Influence of Cardiometabolic and Alzheimer Disease Genetics on Cognitive-Related Outcomes in a Diverse Population

Abstract

BACKGROUND AND OBJECTIVES: Late-life dementia has a strong heritable component. While the APOE ɛ4 allele is the strongest genetic risk factor of Alzheimer disease (AD), polygenic risk factors also play a significant role. Cardiometabolic diseases (CMDs) and their risk factors predict dementia in many cohorts, but how genetic predictors of CMD influence dementia-related outcomes remains unclear. In this study, we examined the associations of polygenic risk scores (PRSs) for major CMDs and the established APOE risk alleles with dementia-related outcomes in the Reasons for Geographic and Racial Differences in Stroke (REGARDS) study, which is enriched for Black participants underrepresented in late-life dementia research. METHODS: We constructed PRSs for AD and related dementias (ADRDs) and CMDs, including stroke, coronary artery disease, venous thromboembolism, type 2 diabetes, atrial fibrillation, blood pressure (systolic blood pressure, diastolic blood pressure, and pulse pressure), and circulating lipid levels (high-density lipoprotein cholesterol, low-density lipoprotein cholesterol, and triglycerides), using lead genome-wide significant genetic variants from the latest genome-wide association studies. We evaluated the predictive performance of each PRS on their primary end point and tested their associations with incident cognitive impairment (ICI), assessed using the 6-item screener and enhanced cognitive battery, and dementia as a contributing cause of death (DCCD). In addition, we explored the impact of APOE variants and, based on previous reports, local genetic ancestry at the APOE locus on these dementia-related outcomes. RESULTS: Up to 8,818 participants were analyzed (mean age 63.7 years; 58.9% female; 83.3% Black). For CMD PRS, we identified a significant association between PRS for pulse pressure and DCCD (hazard ratio [HR] 1.16, 95% CI 1.06-1.28, p < 1.76 × 10-3). We observed that the PRS for ADRD was associated nominally with DCCD (HR 1.12, 95% CI 1.00-1.25), but not with ICI among REGARDS participants, suggesting heterogeneity in associations across different cognitive-related end points. Furthermore, we identified a nominally significant (p < 0.1) attenuated effect of APOE risk alleles on ICI among those with African ancestry at the APOE locus, similar to previous reports. DISCUSSION: Higher PRS for pulse pressure was associated with increased DCCD risk, supporting shared genetic underpinnings between CMD and dementia.