Hypertrophic cardiomyopathy: a genome-wide association meta-analysis and polygenic risk score.

Abstract

Hypertrophic cardiomyopathy (HCM) is a heritable trait with marked variability in expression and outcomes. Our aims were to discover new genetic loci associated with HCM and to test the effect of a new polygenic risk score (PRS) on incidence, phenotype and outcomes stratified by genotype status. A discovery genome-wide association study (GWAS) was performed on 2284 HCM cases and 4525 controls. Two fixed-effects meta-analyses combined our discovery GWAS with single-trait and multi-trait results from a published study. Discovered loci underwent comprehensive bioinformatic analysis including functional and druggability annotations. A PRS using loci from the two meta-analyses was evaluated for association with HCM diagnosis in 411 213 individuals from UK Biobank (UKBB); imaging phenotypes in individuals without HCM; a composite endpoint (including all-cause mortality and transplantation); and sudden cardiac death (SCD) in 1756 HCM cases. PRS analyses were stratified by genotype status. Three loci were found in the discovery GWAS (<i>BAG3</i>, <i>FHOD3</i> and novel locus <i>PPP1R3A</i>). In the meta-analyses, 70 unique loci were identified, four novel (<i>MYPN, YWHAE, NOS1AP</i> and <i>OBSCN</i>). Bioinformatic analyses identified <i>NOS1AP</i> as a candidate HCM gene. A new PRS was significantly associated with HCM diagnosis (HR=3.19, 95% CI 2.46 to 4.14 for top 5% vs lower 95%; HR=1.88, 95% CI 1.72 to 2.06 per SD increase). Significant associations were found between PRS and greater left ventricular (LV) wall thickness and higher LV ejection fraction in UKBB participants without HCM. Genotype-negative HCM cases in the top 20% of the PRS distribution had an increased risk of SCD (HR=2.72, 95% CI 1.03 to 7.17). We report novel HCM loci. A new PRS predicted the risk of HCM development and associated imaging characteristics in the UKBB and outcomes in an HCM cohort.