Background: While plasma glial fibrillary acidic protein (GFAP) is a promising biomarker for early Alzheimer’s disease (AD), its prognostic value in predicting the conversion from cognitively normal (CN) to mild cognitive impairment (MCI) and its interactions with other biomarkers remain unclear. This study evaluated whether baseline plasma GFAP predicts this conversion and examined whether this relationship is moderated by other biomarkers, including amyloid beta (Aβ) 42/40 ratio, neurofilament light chain (NfL), and phosphorylated tau at threonine 217 (p-tau217).

 

Methods: From the Alzheimer's Disease Neuroimaging Initiative database, we initially identified 1,303 participants with available baseline blood plasma analysis data. The following inclusion criteria were then applied: (1) a minimum follow-up period of 6 months with annually collected longitudinal data available; (2) a clinical diagnosis of CN at baseline; and (3) a clinical status that either remained CN at the final follow-up or progressed to MCI. This selection process resulted in 366 participants, with a maximum follow-up period of up to 15 years. The plasma samples were analyzed using Fujirebio and Quanterix assays. The blood plasma values were log-transformed and z-score standardized. We evaluated the effect of baseline GFAP on the risk of MCI progression and the moderation effect of other biomarkers on GFAP using Cox proportional hazards model, adjusted for age, sex, and apolipoprotein ε4 allele positivity.

 

Results: Baseline GFAP significantly predicted MCI conversion; each 1 SD increase was associated with a 90% higher risk (HR=1.904, 95% CI 1.527-2.373). These associations remained robust (HR=1.689, 95% CI 1.394-2.045) after adjusting for Aβ42/40 ratio, p-tau217, and NfL. Moreover, high GFAP levels (median split) were associated with a 2.2-fold higher risk compared with low GFAP. Notably, p-tau217 positively moderated the GFAP-MCI association (p=.033). In contrast, the plasma NfL and Aβ42/40 ratio did not moderate the association.

 

Conclusions: Plasma GFAP serves as a potent independent predictor of MCI conversion, with its prognostic strength moderated by p-tau217. These findings suggest that GFAP can be utilized as a clinically valuable biomarker for identifying high-risk individuals in the preclinical stage and implementing personalized monitoring strategies for AD.