(166) Model-based Meta-analysis: An Innovative and Robust Methodology for Clinical Decision Making

Background: Various meta-analytic methods attempt to compare data outside of a head-to-head clinical trial. Despite their limitations, these methods (e.g., pairwise meta-analysis, network meta-analysis) have been widely used by health economists, regulatory decision makers, and reimbursement authorities for decision making.

Objectives: To highlight the utility and methodological benefits of model-based meta-analysis (MBMA).

Methods: MBMA addresses many limitations of traditional meta-analytic methods by providing a quantitative framework leveraging pharmacological information (dose, time [longitudinal data], population characteristics, endpoint correlations, etc.) to explain observed variability in trial outcomes and allowing for appropriate comparisons of treatment options within a disease area. The authors present published examples demonstrating the utility of this methodology.

Results: MBMA was used in four primary ways: (1) to quantify the comparative efficacy and risk; (2) to develop a framework to quantitatively predict efficacy of regimens that combine novel agents; (3) to compare outcomes using data from randomized and non-randomized studies; and (4) to quantify relationships between surrogate markers and clinical endpoints. For example, MBMA was able to predict Phase 3 trial outcomes for risankizumab versus an active comparator in psoriasis, based solely on placebo-controlled trials; confirm that angiotensin receptor blockers and fixed dose thiazide diuretic combinations are safer than either as a monotherapy; leverage Kaplan–Meier overall survival time course data in non-small-cell-lung cancer with early response outcomes to benchmark standard of care (SoC) efficacy and predict long-term clinical outcomes from early response data for PD-1 inhibitors; inform post-approval promotional materials for ponesimod in multiple sclerosis by simulating the efficacy landscape based on indirect treatment comparisons; and solidify the relationship between antibody concentrations of respiratory syncytial virus (RSV) serum neutralizing activity and protection from RSV disease.

Conclusions: MBMA is an innovative approach that can help predict how a new drug’s safety and efficacy profile might compare to SoC and other competitor drugs without directly comparing them, thereby informing many types of critical decisions in early drug development, as well as contributing to post-marketing materials and clinical decision making.