Genetic-risk assessment of GWAS-derived susceptibility loci for type 2 diabetes in a 10 year follow-up of a population-based cohort study.

To date, genome-wide meta-analyses have identified genetic susceptibility to type 2 diabetes mellitus (T2D) predominantly in populations of European ancestry. However, comprehensive genetic-risk assessment based on previous GWAS loci has not been fully tested in non-European populations. To evaluate whether a genetic-risk score (GRS) could improve T2D-risk prediction in the Korean population, a GRS (GRS-55) was constructed by summing 55 risk alleles based on the 1000 Genomes imputation in the Korean Association Resource study (T2D cases=1042 and controls=2943 at baseline). We also constructed another GRS (GRS-19) based on nominal significance and consistent direction of effect. In mean difference tests, the mean value of the GRS-19 was significantly higher in T2D cases than in controls at baseline examination. In a model adjusted for area, age, sex and body mass index, weighted GRS-19 was found to be associated with enhanced effect sizes of T2D risk under consistent C-statistics. In addition, we confirmed cumulative risk effects on incidence rates of T2D, fasting plasma glucose and glycated hemoglobin (HbA1c) levels in a longitudinal 10 year of follow-up study. These findings highlight that a genotype score comprised of 19 common variants contributed to T2D-risk prediction in the Korean population. Further multi-locus epistatic interactions may provide the possibility to improve risk prediction in C-statistics for discrimination or reclassification.