Total cerebrospinal fluid IgA as a candidate adjunctive biomarker associated with autoimmune encephalitis: a...

Autoimmune encephalitis (AE) is a rare but severe immune-mediated disease of the central nervous system. Early diagnosis remains challenging, and total cerebrospinal fluid (CSF) IgA is a routinely measurable parameter. However, its association with AE and its potential value as a candidate adjunctive biomarker remain unclear.

To investigate whether total CSF IgA is independently associated with AE and to explore its potential as a candidate adjunctive biomarker for identifying AE cases.

We conducted a single-center retrospective case-control study at the Second Affiliated Hospital of Soochow University (January 2010-June 2025), enrolling patients with autoimmune encephalitis and symptomatic controls with non-inflammatory, non-organic neurological disorders. Clinical, laboratory, and neuroimaging data were collected. Logistic regression was used to assess the independent association of total CSF IgA with autoimmune encephalitis and to construct a prediction model. Model performance was evaluated using receiver operating characteristic curves, calibration plots, decision curve analysis, and bootstrap internal validation.

A total of 101 patients with AE and 80 controls were included. Total CSF IgA levels were significantly higher in patients with AE than in controls (median 6.03 vs. 1.46 mg/L, p < 0.001). In the initial multivariable logistic regression model, total CSF IgA was independently associated with AE (OR = 31.63, 95% CI: 2.13-469.09, p = 0.011). The final model incorporated CSF IgA, CSF WBC count, CSF total protein, CSF albumin, blood WBC count, and serum globulin, with CSF IgA remaining independently associated with AE (adjusted OR, 20.53; 95% CI, 4.77-167.45). Although the wide confidence interval warranted caution in interpreting the magnitude of the association. The overall model showed excellent discrimination (AUC = 0.990, 95% CI: 0.982-0.999), good calibration (Brier score, 0.037), and a clear net clinical benefit on decision curve analysis.

Total CSF IgA was independently associated with AE, and a multivariable model combining CSF IgA with routine laboratory parameters showed excellent discrimination and good calibration in distinguishing AE from symptomatic controls. As a rapidly measurable parameter, it shows promise as a candidate adjunctive biomarker. However, its specificity in distinguishing AE from clinical mimics remains uncertain, warranting validation in larger, independent studies.