Search

Background: The ability to predict the disease course of individuals with major depressive disorder (MDD) is essential for optimal treatment planning. Here, we used a data-driven machine learning approach to assess the predictive value of different sets of biological data (whole-blood proteomics, lipid metabolomics, transcriptomics, genetics), both separately and added to clinical baseline variables, for the longitudinal prediction of 2-year remission status in MDD at the individual-subject level., Methods: Prediction models were trained and cross-validated in a sample of 643 patients with current MDD (2-year remission n = 325) and subsequently tested for performance in 161 individuals with MDD (2-year remission n = 82)., Results: Proteomics data showed the best unimodal data predictions (area under the receiver operating characteristic curve = 0.68). Adding proteomic to clinical data at baseline significantly improved 2-year MDD remission predictions (area under the receiver operating characteristic curve = 0.63 vs. 0.78, p = .013), while the addition of other omics data to clinical data did not yield significantly improved model performance. Feature importance and enrichment analysis revealed that proteomic analytes were involved in inflammatory response and lipid metabolism, with fibrinogen levels showing the highest variable importance, followed by symptom severity. Machine learning models outperformed psychiatrists' ability to predict 2-year remission status (balanced accuracy = 71% vs. 55%)., Conclusions: This study showed the added predictive value of combining proteomic data, but not other omics data, with clinical data for the prediction of 2-year remission status in MDD. Our results reveal a novel multimodal signature of 2-year MDD remission status that shows clinical potential for individual MDD disease course predictions from baseline measurements., (Copyright © 2023 Society of Biological Psychiatry. Published by Elsevier Inc. All rights reserved.)

Background: The heterogeneity of genetic effects on major depressive disorder (MDD) may be partly attributable to moderation of genetic effects by environment, such as exposure to childhood trauma (CT). Indeed, previous findings in two independent cohorts showed evidence for interaction between polygenic risk scores (PRSs) and CT, albeit in opposing directions. This study aims to meta-analyze MDD-PRS × CT interaction results across these two and other cohorts, while applying more accurate PRSs based on a larger discovery sample., Methods: Data were combined from 3024 MDD cases and 2741 control subjects from nine cohorts contributing to the MDD Working Group of the Psychiatric Genomics Consortium. MDD-PRS were based on a discovery sample of ∼110,000 independent individuals. CT was assessed as exposure to sexual or physical abuse during childhood. In a subset of 1957 cases and 2002 control subjects, a more detailed five-domain measure additionally included emotional abuse, physical neglect, and emotional neglect., Results: MDD was associated with the MDD-PRS (odds ratio [OR] = 1.24, p = 3.6 × 10 -5 , R 2  = 1.18%) and with CT (OR = 2.63, p = 3.5 × 10 -18 and OR = 2.62, p = 1.4 ×10 -5 for the two- and five-domain measures, respectively). No interaction was found between MDD-PRS and the two-domain and five-domain CT measure (OR = 1.00, p = .89 and OR = 1.05, p = .66)., Conclusions: No meta-analytic evidence for interaction between MDD-PRS and CT was found. This suggests that the previously reported interaction effects, although both statistically significant, can best be interpreted as chance findings. Further research is required, but this study suggests that the genetic heterogeneity of MDD is not attributable to genome-wide moderation of genetic effects by CT., (Copyright © 2017 Society of Biological Psychiatry. Published by Elsevier Inc. All rights reserved.)

Background: Limited successes of gene finding for major depressive disorder (MDD) may be partly due to phenotypic heterogeneity. We tested whether the genetic load for MDD, bipolar disorder, and schizophrenia (SCZ) is increased in phenotypically more homogenous MDD patients identified by specific clinical characteristics., Methods: Patients (n = 1539) with a DSM-IV MDD diagnosis and control subjects (n = 1792) were from two large cohort studies (Netherlands Study of Depression and Anxiety and Netherlands Twin Register). Genomic profile risk scores (GPRSs) for MDD, bipolar disorder, and SCZ were based on meta-analysis results of the Psychiatric Genomics Consortium. Regression analyses (adjusted for year of birth, sex, three principal components) examined the association between GPRSs with characteristics and GPRSs with MDD subgroups stratified according to the most relevant characteristics. The proportion of liability variance explained by GPRSs for each MDD subgroup was estimated., Results: GPRS-MDD explained 1.0% (p = 4.19e -09 ) of MDD variance, and 1.5% (p = 4.23e -09 ) for MDD endorsing nine DSM symptoms. GPRS-bipolar disorder explained 0.6% (p = 2.97e -05 ) of MDD variance and 1.1% (p = 1.30e -05 ) for MDD with age at onset <18 years. GPRS-SCZ explained 2.0% (p = 6.15e -16 ) of MDD variance, 2.6% (p = 2.88e -10 ) for MDD with higher symptom severity, and 2.3% (p = 2.26e -13 ) for MDD endorsing nine DSM symptoms. An independent sample replicated the same pattern of stronger associations between cases with more DSM symptoms, as compared to overall MDD, and GPRS-SCZ., Conclusions: MDD patients with early age at onset and higher symptom severity have an increased genetic risk for three major psychiatric disorders, suggesting that it is useful to create phenotypically more homogenous groups when searching for genes associated with MDD., (Copyright © 2016 Society of Biological Psychiatry. Published by Elsevier Inc. All rights reserved.)