Brain imaging reveals hierarchical topology changes and stage-dependent impairments in autoimmune encephalitis.

BACKGROUND

Autoimmune encephalitis is a rapidly progressing neurological disorder caused by aberrant immune responses against neural antigens. It presents with severe neuropsychiatric symptoms such as seizures and cognitive decline, highlighting the need to clarify its neural mechanisms.

OBJECTIVES

To investigate functional and structural brain network alterations across autoimmune encephalitis clinical phases and explore their potential as diagnostic and prognostic biomarkers.

METHODS

Resting-state functional MRI and diffusion tensor imaging were used to analyze brain network topology in 52 patients, including 30 patients who underwent longitudinal follow-up and 32 age- and sex-matched healthy controls. Functional and structural brain networks were constructed using graph-theoretical approaches, and global and local network measures were compared across groups. Machine learning models classified disease status and disease phase.

RESULTS

Patients with autoimmune encephalitis showed significant disruptions in global and local network efficiencies, particularly in the medial occipital and inferior temporal lobes, more pronounced during the acute phase. Classification models achieved high accuracy distinguishing patients from controls (AUC = 0.97 functional, 0.85 structural) and acute from convalescent phases (AUC = 0.98, 0.83).

CONCLUSIONS

Autoimmune encephalitis involves stage-dependent network impairments reflecting disrupted connectivity. Network efficiency may serve as a biomarker for diagnosis and prognosis, supporting multimodal imaging to guide personalized therapeutic strategies.