Pharmacogenetic profiling of dihydropyrimidine dehydrogenase (DPYD) variants in the Indian population.

BACKGROUND To delineate the pharmacologically relevant dihydropyrimidine dehydrogenase (DPYD) variants in the Indian population. METHODS We have screened 2000 Indian subjects for DPYD variants using Infinium Global Screening Array (GSA). RESULTS The GSA analysis has identified seven coding, two intronic, and three synonymous DPYD variants. Level 1A alleles (rs75017182, rs3918290, P633Qfs*5, D949V) were found to be rare (minor allele frequency: 1.889%), while Level 3 alleles were observed to be predominant (C29R: 24.91%, I543V: 9.047%, M166V: 8.993%, V732I: 8.44%). In silico predictions have revealed that all Level 1A alleles were deleterious, while three (M166V, S534N, V732I) of seven Level 3 alleles were damaging. CUPSAT analysis has revealed that two Level 1A (P633Qfs*, D949V) and three Level 3 (I543V, V732I, S534N) variants were thermolabile. Pooled Indian data has showed that V732I, S534N, and rs3918290 variants are associated with 5-FU/Capecitabine toxicity, while C29R, I543V, and M166V variants exhibited the null association. A comparison of our data with the other population data from the "Allele Frequency Aggregator" database showed similarities with the South Asian data. CONCLUSIONS We have identified four Level 1A (non-functional/dysfunctional) and the seven Level 3 variants in the DPYD gene. Pooled Indian data has revealed the association of V732I, S534N, and rs3918290 variants with 5-FU/Capecitabine toxicity. Clustering analysis has shown the similarities in the DPYD profiles of the Indian and South Asian populations.