Model-Based Meta-Analysis (MBMA) at GSK: Real-World Impact on Strategy and Trials

Matt Zierhut, VP and MBMA Capability Lead at Certara, began with a clear message: MBMA is not just an add-on—it’s becoming a critical part of any robust Model-Informed Drug Development (MIDD) strategy.

Unlike traditional pharmacometric analyses, which typically rely on proprietary patient-level data, MBMA harnesses summary-level data from published clinical trial results. The output is a more comprehensive view of the treatment landscape and a predictive framework that can:

• Quantify comparative efficacy and safety
• Build synthetic control arms to replace or supplement traditional comparators
• Predict late-stage clinical outcomes using early markers
• Support dose selection, go/no-go decisions, and trial optimization
• Enable cross-indication learning and trial enrichment strategies

And with Certara’s rigorously developed clinical outcomes databases, sponsors can skip the most time-consuming step—data curation—and move directly to modeling and simulation. This dramatically accelerates the path to insight and impact.

Monica Simeoni, MBMA Lead at GSK, demonstrated how MBMA enabled her team to bridge dose-response relationships across multiple immune-mediated inflammatory diseases (IMIDs), including IBD. Instead of launching redundant dose-ranging studies, they leveraged existing data across related indications to justify dose selection and accelerate early development. The team also modeled early biomarker responses to predict late-stage outcomes in asthma, allowing for more informed trial design decisions.

Nathan Hanan, Director at GSK, shared two case studies using MBMA to model surface antigen loss—a key marker for functional cure—in chronic hepatitis B. His team:

• Predicted HBsAg loss outcomes from peginterferon-alpha monotherapy and combination treatment
• Evaluated expected cure rates following cessation of nucleotide analog therapy

These models supported internal decision-making and regulatory interactions while laying the groundwork for virtual control arms, reducing patient exposure to potentially less effective or poorly tolerated treatments.

Richard Franzese, Director at GSK, presented a case study in metastatic non-small cell lung cancer, where MBMA was used to connect early response rates (e.g., tumor shrinkage) with long-term outcomes like overall survival. By incorporating covariates such as PD-(L)1 expression and ECOG performance status, the team was able to:

• Simulate head-to-head trials across treatment regimens
• Quantify probability of trial success under different design scenarios
• Inform go/no-go decisions and regulatory strategies

The ability to simulate entire trials based on early-phase data provided a significant strategic edge—helping GSK design smarter trials and avoid costly missteps.

Across therapeutic areas and development stages, MBMA is becoming an essential part of how modern teams de-risk decisions, accelerate development, and maximize every data point.

Whether your goal is to:

• Navigate a crowded indication with confidence
• Make evidence-based dose and trial design choices
• Engage regulators with credible comparative data
• Or accelerate timelines through virtual control arms…

MBMA delivers a scalable, data-driven path forward.