Abstract 13280: Proteomic Indicators of Cardiovascular Health Relate to Brain Volume and Dementia Risk

Abstract

Introduction: Several studies demonstrate that cardiovascular (CV) health - even among individuals without overt CV disease - is associated with brain function and may influence Alzheimer’s disease (AD) and dementia progression. Using a proteomic score to quantify the full spectrum of CV health, the current study examined the association of protein-based indicators of CV risk and CV-related mortality with brain structure and dementia risk among older adults. Methods: The SomaScan platform was used for plasma proteomic phenotyping in the Baltimore Longitudinal Study of Aging (BLSA) (n=984) and the Atherosclerosis Risk in Communities (ARIC) study (n=7901). SomaSignal Test measures of risk for CV events, and heart failure-related mortality were derived from the proteome. Concurrent with blood-draw for proteomic assessments, BLSA participants received 3T brain MRI scans, which quantified AD-relevant brain atrophy (SPARE-AD) and total brain volume (TBV). Following protein measurement at Visit 3, ARIC participants were followed for 20+ years, during which time incident dementia was assessed. We used demographic-, intracranial volume, and APOE ε4-adjusted linear regression and Cox models to relate proteomic indicators of CV health to brain volume and dementia risk, respectively. Results: Protein-based risk for CV events and heart-failure mortality were each associated with greater brain atrophy in regions vulnerable to AD and lower total brain volume ( Figure 1 ). Higher risk scores for CV events and both preserved and reduced ejection fraction heart failure mortality were associated with incident dementia (HR per z-score increase (95% CI): 1.26 (1.20-1.33), 1.18 (1.12-1.23) and 1.27 (1.22-1.32), respectively, all p < 0.001). Conclusions: Proteomic indicators of risk for CV disease/mortality are associated with neurodegeneration and future dementia risk. Proteomic indicators of CV health may aid in monitoring and prediction of neurodegenerative disease.