This cohort study investigated DNA methylation changes associated with perinatal suicidal ideation within the Swedish BASIC cohort. The investigation assessed samples collected at 17 and 38 weeks gestation, as well as 8 weeks post-partum. The primary objective was to identify DNA methylation signatures linked to suicidal ideation and predict post-partum ideation risk.
Analysis revealed distinct numbers of differentially methylated probes and regions across the three time points. Targeted analysis identified 1, 10, and 4 significantly differentially methylated probes, while genome-wide analysis detected 465, 2,880, and 510 probes, respectively. Additionally, 7, 25, and 12 differentially methylated regions were observed at the respective time points. A top pathway term identified at 38 weeks gestation was vitamin digestion and absorption.
Prediction models demonstrated varying accuracy depending on the input variables. Using the top ten genome-wide differentially methylated probes at 17 weeks yielded an AUC of 66.9% for predicting post-partum ideation risk. When DNA methylation and depression severity were combined, the AUC increased to 93.2%. Furthermore, 17-week biomarkers achieved 86.2% accuracy in identifying novel suicidal ideation in the post-partum period.
No safety data, adverse events, or discontinuations were reported in the provided evidence. The study limitations are not explicitly detailed in the input data. While these results highlight associations between methylation changes and ideation, the observational nature of the cohort study precludes causal conclusions. Clinical translation remains uncertain until further validation is conducted.
View Original Abstract ↓
Mental health conditions, including perinatal suicidality, remains a significant health burden representing a leading cause of maternal mortality in the United States. Although the etiology of perinatal suicidal ideation (SI) is not well understood, DNA methylation may provide meaningful mechanistic insights and/or serve as clinical biomarkers during the peripartum period. Using data provided by the Swedish BASIC cohort, we performed a retrospective analysis of DNA methylation changes associated with perinatal SI at three perinatal timepoints (17- and 38-weeks gestation and 8 weeks post-partum) through a targeted and genome-wide approach. Targeted analysis of a priori genes revealed 1, 10, and 4 significantly differentially methylated probes at each timepoint and implicated genes associated with the hypothalamic-pituitary-adrenal axis. Genome-wide results identified 465, 2,880, and 510 differentially methylated probes and 7, 25, and 12 differentially methylated regions at each timepoint. Pathway analysis at 38-weeks gestation identified vitamin digestion and absorption as the top term differentially methylated in perinatal SI. Additionally, genes implicated in estrogen and oxytocin signaling were also significantly differentially methylated. Post-partum ideation-risk was successfully predicted using the top ten genome-wide differentially methylated probes at 17 weeks (AUC=66.9%), with prediction accuracy highest when DNA methylation and depression severity were combined (AUC=93.2%). Furthermore, the prediction accuracy for identifying novel SI in the post-partum period increased to 86.2% with 17-week biomarkers. Our results deliver novel insights regarding the role of DNA methylation and perinatal SI, with biomarkers providing both mechanistic insights and clinical usefulness, contributing to the field of perinatal psychiatry and epigenetics.