A polygenic risk score for coronary artery disease predicts early-onset myocardial infarction and mortality in men

Abstract

Many risk factors for coronary artery disease (CAD) have been described, some of which are genetic. Since the genetic profile can be generated at an early age, its use as a polygenic risk score (PRS) could improve the prediction of CAD risk. To develop a PRS from genetic variants (SNPs) associated with CAD and to assess its association with the incidence of myocardial infarction and mortality, by age and sex. A PRS including the weighted effects of > 1.14 millions of SNPs associated with CAD (CARDIoGRAMplusC4D data) was calculated in the UK Biobank cohort ( n = 408,422), using the LDpred software. Cox regressions were performed, stratified by age quartiles and sex, for MI and mortality, with a median follow-up of 11.1 years. Improvement in risk prediction of MI was assessed over a 10-year period by comparing the PRS to a clinical score, the pooled cohort equation, with categorical net reclassification index for a 2% threshold (NRI 0.02 ) and continuous NRI (NRI > 0 ). From 7744 incident MI cases and 393,751 controls, hazard ratio (HR) for MI reaches 1.53 (95% CI [1.49–1.56], P = 1.3e-296) by SD of PRS. PRS is significantly associated with MI incidence in both sexes, with a stronger association for men (interaction P = 0.002), particularly those aged between 40-51 years (1st quartile) (HR = 2.01, 95% CI [1.86–2.17], P = 4.3e-73). This group presents the highest reclassification improvement (NRI 0.02 = 0.26, 95% CI [0.20–0.29] and NRI > 0 = 0.61, 95% CI [0.52–0.69]), mainly driven by the reclassification of cases to a higher risk. From 23,982 deaths, HR for mortality is 1.08 (95% CI [1.06–1.09], P = 5.5e-30) per SD of PRS, with a stronger association in men (interaction P = 1.6e-06). Our PRS is associated with MI incidence and all-cause mortality, especially in men aged between 40–51 years. The use of PRS could optimize the identification and management of patients at risk for CAD.