To undertake a genome-wide association study (GWAS) to identify genetic variants for stroke in an Indian population.
In a hospital-based case-control study, 8 teaching hospitals in India recruited 4,088 participants, including 1,609 stroke cases. Imputed genetic variants were tested for association with stroke subtypes using both single-marker and gene-based tests. Association with vascular risk factors was performed with logistic regression. Various databases were searched for replication, functional annotation, and association with related traits. Status of candidate genes previously reported in the Indian population was also checked.
Associations of vascular risk factors with stroke were similar to previous reports and show modifiable risk factors such as hypertension, smoking, and alcohol consumption as having the highest effect. Single-marker–based association revealed 2 loci for cardioembolic stroke (1p21 and 16q24), 2 for small vessel disease stroke (3p26 and 16p13), and 4 for hemorrhagic stroke (3q24, 5q33, 6q13, and 19q13) at p < 5 x 10–8. The index single nucleotide polymorphism of 1p21 is an expression quantitative trait locus (plowest = 1.74 x 10–58) for RWDD3 involved in SUMOylation and is associated with platelet distribution width (1.15 x 10–9) and 18-carbon fatty acid metabolism (p = 7.36 x 10–12). In gene-based analysis, we identified 3 genes (SLC17A2, FAM73A, and OR52L1) at p < 2.7 x 10–6. Eleven of 32 candidate gene loci studied in an Indian population replicated (p < 0.05), and 21 of 32 loci identified through previous GWAS replicated according to directionality of effect.
This GWAS of stroke in an Indian population identified novel loci and replicated previously known loci. Genetic variants in the SUMOylation pathway, which has been implicated in brain ischemia, were identified for association with stroke.