Abstract INTRODUCTION Alzheimer's disease (AD) heterogeneity complicates early detection and trial design. Scalable predictors may aid risk stratification. We assessed whether scalable baseline plasma biomarkers and neuropsychological measures predict 5‑year cognitive and functional decline in cognitively unimpaired older adults. METHODS We analyzed 866 amyloid‐positive participants from the Anti‐Amyloid Treatment in Asymptomatic Alzheimer's Disease (A4) trial and 343 amyloid‐negative individuals from the Longitudinal Evaluation of Amyloid Risk and Neurodegeneration (LEARN) study. Decline was defined as a ≥0.5 increase in Clinical Dementia Rating–Global Score over 240 weeks. The separate and joint value of demographics, apolipoprotein E ( APOE ) ε4, amyloid positron emission tomography (PET) standardized uptake value ratio (SUVR), plasma phosphorylated tau‐217 (p‐tau217), and Preclinical Alzheimer's Cognitive Composite (PACC) were assessed. A sub‐study of 656 participants evaluated added value of plasma amyloid beta (Aβ)42/Aβ40, glial fibrillary acidic protein (GFAP), and neurofilament light chain (NfL). RESULTS The p‐tau217 and PACC significantly improved prediction. Full models achieved areas under the curve (AUCs) of 0.78–0.80 across cohorts. Additional plasma biomarkers offered modest AUC gains (1%–3%). DISCUSSION The p‐tau217 and PACC enhanced prediction of preclinical decline, supporting their utility in early identification and trial enrichment in AD.
Khorsand et al. (Wed,) studied this question.