Background: Previous genome-wide association studies revealed that the chromosome 9p21.3 locus is associated with an increased risk of myocardial infarction (MI) and diabetes mellitus (DM). However, it is unclear whether the 9p21.3-MI association is direct or mediated by pathways related to DM. Study Design: We applied mediation analysis to examine the potential mediating effect of DM on the association between the 9p21.3 genetic risk score (GRS; ranged from 0 to 8) and MI in a case-control study of 865 MI patients and 927 controls without coronary artery disease (CAD). The GRS combining 4 lead 9p21.3 single nucleotide polymorphisms (rs1333040, rs4987574, rs2383207, and rs1333049) was constructed. Results: Each 1 unit increase in weighted 9p21.3 GRS was associated with a 9% increased DM risk (95% confidence interval (CI) 1.00, 1.19) in the control group and a 14% increased MI risk (95% CI 1.09, 1.20). Mediation analyses yielded a direct-effect odds ratio (OR) of 1.06 (95% CI 1.04, 1.08) and an indirect-effect OR of 1.00 (95% CI 0.99, 1.01) for weighted GRS. DM mediated 4.40% (95% CI 3.42, 6.52) of the total association between the weighted GRS and MI risk. Individuals with the highest quartile of 9p21.3 GRS and DM had a 6-fold higher MI risk than those with the lowest quartile of 9p21.3 GRS and non-DM (OR 6.03, 95% CI 3.48, 10.5). Conclusion: DM is a weak mediator that explains a small fraction of the 9p21. 3-MI association in Chinese adults. Nevertheless, there is a strong synergistic effect between DM and the 9p21.3 GRS on MI risk.
Keywords: diabetes mellitus; genetic risk score; mediation analysis; metabolic disease; myocardial infarction.