Differential associations of various depression-related phenotypes with cardiometabolic risks: Identification of shared genetic factors and implications for drug repositioning

Abstract

Numerous studies have suggested associations between depression and cardiometabolic abnormalities or diseases, such as coronary artery disease and type 2 diabetes. However, little is known about the mechanism underlying this comorbidity, and whether the relationship differs by depression subtypes. Using the polygenic risk score (PRS) approach and linkage disequilibrium (LD) score regression, we investigated the genetic overlap of various depression-related phenotypes with a comprehensive panel of 20 cardiometabolic traits. GWAS results for major depressive disorder (MDD) were taken from the PGC and CONVERGE studies, with the latter focusing on severe melancholic depression. GWAS results on general depressive symptoms (DS) and neuroticism were also included. We also identified the shared genetic variants and inferred enriched pathways. In addition, we looked for drugs over-represented among the top shared genes, with an aim to finding repositioning opportunities for comorbidities.\n\nWe found significant polygenic sharing between MDD, DS and neuroticism with various cardiometabolic traits. In general, positive polygenic associations with CV risks were observed for most depression phenotypes except MDD-CONVERGE. Counterintuitively, PRS representing severe melancholic depression was associated with reduced CV risks. Enrichment analyses of shared SNPs revealed many interesting pathways, such as those related to inflammation, that underlie the comorbidity of depressive and cardiometabolic traits. Using a gene-set analysis approach, we also revealed a number of repositioning candidates, some of which were supported by prior studies, such as bupropion and glutathione. Our study highlights shared genetic bases of depression with cardiometabolic traits, and suggests the associations vary by depression subtypes. To our knowledge, this is the also first study to make use of human genomic data to guide drug discovery or repositioning for comorbid disorders.