Meta-analysis of effectiveness of steroid-sparing attack prevention in MOG-IgG-associated disorder

Abstract Objective To estimate the efficacy of the commonly used long-term immunotherapies in myelin oligodendrocyte glycoprotein IgG associated disorder (MOGAD) Method A comprehensive search of the databases including PubMed/MEDLINE, EMBASE, and Cochrane database was performed for all studies that assessed the efficacy of azathioprine (AZA), mycophenolate mofetil (MMF), rituximab (RTX), and maintenance intravenous immunoglobulin (mIVIG) in MOGAD. The random-effect model is used to estimate the standard mean difference (SMD) of annualized relapse rate (ARR) and expanded disability status scale (EDSS), mean ARR, probabilities of relapse and worsening EDSS during treatment. Results The initial search identified 714 articles, and 21 satisfied eligibility criteria. All immunotherapies significantly reduced ARR in both pediatric and adult populations. Relapse probabilities and pooled mean ARR (SE: standard error) during therapies were as follow: AZA 53.1% [95%CI 37.4% to 68.2%; ARR 0.291 (0.134)], MMF 38.5% [95%CI 19.4% to 62.0%; ARR 0.836 (0.176)], RTX 48.9% [95%CI 37.8% to 60.2%; ARR 0.629(0.162)], and mIVIG 25.3% [95%CI 14.0% to 41.3%; ARR 0.081 (0.058)]. Only RTX significantly improved EDSS, SMD -0.499 (95%CI -0.996 to -0.003). The proportion of worsening EDSS with immunotherapies were 20.7% (95%CI 8.8% to 41.6%), 8.1% (95%CI 1.1% to 41.2%), and 10.8% (95%CI 3.8% to 26.8%) for AZA, MMF, and RTX, respectively. Conclusion These commonly used immunotherapies significantly reduced ARR in MOGAD. Only RTX had a significant benefit in EDSS improvement. However, a substantial portion of patients continued to relapse with treatment. Randomized controlled studies are needed to verify these findings and perform head-to-head comparisons among these treatment options.