학술논문

Abstract 14595: Genetic Variants Associated With Angiographic Burden of CAD in Europeans and African Americans: The Million Veteran Program
Document Type
Academic Journal
Source
Circulation. Nov 14, 2017 136(Suppl_1 Suppl 1):A14595-A14595
Subject
Language
English
ISSN
0009-7322
Abstract
Introduction: Recent genome-wide association studies (GWAS) have identified 78 loci associated with the presence of clinically manifest coronary artery disease (CAD). However, no large-scale GWAS on angiographic CAD burden has been reported to date.Hypothesis: A large-scale GWAS of angiographic CAD burden will identify novel genetic determinants of coronary atherosclerosis which may help elucidate mechanisms for accelerated atherosclerosis.Methods: We identified 20,815 European and 4,817 African American participants of the U.S. Million Veteran Program who underwent at least one coronary angiogram documented by the Veterans Affairs Clinical Assessment, Reporting, and Tracking program, a national cardiac cath lab quality program. We genotyped individuals with a customized Affymetrix biobank array and performed standard imputation using the 1000 genomes panel. We modeled angiographic CAD burden as a variable ranging from 0 to 4, representing no, non-obstructive, 1-vessel, 2-vessel, and 3-vessel/left main-vessel CAD burden, respectively. We then performed a GWAS for CAD burden using linear and ordinal logistic regression stratified by race and adjusting for age, sex, and the first 10 principal components of ancestry. We tested for enrichment of associations of clinical CAD loci from the CARDIoGRAM+C4D 1000 genomes meta-analysis in our CAD burden study and vice-versa.Results: Six loci (PHACTR1, LPA, CDKN2B, ADAMTS7, LDLR, APOE) in Europeans and one locus (LDLR) in African Americans reached genome-wide significance. Among 78 established loci for CAD, we observed a 9.2-fold enrichment in nominal associations (P<0.05) with angiographic CAD burden (36 observed vs 3.9 expected, P<1x10). Furthermore, among 50 loci demonstrating suggestive association with angiographic CAD burden (P<1x10), we observed 9.6-fold enrichment in nominal associations for established loci in CARDIoGRAM+C4D (24 observed vs 2.5 expected, P<1x10).Conclusion: In this largest GWAS to date of angiographic CAD burden, we identified multiple genome wide significant loci for CAD burden in two racial/ethnic groups for the first time. We further documented a high bi-directional enrichment of variants associated with clinical CAD and angiographic CAD burden.