> ## Content Index
> Fetch the complete content index at: https://www.fdaweb.com/llms.txt
> Use this file to discover other available public pages before exploring further.

# ICH Finalizes Guide on Model-Informed Drug Development
- URL: https://www.fdaweb.com/ich-finalizes-guide-on-model-informed-drug-development/
- Published: 2026-06-02T12:00:00.000Z
- Updated: 2026-09-14T13:40:21.000Z
- Author: David McFarland
- Tags: Drugs, #legacy-id-D5161241

FDA has released an International Council for Harmonization (ICH) guidance that establishes a global framework for using computational modeling and simulation in drug development, a move regulators hope will increase confidence in model-based evidence and expand its role in regulatory decision-making.

The new [guidance](https://www.fda.gov/media/184747/download?ref=fdaweb.com), *M15 General Principles for Model-Informed Drug Development (MIDD)*, provides recommendations for planning, evaluating, documenting, and submitting evidence generated through modeling and simulation approaches. It reflects the growing importance of sophisticated computer models in pharmaceutical development, where sponsors increasingly rely on simulations to predict drug behavior, optimize dosing strategies, assess safety risks, support clinical trial design, and answer regulatory questions that may otherwise require additional studies.

"MIDD" is defined in the guidance as the use of computational modeling and simulation methods that integrate nonclinical data, clinical data, prior knowledge, and information about drugs and diseases to generate evidence that informs development and regulatory decisions.

The document is notable because it creates a common framework for assessing the credibility of model-generated evidence across major regulatory jurisdictions, including the U.S., Europe, Japan, and other ICH member regions.

At the center of the guidance is a structured framework regulators and drug developers can use to determine whether model outputs are sufficiently reliable to support regulatory decisions. The framework requires sponsors to define a specific "question of interest" that a model is intended to answer and establish the model's "context of use," including its role in decision-making and the data used to build it.

The guidance then introduces several new concepts for evaluating model-based evidence, including:

- **Model influence** — the degree to which regulatory decisions depend on model results.
- **Consequence of wrong decision** — the potential impact on patient safety or efficacy if a decision based on the model proves incorrect.
- **Model risk** — a combined assessment of model influence and the consequences of an incorrect decision.
- **Model impact** — the extent to which a modeling approach departs from existing regulatory standards or expectations.

Under the framework, models that play a larger role in decision-making or carry greater potential consequences if wrong will be expected to undergo more extensive evaluation.

The guidance applies to both established and emerging modeling technologies. Examples cited in the document include population pharmacokinetic and pharmacodynamic modeling, physiologically based pharmacokinetic modeling, exposure-response analyses, model-based meta-analyses, quantitative systems pharmacology, disease progression models, agent-based simulations, and artificial intelligence and machine learning approaches.

ICH emphasized that the principles are intended to remain applicable as new computational technologies emerge.

A key theme throughout the guidance is the importance of early communication between sponsors and regulators. The document encourages companies to incorporate modeling strategies early in development and engage regulators before analyses are conducted to ensure that necessary data are collected and that proposed approaches align with regulatory expectations.

To facilitate these discussions, the guideline introduces a standardized "assessment table" designed to document key elements of a modeling strategy, including the regulatory question being addressed, model risk assessments, technical evaluation criteria, and conclusions regarding whether model outputs qualify as evidence for decision-making.

The guidance also outlines expectations for model evaluation, emphasizing three core elements: verification, validation, and applicability assessment.

Sponsors are expected to demonstrate that computer code is functioning correctly, that mathematical assumptions are properly implemented, and that calculations are accurate. Models must also be validated against available data and shown to be appropriate for their intended purpose.

The guideline recommends assessing model robustness through sensitivity analyses, evaluating uncertainty, documenting limitations, and, where feasible, conducting external validation using independent datasets.

For artificial intelligence and machine learning applications, sponsors are advised to address risks such as overfitting and other model-specific concerns.

To improve regulatory review, the guidance recommends that sponsors prepare detailed Model Analysis Plans before conducting analyses and submit comprehensive Model Analysis Reports documenting methodology, assumptions, results, validation activities, and conclusions.

Companies are also encouraged to provide regulators with underlying datasets, modeling code, simulation files, and other supporting materials to enable independent assessment.