To assess glial fibrillary acidic protein (GFAP) plasma levels in the visual and language atypical variants of Alzheimer’s disease (AD) and its association with cross-sectional and longitudinal amyloid-β (Aβ) and tau-PET.

GFAP is a proxy for astrocytic activity, and it is elevated in Aβ-positive individuals, making it a potential, although non-specific, blood-based biomarker for AD.

We assessed plasma GFAP in 72 Aβ-positive participants diagnosed with the visual (n=40) or language (n=32) variant of AD who underwent Aβ- and tau-PET. Fifty-four participants had follow-up PET scans. Tau-PET SUVR was calculated in a meta-region of interest (ROI) specific for atypical Alzheimer’s disease (aty-AD). Linear regressions were run to investigate the association between plasma GFAP and cross-sectional Aβ-PET, tau-PET, and their annualized rate of change, both at the region- and voxel-levels, covarying for age and sex.

Plasma GFAP levels ranged between 133 and 658 pg/mL (median: 266 pg/mL, standard deviation: 110 pg/mL). GFAP did not correlate with age or white matter hyperintensity volume. GFAP also did not correlate with cross-sectional Aβ-PET or tau-PET or with rates of change in Aβ-PET. In the language variant of AD, GFAP correlated with rates of tau accumulation in the aty-AD meta-ROI (p=0.004), covarying for baseline Aβ and tau pathology. On voxel-level analyses, there was a limited positive correlation between GFAP and rates of tau accumulation, particularly in the language variant of AD and in participants with clinical dementia rating sum of boxes <= 4. These associations became weaker after removing one outlier patient with the highest GFAP level (658 pg/mL).

Among Aβ-positive AD participants with atypical presentations, plasma GFAP did not correlate strongly with levels of AD pathology on PET, suggesting that the associations between GFAP and AD pathology might plateau after participants become Aβ-positive. Our findings need to be confirmed in larger cohorts.