Association of Plasma and Cerebrospinal Fluid Alzheimer Disease Biomarkers With Race and the Role of Genetic Ancestry, Vascular Comorbidities, and Neighborhood Factors

IMPORTANCE: Differences in cerebrospinal fluid (CSF) tau Alzheimer dementia (AD) biomarkers by self-identified race have been observed in prior studies. More recently, plasma biomarkers have been gaining recognition, but whether they exhibit similar differences is unclear. Furthermore, the underlying explanation for these differences in AD biomarkers is still unexplored. OBJECTIVES: To investigate differences in plasma biomarkers by race and genetic ancestry and explore potential underlying explanations for these differences. DESIGN, SETTING, AND PARTICIPANTS: This cross-sectional study used participant data from the Brain, Stress, Hypertension, and Aging Research Program (B-SHARP), an observational study conducted in the greater Atlanta metropolitan area. Participants were enrolled from March 1, 2016, to January 1, 2020. MAIN OUTCOMES AND MEASURES: Main outcomes were plasma and CSF amyloid-β (Aβ) 42, Aβ40, phosphorylated tau(181) (p-tau(181)), and neurofilament light. General linear models were used for key comparisons. EXPOSURES: Main independent variables were self-identified race and genetic ancestry. Additional variables were cardiovascular factors, APOE4, educational attainment, Area Deprivation Index, and C-reactive protein (reflecting systemic inflammation state). RESULTS: This analysis included 617 participants (mean [SD] age, 66 [7.9] years; 300 [49%] African American and 317 [51%] White; 429 [70%] with mild cognitive impairment). On the basis of self-reported race, plasma levels of Aβ42 (adjusted mean difference, −1.20 pg/mL; 95% CI, −2.33 to −0.07 pg/mL), Aβ40 (adjusted mean difference, −37.78 pg/mL; 95% CI, −60.16 to −15.39 pg/mL), p-tau(181) (adjusted mean difference, −4.66 pg/mL; 95% CI, −7.05 to −1.90 pg/mL), and neurofilament light (adjusted mean difference, −1.58; 95% CI, −2.83 to −0.19 pg/mL) were consistently lower in African American individuals after adjusting for demographic characteristics, educational attainment, cognition, APOE4, and cardiovascular factors. A similar pattern was observed in the CSF biomarkers except for Aβ42 and Aβ40. Although unadjusted analyses revealed an association between these biomarkers and African ancestry, these associations were not significant after adjusting for the same covariates. Differences by self-reported race were not explained by varied cardiovascular risk factors, C-reactive protein, educational attainment, or Area Deprivation Index. CONCLUSIONS AND RELEVANCE: In this cross-sectional study of plasma biomarkers by race and genetic ancestry, the results indicated that plasma p-tau(181), Aβ40, and NFL were lower in African American individuals based on self-reported race but not genetic ancestry. These differences were not explained by cardiovascular risks or clinical stage differences. These racial differences should be considered in clinical interpretations and clinical trial screenings to avoid an additional increase in underrepresentation of African American individuals in AD trials.