A Phenome-wide association study of a contemporary polygenic risk score for Coronary Artery Disease in the Women’s Health Initiative

Introduction
Polygenic risk scores (PRS) show promise in their ability to predict and improve the treatment of many facets of coronary artery disease (CAD). The basic components of the PRS are known to involve genetic risk related to traditional risk factors. However, associations with carefully adjudicated subtypes of CAD and other health traits have not been adequately tested or documented.
Methods
We conducted a phenome wide association study in 24,693 participants of the Women’s Health Initiative with genetic data and a host of health outcomes, traits, and biomarkers related to cardiovascular disease, cancers, and fractures. Our main exposure variable is a PRS for CAD named ‘meta-GRS’ incorporating 1.7 million genetic variants. We conducted linear and logistic regressions according to variable type.
Results
We found strong significant associations between the PRS and clinical CAD whose magnitude was proportional to the severity of presentation of CAD. We also found statistically significant associations of lower magnitude with other complications of the atherosclerosis. We were able to convincingly extend risk factor observations to include 3 important lifestyle factors (diet, physical activity and smoking), 2 EKG measures that have well-established associations with CAD (LVH and QT interval), an extensive array of LDL and HDL subfractions, multiple biomarkers related to insulin resistance (insulin, glucose, SHGB, leptin), a biomarker of plaque inflammation (Lp-PLA2), and a family history of risk factors of CAD combined with a decrease in parental longevity. Unexpectedly, we found an elevated genetic risk of CAD modestly protects against cancers and hip fractures and increases red blood cell measures.
Conclusion
We confirm that the meta-GRS PRS largely reflects genetic predisposition to CAD and risk factors for CAD. A graded association with presentation subtypes of CAD suggest a strong relationship with burden of coronary atherosclerosis. Further investigation is required to determine the mechanism behind other associations observed including a somewhat unexpected modest protection against all cancers.