Plasma glial fibrillary acidic protein is raised in progranulin-associated frontotemporal dementia.

Heller, Carolin; Foiani, Martha S; Moore, Katrina; Convery, Rhian; Bocchetta, Martina; Neason, Mollie; Cash, David M; Thomas, David; Greaves, Caroline V; Woollacott, Ione Oc; +29 more... Shafei, Rachelle; Van Swieten, John C; Moreno, Fermin; Sanchez-Valle, Raquel; Borroni, Barbara; Laforce, Robert; Masellis, Mario; Tartaglia, Maria Carmela; Graff, Caroline; Galimberti, Daniela; Rowe, James B; Finger, Elizabeth; Synofzik, Matthis; Vandenberghe, Rik; de Mendonca, Alexandre; Tagliavini, Fabrizio; Santana, Isabel; Ducharme, Simon; Butler, Christopher R; Gerhard, Alex; Levin, Johannes; Danek, Adrian; Frisoni, Giovanni; Sorbi, Sandro; Otto, Markus; Heslegrave, Amanda J; Zetterberg, Henrik; Rohrer, Jonathan D; GENFI; GENFI; (2020) Plasma glial fibrillary acidic protein is raised in progranulin-associated frontotemporal dementia. Journal of neurology, neurosurgery, and psychiatry, 91 (3). pp. 263-270. ISSN 0022-3050 DOI: https://doi.org/10.1136/jnnp-2019-321954

Abstract

BACKGROUND: There are few validated fluid biomarkers in frontotemporal dementia (FTD). Glial fibrillary acidic protein (GFAP) is a measure of astrogliosis, a known pathological process of FTD, but has yet to be explored as potential biomarker. METHODS: Plasma GFAP and neurofilament light chain (NfL) concentration were measured in 469 individuals enrolled in the Genetic FTD Initiative: 114 C9orf72 expansion carriers (74 presymptomatic, 40 symptomatic), 119 GRN mutation carriers (88 presymptomatic, 31 symptomatic), 53 MAPT mutation carriers (34 presymptomatic, 19 symptomatic) and 183 non-carrier controls. Biomarker measures were compared between groups using linear regression models adjusted for age and sex with family membership included as random effect. Participants underwent standardised clinical assessments including the Mini-Mental State Examination (MMSE), Frontotemporal Lobar Degeneration-Clinical Dementia Rating scale and MRI. Spearman's correlation coefficient was used to investigate the relationship of plasma GFAP to clinical and imaging measures. RESULTS: Plasma GFAP concentration was significantly increased in symptomatic GRN mutation carriers (adjusted mean difference from controls 192.3 pg/mL, 95% CI 126.5 to 445.6), but not in those with C9orf72 expansions (9.0, -61.3 to 54.6), MAPT mutations (12.7, -33.3 to 90.4) or the presymptomatic groups. GFAP concentration was significantly positively correlated with age in both controls and the majority of the disease groups, as well as with NfL concentration. In the presymptomatic period, higher GFAP concentrations were correlated with a lower cognitive score (MMSE) and lower brain volume, while in the symptomatic period, higher concentrations were associated with faster rates of atrophy in the temporal lobe. CONCLUSIONS: Raised GFAP concentrations appear to be unique to GRN-related FTD, with levels potentially increasing just prior to symptom onset, suggesting that GFAP may be an important marker of proximity to onset, and helpful for forthcoming therapeutic prevention trials.