American Diabetes Association - 75th Scientific Meeting; Section: Epidemiology/Genetics; Poster n. 1788-P: "A Genetic Risk Score of 96 Variants Linked with Type 2 Diabetes and Cardiometabolic Risk Traits Is Associated with Cardiovascular Mortality in 29-Years Follow-up of the Framingham Heart Study (FHS)"

Cardiovascular diseases (CVD) are a major cause of death and are often associated with type 2 diabetes (T2D). Genome-wide studies (GWAS) identifi ed loci associated with T2D, CVD and traits leading to early death. We investigated whether these loci in aggregate carry a higher risk of all-cause and CVD mortality in the FHS. We computed an unweighted genetic risk score (GRS) of 96 variants selected by effect-size within respective GWAS to represent the top 25% of GWAS variants for the following traits: T2D, coronary artery disease, myocardial infarction (MI), stroke, sudden cardiac death, heart rate, long QT-interval, heavy smoking and 15-years all-cause mortality. We used pooled logistic regressions with genetic-only (GRS adjusted for sex) and full CVD risk factors adjusted models (sex, age, smoking, prevalent non-fatal CVD) to test the association of 96-GRS with all-cause and MI/stroke mortality in 3,426 FHS participants across 29 years follow-up (p<0.025 (p=0.05/2) for signifi cance). Prevalence of non-fatal CVD, T2D and smoking was 7.5, 6.1 and 26.4% at baseline and 18.5, 15.9 and 13.2%, respectively, at the beginning of the last period considered. Cumulative incidence of fatal MI/stroke and all-cause mortality was 5.1 and 22.5%, respectively. The 96-GRS was associated with MI/stroke mortality in both genetic-only (OR[95% CI]: 1.04[1.0-1.1], p=0.006) and fully adjusted model (1.04[1-1.1], p=0.009). Association with all-cause mortality did not reach our statistical signifi cance criteria (1.01[1-1.03], p=0.029, geneticonly; 1.02[1-1.03], p=0.034, fully adjusted). An aggregate burden of 96 GWAS variants with the largest effect size on cardiometabolic traits is predictor of MI/stroke death in longitudinal analysis of a large population of European ancestry. Further studies need to specify the impact of cardiometabolic disease genetics on current mortality prediction models. Supported By: R01DK78616