Astrocyte reactivity identifies p-tau217-positive individuals at elevated risk of cognitive decline across the Alzheimer’s disease continuum
Abstract
Plasma phosphorylated tau at threonine 217 (p-tau217) is a robust marker of Alzheimer’s disease (AD) pathology and progression. However, individuals positive for p-tau217 show substantial clinical heterogeneity, suggesting the influence of modifying biological processes. Here, we investigated whether plasma glial fibrillary acidic protein (GFAP) modifies the association between p-tau217 and clinical progression. We analyzed 3,505 individuals across the aging and AD spectrum from six independent cohorts. All participants had baseline Aβ PET, plasma GFAP, plasma p-tau217, and CDR-SB. Participants were classified into four biomarker groups based on GFAP and p-tau217 positivity. Cross-sectional associations with cognitive severity were examined using linear regression and ANCOVA. Longitudinal cognitive trajectories over five years were modeled using linear mixed-effects models. Variance partitioning decomposed biomarker-attributable variance in individual-level annual CDR-SB change rates. Time to clinical worsening (CDR-SB increase ≥ 0.5 points) was assessed using Kaplan–Meier survival analysis and Cox proportional hazards regression. We show that elevated plasma GFAP significantly modulates the association between p-tau217 and both cross-sectional cognitive severity and longitudinal clinical decline. Compared to individuals p-tau217-positive-only, individuals with co-elevated GFAP and p-tau217 exhibited worse baseline cognitive performance, faster annual CDR-SB progression (62% higher rate; p = 0.008), and an increased hazard of cognitive worsening (HR 1.28; 95% CI 1.01–1.61). The interaction between GFAP and p-tau217 accounted for a large proportion of biomarker-related variance in longitudinal cognitive decline (68.6%), far exceeding the independent contributions of p-tau217 (25.3%) and GFAP (1.1%). These findings support that astrocyte reactivity is associated with greater cognitive impairment and faster clinical decline among individuals with elevated p-tau217, providing additional prognostic stratification within the p-tau217-positive population. Combining plasma GFAP and p-tau217 may therefore have value for identifying individuals at higher risk of progression, with potential implications for risk stratification and clinical trial enrichment.