Predictive improvement of adding coronary calcium score and a genetic risk score to a traditional risk model for cardiovascular event prediction.

Abstract

Coronary artery calcium score (CACS) and polygenic risk score have been used as novel markers to predict cardiovascular (CV) events of asymptomatic individuals compared with traditional scores. No previous studies have directly compared the additive capacity of these two markers relative to conventional scores. The aim of the study was to evaluate the change in CV risk prediction ability when CACS, genetic risk score (GRS), or both are added to Systematic Coronary Risk Evaluation 2 (SCORE2). In a prospective, observational population-based study, 1002 asymptomatic subjects (mean age 53.1 ± 6.8 years, 73.8% male), free of clinical coronary disease and diabetes, were selected from GENEMACOR-study controls. SCORE2, CACS, and GRS were estimated to evaluate CV events' predictive and discriminative ability through Harrell's C-statistics. Net reclassification improvement (NRI) and integrated discrimination index were used to reclassify the population. Multivariable Cox proportional hazard ratio (HR) analysis assessed the variables independently associated with CV events. C-statistic demonstrated that the discriminative value for CV event occurrence was 0.608 for SCORE2, increasing to 0.749 (P = 0.001) when CACS was added, and improved to 0.802 (P = 0.0008) with GRS, showing a better discriminative capacity for CV events. Continuous NRI reclassified >70% of the population. Cox proportional analysis showed that the highest categories of SCORE2, CACS, and GRS remained in the equation with an HR of 2.9 (P = 0.003), 5.0 (P < 0.0001), and 3.2 (P = 0.003), respectively, when compared with the lowest categories. In our population, CACS added to SCORE2 had better ability than GRS in CV event risk prediction, discrimination, and reclassification. However, adding the three scores can become clinically relevant, especially in intermediate-risk persons.