Contribution of Lipoprotein(a) to polygenic risk of coronary artery disease: A prospective UK Biobank analysis

Abstract

Abstract Background Lipoprotein(a) (Lp[a]) is a highly atherogenic lipoprotein subfraction that may contribute to polygenic risk of coronary artery disease (CAD), but the extent of this contribution is unknown. Purpose Our objective was to estimate the contribution of Lp(a) to polygenic risk of CAD and to evaluate the respective contributions of Lp(a) and a CAD polygenic risk score (PRS) to CAD. Methods 372,385 UK Biobank participants of European ancestry free of CAD at baseline were included. Plasma Lp(a) levels were measured and a CAD-PRS was calculated using LDpred2. Over the median follow-up of 12.6 years, 13,538 participants had incident CAD (myocardial infarction, coronary artery bypass grafting or coronary angioplasty). Results The LPA region contribution to the CAD-PRS-mediated CAD risk was modest (6.5%). Lp(a) levels significantly increased the predictive performance of a CAD-PRS including age and sex in Cox regression (C-statistic 0.751 versus 0.746, difference = 0.005 (95% CI, 0.004-0.006). Compared to participants in the top CAD-PRS quintile with Lp(a) levels ≥125 nmol/L (CAD event rate 6.6%), the hazard ratio for incident CAD in participants in the bottom CAD-PRS quintile with Lp(a) levels <25 nmol/L was 0.18 (95% CI, 0.17-0.20, p=6.41e-241, CAD event rate 2.0%). Conclusions Compared to individuals with a high genetic risk of CAD (high CAD-PRS and high Lp(a) levels), those with a low genetic risk had an 82% lower CAD risk. These results highlight a substantial contribution of genetic risk factors to CAD and that accurate estimation of genetic risk of CAD should consider blood levels of Lp(a).