Population-Specific Coronary Artery Disease Genetic Risk Stratification in South Asians: Quantitative Evidence Synthesis and a Precision-Medicine Validation Framework

Abstract

Background: Coronary artery disease (CAD) occurs at a particularly high and often early burden in South Asian populations, yet many genetic risk models used in cardiovascular research were derived mainly from European-ancestry cohorts. Objectives: This review examines a six-variant candidate panel for South-Asian CAD research, summarizes published performance of South-Asian polygenic scores, and sets out a validation strategy that clearly separates a targeted candidate-SNP genetic risk score (GRS) from contemporary genome-wide polygenic risk scores (PRSs). Methods: We conducted a focused narrative evidence synthesis with quantitative re-analysis of published summary estimates; this was not a systematic review or a clinical validation study, and no new participant-level genotype or phenotype data were generated. The evidence base drew on landmark CAD genome-wide association studies, Asian replication studies, South-Asian genome-wide score validation, and multi-ancestry PRS research. For the three South-Asian cohorts reported by Wang et al., published odds ratios (ORs) and 95% confidence intervals (CIs) were transformed to the log scale, standard errors were reconstructed from those published intervals, and the estimates were combined using inverse-variance fixed-effect descriptive synthesis; Cochran's Q and I2 were used to summarize heterogeneity. Results: The six variants map to the 9p21/CDKN2A-B-AS1 region, APOA5, PHACTR1, and ADAMTS7, covering vascular, lipid, endothelial, and extracellular-matrix biology. Across the Wang et al. cohorts, genome-wide polygenic score (GPS) estimates were OR/SD 1.58, 1.60, and 1.66, giving a descriptive pooled estimate of 1.63 (95% CI 1.53-1.73; I2=0%). For the top 5% versus the middle quintile, the pooled OR was 3.46 (2.67-4.50; I2=0%). In South Asians, the multi-ancestry GPSMult score also showed strong external associations with prevalent CAD (OR/SD 1.83, 1.69-1.99) and incident CAD (HR/SD 1.62, 1.49-1.77). Conclusion: Genome-wide genetic scores show reproducible associations with CAD in the South-Asian settings studied, but a six-SNP GRS is best viewed as a transparent, low-cost research instrument rather than a replacement for genomewide PRS. Clinical interpretation requires LD-aware modeling, integration of clinical covariates, calibration, and independent external validation.