Abstract
BACKGROUND AND PURPOSE: Biologic therapies are anticipated to dominate the treatment landscape for neuromyelitis optica spectrum disorders (NMOSD) in the future. Despite this, many patients in China continue to use off-label medications due to economic and other constraints. A multicenter NMOSD cohort study was conducted to compare the efficacy and safety of tacrolimus (TAC), mycophenolate mofetil (MMF), azathioprine (AZA), and rituximab (RTX). The objective of this study is to provide a clinical evidence-based reference for patients who still require the use of these off-label medications. METHODS: This retrospective study included NMOSD patients treated with TAC (n = 24), MMF (n = 74), AZA (n = 34), and RTX (n = 81). Of these, 81 underwent magnetic resonance imaging (MRI) activity analysis during follow-up. The observation period commenced with the treatment initiation and extended until August 31, 2023. The primary efficacy outcome was the time to the first relapse post-immunotherapy initiation, The hazard ratio (HR) was analyzed using the Cox proportional hazards model to compare the relative risk of the first relapse between different treatment groups (e.g., RTX, MMF, TAC, and AZA). Secondary outcomes encompassed annualized relapse rate (ARR), MRI activity, drug persistence, and relapse rate (RR). The safety outcome was the occurrence of severe adverse drug reaction events. RESULTS: A total of 213 patients were included in the study. During the first year of immunotherapy, patients treated with RTX (HR = 18.41, 95% CI: 4.039-83.87; p < 0.05) and MMF (HR = 22.72, 95% CI: 4.783-108.0; p < 0.0001) experienced a significantly lower risk of relapse compared to those treated with tacrolimus (TAC). The risk of first relapse in the AZA group was higher compared to the RTX group (HR = 2.786, 95% CI: 0.4771-16.27; p = 0.2551) and the MMF group (HR = 4.005, 95% CI: 0.5973-26.86; p = 0.1529), although the differences were not statistically significant. In the second year, this trend continued with RTX (HR = 6.200, 95% CI: 1.825-21.06; p = 0.0034) and MMF (HR = 6.017, 95% CI: 1.782-20.32; p = 0.0039) demonstrating a lower relapse risk compared to oral TAC. Similarly, RTX and MMF were more effective than oral AZA in reducing relapse risk (RTX: HR = 3.510, 95% CI: 1.202-10.25; p = 0.0216; MMF: HR = 3.909, 95% CI: 1.318-11.59; p = 0.0140). The difference in the risk of the first relapse between the MMF and RTX groups was not statistically significant (HR = 0.7217, p = 0.7156 in the first year; HR = 0.9351, p = 0.9003 in the second year) although the difference was not statistically significant. The risk of first relapse was higher in the group treated with oral conventional immunosuppressants (ISTs) compared to the RTX group, (HR = 2.170, p = 0.1449 in the first year; HR = 1.820, p = 0.1091 in the second year). The annual relapse rate (ARR) significantly decreased after treatment with all four drugs. RTX and MMF were more effective in controlling disease relapse compared to TAC and AZA, though these differences were not statistically significant (RTX: ARR = 0.12, 95% CI: 0.03-0.21; MMF: ARR = 0.15, 95% CI: 0.07-0.23; TAC: ARR = 0.21, 95% CI: 0.03-0.39; AZA: ARR = 0.19, 95% CI: 0.08-0.3; p = 0.81). When combining clinical and relapse-independent MRI activity analyses in 81 NMOSD patients, RTX demonstrated superior control of disease activity, with a statistically significant difference (p = 0.036). No hospitalization events related to severe drug adverse effects were reported in either the IST or RTX groups. CONCLUSION: The study provides data comparing the efficacy of various off-label treatments in a Chinese NMOSD cohort, illustrating that RTX is more effective than traditional immunosuppressants in controlling NMOSD relapses and disease activity but no superiority in the time to the first relapse post-immunotherapy initiation. RTX and MMF may offer superior treatment alternatives for NMOSD patients compared to TAC and AZA.