AI-generated research brief — verify at source
FcRn Inhibitors, C5 Inhibitors, and B-Cell Depletion Show Comparable Efficacy in Generalized Myasthenia Gravis
A Bayesian network meta-analysis of 27 randomized trials found that FcRn inhibitors, C5 complement inhibitors, and CD19+ B-cell depletion therapy each produced clinically meaningful reductions in myasthenia gravis severity versus placebo, with broadly comparable efficacy across classes (QMG reductions of approximately 3.6, 2.6, and 2.5 points, respectively). However, FcRn inhibitors were associated with roughly twice the odds of treatment-related adverse events compared with placebo (OR 2.20, 95% CrI 1.52–3.38), a safety signal not observed for the other two classes.
What Was Studied
The analysis investigated whether novel treatment classes for generalized myasthenia gravis differ meaningfully in efficacy and safety, using changes in the Quantitative Myasthenia Gravis (QMG) score and the MG-Activities of Daily Living (MG-ADL) scale as the primary outcome measures. Because no direct head-to-head trials exist, the study aimed to provide indirect comparative evidence across FcRn inhibitors, complement inhibitors, B-cell targeting agents, CD40 inhibitors, IL-6 signaling inhibitors, immunoglobulin G preparations, and broad immunosuppressants.
How It Was Studied
This was a pre-registered Bayesian network meta-analysis (PROSPERO ID: CRD42023428733) drawing on searches of Scopus, Embase, Web of Science, CENTRAL, and MEDLINE up to May 2025. Two independent reviewers screened for prospective, multiarm randomized trials enrolling adults aged 18 or older with generalized myasthenia gravis that reported mean treatment differences with associated uncertainty. Of 4,823 initially identified records, 27 placebo-controlled randomized trials met inclusion criteria, contributing 1,086 placebo-treated participants and 1,232 treatment-arm participants. The common comparator across all included trials was placebo plus standard of care.
What Was Observed
- FcRn inhibitors produced the largest average QMG reduction versus placebo: a mean decrease of approximately 3.6 points more than placebo (–3.63, 95% CrI –4.43 to –2.77). On the MG-ADL scale, the advantage over placebo was approximately 1.9 points (–1.93, 95% CrI –2.21 to –1.65). Both credible intervals excluded zero, indicating a consistent treatment effect.
- C5 complement inhibitors showed a similar degree of benefit, reducing QMG scores by roughly 2.6 points more than placebo (–2.59, 95% CrI –3.92 to –1.28) and MG-ADL scores by approximately 1.8 points (–1.84, 95% CrI –2.48 to –1.19). The credible intervals for both outcomes excluded zero, supporting a reliable effect.
- CD19+ B-cell depletion therapy (BCDT) demonstrated comparable MG-ADL improvement to the other classes (–1.91, 95% CrI –2.97 to –0.84), but the QMG estimate (–2.50, 95% CrI –5.11 to 0.12) carried a wide credible interval that just crossed zero, introducing greater uncertainty about the magnitude of this effect on that particular scale.
- FcRn inhibitors were associated with a markedly higher rate of treatment-related adverse events compared with placebo — about twice the odds (OR 2.20, 95% CrI 1.52–3.38). By contrast, C5 complement inhibitors and CD19+ BCDT showed adverse event rates statistically comparable to placebo, suggesting a differentiated safety profile across classes despite similar efficacy.
Why This Matters
Until now, clinicians treating generalized myasthenia gravis have lacked direct comparative data to guide selection among newer mechanistic treatment classes. This analysis is the first Bayesian network meta-analysis to synthesize evidence across these multiple therapeutic targets simultaneously, revealing that efficacy differences between the leading classes are not statistically meaningful while highlighting that their safety profiles diverge. The authors also demonstrate that trial population characteristics — particularly age and sex — may confound cross-trial comparisons, a methodological consideration that future studies in this disease area will need to address explicitly.
How to Read This Result
Although this meta-analysis is methodologically rigorous and draws on a substantial body of randomized trial data, all comparisons are indirect, the credible interval for CD19+ BCDT on QMG crosses zero, and sensitivity analyses flagged age and sex as potential confounders — meaning the apparent equivalence across classes should be interpreted as a provisional signal rather than a definitive conclusion pending prospective head-to-head trials.
Limitations
No prospective head-to-head trials comparing these treatment classes currently exist, so all effect estimates rest on indirect network comparisons. Sensitivity analyses identified age and sex as potential sources of confounding across trial populations, meaning differences in patient demographics between studies may distort the apparent relative efficacy of each class. Additionally, the wide credible interval for CD19+ BCDT on the QMG outcome limits confidence in the precise magnitude of that treatment’s effect on this scale.