The association of blood-based biomarkers of neuropathology with cognitive performance and incident dementia in a diverse, nationally-representative sample of US adults.

INTRODUCTION: The association of blood-based biomarkers of neuropathology with cognition, dementia, and mortality and how these association potentially differ by race/ethnicity, has not been examined in large, diverse, nationally-representative samples of adults.

METHODS: The sample included Health and Retirement Study (HRS) respondents over age 50 with blood-based neuropathology biomarker, demographic, and cognitive data (n=4,214). A -40, A -42, neurofilament light chain (NfL), and glial fibrillary acidic protein (GFAP) were measured in plasma (Quanterix Neurology 4-Plex E kit), and phosphorylated tau (pTau-181) was measured in serum (Quanterix Advantage V2.0 kit). Cognitive tests included immediate and delayed word recall, serial 7s, and backward counting (total score 0-27). Dementia classification relied on a diagnostic algorithm previously validated in the HRS.

RESULTS: When each biomarker was analyzed individually, higher A -42/A -40 ratio was associated with better cognitive function among non-Hispanic (NH) whites. Higher NfL was associated with worse cognitive function in the total sample and in each race/ethnic group (NH white, NH black, and Hispanic). Higher pTau-181 was associated with worse cognitive function in the total and NH white sample. Higher GFAP was related to worse cognitive function in the total sample only. In a model that included all four biomarkers, NfL remained significantly related to cognitive performance in the total sample and in each race/ethnic group, and irrespective of APOE status. NfL was predictive of 6-year incident dementia in our sample (OR=1.33). All four markers significantly predicted 6-year mortality.

DISCUSSION: In a large nationally-representative sample of US adults, we found that NfL was the most consistent predictor of cross-sectional and incident dementia 6 years post blood collection. NfL was also the most consistent predictor across race/ethnic groups examined in our study.

HIGHLIGHTS: There are currently limited data on blood-based biomarkers of neuropathology as predictors of cognitive performance and incident dementia in diverse, population-based cohort studies.We used data from the Health and Retirement Study (n-=4,214) to examine the association between blood-based biomarkers of neuropathology and cognitive function, as well as their association with incident dementia and mortality 4 years after measurement. Mean levels of A -42/A -40 were similar across race/ethnic groups and age groups in this US population-representative sample where selection effects have been minimized. Average NfL was higher among non-Hispanic blacks and Hispanics; GFAP was higher among non-Hispanic blacks as compared to non-Hispanic whites. In a model that included all four biomarkers, NfL remained significantly related to cognitive performance in the total sample and in each race/ethnic group. NfL was associated with incident dementia 6 years after measurement in the total sample. A -42/A -40 ratio was predictive of 6-year incident dementia among those with at least one APOE e4 allele.