Cerebrospinal fluid biomarkers for predicting immunotherapy response in autoimmune encephalitis

This study aimed to investigate the predictive value of cerebrospinal fluid (CSF) biomarkers for immunotherapy response in patients with autoimmune encephalitis (AE).

We conducted a retrospective analysis of 80 patients with AE who were hospitalized at Zhongshan Hospital of Xiamen University between January 2014 and July 2024. Treatment response was evaluated using a combined assessment of the modified Rankin Scale (mRS) and the Clinical Assessment Scale in Autoimmune Encephalitis (CASE) scores. Patients were classified into good response (n = 52, 65.0%) and poor response (n = 28, 35.0%) groups according to predefined criteria. Feature selection was performed using the Boruta algorithm and least absolute shrinkage and selection operator regression, followed by multivariable logistic regression to identify independent predictors. Diagnostic performance was evaluated using receiver operating characteristic curves and area under the curve (AUC) analysis.

Following systematic feature selection and multicollinearity assessment, three independent predictors of poor immunotherapy response were identified: intrathecal immunoglobulin G (IgG) synthesis rate (OR = 1.57, 95% CI: 1.03-2.45, P = 0.037), peak CASE score (OR = 1.22, 95% CI: 1.07-1.42, P = 0.004), and CSF albumin concentration (OR = 1.00, 95% CI: 1.00-1.01, P = 0.005). Among individual biomarkers, intrathecal IgG synthesis rate demonstrated the highest discriminatory ability (AUC = 0.734), followed by CSF albumin (AUC = 0.722). The combined predictive model showed significantly superior diagnostic performance compared to any single biomarker (AUC = 0.896).

Elevated intrathecal IgG synthesis, peak CASE score, and CSF albumin concentration are independent predictors of poor response in patients with AE.