Abstract
Introduction: T2D is a widely-recognized risk factor for CVD, and leads to worse CVD outcomes. However, there remains considerable heterogeneity among individuals with T2D with respect to who will develop CVD. To develop more effective strategies for risk prediction of T2D-associated heart and vascular disease, we leveraged the ASCVD Pooled Cohort Risk Equation (PCE), a cohort optimized approach and polygenic risk scores (PRS) based on various CVD traits to generate 10-year CVD risk. Hypothesis: We hypothesize that a cohort optimized approach as well as the addition of PRS to traditional risk factors will allow for improved identification of individuals with T2D who are at risk for CVD. Methods: Among our cohort of 1845 T2D patients of African and European ancestries. PRS were derived using ischemic stroke (IS) from the MEGASTROKE consortium and coronary artery disease (CAD) from CardioGRAMplusC4D and calculated in the PMBB dataset using LDPred. A cox proportional hazards model was used for time-to-event analyses within our cohort and we compared model discrimination using AUC. Results: The PCE demonstrated an inadequate discrimination for the study population with an AUC of 0.494 (SD = 0.002), 0.41 (SD = 0.002), 0.609 (SD = 0.004) and 0.514 (SD = 0.004) for female and male Africans and Europeans respectively. A cohort optimized approach demonstrated an AUC of 0.598 (SD = 0.002), 0.440 (SD = 0.002), 0.564 (SD = 0.004) and 0.388 (SD = 0.002) for female and male Africans and Europeans respectively. The addition of CAD and IS PRS to the PCE demonstrated an AUC of 0.519 (SD = 0.002), 0.414 (SD = 0.002), 0.610 (SD = 0.007) and 0.505 (SD = 0.004) for female and male Africans and Europeans respectively. The addition of CAD and IS PRS to the cohort optimized approach demonstrated an AUC of 0.610 (SD = 0.001), 0.478 (SD = 0.001), 0.539 (SD = 0.007) and 0.403 (SD = 0.003) for female and male Africans and Europeans respectively. Conclusions: This study found that the PCE had an inadequate model discrimination across all race/gender categories, however, was improved by using a cohort specific approach and the incorporation of CAD and IS PRS. These findings support further study into the implementation of cohort specific and PRS based approaches in CVD risk prediction of T2D patients.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call