Determining the risk of coronary artery disease using genetic markers in Asian Indians

Abstract

Jayashree Shanker, Prathima Arvind, Jiny Nair, Vandana Ravindran, Srikarthika Jambunathan, Vijay V Kakkar

Objective: Recent discoveries of single-nucleotide polymorphisms (SNPs) have spurred the development of risk prediction models for coronary artery disease (CAD). We sought to generate a genetic risk score (GRS) for CAD in a representative cohort of Asian Indians. Methods and Results: In Stage 1, 88 variants belonging to 65 genes and the 9p21.3 locus, identified from genome-wide association studies and internal findings were genotyped by Taqman assay in 500 CAD patients (cases) and 500 controls. Twelve SNPs showed significant independent association with CAD. In Stage 2, seven out of the twelve SNPs, analyzed in 534 cases and 534 controls, showed persistent association and belonged to two loci - 9p21.3 (rs10757278, rs2383206, rs10757274, rs1333049, rs4977574) and CELSR2-PSRC1-SORT1 (rs646776, rs599839). Two representative variants, rs10757274 (odds ratio [OR] 1.28, 95% confidence interval [CI] 1.08-1.52) and rs599839 (OR 1.33, 95% CI 1.1-1.62) constituted our pilot GRS. Subjects in the 4th GRS quartile showed a higher risk of CAD as compared to the 1st quartile after adjusting for the classical risk factors (RFs) (OR 2.51, 95% CI 1.82 - 3.45, P < 0.001). In receiver operating characteristic analysis, the two GRS SNPs along with all the conventional RFs (Model 3, C = 0.837) showed better discrimination of CAD than either conventional RFs (Model 1, C = 0.766) or SNPs (Model 2, C = 0.576) alone (Z = −6.6046, P = 3.98 × 10−11), with 24% net reclassification of subjects in the intermediate risk group. Conclusion: A pilot 2-SNP GRS showed 2.5-fold higher risk of CAD in Asian Indians with a modest discrimination, offering scope for further improvement with the addition of newer genetic variants.