ICH M15 Guideline: What the Finalized Guideline Means for Model-Informed Drug Development

The ICH M15 guideline defines MIDD as the use of computational modeling and simulation methods and approaches, including population pharmacokinetics (PopPK), exposure-response (ER) analyses, physiologically based pharmacokinetics (PBPK), model-based meta-analysis (MBMA), and quantitative systems pharmacology (QSP), to integrate nonclinical and clinical data, prior knowledge, and other evidence to inform drug development and regulatory decisions.

Rather than prescribing specific methodologies, the guideline emphasizes early planning, structured thinking, clear communication, and proactive alignment with regulatory authorities to ensure regulatory clarity. It introduces key assessment elements, including:

• Question of interest
• Context of use
• Model influence
• Consequence of wrong decision
• Model risk
• Model impact

Together, these elements formalize expectations for model credibility based on the role the model plays and the decision it supports, within the context of the total evidence package. Modeling is no longer evaluated in isolation; it is assessed based on its impact on development and regulatory decisions.

As Rik de Greef, SVP, Quantitative Science Services, explains:

ICH M15 brings much-needed consistency to how modeling evidence is interpreted across global regulatory agencies. By harmonizing terminology and expectations, it reduces regional variability and enables sponsors to design modeling strategies that can be applied more consistently across FDA, EMA, PMDA, and other authorities.

Eva Gil Berglund, Sr. Director, Clinical Pharmacology and Regulatory Strategy, notes:

MIDD already supports high-impact decisions:

• Replacing certain clinical DDI studies with PBPK modeling
• Informing FIH and dose-escalation strategies in oncology through QSP
• Supporting dose optimization and labeling through exposure-response analyses
• Strengthening dose selection and pediatric extrapolation strategies

The guideline does not introduce new science; it formalizes expectations around credibility, transparency, and how modeling is integrated into development plans.

A central message of the ICH M15 guideline is that MIDD must be embedded early, not appended at submission.

Sponsors are encouraged to define, from the outset:

• The key questions to be addressed
• The context of use, including the role and scope of the model(s) and supporting evidence
• The anticipated model influence, or weight of model outputs relative to other data
• The consequences of a wrong decision, including the severity and likelihood of potential negative effects
• The model risk, or contribution of the model to potential decision error
• The model impact, or how the strategy may differ from standard regulatory expectation

Early integration enables:

• Stronger cross-functional alignment
• More efficient regulatory dialogue
• Generation of decision-ready evidence
• Reduced risk of late-stage rework

The guideline also reinforces the importance of Model Analysis Plans (MAPs), rigorous evaluation and validation, and transparent reporting, all critical to strengthening regulatory confidence.

ICH M15 is part of a broader evolution across the ICH landscape, where multiple guidelines are collectively shaping how evidence is generated and evaluated.

In parallel with M15, several guidelines, both recently finalized and currently in development, highlight how modeling is being applied more consistently across drug development:

• ICH E20 (Adaptive Clinical Trials) is expected to further define how modeling and simulation support more flexible, efficient, and informative trial designs
• ICH E21 (Pregnant and Breastfeeding Populations) reflects increasing emphasis on inclusivity and highlights modeling, particularly PBPK, as a key tool to address data gaps where clinical studies are limited
• ICH E23 (Real World Evidence) signals growing regulatory focus on incorporating real world data, with modeling playing an important role in interpreting and integrating these data sources
• ICH E22 (Patient Preference Studies) introduces structured approaches for incorporating patient perspectives into benefit risk decision making

There are more technical guidelines to come. Importantly, according to the ICH MIDD roadmap^¹, ICH E4, Dose-Response Information to Support Drug Registration, is prioritized to be the next guidance for update or replacement after the MIDD general principles guideline. Taken together, these efforts reinforce a broader shift toward a more integrated, quantitative approach to evidence generation, one that extends beyond traditional clinical data alone.

While ICH M15 provides a framework for how models are evaluated, it is only one part of this broader shift. The guidelines evolving alongside it help clarify how modeling will be applied across development.

For example, ICH E20 is shaping how modeling supports trial design and adaptive decision making. ICH E21 highlights where modeling can help address critical data gaps in underrepresented populations. And with ICH E23, there is increasing focus on how models support the use and interpretation of real-world data.

These are not separate conversations.

ICH M15 defines how regulators evaluate model credibility.
E20, E21, and E23 begin to define where and how those models are applied.

As Rik de Greef notes:

Modeling is no longer a standalone analysis. It is becoming part of the core evidence package, alongside clinical and real-world data used to inform regulatory and strategic decisions.

The ICH M15 guideline represents more than a guidance document. It reflects a broader regulatory expectation that MIDD will be applied systematically, transparently, and strategically.

Organizations that adopt structured planning, multidisciplinary alignment, and robust model credibility practices will be better positioned to:

• Accelerate approvals
• Reduce regulatory risk
• Strengthen global submissions
• Improve development efficiency

For a deeper dive into the key questions sponsors are asking, explore our FAQs on ICH M15 and Model-Informed Drug Development.