Plasma GFAP accuracy and its association with brain pathology

Abstract

AbstractBackgroundGlial fibrillary acidic protein (GFAP) is a biomarker of reactive astrogliosis that increases in the cerebrospinal fluid (CSF) and blood of many neurological and neurodegenerative conditions, including AD. Recent studies have shown that its determination in blood could be especially helpful in AD diagnosis. However, the neuropathological validation of GFAP as a biomarker of AD is still scarce, and only communications to scientific meetings have shown data with pathological validation.MethodIn our study, we aim to evaluate the effectiveness of plasma GFAP in detecting AD neuropathologically defined compared to cognitively unimpaired individuals and the correlation between GFAP levels and AD pathological lesions, estimated by their ADNC score, and other co‐pathologies. Plasma GFAP from cognitively unimpaired controls (n = 118) from the Vallecas cohort and sequential samples of definite AD cases (n = 101) from the Reina Sofia Centre were quantified by the SIMOA platform. On average, the neuropathological assessment was performed on AD patients 139 days after the last plasma collection.ResultPlasma GFAP concentrations were significantly increased in AD cases compared to controls, which allowed high discriminative power (area under the curve for plasma GFAP, 0.88‐0.96). Moreover, longitudinal increases in plasma GFAP levels of AD cases were significantly associated with post‐mortem brain pathology. The individuals who evidenced a higher ADNC stage at the post‐mortem experienced in life a significant increase in the rate of change of plasma GFAP across time, adjusted by age, extraction time of GFAP plasma, and estimated age of disease onset.ConclusionFor the first time, we correlated plasma GFAP levels with post‐mortem brain pathology with exceptionally low time intervals. This study confirms with autopsy data that plasma GFAP is a sensitive biomarker for detecting and tracking AD pathological changes.