Skip to main content
Home / Resources / Blog / That’s So Meta: How Model-based Meta-analysis Informs Drug Development

That’s So Meta: How Model-based Meta-analysis Informs Drug Development

Making the right choices in drug development often means the difference between getting a new medication to patients and it ending up in the scrap heap of failed programs. There is a surfeit of publically available information on approved drugs as well as those currently in development. How can sponsors turn clinical trial data into understanding that helps chart the course for investigational drugs? Model-based meta-analysis (MBMA) is an emerging methodology that quantifies clinical trial efficacy, tolerability, and safety information to enable strategic drug development decisions.

What is MBMA?

Since a landmark presentation at the Clinical Pharmacology Subcommittee Meeting at the US Food and Drug Administration in 2006, MBMA has become an accepted innovative strategy to make better use of available data, resulting in increased knowledge and better decision making in clinical development. The strategy involves a systematic search and tabulation of summary results from public sources which may be combined with proprietary clinical trial data. These data are then analyzed using nonlinear regression models which characterize the impacts of drug class, drug, dose, and time on the response(s) of interest. In addition, the potential influence of study population characteristics or the trial conduct may also be explored. The MBMA approach offers two key advantages over classical meta-analyses. First, it supports bridging across studies, thereby enabling comparison of treatments that may never have been tested together in the same clinical trial. Second, MBMA models are based on pharmacologic principles which facilitate incorporating wider spectrum data with regard to dose, observation time, and clinical trial design. In contrast, traditional meta-analysis generally focuses only on treatments that were compared within the same trial, and on a particular dose level for each drug.

How does MBMA support strategic decision making?

Drug development decisions are usually made with in-depth quantitative analysis of internal data from the drug candidate and a comprehensive, but less quantitative, review of public data or data from other candidates. Most decisions cannot be made with internal data alone. Model-based meta-analysis provides a quantitative framework to leverage valuable external data during drug development decision-making.

MBMA can help answer a number of important questions in areas including:

  • Compare your drug vs the competition: What are the characteristics of the dose-response curves for existing drugs that are in the same class as a new compound? What are typical ranges? How does onset of effect differ between drug classes? How do baseline characteristics or background treatments impact drug response?
  • Optimize trial design: What is the impact of trial design features (e.g. time, endpoints) on treatment effects? How are specific subsets of the population represented? What is the impact of region? How do biomarker and clinical endpoint results compare? Can we predict trial results? How can we optimize dosing to maximize safety and efficacy?
  • Inform go/no go, portfolio, marketing decisions: What are the safety and efficacy profiles for competitor drugs for a given therapeutic indication? Can we differentiate the drug as best-in-class? Where is the therapeutic window of the new drug in comparison to competitor/SOC (standard of care) benchmarks? How can we best position a drug between existing and developing competitors?

Gaining insight into the comparative safety and efficacy profile of your drug

There are very few active comparison trials in drug development, however it is often important to assess a compound’s safety and efficacy profile in comparison to SOC and/or competitor drugs in development. Model-based meta-analysis enables indirect comparison, taking into account the impact of treatment, patient population, and trial characteristics. This type of analysis can help estimate the probability that a drug is superior to its competitors in the same drug class or across drug classes.

Elucidating endpoint-to-endpoint relationships

Our clinical outcomes databases contain large amounts of data from published sources, which enables the applications of MBMA to make biomarker to clinical, and short-term to long-term endpoint predictions. Model-based meta-analysis can also be applied to scale across indications. These analyses help predict drug performance in later stage development, or in a different indication.

Interested in learning more about MBMA?

I’ve provided some resources that provide a deeper examination of this topic:

Could MBMA be a valuable tool for helping you optimize decision making for your drug program? Read our whitepaper “Model-based Meta-analysis: An Innovative Methodology Comes of Age” to learn more!

About the author

By: Mark Lovern