Combining information from multiple common susceptibility polymorphisms increases the predictive power of genetic information: a study of replicated type 2 diabetes variants. M.N. Weedon¹, M.I. McCarthy², G. Hitman³, M. Walker⁴, C.J. Groves², E. Zeggini², N.W. Rayner², B. Shields¹, K.R. Owen¹, A.T. Hattersley¹, T.M. Frayling¹. 1) Peninsula Medical School, Exeter; 2) Oxford Centre for Diabetes, Oxford; 3) Centre of Diabetes & Metabolic Medicine, University of London; 4) School of Medicine, Newcastle upon-Tyne.
   A limited number of studies have assessed the risk of common diseases when combining information from several predisposing polymorphisms. In most cases, individual polymorphisms only moderately increase risk (~20%) and they are thought to be unhelpful in assessing subjects risk clinically. The impact of looking at multiple alleles simultaneously is not well studied. This is often because, for any given disease, there are very few confirmed, common risk alleles. Three common variants (K23 of KCNJ11, P12 of PPARG, and the T allele at rs7903146 of TCF7L2) predispose to type 2 diabetes mellitus (T2D) across many large studies. Risk allele frequencies range from 0.30 to 0.88 in controls. To assess the combined effect of multiple susceptibility alleles we genotyped these variants in a large case/control study (3668 controls v 2409 cases). Individual allele odds ratios (OR) ranged from 1.14 (95% CI: 1.05,1.23) to 1.48 (95% CI: 1.36, 1.60). We found no evidence of gene-gene interaction and the risks of multiple alleles were consistent with a multiplicative model. Each additional risk allele increased odds of type 2 diabetes by 1.28 (1.21, 1.35) times. Subjects with all 6 risk alleles had an OR of 5.71 (95% CI: 1.15, 28.3) compared to subjects with 0 risk alleles. The 8.1% of subjects double homozygous for the risk alleles at TCF7L2 and Pro12Ala had an odds ratio of 3.16 (95% CI: 2.22, 4.50), against the 4.3% of subjects with no TCF7L2 risk alleles and either 0 or 1 E23K or Pro12Ala risk alleles. In conclusion, combining information from several known common risk polymorphisms allows the identification of subgroups of the population with markedly differing risks of developing type 2 diabetes compared to when using single polymorphisms. This approach may have a role in future preventative measures for common, polygenic diseases.