Abstract 10264: Additional Value of a Combined Genetic Score to Framingham Risk Score

Abstract

Introduction: Previous studies consistently reported statistically significant relative risk for coronary artery disease (CAD) with multiple genes. The utility of genetic risk scores (GRS) as additive risk predictors to standard stratification like Framingham risk score (FS) remains inconclusive. Objectives: To evaluate the ability of a multiloci GRS to provide additive value to 10 year risk FS subgroups (low risk <5%, intermediate risk 5-20%, high risk >20%) in a south European sample. Methods: Case-control study of 2555 individuals, 1321 (51.7%) coronary patients and 1234 (48.3%) of controls without coronary disease with a mean age of 52±8.3 years divided in three groups according to FS (FS<5, n=403, 53.3% male, FS 5-20, n=1657, 78% male and FS>20, n= 495, 89.7% male). The multiloci GRS was determined with specific primers of 29 different genetic variants associated to atherosclerotic/CAD disease. A multiplicative model was used based on risk multiplication (odds ratio - OR) of each genotype. Multivariate analysis and respective ROC curves and area under curve (AUC) were performed for each Framingham risk subgroup. The analysis was repeated adding GRS and pairwise comparison of the two ROC Curve was done by the DeLong test. Results: By multivariate analysis GRS was an independent predictor for CAD (OR=2.05; 1.66-2.54, p<0.0001). Diabetes, arterial hypertension, dyslipidemia and smoking (OR= 3.07 (2.47-3.80); OR=2.07 (1.73-2.49); OR=3.1 (2.37-4.07); OR=3.11 (2.58-3.74); all p<0.0001) were also independent CAD predictors. GRS added significant predictive value to Framingham score across the intermediate (0.70 to 0.73 (p<0.0001) and high risk group from 0.68 to 0.72 (p=0.005). Conclusion: GRS proved to add significant incremental value only in intermediate and high risk subgroups. In these subgroups, the inclusion of genotyping may be considered to better clarify cardiovascular risk.