Plasma Glial Fibrillary Acidic Protein in the Alzheimer Disease Continuum: Relationship to Other Biomarkers, Differential Diagnosis, and Prediction of Clinical Progression

Abstract

BACKGROUND: Plasma glial fibrillary acidic protein (GFAP) has emerged as a promising biomarker in neurological disorders, but further evidence is required in relation to its usefulness for diagnosis and prediction of Alzheimer disease (AD). METHODS: Plasma GFAP was measured in participants with AD, non-AD neurodegenerative disorders, and controls. Its diagnostic and predictive value were analyzed alone or combined with other indicators. RESULTS: A total of 818 participants were recruited (210 followed). Plasma GFAP was significantly higher in AD than in non-AD dementia and non-demented individuals. It increased in a stepwise pattern from preclinical AD, through prodromal AD to AD dementia. It effectively distinguished AD from controls [area under the curve (AUC) > 0.97] and non-AD dementia (AUC > 0.80) and distinguished preclinical (AUC > 0.89) and prodromal AD (AUC > 0.85) from Aβ-normal controls. Adjusted or combined with other indicators, higher levels of plasma GFAP displayed predictive value for risk of AD progression (adjusted hazard radio= 4.49, 95%CI, 1.18–16.97, P = 0.027 based on the comparison of those above vs below average at baseline) and cognitive decline (standard-β=0.34, P = 0.002). Additionally, it strongly correlated with AD-related cerebrospinal fluid (CSF)/neuroimaging markers. CONCLUSIONS: Plasma GFAP effectively distinguished AD dementia from multiple neurodegenerative diseases, gradually increased across the AD continuum, predicted the individual risk of AD progression, and strongly correlated with AD CSF/neuroimaging biomarkers. Plasma GFAP could serve as both a diagnostic and predictive biomarker for AD.