Presentation (CTI)

OBJECTIVES: To assess the feasibility of conducting a network meta-analysis (NMA) of treatments for generalized myasthenia gravis (gMG) and to identify methodological challenges impacting comparative evidence synthesis.
METHODS: A rapid literature review identified randomized controlled trials in adults with gMG. Extracted data included study design, population characteristics, interventions, comparators, outcome definitions, follow-up duration, and assessment time points. NMA feasibility was evaluated based on network connectivity, similarity of trial populations, consistency of outcome definitions, comparability of assessment time points, and the presence of potential treatment effect modifiers. Additionally, variability in placebo response rates was assessed as an indicator of unmeasured confounding.
RESULTS: The feasibility assessment identified substantial heterogeneity across the trials, suggesting that conventional NMA approaches may yield highly uncertain comparative estimates. Key methodological challenges included variability in follow-up duration and assessment time points; heterogeneity in baseline disease severity, antibody status, prior treatment exposure and background therapy, and inconsistencies in outcome definitions and responder thresholds for Myasthenia Gravis Activities of Daily Living (MG-ADL), Quantitative Myasthenia Gravis score (QMG), Myasthenia Gravis Composite (MGC) outcomes. These differences raise concerns regarding the transitivity assumption required for valid indirect comparisons. In addition, treatment networks for several outcomes and time points were sparse or disconnected, while evidence of non-proportional or time-varying treatment effects further complicated comparative assessment, all of which pose challenges to a valid indirect comparison.
CONCLUSIONS: Substantial methodological and clinical heterogeneity currently limits the feasibility and robustness of conventional NMA in gMG. These findings underscore the need for greater harmonization of outcome definitions, standardized assessment time points, and improved reporting of patient characteristics to enhance the reliability and interpretability of future comparative NMA in gMG.