The relationship between five widely-evaluated variants in CDKN2A/B and CDKAL1 genes and the risk of type 2 diabetes: a meta-analysis.

The genes encoding two cyclin-dependent kinases-inhibitor-2A/B (CDKN2A/B) and 5 regulatory subunit-associated protein-like 1 (CDKAL1) have been investigated extensively in associations with type 2 diabetes; the results, however, are often irreproducible. We therefore sought to evaluate these associations by performing a meta-analysis on five widely-evaluated variants from the two genes. There were 38 studies (patients/controls: 51,940/52,234) for rs10811661, 16 studies (20,029/24,419) for rs564398 in CDKN2A/B gene, and 27 studies (28,383/47,635) for rs7756992, 26 studies (28,816/31,713) for rs7754840, 21 studies (29,260/38,400) for rs10946398 in CDKAL1 gene. Overall risk estimates for type 2 diabetes conferred by rs10811661-T, rs564398-A, rs7754840-C, rs7756992-G, and rs10946398-C alleles were 1.17 (95% CI: 1.10-1.23; P<0.0005; I(2)=83.9%), 1.1 (95% CI: 1.0-1.21; P=0.051; I(2)=88.3%), 1.24 (95% CI: 1.18-1.3; P<0.0005; I(2)=74.3%), 1.2 (95% CI: 1.11-1.3; P<0.0005; I(2)=92.0%), and 1.19 (95% CI: 1.1-1.29; P<0.0005; I(2)=90.8%), respectively. There was evident publication bias for rs564398 and rs7754840. Subgroup analyses by ethnicity showed remarkable divergences in risk estimate for rs564398 between Asians (odds ratio [OR]=1.01; 95% CI: 0.86-1.19; P=0.868) and Caucasians (OR=1.19; 95% CI: 1.03-1.35; P=0.012) (P<0.05). For all variants examined, the results of studies in retrospective design or with population-based controls were comparative with that of overall studies. In meta-regression analyses, age was found to exert a significant influence on the association between rs10811661 and type 2 diabetes (P=0.003), as well as between rs7754840 and gender (P=0.034). Taken together, our findings provide evidence for a significant contribution of CDKN2A/B gene rs10811661 and CDKAL1 gene rs7756992 and rs10946398 to type 2 diabetes.