Beyond the Forest Plot: Redefining Model‐Based Meta‐Analysis as a Quantitative Engine for Model‐Informed Drug Development

Despite the fact that meta-analysis is a widely utilized methodology in synthesis of evidence from clinical trials, traditional methodologies have limitations with regards to informing dose selection, time-course dynamics and quantitative decision-making across the drug-development continuum. To overcome these limitations, a complex evidence-synthesis model, Model-Based Meta-analysis (MBMA) was proposed to combine pharmacometric modeling and the concepts of meta-analysis. MBMA provides predictive information, by explicitly considering dose-response relationships and longitudinal treatment effects, and by considering between-study heterogeneity, that goes beyond the information provided by a single-time-point summary. Although proven valuable, MBMA remains underutilized and inconsistently applied, partly due to methodological complexity, lack of standardization, and a continued misunderstanding with traditional meta-analysis. The aim of the present review is to critically assess MBMA as a quantitative engine of model-informed drug development (MIDD) and demystify the methodological underpinnings of this technique and its practical applicability. The major MBMA techniques are summarized and synthesized, namely nonlinear mixed-effects modelling, Bayesian hierarchical models and the estimation of models and validation as well as the quantification of uncertainty. The use in therapeutic areas is discussed to demonstrate how MBMA can be used in dose optimization, comparison of efficacy, paediatric extrapolation, and trial design. Regulatory considerations and emerging guidance supporting model-based approaches are also addressed. Landmarking MBMA as a significant distinction from traditional meta-analysis and highlighting its influence on clinical and regulatory decision-making, the review establishes MBMA as an undeniably imperative instrument in translating the heterogeneous clinical data into actionable insights across the lifecycle of drug-development.