Genetic and Lifestyle Risks for Coronary Artery Disease and Long-Term Risk of Incident Dementia Subtypes.

Abstract

Background: Shared genetic and lifestyle risk factors may underlie the development of both coronary artery disease (CAD) and dementia. We examined whether an increased genetic risk for CAD is associated with long-term risk of developing all-cause, Alzheimer's, or vascular dementia, and investigated whether differences in potentially modifiable lifestyle factors in the mid- to late-life period may attenuate this risk.

Methods: A prospective cohort study of 365 782 participants free from dementia for at least 5 years after baseline assessment was conducted within the UK Biobank cohort. Genetic risk was assessed using a genomewide polygenic risk score (PRS) for CAD and lifestyle risk using a modified version of the American Heart Association's Life's Essential 8 Lifestyle Risk Score (LRS). Higher values for both scores were deemed to represent increased risk. Primary outcomes were incident all-cause, Alzheimer's, and vascular dementia diagnoses obtained from electronic health records. Secondary outcomes were neuroimaging phenotypes measured in 32 028 participants recalled for magnetic resonance imaging. Sensitivity analyses were conducted to test the extent by which biological and behavioral risk factors contributed to observed associations.

Results: A total of 8870 cases of all-cause dementia were observed over a median 13.9-year follow-up. Both genetic (PRS) and lifestyle (LRS) risk scores for CAD were associated with a modestly elevated risk of all-cause dementia (subhazard ratio per SD increase, 1.10 [1.08, 1.12], P<0.001, for PRS and 1.04 [1.02, 1.06], P=0.006, for LRS). This risk appeared largely attributable to underlying vascular dementia diagnoses (subhazard ratio, 1.16 [1.11, 1.21], P<0.001 for PRS and 1.15 [1.09, 1.22], P<0.001, for LRS), because Alzheimer's disease was found to demonstrate moderate associations with PRS alone (subhazard ratio, 1.09 [1.06, 1.13]; P<0.001). LRS was found to have an additive rather than interactive effect with PRS, with individuals in the highest tertiles for both genetic and lifestyle risk for CAD ≈70% more likely to develop vascular dementia during follow-up compared with those in the lowest tertiles for both (subhazard ratio, 1.71 [1.39, 2.11]; P<0.001). This was substantially attenuated in those with a low LRS at baseline, however, regardless of underlying genetic risk (40% to 50% reduction for low versus high LRS tertile regardless of PRS tertile; P<0.001 for all). In a subset of individuals recalled for neuroimaging assessments, those in the highest tertiles for genetic and lifestyle risk for CAD demonstrated a ≈25% greater volume of white matter hyperintensities than those in the lowest risk tertiles, but showed little difference in gray matter or hippocampal volumes. Sensitivity analyses identified associations between both biological and behavioral risk scores with white matter hyperintensity burden and vascular dementia, whereas some Alzheimer's dementia associations showed seemingly paradoxical relationships.

Conclusions: Individuals who are genetically predisposed to developing CAD also face an increased risk of developing dementia in old age. This risk is reduced in those demonstrating healthy lifestyle profiles earlier in the lifespan, particularly in those who may be at an increased risk of developing dementia caused by an underlying vascular pathology.