Examining the multivariate genetic architectures of posttraumatic stress disorder and cardiovascular disease phenotypes

Posttraumatic stress disorder (PTSD) is associated with elevated cardiovascular disease (CVD) risk, but the extent of shared inherited liability is unclear. Large genome-wide association studies (GWAS) allow for multivariate evaluation of common genetic architecture. We analyzed European-ancestry GWAS summary statistics for PTSD (N_eff ≈ 1.22 million), CVD outcomes (CVD_outcomes; coronary artery disease, stroke, hypertension; N_eff = 126,000-574,000), and CVD risk factors (CVD_risk; high-density lipoprotein, triglycerides, fasting glucose, waist circumference; N = 211,000-526,000). Using genomic structural equation modeling, we modeled latent genetic structure across PTSD and CVD phenotypes and conducted multivariate GWAS of latent factors with gene and gene-set annotation. A three-factor model distinguishing PTSD, CVD_risk, and CVD_outcomes best fit the data, χ² = 256.9, CFI = .990, SRMR = .040. PTSD showed a moderate genetic correlation with CVD_risk, rg = .340, p < .001, and a significant but negligible genetic correlation with CVD_outcomes, rg = 0.030, p < .001, suggesting minimal shared genetic liability at the latent-factor level. Multivariate GWAS identified seven loci for PTSD, 96 for CVD_risk, and 51 for CVD_outcomes. Gene-based tests implicated 85, 910, and 390 genes, respectively, with 18 overlapping between PTSD and CVD_risk. Enrichment highlighted RNA metabolism for PTSD and lipid/vascular pathways for CVD traits. Additional locus-level overlaps emerged, including a genomic hub in the major histocompatibility complex region on Chromosome 6 for both phenotypes. PTSD shares moderate polygenic overlap with CVD, particularly CVD_risk, but many loci are trait-specific, implicating lipid metabolism, vascular tone, immune/inflammatory pathways, and RNA regulation as potential links.