Quantitative Model Informed Drug Development

Model-Informed Drug Development (MIDD) is an approach that uses quantitative modeling and simulation to integrate nonclinical, clinical, drug, and disease data to inform drug development decisions. MIDD approaches such as physiologically based pharmacokinetic (PBPK), population pharmacokinetic (PopPK), PK/PD, exposure-response, and quantitative systems pharmacology (QSP) modeling can help predict drug behavior, optimize dosing, design clinical trials, and support regulatory decision-making.

MIDD improves drug development by using existing data to reduce uncertainty and inform decisions throughout a program. Modeling and simulation can help identify appropriate doses, predict drug exposure in different patient populations, evaluate drug-drug interactions, optimize clinical trial design, and assess scenarios that may be difficult or impractical to study directly. By answering important development questions earlier, MIDD can help sponsors design more informative studies and make evidence-based decisions about next steps.

PBPK, PopPK, and PK/PD modeling answer different but complementary drug development questions. PBPK modeling uses drug-specific properties and physiological information to predict how a drug is absorbed, distributed, metabolized, and eliminated (ADME) in the body. Population PK (PopPK) modeling analyzes pharmacokinetic variability across patients and identifies factors such as age, body weight, organ function, or concomitant medications that may affect drug exposure. PK/PD modeling links drug exposure to pharmacological, efficacy, or safety responses.

MIDD should be considered as early as possible in drug development and updated as new data become available. Modeling can begin with preclinical and prior knowledge to inform first-in-human and early clinical strategies, then incorporate emerging clinical data to refine dose selection, trial design, exposure-response relationships, and predictions for specific populations. Early MIDD planning also helps ensure that studies generate the data needed to support future modeling and regulatory questions.

MIDD can provide quantitative evidence to support regulatory decisions throughout development and in submissions such as INDs, NDAs, and BLAs. Depending on the context of use, modeling may support dose and regimen selection, exposure-response analyses, assessment of intrinsic and extrinsic factors, drug-drug interaction evaluations, predictions in specific populations, and clinical trial design. Regulatory acceptance depends on the model’s intended use, the quality and relevance of the supporting data, and whether the model is sufficiently evaluated for the question it is intended to address.

The data required for MIDD depend on the modeling approach and the drug development question being addressed. Inputs may include physicochemical and in vitro ADME data, nonclinical pharmacokinetic and pharmacodynamic data, clinical PK and PD data, efficacy and safety endpoints, patient characteristics, disease information, and prior knowledge from related compounds or published literature. High-quality, fit-for-purpose data and a clearly defined question are essential for developing and evaluating models that can reliably inform decisions.

ICH M15 establishes a harmonized framework for planning, evaluating, documenting, and communicating Model-Informed Drug Development evidence. The guideline emphasizes defining the question of interest and context of use, assessing model risk and impact, establishing an appropriate model analysis plan, evaluating model credibility, and clearly reporting the resulting MIDD evidence. For drug developers, ICH M15 makes early planning and alignment between modeling strategy, data generation, and regulatory objectives increasingly important when MIDD evidence will support regulatory decision-making.