment which is approved for specific hematologic malignancy management, specifically MM, B cell NHL, or ALL . CAR T cell therapy represents a largely unexplored area of palliative-hemato-oncologic care. Little is known about CAR T cell-associated PC needs. Anecdotally, however, these patients often have advanced disease, a generally poor prognosis, and significant symptoms, while awaiting the receipt of CAR T cell therapy. Yet CAR-T therapy can be highly successful at achieving short-term and even sometimes long-term remissions. Manufacturing turnaround time and arduous cell collection requirements for CAR T cell therapy create significant delays and logistical issues in caring for these patients. Symptom burden and palliative care needs CAR T cell therapy carries a risk for cytokine release syndrome, neurotoxicity, and other physical symptoms from treatment including pain, fatigue, and anorexia which may last months after treatment . Early studies of CAR T cell therapy show that these patients experience substantial healthcare utilization, especially at the end-of-life. One study reported that among descendants of CAR T cell therapy, most were hospitalized within 30 days of death, died in a hospital setting, and did not receive PC or hospice services . Furthermore, a recent study found that CAR T cell therapy patients report overly optimistic prognostic impressions and have high rates of psychological distress . Palliative care integration and next steps Integration of PC interventions for patients receiving CAR-T lack current evidence, perhaps owing to its novelty. Our clinical experience is that CAR T cell patients have similar health-related experiences to those patients with acute leukemia in that they are experiencing a “high-risk, high-reward” treatment, which is associated with prolonged hospitalizations, iatrogenic symptoms, and the potential for psychological distress. Available, though limited, data suggests a need for psychosocial interventions to support patient coping . Future research integrating PC into CAR T therapy from treatment planning through survivorship or death could help mitigate the substantial burden of treatment toxicity, prognostic uncertainty, and prolonged hospitalization.
Chimeric Antigen Receptor (CAR) T cell therapy is an exciting new treatment which is approved for specific hematologic malignancy management, specifically MM, B cell NHL, or ALL . CAR T cell therapy represents a largely unexplored area of palliative-hemato-oncologic care. Little is known about CAR T cell-associated PC needs. Anecdotally, however, these patients often have advanced disease, a generally poor prognosis, and significant symptoms, while awaiting the receipt of CAR T cell therapy. Yet CAR-T therapy can be highly successful at achieving short-term and even sometimes long-term remissions. Manufacturing turnaround time and arduous cell collection requirements for CAR T cell therapy create significant delays and logistical issues in caring for these patients.
CAR T cell therapy carries a risk for cytokine release syndrome, neurotoxicity, and other physical symptoms from treatment including pain, fatigue, and anorexia which may last months after treatment . Early studies of CAR T cell therapy show that these patients experience substantial healthcare utilization, especially at the end-of-life. One study reported that among descendants of CAR T cell therapy, most were hospitalized within 30 days of death, died in a hospital setting, and did not receive PC or hospice services . Furthermore, a recent study found that CAR T cell therapy patients report overly optimistic prognostic impressions and have high rates of psychological distress .
Integration of PC interventions for patients receiving CAR-T lack current evidence, perhaps owing to its novelty. Our clinical experience is that CAR T cell patients have similar health-related experiences to those patients with acute leukemia in that they are experiencing a “high-risk, high-reward” treatment, which is associated with prolonged hospitalizations, iatrogenic symptoms, and the potential for psychological distress. Available, though limited, data suggests a need for psychosocial interventions to support patient coping . Future research integrating PC into CAR T therapy from treatment planning through survivorship or death could help mitigate the substantial burden of treatment toxicity, prognostic uncertainty, and prolonged hospitalization.
Treatment paradigms Hematopoietic stem cell transplantation (HSCT) is an intensive and potentially curative treatment for many HM. Between 1957 and 2019, there have been more than 1.5 million HSCT procedures performed worldwide . HSCT requires preparative chemotherapy, which is typically delivered during an often prolonged and intensive index hospitalization. Many patients who undergo the procedure, especially those receiving allogeneic transplants, develop complications including graft-versus-host-disease (GVHD) . Autologous HSCT is less risky and poses no risk of GVHD, but still requires high-dose chemotherapy prior to stem cell rescue. Of the diseases specifically discussed in this article, HSCT is commonly performed for patients with AML, ALL, aggressive lymphomas, MM, or MDS. Autologous HSCT is mostly performed in MM and NHL. We have opted to discuss HSCT separately from these other diseases because these patients have considerable symptoms and specific PC needs. Furthermore, PC integration in stem cell transplant care is an active area of study. Symptom burden and palliative care needs Patients undergoing HSCT have unmet PC needs . HSCT is associated with low health-related QOL and high physical and psychological symptom burden . Physical symptoms are common and sometimes debilitating, perhaps comparable to patients with acute leukemias undergoing intensive high-dose chemotherapies. Psychologically, patients who undergo HSCT are highly likely to develop post-traumatic stress symptoms due to their treatment experience . The prolonged, socially isolating hospitalizations are associated with decreased patient-reported QOL, elevated levels of anxiety, and depressive symptoms including pronounced anhedonia . Thirty-seven percent (37%) of HSCT patients meet criteria for clinically significant depressive symptoms the week after transplant . Patients describe feeling trapped, fearful, discouraged, and powerless . Psychological stressors have been linked to higher risks of GVHD and decreased overall survival . Social isolation has only been intensified by the COVID-19 pandemic . We are just beginning to understand the effect of HSCT on patients’ families and caregivers. Caregivers of HSCT recipients have prolonged and intensive caregiving burden, which has been shown to negatively impact QOL, physical well-being, and mood . Even prior to the procedure, caregivers experience immense anticipatory psychological distress . During HSCT, caregiver distress remains elevated as their loved ones experience treatment toxicities, physical and psychological symptoms, and the prolonged hospitalization and prognostic uncertainty . Palliative care integration and next steps PC integration into HSCT care is an active area of study. Several randomized clinical trials examined the feasibility and efficacy of PC-HSCT integration. A 2016 trial showed that specialty PC services improve psychological well-being and reduce symptom burdens during HSCT [ ••]. Outcomes from the same HSCT cohort six months after transplantation showed longitudinal benefits of PC on QOL, physical symptoms, anxiety, depression, and post-traumatic stress symptoms [ ••, ]. There was also an observed benefit in caregiver QOL and psychosocial outcomes, which prompted a subsequent unblinded, randomized trial conducted in 2020 examining a psychological intervention for caregivers of HSCT patients . A multisite randomized clinical trial of integrated specialist palliative care during the initial transplant hospitalization is ongoing (NCT# NCT03641378). PC integration at the index hospitalization for HSCT is beneficial and necessary. Future work is needed to improve psychological outcomes in patients who undergo HSCT and their caregivers. While recent small studies have examined various stress management interventions and treatment modalities, we must develop and implement PC interventions traversing the continuum of HSCT care which promote coping, improve QOL, reduce symptom burdens, and alleviate distress in HSCT patients and families .
Hematopoietic stem cell transplantation (HSCT) is an intensive and potentially curative treatment for many HM. Between 1957 and 2019, there have been more than 1.5 million HSCT procedures performed worldwide . HSCT requires preparative chemotherapy, which is typically delivered during an often prolonged and intensive index hospitalization. Many patients who undergo the procedure, especially those receiving allogeneic transplants, develop complications including graft-versus-host-disease (GVHD) . Autologous HSCT is less risky and poses no risk of GVHD, but still requires high-dose chemotherapy prior to stem cell rescue. Of the diseases specifically discussed in this article, HSCT is commonly performed for patients with AML, ALL, aggressive lymphomas, MM, or MDS. Autologous HSCT is mostly performed in MM and NHL. We have opted to discuss HSCT separately from these other diseases because these patients have considerable symptoms and specific PC needs. Furthermore, PC integration in stem cell transplant care is an active area of study.
Patients undergoing HSCT have unmet PC needs . HSCT is associated with low health-related QOL and high physical and psychological symptom burden . Physical symptoms are common and sometimes debilitating, perhaps comparable to patients with acute leukemias undergoing intensive high-dose chemotherapies. Psychologically, patients who undergo HSCT are highly likely to develop post-traumatic stress symptoms due to their treatment experience . The prolonged, socially isolating hospitalizations are associated with decreased patient-reported QOL, elevated levels of anxiety, and depressive symptoms including pronounced anhedonia . Thirty-seven percent (37%) of HSCT patients meet criteria for clinically significant depressive symptoms the week after transplant . Patients describe feeling trapped, fearful, discouraged, and powerless . Psychological stressors have been linked to higher risks of GVHD and decreased overall survival . Social isolation has only been intensified by the COVID-19 pandemic . We are just beginning to understand the effect of HSCT on patients’ families and caregivers. Caregivers of HSCT recipients have prolonged and intensive caregiving burden, which has been shown to negatively impact QOL, physical well-being, and mood . Even prior to the procedure, caregivers experience immense anticipatory psychological distress . During HSCT, caregiver distress remains elevated as their loved ones experience treatment toxicities, physical and psychological symptoms, and the prolonged hospitalization and prognostic uncertainty .
PC integration into HSCT care is an active area of study. Several randomized clinical trials examined the feasibility and efficacy of PC-HSCT integration. A 2016 trial showed that specialty PC services improve psychological well-being and reduce symptom burdens during HSCT [ ••]. Outcomes from the same HSCT cohort six months after transplantation showed longitudinal benefits of PC on QOL, physical symptoms, anxiety, depression, and post-traumatic stress symptoms [ ••, ]. There was also an observed benefit in caregiver QOL and psychosocial outcomes, which prompted a subsequent unblinded, randomized trial conducted in 2020 examining a psychological intervention for caregivers of HSCT patients . A multisite randomized clinical trial of integrated specialist palliative care during the initial transplant hospitalization is ongoing (NCT# NCT03641378). PC integration at the index hospitalization for HSCT is beneficial and necessary. Future work is needed to improve psychological outcomes in patients who undergo HSCT and their caregivers. While recent small studies have examined various stress management interventions and treatment modalities, we must develop and implement PC interventions traversing the continuum of HSCT care which promote coping, improve QOL, reduce symptom burdens, and alleviate distress in HSCT patients and families .
Patients with HM undergo intensive and often chronic treatments. They experience prolonged hospitalizations, undergo invasive procedures, and endure toxicities with long-lasting physical and psychological impact. Patients with HM and their families are perhaps the most psychologically distressed of all patients with cancer. Despite the high burden of unmet palliative-related needs, patients with blood cancers are substantially less likely to access PC than are patients with solid tumors. The evidence is clear that the way forward includes standard-of-care PC integration into routine hematologic malignancy care to improve patient and caregiver outcomes, but this may not be required or helpful for all patients and situations. More research is needed to inform the highest need populations and the highest impact interventions. As the PC needs for patients with blood cancer vary significantly by disease, a disease-specific PC integration strategy is needed, allowing for serious illness care interventions to be individualized to the specific needs of each patient and situation. As we have outlined throughout this article and summarized in Table , we are beginning to see the development of a robust evidence base for the integration of PC into standard practice AML, MM, and HSCT care. On the other hand, evidence has lagged in other hematologic malignancy conditions such as lymphoma, chronic leukemias, and MDS/MPN, despite high symptom burden, psychological distress, and poor QOL among these patients and their families. High-quality randomized clinical trials are needed for these specific patient populations to build upon the existing evidence and guide us forward in the care of these patients with serious illness.
|
Newly developed 3D in vitro models to study tumor–immune interaction | bfeacae1-300b-4e00-8655-195316b710e8 | 10074642 | Internal Medicine[mh] | Tumor microenvironment (TME) is a complex and continuously evolving entity . It includes not only the tumor cells, but also diverse supporting cell types, such as activated fibroblasts, blood vessels, infiltrating immune cells, and extracellular matrix . The TME is regulated by a variety of cells, hormones and inflammatory responses, and is important for the understanding of tumorigenesis, as well as the development and metastasis of tumor. So it plays a key role in the prevention, diagnosis and prognosis of tumors and has long been a promising direction in oncology research. For traditional tumor therapeutic approaches, tumor cells are the direct targets of therapy. However, as the role of immunity in tumor development has been constantly illustrated, such as the generation and progression of tumors in immunodeficiency or inflammatory response states, other components of the TME are also emerging as new targets for tumor therapy . These immunotherapeutic approaches enhance the immunosurveillance role of the immune system or locally modulate the tumor immune microenvironment. Tumor immunotherapy, represented by immune checkpoint inhibitor (ICI) therapy and adoptive T cell therapy (ACT), has become an important tool for the treatment of tumor patients, especially those with advanced tumors, in addition to surgery, radiotherapy and targeted therapy . Therefore, basic and clinical translational research on tumor immune microenvironment is necessary . Co-culture of 2D tumor cells with exogenous immune cells are the commonly used tools for basic research on tumor immunology and preclinical trials of relevant immunotherapeutic drugs. It is not only easy and cost effective, but also allows easy gene editing and drug intervention. Besides, the model has a wide range of assays and well-established evaluation techniques, making the experimental process very convenient and reproducible. However, tumor cell linages do not reflect the growth pattern of tumor cells in vivo and the heterogeneity, diversity and individuality of patient-derived tumors. The additional immune cells cannot restore the spatial and temporal characteristics and interaction patterns of the TME, and genetic mutations and phenotypic variations may occur during long-term passaging cultures . As for mouse models, humanized mouse models are increasingly used in tumor immunology research in recent years. Humanized mouse models of the immune system co-transplanted with human tumors are effective tools to study the interactions between immune components and tumors of human origin. Commonly used humanized mouse models include Hu-PBL mouse model, Hu-HSC mouse model and Hu-BLT mouse model . Sanmamed MF et al. inoculated human colorectal HT-29 cells and gastric cancer tissues in Hu-PBL mice and administered urelumab (anti-hCD137) or/and nivolumab (anti-PD-1) treatment. Tumor growth was suppressed by combination therapy or monoclonal antibodies alone, but combination therapy did not significantly improve the efficacy. A humanized mouse model of breast cancer was constructed by transferring HSPC, peripheral T cells from the same donor and breast cancer cell lines into NOD-SCID B2m -/- mice, in which CD4 + T cells were found to promote early tumor development through a DC-dependent pattern and the process could be partially inhibited by antagonists of IL-13 . These studies demonstrated that the humanized mouse model can simulate the complex systemic immune response and the rich variety of immune cell interactions in human, reflecting of tumor growth and drug sensitivity in vivo. However, the simulation of human immune system is not perfect, and the time window is often limited. The high incidence of graft versus host disease also affects the success rate. Moreover, the construction process is complicated, with high technical threshold and cost, which is not suitable for large-scale experimental studies and high-throughput drug screening . Therefore, new models need to be constructed for tumor immunology studies. 3D in vitro models are newly developed techniques that can simulate the 3D structure in vivo, thus reproducing the physiology and physiopathological characteristics of the original tissues . These models can partially restore the complex structure of the tumor, preserve tumor heterogeneity as well as genotypic and phenotypic characteristics. This kind of technology is represented by organoid and tumor spheroid model, but also includes air-liquid interface (ALI) culture model, microfluidic culture model, and tissue engineering based 3D-bioprinting model. However, some 3D in vitro models, such as traditional organoid culture model, lack matrix components including immune cells, thus limiting their applications in the study of tumor immune microenvironment. In recent years, with the continuous progress of 3D in vitro culture technology and the model optimization, there are more and more models that can partially remodel the immune microenvironment in vitro and thus be used in the study of tumor immunology, providing a new pathway in this research field . They have some unique advantages over the 2D tumor co-culture models and humanized mouse models but also exists some limits (Table ). Combining 3D in vitro culture technology and other research models will greatly promote the tumor immunology research. This review introduces the commonly used 3D in vitro models that can remodel the immune microenvironment of tumors, compares their culture methods and applications, and provides a preliminary discussion on the current problems and possible future development directions, hoping to provide a reference for further research (Fig. ). Part 1: 3D in vitro models for rebuilding the tumor immune microenvironment Organoid/immune cell co-culture models An organoid initially refers to a 3D structure grown from stem and progenitor cells. It consists of variant organ-specific cell types and can self-organize via cell differentiation and spatially restricted lineage commitment . Organoid can better preserve the genotypic and phenotypic characteristics of the original tumor tissue and retain the heterogeneity of tumor cells. Therefore, it is a more optimal model for tumor research compared to the traditional 2D tumor cell lines . In recent years, organoid technology has been more and more widely used in tumor research. Organoid construction methods can be divided into two types: scaffold-dependent methods and scaffold-free methods. In the scaffold-dependent system, scaffolds provide physical support for cell growth in which cells aggregate and self-assemble to form 3D structures. Depending on the different sources and components, extracellular scaffold can be further divided into matrix derived from decellularized tissue (e.g. Matrigel, de-cellularized ECM from porcine liver tissue), natural polymer-based hydrogel (e.g. methylcellulose, hyaluronic acid, chitosan) and synthetic hydrogel (e.g. PEG, PLGA, PLLA, PVA) . As for scaffold-free methods, hanging drop method, magnetic levitation, rotary cell culture and micromolding technique are also commonly used. In these models, cells spontaneously aggregate to form three-dimensional structures under the action of various external environments such as gravity, magnetic force and rotational agitation. Without the presence of extracellular matrix, these methods allow for precise control of organoid size and morphology and allow for easier co-culture experiments . In the above culture models, tumor stem cells can proliferate and differentiate to form tumor-like structures, but the stromal component of the original tumor tissue cannot be preserved. To apply organoid to tumor immunology research, exogenous immune cells need to be added to interact with tumor organoids. Here we introduce the commonly used methods for co-culturing tumor organoids and immune cells (Fig. A). The easiest way to co-culture organoid with immune cells is by directly adding exogenous immune cells into the culture medium (Fig. A). In this way, immune cells secret cytokine to act on tumor cells, and when immune cells have a high infiltration capacity, they may also enter the matrigel and contact directly with tumor cells. For example, CD3 + T cells isolated from peripheral blood mononuclear cells (PBMCs) were added to the medium of pancreatic cancer organoids. After co-cultured for 72 h, it was found that T cells were viable and Ki67 positive in organoid growth medium but the cell activity was weaker than in RPMI medium. When tumor organoids were present in the matrigel, T cells migrated and infiltrated into matrigel in response to pancreatic cancer cells. Otherwise, they mostly gathered at the edge to form clear boundaries . T cell activation and tumor cell apoptosis could also be observed when different kind of tumor organoids were co-cultured with homologous PBMCs and the addition of PD-1 monoclonal antibody or other immune activators could enhance T cell cytotoxicity, providing a new platform for personalized screening of potential antitumor immunotherapeutic agents . To further increase the contact between tumor organoids and immune cells Lei Yu et al. diluted matrigel to 50% concentration before using it for bladder cancer organoids culture and added specific CAR-T cells to the culture medium. In this case CAR-T cells could better invade the matrigel, thus the killing ability and specificity of CAR-T cells could be better detected. However, the above methods are more suitable for studying the indirect interaction between immune cells and tumor cells. The direct contact is still not sufficient, even if the immune cells have a strong migration and infiltration ability and the matrigel is diluted. Moreover, the components in matrigel may affect the function of immune cells, leading to non-specific activation. Therefore, when considering direct intercellular interactions, mature organoids can also be isolated from the matrigel and cultured in suspension with immune cells (Fig. B). Based on this method, the researchers co-cultured patient-derived MSI-H CRC organoids and NSCLC organoids with homologous PBMCs. The tumoriods were first separated from matrigel and digested into single cell suspensions, which were treated with IFN? to increase the expression of MHC-I molecules on the surface of tumor cells and thus promote antigen presentation. PBMCs isolated from peripheral blood were simultaneously incubated with anti-CD28 and anti-PD-1 antibodies to provide a co-stimulatory signal and counteract the upregulation of PD-L1 expression caused by IFN? treatment. The treated tumoroid single cells were mixed with PBMCs at a certain ratio and co-cultured in T-cell medium containing IL-2 for 7 days. After 2 rounds of co-culture, the T cells were activated and can specifically kill the corresponding tumoroids. In this way, tumor-reactive T cells could be enriched in peripheral blood and their tumor-killing efficiency could be measured at the individual level, providing a good model for cell therapy optimization . Qingda Meng et al. applied a similar method to isolate human pancreatic cancer organoids and mixed them with PBMCs in T-cell medium with IL-2, IL-15, IL-21 and other factors required for immune cell growth. T cells were found to kill tumor cells in a granzyme B or Fas-FasL-dependent manner during co-culture, and the co-cultured T cells exhibited markers of tissue-resident memory phenotype. Each patient’s co-cultured T cells exhibited their own unique expression of immune checkpoint proteins. But this method is only appropriate for studies of short duration, and as the separation time of the organoids from the matrigel increases, adherence or death of the organoids may occur. In order to prevent this situation, the co-culture protocol can be further optimized. Mei Song et al. pre-coated 24-well plates with matrigel and then mixed F4/80 + CD11b + tumor associated macrophages (TAMs) from ovarian cancer-bearing mice with ID8 tumor cells at a ratio of 1:10 in medium containing 2% matrigel. The mixture was planted in the pre-coated plates to investigate the role and mechanism of TAMs in the growth and metastasis of ovarian cancer. This method allowed direct contact between tumor cells and immune cells, and also prevented the adherence of organoids, making the co-culture system more stable and suitable for long-term studies (Fig. B lower part). A similar approach has also been applied to the co-culture of gastric tumoriods with CD8 + T cells and ER + breast tumor organoids with breast fibroblasts . Besides, as another widely used 3D in vitro culture model, tumor spheroids are typically created in a scaffold-free environment by placing cells into suspension colonies without the aid of extracellular matrices (ECMs) or other physical supports . Due to the lack of physical support from the ECMs, tumor spheroids are less complex and cannot completely rebuild the structure in vivo, but they are more convenient for co-culture with immune cells. Tristan Courau et al. cultured tumor cells into tumor spheroids and added T cells and NK cells isolated from peripheral blood to the medium, where immune cells could make direct contact with the tumor spheroids and infiltrate to produce killing effect. Further examination of molecule expression patterns of infiltrated immune cells and tumor cells revealed that immunomodulatory antibodies targeting MICA/B and NKG2A had anti-tumor potential. Thus the tumor spheroid and immune cell co-culture model is a usable tool to study tumor-lymphocyte interactions. In addition, Monocyte-derived macrophages were co-cultured with patient-derived colorectal cancer tumor spheroids in ultra-low attachment plates, and it was found that the co-cultured tumor spheroids were more prone to “budding” (Fig. B upper part). Immunohistochemical staining showed obvious macrophage aggregation at the budding site and the budding was accompanied by a reduction of intercellular tight junctions. Therefore, macrophages may play a role in the invasion and metastasis of colon cancer . In addition to the two co-culture methods mentioned above, it is also possible to co-culture tumor cells and immune cells by mixing them directly in matrix (Fig. C). Following this idea, Jayati Chakrabarti et al. co-cultured mouse gastric tumoriods expressing the activated GLI2 allele GLI2A with DC cells and cytotoxic T lymphocytes (CTLs) in matrigel, and added PD-L1 inhibitors to the co-culture system. They found that CTLs had a significant killing effect on tumor cells. Vivien Koh et al. found that immunotherapy failed to activate CTLs to kill tumor cells in the presence of myeloid-derived suppressor cells (MDSCs). Xi He et al. used the same method to investigate the effect of myeloid-derived cells (MDCs) on the growth of the mouse small intestine adenoma. After co-culturing MDCs with organoids in equal propotions in matrigel for 8 days, it was found that most of the MDCs were transformed into TAMs and stimulated the growth of adenoma organoids. Together, the above studies demonstrate the usability of such co-culture models in reconstructing tumor immune microenvironment in vitro and in studying the intercellular interactions. ALI culture model The ALI culture is a special organoid culture model. In this model, tumor tissues are mechanically minced into pieces of about 1 mm in diameter without enzymatic digestion, mixed with collagen gel and layered on top of pre-solidified collagen gel within a permeable, membranous inner transwell insert. The insert is placed in a cell culture dish, and culture medium is added between the dish and the insert to form the double dish culture system. This way, the upper part of mixture of tumor tissues and collagen is directly exposed to air, and the culture medium infiltrates the bottom of the collagen through the micropores to create an air-liquid interaction model (Fig. B). The ALI culture method can increase the oxygen supply in system, and for respiratory, gastrointestinal or body surface tumors, the air-liquid interaction also better mimics the tumor growth environment in vivo. In this system, tumor cells can grow normally to form tumoroids, which retain the pathological features and genetic alterations of the original tumor. Also, the immune cells and fibroblast stroma of the tumor tissue can be maintained for a certain period of time. Using this approach, James T. Neal et al. established organoids of various origins, such as colorectal, kidney, lung, pancreatic, and thyroid cancers, and found that a variety of immune cells including T cells, B cells, NK cells, and macrophages from the tumor tissues were preserved in these organoids, and the types and proportions of immune cells were highly consistent with the original tumor. Although the amount of immune cells gradually decreased with the extension of culture time, T cells could be retained for more than 30 days and even after organoid passaging with the addition of IL-2 in the culture medium. TCR sequencing revealed a high similarity between the retained T cells in organoids and the T cells contained in original tumors. After the addition of nivolumab to this model, T cell activation marker IFNG, GZMB and PRF1 were significantly increased, and the tumor cells also underwent significant apoptosis, indicating that this model could successfully mimic the response process of tumor immunotherapy and was expected to be applied to evaluate the efficacy and sensitization methods of ICB therapy. Similarly, Laura K. Esser et al. cultured 42 surgically resected renal clear cell carcinoma specimens using the ALI method. Immunohistochemical staining and RNA sequencing verified that the ALI cultured patient derived organoids (PDOs) were highly similar to the characteristics of the tumor specimens and that the immune and stromal cells in PDOs could be preserved to some extent. The authors treated 10 cases of ALI PDO with the targeted cancer drug cabozantinib or the nivolumab and found that PDOs of different patient origins showed very different responses to the two treatments, showing that this model could reflect the different treatment effects of patients. In this model, the survival time of primary immune cells in tumor tissues is the key issue due to the lack of exogenous immune cells. In another study, patient-derived CRC organoids were cultured using ALI method and the survival time of immune cells were evaluated. The results showed that a certain percentage of CD45 + hematopoietic cells were still present after 8 days of culture but the number of CD3 + T cells decreased significantly, suggesting that the culture environment with organoid medium might not be the most suitable for T cell survival and further addition of T cell-promoting growth factors was needed to adjust the culture conditions . In addition to tumor organoids, Lisa E. Wagar et al. also used the ALI method to culture lymphoid organs including lymph nodes, tonsils, and spleens. They minced and digested these lymphoid organs into single cells, which were plated into pre-coated transwell chambers along with relevant antigens. Complete culture medium was added in the lower chamber supplemented with a small amount of recombinant human B cell-activating factor (BAFF) to improve total B cell survival. After several days in culture, clustered cells formed the reaggregated regions. They accessed the organoids and found that they could recapitulate key germinal center features of lymphoid organs in vitro, including the affinity maturation, somatic hyper-mutation and the production of antigen-specific antibodies. Using this system, the authors studied the immune responses against pathogens, such as the identification of key cellular components in the generation of anti-influenza virus immune responses. For further study, lymphoid organoids co-cultured with tumor cells also holds promise for applications related to the study of adaptive immune responses against tumors, the identification and assessment of tumor vaccines, and the investigation of lymph node metastasis of tumors. 3D-Microfluidic based culture method Microfluidics refers to a technique for controlling the flow of micro fluids (10 - 9 to 10 - 18 L) in microscopic channels (10 - 4 to 10 - 5 m in size) . In this system, the size of the microchannels and the volume ratio of cell to extracellular fluid are very similar to those in the TME. And because of the low Reynolds coefficient, the fluid in this system flows in a laminar way, allowing the soluble factors to form a concentration gradient similar to that in vivo. So microfluidic culture system acts as a good model for simulating the TME . Combining microfluidic devices and organoid technology not only improves the homogeneity and controllability of organoids through the design of microfluidic channels and the control of liquid flow rate, but also makes the high-throughput production of organoids possible (Fig. C). Microfluidic devices have already had a wide range of applications in modeling the TME in vitro. First, because multiple different cell types can be manipulated independently in microchannels, co-culture of tumor organoids with additionally added cells is possible. Muhammad R. Haque et al. developed a tumor microarray device to mimic the TME of PDAC by merging PDOs and stromal cells, including pancreatic stellate cells (PSCs) and macrophages. In this multicellular microfluidic system, they successfully established a complex organotypic tumor environment containing connective tissue proliferating stroma and immune cells. Drugs targeting PSCs or macrophages in this model significantly increased the ability of chemotherapeutic agents to kill tumor cells, whereas this response was absent in the tumor cell culture system without stromal cells. This indicates that this system partially mimics the tumor microenvironment and can be used for drug screening targeting the TME. On the other hand, the original immune cells and stromal components are also better preserved in this system due to the higher similarity of the microfluidic device to the in vivo states. Using the microfluidic system to create patient- and mouse-derived organoids, it was found that immune cells native to the patient or mouse tumor tissues could be retained and could respond to PD-1 monoclonal antibody treatment in vitro. Organoids from patients whose tumors were sensitive or resistant to anti PD-1 therapy maintained the same drug responsiveness in vitro. In addition, investigators have used this system to screen for small molecule drugs to sensitize PD-1 monoclonal therapy and found that both CDK4/6 and TBK1/IKKe inhibitors in combination with PD-1 monoclonal antibodies enhanced antitumor immune response. The drug efficacy was further confirmed in in vivo models . Recently, Shengli Ding et al. developed a novel droplet-based microfluidic 3D culture platform to generate a large number of micro-organospheres (MOSs) from a small amount of tissues of cancer patients. The key step is to prepare the single cell suspension of primary tissues from tumor patients, add it to 3D matrigel, and then mix with oil phase liquid to generate MOSs, which can be cultured in suspension after demulsification to remove excess oil. First, the authors evaluated the consistency of MOSs and immune cells in the corresponding tumor samples using single cell transcriptomics and found that MOSs retained tumor-associated fibroblasts as well as myeloid and lymphoid immune cells similar to the original tissues. The expression of immunosuppression-related markers also had high consistency. In addition, this model can be used for co-culture with exogenous immune cells. Due to the smaller size and larger surface area-to-volume ratio of MOSs, both additional TIL and PBMC can more easily infiltrate into MOSs to contact and effectively kill tumor cells, providing a useful tool to assess the efficacy of ACT. 3D-bioprinting model 3D-bioprinting as an emerging technology, is an in vitro 3D structural model manufactured with biological units (cells/ proteins/ DNA etc.) and biological materials according to the requirements of bionic morphology and organism function using 3D printing techniques . Conventional organoids are formed by proliferation, differentiation and self-assembly of stem cells, so they lack control over cell number, cell type ratio and microenvironment. The 3D-bioprinting technology can reconstruct the complex structure of organoids through accurate and stable model construction and multi-cell controlled organoid printing, which can simultaneously print multiple cell components, ECMs and cell growth factors, thus effectively improve the reconstruction of microenvironment in organoids (Fig. D). In 2020 Kunyoo Shin’s team introduced a new concept of a mini-organ called an assembloid based on 3D-bioprinting technology to mimic human tumor tissues in terms of structure and function. The team constructed patient-specific bladder assembloids by 3D-bioprinting. This model not only maintained the genetic changes of the parental tumor but also introduced TME components, revealing that signaling between tumor cells and stromal cells played a key role in controlling tumor plasticity . Marcel Alexander Heinrich et al. used 3D-bioprinting to construct mini-brain by mixing glioma cells (GL261) with macrophages (RAW264.7). On the basis of this model, they investigated the interaction between macrophages and glioma cells and found that glioma cells could recruit macrophages and cause them to generate a glioma-associated macrophage phenotype. Macrophages also promoted the proliferation and invasion of glioma cells. Similarly, Hermida et al. constructed an in vitro brain glioblastoma model using extrusion-based bioprinting. The model was based on alginate modified with RGDS cell adhesion peptide, hyaluronic acid and type I collagen, and integrated multiple cell types including tumor cells, microglias and tumor stromal cells. Compared with tumor cells alone, the 3D-printed tumor model was more resistant to chemotherapeutic drugs, reflecting the role of TME on chemotherapy. The combination of different models The above 3D in vitro models have been widely used in simulating TME and tumor immune-related studies. They all have their own advantages and disadvantages and are suitable for different application scenarios. The following table summarizes and compares the characteristics of different models (Table ). However, the ability to reconstruct the TME in vitro using a single model is limited due to the defects of each model, and therefore the combination of multiple models represents more promising applications. Zhiyi Gong et al. designed a platform applying acoustic droplet printing to fabricate mouse bladder cancer organoids. The organoid produced by this platform could retain the original immune cells in the tumor tissues within 2 weeks. In addition, by placing the manufactured organoids in a microfluidic chip, the size and morphology of the organoids could be further controlled. High-throughput organoid manufacturing, drug screening, and real-time imaging and evaluation of organoids could also be achieved. Subsequently, the authors co-cultured the organoids with homologous spleen-derived immune cells for 2 days, and the infiltration of lymphocytes into the organoids was observed. This can be used to better screen for tumor-responsive T cells and the cells expanded in vitro also have the ability to kill the tumor organoids. Similarly, Konstantinos I. Votanopoulos et al. first mixed patient’s melanoma cells with immune cells derived from the same patient’s lymph nodes in matrigel to form immune-enhanced tumor organoids (iPTOs). Then they applied the 3D-microfluidic system to circulate peripheral blood T cells of the same patient origin around the iPTOs and found that the co-cultured T cells also had the ability to kill tumor cells, reflecting the possible role of iPTOs in the induction of adaptive immunity. The above two studies combined multiple in vitro models to integrate immune cells within the TME and the different species of exogenous immune cells in one system, thus more comprehensively mimic the process of antitumor immunity and have more diverse applications in tumor immunology research. Part 2: Application of 3D in vitro models in tumor immunology research Application of 3D in vitro models in mechanism research of the TME Tumor cells and their microenvironment are a functional entity. The microenvironment interacts and co-evolves with tumor cells and play important roles in multiple processes of tumor development. Therefore, studying the interaction between tumor and microenvironment can help us understand the biological behavior of tumor and lay the theoretical foundation for finding new therapeutic targets and exploring new methods of tumor immunotherapy. 3D in vitro models are good models for studying the TME because of their simplicity, convenience, flexibility and high similarity to original tumor tissues (Fig. E). For example, in tumor microenvironment, T cells and NK cells are usually the main ones that can directly kill tumor cells. Enhancing their killing ability contributes to tumor control. Using a co-culture model of CRC organoids and T cells, researchers screened a series of small molecule inhibitors and found that DKK1 inhibitors could significantly enhance the killing effect of T cells and promote apoptosis of tumor cells by regulating the GSK3ß-E2F1-T-bet axis in CD8 + T cells. DKK1 inhibitors combined with PD-1 monoclonal antibodies could achieve better tumor control . Similarly, BRD1 inhibitors were able to convert NK cells and naïve CD8 + T cells to a more activated and cytotoxic phenotype, helping anti-PD-1/PD-L1 bispecific antibodies to exert a more effective antitumor immune effect in HGSC . By detecting the expression of surface biomarkers on tumor cells and immune cells before and after co-culture, antibodies targeting MICA/B and NKG2A were also found to enhance the killing effect of T cells and NK cells on colorectal cancer organoids, thus providing a potential target for the treatment of CRC . In addition to immune cells that play a major role in tumor killing, there are multiple immunosuppressive components in the TME that can promote tumor progression. Co-culture of mouse MDCs with small intestinal adenoma organoids revealed that MDCs were transformed into TAMs and stimulated adenoma growth via the COX-2-PEG2-EP4 pathway . Using mouse ovarian cancer tumor spheroids co-cultured with TAMs, it was found that when UBR5 was knocked down, the tumor spheroids had slower growth and smaller size. The ability to recruit TAMs was also reduced, as was the expression of cytokines and chemokines associated with TAMs recruitment. This demonstrated that targeting UBR5 could help control the growth of ovarian cancer by modulating the tumor immune microenvironment . Using the similar co-culture method, researchers found that breast fibroblasts secreted cytokines such as IL-1ß, which acted on breast cancer cells through a paracrine pathway to promote their proliferation. Blocking this paracrine pathway enhanced the therapeutic effect of tomaxifen on breast cancer, suggesting that fibroblasts in the remaining breast tissue after breast-conserving surgery may increase the risk of breast cancer recurrence . These studies have explored the interaction of tumor cells with different components of the microenvironment through co-culture models, providing a direction for the development of immunotherapy. Application of 3D in vitro models in assessing the efficacy of ICI treatment for personalized treatment With the increasing research on the immune microenvironment, immunotherapy has been more and more widely used in the treatment of tumors, among which the most frequently applied is immune checkpoint inhibitor therapy. ICI therapy targeting PD1/PD-L1 and CTLA-4 has been clinically employed in progressive melanoma , squamous cell skin cancer , non-small cell lung cancer , renal cell carcinoma , head and neck tumors, and tumors with mismatch repair defects of various tissue types. However, only some of the patients treated with ICI have benefited from this therapy. Considering the adverse effects and financial burden, it is necessary to carefully select patients who may benefit from ICI treatment and personalize the treatment with precision. For precise patient selection, a number of biomarkers have been demonstrated to predict the efficacy of ICI therapy, such as MSI status , TMB , neoantigen expression levels , CD8 + T cell counts , and tumor cell surface PD-L1 expression levels , but none of these indicators have sufficiently high positive and negative predictive values to allow effective screening of patients . Other studies have analyzed immunobiological indicators of the tumor, such as genomic, transcriptomic and proteomic sequencing analysis, or have integrated multiple biomarkers, but even these more complicated analyze methods do not fully reflect the intratumoral and individual heterogeneity of human tumors and cannot accurately predict ICI treatment efficacy . Therefore, static predictors cannot meet clinical needs. It is important to establish an efficacy assessment model that can be monitored dynamically, and the in vitro 3D culture may fill the gap in this field (Fig. F). Paula Voabil et al. cultured tumor tissues of multiple patient sources in vitro and added PD-1 monoclonal antibody to the culture system for 48 h of co-incubation. Changes in 13 cytokines, 13 chemokines and 4 T cell activation markers were examined before and after the addition of the drug to generate a response score, and they found that the response score of the tumor tissues to the drug was highly consistent with the clinical response of patients, indicating that this model had a potential to predict the efficacy of early treatment with PD-1 monoclonal antibody. In addition, the authors performed a multifaceted analysis of tumor tissues in the untreated state, including the proportion and spatial distribution of immune cells and various cytokines and chemokines such as CXCL9, CXCL10, CXCL13, IL-8, and established a method to predict the efficacy of PD-1 monoclonal antibody therapy based on the baseline tumor condition. In another co-culture model, melanoma organoids were mixed with lymph node immune cells of the same patient origin in matrigel. 85% (6/7) of the organoids responded to immunotherapy with nivolumab, pembrolizumab, ipilimumab, and dabrafenib/trametinib in the same way as the actual dosing response in the clinic . Similarly, Myriam Chalabi et al. selected six nonresponder and six responder patients in a cohort of early-stage colon cancer patients receiving anti-PD-1 combined with anti-CTLA-4 neoadjuvant immunotherapy and constructed a co-culture model with PBMC. The results of the in vitro experiments reflected the drug efficacy of the patients to some extent, as T cells were activated and exhibited tumor cell killing in three responder patients but showed no reactivity to tumor organoids in nonresponder patients. The fact that three of the six responder patients did not show T-cell reactivity suggested that this model needs to be further optimized for predicting the efficacy of ICI therapy more accurately. Although 3D in vitro models has the potential to be used as a preliminary predictor of efficacy, the published studies are mainly small sample size studies. More large scale studies are needed in the future to verify the accuracy of efficacy prediction. Application of 3D in vitro models in drug screening and immunotherapy optimizing Despite the increasing application of immunotherapy, its effectiveness is still limited. Only a minority of patients can benefit from immunotherapy, and even in patients who are sensitive, problems of acquired drug resistance may occur. Therefore, new strategies are needed to optimize the effectiveness of immunotherapy. 3D in vitro models have been widely used in the evaluation of different treatment options and in vitro drug screening because of their ability to better reflect tumor characteristics and manipulability (Fig. G). For example, the response of patients to neoadjuvant chemoradiotherapy for colorectal cancer is highly consistent with the in vitro responsiveness of their corresponding organoids to radiotherapy and the same chemotherapeutic agents, and the in vitro drug sensitivity results are expected to guide the clinical treatment of patients . High-throughput screening of drugs using organoids can also help to find new therapeutic options for tumors that are resistant to conventional therapy . However, due to the lack of immune components, traditional organoid models cannot play a role in exploring more optimal immunotherapy regimens. In contrast, multiple 3D in vitro models that integrate immune cells can be useful in the optimization of immunotherapy. PD-1/PD-L1 inhibitors are commonly used immunotherapies in clinical practice, so enhancing the efficacy of PD-1/PD-L1 therapy through different pathways is at the forefront of research. Carminia Maria Della Corte et al. used a co-culture system of NSCLC to verify the synergistic effect of PD-L1 monoclonal antibody combined with MEK inhibitors in the treatment of NSCLC. They found that MEK inhibitors not only had a direct killing effect on tumor cells, but also promoted tumor recognition by CD8 + T cells, increased the expression of cytokines such as IFN?, IL12, IL6 and TNFa, and prevented T cell depletion by downregulating PD-L1, CTLA-4, TIM-3 and LAG-3. These results showed that MEK inhibitor had PD-L1 monoclonal sensitizing effects. Small molecule drugs were screened using a 3D microfluidic organoid culture system, and CDK4/6 inhibitors were found to significantly increase T cell activation and infiltration. Simultaneous application of PD-1 monoclonal antibodies and CDK4/6 inhibitors also showed enhanced T cell killing of tumor cells . Meanwhile, screening of epigenetic inhibitor library and herbal small molecule compound library using 3D in vitro model revealed that GSK-LSD1, CUDC-101, BML-210 and ATT-1 could increase the expression of MHC-I on tumor cell and promote tumor antigen presentation respectively, and the combination with PD-1 antibody could enhance the killing toxicity of CD8 + T cells . In addition to anti PD-1/PD-L1 therapy, other immunotherapeutic approaches can also be studied using 3D in vitro models. Qingda Meng et al. co-cultured pancreatic tumor organoids with PBMCs and added inhibitors of various immune checkpoints such as PD-1, PD-L1, TIM3, TIGIT, LAG3, and NKG2A to the co-culture system. NKG2A inhibitors were found to significantly elevate IFN-? expression in T cells, and the blockade of NKG2A-HLA-E axis was found to be a potential target for enhancing the killing capacity of CD8 + T cell for the treatment of pancreatic cancer. Marcel Alexander Heinrich et al. applied a 3D-bioprinted glioma model to study two immunomodulatory drugs AS1517449 (Stat6 inhibitor) and BLZ945 (Csf-1r inhibitor), which target macrophages. It was found that the function of macrophages was significantly inhibited after BLZ945 treatment, as evidenced by the decrease in Fgf2 and Mmp2 expression in macrophages. In addition, the growth rate of glioma cells was significantly slowed down. This reflected that targeting macrophages may also play a role in tumor immunotherapy. Application of 3D in vitro models in adoptive cell therapy In the field of tumor immunotherapy, apart from immune checkpoint inhibitors, adoptive cell transfer therapy (ACT) is also developing rapidly. This treatment involves isolating immunologically active cells from tumor patients, expanding and functionally characterizing them in vitro, and then re-infusing them into patients for the purpose of killing the tumor directly or stimulating immune response to eliminate tumor cells. According to the development of ACT, the adopted cells include lymphokine-activated killer cell (LAK), cytokine-induced killer cell (CIK), tumor infiltrating lymphocyte (TIL), natural killer cell (NK), cytotoxic T lymphocyte (CTL), chimeric antigen receptor T cell (CAR-T) and T cell engineered with T-cell receptor (TCR-T) . Although ACT has been successful in hematologic malignancies, especially CAR-T cells targeting CD19 in B cell lymphoma and acute lymphocytic leukemia , breakthrough in solid tumors has yet to be achieved, and 3D in vitro models are being broadly used in this field (Fig. H). Firstly, 3D in vitro models can be used as a source of tumor antigens for the preparation of tumor-specific T cells because of their high consistency with the original tumor. Krijn K. Dijkstra et al. co-cultured and screened PBMCs with CRC and NSCLC organoids for several rounds to obtain tumor-reactive T cells for T cell therapy. The iPTOs model constructed by mixing organoids and homologous immune cells could also be used to induce peripheral blood T cells to produce killing effects on tumor cells . Qingda Meng et al. performed TCR sequencing on peripheral blood T cells after co-cultured with pancreatic cancer organoids and found that a specific subpopulation of T cell clones expanded significantly. Cloning and transferring TCRs from this subpopulation into heterologous T cells could enable T cells to acquire the ability to specifically recognize and kill patient tumor cells, providing a potential idea for advanced adoptive cell therapy. Secondly, 3D in vitro models can also be used to assess the efficacy of adoptive cell therapy. By co-culturing the specific immune cells used in the adoptive cell therapy with organoids or other 3D in vitro cultures, the activation of immune cells and the apoptosis of tumors can be assessed. Fadi Jacob et al. co-cultured six glioma organoids which had different levels of EGFRvIII expression with 2173BBz CAR-T cells targeting EGFRvIII. After 72 h co-culture, they found that CAR-T cells infiltrated the organoids in all groups, but only when co-cultured with EGFRvIII positive organoids, CAR-T cells showed significant expansion, increased expression of granzyme and secretion of various cytokines. Increased apoptosis was found in EGRFvIII + tumor cells but not negative cells. This indicated that CAR-T cells were able to specifically kill target cells instead of complete elimination of all tumor cells. Such co-culture models provide a feasible way to test the efficacy of CAR-T therapy. Similarly, co-culture of human bladder cancer organoids with specific CAR-T cells allows detecting the killing ability and specificity of CAR-T cells . CAR-NK92 cells targeting EGFRvIII or FRIZZLED can also exhibit tumor antigen-specific cytotoxicity when co-cultured with CRC organoids in the ALI system . Evaluation of cell therapy efficacy can help optimize the therapeutic approach and guide personalized and precise treatment of patients. Part 3: Current problems and shortcomings of 3D in vitro models Problems in the construction and analysis of 3D in vitro models Although 3D in vitro models have an increasingly wide range of applications in various research fields, there are still some problems in the process of model construction. First, there is no uniform standard for the culture of in vitro models. For example, the culture protocols of the same tumor-derived organoids in different laboratories often differ, such as different culture media components. Some niche factors commonly used in organoid culture are produced by different cell lines and often have batch effects. These unstable factors can have an impact on the experimental results . In addition, extracellular matrices are essential during the construction of 3D in vitro models. Matrigel or other animal-based matrix extract components are most commonly used. These matrix components often have batch-to-batch variation, which can affect the reproducibility of experiments and lead to unstable results. In addition, they may carry unknown pathogens or other immunogenic components that may not only affect the growth of tumor organoids, but may also lead to non-specific activation of immune cells in the presence of immune components in the model, thus affecting the stability of the in vitro simulated -TME . In addition, during tumor organoid culture, normal epithelial cells grow faster than tumor cells and therefore dominate the culture process and inhibit the growth of tumor organoids. To solve this problem, it is necessary to correctly identify the tumor tissues when sampling after surgery and avoid taking the normal tissues as much as possible. During the culture process, tumor organoids can also be selected by adjusting the composition of the culture medium. For example, some colorectal cancers have mutations in the Wnt pathway and can survive in environments without Wnt3a, so colorectal cancer organoids can be picked from normal organoids by removing the Wnt3a factor from the culture medium during the culture process . There are also some tumor organoids that are morphologically different from their corresponding normal tissue organoids, so the normal tissue organoids can be manually removed to ensure the better survival of tumor organoids . Finally, the tumor tissues themselves are very complex in structure and have intratumoral heterogeneity. In the process of in vitro model construction, due to the limitation of culture conditions, the prolongation of culture time and the increase of the number of passages, only a part of tumor cells adapted to the culture conditions can survive, resulting in the loss of intratumoral heterogeneity . During the long time of in vitro culture, new mutations will also be accumulated, which leads to the increase of differences between in vitro models and in vivo tumors. So more optimized culture systems and conditions are needed to solve these problems in the future. Another important difficulty faced when using 3D models for research is the frequent lack of validated evaluation tools. Evaluation methods commonly used in 2D cultures may not be applicable to 3D culture systems due to the unique construction methods and growth characteristics of 3D cell clusters. In various Label-dependent and reader-based assays for cell viability and cytotoxicity, such as MTT assay, LDH activity assay, CellTiter-Glo®3D Cell Viability and fluorescence imaging, the presence of extracellular matrix and the tight intercellular junctions in 3D aggregated cell clusters lead to difficulties in cell lysis and poor penetration of dyes and reagents, resulting in reduced sensitivity of the assay and biased results . In addition, when the organoids are large, there is often cell necrosis in the central region, which produce stronger apoptotic signals and may mislead the experimental results . Therefore, the use of such assays requires improvements in the timing of cell lysis and the diffusion penetration of the dye and reagents, as well as stricter control of the size of organoids. Evaluating changes in the size and morphology of organoids using microscope is also commonly used in assessing the growth state and drug effect. However, organoids also suffer from poor light transmission during imaging, light scattering, and increased background fluorescence intensity due to fluorescence outside the focal plane . In addition, in 2D culture, xy images obtained by microscopy can be used to evaluate the cell growth. But since the 3D structure, the images used to evaluate organoids need to include a series of xy images obtained on the same z-axis, forming a z-stack. Z-stack images also need to be processed by special software, such as z-projection in imagej, to integrate the series of images into a maximal projection image . Different analysis methods need to be applied in combination to overcome these shortcomings. Problems in the reconstruction of tumor immune microenvironment In addition to the problems in the construction of 3D in vitro models, there are also some challenges in using these models to reconstruct the in vitro tumor immune microenvironment. The tumor immune microenvironment is a complex system composed of multiple immune cells and stromal cells, which is highly heterogeneous and dynamic. How to simulate the real state of the in vivo tumor microenvironment as much as possible in vitro needs to be further optimized and refined. As in the organoid/immune cell co-culture model, the type of immune cells added externally is more limited, and the status of the additional immune cells is not consistent with the original cell in the immune microenvironment, and therefore differs significantly from the original microenvironment of the tumor. Moreover, the organoid culture medium cannot maintain the function of immune cells in an optimal condition, so the cellular activity and survival time of immune cells may be affected . In contrast, in ALI or microfluidic culture models a portion of the original tumor infiltrating immune cells are retained in the organoid. These immune cell fractions are maintained for only a limited period of time and are subject to greater loss during passaging or freezing and thawing. Therefore these methods are only suitable for short-term studies . In addition, such systems require high quality of tumor samples. Tumor biopsy samples or puncture specimens do not have sufficient microenvironmental components due to small sample size, therefore it is difficult to use these specimens for in vitro construction of TME. Considering that tumor tissues are spatially heterogeneous, the microenvironment of tumor margins and central sites are often different. Attention should be paid to the representativeness of the samples taken for in vitro culture, and samples should be taken from different regions of the tumor to improve the representation of the samples to tumor tissues . What’s more, the tumor immune microenvironment is in a dynamic state of change, and circulating immune cells have complex interactions with the tumors and tumor infiltrating immune cells. These models only take into account the local immune situation of the TME but not the circulating immune cells. In summary, different in vitro culture models have their unique advantages and corresponding shortcomings, and therefore the most suitable model needs to be selected according to the research objectives. Combining multiple models, integrating their advantages, and further optimizing the culture conditions are ways to refine in vitro 3D culture models for better application in tumor immunology research in the future. Part 4: Future developments of 3D in vitro models The emerging 3D in vitro models have greatly facilitated tumor immunology research. However these models can be further optimized to improve the clinical translation capability and to expand the application. More integrated 3D in vitro models are expected in the future. First is the integration of multiple models. The advantages of multiple models should be utilized to establish a more complete immune system in vitro. For example, in the ALI model, the original immune cells in the TME can be preserved. In contrast, the infiltration of immune cells into tumor organoids in the co-culture model better mimics the process of circulating immune cells trafficking into the tumor and undergoing phenotypic changes. If the two models are combined together, a more comprehensive simulation of the immune response process can be achieved. Second is the functional integration of 3D in vitro models. Simulating the immune microenvironment in vitro does not simply mean putting tumor cells and immune cells in the same spatial space, but more critically, allowing them to interact more functionally. In order to achieve this, it is necessary to restore the in vivo physiological state as much as possible. For example, cells can be exposed to mechanical stress, substrate stiffness and physiologic shear flow. It is also necessary to control the shape and size of the model as well as the ratio of each cell according to the characteristics of different tumor tissues. In these areas, 3D-microfluidic culture systems may have a strong potential. Finally is the integration of other components of the TME. In the TME, there are various components besides immune cells that play important roles, such as tumor-associated fibroblasts, blood vessels and neurons. The microbiota also has close interactions with the TME. Therefore, if multiple microenvironment components can be integrated into the same system in future studies, 3D in vitro models will have more in-depth applications in oncology research.
in vitro models for rebuilding the tumor immune microenvironment Organoid/immune cell co-culture models An organoid initially refers to a 3D structure grown from stem and progenitor cells. It consists of variant organ-specific cell types and can self-organize via cell differentiation and spatially restricted lineage commitment . Organoid can better preserve the genotypic and phenotypic characteristics of the original tumor tissue and retain the heterogeneity of tumor cells. Therefore, it is a more optimal model for tumor research compared to the traditional 2D tumor cell lines . In recent years, organoid technology has been more and more widely used in tumor research. Organoid construction methods can be divided into two types: scaffold-dependent methods and scaffold-free methods. In the scaffold-dependent system, scaffolds provide physical support for cell growth in which cells aggregate and self-assemble to form 3D structures. Depending on the different sources and components, extracellular scaffold can be further divided into matrix derived from decellularized tissue (e.g. Matrigel, de-cellularized ECM from porcine liver tissue), natural polymer-based hydrogel (e.g. methylcellulose, hyaluronic acid, chitosan) and synthetic hydrogel (e.g. PEG, PLGA, PLLA, PVA) . As for scaffold-free methods, hanging drop method, magnetic levitation, rotary cell culture and micromolding technique are also commonly used. In these models, cells spontaneously aggregate to form three-dimensional structures under the action of various external environments such as gravity, magnetic force and rotational agitation. Without the presence of extracellular matrix, these methods allow for precise control of organoid size and morphology and allow for easier co-culture experiments . In the above culture models, tumor stem cells can proliferate and differentiate to form tumor-like structures, but the stromal component of the original tumor tissue cannot be preserved. To apply organoid to tumor immunology research, exogenous immune cells need to be added to interact with tumor organoids. Here we introduce the commonly used methods for co-culturing tumor organoids and immune cells (Fig. A). The easiest way to co-culture organoid with immune cells is by directly adding exogenous immune cells into the culture medium (Fig. A). In this way, immune cells secret cytokine to act on tumor cells, and when immune cells have a high infiltration capacity, they may also enter the matrigel and contact directly with tumor cells. For example, CD3 + T cells isolated from peripheral blood mononuclear cells (PBMCs) were added to the medium of pancreatic cancer organoids. After co-cultured for 72 h, it was found that T cells were viable and Ki67 positive in organoid growth medium but the cell activity was weaker than in RPMI medium. When tumor organoids were present in the matrigel, T cells migrated and infiltrated into matrigel in response to pancreatic cancer cells. Otherwise, they mostly gathered at the edge to form clear boundaries . T cell activation and tumor cell apoptosis could also be observed when different kind of tumor organoids were co-cultured with homologous PBMCs and the addition of PD-1 monoclonal antibody or other immune activators could enhance T cell cytotoxicity, providing a new platform for personalized screening of potential antitumor immunotherapeutic agents . To further increase the contact between tumor organoids and immune cells Lei Yu et al. diluted matrigel to 50% concentration before using it for bladder cancer organoids culture and added specific CAR-T cells to the culture medium. In this case CAR-T cells could better invade the matrigel, thus the killing ability and specificity of CAR-T cells could be better detected. However, the above methods are more suitable for studying the indirect interaction between immune cells and tumor cells. The direct contact is still not sufficient, even if the immune cells have a strong migration and infiltration ability and the matrigel is diluted. Moreover, the components in matrigel may affect the function of immune cells, leading to non-specific activation. Therefore, when considering direct intercellular interactions, mature organoids can also be isolated from the matrigel and cultured in suspension with immune cells (Fig. B). Based on this method, the researchers co-cultured patient-derived MSI-H CRC organoids and NSCLC organoids with homologous PBMCs. The tumoriods were first separated from matrigel and digested into single cell suspensions, which were treated with IFN? to increase the expression of MHC-I molecules on the surface of tumor cells and thus promote antigen presentation. PBMCs isolated from peripheral blood were simultaneously incubated with anti-CD28 and anti-PD-1 antibodies to provide a co-stimulatory signal and counteract the upregulation of PD-L1 expression caused by IFN? treatment. The treated tumoroid single cells were mixed with PBMCs at a certain ratio and co-cultured in T-cell medium containing IL-2 for 7 days. After 2 rounds of co-culture, the T cells were activated and can specifically kill the corresponding tumoroids. In this way, tumor-reactive T cells could be enriched in peripheral blood and their tumor-killing efficiency could be measured at the individual level, providing a good model for cell therapy optimization . Qingda Meng et al. applied a similar method to isolate human pancreatic cancer organoids and mixed them with PBMCs in T-cell medium with IL-2, IL-15, IL-21 and other factors required for immune cell growth. T cells were found to kill tumor cells in a granzyme B or Fas-FasL-dependent manner during co-culture, and the co-cultured T cells exhibited markers of tissue-resident memory phenotype. Each patient’s co-cultured T cells exhibited their own unique expression of immune checkpoint proteins. But this method is only appropriate for studies of short duration, and as the separation time of the organoids from the matrigel increases, adherence or death of the organoids may occur. In order to prevent this situation, the co-culture protocol can be further optimized. Mei Song et al. pre-coated 24-well plates with matrigel and then mixed F4/80 + CD11b + tumor associated macrophages (TAMs) from ovarian cancer-bearing mice with ID8 tumor cells at a ratio of 1:10 in medium containing 2% matrigel. The mixture was planted in the pre-coated plates to investigate the role and mechanism of TAMs in the growth and metastasis of ovarian cancer. This method allowed direct contact between tumor cells and immune cells, and also prevented the adherence of organoids, making the co-culture system more stable and suitable for long-term studies (Fig. B lower part). A similar approach has also been applied to the co-culture of gastric tumoriods with CD8 + T cells and ER + breast tumor organoids with breast fibroblasts . Besides, as another widely used 3D in vitro culture model, tumor spheroids are typically created in a scaffold-free environment by placing cells into suspension colonies without the aid of extracellular matrices (ECMs) or other physical supports . Due to the lack of physical support from the ECMs, tumor spheroids are less complex and cannot completely rebuild the structure in vivo, but they are more convenient for co-culture with immune cells. Tristan Courau et al. cultured tumor cells into tumor spheroids and added T cells and NK cells isolated from peripheral blood to the medium, where immune cells could make direct contact with the tumor spheroids and infiltrate to produce killing effect. Further examination of molecule expression patterns of infiltrated immune cells and tumor cells revealed that immunomodulatory antibodies targeting MICA/B and NKG2A had anti-tumor potential. Thus the tumor spheroid and immune cell co-culture model is a usable tool to study tumor-lymphocyte interactions. In addition, Monocyte-derived macrophages were co-cultured with patient-derived colorectal cancer tumor spheroids in ultra-low attachment plates, and it was found that the co-cultured tumor spheroids were more prone to “budding” (Fig. B upper part). Immunohistochemical staining showed obvious macrophage aggregation at the budding site and the budding was accompanied by a reduction of intercellular tight junctions. Therefore, macrophages may play a role in the invasion and metastasis of colon cancer . In addition to the two co-culture methods mentioned above, it is also possible to co-culture tumor cells and immune cells by mixing them directly in matrix (Fig. C). Following this idea, Jayati Chakrabarti et al. co-cultured mouse gastric tumoriods expressing the activated GLI2 allele GLI2A with DC cells and cytotoxic T lymphocytes (CTLs) in matrigel, and added PD-L1 inhibitors to the co-culture system. They found that CTLs had a significant killing effect on tumor cells. Vivien Koh et al. found that immunotherapy failed to activate CTLs to kill tumor cells in the presence of myeloid-derived suppressor cells (MDSCs). Xi He et al. used the same method to investigate the effect of myeloid-derived cells (MDCs) on the growth of the mouse small intestine adenoma. After co-culturing MDCs with organoids in equal propotions in matrigel for 8 days, it was found that most of the MDCs were transformed into TAMs and stimulated the growth of adenoma organoids. Together, the above studies demonstrate the usability of such co-culture models in reconstructing tumor immune microenvironment in vitro and in studying the intercellular interactions. ALI culture model The ALI culture is a special organoid culture model. In this model, tumor tissues are mechanically minced into pieces of about 1 mm in diameter without enzymatic digestion, mixed with collagen gel and layered on top of pre-solidified collagen gel within a permeable, membranous inner transwell insert. The insert is placed in a cell culture dish, and culture medium is added between the dish and the insert to form the double dish culture system. This way, the upper part of mixture of tumor tissues and collagen is directly exposed to air, and the culture medium infiltrates the bottom of the collagen through the micropores to create an air-liquid interaction model (Fig. B). The ALI culture method can increase the oxygen supply in system, and for respiratory, gastrointestinal or body surface tumors, the air-liquid interaction also better mimics the tumor growth environment in vivo. In this system, tumor cells can grow normally to form tumoroids, which retain the pathological features and genetic alterations of the original tumor. Also, the immune cells and fibroblast stroma of the tumor tissue can be maintained for a certain period of time. Using this approach, James T. Neal et al. established organoids of various origins, such as colorectal, kidney, lung, pancreatic, and thyroid cancers, and found that a variety of immune cells including T cells, B cells, NK cells, and macrophages from the tumor tissues were preserved in these organoids, and the types and proportions of immune cells were highly consistent with the original tumor. Although the amount of immune cells gradually decreased with the extension of culture time, T cells could be retained for more than 30 days and even after organoid passaging with the addition of IL-2 in the culture medium. TCR sequencing revealed a high similarity between the retained T cells in organoids and the T cells contained in original tumors. After the addition of nivolumab to this model, T cell activation marker IFNG, GZMB and PRF1 were significantly increased, and the tumor cells also underwent significant apoptosis, indicating that this model could successfully mimic the response process of tumor immunotherapy and was expected to be applied to evaluate the efficacy and sensitization methods of ICB therapy. Similarly, Laura K. Esser et al. cultured 42 surgically resected renal clear cell carcinoma specimens using the ALI method. Immunohistochemical staining and RNA sequencing verified that the ALI cultured patient derived organoids (PDOs) were highly similar to the characteristics of the tumor specimens and that the immune and stromal cells in PDOs could be preserved to some extent. The authors treated 10 cases of ALI PDO with the targeted cancer drug cabozantinib or the nivolumab and found that PDOs of different patient origins showed very different responses to the two treatments, showing that this model could reflect the different treatment effects of patients. In this model, the survival time of primary immune cells in tumor tissues is the key issue due to the lack of exogenous immune cells. In another study, patient-derived CRC organoids were cultured using ALI method and the survival time of immune cells were evaluated. The results showed that a certain percentage of CD45 + hematopoietic cells were still present after 8 days of culture but the number of CD3 + T cells decreased significantly, suggesting that the culture environment with organoid medium might not be the most suitable for T cell survival and further addition of T cell-promoting growth factors was needed to adjust the culture conditions . In addition to tumor organoids, Lisa E. Wagar et al. also used the ALI method to culture lymphoid organs including lymph nodes, tonsils, and spleens. They minced and digested these lymphoid organs into single cells, which were plated into pre-coated transwell chambers along with relevant antigens. Complete culture medium was added in the lower chamber supplemented with a small amount of recombinant human B cell-activating factor (BAFF) to improve total B cell survival. After several days in culture, clustered cells formed the reaggregated regions. They accessed the organoids and found that they could recapitulate key germinal center features of lymphoid organs in vitro, including the affinity maturation, somatic hyper-mutation and the production of antigen-specific antibodies. Using this system, the authors studied the immune responses against pathogens, such as the identification of key cellular components in the generation of anti-influenza virus immune responses. For further study, lymphoid organoids co-cultured with tumor cells also holds promise for applications related to the study of adaptive immune responses against tumors, the identification and assessment of tumor vaccines, and the investigation of lymph node metastasis of tumors. 3D-Microfluidic based culture method Microfluidics refers to a technique for controlling the flow of micro fluids (10 - 9 to 10 - 18 L) in microscopic channels (10 - 4 to 10 - 5 m in size) . In this system, the size of the microchannels and the volume ratio of cell to extracellular fluid are very similar to those in the TME. And because of the low Reynolds coefficient, the fluid in this system flows in a laminar way, allowing the soluble factors to form a concentration gradient similar to that in vivo. So microfluidic culture system acts as a good model for simulating the TME . Combining microfluidic devices and organoid technology not only improves the homogeneity and controllability of organoids through the design of microfluidic channels and the control of liquid flow rate, but also makes the high-throughput production of organoids possible (Fig. C). Microfluidic devices have already had a wide range of applications in modeling the TME in vitro. First, because multiple different cell types can be manipulated independently in microchannels, co-culture of tumor organoids with additionally added cells is possible. Muhammad R. Haque et al. developed a tumor microarray device to mimic the TME of PDAC by merging PDOs and stromal cells, including pancreatic stellate cells (PSCs) and macrophages. In this multicellular microfluidic system, they successfully established a complex organotypic tumor environment containing connective tissue proliferating stroma and immune cells. Drugs targeting PSCs or macrophages in this model significantly increased the ability of chemotherapeutic agents to kill tumor cells, whereas this response was absent in the tumor cell culture system without stromal cells. This indicates that this system partially mimics the tumor microenvironment and can be used for drug screening targeting the TME. On the other hand, the original immune cells and stromal components are also better preserved in this system due to the higher similarity of the microfluidic device to the in vivo states. Using the microfluidic system to create patient- and mouse-derived organoids, it was found that immune cells native to the patient or mouse tumor tissues could be retained and could respond to PD-1 monoclonal antibody treatment in vitro. Organoids from patients whose tumors were sensitive or resistant to anti PD-1 therapy maintained the same drug responsiveness in vitro. In addition, investigators have used this system to screen for small molecule drugs to sensitize PD-1 monoclonal therapy and found that both CDK4/6 and TBK1/IKKe inhibitors in combination with PD-1 monoclonal antibodies enhanced antitumor immune response. The drug efficacy was further confirmed in in vivo models . Recently, Shengli Ding et al. developed a novel droplet-based microfluidic 3D culture platform to generate a large number of micro-organospheres (MOSs) from a small amount of tissues of cancer patients. The key step is to prepare the single cell suspension of primary tissues from tumor patients, add it to 3D matrigel, and then mix with oil phase liquid to generate MOSs, which can be cultured in suspension after demulsification to remove excess oil. First, the authors evaluated the consistency of MOSs and immune cells in the corresponding tumor samples using single cell transcriptomics and found that MOSs retained tumor-associated fibroblasts as well as myeloid and lymphoid immune cells similar to the original tissues. The expression of immunosuppression-related markers also had high consistency. In addition, this model can be used for co-culture with exogenous immune cells. Due to the smaller size and larger surface area-to-volume ratio of MOSs, both additional TIL and PBMC can more easily infiltrate into MOSs to contact and effectively kill tumor cells, providing a useful tool to assess the efficacy of ACT. 3D-bioprinting model 3D-bioprinting as an emerging technology, is an in vitro 3D structural model manufactured with biological units (cells/ proteins/ DNA etc.) and biological materials according to the requirements of bionic morphology and organism function using 3D printing techniques . Conventional organoids are formed by proliferation, differentiation and self-assembly of stem cells, so they lack control over cell number, cell type ratio and microenvironment. The 3D-bioprinting technology can reconstruct the complex structure of organoids through accurate and stable model construction and multi-cell controlled organoid printing, which can simultaneously print multiple cell components, ECMs and cell growth factors, thus effectively improve the reconstruction of microenvironment in organoids (Fig. D). In 2020 Kunyoo Shin’s team introduced a new concept of a mini-organ called an assembloid based on 3D-bioprinting technology to mimic human tumor tissues in terms of structure and function. The team constructed patient-specific bladder assembloids by 3D-bioprinting. This model not only maintained the genetic changes of the parental tumor but also introduced TME components, revealing that signaling between tumor cells and stromal cells played a key role in controlling tumor plasticity . Marcel Alexander Heinrich et al. used 3D-bioprinting to construct mini-brain by mixing glioma cells (GL261) with macrophages (RAW264.7). On the basis of this model, they investigated the interaction between macrophages and glioma cells and found that glioma cells could recruit macrophages and cause them to generate a glioma-associated macrophage phenotype. Macrophages also promoted the proliferation and invasion of glioma cells. Similarly, Hermida et al. constructed an in vitro brain glioblastoma model using extrusion-based bioprinting. The model was based on alginate modified with RGDS cell adhesion peptide, hyaluronic acid and type I collagen, and integrated multiple cell types including tumor cells, microglias and tumor stromal cells. Compared with tumor cells alone, the 3D-printed tumor model was more resistant to chemotherapeutic drugs, reflecting the role of TME on chemotherapy. The combination of different models The above 3D in vitro models have been widely used in simulating TME and tumor immune-related studies. They all have their own advantages and disadvantages and are suitable for different application scenarios. The following table summarizes and compares the characteristics of different models (Table ). However, the ability to reconstruct the TME in vitro using a single model is limited due to the defects of each model, and therefore the combination of multiple models represents more promising applications. Zhiyi Gong et al. designed a platform applying acoustic droplet printing to fabricate mouse bladder cancer organoids. The organoid produced by this platform could retain the original immune cells in the tumor tissues within 2 weeks. In addition, by placing the manufactured organoids in a microfluidic chip, the size and morphology of the organoids could be further controlled. High-throughput organoid manufacturing, drug screening, and real-time imaging and evaluation of organoids could also be achieved. Subsequently, the authors co-cultured the organoids with homologous spleen-derived immune cells for 2 days, and the infiltration of lymphocytes into the organoids was observed. This can be used to better screen for tumor-responsive T cells and the cells expanded in vitro also have the ability to kill the tumor organoids. Similarly, Konstantinos I. Votanopoulos et al. first mixed patient’s melanoma cells with immune cells derived from the same patient’s lymph nodes in matrigel to form immune-enhanced tumor organoids (iPTOs). Then they applied the 3D-microfluidic system to circulate peripheral blood T cells of the same patient origin around the iPTOs and found that the co-cultured T cells also had the ability to kill tumor cells, reflecting the possible role of iPTOs in the induction of adaptive immunity. The above two studies combined multiple in vitro models to integrate immune cells within the TME and the different species of exogenous immune cells in one system, thus more comprehensively mimic the process of antitumor immunity and have more diverse applications in tumor immunology research.
An organoid initially refers to a 3D structure grown from stem and progenitor cells. It consists of variant organ-specific cell types and can self-organize via cell differentiation and spatially restricted lineage commitment . Organoid can better preserve the genotypic and phenotypic characteristics of the original tumor tissue and retain the heterogeneity of tumor cells. Therefore, it is a more optimal model for tumor research compared to the traditional 2D tumor cell lines . In recent years, organoid technology has been more and more widely used in tumor research. Organoid construction methods can be divided into two types: scaffold-dependent methods and scaffold-free methods. In the scaffold-dependent system, scaffolds provide physical support for cell growth in which cells aggregate and self-assemble to form 3D structures. Depending on the different sources and components, extracellular scaffold can be further divided into matrix derived from decellularized tissue (e.g. Matrigel, de-cellularized ECM from porcine liver tissue), natural polymer-based hydrogel (e.g. methylcellulose, hyaluronic acid, chitosan) and synthetic hydrogel (e.g. PEG, PLGA, PLLA, PVA) . As for scaffold-free methods, hanging drop method, magnetic levitation, rotary cell culture and micromolding technique are also commonly used. In these models, cells spontaneously aggregate to form three-dimensional structures under the action of various external environments such as gravity, magnetic force and rotational agitation. Without the presence of extracellular matrix, these methods allow for precise control of organoid size and morphology and allow for easier co-culture experiments . In the above culture models, tumor stem cells can proliferate and differentiate to form tumor-like structures, but the stromal component of the original tumor tissue cannot be preserved. To apply organoid to tumor immunology research, exogenous immune cells need to be added to interact with tumor organoids. Here we introduce the commonly used methods for co-culturing tumor organoids and immune cells (Fig. A). The easiest way to co-culture organoid with immune cells is by directly adding exogenous immune cells into the culture medium (Fig. A). In this way, immune cells secret cytokine to act on tumor cells, and when immune cells have a high infiltration capacity, they may also enter the matrigel and contact directly with tumor cells. For example, CD3 + T cells isolated from peripheral blood mononuclear cells (PBMCs) were added to the medium of pancreatic cancer organoids. After co-cultured for 72 h, it was found that T cells were viable and Ki67 positive in organoid growth medium but the cell activity was weaker than in RPMI medium. When tumor organoids were present in the matrigel, T cells migrated and infiltrated into matrigel in response to pancreatic cancer cells. Otherwise, they mostly gathered at the edge to form clear boundaries . T cell activation and tumor cell apoptosis could also be observed when different kind of tumor organoids were co-cultured with homologous PBMCs and the addition of PD-1 monoclonal antibody or other immune activators could enhance T cell cytotoxicity, providing a new platform for personalized screening of potential antitumor immunotherapeutic agents . To further increase the contact between tumor organoids and immune cells Lei Yu et al. diluted matrigel to 50% concentration before using it for bladder cancer organoids culture and added specific CAR-T cells to the culture medium. In this case CAR-T cells could better invade the matrigel, thus the killing ability and specificity of CAR-T cells could be better detected. However, the above methods are more suitable for studying the indirect interaction between immune cells and tumor cells. The direct contact is still not sufficient, even if the immune cells have a strong migration and infiltration ability and the matrigel is diluted. Moreover, the components in matrigel may affect the function of immune cells, leading to non-specific activation. Therefore, when considering direct intercellular interactions, mature organoids can also be isolated from the matrigel and cultured in suspension with immune cells (Fig. B). Based on this method, the researchers co-cultured patient-derived MSI-H CRC organoids and NSCLC organoids with homologous PBMCs. The tumoriods were first separated from matrigel and digested into single cell suspensions, which were treated with IFN? to increase the expression of MHC-I molecules on the surface of tumor cells and thus promote antigen presentation. PBMCs isolated from peripheral blood were simultaneously incubated with anti-CD28 and anti-PD-1 antibodies to provide a co-stimulatory signal and counteract the upregulation of PD-L1 expression caused by IFN? treatment. The treated tumoroid single cells were mixed with PBMCs at a certain ratio and co-cultured in T-cell medium containing IL-2 for 7 days. After 2 rounds of co-culture, the T cells were activated and can specifically kill the corresponding tumoroids. In this way, tumor-reactive T cells could be enriched in peripheral blood and their tumor-killing efficiency could be measured at the individual level, providing a good model for cell therapy optimization . Qingda Meng et al. applied a similar method to isolate human pancreatic cancer organoids and mixed them with PBMCs in T-cell medium with IL-2, IL-15, IL-21 and other factors required for immune cell growth. T cells were found to kill tumor cells in a granzyme B or Fas-FasL-dependent manner during co-culture, and the co-cultured T cells exhibited markers of tissue-resident memory phenotype. Each patient’s co-cultured T cells exhibited their own unique expression of immune checkpoint proteins. But this method is only appropriate for studies of short duration, and as the separation time of the organoids from the matrigel increases, adherence or death of the organoids may occur. In order to prevent this situation, the co-culture protocol can be further optimized. Mei Song et al. pre-coated 24-well plates with matrigel and then mixed F4/80 + CD11b + tumor associated macrophages (TAMs) from ovarian cancer-bearing mice with ID8 tumor cells at a ratio of 1:10 in medium containing 2% matrigel. The mixture was planted in the pre-coated plates to investigate the role and mechanism of TAMs in the growth and metastasis of ovarian cancer. This method allowed direct contact between tumor cells and immune cells, and also prevented the adherence of organoids, making the co-culture system more stable and suitable for long-term studies (Fig. B lower part). A similar approach has also been applied to the co-culture of gastric tumoriods with CD8 + T cells and ER + breast tumor organoids with breast fibroblasts . Besides, as another widely used 3D in vitro culture model, tumor spheroids are typically created in a scaffold-free environment by placing cells into suspension colonies without the aid of extracellular matrices (ECMs) or other physical supports . Due to the lack of physical support from the ECMs, tumor spheroids are less complex and cannot completely rebuild the structure in vivo, but they are more convenient for co-culture with immune cells. Tristan Courau et al. cultured tumor cells into tumor spheroids and added T cells and NK cells isolated from peripheral blood to the medium, where immune cells could make direct contact with the tumor spheroids and infiltrate to produce killing effect. Further examination of molecule expression patterns of infiltrated immune cells and tumor cells revealed that immunomodulatory antibodies targeting MICA/B and NKG2A had anti-tumor potential. Thus the tumor spheroid and immune cell co-culture model is a usable tool to study tumor-lymphocyte interactions. In addition, Monocyte-derived macrophages were co-cultured with patient-derived colorectal cancer tumor spheroids in ultra-low attachment plates, and it was found that the co-cultured tumor spheroids were more prone to “budding” (Fig. B upper part). Immunohistochemical staining showed obvious macrophage aggregation at the budding site and the budding was accompanied by a reduction of intercellular tight junctions. Therefore, macrophages may play a role in the invasion and metastasis of colon cancer . In addition to the two co-culture methods mentioned above, it is also possible to co-culture tumor cells and immune cells by mixing them directly in matrix (Fig. C). Following this idea, Jayati Chakrabarti et al. co-cultured mouse gastric tumoriods expressing the activated GLI2 allele GLI2A with DC cells and cytotoxic T lymphocytes (CTLs) in matrigel, and added PD-L1 inhibitors to the co-culture system. They found that CTLs had a significant killing effect on tumor cells. Vivien Koh et al. found that immunotherapy failed to activate CTLs to kill tumor cells in the presence of myeloid-derived suppressor cells (MDSCs). Xi He et al. used the same method to investigate the effect of myeloid-derived cells (MDCs) on the growth of the mouse small intestine adenoma. After co-culturing MDCs with organoids in equal propotions in matrigel for 8 days, it was found that most of the MDCs were transformed into TAMs and stimulated the growth of adenoma organoids. Together, the above studies demonstrate the usability of such co-culture models in reconstructing tumor immune microenvironment in vitro and in studying the intercellular interactions.
The ALI culture is a special organoid culture model. In this model, tumor tissues are mechanically minced into pieces of about 1 mm in diameter without enzymatic digestion, mixed with collagen gel and layered on top of pre-solidified collagen gel within a permeable, membranous inner transwell insert. The insert is placed in a cell culture dish, and culture medium is added between the dish and the insert to form the double dish culture system. This way, the upper part of mixture of tumor tissues and collagen is directly exposed to air, and the culture medium infiltrates the bottom of the collagen through the micropores to create an air-liquid interaction model (Fig. B). The ALI culture method can increase the oxygen supply in system, and for respiratory, gastrointestinal or body surface tumors, the air-liquid interaction also better mimics the tumor growth environment in vivo. In this system, tumor cells can grow normally to form tumoroids, which retain the pathological features and genetic alterations of the original tumor. Also, the immune cells and fibroblast stroma of the tumor tissue can be maintained for a certain period of time. Using this approach, James T. Neal et al. established organoids of various origins, such as colorectal, kidney, lung, pancreatic, and thyroid cancers, and found that a variety of immune cells including T cells, B cells, NK cells, and macrophages from the tumor tissues were preserved in these organoids, and the types and proportions of immune cells were highly consistent with the original tumor. Although the amount of immune cells gradually decreased with the extension of culture time, T cells could be retained for more than 30 days and even after organoid passaging with the addition of IL-2 in the culture medium. TCR sequencing revealed a high similarity between the retained T cells in organoids and the T cells contained in original tumors. After the addition of nivolumab to this model, T cell activation marker IFNG, GZMB and PRF1 were significantly increased, and the tumor cells also underwent significant apoptosis, indicating that this model could successfully mimic the response process of tumor immunotherapy and was expected to be applied to evaluate the efficacy and sensitization methods of ICB therapy. Similarly, Laura K. Esser et al. cultured 42 surgically resected renal clear cell carcinoma specimens using the ALI method. Immunohistochemical staining and RNA sequencing verified that the ALI cultured patient derived organoids (PDOs) were highly similar to the characteristics of the tumor specimens and that the immune and stromal cells in PDOs could be preserved to some extent. The authors treated 10 cases of ALI PDO with the targeted cancer drug cabozantinib or the nivolumab and found that PDOs of different patient origins showed very different responses to the two treatments, showing that this model could reflect the different treatment effects of patients. In this model, the survival time of primary immune cells in tumor tissues is the key issue due to the lack of exogenous immune cells. In another study, patient-derived CRC organoids were cultured using ALI method and the survival time of immune cells were evaluated. The results showed that a certain percentage of CD45 + hematopoietic cells were still present after 8 days of culture but the number of CD3 + T cells decreased significantly, suggesting that the culture environment with organoid medium might not be the most suitable for T cell survival and further addition of T cell-promoting growth factors was needed to adjust the culture conditions . In addition to tumor organoids, Lisa E. Wagar et al. also used the ALI method to culture lymphoid organs including lymph nodes, tonsils, and spleens. They minced and digested these lymphoid organs into single cells, which were plated into pre-coated transwell chambers along with relevant antigens. Complete culture medium was added in the lower chamber supplemented with a small amount of recombinant human B cell-activating factor (BAFF) to improve total B cell survival. After several days in culture, clustered cells formed the reaggregated regions. They accessed the organoids and found that they could recapitulate key germinal center features of lymphoid organs in vitro, including the affinity maturation, somatic hyper-mutation and the production of antigen-specific antibodies. Using this system, the authors studied the immune responses against pathogens, such as the identification of key cellular components in the generation of anti-influenza virus immune responses. For further study, lymphoid organoids co-cultured with tumor cells also holds promise for applications related to the study of adaptive immune responses against tumors, the identification and assessment of tumor vaccines, and the investigation of lymph node metastasis of tumors.
Microfluidics refers to a technique for controlling the flow of micro fluids (10 - 9 to 10 - 18 L) in microscopic channels (10 - 4 to 10 - 5 m in size) . In this system, the size of the microchannels and the volume ratio of cell to extracellular fluid are very similar to those in the TME. And because of the low Reynolds coefficient, the fluid in this system flows in a laminar way, allowing the soluble factors to form a concentration gradient similar to that in vivo. So microfluidic culture system acts as a good model for simulating the TME . Combining microfluidic devices and organoid technology not only improves the homogeneity and controllability of organoids through the design of microfluidic channels and the control of liquid flow rate, but also makes the high-throughput production of organoids possible (Fig. C). Microfluidic devices have already had a wide range of applications in modeling the TME in vitro. First, because multiple different cell types can be manipulated independently in microchannels, co-culture of tumor organoids with additionally added cells is possible. Muhammad R. Haque et al. developed a tumor microarray device to mimic the TME of PDAC by merging PDOs and stromal cells, including pancreatic stellate cells (PSCs) and macrophages. In this multicellular microfluidic system, they successfully established a complex organotypic tumor environment containing connective tissue proliferating stroma and immune cells. Drugs targeting PSCs or macrophages in this model significantly increased the ability of chemotherapeutic agents to kill tumor cells, whereas this response was absent in the tumor cell culture system without stromal cells. This indicates that this system partially mimics the tumor microenvironment and can be used for drug screening targeting the TME. On the other hand, the original immune cells and stromal components are also better preserved in this system due to the higher similarity of the microfluidic device to the in vivo states. Using the microfluidic system to create patient- and mouse-derived organoids, it was found that immune cells native to the patient or mouse tumor tissues could be retained and could respond to PD-1 monoclonal antibody treatment in vitro. Organoids from patients whose tumors were sensitive or resistant to anti PD-1 therapy maintained the same drug responsiveness in vitro. In addition, investigators have used this system to screen for small molecule drugs to sensitize PD-1 monoclonal therapy and found that both CDK4/6 and TBK1/IKKe inhibitors in combination with PD-1 monoclonal antibodies enhanced antitumor immune response. The drug efficacy was further confirmed in in vivo models . Recently, Shengli Ding et al. developed a novel droplet-based microfluidic 3D culture platform to generate a large number of micro-organospheres (MOSs) from a small amount of tissues of cancer patients. The key step is to prepare the single cell suspension of primary tissues from tumor patients, add it to 3D matrigel, and then mix with oil phase liquid to generate MOSs, which can be cultured in suspension after demulsification to remove excess oil. First, the authors evaluated the consistency of MOSs and immune cells in the corresponding tumor samples using single cell transcriptomics and found that MOSs retained tumor-associated fibroblasts as well as myeloid and lymphoid immune cells similar to the original tissues. The expression of immunosuppression-related markers also had high consistency. In addition, this model can be used for co-culture with exogenous immune cells. Due to the smaller size and larger surface area-to-volume ratio of MOSs, both additional TIL and PBMC can more easily infiltrate into MOSs to contact and effectively kill tumor cells, providing a useful tool to assess the efficacy of ACT.
3D-bioprinting as an emerging technology, is an in vitro 3D structural model manufactured with biological units (cells/ proteins/ DNA etc.) and biological materials according to the requirements of bionic morphology and organism function using 3D printing techniques . Conventional organoids are formed by proliferation, differentiation and self-assembly of stem cells, so they lack control over cell number, cell type ratio and microenvironment. The 3D-bioprinting technology can reconstruct the complex structure of organoids through accurate and stable model construction and multi-cell controlled organoid printing, which can simultaneously print multiple cell components, ECMs and cell growth factors, thus effectively improve the reconstruction of microenvironment in organoids (Fig. D). In 2020 Kunyoo Shin’s team introduced a new concept of a mini-organ called an assembloid based on 3D-bioprinting technology to mimic human tumor tissues in terms of structure and function. The team constructed patient-specific bladder assembloids by 3D-bioprinting. This model not only maintained the genetic changes of the parental tumor but also introduced TME components, revealing that signaling between tumor cells and stromal cells played a key role in controlling tumor plasticity . Marcel Alexander Heinrich et al. used 3D-bioprinting to construct mini-brain by mixing glioma cells (GL261) with macrophages (RAW264.7). On the basis of this model, they investigated the interaction between macrophages and glioma cells and found that glioma cells could recruit macrophages and cause them to generate a glioma-associated macrophage phenotype. Macrophages also promoted the proliferation and invasion of glioma cells. Similarly, Hermida et al. constructed an in vitro brain glioblastoma model using extrusion-based bioprinting. The model was based on alginate modified with RGDS cell adhesion peptide, hyaluronic acid and type I collagen, and integrated multiple cell types including tumor cells, microglias and tumor stromal cells. Compared with tumor cells alone, the 3D-printed tumor model was more resistant to chemotherapeutic drugs, reflecting the role of TME on chemotherapy.
The above 3D in vitro models have been widely used in simulating TME and tumor immune-related studies. They all have their own advantages and disadvantages and are suitable for different application scenarios. The following table summarizes and compares the characteristics of different models (Table ). However, the ability to reconstruct the TME in vitro using a single model is limited due to the defects of each model, and therefore the combination of multiple models represents more promising applications. Zhiyi Gong et al. designed a platform applying acoustic droplet printing to fabricate mouse bladder cancer organoids. The organoid produced by this platform could retain the original immune cells in the tumor tissues within 2 weeks. In addition, by placing the manufactured organoids in a microfluidic chip, the size and morphology of the organoids could be further controlled. High-throughput organoid manufacturing, drug screening, and real-time imaging and evaluation of organoids could also be achieved. Subsequently, the authors co-cultured the organoids with homologous spleen-derived immune cells for 2 days, and the infiltration of lymphocytes into the organoids was observed. This can be used to better screen for tumor-responsive T cells and the cells expanded in vitro also have the ability to kill the tumor organoids. Similarly, Konstantinos I. Votanopoulos et al. first mixed patient’s melanoma cells with immune cells derived from the same patient’s lymph nodes in matrigel to form immune-enhanced tumor organoids (iPTOs). Then they applied the 3D-microfluidic system to circulate peripheral blood T cells of the same patient origin around the iPTOs and found that the co-cultured T cells also had the ability to kill tumor cells, reflecting the possible role of iPTOs in the induction of adaptive immunity. The above two studies combined multiple in vitro models to integrate immune cells within the TME and the different species of exogenous immune cells in one system, thus more comprehensively mimic the process of antitumor immunity and have more diverse applications in tumor immunology research.
in vitro models in tumor immunology research Application of 3D in vitro models in mechanism research of the TME Tumor cells and their microenvironment are a functional entity. The microenvironment interacts and co-evolves with tumor cells and play important roles in multiple processes of tumor development. Therefore, studying the interaction between tumor and microenvironment can help us understand the biological behavior of tumor and lay the theoretical foundation for finding new therapeutic targets and exploring new methods of tumor immunotherapy. 3D in vitro models are good models for studying the TME because of their simplicity, convenience, flexibility and high similarity to original tumor tissues (Fig. E). For example, in tumor microenvironment, T cells and NK cells are usually the main ones that can directly kill tumor cells. Enhancing their killing ability contributes to tumor control. Using a co-culture model of CRC organoids and T cells, researchers screened a series of small molecule inhibitors and found that DKK1 inhibitors could significantly enhance the killing effect of T cells and promote apoptosis of tumor cells by regulating the GSK3ß-E2F1-T-bet axis in CD8 + T cells. DKK1 inhibitors combined with PD-1 monoclonal antibodies could achieve better tumor control . Similarly, BRD1 inhibitors were able to convert NK cells and naïve CD8 + T cells to a more activated and cytotoxic phenotype, helping anti-PD-1/PD-L1 bispecific antibodies to exert a more effective antitumor immune effect in HGSC . By detecting the expression of surface biomarkers on tumor cells and immune cells before and after co-culture, antibodies targeting MICA/B and NKG2A were also found to enhance the killing effect of T cells and NK cells on colorectal cancer organoids, thus providing a potential target for the treatment of CRC . In addition to immune cells that play a major role in tumor killing, there are multiple immunosuppressive components in the TME that can promote tumor progression. Co-culture of mouse MDCs with small intestinal adenoma organoids revealed that MDCs were transformed into TAMs and stimulated adenoma growth via the COX-2-PEG2-EP4 pathway . Using mouse ovarian cancer tumor spheroids co-cultured with TAMs, it was found that when UBR5 was knocked down, the tumor spheroids had slower growth and smaller size. The ability to recruit TAMs was also reduced, as was the expression of cytokines and chemokines associated with TAMs recruitment. This demonstrated that targeting UBR5 could help control the growth of ovarian cancer by modulating the tumor immune microenvironment . Using the similar co-culture method, researchers found that breast fibroblasts secreted cytokines such as IL-1ß, which acted on breast cancer cells through a paracrine pathway to promote their proliferation. Blocking this paracrine pathway enhanced the therapeutic effect of tomaxifen on breast cancer, suggesting that fibroblasts in the remaining breast tissue after breast-conserving surgery may increase the risk of breast cancer recurrence . These studies have explored the interaction of tumor cells with different components of the microenvironment through co-culture models, providing a direction for the development of immunotherapy. Application of 3D in vitro models in assessing the efficacy of ICI treatment for personalized treatment With the increasing research on the immune microenvironment, immunotherapy has been more and more widely used in the treatment of tumors, among which the most frequently applied is immune checkpoint inhibitor therapy. ICI therapy targeting PD1/PD-L1 and CTLA-4 has been clinically employed in progressive melanoma , squamous cell skin cancer , non-small cell lung cancer , renal cell carcinoma , head and neck tumors, and tumors with mismatch repair defects of various tissue types. However, only some of the patients treated with ICI have benefited from this therapy. Considering the adverse effects and financial burden, it is necessary to carefully select patients who may benefit from ICI treatment and personalize the treatment with precision. For precise patient selection, a number of biomarkers have been demonstrated to predict the efficacy of ICI therapy, such as MSI status , TMB , neoantigen expression levels , CD8 + T cell counts , and tumor cell surface PD-L1 expression levels , but none of these indicators have sufficiently high positive and negative predictive values to allow effective screening of patients . Other studies have analyzed immunobiological indicators of the tumor, such as genomic, transcriptomic and proteomic sequencing analysis, or have integrated multiple biomarkers, but even these more complicated analyze methods do not fully reflect the intratumoral and individual heterogeneity of human tumors and cannot accurately predict ICI treatment efficacy . Therefore, static predictors cannot meet clinical needs. It is important to establish an efficacy assessment model that can be monitored dynamically, and the in vitro 3D culture may fill the gap in this field (Fig. F). Paula Voabil et al. cultured tumor tissues of multiple patient sources in vitro and added PD-1 monoclonal antibody to the culture system for 48 h of co-incubation. Changes in 13 cytokines, 13 chemokines and 4 T cell activation markers were examined before and after the addition of the drug to generate a response score, and they found that the response score of the tumor tissues to the drug was highly consistent with the clinical response of patients, indicating that this model had a potential to predict the efficacy of early treatment with PD-1 monoclonal antibody. In addition, the authors performed a multifaceted analysis of tumor tissues in the untreated state, including the proportion and spatial distribution of immune cells and various cytokines and chemokines such as CXCL9, CXCL10, CXCL13, IL-8, and established a method to predict the efficacy of PD-1 monoclonal antibody therapy based on the baseline tumor condition. In another co-culture model, melanoma organoids were mixed with lymph node immune cells of the same patient origin in matrigel. 85% (6/7) of the organoids responded to immunotherapy with nivolumab, pembrolizumab, ipilimumab, and dabrafenib/trametinib in the same way as the actual dosing response in the clinic . Similarly, Myriam Chalabi et al. selected six nonresponder and six responder patients in a cohort of early-stage colon cancer patients receiving anti-PD-1 combined with anti-CTLA-4 neoadjuvant immunotherapy and constructed a co-culture model with PBMC. The results of the in vitro experiments reflected the drug efficacy of the patients to some extent, as T cells were activated and exhibited tumor cell killing in three responder patients but showed no reactivity to tumor organoids in nonresponder patients. The fact that three of the six responder patients did not show T-cell reactivity suggested that this model needs to be further optimized for predicting the efficacy of ICI therapy more accurately. Although 3D in vitro models has the potential to be used as a preliminary predictor of efficacy, the published studies are mainly small sample size studies. More large scale studies are needed in the future to verify the accuracy of efficacy prediction. Application of 3D in vitro models in drug screening and immunotherapy optimizing Despite the increasing application of immunotherapy, its effectiveness is still limited. Only a minority of patients can benefit from immunotherapy, and even in patients who are sensitive, problems of acquired drug resistance may occur. Therefore, new strategies are needed to optimize the effectiveness of immunotherapy. 3D in vitro models have been widely used in the evaluation of different treatment options and in vitro drug screening because of their ability to better reflect tumor characteristics and manipulability (Fig. G). For example, the response of patients to neoadjuvant chemoradiotherapy for colorectal cancer is highly consistent with the in vitro responsiveness of their corresponding organoids to radiotherapy and the same chemotherapeutic agents, and the in vitro drug sensitivity results are expected to guide the clinical treatment of patients . High-throughput screening of drugs using organoids can also help to find new therapeutic options for tumors that are resistant to conventional therapy . However, due to the lack of immune components, traditional organoid models cannot play a role in exploring more optimal immunotherapy regimens. In contrast, multiple 3D in vitro models that integrate immune cells can be useful in the optimization of immunotherapy. PD-1/PD-L1 inhibitors are commonly used immunotherapies in clinical practice, so enhancing the efficacy of PD-1/PD-L1 therapy through different pathways is at the forefront of research. Carminia Maria Della Corte et al. used a co-culture system of NSCLC to verify the synergistic effect of PD-L1 monoclonal antibody combined with MEK inhibitors in the treatment of NSCLC. They found that MEK inhibitors not only had a direct killing effect on tumor cells, but also promoted tumor recognition by CD8 + T cells, increased the expression of cytokines such as IFN?, IL12, IL6 and TNFa, and prevented T cell depletion by downregulating PD-L1, CTLA-4, TIM-3 and LAG-3. These results showed that MEK inhibitor had PD-L1 monoclonal sensitizing effects. Small molecule drugs were screened using a 3D microfluidic organoid culture system, and CDK4/6 inhibitors were found to significantly increase T cell activation and infiltration. Simultaneous application of PD-1 monoclonal antibodies and CDK4/6 inhibitors also showed enhanced T cell killing of tumor cells . Meanwhile, screening of epigenetic inhibitor library and herbal small molecule compound library using 3D in vitro model revealed that GSK-LSD1, CUDC-101, BML-210 and ATT-1 could increase the expression of MHC-I on tumor cell and promote tumor antigen presentation respectively, and the combination with PD-1 antibody could enhance the killing toxicity of CD8 + T cells . In addition to anti PD-1/PD-L1 therapy, other immunotherapeutic approaches can also be studied using 3D in vitro models. Qingda Meng et al. co-cultured pancreatic tumor organoids with PBMCs and added inhibitors of various immune checkpoints such as PD-1, PD-L1, TIM3, TIGIT, LAG3, and NKG2A to the co-culture system. NKG2A inhibitors were found to significantly elevate IFN-? expression in T cells, and the blockade of NKG2A-HLA-E axis was found to be a potential target for enhancing the killing capacity of CD8 + T cell for the treatment of pancreatic cancer. Marcel Alexander Heinrich et al. applied a 3D-bioprinted glioma model to study two immunomodulatory drugs AS1517449 (Stat6 inhibitor) and BLZ945 (Csf-1r inhibitor), which target macrophages. It was found that the function of macrophages was significantly inhibited after BLZ945 treatment, as evidenced by the decrease in Fgf2 and Mmp2 expression in macrophages. In addition, the growth rate of glioma cells was significantly slowed down. This reflected that targeting macrophages may also play a role in tumor immunotherapy. Application of 3D in vitro models in adoptive cell therapy In the field of tumor immunotherapy, apart from immune checkpoint inhibitors, adoptive cell transfer therapy (ACT) is also developing rapidly. This treatment involves isolating immunologically active cells from tumor patients, expanding and functionally characterizing them in vitro, and then re-infusing them into patients for the purpose of killing the tumor directly or stimulating immune response to eliminate tumor cells. According to the development of ACT, the adopted cells include lymphokine-activated killer cell (LAK), cytokine-induced killer cell (CIK), tumor infiltrating lymphocyte (TIL), natural killer cell (NK), cytotoxic T lymphocyte (CTL), chimeric antigen receptor T cell (CAR-T) and T cell engineered with T-cell receptor (TCR-T) . Although ACT has been successful in hematologic malignancies, especially CAR-T cells targeting CD19 in B cell lymphoma and acute lymphocytic leukemia , breakthrough in solid tumors has yet to be achieved, and 3D in vitro models are being broadly used in this field (Fig. H). Firstly, 3D in vitro models can be used as a source of tumor antigens for the preparation of tumor-specific T cells because of their high consistency with the original tumor. Krijn K. Dijkstra et al. co-cultured and screened PBMCs with CRC and NSCLC organoids for several rounds to obtain tumor-reactive T cells for T cell therapy. The iPTOs model constructed by mixing organoids and homologous immune cells could also be used to induce peripheral blood T cells to produce killing effects on tumor cells . Qingda Meng et al. performed TCR sequencing on peripheral blood T cells after co-cultured with pancreatic cancer organoids and found that a specific subpopulation of T cell clones expanded significantly. Cloning and transferring TCRs from this subpopulation into heterologous T cells could enable T cells to acquire the ability to specifically recognize and kill patient tumor cells, providing a potential idea for advanced adoptive cell therapy. Secondly, 3D in vitro models can also be used to assess the efficacy of adoptive cell therapy. By co-culturing the specific immune cells used in the adoptive cell therapy with organoids or other 3D in vitro cultures, the activation of immune cells and the apoptosis of tumors can be assessed. Fadi Jacob et al. co-cultured six glioma organoids which had different levels of EGFRvIII expression with 2173BBz CAR-T cells targeting EGFRvIII. After 72 h co-culture, they found that CAR-T cells infiltrated the organoids in all groups, but only when co-cultured with EGFRvIII positive organoids, CAR-T cells showed significant expansion, increased expression of granzyme and secretion of various cytokines. Increased apoptosis was found in EGRFvIII + tumor cells but not negative cells. This indicated that CAR-T cells were able to specifically kill target cells instead of complete elimination of all tumor cells. Such co-culture models provide a feasible way to test the efficacy of CAR-T therapy. Similarly, co-culture of human bladder cancer organoids with specific CAR-T cells allows detecting the killing ability and specificity of CAR-T cells . CAR-NK92 cells targeting EGFRvIII or FRIZZLED can also exhibit tumor antigen-specific cytotoxicity when co-cultured with CRC organoids in the ALI system . Evaluation of cell therapy efficacy can help optimize the therapeutic approach and guide personalized and precise treatment of patients.
Tumor cells and their microenvironment are a functional entity. The microenvironment interacts and co-evolves with tumor cells and play important roles in multiple processes of tumor development. Therefore, studying the interaction between tumor and microenvironment can help us understand the biological behavior of tumor and lay the theoretical foundation for finding new therapeutic targets and exploring new methods of tumor immunotherapy. 3D in vitro models are good models for studying the TME because of their simplicity, convenience, flexibility and high similarity to original tumor tissues (Fig. E). For example, in tumor microenvironment, T cells and NK cells are usually the main ones that can directly kill tumor cells. Enhancing their killing ability contributes to tumor control. Using a co-culture model of CRC organoids and T cells, researchers screened a series of small molecule inhibitors and found that DKK1 inhibitors could significantly enhance the killing effect of T cells and promote apoptosis of tumor cells by regulating the GSK3ß-E2F1-T-bet axis in CD8 + T cells. DKK1 inhibitors combined with PD-1 monoclonal antibodies could achieve better tumor control . Similarly, BRD1 inhibitors were able to convert NK cells and naïve CD8 + T cells to a more activated and cytotoxic phenotype, helping anti-PD-1/PD-L1 bispecific antibodies to exert a more effective antitumor immune effect in HGSC . By detecting the expression of surface biomarkers on tumor cells and immune cells before and after co-culture, antibodies targeting MICA/B and NKG2A were also found to enhance the killing effect of T cells and NK cells on colorectal cancer organoids, thus providing a potential target for the treatment of CRC . In addition to immune cells that play a major role in tumor killing, there are multiple immunosuppressive components in the TME that can promote tumor progression. Co-culture of mouse MDCs with small intestinal adenoma organoids revealed that MDCs were transformed into TAMs and stimulated adenoma growth via the COX-2-PEG2-EP4 pathway . Using mouse ovarian cancer tumor spheroids co-cultured with TAMs, it was found that when UBR5 was knocked down, the tumor spheroids had slower growth and smaller size. The ability to recruit TAMs was also reduced, as was the expression of cytokines and chemokines associated with TAMs recruitment. This demonstrated that targeting UBR5 could help control the growth of ovarian cancer by modulating the tumor immune microenvironment . Using the similar co-culture method, researchers found that breast fibroblasts secreted cytokines such as IL-1ß, which acted on breast cancer cells through a paracrine pathway to promote their proliferation. Blocking this paracrine pathway enhanced the therapeutic effect of tomaxifen on breast cancer, suggesting that fibroblasts in the remaining breast tissue after breast-conserving surgery may increase the risk of breast cancer recurrence . These studies have explored the interaction of tumor cells with different components of the microenvironment through co-culture models, providing a direction for the development of immunotherapy.
With the increasing research on the immune microenvironment, immunotherapy has been more and more widely used in the treatment of tumors, among which the most frequently applied is immune checkpoint inhibitor therapy. ICI therapy targeting PD1/PD-L1 and CTLA-4 has been clinically employed in progressive melanoma , squamous cell skin cancer , non-small cell lung cancer , renal cell carcinoma , head and neck tumors, and tumors with mismatch repair defects of various tissue types. However, only some of the patients treated with ICI have benefited from this therapy. Considering the adverse effects and financial burden, it is necessary to carefully select patients who may benefit from ICI treatment and personalize the treatment with precision. For precise patient selection, a number of biomarkers have been demonstrated to predict the efficacy of ICI therapy, such as MSI status , TMB , neoantigen expression levels , CD8 + T cell counts , and tumor cell surface PD-L1 expression levels , but none of these indicators have sufficiently high positive and negative predictive values to allow effective screening of patients . Other studies have analyzed immunobiological indicators of the tumor, such as genomic, transcriptomic and proteomic sequencing analysis, or have integrated multiple biomarkers, but even these more complicated analyze methods do not fully reflect the intratumoral and individual heterogeneity of human tumors and cannot accurately predict ICI treatment efficacy . Therefore, static predictors cannot meet clinical needs. It is important to establish an efficacy assessment model that can be monitored dynamically, and the in vitro 3D culture may fill the gap in this field (Fig. F). Paula Voabil et al. cultured tumor tissues of multiple patient sources in vitro and added PD-1 monoclonal antibody to the culture system for 48 h of co-incubation. Changes in 13 cytokines, 13 chemokines and 4 T cell activation markers were examined before and after the addition of the drug to generate a response score, and they found that the response score of the tumor tissues to the drug was highly consistent with the clinical response of patients, indicating that this model had a potential to predict the efficacy of early treatment with PD-1 monoclonal antibody. In addition, the authors performed a multifaceted analysis of tumor tissues in the untreated state, including the proportion and spatial distribution of immune cells and various cytokines and chemokines such as CXCL9, CXCL10, CXCL13, IL-8, and established a method to predict the efficacy of PD-1 monoclonal antibody therapy based on the baseline tumor condition. In another co-culture model, melanoma organoids were mixed with lymph node immune cells of the same patient origin in matrigel. 85% (6/7) of the organoids responded to immunotherapy with nivolumab, pembrolizumab, ipilimumab, and dabrafenib/trametinib in the same way as the actual dosing response in the clinic . Similarly, Myriam Chalabi et al. selected six nonresponder and six responder patients in a cohort of early-stage colon cancer patients receiving anti-PD-1 combined with anti-CTLA-4 neoadjuvant immunotherapy and constructed a co-culture model with PBMC. The results of the in vitro experiments reflected the drug efficacy of the patients to some extent, as T cells were activated and exhibited tumor cell killing in three responder patients but showed no reactivity to tumor organoids in nonresponder patients. The fact that three of the six responder patients did not show T-cell reactivity suggested that this model needs to be further optimized for predicting the efficacy of ICI therapy more accurately. Although 3D in vitro models has the potential to be used as a preliminary predictor of efficacy, the published studies are mainly small sample size studies. More large scale studies are needed in the future to verify the accuracy of efficacy prediction.
Despite the increasing application of immunotherapy, its effectiveness is still limited. Only a minority of patients can benefit from immunotherapy, and even in patients who are sensitive, problems of acquired drug resistance may occur. Therefore, new strategies are needed to optimize the effectiveness of immunotherapy. 3D in vitro models have been widely used in the evaluation of different treatment options and in vitro drug screening because of their ability to better reflect tumor characteristics and manipulability (Fig. G). For example, the response of patients to neoadjuvant chemoradiotherapy for colorectal cancer is highly consistent with the in vitro responsiveness of their corresponding organoids to radiotherapy and the same chemotherapeutic agents, and the in vitro drug sensitivity results are expected to guide the clinical treatment of patients . High-throughput screening of drugs using organoids can also help to find new therapeutic options for tumors that are resistant to conventional therapy . However, due to the lack of immune components, traditional organoid models cannot play a role in exploring more optimal immunotherapy regimens. In contrast, multiple 3D in vitro models that integrate immune cells can be useful in the optimization of immunotherapy. PD-1/PD-L1 inhibitors are commonly used immunotherapies in clinical practice, so enhancing the efficacy of PD-1/PD-L1 therapy through different pathways is at the forefront of research. Carminia Maria Della Corte et al. used a co-culture system of NSCLC to verify the synergistic effect of PD-L1 monoclonal antibody combined with MEK inhibitors in the treatment of NSCLC. They found that MEK inhibitors not only had a direct killing effect on tumor cells, but also promoted tumor recognition by CD8 + T cells, increased the expression of cytokines such as IFN?, IL12, IL6 and TNFa, and prevented T cell depletion by downregulating PD-L1, CTLA-4, TIM-3 and LAG-3. These results showed that MEK inhibitor had PD-L1 monoclonal sensitizing effects. Small molecule drugs were screened using a 3D microfluidic organoid culture system, and CDK4/6 inhibitors were found to significantly increase T cell activation and infiltration. Simultaneous application of PD-1 monoclonal antibodies and CDK4/6 inhibitors also showed enhanced T cell killing of tumor cells . Meanwhile, screening of epigenetic inhibitor library and herbal small molecule compound library using 3D in vitro model revealed that GSK-LSD1, CUDC-101, BML-210 and ATT-1 could increase the expression of MHC-I on tumor cell and promote tumor antigen presentation respectively, and the combination with PD-1 antibody could enhance the killing toxicity of CD8 + T cells . In addition to anti PD-1/PD-L1 therapy, other immunotherapeutic approaches can also be studied using 3D in vitro models. Qingda Meng et al. co-cultured pancreatic tumor organoids with PBMCs and added inhibitors of various immune checkpoints such as PD-1, PD-L1, TIM3, TIGIT, LAG3, and NKG2A to the co-culture system. NKG2A inhibitors were found to significantly elevate IFN-? expression in T cells, and the blockade of NKG2A-HLA-E axis was found to be a potential target for enhancing the killing capacity of CD8 + T cell for the treatment of pancreatic cancer. Marcel Alexander Heinrich et al. applied a 3D-bioprinted glioma model to study two immunomodulatory drugs AS1517449 (Stat6 inhibitor) and BLZ945 (Csf-1r inhibitor), which target macrophages. It was found that the function of macrophages was significantly inhibited after BLZ945 treatment, as evidenced by the decrease in Fgf2 and Mmp2 expression in macrophages. In addition, the growth rate of glioma cells was significantly slowed down. This reflected that targeting macrophages may also play a role in tumor immunotherapy.
In the field of tumor immunotherapy, apart from immune checkpoint inhibitors, adoptive cell transfer therapy (ACT) is also developing rapidly. This treatment involves isolating immunologically active cells from tumor patients, expanding and functionally characterizing them in vitro, and then re-infusing them into patients for the purpose of killing the tumor directly or stimulating immune response to eliminate tumor cells. According to the development of ACT, the adopted cells include lymphokine-activated killer cell (LAK), cytokine-induced killer cell (CIK), tumor infiltrating lymphocyte (TIL), natural killer cell (NK), cytotoxic T lymphocyte (CTL), chimeric antigen receptor T cell (CAR-T) and T cell engineered with T-cell receptor (TCR-T) . Although ACT has been successful in hematologic malignancies, especially CAR-T cells targeting CD19 in B cell lymphoma and acute lymphocytic leukemia , breakthrough in solid tumors has yet to be achieved, and 3D in vitro models are being broadly used in this field (Fig. H). Firstly, 3D in vitro models can be used as a source of tumor antigens for the preparation of tumor-specific T cells because of their high consistency with the original tumor. Krijn K. Dijkstra et al. co-cultured and screened PBMCs with CRC and NSCLC organoids for several rounds to obtain tumor-reactive T cells for T cell therapy. The iPTOs model constructed by mixing organoids and homologous immune cells could also be used to induce peripheral blood T cells to produce killing effects on tumor cells . Qingda Meng et al. performed TCR sequencing on peripheral blood T cells after co-cultured with pancreatic cancer organoids and found that a specific subpopulation of T cell clones expanded significantly. Cloning and transferring TCRs from this subpopulation into heterologous T cells could enable T cells to acquire the ability to specifically recognize and kill patient tumor cells, providing a potential idea for advanced adoptive cell therapy. Secondly, 3D in vitro models can also be used to assess the efficacy of adoptive cell therapy. By co-culturing the specific immune cells used in the adoptive cell therapy with organoids or other 3D in vitro cultures, the activation of immune cells and the apoptosis of tumors can be assessed. Fadi Jacob et al. co-cultured six glioma organoids which had different levels of EGFRvIII expression with 2173BBz CAR-T cells targeting EGFRvIII. After 72 h co-culture, they found that CAR-T cells infiltrated the organoids in all groups, but only when co-cultured with EGFRvIII positive organoids, CAR-T cells showed significant expansion, increased expression of granzyme and secretion of various cytokines. Increased apoptosis was found in EGRFvIII + tumor cells but not negative cells. This indicated that CAR-T cells were able to specifically kill target cells instead of complete elimination of all tumor cells. Such co-culture models provide a feasible way to test the efficacy of CAR-T therapy. Similarly, co-culture of human bladder cancer organoids with specific CAR-T cells allows detecting the killing ability and specificity of CAR-T cells . CAR-NK92 cells targeting EGFRvIII or FRIZZLED can also exhibit tumor antigen-specific cytotoxicity when co-cultured with CRC organoids in the ALI system . Evaluation of cell therapy efficacy can help optimize the therapeutic approach and guide personalized and precise treatment of patients.
in vitro models Problems in the construction and analysis of 3D in vitro models Although 3D in vitro models have an increasingly wide range of applications in various research fields, there are still some problems in the process of model construction. First, there is no uniform standard for the culture of in vitro models. For example, the culture protocols of the same tumor-derived organoids in different laboratories often differ, such as different culture media components. Some niche factors commonly used in organoid culture are produced by different cell lines and often have batch effects. These unstable factors can have an impact on the experimental results . In addition, extracellular matrices are essential during the construction of 3D in vitro models. Matrigel or other animal-based matrix extract components are most commonly used. These matrix components often have batch-to-batch variation, which can affect the reproducibility of experiments and lead to unstable results. In addition, they may carry unknown pathogens or other immunogenic components that may not only affect the growth of tumor organoids, but may also lead to non-specific activation of immune cells in the presence of immune components in the model, thus affecting the stability of the in vitro simulated -TME . In addition, during tumor organoid culture, normal epithelial cells grow faster than tumor cells and therefore dominate the culture process and inhibit the growth of tumor organoids. To solve this problem, it is necessary to correctly identify the tumor tissues when sampling after surgery and avoid taking the normal tissues as much as possible. During the culture process, tumor organoids can also be selected by adjusting the composition of the culture medium. For example, some colorectal cancers have mutations in the Wnt pathway and can survive in environments without Wnt3a, so colorectal cancer organoids can be picked from normal organoids by removing the Wnt3a factor from the culture medium during the culture process . There are also some tumor organoids that are morphologically different from their corresponding normal tissue organoids, so the normal tissue organoids can be manually removed to ensure the better survival of tumor organoids . Finally, the tumor tissues themselves are very complex in structure and have intratumoral heterogeneity. In the process of in vitro model construction, due to the limitation of culture conditions, the prolongation of culture time and the increase of the number of passages, only a part of tumor cells adapted to the culture conditions can survive, resulting in the loss of intratumoral heterogeneity . During the long time of in vitro culture, new mutations will also be accumulated, which leads to the increase of differences between in vitro models and in vivo tumors. So more optimized culture systems and conditions are needed to solve these problems in the future. Another important difficulty faced when using 3D models for research is the frequent lack of validated evaluation tools. Evaluation methods commonly used in 2D cultures may not be applicable to 3D culture systems due to the unique construction methods and growth characteristics of 3D cell clusters. In various Label-dependent and reader-based assays for cell viability and cytotoxicity, such as MTT assay, LDH activity assay, CellTiter-Glo®3D Cell Viability and fluorescence imaging, the presence of extracellular matrix and the tight intercellular junctions in 3D aggregated cell clusters lead to difficulties in cell lysis and poor penetration of dyes and reagents, resulting in reduced sensitivity of the assay and biased results . In addition, when the organoids are large, there is often cell necrosis in the central region, which produce stronger apoptotic signals and may mislead the experimental results . Therefore, the use of such assays requires improvements in the timing of cell lysis and the diffusion penetration of the dye and reagents, as well as stricter control of the size of organoids. Evaluating changes in the size and morphology of organoids using microscope is also commonly used in assessing the growth state and drug effect. However, organoids also suffer from poor light transmission during imaging, light scattering, and increased background fluorescence intensity due to fluorescence outside the focal plane . In addition, in 2D culture, xy images obtained by microscopy can be used to evaluate the cell growth. But since the 3D structure, the images used to evaluate organoids need to include a series of xy images obtained on the same z-axis, forming a z-stack. Z-stack images also need to be processed by special software, such as z-projection in imagej, to integrate the series of images into a maximal projection image . Different analysis methods need to be applied in combination to overcome these shortcomings. Problems in the reconstruction of tumor immune microenvironment In addition to the problems in the construction of 3D in vitro models, there are also some challenges in using these models to reconstruct the in vitro tumor immune microenvironment. The tumor immune microenvironment is a complex system composed of multiple immune cells and stromal cells, which is highly heterogeneous and dynamic. How to simulate the real state of the in vivo tumor microenvironment as much as possible in vitro needs to be further optimized and refined. As in the organoid/immune cell co-culture model, the type of immune cells added externally is more limited, and the status of the additional immune cells is not consistent with the original cell in the immune microenvironment, and therefore differs significantly from the original microenvironment of the tumor. Moreover, the organoid culture medium cannot maintain the function of immune cells in an optimal condition, so the cellular activity and survival time of immune cells may be affected . In contrast, in ALI or microfluidic culture models a portion of the original tumor infiltrating immune cells are retained in the organoid. These immune cell fractions are maintained for only a limited period of time and are subject to greater loss during passaging or freezing and thawing. Therefore these methods are only suitable for short-term studies . In addition, such systems require high quality of tumor samples. Tumor biopsy samples or puncture specimens do not have sufficient microenvironmental components due to small sample size, therefore it is difficult to use these specimens for in vitro construction of TME. Considering that tumor tissues are spatially heterogeneous, the microenvironment of tumor margins and central sites are often different. Attention should be paid to the representativeness of the samples taken for in vitro culture, and samples should be taken from different regions of the tumor to improve the representation of the samples to tumor tissues . What’s more, the tumor immune microenvironment is in a dynamic state of change, and circulating immune cells have complex interactions with the tumors and tumor infiltrating immune cells. These models only take into account the local immune situation of the TME but not the circulating immune cells. In summary, different in vitro culture models have their unique advantages and corresponding shortcomings, and therefore the most suitable model needs to be selected according to the research objectives. Combining multiple models, integrating their advantages, and further optimizing the culture conditions are ways to refine in vitro 3D culture models for better application in tumor immunology research in the future.
Although 3D in vitro models have an increasingly wide range of applications in various research fields, there are still some problems in the process of model construction. First, there is no uniform standard for the culture of in vitro models. For example, the culture protocols of the same tumor-derived organoids in different laboratories often differ, such as different culture media components. Some niche factors commonly used in organoid culture are produced by different cell lines and often have batch effects. These unstable factors can have an impact on the experimental results . In addition, extracellular matrices are essential during the construction of 3D in vitro models. Matrigel or other animal-based matrix extract components are most commonly used. These matrix components often have batch-to-batch variation, which can affect the reproducibility of experiments and lead to unstable results. In addition, they may carry unknown pathogens or other immunogenic components that may not only affect the growth of tumor organoids, but may also lead to non-specific activation of immune cells in the presence of immune components in the model, thus affecting the stability of the in vitro simulated -TME . In addition, during tumor organoid culture, normal epithelial cells grow faster than tumor cells and therefore dominate the culture process and inhibit the growth of tumor organoids. To solve this problem, it is necessary to correctly identify the tumor tissues when sampling after surgery and avoid taking the normal tissues as much as possible. During the culture process, tumor organoids can also be selected by adjusting the composition of the culture medium. For example, some colorectal cancers have mutations in the Wnt pathway and can survive in environments without Wnt3a, so colorectal cancer organoids can be picked from normal organoids by removing the Wnt3a factor from the culture medium during the culture process . There are also some tumor organoids that are morphologically different from their corresponding normal tissue organoids, so the normal tissue organoids can be manually removed to ensure the better survival of tumor organoids . Finally, the tumor tissues themselves are very complex in structure and have intratumoral heterogeneity. In the process of in vitro model construction, due to the limitation of culture conditions, the prolongation of culture time and the increase of the number of passages, only a part of tumor cells adapted to the culture conditions can survive, resulting in the loss of intratumoral heterogeneity . During the long time of in vitro culture, new mutations will also be accumulated, which leads to the increase of differences between in vitro models and in vivo tumors. So more optimized culture systems and conditions are needed to solve these problems in the future. Another important difficulty faced when using 3D models for research is the frequent lack of validated evaluation tools. Evaluation methods commonly used in 2D cultures may not be applicable to 3D culture systems due to the unique construction methods and growth characteristics of 3D cell clusters. In various Label-dependent and reader-based assays for cell viability and cytotoxicity, such as MTT assay, LDH activity assay, CellTiter-Glo®3D Cell Viability and fluorescence imaging, the presence of extracellular matrix and the tight intercellular junctions in 3D aggregated cell clusters lead to difficulties in cell lysis and poor penetration of dyes and reagents, resulting in reduced sensitivity of the assay and biased results . In addition, when the organoids are large, there is often cell necrosis in the central region, which produce stronger apoptotic signals and may mislead the experimental results . Therefore, the use of such assays requires improvements in the timing of cell lysis and the diffusion penetration of the dye and reagents, as well as stricter control of the size of organoids. Evaluating changes in the size and morphology of organoids using microscope is also commonly used in assessing the growth state and drug effect. However, organoids also suffer from poor light transmission during imaging, light scattering, and increased background fluorescence intensity due to fluorescence outside the focal plane . In addition, in 2D culture, xy images obtained by microscopy can be used to evaluate the cell growth. But since the 3D structure, the images used to evaluate organoids need to include a series of xy images obtained on the same z-axis, forming a z-stack. Z-stack images also need to be processed by special software, such as z-projection in imagej, to integrate the series of images into a maximal projection image . Different analysis methods need to be applied in combination to overcome these shortcomings.
In addition to the problems in the construction of 3D in vitro models, there are also some challenges in using these models to reconstruct the in vitro tumor immune microenvironment. The tumor immune microenvironment is a complex system composed of multiple immune cells and stromal cells, which is highly heterogeneous and dynamic. How to simulate the real state of the in vivo tumor microenvironment as much as possible in vitro needs to be further optimized and refined. As in the organoid/immune cell co-culture model, the type of immune cells added externally is more limited, and the status of the additional immune cells is not consistent with the original cell in the immune microenvironment, and therefore differs significantly from the original microenvironment of the tumor. Moreover, the organoid culture medium cannot maintain the function of immune cells in an optimal condition, so the cellular activity and survival time of immune cells may be affected . In contrast, in ALI or microfluidic culture models a portion of the original tumor infiltrating immune cells are retained in the organoid. These immune cell fractions are maintained for only a limited period of time and are subject to greater loss during passaging or freezing and thawing. Therefore these methods are only suitable for short-term studies . In addition, such systems require high quality of tumor samples. Tumor biopsy samples or puncture specimens do not have sufficient microenvironmental components due to small sample size, therefore it is difficult to use these specimens for in vitro construction of TME. Considering that tumor tissues are spatially heterogeneous, the microenvironment of tumor margins and central sites are often different. Attention should be paid to the representativeness of the samples taken for in vitro culture, and samples should be taken from different regions of the tumor to improve the representation of the samples to tumor tissues . What’s more, the tumor immune microenvironment is in a dynamic state of change, and circulating immune cells have complex interactions with the tumors and tumor infiltrating immune cells. These models only take into account the local immune situation of the TME but not the circulating immune cells. In summary, different in vitro culture models have their unique advantages and corresponding shortcomings, and therefore the most suitable model needs to be selected according to the research objectives. Combining multiple models, integrating their advantages, and further optimizing the culture conditions are ways to refine in vitro 3D culture models for better application in tumor immunology research in the future.
in vitro models The emerging 3D in vitro models have greatly facilitated tumor immunology research. However these models can be further optimized to improve the clinical translation capability and to expand the application. More integrated 3D in vitro models are expected in the future. First is the integration of multiple models. The advantages of multiple models should be utilized to establish a more complete immune system in vitro. For example, in the ALI model, the original immune cells in the TME can be preserved. In contrast, the infiltration of immune cells into tumor organoids in the co-culture model better mimics the process of circulating immune cells trafficking into the tumor and undergoing phenotypic changes. If the two models are combined together, a more comprehensive simulation of the immune response process can be achieved. Second is the functional integration of 3D in vitro models. Simulating the immune microenvironment in vitro does not simply mean putting tumor cells and immune cells in the same spatial space, but more critically, allowing them to interact more functionally. In order to achieve this, it is necessary to restore the in vivo physiological state as much as possible. For example, cells can be exposed to mechanical stress, substrate stiffness and physiologic shear flow. It is also necessary to control the shape and size of the model as well as the ratio of each cell according to the characteristics of different tumor tissues. In these areas, 3D-microfluidic culture systems may have a strong potential. Finally is the integration of other components of the TME. In the TME, there are various components besides immune cells that play important roles, such as tumor-associated fibroblasts, blood vessels and neurons. The microbiota also has close interactions with the TME. Therefore, if multiple microenvironment components can be integrated into the same system in future studies, 3D in vitro models will have more in-depth applications in oncology research.
In a few years of recent, researchers have taken great efforts to reconstruct tumor immune microenvironment using various 3D models in vitro. The newly developed models including Organoid/immune cell co-culture models, ALI culture model, 3D-Microfluidic based culture method and 3D-bioprinting model have been used in different fields of tumor immunology research and greatly helped the selection of patients benefit from immunotherapy, the development of novel treatment approach and the research of resistance to immunotherapy. In the future, 3D in vitro models can be further enhanced to simulate the tumor microenvironment in a more integrated way both structurally and functionally, to better facilitate tumor immunology research.
|
Patterns and predictors of public dental service utilisation among refugees in Victoria, Australia: a latent profile and multilevel analysis | 6a27511a-cf0f-4f3b-bcca-5ce221795710 | 10074673 | Dental[mh] | Poor oral health is considered a major burden among the resettled refugee population around the world . This can be attributed to a myriad of past traumatic experiences in their home country, compounded by the challenges related to resettlement in the host country. In Australia, refugees have been shown to demonstrate higher rates of dental caries and periodontal disease compared to the general population [ – ]. This is also acknowledged in the most recent National Oral Health Plan of Australia, 2015–2024 , which identifies refugees as a vulnerable group. Timely use of appropriate dental services would contribute to promoting oral health of refugees by providing diagnosis, prevention, and treatment of oral diseases. In Australia, the Commonwealth government entitles all humanitarian migrants, including refugees, eligibility to access universal healthcare. However, dental services are not included in this. Recognising the vulnerability of the refugee group, the Department of Health and Human Services in the state of Victoria, extended the eligibility for public dental services (PDS) to refugees and asylum seekers . Furthermore, to overcome the frequently noted barriers such as long waiting times and financial constraints, additional policies were introduced to provide refugee populations with ‘priority access,’ where they are accommodated in the next available appointment without being placed on a waiting list, and fee exemption for all services . Despite these measures, data from a 2016 audit suggests limited participation rates among refugees within the Victorian public dental system; approximately 17% of Victorian refugees attended PDS in 2015-16 . Little is known about the characteristics of, and factors associated with dental service utilisation among refugees in Australia, in general, and in Victoria, in particular. Dental service use is considered an important indicator of dental health-related behaviour . Understanding of the pattern of PDS use among the refugee populations provides a valuable insight into their access to dental care; for example, whether their pattern of use comprise routine check-ups and preventive care or is it primarily for the treatment of existing dental problems. In addition, it is also critical to analyse the nature of services received by refugees when visiting a dental service provider (e.g., routine check-up, preventive services or specialist treatment) to identify disparities in the utilisation of particular types of services, investigate problems associated with access to these services, and further examine factors determining their utilisation behaviour. Together, these inform the development of targeted strategies that would enable efficient use of the existing resources by the public dental system to serve this population group. Previous research examining dental service utilisation behaviour of refugees were primarily conducted using self-reported surveys . Use of survey data in this regard may be limited by the sample size and characteristics, individual’s recollection of past events such as dental attendance . As a result, significant discrepancies were found between self-reported and actual utilisation . Administrative data provides more accurate information on service utilisation, as the treatments received by the patient represent real-life patterns of care and is precisely recorded at the time of care provision . One study used administrative data to investigate public and private dental service utilisation among refugees in Sweden, the results of which showed a low overall use . However, there are no such studies in Australia, warranting research in this space. This study uses administrative data across a four-year period to examine the utilisation of publicly funded dental services and develop profiles of PDS use among the Victorian refugees. The specific objectives of this study were to: (1) explore patterns of the use of different types of care and services provided through the Victorian public dental health system among the refugee population, (2) analyse and describe the characteristics of refugees with similar patterns of PDS use, (3) examine the association between individual and contextual factors of the refugees and their identified patterns of PDS use.
Data source This retrospective observational study used secondary data analysis of de-identified individual-level data of refugees who have accessed publicly funded dental services in Victoria. The data was obtained from the electronic dental records sourced from the Victorian Dental Health Program dataset (DHPDS). All data were provided by Dental Health Services Victoria (DHSV). Population characteristics The study population included all refugees, irrespective of their age, gender, or ethnicity, for whom a record was created within the Titanium® patient management system between 01 July 2016 and 30 June 2020. Eligible patients attended the Royal Dental Hospital Melbourne (RDHM) or any of the community dental clinics to avail PDS. Services received were any of the three types of courses of care (CoC) – general, emergency, and denture – and one or more service areas provided as part of any of these three CoCs, including diagnostic, preventive and specialist services. Study variables Measures of dental service utilisation included the number and type of CoCs, services received in each visit during the CoC, and date and clinic of the visit. The data collected from the electronic records contain one record per client per visit during a CoC, which allowed for the measurement of frequency of CoCs and visits per service area per client per year. Individual service items were coded using the coding scheme outlined by the Australian Dental Association , and grouped into eleven major service areas – consultations; oral and radiographic examinations; prophylactic and preventive; periodontics; extractions; minor, major and other surgery; endodontics; restorative; crown, bridge and implants; complete and partial dentures; orthodontics. Other variables collected were the type of referral and address (including suburb and postcode) of the clinic site where services were availed. Individual-level variables collected included client demographics as recorded in their first visit - age, gender, country of birth (stratified according to United Nations geographic regions) , preferred language for communication, request for language interpreter service (dummy coded – Yes/No), and type of eligibility card held (health care card, pensioner card, no card). Context was defined as residential neighbourhood of refugee clients at statistical area level 2 (SA2), corresponding to their residential suburb and postcode collected from the electronic records . SA2s are geographical units defined by the Australian Bureau of Statistics which closely align with the boundaries of suburbs in the metropolitan areas and represent ‘functional communities’ that are socially and economically interactive in outside metropolitan areas . All contextual variables were gathered at SA2 level. These included urbanicity of residence (metropolitan, regional, rural) , measure of area-level socioeconomic disadvantage for refugees, whether the Victorian Refugee Health Program (RHP) was delivered at the community health centre (CHC) in the clients’ residential SA2 (dummy coded, Yes/No), and spatial accessibility to PDS via driving and public transit travel modes. Data on variables indicating refugee socioeconomic disadvantage were obtained from the Australian Census and Migrants Integrated Dataset, 2016 . These included proportion of total resident SA2 population who are refugees and proportion of total refugees in each SA2 who – moved to Australia during the last 5 years (as of 2016), did not complete Year 12, are not proficient in English, are above 15 years and unemployed, have an annual income <$25,999, need assistance with core activities, and live in households without a motor vehicle . Principal Components Analysis was conducted to reduce the dimensionality of these variables and obtain a unified measure of SA2-level socioeconomic disadvantage for the refugee population. Based on the Kaiser’s criterion, five components (capturing 80.02% total variance) with eigenvalues greater than one were combined using their respective eigenvalues as weightings (see Additional file 1 for details). The resulting scores were classified into tertiles. List of RHP sites were gathered from the Victorian Department of Health and were assigned to their respective SA2s based on their suburb and postcode in the postal address . Spatial accessibility to PDS was calculated using the enhanced two-step floating catchment area method individually via road network (for driving mode including car or other motor vehicle) and public transit network (for various public transit modes including bus, tram, metro, or train), as detailed elsewhere . The spatial accessibility index scores obtained from these calculations represent the ratio of full-time equivalent dental professionals to the population eligible for PDS within each SA2, weighted by the travel time between their respective locations via driving or public transit mode; these scores were used as continuous variables. Statistical analysis Latent profile analysis (LPA) was used to identify distinct subgroups that characterise the utilisation of PDS among the Victorian refugee population for the period 2016-17 to 2019-20. LPA is a person-centered model-based approach to identifying underlying subgroups (called latent profiles) based on an unobservable attribute (called latent variable) by assessing multiple dimensions indicated by measured variables (called indicator variables) pertaining to this attribute . The indicator variables can have a continuous, count, or a combination of these distributions. Individuals are probabilistically assigned to the latent profiles based on two model parameters estimated on a maximum-likelihood basis : (a) profile membership probabilities; (b) means and variances of indicator variables, conditional on profile membership. Profiles of individuals sharing similar patterns of the means and variances of each indicator variable are identified and grouped. This enables the distinctness of each identified profiles to be assessed and qualitatively described. Indicator variables of refugees’ PDS utilisation used in the LPA included attributes of dental services received and clinic sites of service provision, during the four-year period. Number of CoCs received per client in each of the three types of care (i.e., general, emergency, and denture) and by each of two referral types (i.e., self-referral or referred by dental professionals, health care professionals, refugee and community support services, and other support services), as well as number of unique visits in each of the eleven service areas within a CoC were indicators of dental service characteristics. For clinic characteristics, number of CoCs received based on urbanicity of the clinic site (i.e., metropolitan, regional, or rural site) , whether the site was within the SA2 of clients’ residence (i.e., within or outside SA2 of residence), and whether the clinic was co-located with CHCs delivering RHP (i.e., co-located or not co-located) were used as indicators. All indicators were counts; hence, a Poisson model was used in the LPA . As there were differences in the number of years in which each client attended PDS, during the study period, it was included as an offset in the Poisson model. Doing so adjusted for any difference in utilisation among the clients by modelling counts as rates (i.e., number of CoCs per year or number of visits per year). Therefore, based on the indicators included in the LPA, the final profile assignment of individuals was based on the combined patterns in the conditional mean rates of utilisation in each of the indicator variables. LPA model building process was done iteratively. First, a model with one latent profile was fitted. Next, the number of profiles were augmented in a step wise manner until the models no longer converged . From thus obtained models, the best fitted model with the optimal number of profiles was selected based on the following criteria : (a) relative fit statistics – Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC); (b) classification diagnostics – entropy values closer to 1, average posterior probability of profile membership > 0.7, odds of correct classification based on posterior probabilities > 5 for each profile group; (c) substantive model interpretability and parsimony. Finally, individuals were assigned to a profile based on their maximum posterior probabilities. Multilevel multinomial logistic regression was performed to examine the role of individual and contextual level variables pertaining to refugees (independent variables) in predicting their profile membership (dependent variable). As clients in the DHPDS data are clustered at the neighbourhood (SA2) level, a two-level random-intercept model was fitted with individuals (level 1) nested within contexts (level 2) . Associations between the independent variables and the profile membership were tested in bivariate analysis, and only variables with significant association (p < .05) were included in the multilevel multivariate analysis. The first model was a null model, which only included the dependent variable with its variance split in the two levels of analysis. Subsequently, individual and contextual level variables were included in blocks in the second and third models, respectively. The association between profile membership and each independent variable was adjusted for in terms of all the other variables included in the models. The amount of contextual-level variation in the patterns of PDS utilisation was determined by estimating intra-class correlation coefficient (ICC) using the null model and the extent to which the independent variables were able to explain this variation was determined by calculating proportional change in variance for Models 2 and 3, in reference to the null model . Coefficients from the regression models were exponentiated to obtain conditional odds ratio (COR) with 95% confidence intervals (CI) . Characteristics of the patients and the identified latent profiles were summarised using frequency and percentages. Chi-squared test was used to examine the differences in distributions of individual and contextual characteristics across the identified profiles. All models were estimated via a Generalised Structural Equation Modelling in Stata 17, using gsem command with poisson log link (LPA models) and multinomial logit link (multilevel models) functions . Less than 1.5% of data values were missing on explanatory variables (see Additional file 1, Table S3), which may be considered insignificant relative to the general standard of 5% . So, listwise deletion approach was used to handle missing data. A p value of < 0.05 was considered statistically significant.
This retrospective observational study used secondary data analysis of de-identified individual-level data of refugees who have accessed publicly funded dental services in Victoria. The data was obtained from the electronic dental records sourced from the Victorian Dental Health Program dataset (DHPDS). All data were provided by Dental Health Services Victoria (DHSV).
The study population included all refugees, irrespective of their age, gender, or ethnicity, for whom a record was created within the Titanium® patient management system between 01 July 2016 and 30 June 2020. Eligible patients attended the Royal Dental Hospital Melbourne (RDHM) or any of the community dental clinics to avail PDS. Services received were any of the three types of courses of care (CoC) – general, emergency, and denture – and one or more service areas provided as part of any of these three CoCs, including diagnostic, preventive and specialist services.
Measures of dental service utilisation included the number and type of CoCs, services received in each visit during the CoC, and date and clinic of the visit. The data collected from the electronic records contain one record per client per visit during a CoC, which allowed for the measurement of frequency of CoCs and visits per service area per client per year. Individual service items were coded using the coding scheme outlined by the Australian Dental Association , and grouped into eleven major service areas – consultations; oral and radiographic examinations; prophylactic and preventive; periodontics; extractions; minor, major and other surgery; endodontics; restorative; crown, bridge and implants; complete and partial dentures; orthodontics. Other variables collected were the type of referral and address (including suburb and postcode) of the clinic site where services were availed. Individual-level variables collected included client demographics as recorded in their first visit - age, gender, country of birth (stratified according to United Nations geographic regions) , preferred language for communication, request for language interpreter service (dummy coded – Yes/No), and type of eligibility card held (health care card, pensioner card, no card). Context was defined as residential neighbourhood of refugee clients at statistical area level 2 (SA2), corresponding to their residential suburb and postcode collected from the electronic records . SA2s are geographical units defined by the Australian Bureau of Statistics which closely align with the boundaries of suburbs in the metropolitan areas and represent ‘functional communities’ that are socially and economically interactive in outside metropolitan areas . All contextual variables were gathered at SA2 level. These included urbanicity of residence (metropolitan, regional, rural) , measure of area-level socioeconomic disadvantage for refugees, whether the Victorian Refugee Health Program (RHP) was delivered at the community health centre (CHC) in the clients’ residential SA2 (dummy coded, Yes/No), and spatial accessibility to PDS via driving and public transit travel modes. Data on variables indicating refugee socioeconomic disadvantage were obtained from the Australian Census and Migrants Integrated Dataset, 2016 . These included proportion of total resident SA2 population who are refugees and proportion of total refugees in each SA2 who – moved to Australia during the last 5 years (as of 2016), did not complete Year 12, are not proficient in English, are above 15 years and unemployed, have an annual income <$25,999, need assistance with core activities, and live in households without a motor vehicle . Principal Components Analysis was conducted to reduce the dimensionality of these variables and obtain a unified measure of SA2-level socioeconomic disadvantage for the refugee population. Based on the Kaiser’s criterion, five components (capturing 80.02% total variance) with eigenvalues greater than one were combined using their respective eigenvalues as weightings (see Additional file 1 for details). The resulting scores were classified into tertiles. List of RHP sites were gathered from the Victorian Department of Health and were assigned to their respective SA2s based on their suburb and postcode in the postal address . Spatial accessibility to PDS was calculated using the enhanced two-step floating catchment area method individually via road network (for driving mode including car or other motor vehicle) and public transit network (for various public transit modes including bus, tram, metro, or train), as detailed elsewhere . The spatial accessibility index scores obtained from these calculations represent the ratio of full-time equivalent dental professionals to the population eligible for PDS within each SA2, weighted by the travel time between their respective locations via driving or public transit mode; these scores were used as continuous variables.
Latent profile analysis (LPA) was used to identify distinct subgroups that characterise the utilisation of PDS among the Victorian refugee population for the period 2016-17 to 2019-20. LPA is a person-centered model-based approach to identifying underlying subgroups (called latent profiles) based on an unobservable attribute (called latent variable) by assessing multiple dimensions indicated by measured variables (called indicator variables) pertaining to this attribute . The indicator variables can have a continuous, count, or a combination of these distributions. Individuals are probabilistically assigned to the latent profiles based on two model parameters estimated on a maximum-likelihood basis : (a) profile membership probabilities; (b) means and variances of indicator variables, conditional on profile membership. Profiles of individuals sharing similar patterns of the means and variances of each indicator variable are identified and grouped. This enables the distinctness of each identified profiles to be assessed and qualitatively described. Indicator variables of refugees’ PDS utilisation used in the LPA included attributes of dental services received and clinic sites of service provision, during the four-year period. Number of CoCs received per client in each of the three types of care (i.e., general, emergency, and denture) and by each of two referral types (i.e., self-referral or referred by dental professionals, health care professionals, refugee and community support services, and other support services), as well as number of unique visits in each of the eleven service areas within a CoC were indicators of dental service characteristics. For clinic characteristics, number of CoCs received based on urbanicity of the clinic site (i.e., metropolitan, regional, or rural site) , whether the site was within the SA2 of clients’ residence (i.e., within or outside SA2 of residence), and whether the clinic was co-located with CHCs delivering RHP (i.e., co-located or not co-located) were used as indicators. All indicators were counts; hence, a Poisson model was used in the LPA . As there were differences in the number of years in which each client attended PDS, during the study period, it was included as an offset in the Poisson model. Doing so adjusted for any difference in utilisation among the clients by modelling counts as rates (i.e., number of CoCs per year or number of visits per year). Therefore, based on the indicators included in the LPA, the final profile assignment of individuals was based on the combined patterns in the conditional mean rates of utilisation in each of the indicator variables. LPA model building process was done iteratively. First, a model with one latent profile was fitted. Next, the number of profiles were augmented in a step wise manner until the models no longer converged . From thus obtained models, the best fitted model with the optimal number of profiles was selected based on the following criteria : (a) relative fit statistics – Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC); (b) classification diagnostics – entropy values closer to 1, average posterior probability of profile membership > 0.7, odds of correct classification based on posterior probabilities > 5 for each profile group; (c) substantive model interpretability and parsimony. Finally, individuals were assigned to a profile based on their maximum posterior probabilities. Multilevel multinomial logistic regression was performed to examine the role of individual and contextual level variables pertaining to refugees (independent variables) in predicting their profile membership (dependent variable). As clients in the DHPDS data are clustered at the neighbourhood (SA2) level, a two-level random-intercept model was fitted with individuals (level 1) nested within contexts (level 2) . Associations between the independent variables and the profile membership were tested in bivariate analysis, and only variables with significant association (p < .05) were included in the multilevel multivariate analysis. The first model was a null model, which only included the dependent variable with its variance split in the two levels of analysis. Subsequently, individual and contextual level variables were included in blocks in the second and third models, respectively. The association between profile membership and each independent variable was adjusted for in terms of all the other variables included in the models. The amount of contextual-level variation in the patterns of PDS utilisation was determined by estimating intra-class correlation coefficient (ICC) using the null model and the extent to which the independent variables were able to explain this variation was determined by calculating proportional change in variance for Models 2 and 3, in reference to the null model . Coefficients from the regression models were exponentiated to obtain conditional odds ratio (COR) with 95% confidence intervals (CI) . Characteristics of the patients and the identified latent profiles were summarised using frequency and percentages. Chi-squared test was used to examine the differences in distributions of individual and contextual characteristics across the identified profiles. All models were estimated via a Generalised Structural Equation Modelling in Stata 17, using gsem command with poisson log link (LPA models) and multinomial logit link (multilevel models) functions . Less than 1.5% of data values were missing on explanatory variables (see Additional file 1, Table S3), which may be considered insignificant relative to the general standard of 5% . So, listwise deletion approach was used to handle missing data. A p value of < 0.05 was considered statistically significant.
Descriptive statistics A total of 25,542 refugee clients attended PDS during the study period receiving a total of 47,919 CoCs, including 31,469 general, 14,887 emergency, and 1563 denture CoCs. In total, clients had 246,119 unique visits across the eleven major service areas. Most of the CoCs were self-referred (88.37%) and were received at a metropolitan clinic (90.62%) located outside clients’ residential SA2 (67.29%). Mean age at the time of first visit was 29.46 years (± 18.15), and females comprised 51.83% of the clientele. Majority were born in countries in the Middle East and North Africa region (38.98%), preferred to communicate in Arabic (28.02%), and resided in the metropolitan region (90.96%). LPA model selection Model selection was based on the statistical fit and the substantive capacity of the model to distinguish between individual PDS use patterns. The model fit was shown to improve with each additional profile; AIC and BIC values continuously decreased as the number of profiles estimated increased (see Additional file 2, Table S4). However, there was only a small improvement in the BIC values (as indicated by ΔBIC) after the six-profile model and signs of model overfitting (with smallest profile comprising < 5% total sample) . Therefore, the six-profile model was selected as the optimal model. The model also adequately differentiates the profiles, as indicated by the entropy value (0.86), high average posterior probability (> 0.8 for each group) and odds of correct classification (> 10 for each group) (see Additional file 2, Table S5). Profiles of refugee public dental services utilisation Figure shows the pattern of mean rates of utilisation in each of the indicators across the six identified profiles (see Additional file 2, Table S6). Based on this, descriptors were used to characterise the PDS utilisation pattern of refugees in each group. The profiles were significantly different with respect to the characteristics of refugees assigned to them, as demonstrated in Table . Profile 1, ‘General – Restorative’ (n = 9732, 38.10%) (herein restorative users), was the largest in size and characterised by higher utilisation rates of general CoC for restorative services. There was also higher uptake of prophylactic and preventive services in this group compared to other groups (except Profile 3 or 5). Clients had the lowest rate of utilisation at clinics co-located with CHCs delivering RHP, located in the regional and rural areas, outside their SA2 of residence, among the profile groups. Individuals were predominantly females, between the ages of 16 and 30, and lived in metropolitan areas most accessible to PDS via driving and least accessible via public transit modes. Profile 2, ‘Denture – Complete and partial denture’ (n = 1619, 6.34%) (herein denture users), was the smallest and was distinctly characterised by having the highest utilisation rates of denture CoC among all the groups. Not surprisingly, the visits for complete and partial denture services were the highest and extraction services were relatively higher than other groups, except Profile 6. Services were predominantly received at co-located clinic sites. Individuals were relatively older (> 45 years), with higher proportions of males, pensioner concession card holders, and from countries in the Europe, Central Asia, Americas, and Caribbean region, than other groups. Profile 3, ‘Emergency – Operative’ (n = 2189, 8.57%) (herein operative users), had the highest mean utilisation rates of emergency CoC for endodontic and restorative services. In comparison to other groups, the services received were by self-referral and at metropolitan clinic sites, outside clients’ SA2 of residence. Individuals comprised a higher proportion of those born in the Middle East and North Africa region, spoke Arabic, Persian, or Dari, and lived in areas with least socioeconomic disadvantage for refugees. Within the group, most individuals were between the ages 16 and 45. Profile 4, ‘General – Orthodontic’ (n = 3671, 14.37%) (herein orthodontic users), had the highest mean rate of general CoCs than other groups, comprising visits predominantly for orthodontic services. Individuals were younger (0–15 years) compared to other groups. The majority had a health care card, spoke Karen, were born in the East Asia and Pacific countries, lived in areas with lowest accessibility to PDS via driving mode, than other groups. Profile 5, ‘General – Preventive’ (n = 2324, 9.10%) (herein preventive users), predominantly received general CoCs with highest rates of prophylactic and preventive, as well as periodontic services. This group had lowest rates of service utilisation at metropolitan clinic sites not co-located with RHP delivering CHCs. Unlike other groups, majority were between 31 and 45 years, born in countries in the Sub-Saharan Africa region, preferred to communicate in English and Other languages (which included all languages other than the five predominant ones), and lived in the regional areas. Profile 6, ‘Emergency – Extractions’ (n = 6007, 23.52%) (herein extractions users), was the second largest group and received higher emergency CoCs primarily for extractions, and at clinics located within the clients’ SA2 of residence. Comparing across the profiles, a higher proportion of individuals in this group were from South Asian countries, did not request interpreter services, and lived in rural areas most accessible via public transit and where RHP was delivered. Predictors of profile membership As profile membership (dependent variable) was a multinomial categorical variable, Profile 5 (i.e., preventive users) was used as the reference category in all regression analyses. This was to enable comparison of preventive users and those with a more treatment-oriented pattern of PDS utilisation, i.e., the estimated CORs represent the odds of using a particular service instead of preventive services . Bivariate analysis showed significant associations between each of the individual and contextual level variables of refugees and their profile membership. (Additional file 2, Table S7) Therefore, all independent variables were included in the subsequent multilevel multivariate analysis. The fitted models along with the estimated effects and their 95% CIs are presented in Table . Model 1 is nested within Model 2, which is in turn nested within Model 3. Comparison of the models showed a significant improvement in the fit, as indicated by the reduced AIC and BIC values. In Model 1, only the profile membership of refugees was included, with no predictors. The variance of the random intercept (1.15, p < .001) indicated a statistically significant difference between SA2s in the likelihood of refugees belonging to a particular PDS utilisation profile. ICC estimated from the null model (ICC = 0.259) indicated that 25.9% of the total variation in this likelihood is attributed to the differences between the refugees’ SA2 of residence (Table ); in other words, 74.1% of the variability was accounted for by the individual differences between refugees and other unknown factors. Model 2 analysed the effect of all individual-level variables. When controlled for the effects of individual variables, the estimated fixed effects continued to remain statistically significant similar to the null model, with a substantial increase in the relative likelihood for restorative and orthodontic users group (COR, 11.16 and 29.82, respectively). Age was a significant predictor of profile membership; the effect of age was positive for denture users and negative for all other profiles. Females relative to males, had a higher odds of belonging to any of the utilisation profiles (COR range, 1.14–1.26), except denture users group. Refugees born in any region were more likely to use extractions and operative services than preventive services (COR range, 1.76–4.46), compared to those born in East Asia and Pacific. Having an eligibility card was significantly associated with a higher odds of orthodontic and denture services use (COR range, 1.44–2.74), than preventive services. The relative likelihood of belonging to the any utilisation profile was higher for Persian and Dari speakers (COR range, 1.33–2.19), compared to those who prefer English. Orthodontic services users were more likely to request interpreter services (COR range, 1.19–1.27), compared to the reference profile group. Together the individual-level predictors explained 23.70% variation in PDS utilisation patterns between SA2s (Table ). In the final model, Model 3, contextual-level variables were added to analyse the combined effect of individual and contextual level variables on profile membership. Most of the individual-level variables from the previous model continued to have a significant effect on profile membership, with only a small change in their effect size (max ΔCOR = 0.88). The relative likelihood of orthodontic services use among the refugees from Sub-Saharan countries (COR, 0.68), operative and extraction services use among those who do not request for interpreter services (COR, 0.83 and 1.23, respectively), compared to their corresponding reference categories, gained significance. Urbanicity of residence was a significant predictor of profile membership. For refugees living in rural areas, the relative likelihood of belonging to extractions group was higher (COR, 1.52; 95% CI 1.21–3.24) and lower for the remaining groups (COR range, 0.02 and 0.34), than those living in metropolitan areas. The odds of using restorative, operative, and extraction services, instead of preventive services, decreased for refugees living in regional areas compared to those in metropolitan areas (COR range, 0.08–0.42). Those in the most socioeconomically disadvantaged tertile had a significantly higher relative likelihood of belonging to extractions group (COR, 2.72; 95% CI 1.54–4.18), than those in the lowest tertile. The effect was insignificant for other groups. Increase in spatial accessibility index scores via driving and public transit modes increased the odds of using extraction services by factors of 1.57 and 1.48, respectively. The estimated proportional change in variance for Model 3 indicated that 40.22% of the variation in the patterns of PDS utilisation between different SA2s was explained by the individual and contextual level predictors (Table ).
A total of 25,542 refugee clients attended PDS during the study period receiving a total of 47,919 CoCs, including 31,469 general, 14,887 emergency, and 1563 denture CoCs. In total, clients had 246,119 unique visits across the eleven major service areas. Most of the CoCs were self-referred (88.37%) and were received at a metropolitan clinic (90.62%) located outside clients’ residential SA2 (67.29%). Mean age at the time of first visit was 29.46 years (± 18.15), and females comprised 51.83% of the clientele. Majority were born in countries in the Middle East and North Africa region (38.98%), preferred to communicate in Arabic (28.02%), and resided in the metropolitan region (90.96%).
Model selection was based on the statistical fit and the substantive capacity of the model to distinguish between individual PDS use patterns. The model fit was shown to improve with each additional profile; AIC and BIC values continuously decreased as the number of profiles estimated increased (see Additional file 2, Table S4). However, there was only a small improvement in the BIC values (as indicated by ΔBIC) after the six-profile model and signs of model overfitting (with smallest profile comprising < 5% total sample) . Therefore, the six-profile model was selected as the optimal model. The model also adequately differentiates the profiles, as indicated by the entropy value (0.86), high average posterior probability (> 0.8 for each group) and odds of correct classification (> 10 for each group) (see Additional file 2, Table S5).
Figure shows the pattern of mean rates of utilisation in each of the indicators across the six identified profiles (see Additional file 2, Table S6). Based on this, descriptors were used to characterise the PDS utilisation pattern of refugees in each group. The profiles were significantly different with respect to the characteristics of refugees assigned to them, as demonstrated in Table . Profile 1, ‘General – Restorative’ (n = 9732, 38.10%) (herein restorative users), was the largest in size and characterised by higher utilisation rates of general CoC for restorative services. There was also higher uptake of prophylactic and preventive services in this group compared to other groups (except Profile 3 or 5). Clients had the lowest rate of utilisation at clinics co-located with CHCs delivering RHP, located in the regional and rural areas, outside their SA2 of residence, among the profile groups. Individuals were predominantly females, between the ages of 16 and 30, and lived in metropolitan areas most accessible to PDS via driving and least accessible via public transit modes. Profile 2, ‘Denture – Complete and partial denture’ (n = 1619, 6.34%) (herein denture users), was the smallest and was distinctly characterised by having the highest utilisation rates of denture CoC among all the groups. Not surprisingly, the visits for complete and partial denture services were the highest and extraction services were relatively higher than other groups, except Profile 6. Services were predominantly received at co-located clinic sites. Individuals were relatively older (> 45 years), with higher proportions of males, pensioner concession card holders, and from countries in the Europe, Central Asia, Americas, and Caribbean region, than other groups. Profile 3, ‘Emergency – Operative’ (n = 2189, 8.57%) (herein operative users), had the highest mean utilisation rates of emergency CoC for endodontic and restorative services. In comparison to other groups, the services received were by self-referral and at metropolitan clinic sites, outside clients’ SA2 of residence. Individuals comprised a higher proportion of those born in the Middle East and North Africa region, spoke Arabic, Persian, or Dari, and lived in areas with least socioeconomic disadvantage for refugees. Within the group, most individuals were between the ages 16 and 45. Profile 4, ‘General – Orthodontic’ (n = 3671, 14.37%) (herein orthodontic users), had the highest mean rate of general CoCs than other groups, comprising visits predominantly for orthodontic services. Individuals were younger (0–15 years) compared to other groups. The majority had a health care card, spoke Karen, were born in the East Asia and Pacific countries, lived in areas with lowest accessibility to PDS via driving mode, than other groups. Profile 5, ‘General – Preventive’ (n = 2324, 9.10%) (herein preventive users), predominantly received general CoCs with highest rates of prophylactic and preventive, as well as periodontic services. This group had lowest rates of service utilisation at metropolitan clinic sites not co-located with RHP delivering CHCs. Unlike other groups, majority were between 31 and 45 years, born in countries in the Sub-Saharan Africa region, preferred to communicate in English and Other languages (which included all languages other than the five predominant ones), and lived in the regional areas. Profile 6, ‘Emergency – Extractions’ (n = 6007, 23.52%) (herein extractions users), was the second largest group and received higher emergency CoCs primarily for extractions, and at clinics located within the clients’ SA2 of residence. Comparing across the profiles, a higher proportion of individuals in this group were from South Asian countries, did not request interpreter services, and lived in rural areas most accessible via public transit and where RHP was delivered.
As profile membership (dependent variable) was a multinomial categorical variable, Profile 5 (i.e., preventive users) was used as the reference category in all regression analyses. This was to enable comparison of preventive users and those with a more treatment-oriented pattern of PDS utilisation, i.e., the estimated CORs represent the odds of using a particular service instead of preventive services . Bivariate analysis showed significant associations between each of the individual and contextual level variables of refugees and their profile membership. (Additional file 2, Table S7) Therefore, all independent variables were included in the subsequent multilevel multivariate analysis. The fitted models along with the estimated effects and their 95% CIs are presented in Table . Model 1 is nested within Model 2, which is in turn nested within Model 3. Comparison of the models showed a significant improvement in the fit, as indicated by the reduced AIC and BIC values. In Model 1, only the profile membership of refugees was included, with no predictors. The variance of the random intercept (1.15, p < .001) indicated a statistically significant difference between SA2s in the likelihood of refugees belonging to a particular PDS utilisation profile. ICC estimated from the null model (ICC = 0.259) indicated that 25.9% of the total variation in this likelihood is attributed to the differences between the refugees’ SA2 of residence (Table ); in other words, 74.1% of the variability was accounted for by the individual differences between refugees and other unknown factors. Model 2 analysed the effect of all individual-level variables. When controlled for the effects of individual variables, the estimated fixed effects continued to remain statistically significant similar to the null model, with a substantial increase in the relative likelihood for restorative and orthodontic users group (COR, 11.16 and 29.82, respectively). Age was a significant predictor of profile membership; the effect of age was positive for denture users and negative for all other profiles. Females relative to males, had a higher odds of belonging to any of the utilisation profiles (COR range, 1.14–1.26), except denture users group. Refugees born in any region were more likely to use extractions and operative services than preventive services (COR range, 1.76–4.46), compared to those born in East Asia and Pacific. Having an eligibility card was significantly associated with a higher odds of orthodontic and denture services use (COR range, 1.44–2.74), than preventive services. The relative likelihood of belonging to the any utilisation profile was higher for Persian and Dari speakers (COR range, 1.33–2.19), compared to those who prefer English. Orthodontic services users were more likely to request interpreter services (COR range, 1.19–1.27), compared to the reference profile group. Together the individual-level predictors explained 23.70% variation in PDS utilisation patterns between SA2s (Table ). In the final model, Model 3, contextual-level variables were added to analyse the combined effect of individual and contextual level variables on profile membership. Most of the individual-level variables from the previous model continued to have a significant effect on profile membership, with only a small change in their effect size (max ΔCOR = 0.88). The relative likelihood of orthodontic services use among the refugees from Sub-Saharan countries (COR, 0.68), operative and extraction services use among those who do not request for interpreter services (COR, 0.83 and 1.23, respectively), compared to their corresponding reference categories, gained significance. Urbanicity of residence was a significant predictor of profile membership. For refugees living in rural areas, the relative likelihood of belonging to extractions group was higher (COR, 1.52; 95% CI 1.21–3.24) and lower for the remaining groups (COR range, 0.02 and 0.34), than those living in metropolitan areas. The odds of using restorative, operative, and extraction services, instead of preventive services, decreased for refugees living in regional areas compared to those in metropolitan areas (COR range, 0.08–0.42). Those in the most socioeconomically disadvantaged tertile had a significantly higher relative likelihood of belonging to extractions group (COR, 2.72; 95% CI 1.54–4.18), than those in the lowest tertile. The effect was insignificant for other groups. Increase in spatial accessibility index scores via driving and public transit modes increased the odds of using extraction services by factors of 1.57 and 1.48, respectively. The estimated proportional change in variance for Model 3 indicated that 40.22% of the variation in the patterns of PDS utilisation between different SA2s was explained by the individual and contextual level predictors (Table ).
The study investigated the patterns and predictive factors of PDS use among refugees in Victoria, using existing administrative data over a four-year period. There was a significant heterogeneity within the study population in terms of the combined patterns and rates of utilisation of different types of CoCs and eleven major service areas as well as the location attributes of clinic sites where they availed PDS. Six distinct profiles of PDS use were identified, described, and subsequently investigated. Together, the findings of this study further the understanding of access and utilisation of PDS among Victorian refugees. This study is the first to employ LPA to develop profiles of refugee population based on their dental service utilisation pattern. According to the LPA model, the majority of refugees (about 52%) who attended PDS during the study period had a higher probability of using restorative or orthodontic services as part of general CoC. Another 32% predominantly used emergency CoC for extractions or endodontic procedures. Notably, only a very small proportion of refugees (about 9%) used prophylactic and preventive services. These identified patterns were consistent with previous studies which found a low use of preventive services , high use of oral surgery and endodontic services , and a high proportion of those seeking emergency dental care among refugee populations. Refugees tend to use dental services only when in severe pain or when self-treatments do not work . Consequently, they have a problem-oriented pattern of dental attendance, wherein services are sought infrequently and primarily for treating dental problems . Evidence suggests that a visiting pattern comprising regular dental check-ups and preventive (or interceptive) care is associated with decrease in the use of emergency dental services and better oral health outcomes . Regrettably, this so-called ‘favourable’ utilisation pattern was observed only among a small proportion of refugees in this study. Examination of refugee characteristics within and across the groups showed a clear distinction between the profiles. Overall, females and young and middle-aged adults (16–45 years) had the most utilisation among the identified profiles, except denture users group (Table ). This compared favourably with dental visits among the general populations in Australia . Whereas these findings are encouraging, considering the higher burden of dental disease among refugee males and children compared to those in general population , a higher uptake among these groups would have been expected. In refugee families, the dental health-related attitude of parents is critical in determining their children’s utilisation pattern, as they are the decision-makers for their children’s dental care needs . So, the lower use among younger age groups may primarily be attributed to their parents. Interestingly though, there was a very high uptake of orthodontic services among children and adolescents (0–15 years), relative to other services, including preventive services. This finding is also substantiated in a study among Australian refugees which reported that the most frequent oral health concern of refugee children or their parents was cosmetic related . Another reason could be that some refugees (with high orthodontic treatment need) can avail orthodontic treatment at no cost via PDS, as opposed to the very expensive private alternatives. Although use of orthodontic services for cosmetic reasons suggest a considerable improvement in oral health attitudes or service awareness among resettled refugees, the findings highlight the need for strategies to improve uptake of preventive services among the 0–15 years group. High utilisation among female refugees is contrary to the literature. It is generally believed that most refugee families have a male dominant structure in which female health-related decisions are made by males . While the study findings suggest otherwise, the reasons for this could be manifold including individual family circumstances, mix of cultural groups within the sample, and differences in the lengths of stay and levels of assimilation to the Australian culture among the study population. In addition to describing the profile characteristics, the study also determined the predictors of refugee PDS utilisation pattern. At the individual-level, the primary correlates were age and gender. Ethnicity of refugees, based on their region of birth, had a consistently positive association with PDS utilisation pattern, except for denture and orthodontic services use among Sub-Saharan refugees (Table ). However, there was a considerable difference in the likelihood across profiles among different ethnic groups. Burden of oral diseases may vary among refugees based on their ethnicity owing to the cultural or religious norms, dietary preferences, oral hygiene practices, oral health related attitude including access to dental care in their home country, and their ability to assimilate to the host country’s culture . To some extent, this might have had a decisive influence on their utilisation pattern. Considering the inclination of refugees to resettle in ethnic clusters , other factors that could explain the differences may be related to the cultural and social support available to each of these ethnic groups in their communities of settlement. The relationship between the remaining individual variables and utilisation patterns were mixed across the groups. The role of refugees’ context in predicting their PDS utilisation pattern was confirmed in the current study. About 26% of the variation in refugees’ PDS utilisation patterns was due to the differences in the characteristics of their place of residence (i.e., SA2). A clear gradient was observed between higher SA2-level socioeconomic disadvantage of refugees and increased likelihood of emergency extraction service use. The association between area-level socioeconomic disadvantage and dental service utilisation pattern among refugees reflects on the importance of contextual-level factors in determining PDS utilisation among refugees. This finding is new and an important one. Refugees living in the rural areas were 52% more likely to use emergency extraction services than their metropolitan and regional counterparts. This effect was evident even after adjusting for socioeconomic disadvantage, physical accessibility to community dental clinics via different travel modes and availability of dental professionals, which are considered primary barriers to access among rural residents in Australia . As such, this finding is particularly significant, as it points to factors associated with higher use of extraction services, beyond those considered in the analysis. Such factors may include, but are not limited to, oral health promotion activities , social, cultural, or religious networks disseminating information on dental services , and presence of community organisations supporting rural refugees in accessing dental care. With refugee resettlement shifting to rural areas , it is critical to reorient the public dental system to address these growing inequalities among rural refugee populations. The study findings add new knowledge on the association between spatial accessibility to dental services and the pattern of utilisation. The latest Australian National Oral Health Plan emphasises the importance of understanding this relationship in order to improve service delivery for vulnerable population groups (including refugees) . The current study found a significant association between potential spatial accessibility to PDS via driving and public transit modes of travel and the utilisation patterns among refugees. Overall, refugees in SA2s least accessible via any travel mode used PDS less than those in most accessible SA2s, which reflects on the impact of potential accessibility on realised service utilisation. Bivariate associations revealed significantly higher likelihood in the use of extraction services than preventive services, with increase in accessibility via any travel mode. When adjusted for the effects of other variables in the multivariate analysis, these associations remained significant. Clearly, this finding implies that irrespective of the level of opportunity to access services, refugees continue to incline toward attending PDS in a problem-oriented manner. Together, the individual and contextual level factors explained about 40% of the total difference in the utilisation pattern across SA2s; meaning that the remaining 60% variation is due to other factors not included in this study. One of the most important individual-level factors is subjective or objective oral health need, which was found to significantly impact refugee dental service utilisation . Among others, oral health literacy, length of stay in the host country and cultural assimilation were also shown to be positively associated with the utilisation . As well, factors related to the dental health organisation such as cultural competence and responsiveness of dental and support staff, and appropriateness of care provided have been noted to be potential in determining refugees’ utilisation behaviour . Future research should examine the role of these factors on their utilisation pattern. Strengths and limitations This study was the first to have comprehensively evaluated utilisation of PDS among a large sample of refugees in Victoria using administrative data over multiple years. Homogeneity in refugees’ patterns of PDS utilisation was demonstrated using LPA based on multiple indicators including the attributes of dental care and clinic of care reception. This enabled capturing meaningful variations in the complex interactions among different dimensions of PDS utilisation, rather than relying on any one dimension (e.g., either CoC or service type). Moreover, refugees were classified into profiles based on model-based cut-off thresholds derived from within the data, minimising any classification errors that may arise from using arbitrary cut-offs for grouping (e.g., above or below a mean value) . Furthermore, the role of individual and contextual level predictors of PDS utilisation pattern was analysed using a multilevel design. There are some limitations, primarily arising from the clinical records data. Refugee clients were identified within the DHPDS based on how these individuals were identified and recorded by the public dental clinic staff in the Titanium® system. Although there are a flexible set of criteria available to them to identify an individual as a refugee , there is no one agreed upon definition. As such, there may be inconsistencies across clinics. The variables included in the LPA and multilevel analysis were restricted by the availability and completeness of clinical records data. This precluded the evaluation of some important factors known to impact dental service use. For example, there was a large amount missing data (missing for about 69% clients) for variables indicating the oral health status, such as decayed, missing and filled teeth. The study results must be interpreted within the context of some methodological limitations. The DHPDS does not capture information on those who do not utilise public dental services. As such, factors influencing non-utilisation of public dental services were not evaluated. Profiles developed through LPA are not exclusive , i.e., there might be overlap in the services used by refugees in different groups. The assignment of individuals was based on their highest probability of belonging to a particular utilisation pattern which may have resulted in certain amount of misclassification. Due to the focus of the study and cross-sectional design, some dependency structures in the DHPDS data (for e.g., clustering of clients based on their date/year of treatment visit and clinic site of visit) were disregarded which may have resulted in some bias in the model estimates. Finally, as with any study based on administrative data analysis, the findings cannot be generalised outside the study population, i.e., refugees attending PDS in Victoria.
This study was the first to have comprehensively evaluated utilisation of PDS among a large sample of refugees in Victoria using administrative data over multiple years. Homogeneity in refugees’ patterns of PDS utilisation was demonstrated using LPA based on multiple indicators including the attributes of dental care and clinic of care reception. This enabled capturing meaningful variations in the complex interactions among different dimensions of PDS utilisation, rather than relying on any one dimension (e.g., either CoC or service type). Moreover, refugees were classified into profiles based on model-based cut-off thresholds derived from within the data, minimising any classification errors that may arise from using arbitrary cut-offs for grouping (e.g., above or below a mean value) . Furthermore, the role of individual and contextual level predictors of PDS utilisation pattern was analysed using a multilevel design. There are some limitations, primarily arising from the clinical records data. Refugee clients were identified within the DHPDS based on how these individuals were identified and recorded by the public dental clinic staff in the Titanium® system. Although there are a flexible set of criteria available to them to identify an individual as a refugee , there is no one agreed upon definition. As such, there may be inconsistencies across clinics. The variables included in the LPA and multilevel analysis were restricted by the availability and completeness of clinical records data. This precluded the evaluation of some important factors known to impact dental service use. For example, there was a large amount missing data (missing for about 69% clients) for variables indicating the oral health status, such as decayed, missing and filled teeth. The study results must be interpreted within the context of some methodological limitations. The DHPDS does not capture information on those who do not utilise public dental services. As such, factors influencing non-utilisation of public dental services were not evaluated. Profiles developed through LPA are not exclusive , i.e., there might be overlap in the services used by refugees in different groups. The assignment of individuals was based on their highest probability of belonging to a particular utilisation pattern which may have resulted in certain amount of misclassification. Due to the focus of the study and cross-sectional design, some dependency structures in the DHPDS data (for e.g., clustering of clients based on their date/year of treatment visit and clinic site of visit) were disregarded which may have resulted in some bias in the model estimates. Finally, as with any study based on administrative data analysis, the findings cannot be generalised outside the study population, i.e., refugees attending PDS in Victoria.
This study represents a significant step towards the development of an evidence-based knowledge around PDS utilisation among the refugee population in Victoria. Profiles of refugees with distinct patterns of PDS utilisation were developed. The findings demonstrated that the characteristics of refugees’ place of residence including urbanicity, socioeconomic disadvantage, delivery of RHP, and potential spatial accessibility to PDS determined their utilisation pattern. Where opportunities to access PDS were present, refugees were more likely to use extraction services than preventive services. Overall, the findings reiterate the critical need for targeted strategies to promote the importance of routine dental visits, oral disease prevention, and timely intervention among refugee groups.
Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 2
|
Measurement of person-centred consultation skills among healthcare practitioners: a systematic review of reviews of validation studies | 53dac6a0-1bc0-4e20-9d91-ced53716e4b2 | 10074817 | Patient-Centered Care[mh] | Person-centred care (also termed patient-centred care ) has been widely acknowledged as an essential element of high-quality health service provision . The concept of person-centredness has been utilized for roughly half a century and has been applied at different levels, from national healthcare policy to skills as specific as non-verbal communication behaviours . Many different perspectives on, and definitions of, person-centredness exist, thus making it a somewhat contested concept to operationalise . Arguably, these are variations in emphasis within a core theme, though they do have implications for valid measurement. Consultations are a key component in health care provision which offer an opportunity for patients to discuss issues with practitioners. Practitioners often have multiple tasks within consultations, including eliciting information to aid assessment, and information-giving. Individual practitioners vary in consultation skills and commitment to make the conversation person-centred in practice . In the past two decades person-centred communication skills acquisition has received much greater attention in training programmes . To evaluate the efficacy of training programmes designed to enhance person-centred skills, validated instruments that objectively measure these skills and their use in practice are needed. Systematic reviews of validation studies of instruments measuring person-centeredness were known to exist prior to undertaking this study, however, it was clear that this literature was diverse, and that such reviews may have different purposes, aims, and inclusion criteria. Reviews have been aimed at identifying and/or appraising instruments for specific conditions (e.g., cancer, ), health care settings (e.g., neonatal intensive care units, ), or professions (e.g., psychiatrists, ). In addition, across existing reviews different conceptualisations of person-centredness frame research questions and selection criteria in distinct ways (e.g., see [ – ]). Consequently, there may be little overlap in the primary studies included in available reviews, and no one review summarises and evaluates the literature as a whole. For these reasons we aimed to provide a high-level synthesis of this complex literature by undertaking a systematic review of reviews. This was intended to provide an overview of how existing systematic reviews are designed and report on validation studies, and to incorporate details of the included instruments. This study thus brings together what is known about available instruments that may be considered for use in training and assessment of person-centred consultation skills among healthcare practitioners, for researchers and research users. This review of reviews was thus not undertaken to identify a particular instrument for a particular purpose, but rather to survey the level of development of, and the strength of the evidence available in, this field of study. Reflecting these aims, the objectives of this review of reviews were to: 1) undertake a critical appraisal of systematic reviews reporting validation studies of instruments aiming to measure person-centred consultation skills among healthcare practitioners, and 2) identify and summarise the range of validated instruments available for measuring person-centred consultation skills in practitioners, including material on the strength of the validation evidence for each instrument.
This review followed the process outlined in this section, which followed the development of a study protocol prior to the conduct of the review. We did not prospectively register or otherwise publish the protocol. Search strategy Systematic searches were conducted in the electronic databases MEDLINE, EMBASE, PsycINFO, and CINAHL. The search strategy combined different search terms for three key search components: ‘person- or patient centredness’ (Block 1), ‘assessment instrument’ (Block 2), and ‘systematic or scoping review’ (Block 3). For Block 1 (the search component ‘person- or patient centredness’) we used an iterative approach. A preliminary search of EMBASE, MEDLINE, and PsychInfo (all in Ovid) was undertaken using the keywords: (person-cent* or patient-cent* or personcent* or patientcent*) and ‘review’ in the title; and ‘measurement or tool or scale or instrument’; from 2010. Full text papers identified ( n = 24) were searched for words used to describe ‘person- or patient centredness’. The resulting search terms were discussed and selected to reflect the scope of the study. The final search included the following terms: person-cent* or patient-cent* or personcent* or patientcent* or person-orient* or person-focus* or person-participation or person-empowerment or person-involvement or patient-orient* or patient-focus* or patient-participation or patient-empowerment or patient-involvement or "person orient*" or "person focus*" or "person participation" or "person empowerment" or "person involvement" or "patient orient*" or "patient focus*" or "patient participation" or "patient empowerment" or "patient involvement"; or (clinician-patient or physician–patient or professional-patient or provider-patient or practitioner-patient or pharmacist-patient or doctor-patient or nurse-patient) adjacent to (communication* or consultation* or practice* or relation* or interaction* or rapport). For Block 2 (the search component ‘assessment instrument’) we used the existing COSMIN filters proposed by Terwee et al. . The COSMIN (COnsensus-based Standards for the selection of health Measurement Instruments) project has developed highly sensitive search filters for finding studies on measurement properties . The search filter was adapted to each database. For Block 3, the search terms (systematic* or scoping) adjacent to review* were used. The search did not include restrictions pertaining to date of publication, and the language was restricted to English. The database search was conducted in September 2020. See appendix 1 for the details of all searches run in all databases. Study selection One author (JG) screened titles and abstracts against preliminary selection criteria, using Rayyan software for systematic reviews . Ideally all parts of the process of undertaking a review are duplicated to in order to avoid errors. Here we relied on one author for screening, with the rationale was that we expected systematic reviews to be readily identifiable in the title and abstract, making screening more straightforward, for example, than in conducting a systematic review of primary studies, which may be described in more heterogeneous ways. Another author (AD) screened 5% independently. The authors met weekly to resolve any problems or questions during the process and no contentious issues were identified in screening. Full text articles of potentially eligible papers were retrieved and assessed for inclusion against the criteria below. Two authors (AD & JM) reviewed all full text papers independently in order to select studies for inclusion. One disagreement was resolved through discussion with a third author (DS) and reasons for exclusion were noted. Inclusion criteria were: a peer-reviewed journal report used systematic review methods to identify primary studies for inclusion (including both a search strategy and explicit selection criteria) stated aims and objectives specifying the measurement of ‘person centredness’ or ‘patient centredness’ or a related construct as defined by search Block 1. concerned assessment of individual practitioner consultation skills or behaviour (i.e., not policy) included only validation studies of instruments reported any measurement properties of the included instruments Reviews of instruments developed for any practitioner group, patient population, or health care setting were included. Studies were excluded unless they met all inclusion criteria. After the full text eligibility check, a backwards search of the references of the included reviews, as well as a forward reference search using Google Scholar was performed. This was last updated in January 2022 and no further reviews were identified. A PRISMA flowchart shows the results of the identification, screening, and eligibility assessment process (Fig. ). Data extraction One author (AD) performed data extraction from the included reviews using a standardised form created in Excel developed by all co-authors in a preliminary phase. A second author (DS) subsequently checked all the extracted information in the form, and screened the paper for any missing information. At the review level, we extracted the stated aims and objectives, definition or conceptualisation of person-centredness used, numbers, names and types of instruments, research questions, dates, databases, and languages included in search strategies, selection criteria regarding health care populations, health care settings, raters of the instruments, other selection criteria, details of the assessment of methodological quality and psychometric properties, and numbers of validation studies. At the validation study level, we extracted the country of origin, the type of validation study, and whether the developers of the instrument validated their own instrument. At the instrument level we extracted who developed the instrument, in what year, in which country and in what language the instrument, how many subscales and items the instruments consisted of, and the response formats used. Other information on validation studies and instruments was not reported consistently enough to be extracted. Quality assessment Two authors (AD & DS) independently assessed the quality of the included reviews using the Joanna Briggs Institute Critical Appraisal Checklist for Systematic Reviews and Research Syntheses checklist . Each of the 11 criteria was given a rating of ‘yes’ (definitely done), ‘no’ (definitely not done), ‘unclear’ (unclear if completed) or ‘not applicable’. Discrepancies in the ratings of the methodological reviews were be resolved by consensus.
Systematic searches were conducted in the electronic databases MEDLINE, EMBASE, PsycINFO, and CINAHL. The search strategy combined different search terms for three key search components: ‘person- or patient centredness’ (Block 1), ‘assessment instrument’ (Block 2), and ‘systematic or scoping review’ (Block 3). For Block 1 (the search component ‘person- or patient centredness’) we used an iterative approach. A preliminary search of EMBASE, MEDLINE, and PsychInfo (all in Ovid) was undertaken using the keywords: (person-cent* or patient-cent* or personcent* or patientcent*) and ‘review’ in the title; and ‘measurement or tool or scale or instrument’; from 2010. Full text papers identified ( n = 24) were searched for words used to describe ‘person- or patient centredness’. The resulting search terms were discussed and selected to reflect the scope of the study. The final search included the following terms: person-cent* or patient-cent* or personcent* or patientcent* or person-orient* or person-focus* or person-participation or person-empowerment or person-involvement or patient-orient* or patient-focus* or patient-participation or patient-empowerment or patient-involvement or "person orient*" or "person focus*" or "person participation" or "person empowerment" or "person involvement" or "patient orient*" or "patient focus*" or "patient participation" or "patient empowerment" or "patient involvement"; or (clinician-patient or physician–patient or professional-patient or provider-patient or practitioner-patient or pharmacist-patient or doctor-patient or nurse-patient) adjacent to (communication* or consultation* or practice* or relation* or interaction* or rapport). For Block 2 (the search component ‘assessment instrument’) we used the existing COSMIN filters proposed by Terwee et al. . The COSMIN (COnsensus-based Standards for the selection of health Measurement Instruments) project has developed highly sensitive search filters for finding studies on measurement properties . The search filter was adapted to each database. For Block 3, the search terms (systematic* or scoping) adjacent to review* were used. The search did not include restrictions pertaining to date of publication, and the language was restricted to English. The database search was conducted in September 2020. See appendix 1 for the details of all searches run in all databases.
One author (JG) screened titles and abstracts against preliminary selection criteria, using Rayyan software for systematic reviews . Ideally all parts of the process of undertaking a review are duplicated to in order to avoid errors. Here we relied on one author for screening, with the rationale was that we expected systematic reviews to be readily identifiable in the title and abstract, making screening more straightforward, for example, than in conducting a systematic review of primary studies, which may be described in more heterogeneous ways. Another author (AD) screened 5% independently. The authors met weekly to resolve any problems or questions during the process and no contentious issues were identified in screening. Full text articles of potentially eligible papers were retrieved and assessed for inclusion against the criteria below. Two authors (AD & JM) reviewed all full text papers independently in order to select studies for inclusion. One disagreement was resolved through discussion with a third author (DS) and reasons for exclusion were noted. Inclusion criteria were: a peer-reviewed journal report used systematic review methods to identify primary studies for inclusion (including both a search strategy and explicit selection criteria) stated aims and objectives specifying the measurement of ‘person centredness’ or ‘patient centredness’ or a related construct as defined by search Block 1. concerned assessment of individual practitioner consultation skills or behaviour (i.e., not policy) included only validation studies of instruments reported any measurement properties of the included instruments Reviews of instruments developed for any practitioner group, patient population, or health care setting were included. Studies were excluded unless they met all inclusion criteria. After the full text eligibility check, a backwards search of the references of the included reviews, as well as a forward reference search using Google Scholar was performed. This was last updated in January 2022 and no further reviews were identified. A PRISMA flowchart shows the results of the identification, screening, and eligibility assessment process (Fig. ).
One author (AD) performed data extraction from the included reviews using a standardised form created in Excel developed by all co-authors in a preliminary phase. A second author (DS) subsequently checked all the extracted information in the form, and screened the paper for any missing information. At the review level, we extracted the stated aims and objectives, definition or conceptualisation of person-centredness used, numbers, names and types of instruments, research questions, dates, databases, and languages included in search strategies, selection criteria regarding health care populations, health care settings, raters of the instruments, other selection criteria, details of the assessment of methodological quality and psychometric properties, and numbers of validation studies. At the validation study level, we extracted the country of origin, the type of validation study, and whether the developers of the instrument validated their own instrument. At the instrument level we extracted who developed the instrument, in what year, in which country and in what language the instrument, how many subscales and items the instruments consisted of, and the response formats used. Other information on validation studies and instruments was not reported consistently enough to be extracted.
Two authors (AD & DS) independently assessed the quality of the included reviews using the Joanna Briggs Institute Critical Appraisal Checklist for Systematic Reviews and Research Syntheses checklist . Each of the 11 criteria was given a rating of ‘yes’ (definitely done), ‘no’ (definitely not done), ‘unclear’ (unclear if completed) or ‘not applicable’. Discrepancies in the ratings of the methodological reviews were be resolved by consensus.
Description of the reviews The search identified 2,215 unique articles with 21 papers selected for a full-text eligibility assessment (see Fig. ). Four studies were included. None of the reviews identified in further searching fulfilled our inclusion criteria. The four included reviews each had different aims and selection criteria, resulting in few primary studies and instruments being included in more than one review. Two reviews targeted different groups of practitioners; nurses for Köberich and Farin and physicians or medical students for Brouwers et al. ). Hudon et al. and Köberich and Farin included only patient rated instruments, while Ekman et al. included only direct observation tools (e.g., checklists or rating scales). In total, the four reviews included 71 validation studies (68 unique studies) of 42 different instruments. Conceptualisations of person-centredness Conceptualisations of person-centredness varied between the included studies. Two reviews used Stewart and colleagues model of interconnecting dimensions: 1) exploring both the disease and the illness experience; 2) understanding the whole person; 3) finding common ground between the physician and patient; 4) incorporating prevention and health promotion; 5) enhancing the doctor–patient relationship, and 6) ‘being realistic’ about personal limitations and issues such as the availability of time and resources. Dimensions 4 and 6 were later dropped . Brouwers et al. included instruments measuring at least three out of the six dimensions, while Hudon et al. included those measuring at least two out of the later version of four dimensions. Köberich and Farin used a framework of three core themes of person centredness based on Kitson et al. : 1) participation and involvement; 2) relationship between the patient and the health professional; and 3) the context where care is delivered. Finally, Ekman et al. used an Institute of Medicine framework of six dimensions: 1) respect for patients’ values, preferences, and expressed needs; 2) coordination and integration of care; 3) information, communication, and education; 4) physical comfort; 5) emotional support, e.g., relieving fear and anxiety; and 6) involvement of family and friends (Table ). Overview of reviews Hudon et al.’s review aimed to identify and compare instruments, subscales, or items assessing patients’ perceptions of patient-centred care used in an ambulatory family medicine setting. Only patient rated instruments were included. Quality assessment of the validation studies was conducted with the Modified Version of Standards for Reporting of Diagnostic Accuracy (STARD) tool . The authors identified two instruments fully dedicated to patient-centred care, and 11 further instruments with subscales or items measuring person-centred care. Köberich and Farin’s review aimed to provide an overview of instruments measuring patients’ perception of patient-centred nursing care, defined as the degree to which the patient’s wishes, needs and preferences are taken into account by nurses when the patient requires professional nursing care. Again, only patient rated instruments were included. The four included instruments were described in detail, including their theoretical background, development processes including consecutive versions and translations, and validity and reliability testing. No quality assessment was undertaken. Brouwers et al. aimed to review all available instruments measuring patient centredness in doctor–patient communication, in the classroom and workplace, for the purposes of providing direct feedback. Instruments for use in health care professionals other than physicians or medical students were thus excluded. The authors used the COSMIN checklist for quality assessment of the instruments . Ekman et al.’s review aimed to identify available instruments for direct observation in assessment of competence in person-centred care. The study then assessed them with respect to underlying theoretical or conceptual frameworks, coverage of recognized components of person-centred care, types of behavioural indicators, psychometric performance, and format (i.e., checklist, rating scale, coding system). The review used the six-dimension framework endorsed by the Institute of Medicine however, they did not use the framework as a selection criterion. No quality assessment was undertaken. The authors group the included instruments in four categories: global person-centred care/person centredness, shared decision-making, person-centred communication, and nonverbal person-centred communication. The critical appraisal of the included reviews using the Joanna Briggs Institute Critical Appraisal Checklist for Systematic Reviews and Research Syntheses is reported in Table . The review by Brouwers et al. scored positively on all but one items. We note that no study assessed publication bias, and this may be a particularly important threat to valid inference in a literature of this nature. There were issues with the methods of critical appraisal in two reviews. Overview of the validation studies Sixty-eight validation studies were included across the four reviews. Hudon et al. described one to three validation studies for each instrument included and was the only review to report specific information on the validation studies in addition to information on the instruments. Köberich and Farin identified several validation studies for each instrument. Brouwers et al. identified one validation study for each included instrument. Ekman et al. describe one validation study for 13 instruments, and two validation studies for three other included instruments. Table provides an overview of the validation studies [ , – ]. The validation studies were published between 1989 and 2015 inclusive. The majority of the studies were done in English speaking countries: 29 originated in the USA, 10 in the UK, 8 in Canada; 4 in Finland; 2 in Australia, the Netherlands, and Turkey; and 1 in Germany, Israel, Norway, and Sweden. The country of origin was not specified for the remaining 7 studies. Overview of the instruments Forty-two instruments were included across the four reviews, with minimal overlap. The Patient-Centred Observation Form (PCOF) was included in two reviews . The original Perceived Involvement in Care Scale (PICS) is included by Hudon , while Brouwers included the modified PICS (M-PICS). The Consultation and Relational Empathy instrument (CARE), and the Patient Perception of Patient Centeredness (PPPC) are included by both Hudon and Brouwers . Hudon included what they referred to as the Consultation Care Measure (CCM), and Brouwers included the same instrument, named differently as the Little instrument. Little et al. do not name the instrument in their validation study, so we decided to refer to this instrument as the ‘Little Instrument’ in this review of reviews. The four reviews reported varying types of information on the included instruments. All reported the year and country of development, the response scale, the number of subscales and items, and the intended rater of the instrument. Table gives an overview of what information about the instrument is included in each review. As with the validation studies, the publication years of the instruments ranged from 1989 up to 2015. The majority of the instruments were developed in English speaking countries: 21 originated from the USA, 7 from the UK, 7 from Canada; 2 from the Netherlands; and 1 from Australia, Finland, Germany, Israel, and Norway. The country of origin was not specified in the review for the remaining 3 instruments. Table summarises the information that is reported in the reviews. The measurement properties of instruments that were reported in the reviews varied considerably.. Table shows which properties were reported in which review, and Table is a literal presentation of all psychometric information reported in the four included reviews.
The search identified 2,215 unique articles with 21 papers selected for a full-text eligibility assessment (see Fig. ). Four studies were included. None of the reviews identified in further searching fulfilled our inclusion criteria. The four included reviews each had different aims and selection criteria, resulting in few primary studies and instruments being included in more than one review. Two reviews targeted different groups of practitioners; nurses for Köberich and Farin and physicians or medical students for Brouwers et al. ). Hudon et al. and Köberich and Farin included only patient rated instruments, while Ekman et al. included only direct observation tools (e.g., checklists or rating scales). In total, the four reviews included 71 validation studies (68 unique studies) of 42 different instruments.
Conceptualisations of person-centredness varied between the included studies. Two reviews used Stewart and colleagues model of interconnecting dimensions: 1) exploring both the disease and the illness experience; 2) understanding the whole person; 3) finding common ground between the physician and patient; 4) incorporating prevention and health promotion; 5) enhancing the doctor–patient relationship, and 6) ‘being realistic’ about personal limitations and issues such as the availability of time and resources. Dimensions 4 and 6 were later dropped . Brouwers et al. included instruments measuring at least three out of the six dimensions, while Hudon et al. included those measuring at least two out of the later version of four dimensions. Köberich and Farin used a framework of three core themes of person centredness based on Kitson et al. : 1) participation and involvement; 2) relationship between the patient and the health professional; and 3) the context where care is delivered. Finally, Ekman et al. used an Institute of Medicine framework of six dimensions: 1) respect for patients’ values, preferences, and expressed needs; 2) coordination and integration of care; 3) information, communication, and education; 4) physical comfort; 5) emotional support, e.g., relieving fear and anxiety; and 6) involvement of family and friends (Table ).
Hudon et al.’s review aimed to identify and compare instruments, subscales, or items assessing patients’ perceptions of patient-centred care used in an ambulatory family medicine setting. Only patient rated instruments were included. Quality assessment of the validation studies was conducted with the Modified Version of Standards for Reporting of Diagnostic Accuracy (STARD) tool . The authors identified two instruments fully dedicated to patient-centred care, and 11 further instruments with subscales or items measuring person-centred care. Köberich and Farin’s review aimed to provide an overview of instruments measuring patients’ perception of patient-centred nursing care, defined as the degree to which the patient’s wishes, needs and preferences are taken into account by nurses when the patient requires professional nursing care. Again, only patient rated instruments were included. The four included instruments were described in detail, including their theoretical background, development processes including consecutive versions and translations, and validity and reliability testing. No quality assessment was undertaken. Brouwers et al. aimed to review all available instruments measuring patient centredness in doctor–patient communication, in the classroom and workplace, for the purposes of providing direct feedback. Instruments for use in health care professionals other than physicians or medical students were thus excluded. The authors used the COSMIN checklist for quality assessment of the instruments . Ekman et al.’s review aimed to identify available instruments for direct observation in assessment of competence in person-centred care. The study then assessed them with respect to underlying theoretical or conceptual frameworks, coverage of recognized components of person-centred care, types of behavioural indicators, psychometric performance, and format (i.e., checklist, rating scale, coding system). The review used the six-dimension framework endorsed by the Institute of Medicine however, they did not use the framework as a selection criterion. No quality assessment was undertaken. The authors group the included instruments in four categories: global person-centred care/person centredness, shared decision-making, person-centred communication, and nonverbal person-centred communication. The critical appraisal of the included reviews using the Joanna Briggs Institute Critical Appraisal Checklist for Systematic Reviews and Research Syntheses is reported in Table . The review by Brouwers et al. scored positively on all but one items. We note that no study assessed publication bias, and this may be a particularly important threat to valid inference in a literature of this nature. There were issues with the methods of critical appraisal in two reviews.
Sixty-eight validation studies were included across the four reviews. Hudon et al. described one to three validation studies for each instrument included and was the only review to report specific information on the validation studies in addition to information on the instruments. Köberich and Farin identified several validation studies for each instrument. Brouwers et al. identified one validation study for each included instrument. Ekman et al. describe one validation study for 13 instruments, and two validation studies for three other included instruments. Table provides an overview of the validation studies [ , – ]. The validation studies were published between 1989 and 2015 inclusive. The majority of the studies were done in English speaking countries: 29 originated in the USA, 10 in the UK, 8 in Canada; 4 in Finland; 2 in Australia, the Netherlands, and Turkey; and 1 in Germany, Israel, Norway, and Sweden. The country of origin was not specified for the remaining 7 studies.
Forty-two instruments were included across the four reviews, with minimal overlap. The Patient-Centred Observation Form (PCOF) was included in two reviews . The original Perceived Involvement in Care Scale (PICS) is included by Hudon , while Brouwers included the modified PICS (M-PICS). The Consultation and Relational Empathy instrument (CARE), and the Patient Perception of Patient Centeredness (PPPC) are included by both Hudon and Brouwers . Hudon included what they referred to as the Consultation Care Measure (CCM), and Brouwers included the same instrument, named differently as the Little instrument. Little et al. do not name the instrument in their validation study, so we decided to refer to this instrument as the ‘Little Instrument’ in this review of reviews. The four reviews reported varying types of information on the included instruments. All reported the year and country of development, the response scale, the number of subscales and items, and the intended rater of the instrument. Table gives an overview of what information about the instrument is included in each review. As with the validation studies, the publication years of the instruments ranged from 1989 up to 2015. The majority of the instruments were developed in English speaking countries: 21 originated from the USA, 7 from the UK, 7 from Canada; 2 from the Netherlands; and 1 from Australia, Finland, Germany, Israel, and Norway. The country of origin was not specified in the review for the remaining 3 instruments. Table summarises the information that is reported in the reviews. The measurement properties of instruments that were reported in the reviews varied considerably.. Table shows which properties were reported in which review, and Table is a literal presentation of all psychometric information reported in the four included reviews.
This review of reviews sought to summarise the range of validated instruments available for measuring practitioners’ person-centred consultation skills, including the strength of the validation evidence for each instrument, and to appraise the systematic reviews examining the validation studies. The reviews varied in quality, and our JBI quality assessment showed only one review which fulfilled all assessment criteria except for the assessment of publication bias . In addition, only one review described several validation studies per instrument, including modifications and translations . We found that the four included systematic reviews used very different inclusion criteria, leading to little overlap in included validation studies and instruments between them. This was because the reviews also differed in aims, appraisal tools used, and conceptual framework used, which limited the consistency of reported information across studies and instruments. These features underline the value of the present study, which in bringing together these literatures offers a guide to a wider set of instruments of interest to researchers than has previously been available. This diversity also underlines a key limitation of this review of reviews, as the included reviews themselves may complicate attention to the primary literature unhelpfully. We make no claim that the list of instruments reported in this review of reviews is exhaustive. Our search was undertaken in September 2020 and although we have checked for citations of the included reviews and the primary studies, we may have missed later published reviews and instruments. There are many more instruments available, varying in aims, objectives, and conceptualisations of person-centredness. In addition, there may be other validation studies available on the instruments the reviews did not include, or which were published after the reviews, and the study findings suggest it is indeed likely that new instruments will have been published. We searched for all reviews meeting our selection criteria and acknowledge the perennial possibility that we may have missed eligible reviews, as well as being clear that there exist other validation studies and instruments that our study was not designed to include. We used an extensive list of keywords for our search, based on published reviews of person-centredness, but as the concept is so scattered, we may have left out search terms that could have led us to other reviews that could have been included. This we regard as a real risk and suggest careful extension of search strategy development in future studies. Procedural issues, particularly reliance on sole author for screening and data extraction, albeit with checks, should be borne in mind as review limitations. There are many instruments available which measure person-centred skills in healthcare practitioners. The reviews point out that the instruments measured person-centredness in various dimensions, emphasising different aspects of the basic concept of person-centredness. This indicates the lack of agreement on what could be considered defining, central or important characteristics, so there are construct validity issues to be considered carefully. Person-centred care is an umbrella term used for many different conceptualisations in many different contexts . Separating consideration of what constitutes person centred care from person centred consultation skills is necessary, as the latter construct is merely one element of the former. Often teaching materials and guidelines on person centredness are not very clear on what person-centred behaviour and communication actually entails, and what skills and behaviours health care professionals are supposed to learn to make their practice person-centred. For example, Kitson and colleagues reported that health policy stakeholders and nurses perceive patient-centred care more broadly than medical professionals. Medical professionals tend to focus on the doctor-patient relationship and the decision-making process, while in the nursing literature there is also a focus on patients’ beliefs and values . Measurement instruments can help us operationalise person-centredness and can help practitioners understand what exactly it is that they are supposed to be doing. Developing the science of measurement in this area may also assist resolution of the construct validity issues by making clear what can be validly measured and what cannot. Three of the four reviews [ , , ] concluded that psychometric evidence is lacking for nearly all of the instruments. This finding may seem unsurprising in light of the foregoing discussion of construct validity. Brouwers used the COSMIN rating scale and found only one instrument rated as ‘excellent’ on all aspects of validity studied (internal consistency, content, and structural validity), but its reliability had not been studied. Köberich specifically mentions test–retest reliability as a neglected domain and adds that all instruments lack evidence of adequate convergent, discriminant, and structural validity testing. Köberich and Farin, Brouwers, and Ekman [ , , ] also highlight the need for further research on validity and reliability of existing instruments in their discussion and conclusion sections. In other reviews, De Silva , Gärtner et al. and Louw et al. attribute the lack of good evidence on the measurement qualities of instruments both to a failure to study their measurement properties and to the overall poor methodological quality of validation studies. Many tools are developed but few are studied sufficiently in terms of their psychometric properties and usefulness for research on and teaching of person-centredness. Often, a tool is “developed, evaluated, and then abandoned” . Researchers and research users may seek instruments of these kinds for many different purposes. Using the most relevant and promising instruments that have already been developed and tested, in however a limited fashion, and rigorously studying and reporting on their psychometric properties, will be useful in building the science of measuring person-centred consultation skills. It may also be useful to develop item banking approaches that combine instruments. Researchers or educators intending to choose an instrument for their purposes also need to know several things to decide whether an instrument is relevant and suitable for their specific needs. For future primary studies and systematic reviews, we suggest paying heed to, and indeed rectifying, the limitations of existing studies identified here and elsewhere. In addition, both Hudon and Ekman found that paradoxically, there is very limited evidence of patients taking part in the evaluation process. This has also been reported in a systematic review by Ree et al. who looked specifically at patient involvement in person centeredness instruments for health professionals. This is painfully ironic. There is thus a further major lesson to be drawn from this study; that in developing the science of measurement of person-centred skills, new forms of partnership need to be formed between researchers and patients.
There are many instruments available which measure person-centred skills in healthcare practitioners and the most relevant and promising instruments that have already been developed, or items within them, should be further studied rigorously. Validation study of existing material is needed, not the development of new measures. New forms of partnership are needed between researchers and patients to accelerate the pace at which further work will be successful.
Additional file 1. Search string for Embase, PsycInfo, MEDLINE. Search string for CINAHL.
|
The relationship between person-centered care in nursing homes and COVID-19 infection, hospitalization, and mortality rates | 46e41f3a-92ba-42fb-b2b5-1135b7a4cdb2 | 10075075 | Patient-Centered Care[mh] | The impact of COVID-19 on nursing home (NH) residents resulted in considerable attention being directed at the organizational structure, practices, and policies associated with NH care. For example, one study by Zimmerman and colleagues explored Green House and other small NH household models in comparison to larger and traditional NHs and found COVID-19 incidence and mortality rates to be lower in the Green House/small home settings. Household models involve more than just redesigned architecture and interiors; they also include operational and organizational approaches to person-centered care which themselves may be beneficial to infection control and resident outcomes. One such model of person-centered NH care being enacted statewide is the Kansas PEAK (Promoting Excellent Alternatives in Kansas Nursing Homes) 2.0 program. PEAK 2.0 began in 2012 as a tiered payment structure to incentivize Kansas NHs to implement person-centered care. Participation in the PEAK program is voluntary. PEAK is a structured program with consistent operationalized definitions for 12 core person-centered care practices, and all participating homes are provided training and education about common expectations for implementation. For example, assigning consistent staff to a reduced number of residents who live in a designated living area is one of the required practices to “get small” and encourage relationship building. After initial training, PEAK homes undergo an annual external evaluation by the PEAK staff to determine the core practices that have been successfully implemented which determines their level in the program. . Levels range from 0 to 5, with 0 representing a foundational year of planning and 5 representing comprehensive and sustained adoption of all of the required domains and cores in the program. Research comparing PEAK homes to non-PEAK homes has demonstrated that PEAK homes have higher satisfaction with quality of life and clinical quality of care. , It is plausible that person-centered care practices (such as creating dedicated teams of consistent staffing for a smaller group of residents, and decentralized dining) similarly benefit infection control and related outcomes. Therefore, this study explored the potential influence of person-centered care practices on the incidence and mortality rates of COVID-19 comparing PEAK to non-PEAK homes. Modeling the methods used in the study of Green House/small NHs by Zimmerman et al., it was hypothesized that PEAK homes would perform better related to COVID-19 infections, admission/readmission, and mortality rates than non-PEAK homes.
This cohort study used NH COVID-19 data published by the Center for Medicare & Medicaid Services (CMS). Beginning May 2020, all U.S. NHs were required to report COVID-19 data to CMS, and the data were made publicly available. COVID-19 data of all Kansas NHs reporting to CMS from January 20, 2020 to July 31, 2020 were downloaded for these analyses (the same dates as in the Zimmerman et al. study); the end date was chosen to precede changes in care such as becoming COVID-19-only NHs. Sample To identify providers in the CMS data who had been active in the PEAK program prior to the start of the pandemic, homes with established PEAK level status (0–5) in 2018 and 2019 were identified. The remaining homes were identified as non-PEAK NHs for comparison. Given geographical differences in COVID-19 infiltration, PEAK and non-PEAK homes were matched and paired within counties for analysis. Of the 314 Kansas NHs that reported sufficient analyzable COVID-19 information to CMS, 170 (54%) were PEAK NHs in 2018 and 2019. Of these, only 37 counties had both PEAK and non-PEAK homes, resulting in 109 PEAK and 112 non-PEAK NHs for analysis. The majority of counties had 5 or fewer PEAK homes, the exception being 3 counties that each had 10–12 PEAK NHs. Variables used from the CMS data included individual NH census (occupied beds), number of residents with confirmed COVID-19 infections (new laboratory positive cases), COVID-19 admission/readmission (person admitted or readmitted or previously hospitalized and treated for COVID-19), and number of resident deaths from COVID-19 (suspected or laboratory positive who died in the NH or another location). Analyses COVID-19 cases and admissions/readmissions were calculated per 1000 resident days; COVID-19 mortality was calculated per 100 suspected and confirmed positive COVID-19 cases. The analysis modeled the method used by Zimmerman et al. to derive rates. To calculate the COVID-19 infection rate, COVID-19 case counts were summed for each NH, divided by total case counts per days of exposures (i.e., resident days), and multiplied by 1000. COVID-19 admission/readmission rates were calculated using the same approach. In contrast, the COVID-19 mortality rate was calculated by dividing the sum of COVID-19 death counts by the sum of suspected plus confirmed COVID-19 cases and multiplying by 100. For all three COVID-19 outcomes—cases, admission/readmission, and mortality—a log-rank test was applied to compare rates between PEAK and non-PEAK NHs. Log-rank tests are nonparametric tests used to detect differences among higher values and censored data. The P -value of <0.05 was considered statistically significant. Analyses were conducted in R version 4.0 (2020-04-24). Secondarily, differences in COVID-19 infection, admission/readmission, and mortality rates were analyzed by the extent of person-centeredness among PEAK homes (i.e., PEAK levels). PEAK NHs at levels 0 and 1 are novice and were combined to represent the early stage of person-centered implementation (Stage 1). Homes at the PEAK levels 0-1 (Stage 1) have implemented very few person-centered practices, so it was important to make this group distinct from other PEAK participating homes. PEAK NHs at levels 2–5 are those with moderate to advanced experience of person-centered implementation and were combined to represent the more advanced stage (Stage 2). PEAK NHs at levels 2–5 (Stage 2) have demonstrated implementation of a broad array of PCC practices, such as creating dedicated teams of consistent staffing for a smaller group of residents, and decentralized dining. This strategy resulted in three groups for purposes of comparison: non-PEAK NHs, Stage 1 PEAK NHs, and Stage 2 PEAK NHs.
To identify providers in the CMS data who had been active in the PEAK program prior to the start of the pandemic, homes with established PEAK level status (0–5) in 2018 and 2019 were identified. The remaining homes were identified as non-PEAK NHs for comparison. Given geographical differences in COVID-19 infiltration, PEAK and non-PEAK homes were matched and paired within counties for analysis. Of the 314 Kansas NHs that reported sufficient analyzable COVID-19 information to CMS, 170 (54%) were PEAK NHs in 2018 and 2019. Of these, only 37 counties had both PEAK and non-PEAK homes, resulting in 109 PEAK and 112 non-PEAK NHs for analysis. The majority of counties had 5 or fewer PEAK homes, the exception being 3 counties that each had 10–12 PEAK NHs. Variables used from the CMS data included individual NH census (occupied beds), number of residents with confirmed COVID-19 infections (new laboratory positive cases), COVID-19 admission/readmission (person admitted or readmitted or previously hospitalized and treated for COVID-19), and number of resident deaths from COVID-19 (suspected or laboratory positive who died in the NH or another location).
COVID-19 cases and admissions/readmissions were calculated per 1000 resident days; COVID-19 mortality was calculated per 100 suspected and confirmed positive COVID-19 cases. The analysis modeled the method used by Zimmerman et al. to derive rates. To calculate the COVID-19 infection rate, COVID-19 case counts were summed for each NH, divided by total case counts per days of exposures (i.e., resident days), and multiplied by 1000. COVID-19 admission/readmission rates were calculated using the same approach. In contrast, the COVID-19 mortality rate was calculated by dividing the sum of COVID-19 death counts by the sum of suspected plus confirmed COVID-19 cases and multiplying by 100. For all three COVID-19 outcomes—cases, admission/readmission, and mortality—a log-rank test was applied to compare rates between PEAK and non-PEAK NHs. Log-rank tests are nonparametric tests used to detect differences among higher values and censored data. The P -value of <0.05 was considered statistically significant. Analyses were conducted in R version 4.0 (2020-04-24). Secondarily, differences in COVID-19 infection, admission/readmission, and mortality rates were analyzed by the extent of person-centeredness among PEAK homes (i.e., PEAK levels). PEAK NHs at levels 0 and 1 are novice and were combined to represent the early stage of person-centered implementation (Stage 1). Homes at the PEAK levels 0-1 (Stage 1) have implemented very few person-centered practices, so it was important to make this group distinct from other PEAK participating homes. PEAK NHs at levels 2–5 are those with moderate to advanced experience of person-centered implementation and were combined to represent the more advanced stage (Stage 2). PEAK NHs at levels 2–5 (Stage 2) have demonstrated implementation of a broad array of PCC practices, such as creating dedicated teams of consistent staffing for a smaller group of residents, and decentralized dining. This strategy resulted in three groups for purposes of comparison: non-PEAK NHs, Stage 1 PEAK NHs, and Stage 2 PEAK NHs.
displays COVID-19 rates of PEAK and non-PEAK NHs, including median rates of COVID-19 cases, COVID-19 admission/readmission per 1000 resident days, and COVID-19 mortality per 100 positive cases; it shows that the median rates and rates at the 75th percentile are 0 in all NHs (PEAK and non-PEAK). Comparatively non-PEAK NHs have higher COVID-19 cases, admission, and mortality than PEAK homes. In NHs that lie above 90% (90th percentile), non-PEAK NHs had 3.9 times more COVID-19 cases than PEAK NHs (31 COVID-19 cases for every 1000 resident days in non-PEAK homes compared to 8 COVID-19 cases in PEAK homes). Similarly, the COVID-19 admission/readmission rate was 2.5 times more in non-PEAK than PEAK NHs. In terms of mortality rates, at the 90th percentile both non-PEAK and PEAK NHs had zero COVID-19 mortality. Above the 95th percentile, however, non-PEAK NHs had a 10.7 COVID-19 mortality rate per 100 positive cases, compared to a PEAK NH rate of 0; the maximum mortality rate in non-PEAK homes was 56.9 per 100 positive cases compared with 16.7 in PEAK NHs. displays lower COVID-19 case rates, admission/readmission rates, and COVID-19 mortality rates among NHs with increased levels of PEAK participation. Each dot represents COVID-19 case, admission, and mortality rates. The majority of the rates fall below the third quartiles, and PEAK homes at levels 2 to 5 (Stage 2) have significantly fewer dots in the fourth quartile compared to NHs that never participated in PEAK. Non-PEAK NHs had scattered and higher rates (case, admission, and mortality) than PEAK NHs levels 0-1 (Stage 1) and levels 2–5 (Stage 2).
Studies across the U.S. and Canada have examined data associated with the incidence of COVID-19 infections, extent of infections, and mortality rates in NHs. In the study by Zimmerman and colleagues, the median (middle value) rates of COVID-19 cases per 1000 resident days were 0 in both Green House/small NHs and NHs <50 beds, while they were 0.06 in NHs >50 beds. This study sought to conduct a comparative analysis using the same research design and time frame, given the related focus on person-centered care and recognition of the low rates and geographic clustering. In other studies, characteristics examined in relation to COVID-19 include design standards and NH size, , , profit status, CMS quality measures, , , , , , geographic factors (e.g., urban versus rural), , , population variables (e.g., racial and ethnic characteristics), , , and staffing levels and mix. , , To affect change, providers must be able to translate results into actions that are within their control to implement and sustain with available resources. Characteristics such as the design of the building (e.g., standalone households versus traditional buildings) are not actionable change for an individual provider, nor is location or numerous other salient characteristics. Person-centered care practices, however, have been shown to be implemented in a broad cross-section of environments, settings, provider types, and geographic locations. , The outcomes of these person-centered efforts have had positive impacts on both quality of health and quality of life. This study suggests that person-centered care may also benefit infection prevention and related outcomes, as witnessed through the COVID-19 data. The PEAK practices most likely to be related to infection prevention and outcomes include changes associated with “getting small.” These changes entail consistent staffing approaches, no more than 30 residents in designated work areas (architecturally defined), removal of large-centralized nursing desks in exchange for dispersed and integrated places to chart, and expectations that these work areas have all the necessary supplies and equipment needed by staff. Additionally, PEAK practices emphasize leadership approaches that engage all staff in decision making and problem-solving. Of course, the design of this study does not allow for a causal examination of the association between person-centered practices and improved infection control and outcomes. These findings are preliminary and more research on infection prevention outcomes are needed. Such research should include co-variates that were not available for these analyses. One confounder relates to the nature of the NHs that adopt PEAK. Of the 314 NHs in the sampling frame for this study, 54% were PEAK homes. Other studies have shown that homes that adopt culture change (typically synonymous with person-centered care) differ from other homes such as by being non-profit and having fewer Medicaid residents, both of which also relate to higher quality care. , Similarly, early PEAK adopters tend to be non-profit and have higher occupancy and quality ratings than non-PEAK homes, but these differences lessen and are nonsignificant over time. Other limitations are that this study is restricted to a sample of homes in one state—although a strength is that it is a state-wide sample—and that the findings may not reflect Kansas NHs that did not report data to CMS. The timeframe for this study is also fairly narrow, but it provides a comparative analysis to the study on Green House/small house which uses the same window of observation and was not confounded by homes choosing to purposefully serve COVID-19 residents. Consequently, the findings remain valid despite the limited window of observation.
Addressing the quality of NH care has gained renewed attention due to the devastation caused by COVID-19. Solutions for safer and more effective care must address a spectrum of approaches because NHs across the country represent a range of settings. Providers around the world are turning their attention to different models of care and operational behaviors that reduce characteristics associated with cross-contamination and other problems with infection control. New building, remodeling, and organizational practices should be carefully considered in response to a continued understanding of how the physical and social environment support improved operational practices and resident outcomes. Addressing the needs of existing skilled care settings may require attention to person-centered practices and re-conceptualizing the use of existing buildings in a manner that answers the call for continuous quality improvement and better outcomes. PEAK homes share similar cultural and environmental priorities found in the household / small house models of care, and results from this analysis demonstrate that there may also be positive implications for managing contagious infections in these settings. The implications are that improved outcomes, particularly around infection control, may perhaps be achieved through affordable adaptations to NH policy and practice promoting person-centeredness. Doing so could improve care that is more broadly accessible to diverse resident populations.
The Kansas State Center on Aging, Richardson Foundation Fund was used to support this work.
MK is the Co-Principal Investigator on the PEAK Project contracted with the Kansas Department for Aging and Disability (KDADs) and LC is the Program Director. Other authors declare no conflict of Interests.
|
Proteome alterations in human autopsy tissues in relation to time after death | 24c197c8-d106-4218-ad66-2e8599c95d7f | 10075177 | Forensic Medicine[mh] | For centuries, the autopsy has provided the best quality assurance for clinical diagnostics and further improvement of therapeutic approaches. Currently, numerous research projects, including the most advanced translational multi-omics studies provide deeper insights than ever before into various disease states, especially in the context of cancer . Naturally, tissue specimens and their quality are of pivotal importance for these studies. To our knowledge, such studies are largely based on biopsy samples or surgically resected specimens, often excluding autopsy samples, which however may yield unique insight into advanced disease states. Leading cancer centers have established Rapid Autopsy Programs (RAPs). These projects aim to obtain the tissue from deceased patients immediately after death to reduce the effect of post-mortem tissue degradation. Several recent studies highlight the usefulness of RAP, e.g. for genomics , transciptomics or proteomics . Nevertheless, RAPs remain confined to specialized institutions and it still poses a challenge to harness the possibilities of non-RAP autopsy specimens for translational research. Some authors have sought to keep the post-mortem interval as short as possible by performing core biopsies after death . However, tissue specimens obtained during traditional autopsies are still much more widely available for both cancer and non-cancerous diseases. Numerous studies highlight the value of autopsy-based samples in biomedical research, with examples covering cancer research , improvement of clinical practice and education , neuroscience research , pathogenesis of COVID-19 disease , as well as genetic and tissue-specific gene expression differences in non-cancer individuals . Moreover, autopsy cases can provide large amounts of tissue samples that can be used for immunohistochemical purposes as control tissue from organs from which only biopsy samples are usually available. Protein expression is a primary area of interest for routine histological diagnostics and tissue-based research projects, but the limitations of its post-mortem applicability remain largely unclear, making it challenging to use for this purpose. It is therefore important to investigate post-mortem temporal kinetics of protein expression patterns to determine the limitations on the relevance of proteomic results based on autopsy samples and their applicability to support more comprehensive conclusions. In the present study, the proteome of human tissue specimens obtained during a routine autopsy at defined time points has been analyzed via liquid-chromatography tandem mass spectrometry (LC–MS/MS). The objectives of the study presented here were (a) to determine the maximum post-mortem interval (PMI) which is still suitable for the characterization of protein expression patterns; (b) to explore organ-specific differences in protein degradation; and (c) to investigate whether certain proteins follow specific degradation kinetics. Here, we will demonstrate that the proteomic results can be used to determine the post-mortem time interval as long as protein degradation is sufficiently limited to allow tissue samples to be used for protein-based diagnostic or research purposes.
Experimental set-up The study protocol is in accordance with the Declaration of Helsinki and has been approved by the Ethics Committee of Semmelweis University, Budapest, Hungary (13-1/2015). All autopsies were performed at the Department of Pathology, Forensic and Insurance Medicine, Semmelweis University, Budapest, Hungary. Tissue specimens from the lung, kidney and liver have been derived from patients with predefined post-mortem time interval (6, 12, 18, 24, 48, 72, 96 h) and a lack of specific diseases significantly affecting the preservation of the investigated organs. Corpses were stored at 4 °C between death and autopsy procedure. For each time interval not less than three deceased patients were included. Tissue lysis 0.1 mg from each tissue specimen were cut into small pieces, mixed with lysis buffer (200 mM HEPES, pH 7.5 (AppliChem, Germany), 1% acid-labile surfactant (ALS, sodium 3-[(2-methyl-2-undecyl-1,3-dioxolan-4-yl) methoxy]-1-propanesulfonate)), and incubated at 90 °C for 10 min with mild agitation (500 rpm). Subsequently, samples were further homogenized using a PreCellys Lysis Kit (Bertin instruments, USA) for three cycles (each 30 secundum, at 6500 rpm) and sonicated with a Biorupture device for 20 cycles (each 30 secundum ON and 30 secundum OFF). After centrifugation (at 4 °C, for 10 min at max speed), the supernatant was collected, the pH value was adjusted to pH 8.0 and samples were reduced (5 mM TCEP (Sigma Aldrich, Switzerland), 10 min at 95 °C) and alkylated (10 mM iodacetamide (Sigma Aldrich, USA) in darkness at room temperature for 30 min). A double digest with Lysyl Endopeptidase (Wako, Japan) in a ratio of 1:100 for 2 h, at 37 °C, at 500 rpm and trypsin TPCK treated (Worthington, USA) in a ratio of 1:50 overnight at 37 °C, at 500 rpm was performed. Sample purification For the ALS deactivation and acidifying of the peptide solution, incubation was performed with 2% Trifluoroacetic acid (TFA) (Thermo Scientific, USA) for 30 min, 37 °C, at 500 rpm. The supernatant was used for purification. The columns [HyperSep SpinTip C18, (Thermo Scientific, USA)] were conditioned by adding elution buffer (0.05% TFA [Thermo Scientific, USA), 68% Acetonitrile (ACN) (Merck, Germany)] and centrifugation. Afterward, the columns were equilibrated by adding wash buffer (0.05% TFA, 8% ACN) followed by a centrifugation step. The peptide solution was loaded twice on the columns. The columns were washed three times with wash solution for 1 min, at 1000× g and peptides were eluted with elution buffer for 2 min, at 1000× g afterwards. Finally, 2 μg of peptides was dried at 45 °C in a vacuum concentrator and submitted for LC–MS/MS. Preparation of isobarically labeled samples was performed essentially as described previously . LC–MS/MS measurements For the mass spectrometry measurements, the dried peptides were re-suspended with 1% formic acid (Fluka, USA). One µg peptides were injected on a Q-Exactive Plus mass spectrometer (Thermo Scientific, San Jose, CA) coupled to an EASY-nLC 1000 UHPLC system (Thermo Scientific). The analytical column was self-packed with silica beads coated with C18 (Reprosil Pur C18-AQ, d = 3 Â) (Dr. Maisch HPLC GmbH, Ammerbusch, Germany). For peptide separation, a linear gradient of increasing buffer B [0.1% formic acid and 1% mono ethylene glycol (Sigma-Aldrich, USA) in 80% ACN, (Merck, Germany)] was applied, ranging from 5 to 40% buffer B over the first 90 min and from 40 to 100% buffer B in the subsequent 30 min (120 min separating gradient length). Peptides were analyzed in data-dependent acquisition mode (DDA). All included samples of each specimen were measured in one run. LC–MS/MS conditions for isobarically labeled samples are as described previously . Protein identification and quantitation Raw data were analyzed with MaxQuant (v 1.6.17.0) allowing for zero missed cleavage sites, no variable modifications, carbamidomethylation of cysteines as fixed modification. Furthermore, the digestion Trypsin and label-free quantification options were selected. For the protein quantification label min ratio count was set to 1. Only unique peptides were used for quantification, match-between-run settings were allowed. The Human-EBI-reference database was used for protein identification. (one protein per gene; downloaded from https://www.ebi.ac.uk/ on Jan 11th 2020; iRT sequences added manually, and common contaminants were included by MaxQuant). Analysis of isobarically labelled samples was performed using MSFragger and TMT-Integrator, allowing for zero missed cleavage sites, no variable modifications, carbamidomethylation of cysteines, and TMTpro-labelling of peptide N-termini and lysines as fixed modification. Proteomics data processing and analysis MaxQuant evidence and protein group files were imported into the European Galaxy server ( https://usegalaxy.eu ) and analyzed with MSstats (v 3.22.0.1) and basic filter tools were applied separately for each tissue. To investigate the degree of time-dependent proteome alterations, the first time point (6 h post-mortem), was used as a reference time point. All other time points were pair-wise compared to the reference time point. Proteins with adjusted p -value < 0.05 were considered as significantly fast-degraded (negative fold change) termed as degraded proteins or slow-degraded termed as pseudo-enriched (positive fold change) proteins and exported as tab-separated files. All further protein identifications, quantitative data processing and statistical analyses were performed using the R programming language (v 4.0.4) running on RStudio (v 1.4.1103-4). For each tissue separately, the protein and peptide identifications, along with their quantitative data, were processed using the MSstats package (v 3.20.3) for summarization of protein quantitation and normalization. Only unique peptides were used for quantitation. Of note, “ Leading Razor Protein” were used as protein IDs. Data were log2-transformed before median equalization for normalization. The resulting data were reformatted into a protein abundance matrix and used for further analyses. The distribution of missing values was visually assessed using the naniar 10 package (v 0.6.1) . Random forest-based imputation ( missForest package v 1.4) was applied as a preprocessing step before data modeling. The limma package (v3.48.1) was used to apply a linear model using time as a continuous variable to detect proteins with either linearly positive or negative associations with time. The p-values arising from these inferential analyses were corrected by multiple testing using the Benjamini–Hochberg method, as applied by limma . Proteins showing an adjusted p -value smaller than 0.05 after modeling, were considered as associated with time. To evaluate the general functional annotation of the proteins detected by the linear model, enrichment analyses against Gene Ontology (Biological Processes and Cellular Component and Protein Family) annotation was performed. We used the Bioconductor Gene Ontology (GO.db;) and Human Ensembl (EnsDb.Hsapiens.v86;) annotation databases for the enrichment tests. The analyses consisted of over-representation tests based hypergeometric models as implemented in the clusterProfiler package (version 3.16.1) . Proteins with either positive (pseudo-enriched) or negative (depleted) association with Time and a non-adjusted p -value < 0.05 after limma were used as an input for the tests. The enrichment analysis was performed individually for either pseudo-enriched or depleted proteins within each tissue. All proteins identified in such tissue were used as background, and functional groups presenting an adjusted p -value < 0.05 were considered as significantly enriched. Data and code availability Raw spectral files and intermediary identification search results were uploaded to the MassIVE repository (part of the ProteomeXchange consortium). Data can be accessed using the MassIVE accession number MSV000090133 or the ProteomeXchange accession number PXD036059. The peptide and protein identification files from MaxQuant, together with sample annotation, code for pre-processing, reproducible reports and visualizations for the analysis of protein abundance association with time, are openly accessible via Zenodo and GitHub (Cosenza-Contreras, 2022; https://doi.org/10.5281/zenodo.7007047 ) .
The study protocol is in accordance with the Declaration of Helsinki and has been approved by the Ethics Committee of Semmelweis University, Budapest, Hungary (13-1/2015). All autopsies were performed at the Department of Pathology, Forensic and Insurance Medicine, Semmelweis University, Budapest, Hungary. Tissue specimens from the lung, kidney and liver have been derived from patients with predefined post-mortem time interval (6, 12, 18, 24, 48, 72, 96 h) and a lack of specific diseases significantly affecting the preservation of the investigated organs. Corpses were stored at 4 °C between death and autopsy procedure. For each time interval not less than three deceased patients were included.
0.1 mg from each tissue specimen were cut into small pieces, mixed with lysis buffer (200 mM HEPES, pH 7.5 (AppliChem, Germany), 1% acid-labile surfactant (ALS, sodium 3-[(2-methyl-2-undecyl-1,3-dioxolan-4-yl) methoxy]-1-propanesulfonate)), and incubated at 90 °C for 10 min with mild agitation (500 rpm). Subsequently, samples were further homogenized using a PreCellys Lysis Kit (Bertin instruments, USA) for three cycles (each 30 secundum, at 6500 rpm) and sonicated with a Biorupture device for 20 cycles (each 30 secundum ON and 30 secundum OFF). After centrifugation (at 4 °C, for 10 min at max speed), the supernatant was collected, the pH value was adjusted to pH 8.0 and samples were reduced (5 mM TCEP (Sigma Aldrich, Switzerland), 10 min at 95 °C) and alkylated (10 mM iodacetamide (Sigma Aldrich, USA) in darkness at room temperature for 30 min). A double digest with Lysyl Endopeptidase (Wako, Japan) in a ratio of 1:100 for 2 h, at 37 °C, at 500 rpm and trypsin TPCK treated (Worthington, USA) in a ratio of 1:50 overnight at 37 °C, at 500 rpm was performed.
For the ALS deactivation and acidifying of the peptide solution, incubation was performed with 2% Trifluoroacetic acid (TFA) (Thermo Scientific, USA) for 30 min, 37 °C, at 500 rpm. The supernatant was used for purification. The columns [HyperSep SpinTip C18, (Thermo Scientific, USA)] were conditioned by adding elution buffer (0.05% TFA [Thermo Scientific, USA), 68% Acetonitrile (ACN) (Merck, Germany)] and centrifugation. Afterward, the columns were equilibrated by adding wash buffer (0.05% TFA, 8% ACN) followed by a centrifugation step. The peptide solution was loaded twice on the columns. The columns were washed three times with wash solution for 1 min, at 1000× g and peptides were eluted with elution buffer for 2 min, at 1000× g afterwards. Finally, 2 μg of peptides was dried at 45 °C in a vacuum concentrator and submitted for LC–MS/MS. Preparation of isobarically labeled samples was performed essentially as described previously .
For the mass spectrometry measurements, the dried peptides were re-suspended with 1% formic acid (Fluka, USA). One µg peptides were injected on a Q-Exactive Plus mass spectrometer (Thermo Scientific, San Jose, CA) coupled to an EASY-nLC 1000 UHPLC system (Thermo Scientific). The analytical column was self-packed with silica beads coated with C18 (Reprosil Pur C18-AQ, d = 3 Â) (Dr. Maisch HPLC GmbH, Ammerbusch, Germany). For peptide separation, a linear gradient of increasing buffer B [0.1% formic acid and 1% mono ethylene glycol (Sigma-Aldrich, USA) in 80% ACN, (Merck, Germany)] was applied, ranging from 5 to 40% buffer B over the first 90 min and from 40 to 100% buffer B in the subsequent 30 min (120 min separating gradient length). Peptides were analyzed in data-dependent acquisition mode (DDA). All included samples of each specimen were measured in one run. LC–MS/MS conditions for isobarically labeled samples are as described previously .
Raw data were analyzed with MaxQuant (v 1.6.17.0) allowing for zero missed cleavage sites, no variable modifications, carbamidomethylation of cysteines as fixed modification. Furthermore, the digestion Trypsin and label-free quantification options were selected. For the protein quantification label min ratio count was set to 1. Only unique peptides were used for quantification, match-between-run settings were allowed. The Human-EBI-reference database was used for protein identification. (one protein per gene; downloaded from https://www.ebi.ac.uk/ on Jan 11th 2020; iRT sequences added manually, and common contaminants were included by MaxQuant). Analysis of isobarically labelled samples was performed using MSFragger and TMT-Integrator, allowing for zero missed cleavage sites, no variable modifications, carbamidomethylation of cysteines, and TMTpro-labelling of peptide N-termini and lysines as fixed modification.
MaxQuant evidence and protein group files were imported into the European Galaxy server ( https://usegalaxy.eu ) and analyzed with MSstats (v 3.22.0.1) and basic filter tools were applied separately for each tissue. To investigate the degree of time-dependent proteome alterations, the first time point (6 h post-mortem), was used as a reference time point. All other time points were pair-wise compared to the reference time point. Proteins with adjusted p -value < 0.05 were considered as significantly fast-degraded (negative fold change) termed as degraded proteins or slow-degraded termed as pseudo-enriched (positive fold change) proteins and exported as tab-separated files. All further protein identifications, quantitative data processing and statistical analyses were performed using the R programming language (v 4.0.4) running on RStudio (v 1.4.1103-4). For each tissue separately, the protein and peptide identifications, along with their quantitative data, were processed using the MSstats package (v 3.20.3) for summarization of protein quantitation and normalization. Only unique peptides were used for quantitation. Of note, “ Leading Razor Protein” were used as protein IDs. Data were log2-transformed before median equalization for normalization. The resulting data were reformatted into a protein abundance matrix and used for further analyses. The distribution of missing values was visually assessed using the naniar 10 package (v 0.6.1) . Random forest-based imputation ( missForest package v 1.4) was applied as a preprocessing step before data modeling. The limma package (v3.48.1) was used to apply a linear model using time as a continuous variable to detect proteins with either linearly positive or negative associations with time. The p-values arising from these inferential analyses were corrected by multiple testing using the Benjamini–Hochberg method, as applied by limma . Proteins showing an adjusted p -value smaller than 0.05 after modeling, were considered as associated with time. To evaluate the general functional annotation of the proteins detected by the linear model, enrichment analyses against Gene Ontology (Biological Processes and Cellular Component and Protein Family) annotation was performed. We used the Bioconductor Gene Ontology (GO.db;) and Human Ensembl (EnsDb.Hsapiens.v86;) annotation databases for the enrichment tests. The analyses consisted of over-representation tests based hypergeometric models as implemented in the clusterProfiler package (version 3.16.1) . Proteins with either positive (pseudo-enriched) or negative (depleted) association with Time and a non-adjusted p -value < 0.05 after limma were used as an input for the tests. The enrichment analysis was performed individually for either pseudo-enriched or depleted proteins within each tissue. All proteins identified in such tissue were used as background, and functional groups presenting an adjusted p -value < 0.05 were considered as significantly enriched.
Raw spectral files and intermediary identification search results were uploaded to the MassIVE repository (part of the ProteomeXchange consortium). Data can be accessed using the MassIVE accession number MSV000090133 or the ProteomeXchange accession number PXD036059. The peptide and protein identification files from MaxQuant, together with sample annotation, code for pre-processing, reproducible reports and visualizations for the analysis of protein abundance association with time, are openly accessible via Zenodo and GitHub (Cosenza-Contreras, 2022; https://doi.org/10.5281/zenodo.7007047 ) .
Experimental setup and proteome coverage The presented prospective study aims to investigate the suitability of autopsy specimens for proteome analyses with such samples having been harvested outside of RAPs, hence representing a more routine setting. The study focuses on three organs, i. e. the kidneys, the liver, and the lungs. The post-mortem interval between the time of death and the autopsy was determined as 6 h, 12 h, 18 h, 24 h, 48 h, 72 h, and 96 h. The storage temperature for each corpse was 4 °C. Tissue specimens for each time point was were collected from 3 to 4 individuals (Fig. ). General cohort characteristics are shown in Table . Tissue lysis and protein extraction were based on acid-labile surfactants and bead grinding; hence representing a rather prototypical workflow to ensure wide applicability. We have employed data-dependent acquisition (DDA) with label-free quantitation (LFQ). Overall, we have identified and quantified the following total protein numbers, when solely considering unique peptides: kidney: 2174 proteins, liver: 2290 proteins, lung: 1785 (Table ). As with most proteomic studies, not all proteins are simultaneously identified in all samples, resulting in some gaps. Further details on proteome coverage are provided in the subsequent sections. The proteome coverage of this study is somewhat limited. The instrumental setup is well suited to identify > 3500 proteins from cell-line derived samples (not shown). The decreased proteome coverage for the tissue-based samples might reflect sample-intrinsic properties, e.g. a less balanced proteome with more overshadowing components. Consequently, the present study is limited to post-mortem-alterations of abundant proteins, while the impact on less abundant proteins and post-translational modifications remains beyond its scope. By using MaxQuant and its LFQ algorithm, we utilize an approach that is robust and well-established . The instrument and measurement setup have been successfully used in previous studies employing label-free quantitative proteomics [ – ]. In preparation for using label-free quantitative proteomics with our setup, we have validated technical reproducibility. Consecutive LC–MS/MS measurements of the same sample (“technical replicates”) achieved correlation coefficients > 0.98 for the LFQ protein intensities (not shown). Moreover, we employed spiked-in retention normalization peptides (iRT) as a quality control measure . In the kidney data, nine peptides of the iRT panel were consistently identified and quantified; in the liver and lung data, this was the case for ten iRT peptides. Alteration of proteome coverage post mortem As initially outlined, the primary aim of the presented study is to investigate at which time point post-mortem, autopsy material has decayed to such an extent that it is no longer amenable to proteomic analysis. To this end, corpses were stored at 4 °C and autopsy time points of 6 h, 12 h, 18 h, 24 h, 48 h, 72 h, and 96 h after the patients’ demise were chosen to collect autopsy specimens of kidney, liver, and lung tissue; with n = 3 or 4 for each organ at each time point. The proteome coverage (three replicates per organ and time-point; a single LC–MS/MS runs per sample) per organ and time point is shown in Fig. . For the kidneys and the liver, we observe a slight decline of protein ID numbers upon prolonged time intervals. When setting 6 h post-mortem as the reference time point, a mild decrease in protein IDs emerges as being significant ( p < 0.05; two-tailed t -test) at 72 h and 96 h for kidney tissue and at each time point for the liver. For the lung tissue, the protein coverage appears rather constant. In the case of the liver, proteome coverage at 24 h still reaches > 90% of the proteome coverage at the reference time point of 6 h. We conclude that for kidney, lung, and—to a lesser extent—liver tissues, autopsy sampling at 24 h post-mortem allows for sufficient proteome coverage. Alteration of proteome composition post mortem Next, we set out to identify the quantity of significantly degraded proteins per organ and the optimal time point for post-mortem sampling. We have chosen 6 h sampling as the reference time point and employed linear models of microarray analysis (limma) as implemented in MSStats . Of note, the same amount of proteome per sample (300 ng) has been injected into the LC–MS/MS setup. Hence, our analysis captures alterations of proteome composition: while rapidly degrading proteins vanish, more stable proteins become more prevalent within the mixture, leading to a pseudo-enrichment despite the absence of widespread protein synthesis post-mortem . The pair-wise comparisons to the 6 h reference time point are shown in Fig. . Proteins with an adjusted p -value < 0.05 have been considered as either being depleted by degradation or as being more stable and hence being pseudo-enriched. For the kidneys and the liver, we have noticed very few significant proteome alterations at 18 h and 24 h while substantial and significant protein degradation becomes apparent at 48 h. For the lungs, substantial protein degradation only becomes apparent at 72 h and the proteome composition appears to be rather static for up to 48 h. The quantities of significantly degraded or pseudo-enriched proteins are summarized in Fig. . We conclude that for the kidneys and the liver, autopsy sampling for up to 24 h allows for quantitative proteome studies that are likely to be representative of the proteome biology at the time of death. For the lungs, this time point is up to 48 h. Of note, this approach focuses on abundance proteomics and does not include post-translational modifications which may present more dynamic alterations. As an additional overview of the proteomic data, we performed principal component analysis (PCA) for proteins that were ubiquitously quantified in all samples for either kidney, lung, or liver using MetaboAnalyst (Supplementary Fig. 1.). Missing values present a complication for approaches such as PCA or hierarchical clustering and typically require an imputation step. In our study, we observe a partially time-dependent increase in missingness. We refrained from imputation for PCA and focused on proteins that were ubiquitously quantified in all samples per organ. A clear segregation of samples with different (short or prolonged) post-mortem times does not become apparent by PCA. Identifying proteins with defined kinetics of degradation and pseudo-enrichment In addition to the previous statistical analysis, we have sought to identify proteins that present a defined (i.e. statistically significant) degradation or pseudo-enrichment during storage for up to 96 h at 4 °C. To this end, we have employed limma statistics with time as a continuous variable. In kidney tissue 402 (Supplementary Fig. 2A), in liver tissue 235 (Supplementary Fig. 2B), and in lung tissue ten (Supplementary Fig. 2C) proteins have been identified as reaching significance (adjusted p -value < 0.05) for a linear model with time as a continuous variable. Of note, linear models—as used in the present study—can only grasp those proteins whose degradation (or pseudo-enrichment) kinetics fits the proposed statistical model. We show this approach as a means to demonstrate that there is a systematic trends in post-mortem protein degradation. We do not imply that protein degradation generally follows a linear trend. The comparison of the identified proteins has demonstrated that 43 proteins of kidney and liver tissues have the same linear behavior. Six proteins have opposite behavior in each organ (Supplementary Tables 1–3). Within the kidney specimens O00468 (agrin), Q04917 (14–3-3 protein eta), and Q92499 (ATP-dependent RNA helicase DDX1) have shown pseudo-enrichment behavior (positive linear correlation) whereas in liver samples they exhibited degradation behavior (negative linear correlation). The proteins O43493, P10636 and P21796 have demonstrated degradation behavior (negative linear correlation) in kidney specimens and pseudo-enrichment behavior (positive linear correlation) in liver specimens. A pairwise comparison of kidney and lung tissues revealed four proteins with pseudo-enrichment behavior and one protein with degradation behavior. When liver and lung samples were compared, two identical proteins were identified that showed the same linear characteristic behavior (one of them decreased, the other one pseudo-enriched). (Supplementary Tables 1–3). We conclude that degradation kinetics appear to be organ specific. For each tissue specimens (kidney, liver and lung) the top proteins with the lowest adjusted p -value (≤ 0.05) have been selected and further analyses of emerging behavior patterns of their abundance over time (6 –96 h) after patients’ demise have been conducted. All degraded and pseudo-enriched proteins show similar behavior over time shown as an example in the top 9 proteins (Fig. ) of each specimen. Organ-Specific gene ontological patterns of degradation and pseudo–enrichment in kidney, liver, and lung tissues Based on the collection of proteins identified by pseudo-enrichment and degradation kinetics, an enrichment analysis has been applied to investigate their distribution by cellular components (Gene Ontology database) and structural motifs (Protein Family) as well as their role in biological processes. Only pseudo-enriched (i.e. stable) proteins have been found to show an over-representation of annotated biological terms. Regarding kidney tissue, these proteins are over-represented in terms associated with developmental processes (GO:BP), cytoskeletal proteins, and supramolecular complexes such as proteasomal units (GO:CC). Pseudo-enrichment of kidney proteins with various structural motifs has also been identified, with the most significant enrichment of coiled-coil proteins. Pseudo-enriched proteins in liver tissue have been associated with organelle membrane and cell projection cellular components. In lung tissue, pseudo-enriched proteins show significant enrichment of basement membrane and collagen trimmer cellular components, together with structural motifs associated with the EGF/laminin domain. In summary, no similar pattern of over-representation of biological motifs for pseudo-enriched or degraded proteins has been found among the studied tissues (Fig. ). Corroboration by isobaric labelling—based quantitative proteomics Isobaric labelling together with prefractionation provides a strategy for enhanced coverage in quantitative proteomics. We probed the samples representing 6 h, 24 h, 48 h, 96 h post-mortem using 16-plex isobaric labelling with separate labelling pools for the kidney, liver, and lung. These samples also represent different sampling areas as compared to the LFQ samples. Scarcity of material reduced the number of fractions yielded by prefractionation to less than six. This approach enhanced proteome coverage to 3818 proteins for the kidney, 3709 proteins for liver, and 3453 proteins for the lung tissue. As described for the LFQ data, we used limma statistics to determine the numbers of the degraded proteins per post-mortem time points compared to 6 h. In good agreement with the LFQ data, kidney stands out as the most affected tissue while the lung is comparably stable (Fig. ).
The presented prospective study aims to investigate the suitability of autopsy specimens for proteome analyses with such samples having been harvested outside of RAPs, hence representing a more routine setting. The study focuses on three organs, i. e. the kidneys, the liver, and the lungs. The post-mortem interval between the time of death and the autopsy was determined as 6 h, 12 h, 18 h, 24 h, 48 h, 72 h, and 96 h. The storage temperature for each corpse was 4 °C. Tissue specimens for each time point was were collected from 3 to 4 individuals (Fig. ). General cohort characteristics are shown in Table . Tissue lysis and protein extraction were based on acid-labile surfactants and bead grinding; hence representing a rather prototypical workflow to ensure wide applicability. We have employed data-dependent acquisition (DDA) with label-free quantitation (LFQ). Overall, we have identified and quantified the following total protein numbers, when solely considering unique peptides: kidney: 2174 proteins, liver: 2290 proteins, lung: 1785 (Table ). As with most proteomic studies, not all proteins are simultaneously identified in all samples, resulting in some gaps. Further details on proteome coverage are provided in the subsequent sections. The proteome coverage of this study is somewhat limited. The instrumental setup is well suited to identify > 3500 proteins from cell-line derived samples (not shown). The decreased proteome coverage for the tissue-based samples might reflect sample-intrinsic properties, e.g. a less balanced proteome with more overshadowing components. Consequently, the present study is limited to post-mortem-alterations of abundant proteins, while the impact on less abundant proteins and post-translational modifications remains beyond its scope. By using MaxQuant and its LFQ algorithm, we utilize an approach that is robust and well-established . The instrument and measurement setup have been successfully used in previous studies employing label-free quantitative proteomics [ – ]. In preparation for using label-free quantitative proteomics with our setup, we have validated technical reproducibility. Consecutive LC–MS/MS measurements of the same sample (“technical replicates”) achieved correlation coefficients > 0.98 for the LFQ protein intensities (not shown). Moreover, we employed spiked-in retention normalization peptides (iRT) as a quality control measure . In the kidney data, nine peptides of the iRT panel were consistently identified and quantified; in the liver and lung data, this was the case for ten iRT peptides.
As initially outlined, the primary aim of the presented study is to investigate at which time point post-mortem, autopsy material has decayed to such an extent that it is no longer amenable to proteomic analysis. To this end, corpses were stored at 4 °C and autopsy time points of 6 h, 12 h, 18 h, 24 h, 48 h, 72 h, and 96 h after the patients’ demise were chosen to collect autopsy specimens of kidney, liver, and lung tissue; with n = 3 or 4 for each organ at each time point. The proteome coverage (three replicates per organ and time-point; a single LC–MS/MS runs per sample) per organ and time point is shown in Fig. . For the kidneys and the liver, we observe a slight decline of protein ID numbers upon prolonged time intervals. When setting 6 h post-mortem as the reference time point, a mild decrease in protein IDs emerges as being significant ( p < 0.05; two-tailed t -test) at 72 h and 96 h for kidney tissue and at each time point for the liver. For the lung tissue, the protein coverage appears rather constant. In the case of the liver, proteome coverage at 24 h still reaches > 90% of the proteome coverage at the reference time point of 6 h. We conclude that for kidney, lung, and—to a lesser extent—liver tissues, autopsy sampling at 24 h post-mortem allows for sufficient proteome coverage.
Next, we set out to identify the quantity of significantly degraded proteins per organ and the optimal time point for post-mortem sampling. We have chosen 6 h sampling as the reference time point and employed linear models of microarray analysis (limma) as implemented in MSStats . Of note, the same amount of proteome per sample (300 ng) has been injected into the LC–MS/MS setup. Hence, our analysis captures alterations of proteome composition: while rapidly degrading proteins vanish, more stable proteins become more prevalent within the mixture, leading to a pseudo-enrichment despite the absence of widespread protein synthesis post-mortem . The pair-wise comparisons to the 6 h reference time point are shown in Fig. . Proteins with an adjusted p -value < 0.05 have been considered as either being depleted by degradation or as being more stable and hence being pseudo-enriched. For the kidneys and the liver, we have noticed very few significant proteome alterations at 18 h and 24 h while substantial and significant protein degradation becomes apparent at 48 h. For the lungs, substantial protein degradation only becomes apparent at 72 h and the proteome composition appears to be rather static for up to 48 h. The quantities of significantly degraded or pseudo-enriched proteins are summarized in Fig. . We conclude that for the kidneys and the liver, autopsy sampling for up to 24 h allows for quantitative proteome studies that are likely to be representative of the proteome biology at the time of death. For the lungs, this time point is up to 48 h. Of note, this approach focuses on abundance proteomics and does not include post-translational modifications which may present more dynamic alterations. As an additional overview of the proteomic data, we performed principal component analysis (PCA) for proteins that were ubiquitously quantified in all samples for either kidney, lung, or liver using MetaboAnalyst (Supplementary Fig. 1.). Missing values present a complication for approaches such as PCA or hierarchical clustering and typically require an imputation step. In our study, we observe a partially time-dependent increase in missingness. We refrained from imputation for PCA and focused on proteins that were ubiquitously quantified in all samples per organ. A clear segregation of samples with different (short or prolonged) post-mortem times does not become apparent by PCA.
In addition to the previous statistical analysis, we have sought to identify proteins that present a defined (i.e. statistically significant) degradation or pseudo-enrichment during storage for up to 96 h at 4 °C. To this end, we have employed limma statistics with time as a continuous variable. In kidney tissue 402 (Supplementary Fig. 2A), in liver tissue 235 (Supplementary Fig. 2B), and in lung tissue ten (Supplementary Fig. 2C) proteins have been identified as reaching significance (adjusted p -value < 0.05) for a linear model with time as a continuous variable. Of note, linear models—as used in the present study—can only grasp those proteins whose degradation (or pseudo-enrichment) kinetics fits the proposed statistical model. We show this approach as a means to demonstrate that there is a systematic trends in post-mortem protein degradation. We do not imply that protein degradation generally follows a linear trend. The comparison of the identified proteins has demonstrated that 43 proteins of kidney and liver tissues have the same linear behavior. Six proteins have opposite behavior in each organ (Supplementary Tables 1–3). Within the kidney specimens O00468 (agrin), Q04917 (14–3-3 protein eta), and Q92499 (ATP-dependent RNA helicase DDX1) have shown pseudo-enrichment behavior (positive linear correlation) whereas in liver samples they exhibited degradation behavior (negative linear correlation). The proteins O43493, P10636 and P21796 have demonstrated degradation behavior (negative linear correlation) in kidney specimens and pseudo-enrichment behavior (positive linear correlation) in liver specimens. A pairwise comparison of kidney and lung tissues revealed four proteins with pseudo-enrichment behavior and one protein with degradation behavior. When liver and lung samples were compared, two identical proteins were identified that showed the same linear characteristic behavior (one of them decreased, the other one pseudo-enriched). (Supplementary Tables 1–3). We conclude that degradation kinetics appear to be organ specific. For each tissue specimens (kidney, liver and lung) the top proteins with the lowest adjusted p -value (≤ 0.05) have been selected and further analyses of emerging behavior patterns of their abundance over time (6 –96 h) after patients’ demise have been conducted. All degraded and pseudo-enriched proteins show similar behavior over time shown as an example in the top 9 proteins (Fig. ) of each specimen.
Based on the collection of proteins identified by pseudo-enrichment and degradation kinetics, an enrichment analysis has been applied to investigate their distribution by cellular components (Gene Ontology database) and structural motifs (Protein Family) as well as their role in biological processes. Only pseudo-enriched (i.e. stable) proteins have been found to show an over-representation of annotated biological terms. Regarding kidney tissue, these proteins are over-represented in terms associated with developmental processes (GO:BP), cytoskeletal proteins, and supramolecular complexes such as proteasomal units (GO:CC). Pseudo-enrichment of kidney proteins with various structural motifs has also been identified, with the most significant enrichment of coiled-coil proteins. Pseudo-enriched proteins in liver tissue have been associated with organelle membrane and cell projection cellular components. In lung tissue, pseudo-enriched proteins show significant enrichment of basement membrane and collagen trimmer cellular components, together with structural motifs associated with the EGF/laminin domain. In summary, no similar pattern of over-representation of biological motifs for pseudo-enriched or degraded proteins has been found among the studied tissues (Fig. ).
Isobaric labelling together with prefractionation provides a strategy for enhanced coverage in quantitative proteomics. We probed the samples representing 6 h, 24 h, 48 h, 96 h post-mortem using 16-plex isobaric labelling with separate labelling pools for the kidney, liver, and lung. These samples also represent different sampling areas as compared to the LFQ samples. Scarcity of material reduced the number of fractions yielded by prefractionation to less than six. This approach enhanced proteome coverage to 3818 proteins for the kidney, 3709 proteins for liver, and 3453 proteins for the lung tissue. As described for the LFQ data, we used limma statistics to determine the numbers of the degraded proteins per post-mortem time points compared to 6 h. In good agreement with the LFQ data, kidney stands out as the most affected tissue while the lung is comparably stable (Fig. ).
The main objectives of our study have been to investigate the process of protein degradation after death using quantitative mass spectrometry and determine the time interval after death beyond which autolytic processes render tissue samples unsuitable for protein expression studies. In recent years, an autopsy has become a focus of interest for several reasons. One reason is that the currently available bioinformatics databases are mainly based on data obtained from surgical resection specimens. Due to the fact that most tumors are diagnosed at an inoperable high/advanced tumor stage the number of available tissue specimens is limited to core needle biopsies. Accordingly, rapid autopsy programs were primarily designed to obtain tissue samples from patients with advanced tumor stage. The longer the post-mortem interval, the greater the chance of degradation of the macromolecules of interest, so these programs describe the procedures specifically designed to retrieve fresh tissue from patients within a short time after death. Depending on the preferences and regulations of a given centre and the specific aspects of the individual case (pre-mortem conditions: e.g. inflammatory conditions, especially septic processes that may accelerate postmortem tissue degradation, or agonal factors such as prolonged hypoxia due to respiratory and/or cardiac failure, microcirculatory disturbance, shock, but febrile conditions or, conversely, death due to hypothermia may also affect postmortem decay; post-mortem conditions (e.g. external temperature, humidity)), the maximum acceptable post-mortem interval varies between 0.5 and 23 h . As rapid autopsy programs have focused primarily on creating conditions that satisfy RNA- and DNA-based molecular approaches, we have considered it important to clarify the issue of post-mortem interval length from a proteomic perspective. As the rate of degradation of macromolecules can vary greatly between different organs, tissue types and subtypes, and between different molecules, we first aimed to determine the kinetics of degradation in various organs using proteomic methods. Although RAP programs are becoming more widely available, it is mainly researchers who benefit from them. However, the diagnostic and pathogenetic questions of diseases that can be investigated by post-mortem sampling may be of interest to a much wider audience, and for the time being, autopsy case samples from routine pathological practice are available to answer these questions. Therefore, the aim of our study was to investigate the potential of protein expression analysis of autopsy tissue samples handled under routine conditions. To simulate these routine diagnostic circumstances, we have selected tissues of organs most commonly examined by post-mortem histology in our autopsy practice, and where the decomposition is neither too fast nor too slow, namely the lungs, the liver and the kidneys. Basically, we explored the issue through the following two approaches: a) post-mortem alteration of proteome coverage, b) post-mortem alteration of proteome composition. Recently, more and more innovative approaches to protein identification and quantification are becoming available that can provide deeper information, e.g. aptamers that also consider the structural features of proteins in their identification [ – ]. We have used LC–MS/MS because it has a solid knowledge base due to its wide use and compatibility with different sample types, including formalin fixation and paraffin embedding. LC–MS/MS technology is not primarily suited for spatially resolved probing of proteomic heterogeneity or for demonstrating proteome changes in different organ regions, and thus no attempt is made to extend it in this direction in the present study. Regarding the proteome coverage, a slight decline was found by the time for kidney and liver tissue, however, it was rather constant up to 72 h in the case of lung tissue. As the most significant changes were found in kidney and liver tissues after 24 h, and proteome coverage still reached > 90% of the 6-h value at 24 h, we conclude that 24 h post-mortem may be the maximum acceptable time interval beyond which samples are not suitable for quantitative protein expression studies. The post-mortem alteration of proteome composition was analyzed in our cohort by testing the same amount of total proteins at different post-mortem time points by LC–MS/MS. In this approach, candidates with decreasing amounts over time represent the degrading proteins, while proteins with pseudo-enrichment kinetics correspond to the (more or less) stable ones. Our proteome composition results also supported that substantial and significant protein degradation becomes apparent after 24 h in liver and kidney tissues and after 48 h in the lung tissue. Consequently, a longer post-mortem period may have a significant impact on the proteome composition (differential degradation), but sampling within 24 h may be appropriate, as degradation is within acceptable limits even in organs with faster autolysis. By analyzing organ-specific gene ontological patterns and identifying specific proteins that show statistically significant degradation or pseudo-enrichment, we can gain deeper insights into these processes. This may be of interest in terms of which proteins or protein families belong to the more degradable or stable subgroups. By identifying such patterns and revealing their similarities and differences between different organs, we can gain a better understanding of the degradation patterns that are well-known from macroscopic and microscopic autopsy pathology, e.g. differential and organ-specific autolysis. Moreover, by further refinement, such as microdissection sampling, it may be possible to answer later why, for example, in the renal cortex the proximal tubules are affected first, followed by the distal tubules and glomeruli. Thus, we have tried to classify these proteins according to the biological processes, cellular components (Gene Ontology database) and structural motifs (Protein Family). However, based on the databases listed, it was not possible to classify proteins that are significantly degradable in any of the three organs. Furthermore, it was not possible to classify liver proteins showing pseudo-enrichment according to these databases. However, enrichment analysis based on the GO Cellular Components database indicated an enrichment of liver proteins localized to organelle membranes, cell projections, and cilia. In the kidney, which is known to have very quickly degrading tissue components, we were able to find numerous pseudo-enriched proteins classified according to the biological processes, cellular components or structural motifs. Interestingly, pseudo-enriched kidney proteins were involved mainly in developmental and morphogenesis processes. More interestingly, classification by cellular components showed pseudo-enrichment of cytoskeletal and supramolecular complex proteins in the kidney tissue, including actin and cortical cytoskeleton proteins, intermediate filaments and microtubules, collagen network, etc., as well as catalytic proteins such as different metabolic and degrading enzymes (endonuclease, peptidases), and proteasome-related proteins. Since hypoxia and degradation at the cellular-subcellular level first affect the integrity of membranes and thus cause degradation of membrane-associated proteins and disorganization of membrane-bound subcellular compartments, the relative conservation and pseudo-enrichment of cytoskeletal proteins is in line with previous observations . Peptidases and the proteasome (a large complex that catalyzes the degradation of proteins) are factors involved in the degradation of proteins and may therefore play a key role in autolytic processes, therefore their pseudo-enrichment may explain the rapid decomposition of some renal tissue components. Classification by structural motifs (protein family) also revealed an enrichment of structural proteins such as cytoskeletal (non-muscle cytoplasmic myosin, spectrin) and basal lamina (laminin) proteins. These results suggest that proteins with similar post-mortem kinetics are not primarily shared in their biological functions but are more likely to be similar in their sensitivity to autolysis. The overrepresentation of protein families with analogous structural motifs in the kidney indicates that structural features may be another common factor in determining similar postmortem stability. Post-mortem proteomics is an emerging field, a similar potential application of this could be the determination of PMI. In the field of forensic pathology, some authors have attempted to estimate the time after death from the postmortem degradation of skeletal muscle proteins . Our results could help to potentially extend this to the examination of tissue samples from internal organs; however, the present study does not investigate the impact of different storage temperatures, which is highly relevant in the field of forensics. Similar studies on post-mortem proteome alterations have been performed in animal tissues [ , , ]. A prime example is the post-mortem aging of beef, which has been the focus of several proteomic studies . However, the primary interest of these has been in muscle proteins rather than proteins of organs such as the kidneys or the liver. Moreover, several studies have investigated the effect of PMI on DNA/RNA degradation kinetics, or DNA/RNA preservation, in human samples. However, despite a careful literature search, we are not aware of any proteomic investigation similar to the study we have performed. Several studies present disease-specific proteome alterations based on autopsy samples, but these studies have not evaluated the impact of PMI on proteome integrity [ , , ]. Some further limitations of the presented study also have to be mentioned. Firstly, the sample size per time-point is limiting. Another major point is, that the tissue specimens withdrawn from the autopsy at different time points are not from one, but from different deceased persons. Thus, it is possible that, although we have attempted to establish and adhere to standard conditions of adequacy for sampling, as we are dealing with different individuals, not only PMI but also individual variation may influence the differences in samples from different time points. It is also important to emphasize that our results only reflect the degradation of normal tissue components, not the neoplastic lesions, which may exhibit a more accelerated degradation process (due to tissue hypoxia and necrosis). Therefore, further studies will also be needed to validate the appropriateness of the 24-h maximum post-mortem period for tumor tissue sampling. Since we have found shorter acceptable PMI for the liver and kidney than for lung in terms of both proteome coverage and proteome composition. These results raise the question of whether the differences may be due to the anatomical location of the organs in addition to their different composition. In fact, the proximity of the colon to these two organs provides an opportunity for the transposition of bowel bacteria and thus an additional heterolytic factor increasing tissue decomposition. A previous study from Hurtado et al. has shown a significant increase in bacterial contamination—especially with the Enterobacteriaceae and Pseudomonas spp.- after 24 h postmortem . A further limiting factor may be that in some animal studies, mRNA expression has been found to be active in certain cell types up to 48 h post-mortem . This could theoretically add to the pseudo-enrichment of proteins by increasing it with additional (real) protein production, but no data on post-mortem translational activity are available, we hypothesise that it may not significantly affect our results—if it even exists. In conclusion, using proteomics as a clock instead of the post-mortem interval, we can see that this clock ticks at different rates in different tissues. Our post-mortem proteomic approach may pave the way to determine the time interval after death until the overall preservation of proteins allows the use of tissue samples for protein-based diagnostics or research, and to determine the same for specific proteins, protein groups or protein families to be tested.
Below is the link to the electronic supplementary material. Supplementary file1 (DOCX 986 KB)
|
Reproductive Scientist Development Program: Bridging the Gap to the Physician Scientist Career | 61ab5d56-06a3-4028-b859-c556bc6bef11 | 10075491 | Gynaecology[mh] | Physician-scientists trained in OBGYN are poised to address gaps in knowledge about health, disease, and delivery of patient care through research to ultimately advance reproductive science and promote women’s healthcare. As physicians, they have received medical and often subspecialty training in OBGYN, but most also receive formal and intensive scientific training to conduct independent investigation in the clinical and/or laboratory setting. This combination of expertise uniquely positions OBGYN physician-scientists to communicate knowledge across disciplines and to lead scientific teams or organizations to guide healthcare and important policy decisions. OBGYN scientists are typically academic medical faculty and often embody the missions of an academic medical center by performing research and providing clinical service. Traditionally, the OBGYN scientist has engaged in laboratory and clinical investigation, but to achieve a sustainable and diverse research infrastructure, this has since evolved to also include research on health services and implementation, population health, community engagement, and health equity. To successfully embark on and transition to a physician-scientist career, early-stage investigators require funding to support their income while maximizing their time for research productivity, prolonged training to obtain competency in clinical and scientific research, balance of work/life demands, mentors that can support and guide a young physician-scientist, and skills to manage the time-consuming and demanding requirements to maintain their clinical credentials. With the present waning National Institutes of Health (NIH) funding landscape and career opportunities available to support physician-scientist trainees, the OBGYN-scientist workforce has continued to diminish. This is a concerning trend that parallels global underinvestment in women’s health research and the progressively smaller pool of early-stage physician-scientists applying for NIH research support . OBGYN subspecialties that have traditionally attracted OBGYN scientists, including maternal–fetal medicine, reproductive endocrine infertility, and gynecology oncology, are also experiencing decreased interest from fellowship applicants to pursue an OBGYN scientist career . A long-term consequence of this is a subsequent decline in OBGYN-scientist mentors that can train the next generation of physician-scientists in obstetrics and gynecology. In response to this and to achieve their goal of improving the health of women, in 2019 the NIH Office of Research on Women’s Health led the development of a “Trans-NIH Plan for Women’s Health Research.” This plan is composed of five strategic goals, which include “advancing rigorous research that is relevant to the health of women” by “developing a well-trained diverse, and robust workforce.” Fostering the pipeline of OBGYN scientists promotes the engine of scientific progress and accelerates the transition of knowledge into improved healthcare for women. Thus, the development and effective implementation of innovative programs that attract, support, retain, and advance OBGYN physician-scientist careers will increase the diverse representation of women’s reproductive health research. To address this need, federal and foundation-funded career-development programs were created to develop the next generation of OBGYN physician-researchers, to promote science focused on the reproductive health of women, and to revitalize and strengthen the scientific base of academic OBGYN departments. The NIH K-series career development awards underwent an expansion between 1990s and 2017 to promote the careers of early-stage biomedical academic researchers . One of the K-series awards is the K12 RSDP, a federally funded program dedicated to promoting women’s health through the development of independent OBGYN scientific investigators. To assess the academic achievements and impact of the RSDP, we surveyed current and former RSDP scholars. Here we describe the history of the RSDP program, report our survey results, and discuss future steps needed to continue to strengthen the research training of OBGYN physicians in order to narrow the deficient gaps in the biomedical workforce. Career-development awards are largely comprised of NIH-funded K-awards, foundation awards, and project grants. The following is a list of career-development awards often pursued by OBGYN scientists. NIH-Funded K-Awards These are NIH grants awarded to the individual and consist of K08, K23, K99/R00, and the K12 RSDP (Table ). The RSDP is unique among most K12 funding programs. It is a national K12 award, administered centrally through the principal investigator’s home institution but granting individual awards to scholars throughout the USA and Canada. NIH-Funded Institutional K-Programs These K-type funding mechanisms differ from the RSDP in that they are not national but instead are institutional. The objective of these awards is to create a pool of junior physician investigators with expertise in women’s health and reproductive biology research. For these mechanisms, institutions have their own institutional K programs that may include a KL2 Clinical and Translational Science Award (CTSA), a K12 Women’s Reproductive Health Research (WRHR) award, and/or the K12 Building Interdisciplinary Research Career in Women’s Health (BIRCWH) award. Foundation Awards These include funding opportunities from the Society for Maternal–Fetal Medicine (sMFM), American Association of Obstetricians and Gynecologists Foundation (AAOGF), American Board of Obstetricians and Gynecologists (ABOG), American College of Obstetricians and Gynecologists (ACOG), American Society of Reproductive Medicine (ASRM), Gynecologic Oncologic Group Foundation (GOG), Foundation for Women’s Cancer, Mary Kay Foundation, Doris Duke Charitable Foundation, Foundation for Women’s Cancer, March of Dimes, Burroughs Welcome Fund (BWF), and the American Heart Association (AHA). The design of these awards is to support future academic physician leaders who seek additional science training in reproductive biology and women’s health. Project Grants Project grant opportunities are offered by the Preeclampsia Foundation, Thrasher Research Fund, Fetal Health Foundation, the Society of Reproductive Investigation (SRI), and the American Diabetic Association (ADA). These grants are often sums of money awarded to fund a specific project that contributes to an individual’s career trajectory but does not provide longer-term protected research time for a physician scientist’s career development. These are NIH grants awarded to the individual and consist of K08, K23, K99/R00, and the K12 RSDP (Table ). The RSDP is unique among most K12 funding programs. It is a national K12 award, administered centrally through the principal investigator’s home institution but granting individual awards to scholars throughout the USA and Canada. These K-type funding mechanisms differ from the RSDP in that they are not national but instead are institutional. The objective of these awards is to create a pool of junior physician investigators with expertise in women’s health and reproductive biology research. For these mechanisms, institutions have their own institutional K programs that may include a KL2 Clinical and Translational Science Award (CTSA), a K12 Women’s Reproductive Health Research (WRHR) award, and/or the K12 Building Interdisciplinary Research Career in Women’s Health (BIRCWH) award. These include funding opportunities from the Society for Maternal–Fetal Medicine (sMFM), American Association of Obstetricians and Gynecologists Foundation (AAOGF), American Board of Obstetricians and Gynecologists (ABOG), American College of Obstetricians and Gynecologists (ACOG), American Society of Reproductive Medicine (ASRM), Gynecologic Oncologic Group Foundation (GOG), Foundation for Women’s Cancer, Mary Kay Foundation, Doris Duke Charitable Foundation, Foundation for Women’s Cancer, March of Dimes, Burroughs Welcome Fund (BWF), and the American Heart Association (AHA). The design of these awards is to support future academic physician leaders who seek additional science training in reproductive biology and women’s health. Project grant opportunities are offered by the Preeclampsia Foundation, Thrasher Research Fund, Fetal Health Foundation, the Society of Reproductive Investigation (SRI), and the American Diabetic Association (ADA). These grants are often sums of money awarded to fund a specific project that contributes to an individual’s career trajectory but does not provide longer-term protected research time for a physician scientist’s career development. Program Overview The RSDP program was established in 1988 by Drs. Larry Longo and Robert Jaffe to train obstetrician-gynecologists committed to academic investigative careers in fundamental biomedical science . It was managed initially through the University of California, San Francisco (1988–2012) until it moved to Washington University in St. Louis under the direction of Dr. Kelle Moley (2013–2018) and subsequently to the University of Missouri under the oversight of Dr. Danny Schust (2019–present). The program is also uniquely supported by an assigned Fertility and Infertility Branch program officer from the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) of the National Institutes of Health (NIH); Dr. Lou DePaolo initially occupied this position until 2022, when Dr. Esther Eisenberg assumed this role. Overall, the RSDP comprised 8–12 present scholars overseen by executive, selection, and evaluation committee members. There are now over 100 past and present RSDP scholars that form a growing network of physician-scientists with shared experiences and broad interests. The RSDP provides career development support for obstetricians and gynecologists who are committed to a basic science career in academic medicine and research. The areas of interest covered by the program are broadly defined, with emphasis in cell and molecular biology as applied to problems in reproductive endocrinology, genetics, maternal–fetal medicine, oncology, immunology, infectious disease, or other aspects of reproduction. The RSDP program serves as a proven stepping stone for a successful career as a physician-scientist. It provides salary and research supply support, protected research time, a network of accomplished researchers in biomedical research, the opportunity for collaboration on future projects, and continuous scientific and career support to the RSDP scholar. Funding Support The program is supported by the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) of the NIH in collaboration with private agencies, professional societies, foundations, and private industry. Program Structure Initially, the RSDP was a 5-year program, but was redesigned as a 4-year program in 2018 with the intent to encourage physician-scientist scholars to achieve scientific independence earlier. The program consists of two phases, each with its own separate application process. Applications for phase 1 are similar to most career-development awards but include an initial letter of intent. If invited for full application, the submission process would include a career statement from the applicant, research plan, budget, letters of support, and resume. Phase I is a 2-year award presently requiring 100% protected time, during which the physician-scientist scholars pursue intensive basic research in the laboratory of a nationally recognized scientific leader, including those of Nobel laureates and members of the National Academy of Science. Note that in the most recent Request for Competitive Renewal Applications (RFA) for the RSDP, the time commitment for phase I of the RSDP is being changed to 75% protected time to align it with other K-type training mechanisms. Upon successful completion of a second competitive application submitted in year 2 of phase I, phase II scholars receive 2 years of additional support at 75% protected time for laboratory-based research. Moving forward, both phases will also include local institutional department sponsorship for additional teaching and/or clinical services, to which the scholars are able to devote up to 25% effort. By the completion of phase II, the goal is for the RSDP scholar to be competitive and, ideally, successful in securing funding from the NIH or other private and public foundations to continue their research program. With nearly 35 years of data available, we here assessed RSDP outcomes via a scholar survey. The RSDP program was established in 1988 by Drs. Larry Longo and Robert Jaffe to train obstetrician-gynecologists committed to academic investigative careers in fundamental biomedical science . It was managed initially through the University of California, San Francisco (1988–2012) until it moved to Washington University in St. Louis under the direction of Dr. Kelle Moley (2013–2018) and subsequently to the University of Missouri under the oversight of Dr. Danny Schust (2019–present). The program is also uniquely supported by an assigned Fertility and Infertility Branch program officer from the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) of the National Institutes of Health (NIH); Dr. Lou DePaolo initially occupied this position until 2022, when Dr. Esther Eisenberg assumed this role. Overall, the RSDP comprised 8–12 present scholars overseen by executive, selection, and evaluation committee members. There are now over 100 past and present RSDP scholars that form a growing network of physician-scientists with shared experiences and broad interests. The RSDP provides career development support for obstetricians and gynecologists who are committed to a basic science career in academic medicine and research. The areas of interest covered by the program are broadly defined, with emphasis in cell and molecular biology as applied to problems in reproductive endocrinology, genetics, maternal–fetal medicine, oncology, immunology, infectious disease, or other aspects of reproduction. The RSDP program serves as a proven stepping stone for a successful career as a physician-scientist. It provides salary and research supply support, protected research time, a network of accomplished researchers in biomedical research, the opportunity for collaboration on future projects, and continuous scientific and career support to the RSDP scholar. The program is supported by the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD) of the NIH in collaboration with private agencies, professional societies, foundations, and private industry. Initially, the RSDP was a 5-year program, but was redesigned as a 4-year program in 2018 with the intent to encourage physician-scientist scholars to achieve scientific independence earlier. The program consists of two phases, each with its own separate application process. Applications for phase 1 are similar to most career-development awards but include an initial letter of intent. If invited for full application, the submission process would include a career statement from the applicant, research plan, budget, letters of support, and resume. Phase I is a 2-year award presently requiring 100% protected time, during which the physician-scientist scholars pursue intensive basic research in the laboratory of a nationally recognized scientific leader, including those of Nobel laureates and members of the National Academy of Science. Note that in the most recent Request for Competitive Renewal Applications (RFA) for the RSDP, the time commitment for phase I of the RSDP is being changed to 75% protected time to align it with other K-type training mechanisms. Upon successful completion of a second competitive application submitted in year 2 of phase I, phase II scholars receive 2 years of additional support at 75% protected time for laboratory-based research. Moving forward, both phases will also include local institutional department sponsorship for additional teaching and/or clinical services, to which the scholars are able to devote up to 25% effort. By the completion of phase II, the goal is for the RSDP scholar to be competitive and, ideally, successful in securing funding from the NIH or other private and public foundations to continue their research program. With nearly 35 years of data available, we here assessed RSDP outcomes via a scholar survey. To assess the academic achievements and impact of RSDP, the RSDP program leaders developed an electronic 30-item survey to assess the demographics and career details of all current and former RSDP scholars. Survey Details The survey was created in Google Documents ( www.docs.google.com ) and distributed via email to all 114 RSDP members from 1988 to 2021. The survey link was sent 5 times via email or e-newsletter between May 5, 2021, and February 28, 2022. Participation was voluntary, and respondents were given the option to skip questions and/or end the survey at any time. There were no duplicate survey submissions. Statistical Analysis This study was deemed exempt by the University of Missouri’s Institutional Review Board. Characteristics (gender, specialty, field, geographic region where program was completed and of current position, whether a MD/PhD degree was held, current position, and current scholar status) of all RSDP scholars ( n = 114) were assessed overall and by 5-year cohorts. Characteristics of survey respondents were compared to nonrespondents using Pearson’s chi-squared or Fisher’s exact test. Descriptive statistics of the survey respondents ( n = 53) were determined using means and standard deviations for age upon entering the program (years) and length of RSDP support (years), medians and interquartile ranges for number of publications overall and directly resulting from the program, and proportions for current research areas. The number and proportion of successful NIH and Department of Defense (DOD) awards were calculated among RSDP alumni ( n = 103) and all scholars ( n = 114, including current scholars), as were the conversion time (years) from a K grant to any NIH grant and from a K grant to an R01 grant. The number and proportion of successful foundation grants were calculated among the survey respondents ( n = 53). Differences in any NIH funding, R01 funding, leadership position (defined as academic or medical director or a chair or dean at any level), and dean or chair position were assessed by scholar gender, subspecialty, and PhD attainment using Pearson’s chi-squared or Fisher’s exact test. All analyses were conducted in Stata 17.0 (StataCorp, College Station, TX). The survey was created in Google Documents ( www.docs.google.com ) and distributed via email to all 114 RSDP members from 1988 to 2021. The survey link was sent 5 times via email or e-newsletter between May 5, 2021, and February 28, 2022. Participation was voluntary, and respondents were given the option to skip questions and/or end the survey at any time. There were no duplicate survey submissions. This study was deemed exempt by the University of Missouri’s Institutional Review Board. Characteristics (gender, specialty, field, geographic region where program was completed and of current position, whether a MD/PhD degree was held, current position, and current scholar status) of all RSDP scholars ( n = 114) were assessed overall and by 5-year cohorts. Characteristics of survey respondents were compared to nonrespondents using Pearson’s chi-squared or Fisher’s exact test. Descriptive statistics of the survey respondents ( n = 53) were determined using means and standard deviations for age upon entering the program (years) and length of RSDP support (years), medians and interquartile ranges for number of publications overall and directly resulting from the program, and proportions for current research areas. The number and proportion of successful NIH and Department of Defense (DOD) awards were calculated among RSDP alumni ( n = 103) and all scholars ( n = 114, including current scholars), as were the conversion time (years) from a K grant to any NIH grant and from a K grant to an R01 grant. The number and proportion of successful foundation grants were calculated among the survey respondents ( n = 53). Differences in any NIH funding, R01 funding, leadership position (defined as academic or medical director or a chair or dean at any level), and dean or chair position were assessed by scholar gender, subspecialty, and PhD attainment using Pearson’s chi-squared or Fisher’s exact test. All analyses were conducted in Stata 17.0 (StataCorp, College Station, TX). Current and Former RSDP Scholar Characteristics The survey was emailed to 114 current and former RSDP scholars; 53 responded, for a response rate of 46%. Survey respondents were more likely to work in academia in the Northeast or Midwest USA, be a current RSDP scholar, and have a role as assistant professor or director (Supplemental Table ). At present, there are 11 active RSDP scholars. The characteristics of current and former RSDP scholars are described in Table . From 1988 to 1997, the program was comprised of a similar number of male (47%) and female (53%) scholars, but since 1998, the program has predominately been comprised of female scholars; in total, there have been 63% female and 37% male scholars. Two-thirds of RSDP recipients were doctors of medicine (MD) and a third of scholars held both MD and doctors of philosophy (PhD) degrees. With the recent redesign of the RSDP program from 5 to 4 years in length, there has been a proportionate decline in the number of RSDP scholars, which occurred equally in both males and females (Table ). Scholar Characteristics While Supported by the RSDP RSDP scholars who responded to the survey started their award at an average age of 34.7 (SD ± 2.8) years old (Table ) and were supported by the RSDP mechanism for an average of 3.8 years (SD ± 1.5). On average, each scholar generated a median of 8 (IQR: 3–12) publications directly from their RSDP work. Location and Subspecialty Training Institutions that scholars resided in during their RSDP training were largely located in the Northeast (36%) or West (31%) (Table ) . Similarly, post-training scholars are currently in the West (34%), Northeast (26%), Midwest (19%), and South (13%). Since the program’s inception, the highest proportion of RSDP scholars are subspecialists in maternal–fetal medicine (37%), reproductive endocrine infertility (32%), and gynecology oncology (21%). Other RSDP scholar subspecialties include genetics (3%), family planning (1%), and urogynecology (1%), with 6% of scholars practicing general OBGYN. Career Practice Pattern and Leadership Positions A large portion of RSDP scholars pursued an academic career (70%), with others in private practice (22%) and industry (6%) (Table ). Current and former RSDP scholars span from instructor (4%), assistant professor (19%), associate professor (21%), to full professor (25%). Many are in leadership roles, including academic director (32%), medical director (10%), department chair or vice chair (15%), and dean or associate dean or vice dean (3%) (Table ). Area of Research and External Research Funding At present, most scholars who responded to the survey engage in translational research (64%), with others conducting basic science (23%), patient-oriented (11%), or health services (4%) research. Some responding former scholars (6%) are not involved in research currently (Table ). Amongst former RSDP scholars, many have secured NIH (46%) (Table ) funding, and of the 53 survey respondents, half had achieved foundation (53%) funding (Table ). In addition to their RSDP K12 award, some scholars were also supported by other K12 mechanisms prior to or after their RSDP training, including the BIRCWH (5%) and WRHR (18%), as well as K08s (15%) (Table ). RSDP alumni have also successfully obtained funding through other federal mechanisms, including NIH R01 (29%), R21 (14%), U54 (9%), U01 (6%), P01 (3%), and P50 (3%), as well as DOD grant awards (3%). The scholar transition time from K-funding to an R01 mechanism is a median of 8 years (IQR: 6–12). The most common foundation funding achieved by RSDP scholars who responded to the survey was from the March of Dimes (17%), the American Society of Reproductive Medicine (11%), the Burroughs Wellcome Fund (9%), the Gates Foundation (6%), and the Gynecologic Oncology Group (4%) (Table ). There was no significant difference in NIH funding success rates by gender, OBGYN subspecialty, or educational degree. However, there was a trend of greater NIH funding success, including R01 funding, for RSDP scholars specializing in reproductive endocrine infertility (50%), gynecology oncology (46%), and maternal–fetal medicine (36%). At this time, 33 RSDP scholars are currently NIH funded as principal investigators. The survey was emailed to 114 current and former RSDP scholars; 53 responded, for a response rate of 46%. Survey respondents were more likely to work in academia in the Northeast or Midwest USA, be a current RSDP scholar, and have a role as assistant professor or director (Supplemental Table ). At present, there are 11 active RSDP scholars. The characteristics of current and former RSDP scholars are described in Table . From 1988 to 1997, the program was comprised of a similar number of male (47%) and female (53%) scholars, but since 1998, the program has predominately been comprised of female scholars; in total, there have been 63% female and 37% male scholars. Two-thirds of RSDP recipients were doctors of medicine (MD) and a third of scholars held both MD and doctors of philosophy (PhD) degrees. With the recent redesign of the RSDP program from 5 to 4 years in length, there has been a proportionate decline in the number of RSDP scholars, which occurred equally in both males and females (Table ). RSDP scholars who responded to the survey started their award at an average age of 34.7 (SD ± 2.8) years old (Table ) and were supported by the RSDP mechanism for an average of 3.8 years (SD ± 1.5). On average, each scholar generated a median of 8 (IQR: 3–12) publications directly from their RSDP work. Institutions that scholars resided in during their RSDP training were largely located in the Northeast (36%) or West (31%) (Table ) . Similarly, post-training scholars are currently in the West (34%), Northeast (26%), Midwest (19%), and South (13%). Since the program’s inception, the highest proportion of RSDP scholars are subspecialists in maternal–fetal medicine (37%), reproductive endocrine infertility (32%), and gynecology oncology (21%). Other RSDP scholar subspecialties include genetics (3%), family planning (1%), and urogynecology (1%), with 6% of scholars practicing general OBGYN. A large portion of RSDP scholars pursued an academic career (70%), with others in private practice (22%) and industry (6%) (Table ). Current and former RSDP scholars span from instructor (4%), assistant professor (19%), associate professor (21%), to full professor (25%). Many are in leadership roles, including academic director (32%), medical director (10%), department chair or vice chair (15%), and dean or associate dean or vice dean (3%) (Table ). At present, most scholars who responded to the survey engage in translational research (64%), with others conducting basic science (23%), patient-oriented (11%), or health services (4%) research. Some responding former scholars (6%) are not involved in research currently (Table ). Amongst former RSDP scholars, many have secured NIH (46%) (Table ) funding, and of the 53 survey respondents, half had achieved foundation (53%) funding (Table ). In addition to their RSDP K12 award, some scholars were also supported by other K12 mechanisms prior to or after their RSDP training, including the BIRCWH (5%) and WRHR (18%), as well as K08s (15%) (Table ). RSDP alumni have also successfully obtained funding through other federal mechanisms, including NIH R01 (29%), R21 (14%), U54 (9%), U01 (6%), P01 (3%), and P50 (3%), as well as DOD grant awards (3%). The scholar transition time from K-funding to an R01 mechanism is a median of 8 years (IQR: 6–12). The most common foundation funding achieved by RSDP scholars who responded to the survey was from the March of Dimes (17%), the American Society of Reproductive Medicine (11%), the Burroughs Wellcome Fund (9%), the Gates Foundation (6%), and the Gynecologic Oncology Group (4%) (Table ). There was no significant difference in NIH funding success rates by gender, OBGYN subspecialty, or educational degree. However, there was a trend of greater NIH funding success, including R01 funding, for RSDP scholars specializing in reproductive endocrine infertility (50%), gynecology oncology (46%), and maternal–fetal medicine (36%). At this time, 33 RSDP scholars are currently NIH funded as principal investigators. OBGYN physician-scientists are uniquely positioned to play a critical role in advancing biomedical research in the reproductive sciences. The background of an OBGYN physician-scientist facilitates the recognition of clinically relevant and novel research ideas and the ability to translate results from the bench to bedside. However, the physician-scientist workforce is diminishing as a result of challenges in achieving adequate funding, clinical demands, paucity of available mentorship, lack of institutional support, educational debt, lower wages compared to fully clinical colleagues, and difficulties with work-life balance . As physician-scientists are becoming endangered at the national level, more comprehensive and supportive research experiences need to be available to OBGYN residents, fellows, and junior faculty so that they can develop an interest in a research-focused career early . In our study of a 33-year-old NIH national K12 program for physician-scientists in OBGYN, gender, OBGYN subspecialty training, and educational degree did not impact an RSDP scholar’s ability to obtain NIH funding or leadership positions. Prior studies of other medical specialties noted gender disparities among early-career K-award-funded physician-scientists . RSDP scholars are mostly women, which likely reflects the recent demographics of the OBGYN field . Similar to a prior study by Okeigwe et al. , we also found having a PhD degree was not a prerequisite to achieving successful independent funding. However, prior NIH reports of K-award recipients have shown that those with an MD/PhD had higher rates of achieving R01 funding compared to those with an MD alone . Compared to K08 (32%) and K23 (18%) recipients from other medical specialties as well as OBGYN K08 (32%) and K23 (17%) recipients, RSDP scholars had comparable R01 success rates (29%) . In addition, the conversion time from RSDP K12 to an R01 (8 years) was comparable to published rates of conversion for K08 (6 years) and K23 (7 years) recipients . Among grant recipients, similar to the sMFM AAOGF award (53) , a 3-year program designed to support the training of OBGYN MD physician-scientists, the RSDP scholars had similar NIH-grant funding success rates (46%). This suggests that early career development programs like the RSDP benefit aspiring OBGYN physician-scientists. Older studies, published mostly 1 to 2 decades ago, also examined the early career outcomes of RSDP scholars and AAOGF fellows , especially in regards to the receipt of a faculty position and external research support, as these are two traditional indicators of a successful academic medicine career. In the prior cohorts examined, timing of data collection and variables of interest differed amongst studies, but the overall results of these studies were similar. Approximately 90–95% of former fellows were reported to be in faculty positions, and between 58 and 80% had successfully obtained external research support during their careers . A more recent study in 2005 by Pion et al. reported that amongst AAOGF fellows, 88% had active academic medicine careers and 22% had been awarded at least one R01 grant, but only 29% were currently NIH funded, likely reflecting the increasing challenges of maintaining a continuous NIH-funded research program . Our current study of RSDP alumni career outcomes aligns with the results of earlier evaluations, with the overwhelming majority (80%) having active academic medicine careers and 29% having achieved R01 funding. The RSDP is successful largely because the program encompasses many predictive factors of a successful transition to research independence by trainees on career development awards . These include fostering a unique and extensive collaborative network, providing rigorous local and national mentorship, and requiring strong institutional support and resources . The RSDP is a scientific community of current and former RSDP scholars who actively collaborate on multi-center studies, grants, and manuscripts as well as share biospecimens, samples, cell lines, data, and ideas. RSDP alumni are also dedicated to mentoring and training future generations of RSDP scholars and continue to play a dynamic role in the program after completing their K12 training. In addition, unlike many other physician-scientist training programs , it provides funding and a large amount of protected research time. Our study is not without limitations; we had a 46% response rate largely because the survey was conducted during the COVID-19 pandemic, when many of the institutions and hospitals employing current and former RSDP scholars were facing acute workforce challenges. Given the degree of continued participation of RSDP scholars and alumni at annual meetings and RSDP-sponsored activities and mentorship from RSDP alumni to current scholars, this lower response rate is unlikely a reflection of the personal connection felt towards the program. Additionally, most survey nonrespondents were older RSDP alumni who had information on file that was not current, and updated information was difficult to obtain. As older RSDP alumni are more likely to have successfully achieved research funding and academic leadership positions secondary to greater time in rank, lack of data on these past scholars most likely negatively biased our results on the RSDP’s success and impact. The declining numbers of physician-scientists, persistent underfunding of reproductive health research, and historic lack of prioritization of women’s health issues by the NIH have impeded the advancement of clinical OBGYN care. These are well-known concerns. To reverse this trend, it is important to assess and understand both the institutional and individual barriers and mechanisms of success to this career pathway. In addition, a greater focus on recruiting and retaining the next generation of OBGYN researchers is crucial. As such, it is critical to continue to recognize and invest in the success of national NIH K programs, such as the RSDP, to perpetuate the pipeline of OBGYN physician-scientists committed to women’s health research. Ours is the most current study to analyze the structure and impact of the RSDP. Prior studies focused on this subject were largely published over 1 to 2 decades ago . In summary, mentored early career development awards, like the RSDP, better position OBGYN physician-scientists to achieve independent research careers. This study demonstrates that the RSDP program has added to the research capability of academic OBGYN departments across the nation with its strong track record of former scholars attaining independent federal funding, pursuing academic careers, and achieving leadership positions. Our study validates that support for programs like the RSDP is critical to continue to improve the long-term recruitment and mentorship of OBGYN physician-scientists and to advance clinical care in OBGYN to better the health of women and their offspring. Below is the link to the electronic supplementary material. Supplementary file1 (DOCX 20 KB) |
Objective assessment for open surgical suturing training by finger tracking can discriminate novices from experts | f62e6510-de5c-4072-b49a-1de840918fe7 | 10075519 | Suturing[mh] | It is important to have effective surgical training devices that can be used regularly to maintain surgical skills. Tools to train surgical skills are investigated and developed frequently . Most of these training tools are used to train minimal invasive skills [ , , ]. For example, box trainers or virtual reality simulators are developed, which can be used to track laparoscopic instruments and some even provide the trainee with haptic feedback [ , , ]. New and more advanced techniques to practice minimal invasive surgical skills continue to develop, while devices to adequately train and assess open surgical skills fall behind. Several tools to enable the practice of open surgical procedures do exist, ranging from basic suturing pads to three-dimensional, layered models and even (live) animals . For all these tools, an expert is necessary to provide the trainee with feedback, which make their use less flexible. This factor makes this approach unsuitable for self-directed continuous training. Furthermore, those experts use observer-based tools, such as OSATS , GERT or UWOMSA , which are always more or less subjective. For adequate assessment a training tool with objective parameters is desired. An available technique with objective parameters that overcomes the need for expert observation, uses electromagnetic motion tracking to track hand movements during an open surgical procedure . Unfortunately, this system is very expensive and consequently not fitted for broad implementation in continuous training. In previous research, we showed that a low-cost tracking tool has potential to be used in open surgical simulated tasks for the tracking of finger movements . This technique has the potential to overcome the need for expert observation, while remaining cost-effective. Therefore, this assessment tool could be useful in the training of open surgical skills. This tracking software has been used in MIS training and has shown be able to discriminate between expertise levels (construct validity), however this has not been evaluated in the use for open surgical training. Prior to using this system as a training device, it is important to demonstrate a good construct validity. The aim of this study is to investigate the ability to discriminate between expertise levels of this low-cost, broadly accessible technique of finger tracking in simple open surgical suturing tasks.
Participants Surgeons, surgical residents and medical students at the Radboud University Medical Center, Nijmegen, The Netherlands, were recruited to participate in this study. Participants were included from September 2020 until September 2021. The medical students were included during the first week of their surgical rotation (internship) and assumed to be novices in surgical suturing skills, because they had no previous surgical rotation. The students had been taught basic suturing skills during education prior to their surgical internship, after which they were included, before any clinical exposure. Therefore, they had knowledge on how to perform the sutures needed for this study and did not need additional guidance during the study. Surgical residents and surgeons were perceived as suturing experts for the selected tasks. First, the participants completed an informed consent form and a short questionnaire, regarding their suturing experience. Novices were only included if they had no previous experience prior to their suturing education and experts had equal to or more than fifty sutures as previous experience. This was based on previous studies stating that expertise should be reached by 50 repetitions for most basic surgical procedures, such as laparoscopic cholecystectomy . By ensuring that the experts had done at least 50 repetitions on all different suturing skills that we evaluated in this study, we regard them as experienced. Participants agreed with anonymous processing of the collected data. A waiver for medical ethical approval was provided, because of the non-medical intervention setup of this study. Equipment To objectively track participant’s finger movements, a tablet with tracking software and an open surgical simulator were used. show a setup of the used materials. To track the participants’ finger movements, SurgTrac software (eoSurgical Ltd., Edinburgh, United Kingdom) was used as a tablet application. The tablet was placed in a stand above the right shoulder of the participant. The distance between the camera and the simulator was set at sixty centimeters to have an adequate overview. SurgTrac software has been developed to track minimally invasive surgical instruments, tagged with a blue and red sticker, in a simulated setting. The software recorded 30 frames per second for accurate constant motion tracking . Recent research has shown that it is possible to use SurgTrac software for finger tracking in open surgical simulation . Participants wore white surgical gloves with the right index finger tagged by a red balloon-tube and the left index finger tagged by a blue balloon-tube. A model by PediatrickBoxx (Nijmegen, The Netherlands) was used for a standardized simulation task of open surgery. The model consists of a wooden cast placed in a position of forty-five degrees angle to the table, with a suturing pad by EduStitch . This achieves an adequate view of the task execution through the camera. Tasks All participants executed four suturing tasks on the given equipment in the following order: Knot tying by hand ( ): participants tied a reef knot consisting of an underhand and overhand throw. Transcutaneous suturing and knot tying with instruments ( ): participants executed one transcutaneous suture on the incision of their pad and tied the suture using their instruments. Vertical mattress suturing (‘Donati’ suture) and knot tying with instruments ( ): participants executed one vertical mattress suture on the incision of their pad and tied the suture using their instruments. Continuous intracutaneous suturing without knot tying ( ): an intracutaneous knot was tied in advance by the researcher at the upper side of the incision. Participants made an intracutaneous suture through the total incision in their pad (four centimeters in length). No intracutaneous or extracutaneous knot was done at the end of the suture Outcomes The SurgTrac software on the tablet tracked the red and blue tag on the index fingers of the participants during the execution of the suturing tasks. The parameters that the SurgTrac software measures are time of executing a task (in seconds), distance traveled by the left and right tag (in meters), distance between hands (average distance between the red and blue tag in centimeters), hands off-screen (in percentage of time), speed (mean speed of right or left hand, in millimeters/second), acceleration (mean acceleration of right or left hand, in millimeters/second , smoothness (mean smoothness of motion of right or left hand, in millimeters/second and handedness (percentage right- and left-hand usage). These parameters, as measured by SurgTrac have been frequently validated for training of minimally invasive skills [ , , ] especially time and distance. Because no other data for open surgery and finger-tracking is available for this software, the proven valid parameters in MIS skills, namely time and distance serve as the primary outcome parameters for the validation of finger-tracking in this open surgical suturing tasks study. The other parameters provided by the software will serve as secondary outcome parameters. All parameters will be included in the construct validation process, barring the parameter off-screen, because it lacks clinical relevance in open surgery (no use of a screen). The parameters are measured separately for the red and blue tag by SurgTrac. The combined total score of the red and blue tag for the parameter distance, consisted of distance travelled by a participant’s right and left hand in total. The combined total scores of the parameters speed, acceleration and smoothness were analysed as a mean score for right and left hand in total. The score of handedness is described as difference between percentage usage of the right hand and percentage usage of the left hand. Statistical analyses Data was analyzed by IBM Statistical Package for Social Sciences (SPSS), version 25. A p-value of<0.05 was assumed statistically significant. To compare outcomes of the novices and expert group, Mann Whitney U tests were used in case of data with non-normal distribution. All other data were represented as descriptive statistics. For a desired power of 0.80 with a power rate of 0.05 a sample size of 19 participants is required. This was based on the expected and clinically relevant differences in time to complete the task. For basic open suturing tasks we assumed 45 seconds as a clinically relevant difference. However, a large variation in results was expected in the novices group. Therefore, more novices were included, to overcome this problem and strengthen the data.
Surgeons, surgical residents and medical students at the Radboud University Medical Center, Nijmegen, The Netherlands, were recruited to participate in this study. Participants were included from September 2020 until September 2021. The medical students were included during the first week of their surgical rotation (internship) and assumed to be novices in surgical suturing skills, because they had no previous surgical rotation. The students had been taught basic suturing skills during education prior to their surgical internship, after which they were included, before any clinical exposure. Therefore, they had knowledge on how to perform the sutures needed for this study and did not need additional guidance during the study. Surgical residents and surgeons were perceived as suturing experts for the selected tasks. First, the participants completed an informed consent form and a short questionnaire, regarding their suturing experience. Novices were only included if they had no previous experience prior to their suturing education and experts had equal to or more than fifty sutures as previous experience. This was based on previous studies stating that expertise should be reached by 50 repetitions for most basic surgical procedures, such as laparoscopic cholecystectomy . By ensuring that the experts had done at least 50 repetitions on all different suturing skills that we evaluated in this study, we regard them as experienced. Participants agreed with anonymous processing of the collected data. A waiver for medical ethical approval was provided, because of the non-medical intervention setup of this study. Equipment To objectively track participant’s finger movements, a tablet with tracking software and an open surgical simulator were used. show a setup of the used materials. To track the participants’ finger movements, SurgTrac software (eoSurgical Ltd., Edinburgh, United Kingdom) was used as a tablet application. The tablet was placed in a stand above the right shoulder of the participant. The distance between the camera and the simulator was set at sixty centimeters to have an adequate overview. SurgTrac software has been developed to track minimally invasive surgical instruments, tagged with a blue and red sticker, in a simulated setting. The software recorded 30 frames per second for accurate constant motion tracking . Recent research has shown that it is possible to use SurgTrac software for finger tracking in open surgical simulation . Participants wore white surgical gloves with the right index finger tagged by a red balloon-tube and the left index finger tagged by a blue balloon-tube. A model by PediatrickBoxx (Nijmegen, The Netherlands) was used for a standardized simulation task of open surgery. The model consists of a wooden cast placed in a position of forty-five degrees angle to the table, with a suturing pad by EduStitch . This achieves an adequate view of the task execution through the camera.
All participants executed four suturing tasks on the given equipment in the following order: Knot tying by hand ( ): participants tied a reef knot consisting of an underhand and overhand throw. Transcutaneous suturing and knot tying with instruments ( ): participants executed one transcutaneous suture on the incision of their pad and tied the suture using their instruments. Vertical mattress suturing (‘Donati’ suture) and knot tying with instruments ( ): participants executed one vertical mattress suture on the incision of their pad and tied the suture using their instruments. Continuous intracutaneous suturing without knot tying ( ): an intracutaneous knot was tied in advance by the researcher at the upper side of the incision. Participants made an intracutaneous suture through the total incision in their pad (four centimeters in length). No intracutaneous or extracutaneous knot was done at the end of the suture
The SurgTrac software on the tablet tracked the red and blue tag on the index fingers of the participants during the execution of the suturing tasks. The parameters that the SurgTrac software measures are time of executing a task (in seconds), distance traveled by the left and right tag (in meters), distance between hands (average distance between the red and blue tag in centimeters), hands off-screen (in percentage of time), speed (mean speed of right or left hand, in millimeters/second), acceleration (mean acceleration of right or left hand, in millimeters/second , smoothness (mean smoothness of motion of right or left hand, in millimeters/second and handedness (percentage right- and left-hand usage). These parameters, as measured by SurgTrac have been frequently validated for training of minimally invasive skills [ , , ] especially time and distance. Because no other data for open surgery and finger-tracking is available for this software, the proven valid parameters in MIS skills, namely time and distance serve as the primary outcome parameters for the validation of finger-tracking in this open surgical suturing tasks study. The other parameters provided by the software will serve as secondary outcome parameters. All parameters will be included in the construct validation process, barring the parameter off-screen, because it lacks clinical relevance in open surgery (no use of a screen). The parameters are measured separately for the red and blue tag by SurgTrac. The combined total score of the red and blue tag for the parameter distance, consisted of distance travelled by a participant’s right and left hand in total. The combined total scores of the parameters speed, acceleration and smoothness were analysed as a mean score for right and left hand in total. The score of handedness is described as difference between percentage usage of the right hand and percentage usage of the left hand.
Data was analyzed by IBM Statistical Package for Social Sciences (SPSS), version 25. A p-value of<0.05 was assumed statistically significant. To compare outcomes of the novices and expert group, Mann Whitney U tests were used in case of data with non-normal distribution. All other data were represented as descriptive statistics. For a desired power of 0.80 with a power rate of 0.05 a sample size of 19 participants is required. This was based on the expected and clinically relevant differences in time to complete the task. For basic open suturing tasks we assumed 45 seconds as a clinically relevant difference. However, a large variation in results was expected in the novices group. Therefore, more novices were included, to overcome this problem and strengthen the data.
Demographics 76 participants were included, 57 participants in the novice group and 19 participants in the expert group. The novice group consisted of medical students without any suturing experience, prior to their basic training, just before this study. The expert group comprised of six residents and thirteen surgeons, with varying, but at least three years of surgical experience. All the experts had executed fifty or more sutures in their medical career. The mean age was 23.8 years (SD 3.0) in the novice group, 43.9 years (SD 9.9) in the expert group and 28.9 years (SD 10.3) overall. The novice group consisted of fourteen male and forty-three female, where the expert group consisted of more male than female, namely fifteen male and four female participants. No significant differences in outcome parameters were found for gender within both groups. Construct validity shows the primary outcome parameters of the four executed tasks in both groups. Novices and experts differed significantly for the parameters time ( p < 0.001) and distance (<0.001< p < 0.041) in all four tasks. The differences between novices and experts for the primary outcome parameters are visualized in . In the secondary outcome parameters are shown. Of those, smoothness differed significantly ( p < 0.001) for all four tasks. Furthermore, for Task 2 and 3 there was a significant difference between novices and experts in handedness (0.001< p < 0.046). While Task 4 shows a significant difference in distance between the hands ( p = 0.015), in addition to the significant differences in time, distance and smoothness.
76 participants were included, 57 participants in the novice group and 19 participants in the expert group. The novice group consisted of medical students without any suturing experience, prior to their basic training, just before this study. The expert group comprised of six residents and thirteen surgeons, with varying, but at least three years of surgical experience. All the experts had executed fifty or more sutures in their medical career. The mean age was 23.8 years (SD 3.0) in the novice group, 43.9 years (SD 9.9) in the expert group and 28.9 years (SD 10.3) overall. The novice group consisted of fourteen male and forty-three female, where the expert group consisted of more male than female, namely fifteen male and four female participants. No significant differences in outcome parameters were found for gender within both groups.
shows the primary outcome parameters of the four executed tasks in both groups. Novices and experts differed significantly for the parameters time ( p < 0.001) and distance (<0.001< p < 0.041) in all four tasks. The differences between novices and experts for the primary outcome parameters are visualized in . In the secondary outcome parameters are shown. Of those, smoothness differed significantly ( p < 0.001) for all four tasks. Furthermore, for Task 2 and 3 there was a significant difference between novices and experts in handedness (0.001< p < 0.046). While Task 4 shows a significant difference in distance between the hands ( p = 0.015), in addition to the significant differences in time, distance and smoothness.
The SurgTrac software, used in this study for tracking index finger movements, was initially validated for tracking laparoscopic instruments during execution of minimally invasive surgical tasks . In a prior experimental study ( accepted manuscript ) the feasibility of tracking index finger movements using SurgTrac was confirmed. Therefore, this study focusses on the construct validation of this assessment method for open surgical training, to evaluate the ability of this assessment method to discriminate between novices and experts. All executed basic suturing tasks in this study showed a significant difference in the outcome parameters time, distance and smoothness when comparing novices to experts. For these parameters construct validity is established. For the other parameters measured by SurgTrac, such as handedness and distance between hands, construct validity could not be established. Although these latter two do not seem to have any clinical relevance in the assessment of the trainee’s skill level. Unequal distribution of using the right and left hand or the distance between both hands, does for itself not affect the quality of a suture and has no clinical relevance in the assessment of it. Smoothness, on the contrary, is an indicator of a participant’s instrument handling and a relevant indicator for the executioner’s level of basic suturing skills. Because the parameter smoothness is calculated using the parameters time and distance, it is debatable whether this parameter as such can be best used to differentiate between skills levels of the trainees. Nevertheless, this calculated parameter did show a significant difference between novices and experts, which means it can potentially be used by a trainee to monitor their own skills level. The current simulation and assessment setting with tracking finger movements can adequately discriminate between a novice or expert level for basic suturing skills. Tracking finger movements using SurgTrac while performing basic suturing skills on a simulator is therefore a promising setting to train and objectively assess basic open suturing skills. The aim of this study is to evaluate whether and which parameters of the SurgTrac application are able to discriminate novices from experts. This had been done in numerous studies as a construct validity study . We acknowledge that this is a basic form of validity testing, however, it is a first step towards the assessment of the true potential and capabilities of this tracking software in the use for open surgical training. The next step is a concurrent validity study, which we are currently doing, based on the results that we have from this study. After that a pass-fail cut of should be established, to be able to truly assess and inform the trainee whether they are proficient or not. The previously mentioned simulation setting is, unlike previous investigated settings, such as using sensory gloves with electromagnets , easily accessible and relatively affordable, because of the use of simple materials like gloves and balloon-tubes. Consequently, this setting is accessible for nearly every potential trainee in virtually every country, because the SurgTrac application can be downloaded to any smartphone or tablet. This improves the possibilities of medical students, residents and doctors to train and assess their skills independently. Furthermore, this tool is very compact, which makes it usable in every desired place, like home or work. Another advantage of this tool can be that no expert is needed for assessment, allowing a trainee to train at any suited time. A simulator for basic open suturing tasks combined with finger motion tracking, provides an easily accessible simulation setting, which is available for everyone to train anytime. This is not only to learn new surgical skills, but also to maintain the optimal level of skills, especially if there is no consistent exposure in the clinical setting. Because continuous training is a necessity to maintain surgical skills, as shown before in MIS studies on this subject . This study showed that time, distance and smoothness are valid parameters to assess technical aspects of performing a suture in open surgery. Yet, these parameters, cannot be extrapolated to assess quality aspects, such as firmness of the knot, distance between sutures or distance between the wound edges. Clinical outcome of any surgical task is the ultimate outcome parameter, therefore further research should focus on correlating clinically relevant parameters with quality of technical performance. Limitations No information was collected about the right- or left-handedness of participants. Analyses were performed using the combined total scores of the trainees right and left hand, in which dominance had no influence on the outcome parameters. Nevertheless, it would be interesting to further explore the differences between novices and experts taken hand dominance into account and analyse their dominant and non-dominant hand separately. Although not all parameters provided by the SurgTrac software seemed to be relevant for open surgery, these were all included in the study to avoid assumption bias. Due to the position of the tablet, and thus camera, with the tracking app, sometimes the fingers were not clearly in view, which could affect the outcome data. This was visible when looking at the out of view data and when this was too much, it was clear that the data could be less reliable. However, this did not seem to affect our results. When further validation studies have been performed, such as concurrent validity, where the results of this assessment method are compared to expert assessment, for example, the true potential will be evident. Based on this kind of studies a pass-fail score could be made, which trainees can use to evaluate their skills independently. However, it could also state that more parameters are needed for a true relevant assessment, which are not accounted for in this method.
No information was collected about the right- or left-handedness of participants. Analyses were performed using the combined total scores of the trainees right and left hand, in which dominance had no influence on the outcome parameters. Nevertheless, it would be interesting to further explore the differences between novices and experts taken hand dominance into account and analyse their dominant and non-dominant hand separately. Although not all parameters provided by the SurgTrac software seemed to be relevant for open surgery, these were all included in the study to avoid assumption bias. Due to the position of the tablet, and thus camera, with the tracking app, sometimes the fingers were not clearly in view, which could affect the outcome data. This was visible when looking at the out of view data and when this was too much, it was clear that the data could be less reliable. However, this did not seem to affect our results. When further validation studies have been performed, such as concurrent validity, where the results of this assessment method are compared to expert assessment, for example, the true potential will be evident. Based on this kind of studies a pass-fail score could be made, which trainees can use to evaluate their skills independently. However, it could also state that more parameters are needed for a true relevant assessment, which are not accounted for in this method.
Tracking index finger movements using SurgTrac software on a tablet, while executing basic suturing skills on a low-cost surgical suturing model, shows excellent construct validity for time, distance and motion smoothness in all four suturing tasks. This new open surgical assessment method can be implemented in any training setting, because it is easily set up as an application on any mobile device, making it a potent objective assessment tool in open surgical training.
|
Purulent Pericarditis Due to Pneumococcal Bacteremia Caused by Acupuncture: An Autopsy Case Report | 81065162-f9c7-456a-93db-ae86b07e3c92 | 10076128 | Forensic Medicine[mh] | Purulent pericarditis is an infection of the pericardial cavity that produces a purulent fluid, commonly caused by Streptococcus pneumoniae ( ). However, pericarditis caused by S. pneumoniae is commonly an intrapleural infection and is rarely caused by bacteremia ( ). Pneumococcal bacteremia seldom occurs due to skin and soft tissue infection ( ). We herein report an autopsy case of a patient who died as a result of purulent pericarditis caused by pneumococcal bacteremia owing to acupuncture-induced skin and soft tissue infection.
A 65-year-old man visited the emergency room with a 1-month history of left thigh and knee pain as well as a 5-day history of dyspnea and a fever. The patient had been receiving regular acupuncture treatment at his lower back and thighs for chronic pain every other month; the last acupuncture treatment had been performed one week prior to the onset of the pain of his left thigh. He denied any chest pain, cough, or sputum. Furthermore, the patient had been taking 4 mg/day of oral prednisolone for a few years to treat ureteral stricture due to retroperitoneal fibrosis, with which he had been diagnosed approximately 4 years prior to admission. Additional medical history included hypertension, hyperlipidemia, and chronic back pain, for which he had received amlodipine, telmisartan, pitavastatin, and indomethacin; he had not received the pneumococcal vaccine. On admission, the patient was conscious, his body temperature was 37.8°C, respiratory rate was 24 breaths/minute, SpO 2 was 90% (room air), blood pressure was 101/68 mmHg, pulse rate was 81 bpm, and a fever and hypoxemia were present. On a physical examination, a coarse crackle of the left lung was heard, but the heart sounds were clear without any abnormal heart murmur or pericardial friction. There was a fever in the left thigh and knee joint, with swelling and tenderness also present. The painful site on his left thigh coincided with the site where the patient had received acupuncture before the onset of the pain. Blood sampling data showed an elevated inflammatory response, with a white blood cell count of 22,500 /μL and a C-reactive protein level of 33.84 mg/dL ( ). We performed cardiocentesis and drained 280 mL of fluid. We also performed arthrocentesis of the patient's left knee and placed a drainage tube. Thoracentesis was not performed because it was difficult to perform due to a low pleural fluid volume. The pericardial effusion and joint fluid were both purulent, with a high white blood cell count and low glucose level ( ). On computed tomography (CT), a left thigh abscess contiguous to the left intra-articular knee abscess was observed ( ). In the chest, bilateral pleural effusions and pericardial effusion without pleural thickening were observed ( ). A chest radiograph showed an enlarged heart and bilateral pleural effusion; an electrocardiogram showed PQ elevation at the I, II, and aVR leads and ST elevation at the I, II, aVF, and precordial leads. Transthoracic echocardiographic findings showed no abnormalities in the left ventricular motion or moderate pericardial effusion. There were no findings of verrucous or valve destruction. We diagnosed the patient with purulent pericarditis, pyogenic arthritis of the left knee, and a left femoral muscle abscess. Therefore, we initiated antibiotic therapy with ceftriaxone and vancomycin. On the second day of admission, blood and joint fluid cultures grew S. pneumoniae ; however, culture of the pericardial effusion was negative. The diagnosis of invasive pneumococcal infection with pericarditis and arthritis was made, and sequential typing later revealed that the pneumococcal serotype was 15A (serotyping by Quellung capsule swelling using Statens Serum Institute antisera). On the same day, vancomycin was discontinued based on the results of the culture. On the fifth day of admission, liver enzymes were elevated, and the antibiotic was switched to penicillin G. On the ninth day of hospitalization, chest radiographs showed that cardiomegaly had improved, bilateral pleural effusions had almost disappeared, and no new lung lesions had appeared. In addition, the inflammatory response had improved according to the blood test results. However, no significant improvement was observed in the patient's dyspnea. Regarding the knee joint drainage, the amount of drained fluid gradually decreased, and the tube was removed on the 10th day of hospitalization. Furthermore, due to strong patient dissatisfaction with frequent penicillin G infusions, the antibiotic was changed to cefazolin on the 11th day of hospitalization. Despite the antibiotics and drainage, the fever and dyspnea worsened, and hypoxia appeared on the 13th day of hospitalization. We performed whole-body CT and found that pericardial effusion, pleural effusion, and left knee joint fluid reappeared. We considered local drainage to be inadequate and repeated pericardial effusion and left knee drainage the following day. We also administered diuretics for pleural effusion and hypoxia. As a result, his hypoxia and pleural effusion improved, but the dyspnea did not. On the 17th day, the patient suddenly experienced cardiopulmonary arrest, and despite resuscitation efforts, the patient unfortunately died. On the same day, at the request of the family, an autopsy was performed, which revealed massive purulent deposits in the epicardium and pericardial sac and partial adhesion of the epicardium and pericardial sac ( ). However, there was no valve destruction or verrucous adherence, suggesting a low probability of purulent pericarditis due to direct spillover from infective endocarditis. A microscopic examination showed marked inflammatory cell infiltration and granulation in the epicardium. This condition was considered similar to constrictive pericarditis. Microscopic findings showed lobular central hepatocyte atrophy and impaired reflux in other organs, suggesting that the direct cause of death was decreased cardiac output due to purulent pericarditis. There was no evidence of inflammation or inflammatory scarring in the pleura or lungs, and there was significant pulmonary congestion. The pleural effusion was culture-negative and transudative, suggesting that it was associated with cardiac failure. Autopsy findings were not suggestive of intrathoracic infection. Therefore, the development of pericarditis from the thoracic cavity was ruled out, and a bloodstream infection developing from left knee arthritis was considered as the cause, which is atypical. The clinical course during hospitalization is shown in .
We encountered a case of purulent pericarditis caused by S. pneumoniae that developed from a left femoral muscle abscess and pyogenic knee arthritis. There are two noteworthy points about this case: the autopsy results revealed that pneumococcal purulent pericarditis had been caused by a bloodstream infection rather than a thoracic cavity infection, and the pneumococcal bacteremia was thought to be caused by a skin and soft tissue infection associated with acupuncture treatment. Purulent pericarditis is an infection in the pericardial space that produces macroscopically or microscopically purulent fluid. There is also the possibility of pericardial stenosis and constrictive pericarditis after improvement. Therefore, surgical treatment, such as pericardiectomy, may be performed if the treatment of purulent pericarditis with pericardial drainage is deemed inadequate. To avoid invasive procedures, intrapericardial thrombolysis may also be performed ( ). However, even with treatment, the mortality rate in patients treated for bacterial endocarditis is 40%, mostly due to cardiac tamponade, toxicity, and constriction ( ). Staphylococcus aureus is the most common cause of purulent pericarditis, although S. pneumoniae is also a common cause ( ). There are several mechanisms by which patients develop purulent pericarditis, including contiguous spread from an intrathoracic site, hematogenous spread, extension from a myocardial site, perforating injury or surgery, and extension from a subdiaphragmatic site ( ). Pneumococcus is commonly associated with contiguous spread from an intrathoracic site, such as the lungs, while S. aureus is often involved in hematogenous spread ( ). The occurrence of pericarditis due to pneumococcal bacteremia is also rare, with a previous report indicating that only 3 out of 844 patients hospitalized for pneumococcal bacteremia developed pneumococcal pericarditis ( ). In the present case, there was no evidence of pleural enhancement suggestive of pleurisy on CT upon admission. In addition, there was no inflammation or inflammatory scarring suggestive of intrathoracic infection at the autopsy, although the possibility of modification due to antibiotic administration prior to the autopsy cannot be ruled out. In contrast, S. pneumoniae was detected in the knee joint fluid and blood. Considering the onset of dyspnea symptoms after the onset of left knee pain in the patient's history, it was likely that bacteremia triggered by the skin and soft tissue infection preceded purulent pericarditis, as described below. Therefore, the cause of purulent pericarditis was more likely to be bloodstream infection with S. pneumoniae than intrathoracic infection. There have been a few cases of S. pneumoniae associated with bacteremia without thoracic infection ( , , ). Pneumococcal purulent pericarditis, as in this case, can occur due to hematogenous infection even in the absence of pneumonia or pleurisy. Therefore, it is important to recognize that pneumococcal purulent pericarditis can occur even in patients with invasive pneumococcal infections without pulmonary involvement. Invasive pneumococcal infection, defined as an infection confirmed by the isolation of S. pneumoniae from a normally sterile site, is a highly lethal disease ( ). In our patient, S. pneumoniae was detected in the pericardial fluid, joint fluid, and bloodstream. The main focal diagnosis of pneumococcal bacteremia in adults is usually due to respiratory tract infections ( ), but there are occasional cases that are not associated with this focal infection ( ). Furthermore, pneumococcal arthritis in adults is often caused by bacteremia, but the primary focus is not identified in many cases. There is a hypothesis that pyogenic arthritis in such cases arises from joint seeding during transient bacteremia with a mucous membrane source ( ). Therefore, it is difficult to completely exclude the possibility that transient bacteremia was the cause of the knee arthritis, thigh muscle abscess, and purulent pericarditis in the present case. However, there are three reasons to believe that the bacteremia in our patient was likely caused by pneumococcal infection of the skin and soft tissues with skin entry. The first rationale is that cutaneous and soft tissue pneumococcal infections with skin entry are not common but are often reported ( ). Risks have been reported with underlying disease, the use of immunosuppressive drugs and local trauma, such as surgical wounds and burns ( ). Both risks were present in our patient: oral corticosteroid use and local trauma caused by acupuncture. The second rationale is the medical history of left thigh pain, which appeared after the acupuncture treatment; it was the initial symptom, followed by dyspnea. This may support a course of evolution from a femoral muscle abscess to pyogenic arthritis and then to pericarditis via bloodstream infection. The third rationale is the physical and imaging findings of a femoral muscle abscess in the same area where acupuncture had been performed several weeks prior and the fact that pyogenic knee arthritis was contiguous with the abscess. These findings support the hypothesis that acupuncture-induced thigh abscesses occurred first, followed by the development of pyogenic arthritis. Acupuncture has been used as an integrative or complementary therapy for pain ( ). However, it can carry the risk of skin and soft tissue infections; acupuncture site infections are mainly caused by inadequate skin disinfection, the use of contaminated needles, and poor adherence to hand hygiene by medical personnel ( ). S. aureus , Mycobacteriaceae , and Enterobacteriaceae pathogens are commonly reported as the causative agents of acupuncture site infections ( , ). S. pneumoniae rarely causes acupuncture site infections, and only one case has been reported in the literature ( ). It is necessary to recognize that S. pneumoniae may be the organism responsible for skin and soft tissue infections caused by trauma, such as acupuncture, in immunosuppressed patients. Conclusion Pneumococcal purulent pericarditis occasionally results from a bloodstream infection instead of a thoracic infection. If pneumococcal pericarditis is found, it may be necessary to confirm bacteremia in addition to confirming thoracic infection. Furthermore, pneumococcal skin and soft tissue infections can occur due to trauma, such as acupuncture, in immunosuppressed patients and can cause pneumococcal bacteremia. For this reason, a case of pneumococcal bacteremia may require confirmation of a skin and soft tissue infection.
Pneumococcal purulent pericarditis occasionally results from a bloodstream infection instead of a thoracic infection. If pneumococcal pericarditis is found, it may be necessary to confirm bacteremia in addition to confirming thoracic infection. Furthermore, pneumococcal skin and soft tissue infections can occur due to trauma, such as acupuncture, in immunosuppressed patients and can cause pneumococcal bacteremia. For this reason, a case of pneumococcal bacteremia may require confirmation of a skin and soft tissue infection.
|
Role of autopsy imaging in veterinary forensic medicine: experiences in 39 cases | ef8f2771-8905-49dd-ade7-0adf6b6b6e97 | 10076196 | Forensic Medicine[mh] | Cases A total of 39 suspicious dead animals were brought into the Veterinary Teaching Hospital, Azabu University from the local police department between 2016 and 2019. All procedures, including
diagnostic imaging and pathological examination, were performed in accordance with established guidelines at the Veterinary Teaching Hospital, Azabu University. The date, place, and
situation in which the cadavers were found were all anonymized based on the agreement between the university and police department. Methods Diagnostic imaging was performed using radiography, computed tomography (CT), or both before necropsy, with magnetic resonance imaging (MRI) being added as needed. Digital Imaging and
Communications in Medicine data were sent to a viewer (Newton OsiriX, Newton-Graphics, Sapporo, Japan) for evaluation of the images, which were then read by a university faculty radiologist.
After diagnostic imaging, pathological examination was performed, with drug testing being added as needed. Pathological examination was performed by a university faculty pathologist. As
pathological examination was performed based on Ai findings, the sensitivity between pathological and Ai results were not compared. The leading causes of death were comprehensively
determined based on Ai reports, pathological reports, and drug test results. Radiography Radiographs were obtained using an X-ray unit (VPX-40, Canon, Ohtawara, Japan) with the following parameters: 60–70 kV, 1.0–1.8 mAs up to animal size, and a 1.0 m film-focus distance.
Images were acquired using a digital radiography unit (Aero DR, Konica Minolta, Tokyo, Japan). CT CT images were obtained using a 16-detector row CT (BrightSpeed16, GE Healthcare, Port Washington, NY, USA). Imaging parameters were as follows: tube voltage, 120−135 kV; tube current,
200−400 mA; scanning slice thickness, 0.5−1.0 mm up to cases, and reconstructed bone, lung, and soft-tissue algorithms. Three-dimensional volume rendering CT images were generated using
Global Illumination software (Vitrea, Canon). MRI To obtain further information, MRI was performed in three cases using a MRI system with head quadrature radio frequency coil (Vantage Titan 3T, Canon), from which T1-, T2-, and T2*-weighted
images were obtained.
A total of 39 suspicious dead animals were brought into the Veterinary Teaching Hospital, Azabu University from the local police department between 2016 and 2019. All procedures, including
diagnostic imaging and pathological examination, were performed in accordance with established guidelines at the Veterinary Teaching Hospital, Azabu University. The date, place, and
situation in which the cadavers were found were all anonymized based on the agreement between the university and police department.
Diagnostic imaging was performed using radiography, computed tomography (CT), or both before necropsy, with magnetic resonance imaging (MRI) being added as needed. Digital Imaging and
Communications in Medicine data were sent to a viewer (Newton OsiriX, Newton-Graphics, Sapporo, Japan) for evaluation of the images, which were then read by a university faculty radiologist.
After diagnostic imaging, pathological examination was performed, with drug testing being added as needed. Pathological examination was performed by a university faculty pathologist. As
pathological examination was performed based on Ai findings, the sensitivity between pathological and Ai results were not compared. The leading causes of death were comprehensively
determined based on Ai reports, pathological reports, and drug test results.
Radiographs were obtained using an X-ray unit (VPX-40, Canon, Ohtawara, Japan) with the following parameters: 60–70 kV, 1.0–1.8 mAs up to animal size, and a 1.0 m film-focus distance.
Images were acquired using a digital radiography unit (Aero DR, Konica Minolta, Tokyo, Japan).
CT images were obtained using a 16-detector row CT (BrightSpeed16, GE Healthcare, Port Washington, NY, USA). Imaging parameters were as follows: tube voltage, 120−135 kV; tube current,
200−400 mA; scanning slice thickness, 0.5−1.0 mm up to cases, and reconstructed bone, lung, and soft-tissue algorithms. Three-dimensional volume rendering CT images were generated using
Global Illumination software (Vitrea, Canon).
To obtain further information, MRI was performed in three cases using a MRI system with head quadrature radio frequency coil (Vantage Titan 3T, Canon), from which T1-, T2-, and T2*-weighted
images were obtained.
Descriptions of all cases were shown in . Among the 39 cases examined, 28, 6, 3, 1, and 1 involved cats, dogs, rabbits, a ferret, and a pigeon, respectively. The major Ai findings included skull fracture, rib fracture,
subcutaneous emphysema, pneumothorax, pneumoperitoneum, diaphragmatic hernia, and abdominal rupture, respectively. The leading causes of death were traumatic impact, blood loss, poisoning,
suffocation, tension pneumothorax, starvation, and drowning. There were three cases with fading kitten syndrome, one with respiratory failure, and one with acute uremia. These five cases
determined to be non-criminal deaths. There were seven cases with an undetermined cause of death. Representative cases are shown below. Skull fracture and diaphragmatic hernia (Case 16) Based on its appearance, the cadaver was from a cat. Skull fracture, diaphragmatic hernia, rib fractures, subcutaneous emphysema, dilated digestive tract, and a fractured tail were observed
by radiography. Additionally, intracranial gas and bleeding, intramuscular bleeding, intraventricular gas, and pneumothorax were noted on CT/MRI. Intracranial bleeding or gas on MRI were
shown as magnetic susceptibility artifacts. These Ai findings, as well as brain liner contusion, were confirmed by necropsy. The skull fracture, rib fracture, and diaphragmatic hernia were
due to cause by traumatic impact. However, a traffic accident causes traumatic impact, tail dislocation and adjacent subcutaneous emphysema can be ruled out as accidental; the animal must
have been swung around by its tail. The leading cause of death was determined to be traumatic impact ( ). Rib fracture and retroperitoneal bleeding (Case 36) Based on its appearance, the cadaver was from a cat. Subcutaneous emphysema and pneumoperitoneum were observed on radiography. In addition, bilateral rib fracture, pneumothorax, abdominal
wall rupture, and retroperitoneal bleeding were identified on CT. These Ai findings were confirmed by necropsy. Subcutaneous bleeding, which was not detected on CT, was uncovered on
necropsy. Given that animals are covered with fur, bruises are often overlooked on appearance. The bilateral fracture is a strong indicator of abuse . The leading cause of death was a combination of traumatic impact and tension pneumothorax ( ). Poisoning (Case 21) Based on its appearance, the cadaver was from a cat. Radiography showed a pellet-form feed in a dilated esophagus. This case must have suffered from sudden death during eating. The feed,
which was stained blue, in the esophagus was confirmed during pathological examination. No other pathological findings were found, suggesting the possibly of poisoning. Carbamate pesticides,
which are widely used as insecticides in many countries where their availability has been associated with malicious animal poisonings , were detected
on drug testing. Hence, the leading cause of death was poisoning ( ). Starvation (Case 38) Based on its appearance, the cadaver was from a ferret. Its body weighed 220 g. Given that adult ferrets weigh around 1,000 g, malnutrition was obvious. A subcutaneous microtip was observed
on radiography and CT. Lack of serosal detail in the abdomen observed on radiography is indicative of emaciation due to loss of abdominal fat. Moreover, radiography and CT showed no food
inside the stomach and no feces in the colon. During in necropsy, the body was found to be skinny with no contents in the digestive tract, which means that this ferret had not been fed for a
long time. Given the lack of specific findings except for severe emaciation, the leading cause of death was most likely starvation due to neglect ( ). Undetermined (Case 32) Based on its appearance, the cadaver was from a dog that had been mummified, with its head having ossified. Pneumocephalus, which is a sign of autolysis , was detected on radiography. Pathological findings associated with death could not determine. As such, the leading cause of death remained undetermined. The body might have
been abandoned after death ( ).
Based on its appearance, the cadaver was from a cat. Skull fracture, diaphragmatic hernia, rib fractures, subcutaneous emphysema, dilated digestive tract, and a fractured tail were observed
by radiography. Additionally, intracranial gas and bleeding, intramuscular bleeding, intraventricular gas, and pneumothorax were noted on CT/MRI. Intracranial bleeding or gas on MRI were
shown as magnetic susceptibility artifacts. These Ai findings, as well as brain liner contusion, were confirmed by necropsy. The skull fracture, rib fracture, and diaphragmatic hernia were
due to cause by traumatic impact. However, a traffic accident causes traumatic impact, tail dislocation and adjacent subcutaneous emphysema can be ruled out as accidental; the animal must
have been swung around by its tail. The leading cause of death was determined to be traumatic impact ( ).
Based on its appearance, the cadaver was from a cat. Subcutaneous emphysema and pneumoperitoneum were observed on radiography. In addition, bilateral rib fracture, pneumothorax, abdominal
wall rupture, and retroperitoneal bleeding were identified on CT. These Ai findings were confirmed by necropsy. Subcutaneous bleeding, which was not detected on CT, was uncovered on
necropsy. Given that animals are covered with fur, bruises are often overlooked on appearance. The bilateral fracture is a strong indicator of abuse . The leading cause of death was a combination of traumatic impact and tension pneumothorax ( ).
Based on its appearance, the cadaver was from a cat. Radiography showed a pellet-form feed in a dilated esophagus. This case must have suffered from sudden death during eating. The feed,
which was stained blue, in the esophagus was confirmed during pathological examination. No other pathological findings were found, suggesting the possibly of poisoning. Carbamate pesticides,
which are widely used as insecticides in many countries where their availability has been associated with malicious animal poisonings , were detected
on drug testing. Hence, the leading cause of death was poisoning ( ).
Based on its appearance, the cadaver was from a ferret. Its body weighed 220 g. Given that adult ferrets weigh around 1,000 g, malnutrition was obvious. A subcutaneous microtip was observed
on radiography and CT. Lack of serosal detail in the abdomen observed on radiography is indicative of emaciation due to loss of abdominal fat. Moreover, radiography and CT showed no food
inside the stomach and no feces in the colon. During in necropsy, the body was found to be skinny with no contents in the digestive tract, which means that this ferret had not been fed for a
long time. Given the lack of specific findings except for severe emaciation, the leading cause of death was most likely starvation due to neglect ( ).
Based on its appearance, the cadaver was from a dog that had been mummified, with its head having ossified. Pneumocephalus, which is a sign of autolysis , was detected on radiography. Pathological findings associated with death could not determine. As such, the leading cause of death remained undetermined. The body might have
been abandoned after death ( ).
Among the 39 cases examined, 31 exhibited some abnormal Ai findings. The combination of Ai, pathological examination, and drug testing complementarily improved determining the leading to
cause of death that have been determined in 32 cases. Criminal animal abuse, which has been largely overlooked to date, can be detected through Ai. Additionally, Ai findings provide
pathologist with pre-necropsy information regarding pathologic location and guidance on areas of concentration and sampling prior to necropsy. Major Ai findings observed herein included skull and rib fractures, subcutaneous emphysema, pneumothorax, pneumoperitoneum, diaphragmatic hernia, and abdominal rupture. CT has the advantage
of detecting fractures involving the skull and ribs . Moreover, gases, which can easily escape during necropsy, are easily detected on CT. Gases can be
attributed to postmortem changes related to putrefaction [ , , ]. The
leading causes of death were traumatic impact, blood loss, poisoning, suffocation, tension pneumonia, starvation, and drowning. Traumatic impact consequently may have resulted in fractures,
diaphragmatic hernia, and abdominal rupture had occurred due to kicking or hitting. All such findings strongly suggested the involvement of animal abuse by humans. There are forensic case
reports of gunshots overseas [ , , ], we have not encountered any gunshot
cases throughout our experience given the strict gun control laws in Japan and the unfamiliarity of the Japanese population with guns. All eight cases of skull fractures and all five of poisoning, including suspected poisoning, were of cats. As the number of dogs and cats in Japan are almost equal, violence against cats may
occur more frequently than dogs. It was reported that cat deaths were two-fold higher than dogs . The fact of targets of animal abuse in cats could
be attributed to criminal psychology. One case (Case 38) exhibited a microtip, 8.2 mm in length and 1.4 mm in diameter, which is a practical tracing system. The microtip has a 15-digit identification number itself, which can be
scanned using a special device to provide information regarding the owner of the animal. The original purpose of a microtip was to identify the owner in cases where it had escaped. Setting a
microtip for cats and dogs had just become mandatory in Japan in 2022, allowing police to trace the cadavers to their owners and helping reduce the number of neglected/released pets. Animals have never been killed in the guise of suicide, which differs from human forensics. As such, the purpose of veterinary forensic Ai is to simply determine whether the death was
accidental or non-accidental, that is, whether animal abuse was positive or not. In the current study, ten fracture cases had been detected through CT, which is useful in judicial
investigations to rule out fractures . Given the short time needed for CT, it is readily available. And there are many CTs, which easily accessible
across veterinary hospitals in Japan. On the other hand, full necropsy examination including histology and toxicology takes a long time and intensive labor. Moreover, the number of certified
pathologists is small. The results of external examination and CT are sufficient to determine whether animal abuse has occurred in most cases. If screening CT confirms animal abuse,
pathological examination may not be necessary . If screening CT does not indicate animal abuse, the case would be further processed by pathological
examination and suspected poisoning should be sent for drug testing. This will reduce the number of cases requiring laborious pathological examination. We believe that using screening CT is a
more practical option in veterinary forensics and increase the number of Ai of suspicious dead animals. The increase numbers of Ai must be contributed to make accuse animal abuse. It is known that animal abuse can associated with various types of interpersonal violence, including child abuse, domestic violence and homicide [ , , ]. Malicious juvenile offenders often have an episode of animal abuse. That is, animal abuse may lead to
other crimes. Animal abuse might be an early indicator for violent behavior, such as child abuse, domestic violence, or psychotic homicide, therefore detecting animal abuse can prevent these
crimes. Although we do not have any criminal information outside of the cadavers examined in the current study and the issue of identifying the perpetrator of abuse is beyond the scope of
veterinary forensic medicine, detecting animal abuse provides useful information that could aid police in accusing animal abuse cases. Moreover, such information is useful not only for the
police but also in the field of criminal psychology. Therefore, veterinary forensics would contribute toward maintaining social safety networks. The use of Ai for veterinary forensics is still in its infancy in Japan. There are several limitations of Ai for veterinary forensics, including financial concerns, availability of
facilities, and manpower. Therefore, we would like to propose the establishment of infrastructures for veterinary forensics. One option is that the Ministry of the Environment and the National
Police Agency work together to establish vetAi centers at veterinary universities. Once the number of forensic cases accumulates at the vetAi center, veterinary forensics in Japan will make
great progress. The autopsy for identifying the cause of death in case of suspicious death in humans is legally enforced, while the necropsy for suspicious death in animals has not been
legitimated. At the least, Ai findings should be recorded for the court as an objective evidence of forensic suspicious animal abuse cadaver offered from police department. Administrative
cooperation is essential for implementing this. In addition, it is important to make strong relationship among administrative agency, police department, and veterinarians. Then, veterinarians
could play an important role to contribute toward the development of well-being society. In conclusion, Ai can be a valuable examination tool for veterinary forensic cases. Screening CT before necropsy is a more practical option for veterinary forensics.
The authors declare that there were no conflicts of interest.
|
Fifty years on: Serotonin and depression | d0cffa41-7d79-4c3d-bb32-c531279a20da | 10076339 | Pharmacology[mh] | Serotonin is synthesised from a precursor amino acid, tryptophan. Tryptophan is an essential amino acid which cannot be produced in the body and its availability to the brain from blood is a rate-limiting step in the production of brain serotonin. Sampling peripheral plasma tryptophan in humans is straightforward and three meta-analyses have concluded that plasma tryptophan is decreased in unmedicated depressed patients, with effect sizes increasing from Hedge’s g = 0.45 in all patients to 0.84 in unmedicated patients , a subsequent meta-analysis finding similar effect sizes in major depressive disorder (MDD) patients (Standardised Mean Difference (SMD) = −0.46) , and recent replication (SMD = −0.51) . Neither the role that diminished plasma tryptophan might play in the causation of depression nor the mechanism that underlies this abnormality is well understood. Current theories have implicated a role for inflammation and the induction of tryptophan metabolising enzymes such as indolamine 2,3-dioxygenase. This could lead to increased metabolism of tryptophan through the kynurenine pathway, potentially giving rise to neurotoxic metabolites such as quinolinic acid. However, whether the kynurenine pathway is activated in patients with major depression is open to question . Tryptophan depletion The critical role that tryptophan availability plays in brain serotonin synthesis has led to some ingenious techniques designed to lower brain serotonin levels acutely through dietary manipulation. The most widely used technique is ‘tryptophan depletion’ where administration of an amino acid load which lacks tryptophan lowers plasma tryptophan levels by about 80% over a few hours . In both animals and humans , this produces a significant decrease in brain serotonin synthesis, though the exact mechanism remains unclear. Nonetheless, tryptophan depletion offers a straightforward way to test the serotonin hypothesis of depression. It is well established that tryptophan depletion in healthy participants, who lack obvious risk factors for depression, does not cause significant lowering of mood . Therefore, the notion that decreasing brain serotonin levels is sufficient to cause depression can be rejected. However, in people recovered from depression and free of pharmacological and psychological treatment for long periods of time, tryptophan depletion produces a clinically significant lowering of mood. Such an effect is apparent in the review by and was reported in a highly cited meta-analysis, cited in the aforementioned ‘umbrella’ review . The effect size for mood symptoms in remitted antidepressant free depression patients was 1.9, from 8 samples, with removal of a potential outlier giving Hedge’s g = −1.06. From this, it seems reasonable to conclude that diminished serotonin levels in the brain are neither necessary nor sufficient to cause clinical depression. However, in those at risk of depression through suffering previous episodes, reductions in brain serotonin levels can lead to clinical relapse. Presumably here, decreased activity of serotonin pathways interacts with important pre-existing neurobiological vulnerabilities, probably in the regulation of key neural networks . The widespread neuromodulatory role of serotonin is likely to be relevant to this (see below). Serotonin-mediated neuroendocrine responses Prior to the advent of brain imaging, serotonin-mediated neuroendocrine responses were employed as a means of assessing the function of brain serotonin neurons involved in anterior pituitary hormone release. The most consistent data in depressed patients are from pharmacological challenges that increase serotonin activity and plasma prolactin levels by occupying the serotonin transporter on serotonin nerve terminals. Two agents have been employed for this purpose, clomipramine and citalopram. The literature reveals four studies, two with clomipramine and two with citalopram; all found a decreased prolactin response in unmedicated depressed patients . Similar findings were reported in a small study of depressed adolescents . It is possible that impaired prolactin release to challenge with serotonin reuptake inhibitors is a trait marker of vulnerability to depression because the abnormality apparently persists in patients recovered from depression and withdrawn from antidepressant medication . Serotonin imaging in the brain Positron emission tomography (PET) and single photon emission tomography (SPET) have allowed more direct assessment of serotonergic mechanisms in the human brain; however, the technical difficulty and expense of this work have limited the number of studies. The most investigated serotonin receptors in depressed patients are the 5-HT 1A receptor and the serotonin transporter. 5-HT 1A receptors are divided into two classes, dependent on location. First, they are located on soma and dendrites of serotonin neurons in the raphe nuclei in the brainstem, where they act as an inhibitory autoreceptor . Second, large numbers of 5-HT 1A receptors are also found post-synaptically (as heteroreceptors) in cortical and limbic regions. Most relevant PET studies have included depressed patients who were either drug naïve or drug free for long periods of time. The majority of studies have reported decreased 5-HT 1A receptor binding. A meta-analysis cited by Moncrieff et al. included 10 studies, half of which included mixed populations of people with bipolar disorder or postpartum depression. Most of these studies used BP ND , which only measures brain activity of ligand, and therefore does not require arterial blood sampling, measuring difference between target and reference region, assuming negligible activity in the reference region. All but one study (in postpartum depression) included people who were drug free. The meta-analysis reported a large reduction in 5-HT 1A receptors in mesiotemporal cortex with smaller decreases in other post-synaptic areas and the raphe nuclei . A problem with interpreting this literature has been the replicated finding by one group of increased 5-HT 1A receptor binding in unmedicated depressed patients across all brain regions . Specifically, Parsey et al. showed higher binding with an arterial input function (obviating need for a reference region) and, using similar methods to the studies cited in the aforementioned meta-analysis, lowered binding when grey matter cerebellum was used as a reference region. Their explanation was that binding in grey matter in patients compared to controls would account for these discrepant findings . Other studies in the meta-analysis that used white matter cerebellum as reference region were in people with bipolar depression, with discrepant findings, and therefore drawing conclusions is difficult. Binding could reflect a number of different causes, such as changes in receptor density or affinity. The advent of new PET ligands (most probably agonists; ) for the 5-HT 1A receptor may enable resolution of this issue. Until then, a degree of agnosticism seems appropriate concerning the status of brain 5-HT 1A receptors in depressed patients revealed by PET. Serotonin transporters In PET and SPET studies, the highest density of serotonin transporters is found in the brainstem and midbrain, where serotonin cell bodies are concentrated. Meta-analyses have shown reliable reductions in transporter binding in these brain regions as well as in the amygdala , the latter study also shows reductions in striatum. Decreases in transporter binding are seen in patients unmedicated for long periods of time and also those who are antidepressant naïve . The mechanism that underlies these reductions in serotonin transporter binding is not established, which may represent decreased activity or decreased numbers of serotonin neurons in people with depression. PET studies have potential to image neurotransmitter release in vivo, involving an appropriate pharmacological challenge to release endogenous neurotransmitter followed by measurement of the amount of ligand displaced from post-synaptic receptors. This approach has been used with success in the investigation of psychosis to show that acute psychotic symptoms are associated with increased dopamine release with amphetamine challenge in people with schizophrenia , dopamine depletion agents revealing increased receptor binding, indicating less endogenous dopamine . However, it has proved challenging to identify a serotonin receptor ligand that is readily displaceable by endogenous serotonin. Pharmacological manipulation studies have historically failed to demonstrate in vivo change in ligand binding for the 5 HT 1A receptor and SERT following tryptophan depletion , and a [( 11 )C]WAY-100635 PET study examining 5 HT 1A receptor binding with tryptophan infusion similarly failed to detect an effect (for detailed review, see ). Recently, however, it has been demonstrated that d-amphetamine administration produces sufficient serotonin release to displace the 5-HT 2A receptor ligand, [ 11 C]Cimbi-36 from frontal cortex, and a study using this technique found diminished serotonin release in 17 unmedicated depressed patients. . However, the sample included a number of people with Parkinson’s disease, and the control group was not matched for age (though the latter should not theoretically impact findings); though the findings were rightfully acknowledged as important, though preliminary. This study requires replication but demonstrates that as techniques advance it should be possible to assess the activity of serotonin neurons in depression with greater precision. Status of serotonin activity in depression The evidence summarised here shows that there are some reliable abnormalities in serotonin activity in unmedicated depressed patients. Currently, a simple synthesis of these abnormalities is elusive, but overall the evidence suggests a decrease in the activity of presynaptic serotonin neurons. This would be consistent with diminished tryptophan availability to the brain, impaired serotonin-mediated endocrine responses to presynaptic challenge, and lowered serotonin transporter binding on raphe cell bodies in the brainstem. Decreased serotonin release in the brain in response to amphetamine challenge, if replicated, would further support this view. While these changes in the serotonin activity in depressed patients are of interest, they do not necessarily represent causal mechanisms. However, the work with tryptophan depletion suggests that in some circumstances, particularly in patients who have experienced recurrent depression, low serotonin activity could be involved in predisposing to clinical relapse. This might be relevant to the role of maintenance antidepressant treatment in people at high risk of recurrent illness. Systems-level role of serotonin in cognition When assessing whether the serotonergic system is involved in the aetiology of depression, it is useful to first consider the role serotonin plays in how the brain processes information. In common with other neuromodulatory neurotransmitters, there are relatively few serotonergic neurons in the brain, their nuclei are concentrated in a small region of the brainstem, and they project widely throughout both the cortex and sub-cortical structures . This anatomical arrangement is well suited to the transmission of relatively simple, globally relevant messages across multiple regions of the brain. While the complexity of the serotonergic system is increased by a broad range of functionally distinct receptors and anatomically specific subnuclei projections , there has been continued interest in determining what the content of these messages might be. The paradigmatic example of this approach has examined dopaminergic neurons originating in the ventral tegmental area. The activity of approximately 50% of these dopaminergic neurons is well described by ‘reward prediction error’ signals , which carry the message ‘that was better than you thought it was going to be’ . To date, the message(s) conveyed by the serotonergic system have not been as convincingly characterised as they have for the dopaminergic system. Candidate serotonergic messages draw on the behavioural effect of serotonergic stimulation in experimental settings, which tends to lead to the withholding or inhibition of behaviour, and the observation that serotonergic neurons seem to be activated by both punishing and rewarding events . Thus, it has been suggested that serotonin carriers a ‘punishment prediction error’ signal (‘that was worse than you thought it was going to be’) , or possibly an estimate of the expected rate at which adversity will be encountered (‘It is risky to act at the moment’) . A second variant of message linked to serotonin function concerns how future events are evaluated and seeks to explain why increasing levels of serotonin cause a reduction in impulsivity. As a rule, rewards that are immediately available are preferred over those which are delayed and serotonin has been argued to reduce ‘delay-discounting’, the rate at which time erodes the value of rewarding events (‘it is worth waiting’) . Lastly, an intriguing suggestion frames the evaluative and prospective dimensions of the serotonergic message in terms of trains of thought. Specifically, it is suggested that when deciding what action to take it is useful to imagine both the immediate outcome of your actions and the outcomes of the subsequent actions that then become available. This process is akin to following the branches of a tree from the trunk to the tips, with each branch representing a series of sequential actions. In many cases, the number of future outcomes to be considered becomes extremely large very quickly (i.e. the tree often has many branches that divide often) making this a challenging task. Serotonin is argued to simplify the problem by halting the search along a specific branch as soon as an unwanted outcome is reached, in effect pruning the tree (‘don’t go there!’) . None of the proposed serotonergic messages enjoy the broad, cross-species empirical support of the dopaminergic reward prediction-error account and should therefore be considered as pointers to serotonin’s overall role in cognitive function rather than definitive descriptions. However, a common factor across all of the potential serotonergic messages is that they contain information about the estimated value of events, a quantity which, when measured, is disordered in depressed patients and is a core component of cognitive accounts of the illness. This observation clearly does not provide compelling evidence that disordered serotonergic function is a significant aetiological factor in the development or maintenance of depression; however, it does suggest an intermediate cognitive mechanism by which altering serotonergic function (e.g. via tryptophan depletion, treatment with serotonergic medication or pathology) might lead to changes in depressive symptoms. The critical role that tryptophan availability plays in brain serotonin synthesis has led to some ingenious techniques designed to lower brain serotonin levels acutely through dietary manipulation. The most widely used technique is ‘tryptophan depletion’ where administration of an amino acid load which lacks tryptophan lowers plasma tryptophan levels by about 80% over a few hours . In both animals and humans , this produces a significant decrease in brain serotonin synthesis, though the exact mechanism remains unclear. Nonetheless, tryptophan depletion offers a straightforward way to test the serotonin hypothesis of depression. It is well established that tryptophan depletion in healthy participants, who lack obvious risk factors for depression, does not cause significant lowering of mood . Therefore, the notion that decreasing brain serotonin levels is sufficient to cause depression can be rejected. However, in people recovered from depression and free of pharmacological and psychological treatment for long periods of time, tryptophan depletion produces a clinically significant lowering of mood. Such an effect is apparent in the review by and was reported in a highly cited meta-analysis, cited in the aforementioned ‘umbrella’ review . The effect size for mood symptoms in remitted antidepressant free depression patients was 1.9, from 8 samples, with removal of a potential outlier giving Hedge’s g = −1.06. From this, it seems reasonable to conclude that diminished serotonin levels in the brain are neither necessary nor sufficient to cause clinical depression. However, in those at risk of depression through suffering previous episodes, reductions in brain serotonin levels can lead to clinical relapse. Presumably here, decreased activity of serotonin pathways interacts with important pre-existing neurobiological vulnerabilities, probably in the regulation of key neural networks . The widespread neuromodulatory role of serotonin is likely to be relevant to this (see below). Prior to the advent of brain imaging, serotonin-mediated neuroendocrine responses were employed as a means of assessing the function of brain serotonin neurons involved in anterior pituitary hormone release. The most consistent data in depressed patients are from pharmacological challenges that increase serotonin activity and plasma prolactin levels by occupying the serotonin transporter on serotonin nerve terminals. Two agents have been employed for this purpose, clomipramine and citalopram. The literature reveals four studies, two with clomipramine and two with citalopram; all found a decreased prolactin response in unmedicated depressed patients . Similar findings were reported in a small study of depressed adolescents . It is possible that impaired prolactin release to challenge with serotonin reuptake inhibitors is a trait marker of vulnerability to depression because the abnormality apparently persists in patients recovered from depression and withdrawn from antidepressant medication . Positron emission tomography (PET) and single photon emission tomography (SPET) have allowed more direct assessment of serotonergic mechanisms in the human brain; however, the technical difficulty and expense of this work have limited the number of studies. The most investigated serotonin receptors in depressed patients are the 5-HT 1A receptor and the serotonin transporter. 5-HT 1A receptors are divided into two classes, dependent on location. First, they are located on soma and dendrites of serotonin neurons in the raphe nuclei in the brainstem, where they act as an inhibitory autoreceptor . Second, large numbers of 5-HT 1A receptors are also found post-synaptically (as heteroreceptors) in cortical and limbic regions. Most relevant PET studies have included depressed patients who were either drug naïve or drug free for long periods of time. The majority of studies have reported decreased 5-HT 1A receptor binding. A meta-analysis cited by Moncrieff et al. included 10 studies, half of which included mixed populations of people with bipolar disorder or postpartum depression. Most of these studies used BP ND , which only measures brain activity of ligand, and therefore does not require arterial blood sampling, measuring difference between target and reference region, assuming negligible activity in the reference region. All but one study (in postpartum depression) included people who were drug free. The meta-analysis reported a large reduction in 5-HT 1A receptors in mesiotemporal cortex with smaller decreases in other post-synaptic areas and the raphe nuclei . A problem with interpreting this literature has been the replicated finding by one group of increased 5-HT 1A receptor binding in unmedicated depressed patients across all brain regions . Specifically, Parsey et al. showed higher binding with an arterial input function (obviating need for a reference region) and, using similar methods to the studies cited in the aforementioned meta-analysis, lowered binding when grey matter cerebellum was used as a reference region. Their explanation was that binding in grey matter in patients compared to controls would account for these discrepant findings . Other studies in the meta-analysis that used white matter cerebellum as reference region were in people with bipolar depression, with discrepant findings, and therefore drawing conclusions is difficult. Binding could reflect a number of different causes, such as changes in receptor density or affinity. The advent of new PET ligands (most probably agonists; ) for the 5-HT 1A receptor may enable resolution of this issue. Until then, a degree of agnosticism seems appropriate concerning the status of brain 5-HT 1A receptors in depressed patients revealed by PET. Serotonin transporters In PET and SPET studies, the highest density of serotonin transporters is found in the brainstem and midbrain, where serotonin cell bodies are concentrated. Meta-analyses have shown reliable reductions in transporter binding in these brain regions as well as in the amygdala , the latter study also shows reductions in striatum. Decreases in transporter binding are seen in patients unmedicated for long periods of time and also those who are antidepressant naïve . The mechanism that underlies these reductions in serotonin transporter binding is not established, which may represent decreased activity or decreased numbers of serotonin neurons in people with depression. PET studies have potential to image neurotransmitter release in vivo, involving an appropriate pharmacological challenge to release endogenous neurotransmitter followed by measurement of the amount of ligand displaced from post-synaptic receptors. This approach has been used with success in the investigation of psychosis to show that acute psychotic symptoms are associated with increased dopamine release with amphetamine challenge in people with schizophrenia , dopamine depletion agents revealing increased receptor binding, indicating less endogenous dopamine . However, it has proved challenging to identify a serotonin receptor ligand that is readily displaceable by endogenous serotonin. Pharmacological manipulation studies have historically failed to demonstrate in vivo change in ligand binding for the 5 HT 1A receptor and SERT following tryptophan depletion , and a [( 11 )C]WAY-100635 PET study examining 5 HT 1A receptor binding with tryptophan infusion similarly failed to detect an effect (for detailed review, see ). Recently, however, it has been demonstrated that d-amphetamine administration produces sufficient serotonin release to displace the 5-HT 2A receptor ligand, [ 11 C]Cimbi-36 from frontal cortex, and a study using this technique found diminished serotonin release in 17 unmedicated depressed patients. . However, the sample included a number of people with Parkinson’s disease, and the control group was not matched for age (though the latter should not theoretically impact findings); though the findings were rightfully acknowledged as important, though preliminary. This study requires replication but demonstrates that as techniques advance it should be possible to assess the activity of serotonin neurons in depression with greater precision. In PET and SPET studies, the highest density of serotonin transporters is found in the brainstem and midbrain, where serotonin cell bodies are concentrated. Meta-analyses have shown reliable reductions in transporter binding in these brain regions as well as in the amygdala , the latter study also shows reductions in striatum. Decreases in transporter binding are seen in patients unmedicated for long periods of time and also those who are antidepressant naïve . The mechanism that underlies these reductions in serotonin transporter binding is not established, which may represent decreased activity or decreased numbers of serotonin neurons in people with depression. PET studies have potential to image neurotransmitter release in vivo, involving an appropriate pharmacological challenge to release endogenous neurotransmitter followed by measurement of the amount of ligand displaced from post-synaptic receptors. This approach has been used with success in the investigation of psychosis to show that acute psychotic symptoms are associated with increased dopamine release with amphetamine challenge in people with schizophrenia , dopamine depletion agents revealing increased receptor binding, indicating less endogenous dopamine . However, it has proved challenging to identify a serotonin receptor ligand that is readily displaceable by endogenous serotonin. Pharmacological manipulation studies have historically failed to demonstrate in vivo change in ligand binding for the 5 HT 1A receptor and SERT following tryptophan depletion , and a [( 11 )C]WAY-100635 PET study examining 5 HT 1A receptor binding with tryptophan infusion similarly failed to detect an effect (for detailed review, see ). Recently, however, it has been demonstrated that d-amphetamine administration produces sufficient serotonin release to displace the 5-HT 2A receptor ligand, [ 11 C]Cimbi-36 from frontal cortex, and a study using this technique found diminished serotonin release in 17 unmedicated depressed patients. . However, the sample included a number of people with Parkinson’s disease, and the control group was not matched for age (though the latter should not theoretically impact findings); though the findings were rightfully acknowledged as important, though preliminary. This study requires replication but demonstrates that as techniques advance it should be possible to assess the activity of serotonin neurons in depression with greater precision. The evidence summarised here shows that there are some reliable abnormalities in serotonin activity in unmedicated depressed patients. Currently, a simple synthesis of these abnormalities is elusive, but overall the evidence suggests a decrease in the activity of presynaptic serotonin neurons. This would be consistent with diminished tryptophan availability to the brain, impaired serotonin-mediated endocrine responses to presynaptic challenge, and lowered serotonin transporter binding on raphe cell bodies in the brainstem. Decreased serotonin release in the brain in response to amphetamine challenge, if replicated, would further support this view. While these changes in the serotonin activity in depressed patients are of interest, they do not necessarily represent causal mechanisms. However, the work with tryptophan depletion suggests that in some circumstances, particularly in patients who have experienced recurrent depression, low serotonin activity could be involved in predisposing to clinical relapse. This might be relevant to the role of maintenance antidepressant treatment in people at high risk of recurrent illness. When assessing whether the serotonergic system is involved in the aetiology of depression, it is useful to first consider the role serotonin plays in how the brain processes information. In common with other neuromodulatory neurotransmitters, there are relatively few serotonergic neurons in the brain, their nuclei are concentrated in a small region of the brainstem, and they project widely throughout both the cortex and sub-cortical structures . This anatomical arrangement is well suited to the transmission of relatively simple, globally relevant messages across multiple regions of the brain. While the complexity of the serotonergic system is increased by a broad range of functionally distinct receptors and anatomically specific subnuclei projections , there has been continued interest in determining what the content of these messages might be. The paradigmatic example of this approach has examined dopaminergic neurons originating in the ventral tegmental area. The activity of approximately 50% of these dopaminergic neurons is well described by ‘reward prediction error’ signals , which carry the message ‘that was better than you thought it was going to be’ . To date, the message(s) conveyed by the serotonergic system have not been as convincingly characterised as they have for the dopaminergic system. Candidate serotonergic messages draw on the behavioural effect of serotonergic stimulation in experimental settings, which tends to lead to the withholding or inhibition of behaviour, and the observation that serotonergic neurons seem to be activated by both punishing and rewarding events . Thus, it has been suggested that serotonin carriers a ‘punishment prediction error’ signal (‘that was worse than you thought it was going to be’) , or possibly an estimate of the expected rate at which adversity will be encountered (‘It is risky to act at the moment’) . A second variant of message linked to serotonin function concerns how future events are evaluated and seeks to explain why increasing levels of serotonin cause a reduction in impulsivity. As a rule, rewards that are immediately available are preferred over those which are delayed and serotonin has been argued to reduce ‘delay-discounting’, the rate at which time erodes the value of rewarding events (‘it is worth waiting’) . Lastly, an intriguing suggestion frames the evaluative and prospective dimensions of the serotonergic message in terms of trains of thought. Specifically, it is suggested that when deciding what action to take it is useful to imagine both the immediate outcome of your actions and the outcomes of the subsequent actions that then become available. This process is akin to following the branches of a tree from the trunk to the tips, with each branch representing a series of sequential actions. In many cases, the number of future outcomes to be considered becomes extremely large very quickly (i.e. the tree often has many branches that divide often) making this a challenging task. Serotonin is argued to simplify the problem by halting the search along a specific branch as soon as an unwanted outcome is reached, in effect pruning the tree (‘don’t go there!’) . None of the proposed serotonergic messages enjoy the broad, cross-species empirical support of the dopaminergic reward prediction-error account and should therefore be considered as pointers to serotonin’s overall role in cognitive function rather than definitive descriptions. However, a common factor across all of the potential serotonergic messages is that they contain information about the estimated value of events, a quantity which, when measured, is disordered in depressed patients and is a core component of cognitive accounts of the illness. This observation clearly does not provide compelling evidence that disordered serotonergic function is a significant aetiological factor in the development or maintenance of depression; however, it does suggest an intermediate cognitive mechanism by which altering serotonergic function (e.g. via tryptophan depletion, treatment with serotonergic medication or pathology) might lead to changes in depressive symptoms. What should we make of Alec Coppen’s insight, over 50 years on? Serotonergic agents continue to be widely employed in the treatment of a range of mental health conditions, particularly anxiety and depression. Indeed, one of the more dramatic demonstrations of the role of serotonin in mood and self-conscious experience comes from the study of psychedelic drugs, now being repurposed for the treatment of resistant depression. As we have described, there are some reliable abnormalities in serotonin mechanisms in depressed patients but their potential role in the causation of illness remains to be determined. A likely aid in resolving this question will be the continued intense interest in pre-clinical studies of the role of serotonin in processes relevant to depression such as reward and punishment learning, decision-making, emotional regulation, and social cognition . Along with this, in clinical studies, there will be improvements in methods of assessing brain serotonin activity, as shown by recent investigations measuring serotonin release in the living human brain. Clearly, the role of serotonin in depression will need to be integrated into more complex neurobiological models than those originally envisaged. Nevertheless, the link between impaired serotonin activity and depression is likely to outlive its recent obituaries. |
Editorial: Recent advances in molecular and structural endocrinology | 15c5c594-a450-480b-8cbf-986f67a45abc | 10076826 | Physiology[mh] | PM wrote the editorial, SC read, corrected, and approved it. |
Subsets and Splits