1478-P: Polygenic Risk Score Predicts Incident Cardiovascular Disease in People with Diabetes at Borderline or Intermediate 10-Year ASCVD Risk

Abstract

People with type 2 diabetes (T2D) and 10-year atherosclerotic cardiovascular disease (ASCVD) of 20% or higher are recommended for intense lipid management. However, the role of genetic predisposition to ASCVD on determining the intensity of statin therapy in patients with T2D at borderline (5-7.5%) or intermediate (7.5-20%) risk for incident ASCVD is unclear. We aimed to investigate the utility of polygenic risk score (PRS) in predicting incident ASCVD in people with T2D at borderline and intermediate risk from UK Biobank. We calculated 10-year ASCVD risk using pooled cohort risk equations and PRS for coronary artery disease by Bayesian regression (PRS-CS) in participants of UK Biobank. We performed time-to-event analysis for incident ASCVD and assessed improvement of C-statistic when PRS was added to 10-year ASCVD risk. In addition, risk of ASCVD was estimated for those who were in the top 20 percentile of PRS compared to the remainders. During a median follow-up of 9 years, the incidence of ASCVD in people with T2D was 5.9% and 8.6% in borderline (N=2511) and intermediate (N=8292) group, respectively. The normalized mean PRS for those who developed ASCVD was higher in each group: 0.20±0.96 vs −0.02±1.01 in borderline and 0.20±1.02 vs −0.02±0.99 in intermediate group (all P<0.01). Adding PRS to 10-year ASCVD risk improved the C statistic from 0.51 (95% CI 0.46-0.55) to 0.57 (0.52-0.62) in borderline and from 0.55 (0.53-0.57) to 0.58 (0.56-0.60) in intermediate group. Standardized PRS was associated with incident ASCVD in each group: HR 1.22 (CI 1.05-1.45, P<0.05) and 1.24 (1.15-1.34, P<0.001). Moreover, those who were in the top 20 percentile had increased risk of ASCVD: HR 1.63 (1.14-2.33, P<0.01) and 1.43 (1.21-1.70, P<0.001), respectively. In conclusion, genetic information in people with T2D at borderline and intermediate ASCVD risk could be used to identify those who are at higher risk for incident ASCVD and require high intensity statin treatment. Disclosure J.Choi: None. H.Lee: None. J.S.Lee: None. T.Rhee: None. K.Park: None. S.Kwak: None. Funding National Human Genome Research Institute (FAIN U01HG011723)