2026 American Academy of Neurology Abstract Website

Objective:

To identify characteristic MRI features of glial fibrillary acidic protein astrocytopathy (GFAP-A) and to establish a diagnostic algorithm that distinguishes GFAP-A from other GFAP-IgG–positive neurological disorders.

Background:

GFAP-IgG in cerebrospinal fluid is essential for diagnosing GFAP-A, but its limited specificity can lead to diagnostic uncertainty. Distinctive neuroimaging features may improve diagnostic accuracy and support early recognition.

Design/Methods:

This retrospective cohort study included 100 CSF GFAP-IgG–positive patients (51 confirmed GFAP-A, 49 with alternative diagnoses) and 442 GFAP-IgG–negative controls evaluated at the Second Affiliated Hospital of Guangzhou Medical University (2015–2025). MRI patterns, clinical data, and CSF profiles were analyzed. Diagnostic performance of an imaging “tetrad” and a clinical–CSF algorithm was assessed by sensitivity, specificity, and area under the ROC curve (AUC). Histopathological correlations were reviewed in biopsy/autopsy samples.

Results:

GFAP-A patients were older, more frequently male, and more often presented with meningoencephalomyelitis than non-GFAP-A controls (P < 0.001). Four MRI signs were characteristic of GFAP-A: claustral “V sign,” posterior thalamic “inverted-brow” sign, pontine “rat’s-eye” sign, and radial perivascular enhancement. In typical meningoencephalomyelitis, CSF GFAP-IgG positivity confirmed diagnosis; presence of any tetrad feature or CSF protein >700 mg/L improved specificity. In atypical cases, ≥1 supporting feature (tetrad imaging, elevated CSF protein, or steroid-responsive symptoms) was suggestive, while ≥2 features achieved high diagnostic accuracy. Histopathology demonstrated venocentric inflammation corresponding to the claustral “V sign,” inverted-brow sign, and radial perivascular enhancement, while Wallerian degeneration underlay the pontine “rat’s-eye” sign.

Conclusions:

An integrated diagnostic algorithm combining clinical, imaging, and CSF data significantly improves specificity over antibody testing alone, enabling earlier and more accurate recognition of GFAP-A.