Factors for Rituximab Refractoriness in AQP4-IgG+ NMOSD: A Cohort Study

Abstract

Objective: Neuromyelitis optica spectrum disorder (NMOSD) is a severe autoimmune condition of the central nervous system (CNS), often associated with aquaporin-4 antibodies (AQP4-IgG). Rituximab, a CD20+ B-cell depleting monoclonal antibody, is widely used as first-line therapy. However, a subset of patients exhibits treatment refractoriness. Our objective is to investigate factors associated with treatment refractoriness in AQP4-IgG-positive NMOSD patients treated with rituximab.

Methods: This retrospective cohort study included 54 AQP4-IgG-positive NMOSD patients treated with rituximab between 2006 and 2023. Clinical, imaging, and genetic data were analyzed. Treatment failure was defined as at least one relapse occurring after 6 months of rituximab initiation. Statistical analyses included multivariate analyses of covariance (MANCOVA) and Cox regression to identify independent predictors of treatment failure.

Results: Among the 54 patients (82.5% female, median age 45 years, range: 34-54.5), 12 (22.2%) exhibited rituximab treatment failure. The presence of asymptomatic lesions during follow-up was significantly associated with treatment failure (p = 0.02) and emerged as an independent predictor in MANCOVA (Wilks' Lambda = 0.01, F = 20.5, η² = 0.357, p < 0.001). These lesions also increased the risk of clinical relapses (HR = 25.9, 95% CI = 3.09-218, p < 0.01). Other variables, including age, sex, baseline EDSS, and persistent gadolinium enhancement, were not significantly associated with treatment failure. Genetic analysis of the FCGR3A-V158F polymorphism did not reveal a significant relationship with treatment outcomes.

Interpretation: Asymptomatic lesions during rituximab treatment are a strong predictor of therapeutic failure in AQP4-IgG-positive NMOSD patients. Early identification of these lesions could guide clinicians in optimizing treatment strategies, including transitioning to alternative therapies.

Keywords: NMOSD; neuromyelitis optica spectrum disorders; rituximab.

FIGURE 1

CONSORT diagram illustrating the patient selection process.

FIGURE 2

Illustrative examples of asymptomatic lesions observed in patients with AQP4‐positive neuromyelitis optica spectrum disorders. Volumetric FLAIR. (A) Tumefactive periventricular lesion. (B) Periventricular hyperintensity, also referred to as the “pencil sign.” (C) Subcortical demyelinating lesions (yellow arrows). (D) Periventricular and subcortical white matter lesions. (E) Corpus callosum lesions and linear hyperintensity adjacent to the third and fourth ventricles. (F) Periaqueductal hyperintensity. Sagittal STIR. (G) Monosegmental spinal cord lesion.

FIGURE 3

Radar chart comparing clinical and imaging characteristics between patients with and without Rituximab treatment failure. The analysis revealed no statistically significant differences between the two groups in the proportion of smokers, the number of relapses 2 years before rituximab initiation, age at treatment onset, or baseline EDSS. A smaller proportion of treatment‐naïve patients was observed in the failure group (28.6%) compared to the no‐failure group (46.5%), though this difference did not reach statistical significance (p > 0.05). Asymptomatic lesions during Rituximab treatment were significantly more frequent in the failure group (57.1%) compared to the no‐failure group (14%) (p = 0.02). In contrast, persistent gadolinium enhancement was slightly more common in the no‐failure group (23.3%) than in the failure group (14.3%), but this difference was not statistically significant (p > 0.05).