Association of genetically predicted lipid traits and lipid-modifying targets with heart failure.

Abstract

To assess the association of genetically predicted lipid traits and lipid-modification via licensed or investigational targets with heart failure (HF). Two-sample Mendelian randomization (MR) study was conducted using summary-level genome-wide association studies (GWASs) from UK Biobank and HERMES Consortium. Genetic variants obtained from UK Biobank GWAS data were selected as instrumental variables to predict the level of lipid traits [LDL cholesterol (LDL-C), HDL cholesterol (HDL-C), triglyceride (TG), apolipoprotein B (ApoB), and apolipoprotein AI (ApoAI)] and lipid-modifying effect of eight drug targets [HMGCR, PCSK9, NPC1L1, PPARA, lipoprotein lipase (LPL), ANGPTL3, APOC3, and cholesteryl ester transfer protein (CETP)]. In this study, we observed that genetically predicted LDL-C, TG, HDL-C or ApoB were significantly related to HF, which were mainly mediated by coronary heart disease (CHD). Drug target MR analyses identified PCSK9, CETP, and LPL as potential targets to prevent HF. The genetic proxy of LDL-C and ApoB increase modified by PCSK9 showed similar evidence in increasing risk of HF (PLDL-C = 1.27*10-4; PApoB = 1.94*10-4); CETP played a role in HF risk via modifying all investigational lipid traits with the strongest evidence though ApoB (P = 5.87*10-6); LPL exerted effects on HF via modifying most lipid traits with the strongest evidence observed via modifying TG (P = 3.73*10-12). This two-sample MR study provided genetic evidence of the associations between lipid traits and HF risk, which were mostly mediated by CHD. Besides, drug target MR studies indicated that PCSK9 inhibition, CETP inhibition, and LPL activation were effective in HF reduction.