Different monoclonal antibodies and immunosuppressants administration in patients with neuromyelitis optica spectrum disorder: a Bayesian network meta-analysis

Abstract

Background: A variety of novel monoclonal antibodies and immunosuppressant have been proved effective in treating Neuromyelitis Optica Spectrum Disorder (NMOSD). This network meta-analysis compared and ranked the efficacy and tolerability of currently used monoclonal antibodies and immunosuppressive agents in NMOSD.

Methods: Electronic database including PubMed, Embase and Cochrane Library were searched for relevant studies evaluating monoclonal antibodies and immunosuppressants in patients with NMOSD. The primary outcome measures were annualized relapse rate (ARR), relapse rate, the Expanded Disability Status Scale (EDSS) score, and total adverse events (AEs).

Results: We identified 25 studies with 2919 patients in our meta-analysis. For the primary outcome, rituximab (RTX) (SUCRA: 0.02) ranked first in reduction ARR with a significant difference compared with azathioprine (AZA) (MD - 0.34, 95% CrI - 0.55 to - 0.12) and mycophenolate mofetil (MMF) (MD -0.38, 95% CrI - 0.63 to - 0.14). Tocilizumab (SUCRA: 0.05) ranked first in relapse rate, which was superior to satralizumab (lnOR - 25.4, 95% CrI - 74.4 to - 2.49) and inebilizumab (lnOR - 24.86, 95% CrI - 73.75 to - 1.93). MMF (SUCRA: 0.27) had the fewest AEs followed by RTX (SUCRA: 0.35), both of which showed a significant difference compared with AZA and corticosteroids (MMF vs AZA: lnOR - 1.58, 95% CrI - 2.48 to - 0.68; MMF vs corticosteroids: lnOR - 1.34, 95% CrI - 2.3 to - 0.37) (RTX vs AZA: lnOR - 1.34, 95% CrI - 0.37 to - 2.3; RTX vs corticosteroids: lnOR - 2.52, 95% CrI - 0.32 to - 4.86). In EDSS score, no statistical difference was found between different interventions.

Conclusion: RTX and tocilizumab showed better efficacy than traditional immunosuppressants in reducing relapse. For safety, MMF and RTX had fewer AEs. However, studies with larger sample size on newly developed monoclonal antibodies are warranted in the future.

Keywords: AZA; Meta-analysis; Neuromyelitis optica spectrum disorder; RTX; Relapse.

Fig. 1

The study search, selection, and inclusion process

Fig. 2

Network of trails comparing different monoclonal antibodies and immunosuppressants of NMOSD treatments. The size of circles represented the number of participants for each intervention and the width of lines represented the number of trials compared between treatments. a ARR. b Relapse. c EDSS score. d Total adverse Events. e Gastrointestinal intolerance f Hepatotoxicity. g Leukopenia

Fig. 3

Network meta-analysis results of NMOSD treatments. a ARR and relapse. b EDSS score and total adverse events. c Gastrointestinal intolerance and hepatotoxicity. d Leukopenia. Values in bold indicate significant difference

Fig. 4

Cumulative probability of each intervention for efficacy outcomes. A smaller SUCRA value indicated a better rank for the intervention. a ARR. b Relapse. c EDSS score

Fig. 5

Cumulative probability of each intervention for safety outcomes. A smaller SUCRA value indicated a better rank for the intervention. a Total adverse events. b Gastrointestinal intolerance c Hepatotoxicity. d Leukopenia

Fig. 6

Risk of bias: a summary table for each risk of bias item for each study