Associations Between Cardiovascular Disease, Cognition, Kidney Function and Polyepigenetic Scores of Atherosclerosis in the Health and Retirement Study

TitleAssociations Between Cardiovascular Disease, Cognition, Kidney Function and Polyepigenetic Scores of Atherosclerosis in the Health and Retirement Study
Publication TypeJournal Article
Year of Publication2024
AuthorsAmmous, F, Mitchell, C, Faul, J
KeywordsCardiovascular disease, Cognition, Kidney function, Polyepigenetic Scores

Introduction: Atherosclerosis, characterized by the accumulation of arterial fatty and fibrous materials, underlies various morbidities, particularly cardiovascular disease (CVD). Epigenetic processes, like DNA methylation (DNAm), may function as mediators or biomarkers of atherosclerosis. In this study, we estimated polyepigenetic scores, which are risk-profile scores calculated from the results of an external epigenome-wide association study, for coronary artery calcification (DNAm-CAC) and carotid plaque (DNAm-cPlaque). We assessed the hypothesis that these scores are associated CVD, cognitive function, and renal function.

Methods: We utilized data from the Health and Retirement Study, a longitudinal survey of U.S. adults over age 50. DNAm was measured on a representative sample of respondents aged 56 years and older who consented to blood draws in 2016 (N=3,875). We defined CVD based on self-reported physician diagnoses, cognitive function/dementia status using the Langa-Weir classification, and kidney function using serum cystatin C. We used weighted linear regression for modeling associations with cognitive and kidney function, logistic regression for prevalent CVD, and Cox proportional hazards models for incident CVD and dementia over a four-year follow-up period.

Results: In 2016, the sample had a mean age of 67.4 years (SD=8.5). Most participants were females (57%), with Whites/Caucasians comprising 75% and Blacks/African Americans 17% of the sample. Both DNAm-CAC and DNAm-cPlaque were higher in smokers, individuals with a history of stroke, and those of Black/African American ancestry, while participants with higher education and greater wealth had lower scores (p<0.05). In unadjusted models, DNAm-CAC was linked to higher CVD (Odds ratio: 1.16, 95% CI: 1.07-1.26, p = 3.7х10-4), worse cognitive function (Beta: -0.50, 95%CI: -0.68 - -0.32, p=8.0х10-9 for DNAm-CAC and -0.38, 95% CI: -0.56 - -0.20, p=1.1х10-5 for DNAm-cPlaque), and declining kidney function (1.02 fold-increase, 95% CI: 1.01-1.03, p=0.004). Adjusting for comorbidities/lifestyle factors attenuated CVD and kidney function associations but not cognition (Beta = -0.30, 95% CI: -0.14 - -0.30 for DNAm-CAC; -0.23, 95% CI:-0.05- -0.23 for DNAm-cPlaque). DNAm-cPlaque was associated with dementia/cognitive impairment incidence (hazard ratio: 1.16, 95% CI: 1.01-1.32, p=0.03) after adjusting for comorbidities/lifestyle factors including APOE-E4 genotype. For both cognitive function and dementia/cognitive impairment incidence, adjusting for race/ethnicity, education, and wealth fully attenuated the associations.

Conclusions: Our findings highlight the significance of polyepigenetic scores of atherosclerosis as potential biomarkers for associated morbidities in a representative U.S. sample and underscore the role of social factors as a pathway influencing these associations.

Citation Key13977