Involvement of TNFRSF9 in the Pathogenesis of Autoimmune Glial Fibrillary Acidic Protein Astrocytopathy

Background and Objectives
Autoimmune glial fibrillary acidic protein (GFAP) astrocytopathy (GFAP-A) is a CNS inflammatory disorder characterized by steroid responsiveness and relapse. Although GFAP immunoglobulin G (GFAP-IgG) in CSF supports diagnosis, biomarkers reflecting inflammatory activity, disease severity, and relapse risk remain unclear. The aim of this study was to identify and validate CSF protein biomarkers associated with GFAP-A.
Methods
This retrospective observational study included patients with GFAP-A who presented with compatible neurologic features, were positive for CSF GFAP-IgG, and had no alternative diagnosis at a single tertiary referral center. The disease control group involved patients with inflammatory and noninflammatory neurologic disorders. Candidate biomarkers were explored using CSF proteomic analysis comparing GFAP-A with idiopathic normal-pressure hydrocephalus. The most significantly altered protein was quantified by ELISA in an expanded cohort, including patients with GFAP-A and multiple control groups. The primary outcome was the between-group differences in CSF biomarker concentrations. Secondary outcomes included the associations between biomarker levels and clinical characteristics, magnetic resonance imaging (MRI) findings, CSF parameters, and inflammatory cytokines. Statistical analyses included nonparametric tests with effect sizes, Spearman correlation coefficients with confidence intervals, and Benjamini-Hochberg correction for multiple testing.
Results
In total, 102 participants were analyzed, including 42 patients with GFAP-A (median age; 61 years, 31% women). Proteomic screening identified tumor necrosis factor receptor superfamily member 9 (TNFRSF9) as the most significantly upregulated CSF protein in GFAP-A. ELISA confirmed significantly higher CSF TNFRSF9 concentrations in patients with GFAP-A compared with all control groups (Cohen d = 0.45–0.78, all adjusted p < 0.05). Within the GFAP-A cohort, TNFRSF9 levels correlated with CSF cell count, CSF protein concentration, CSF adenosine deaminase, and admission modified Rankin Scale score. TNFRSF9 levels were higher in patients with impaired consciousness and perivascular linear contrast enhancement on MRI, were associated with subsequent clinical relapse, and correlated with proinflammatory cytokines (interleukin-6 and tumor necrosis factor α) and GFAP.DiscussionCSF TNFRSF9 may serve as a biomarker of inflammatory activity, disease severity, and relapse risk in patients with GFAP-A. The major limitations of this study include its retrospective design and modest sample size, which highlight the need for prospective validation.