DOI: 10.1002/alz.074993 ISSN: 1552-5260

Alzheimer’s Disease Biomarkers: Roles of Type 2 Diabetes from the HABS‐HD Study

Fang Yu, Keenan A Pituch, Molly Maxfield, Elsa Basena, Yonas E. Geda, Jeremy J. Pruzin, David W. Coon, Gabriel Q Shaibi,
  • Psychiatry and Mental health
  • Cellular and Molecular Neuroscience
  • Geriatrics and Gerontology
  • Neurology (clinical)
  • Developmental Neuroscience
  • Health Policy
  • Epidemiology

Abstract

Background

Alzheimer’s disease (AD) disproportionately affects Mexican Americans, however, reasons underlying this health disparity remain unknown. Type 2 diabetes (T2D) is a risk factor for AD and may contribute to AD pathogenesis. The purpose of this study was to examine the associations between T2D and AD blood biomarkers among Mexican Americans and non‐Hispanic Whites.

Methods

This study analyzed baseline data from the Health and Aging Brain Study: Health Disparities (HABS‐HD) that investigated AD differences among Mexican Americans and non‐Hispanic White. HABS‐AD participants were excluded from this study if they had missing data or were outliers (z‐scores >|4|) on a given AD biomarker (N = 1,552 for Aβ42/40 ratio, 1,570 for t‐tau, and 1,553 for NfL). AD biomarkers included plasma Aβ42/42 ratio, total tau (t‐tau), and neurofilament light (NfL) analyzed using Simoa. Predictors were blood glucose, HbA1c, and T2D diagnosis. Data were analyzed with regression analyses controlling for covariates (e.g., demographics, smoking, BMI, health status, diseases, cognition) and treating missing predictor data using Bayesian MCMC estimation.

Results

Mexican Americans differed significantly from non‐Hispanic Whites in age (66.6±8.7 vs. 69.5±8.6), sex (64.9% female vs. 55.1%), education (9.5±4.6 vs. 15.6±2.5), blood glucose (113.5±36.6 vs. 99.2±17.0), HbA1c (6.33±1.4 vs. 5.51±0.6), T2D diagnosis (35.3% vs. 11.1%), Aβ42/40 ratio (.051±.012 vs. .047±.011), t‐tau (2.56±.95 vs. 2.33±.90), and NfL (16.3±9.5 vs. 20.3±10.3). The models with the covariates and predictors accounted for 6% (p<.001), 15% (p<.001), and 36% (p<.001) of the variation in Aβ42/40 ratio, t‐tau, and NfL, respectively. Blood glucose, HbA1c, and T2D diagnosis were not related to Aβ42/40 ratio and t‐tau but explained 3.7% of the variation in NfL (p<.001). Among non‐Hispanic Whites, blood glucose, HbA1c, and T2D diagnosis were negatively (b = ‐0.09, p<.01, β = ‐0.26), not (b = 0.34, p = .71, β = 0.04), and positively (b = 3.32, p<.01, β = 0.33) associated with NfL, respectively. In contrast, blood glucose and T2D diagnosis were not, while HbA1c was positively (b = 2.31, p<.001, β = 0.26), associated with NfL among Mexican Americans.

Conclusion

Blood glucose, HbA1c, and T2D may contribute to differences in NfL levels, but not Aβ42/40 ratio and t‐tau, in an ethnicity‐specific manner. Future studies are needed to corroborate our findings in longitudinal cohorts.

More from our Archive