As computational models play an increasingly influential role in drug development, the ability to demonstrate their credibility has become as important as the models themselves.
With the FDA‘s adoption of ICH M15: General Principles for Model–Informed Drug Development (MIDD), the industry now has a harmonized framework for evaluating the credibility and regulatory applicability of model-derived evidence.
Key regulatory elements reinforced through the June 2026 implementation include:
🔹 A structured MIDD Evidence Assessment Framework built around six elements: Question of Interest, Context of Use, Model Influence, Consequence of Wrong Decision, Model Risk, and Model Impact.
🔹 A risk-based approach that aligns the extent of model verification, validation, and applicability assessment with the potential impact of model-informed decisions.
🔹 Introduction of a standardized MIDD Assessment Table to support transparent communication and alignment between sponsors and regulatory authorities.
🔹 Formal incorporation of Technical Criteria and Appropriateness of Proposed MIDD as essential considerations in evaluating model-derived evidence.
🔹 Standardized documentation through Model Analysis Plans (MAPs) and Model Analysis Reports (MARs) to improve consistency, transparency, and reproducibility.
🔹 Recognition of both established and emerging methodologies, including PBPK, population PK/PD, quantitative systems pharmacology, disease progression modeling, and AI/ML-enabled approaches.
The significance of ICH M15 lies not in introducing new modeling methodologies, but in defining the evidentiary standards by which model–derived outputs are assessed and trusted.
As quantitative evidence becomes increasingly central to development and regulatory decisions, ICH M15 provides the framework needed to transform computational insights into regulatory-grade evidence. At InSilicoMinds, we view this as an important milestone in the evolution of model-informed drug development. As regulatory expectations continue to evolve, success will increasingly depend not only on developing sophisticated models, but also on demonstrating their credibility, transparency, and fitness-for-purpose through scientifically rigorous evidence.
