Appraisal of multiple polygenic risk scores to estimate the risk of myocardial infarction and coronary artery lesions.

BACKGROUND

Polygenic risk scores (PRS) could help to identify individuals with a high genetic risk profile for coronary artery disease (CAD). We aimed to evaluate the association between previously reported PRS and myocardial infarction (MI) as well as the extent and recurrence of coronary artery lesions.

METHODS

We validated previously reported CAD-PRS and 6 cardiovascular (CV) risk factors PRS (systolic blood pressure [SBP], type 2 diabetes [T2D], body-mass index [BMI], low-density lipoprotein cholesterol [LDL], triglycerides [TG], and lipoprotein-[a][Lp(a)]) in individuals of European ancestry from two Canadian population-based cohorts, the Canadian Longitudinal Study on Aging (CLSA, N = 24,599) and CARTaGENE (N = 26,806). Using a stepwise model, we determined an optimal combination of PRS to identify MI. We tested the selected PRS for association with the severity and recurrence of atherosclerotic CAD evaluated by coronary angiography in patients undergoing cardiac surgery (QUEBEC-ANGIO, N = 4108).

RESULTS

We show that the CAD-PRS most strongly associated with MI has odds ratios per standard deviation increment of 1.75 [1.64-1.86] (P = 1.57E-70) in CLSA and 1.87 [1.73-2.03] (P = 3.06E-53) in CARTaGENE. In CLSA, the optimal model includes CAD-PRS, SBP-PRS, BMI-PRS, LDL-PRS, TG-PRS and Lp(a)-PRS. Adding these PRS increases modestly yet significantly the discriminative capacity when compared to traditional risk factors (difference of AUC = 0.025 [0.019-0.031] in CLSA, 0.018 [0.012-0.024] in CARTaGENE). In QUEBEC-ANGIO, the CAD-PRS is gradually and significantly associated with the extent and recurrence of CAD.

CONCLUSIONS

Screening multiple validated PRS may significantly improve genetic risk estimation of MI as well as the extent and recurrence of coronary artery lesions.