Objective:

To examine whether plasma glial fibrillary acidic protein (GFAP), a marker of astrocytic activation, predicts cognitive decline, clinical progression, and biomarker changes in preclinical Alzheimer's disease (AD), and whether these associations differ by sex or amyloid status.

Background:

GFAP reflects astrocyte reactivity associated with amyloid pathology and neurodegeneration in AD, but its prognostic role in the preclinical stage remains unclear.

Design/Methods:

Data were analyzed from 949 cognitively unimpaired participants in the Anti-Amyloid Treatment in Asymptomatic Alzheimer’s Disease (A4) Study, a 240-week, multicenter, placebo-controlled trial of solanezumab in amyloid-positive older adults, and the companion amyloid-negative LEARN cohort. Associations between baseline plasma GFAP and longitudinal cognitive trajectories (Preclinical Alzheimer’s Cognitive Composite [PACC]), clinical progression (defined as conversion from global CDR = 0 to ≥ 0.5), amyloid PET, and structural MRI measures of neurodegeneration were examined.

Results:

In the A4 placebo group (Aβ+), higher baseline GFAP predicted faster PACC decline (β = –0.21 ± 0.06, p = 0.004) and greater cortical atrophy in AD-signature regions (β = –5.7 ± 2.4, p = 0.02). In the LEARN cohort (Aβ-), GFAP predicted progression to CDR > 0 (HR = 1.45, 95% CI 1.13–1.87, p = 0.008) and amyloid conversion (OR = 1.40, 95% CI 1.00–1.97, p = 0.048). Among females, GFAP was associated with both lower baseline PACC (β = –5.35 ± 1.96, p = 0.007), faster cognitive decline (β = –0.21 ± 0.05, p < 0.001), and increased risk of CDR progression (HR = 1.31, 95% CI 1.08–1.58, p = 0.005). Among males, associations were weaker, limited to PACC slope (β = –0.14 ± 0.06, p = 0.018), and not significant for clinical progression or imaging outcomes.

Conclusions: