Association of genetic variation with age at diagnosis in type 1 diabetes

Introduction Type 1 diabetes is an autoimmune disease with a strong genetic basis. The aim of this study was to identify additional single-nucleotide polymorphisms (SNPs) for type 1 diabetes age at diagnosis and to replicate previously identified loci. Research design and methods Meta genome-wide association studies of age at diagnosis from eight cohorts (n=5910 in total) were performed in three models. Model 1 was age at diagnosis with no covariates. Model 2 was age at diagnosis adjusted for DR3/DR4 genotype categories. Model 3 was similar to model 2, including the most significant SNP from model 2 (coded additively). Models 1 and 2 were also performed for major histocompatibility complex (MHC) imputed data. In addition, we tested previously identified loci for age at diagnosis and type 1 diabetes risk for association with age at diagnosis in model 1. Results In model 1, we identified a genome-wide significant locus (rs2856721, p=3.3×10 −11 ) in the MHC region whose effect was attenuated in model 2 (p=0.03). In model 2, we identified another locus in the MHC region, rs76730244, p=4.9×10 −9 , which was associated with age at diagnosis adjusted for DR3/DR4 genotypes. Model 3 and analysis of the MHC region did not reveal novel loci. Among 14 previously identified SNPs for age at diagnosis, 6 were confirmed; in addition, 11 out of 78 non-HLA loci for type 1 diabetes risk were associated with age at diagnosis. Conclusions We identified rs76730244 in the MHC region for age at diagnosis of type 1 diabetes, which was independent of the HLA-DR3/DR4 genotype categories. We also confirmed 6 previously identified SNPs and showed that 11 non-HLA loci for type 1 diabetes risk are associated with age at diagnosis.