Aims: Investigate the role of 33 genetic variants emerged from GWAS apart from phenotypic and behaviour information, in prediction and discrimination of CAD.
Methods: A case-control study was performed with 3050 subjects (1619 coronary patients and 1431 controls) from GENEMACOR. Traditional and new risk factors (TRF) such as smoking, dyslipidemia, diabetes, family history, hypertension, body mass index, physical inactivity, heat rate, creatinine clearance, alcohol, pulse wave velocity, homocysteine, glucose, fibrinogen, lipoprotein (a), APO B lipoprotein, CRP (as) were investigated as well as the 33 genetic variants (Fig 1) previously associated with CAD. Genotyping was performed by TaqMan Real-Time PCR method. A multiplicative genetic risk score (mGRS) with these 33 variants has been created. Multiple logistic regression models were used to estimate the ORs and 95% CI, without and with GRS (5th quintile) adjusted for potential confounders. Area under the ROC curve (AUC) of each one was compared using Delong test.
Results: In our population, the GRS mean was 0.64±0.75 (patients) and 0.46±0.51 (controls). After multivariate model with all the studied risk factors, the following: alcohol, pulse wave velocity, body mass index, lipoprotein APO B, PCR (as) did not remain in the equation and all others showed independency and significance for CAD. When the last quintile of the score is added to the model, CAD risk is 1.93 95% CI 1.56-2.40 (p<0.0001). In the first ROC curve with all risk factors, the AUC was 0.80 and adding the last quintile of mGRS the AUC increased slightly to 0.81 with statistical significance (p=0.002).
Conclusions: Although with statistical significance, the genetic information together with the nongenetic did not add a strong evidence for CAD risk. Future knowledge about rare genetic variants and other SNPs as well as their complex interactions both with genetic and environmental factors will provide an improved clinical utility of the GRS.