🔎︎
Plasma phosphorylated tau 217 (P-tau217) showed the strongest association with clinical Alzheimer disease (AD) among five blood biomarkers tested in a large community cohort of Caribbean Hispanic adults, but more than a third of clinically diagnosed patients tested negative, according to a cross-sectional study published in Neurology Open Access by Yian Gu, Lawrence S. Honig, Richard Mayeux, and colleagues at Columbia University Irving Medical Center.
P-tau217 has moved quickly toward routine use. Cerebrospinal fluid analysis and positron emission tomography remain the gold standards but are invasive and costly, while blood-based markers are noninvasive and scalable. Data in Caribbean Hispanic populations have been scarce, and earlier work in this group reported high rates of mixed pathology, including cerebrovascular injury and non-AD neurodegenerative processes.
Investigators drew on the Estudio Familiar de Influencia Genética en Alzheimer, enrolling 1,817 adults living in the Dominican Republic and New York City. The mean age was 71.3 years, 67% were women, and the mean education was 6.5 years. Consensus clinical diagnoses were made by neuropsychologists and neurologists without reference to biomarker levels, and 407 participants (22.4%) met criteria for clinical AD. A positivity threshold of 0.39 pg/mL was derived from a separate validation group of 44 autopsy-confirmed and amyloid PET cases.
P-tau217 carried the strongest association with clinical AD, with an odds ratio of 1.74 (95% CI, 1.47-2.07). Biological and clinical classification did not align cleanly. Among the 407 patients with clinical AD, 152, or 37.3%, were P-tau217-negative. Among the 1,410 participants without dementia, 448 were P-tau217-positive.
Stratifying by P-tau217 status changed which marker carried the signal. In P-tau217-negative individuals, neurofilament light chain was associated with clinical AD (odds ratio, 1.66; 95% CI, 1.28-2.14). In P-tau217-positive individuals, glial fibrillary acidic protein carried the association instead (odds ratio, 1.87; 95% CI, 1.44-2.43). Post hoc interaction terms confirmed both patterns. Among P-tau217-positive participants, more years of education were associated with lower odds of clinical AD, which the authors read as consistent with cognitive reserve.
The authors concluded that considering multiple biomarkers simultaneously may improve diagnostic evaluation in Caribbean Hispanic populations, help clinicians avoid false reassurance from a negative AD-specific marker, and support the development of cost-effective, targeted biomarker panels for early detection and differential diagnosis.
They noted that the cut-point came from a small sample drawn from a different population, that the cross-sectional design cannot establish temporal order, and that kidney function was assessed in only a subset, a potentially meaningful confounder given the prevalence of kidney disease in Hispanic populations, though sensitivity analysis in that subset showed little change.
