Inflammatory ratios as early predictors of disease severity and immunotherapy response in autoimmune encephalitis

This study aimed to evaluate and compare the predictive value of systemic inflammatory ratios-including the neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), systemic inflammation index (SII), and C-reactive protein (CRP)-for disease severity, immunotherapy response, and in-hospital mortality in patients with autoimmune encephalitis (AE).

A total of 389 AE patients diagnosed between December 2014 and January 2024 were retrospectively enrolled. Demographic, clinical, and laboratory data were collected. Inflammatory markers were measured prior to immunotherapy. Patients were stratified based on ICU admission and treatment response. Receiver operating characteristic (ROC) curves were used to assess the predictive performance of each biomarker.

ICU-admitted patients (n = 86) exhibited significantly higher NLR, PLR, SII, and CRP levels compared to non-ICU patients (all p < 0.05). NLR showed the highest predictive accuracy for ICU admission (AUC = 0.786). Combined NLR and CRP further improved prediction (AUC = 0.804). For immunotherapy response, NLR and PLR were significantly elevated in poor responders (p < 0.05), with PLR demonstrating an AUC of 0.728. Regarding mortality, PLR and SII were the strongest predictors (AUC = 0.850 and 0.799, respectively), and their combination achieved an AUC of 0.856.

Inflammatory ratios, particularly NLR and PLR, serve as accessible and reliable biomarkers for predicting critical illness, treatment response, and mortality in AE patients. Their integration into clinical practice may facilitate early risk stratification and personalized treatment strategies.