Science

Rachel F. Buckley, PhD; Diana L. Townsend, PhD; Colin J. Birkenbihl, PhD;
Madison Cuppels, BSc; Gillian T. Coughlan, PhD; Mabel T. Seto, PhD; Jane A. Brown, PhD; Michael J. Properzi, BEng, BCompSci; Merle C. Hönig, PhD; Annie Li, BSc; Aaron P. Schultz, PhD; Jasmeer Chhatwal, MD, PhD; Hyun-Sik Yang, MD; Steven Arnold, MD; Pia Kivisäkk, PhD; Bryan D. James, PhD; Sid O’Bryant, PhD; Robert A. Rissman, PhD; Melissa Petersen, PhD; Jessica Z. K. Caldwell, PhD; Tobey Betthauser, PhD; Julie Elisabeth Oomens, PhD; Maria Carrigan, MSc; Brian Healy, PhD; Jorge Garcia Condado, PhD; Sterling C. Johnson, PhD; Wai-Ying Wendy Yau, MD; Oliver Langford, MSc; Michelle Farrell, PhD; Rebecca E. Amariglio, PhD; Dorene M. Rentz, PsyD; Kathryn V. Papp, PhD; Ron Brookmeyer, PhD; Timothy J. Hohman, PhD; Michael Donohue, PhD; Paul S. Aisen, MD; Keith A. Johnson, MD; Reisa A. Sperling, MD

Importance Blood-based biomarkers for Alzheimer disease, particularly plasma phosphorylated tau 217 (p-tau217), accurately reflect early Alzheimer disease brain pathology in cognitively unimpaired individuals, but estimates of absolute risk of progression to cognitive impairment across multiple cohorts are needed.

Objective To estimate absolute risk of progression to cognitive impairment and rates of cognitive decline based on plasma p-tau217 across cognitively unimpaired older adults.

Design, Setting, and Participants Longitudinal cohort study using harmonized data from 2684 cognitively unimpaired older adults (defined within cohort) across 6 observational and clinical trial cohorts based in North America, Japan, and Australia. The earliest enrollment was in 2004, with most recent follow-up in 2025.

Exposure Baseline plasma p-tau217.

Main Outcomes and Measures The primary outcome was time to progression to cognitive impairment (mild cognitive impairment, dementia, or 2 consecutive global Clinical Dementia Rating scores ≥0.5). The secondary outcome was longitudinal change on the latent Preclinical Alzheimer Cognitive Composite (PACC; higher values indicate better performance).

Results Among the 2684 participants (median [IQR] age, 69.6 [66.2-74.2] years; 1697 [63%] female), there were 478 events of progression to cognitive impairment over a median follow-up of 5.4 years (maximum follow-up of 13.5 years). Each 1-SD increase in baseline p-tau217 level was associated with an increased risk of progression to cognitive impairment (hazard ratio, 1.38 [95% CI, 1.30-1.46]), and the association remained significant after adjustment, including β-amyloid positron emission tomography scan Centiloids (hazard ratio, 1.32 [95% CI, 1.24-1.41]). Participants with high (1.1-2.4 SD) and very high (>2.5 SD) baseline p-tau217 had 24% (95% CI, 20%-28%) and 38% (95% CI, 33%-43%) absolute risk of progression over 5 years, respectively, and risk was markedly higher over 10 years, although longer-term estimates were constrained by limited data. Elevated p-tau217 was also associated with faster cognitive decline based on change in latent PACC score. Among the overall sample, baseline latent PACC scores ranged from −0.8 to 2.7. The 5-year annualized decline for the very high p-tau217 group was −0.07 latent PACC units/y (95% CI, −0.10 to −0.05), relative to 0.03 units/y (95% CI, 0.02-0.04) in the low p-tau217 group.

Conclusions and Relevance In a pooled sample of multiple selected cohorts of cognitively unimpaired older adults, higher plasma p-tau217 levels were consistently associated with increased risk of clinical progression and accelerated cognitive decline. By providing time-specific absolute risk estimates, these findings support the potential of p-tau217 for prognostic model development, with direct implications for future trial design. Further validation in unselected populations is needed to inform individual prognosis and clinical decision-making in cognitively unimpaired individuals.