SHARED CARDIOVASCULAR RISK FACTORS AND GENETIC OVERLAP OF POSTPARTUM DEPRESSION AND MAJOR DEPRESSIVE DISORDER

Postpartum depression has short- and long-term consequences for maternal mental health, increasing risk for the development of later episodes of major depressive disorder. Previous studies show highly intertwined pathophysiologies between postpartum depression and major depressive disorder, with high genetic correlations. Further, disorders share underlying cardiovascular risk factors, such as blood lipids and BMI. However, the genetic mechanisms linking these risk factors to the disorders as well as the polygenic overlap between disorders have not yet been comprehensively investigated. Further, previous genetic studies have mostly used mixed-sex samples, potentially overlooking sex-specific effects. Therefore, we used summary statistics from female-specific genome-wide association studies (GWAS), leveraging publicly available data when available. Further, we ran a GWAS meta-analysis on diagnosed major depressive disorder in females using data from the UK Biobank, the Norwegian Mother and Child Cohort Study, and a previously conducted GWAS from the Psychiatric Genomics Consortium (N cases = 44,610; N controls = 284,866). We estimated genetic correlations, using linkage disequilibrium score regression, and causal associations, using univariable and multivariable Mendelian randomization. We investigated associations between cardiovascular risk factors (i.e., c-reactive protein, HDL-cholesterol, LDL-cholesterol, triglycerides, and body mass index [BMI]), postpartum depression, and major depressive disorder. Further, we applied bivariate and trivariate causal mixture models (MiXeR) to explore polygenic overlap between postpartum depression, major depressive disorder, and BMI. We found widespread genetic correlations between cardiovascular risk factors, postpartum depression, and major depressive disorder, as well as a very high genetic correlation between the disorders (rg = 1.00, SE = 0.07, p = 5.61e-44). Using Mendelian randomization, LDL-cholesterol as an exposure was associated with postpartum depression as an outcome, and postpartum depression as an exposure was related to major depressive disorder as an outcome. However, these associations were not significant after adjusting for multiple comparisons. When including BMI in multivariable analyses, the effects of BMI (b = 0.14, SE = 0.04, p = 2.36e-3) and postpartum depression (b = 0.11, SE = 0.05, p = 4.26e-2) as exposures on major depressive disorder as an outcome were significant. Both bivariate and trivariate MiXeR estimated high polygenic overlap of postpartum depression with major depressive disorder. Further, trivariate MiXeR estimated 11% overlap of both disorders with BMI. Using multiple genetic approaches, our findings support a strong shared genetic architecture of postpartum depression and major depressive disorder. Further, BMI may partly explain shared disease mechanisms, emphasizing its role as a modifiable risk factor across disorders. Moreover, our results highlight the importance of identifying at-risk females to prevent highly interconnected disorders that share risk factors and can occur in different periods across the female lifespan.