Objective: To examine the association between genomewide association study-based diabetes mellitus-related single-nucleotide polymorphisms (SNPs) and coronary artery calcification (CAC), a valid risk factor for coronary heart disease, in a large, unselected, population-based cohort.
Methods and results: We genotyped 11 validated genomewide association study-based diabetes SNPs in 4459 participants of the Heinz Nixdorf Recall Study. We applied generalized linear regression models to explore the impact of the diabetes SNPs on CAC and to jointly model the effect of the SNPs and CAC on diabetes status. We observed a significant association between cyclin-dependent kinase inhibitor 2A/2B (CDKN2A/2B) variant rs564398 and quantitative CAC (P=1.81 x 10(-5) and adjusted P=4.02 x 10(-4); odds ratio for the presence of CAC, 1.12 [95% CI, 1.02 to 1.25]). Moreover, we observed no strong impact of CAC on diabetes risk in the presence of the other genetic variants.
Conclusions: We show that a genetic variant near CDKN2A/2B that has been reported to be strongly associated with diabetes is strongly associated with CAC. In contrast, variants near insulin-like growth factor-binding protein 2 (IGFBP2), CDK5 regulatory subunit associated protein 1-like 1 (CDKAL1), solute carreir family 30 (zinc transporter), member 8 (SLC30A8), hematopoietically-expressed homeobox (HHEX), and transcription factor 7-like2 (TCF7L2) were clearly associated with diabetes; no evidence for an association to CAC was observable. This differential association pattern underlines the potential of endophenotypes, such as CAC, to extend the scope of disease outcome associations.