Relapses during treatment with monoclonal antibodies targeting B-cells in NMOSD

While sustained B-cell depletion has proven effective in maintaining a relapse-free state in patients with neuromyelitis optica spectrum disorder (NMOSD), B-cell-depleting therapies do not demonstrate efficacy in all cases. This study aims to investigate the risk factors associated with relapses during B-cell-targeting monoclonal antibody (mAb) treatment in patients with NMOSD.

We retrospectively analyzed baseline clinical characteristics, B-cell test results, and relapse data from NMOSD patients who received B-cell-targeting mAb treatment at four medical centers from July 2014 to September 2024. The annualized relapse rates (ARR) before and after mAb treatment were compared, and potential risk factors for relapses during mAb treatment were evaluated using a Cox proportional hazard model.

This study included 101 NMOSD patients, comprising 14 patients treated with inebilizumab (INEB) and 87 with rituximab (RTX). The patients' mean age at mAb initiation was 49 years (range: 10-74), with a median follow-up duration of 21.5 months (9.4, 37.6). During the study period, two patients were lost to follow-up, one died, and 17 experienced relapses, yielding a relapse-free rate of 83.2%. The ARR showed a significant reduction from 0.3 (0, 0.9) before mAb initiation to 0 (0, 0) after treatment (p = 0.002). Kaplan-Meier curve analysis demonstrated that patients with clustered attacks had a significantly higher risk of relapses compared to those without clustered attacks before mAb initiation (p < 0.001). Similarly, patients with EDSS scores ≥ 6 showed a higher relapse risk than those with EDSS scores < 6 (p = 0.006). Cox proportional hazard models identified several significant risk factors for relapses during mAb treatment, including the total number of attacks (including 2-year, 1-year, and clustered attacks), disease duration, ARR, and EDSS scores ≥ 6 before mAb initiation. Furthermore, patients who experienced relapses with CD19⁺ B-cell levels < 1% demonstrated a significantly higher number of attacks 1 year before mAb initiation (p = 0.015) and a greater proportion of clustered attacks compared to those with CD19⁺ B-cell levels ≥ 1% at relapse (p = 0.035).

Attack frequency and cumulative disability before mAb initiation are important predictors for relapses during targeted B-cell mAb treatment for NMOSD. Future therapeutic strategies should focus on preventing clustered attacks to reduce the likelihood of mAb treatment failure.