Modelling and simulation now support dose selection, paediatric extrapolation, drug interaction claims and trial design across most development programmes. ICH M15, adopted at Step 4 on 29 January 2026, gives regulators and sponsors a shared framework for planning, evaluating and documenting model-informed drug development (MIDD). Its core message is simple: the weight a model can carry depends on the question it answers and on what happens if it is wrong.
What M15 covers
M15 sets out general principles and good practice for MIDD, rather than technical rules for individual methods. It applies across approaches such as population pharmacokinetics, physiologically based pharmacokinetic (PBPK) modelling, exposure-response analysis and quantitative systems pharmacology, and it recognises that machine learning methods may be used alone or in combination with them. It also gives recommendations on how MIDD evidence should be reported and submitted.
The assessment framework
At the heart of M15 is a structured way to decide how much evidence a model needs. Sponsors are expected to define:
- The question of interest: the specific development or regulatory question the model is meant to answer.
- The context of use: how the model output will be used to answer that question.
- Model influence: how much weight the model carries relative to other evidence. The final guideline rates this as low, medium or high, with high influence where the model is the sole support for a decision.
- Consequence of a wrong decision: the impact on patients or on the decision if the model’s prediction is wrong.
- Model risk: the combination of influence and consequence, which drives how much verification, validation and documentation is needed.
What changes for sponsors
- MIDD planning should begin early, with the question and context of use agreed before the analysis is run
- Modelling analysis plans and reports should follow a consistent, assessable structure
- High-risk uses, such as replacing a clinical study, need more evidence of model credibility
- Scientific advice is the natural place to agree context of use and model risk with EMA or FDA
How it connects to other guidance
M15 sits alongside existing regional guidance, such as EMA’s guidance on PBPK reporting and the paediatric extrapolation principles of ICH E11A. It provides the common language; method-specific expectations continue to come from those documents and from regulators’ feedback in scientific advice.
How PQRA helps
PQRA helps sponsors plan regulatory strategy around modelling evidence, prepare scientific advice requests to EMA and national authorities, and position MIDD outputs correctly within Module 2 and Module 5 of the CTD.


No responses yet