Importance to FDA

Model-informed product development (MIPD) aims to integrate information from diverse data sources to help decrease uncertainty and lower failure rates, and to develop information that cannot or would not be generated experimentally. MIPD encompasses model-informed drug development (MIDD), an approach that involves developing and applying exposure-based biological and statistical models derived from preclinical and clinical data sources to inform drug development or regulatory decision-making. FDA’s MIDD Pilot Program facilitates integrating MIDD into more drug applications and advancing its use, and addresses some of FDA’s goals under the Prescription Drug User Fee Act VI (PDUFA VI), included as part of FDA Reauthorization Act of 2017 (Public Law 115-52).

MIDD applications include potential contributions toward predicting clinical outcomes; informing clinical trial designs and efficiency; supporting evidence for efficacy; optimizing drug dosing/therapeutic individualization; predicting product safety and evaluating potential adverse event mechanisms; product performance optimization; and informing policy.

FDA has committed resources to transforming computational modeling from a valuable scientific tool to a valuable medical device regulatory tool and to developing mechanisms to rely more on digital evidence. FDA continues to advance these methodologies and techniques to ensure the benefits of product innovation and more rapid introduction of life-saving technology to our nation’s patients.

MIPD applies to innovations in processing of foods, which rely on modeling and simulation to ensure foods are safe and wholesome for consumption. Using modeling and simulation-based approaches helps to examine situations that cannot easily be studied experimentally, such as retroactive studies of foodborne outbreaks or contamination events; prospective studies of intended or unintended changes in the food safety or nutrition system (e.g., food, environment, processing, handling, consumption, or compliance); or system sensitivity and vulnerability assessments.

Mathematical models developed using modeling and simulation-based approaches evaluate specific conditions to study systems based on different levels of exposure, chemical toxicities, growing, harvesting, or processing practices, levels of compliance with Good Agricultural Practices, Current Good Manufacturing Practices, Food Code, or other regulations, proposed mitigations or controls and various failure/outbreak scenarios. Applying modeling approaches to food processing improves risk assessment of pathogens and toxins in foods and predicting risks of illness in food categories. Compliance activities rely on modeling to inform regulatory decisions and to ensure regulated stakeholders meet legal requirements. In addition, FDA’s Catalog of Regulatory Science Tools collates innovative science-based approaches to help improve the development and assessment of emerging medical technologies. Tools in the catalog include phantoms, methods, and computational models and simulations.

Examples

FDA advances the use of modeling and simulation in product development:

• Evaluating dose selection and refinement, treatment duration, response measures, safety evaluations and assessing the combined effect of drug interactions, kidney and liver failure in patients in the absence of dedicated trials.
• Developing means to facilitate software development that assists in analyzing medical imaging and diagnostics. For example, FDA developed the Virtual Imaging Clinical Trials for Regulatory Evaluation (VICTRE) multi-modality anthropomorphic breast phantom. VICTRE is a digital breast phantom with modifiable parameters, including phantom voxel size (resolution) and breast density in the area of medical imaging and diagnostics.
• Making available The Virtual Family: A set of anatomically correct whole-body computational models based on multimodal imaging.
• Credibility of Computational Models Program: Research on Computational Models and Simulation Associated with Medical Devices
• Developing multiple (Quantitative) Structure-Activity Relationship models that use a range of in silico tools to predict toxicological outcomes, such as genotoxicity, carcinogenicity, and drug-induced liver injury. Other research will explore additional model endpoints and expand previous models with newly published data.
• Using many in vitro techniques to identify drug‒drug interactions and drug‒target interactions that may be clinically relevant. For example, patch clamp techniques evaluate the effects of drugs on cardiac ion channels and provide physical evidence of drug interactions with a variety of transporters, enzymes, and receptors that may be used in regulatory decision-making.
• Developing mechanistically informed models based on pharmacokinetics that predict the disposition of chemicals, medical products, and their metabolites in the body and could be used for examining their potential for biopersistence.
• Projecting population level effects of a potential public health standard (i.e., nicotine standard) on the prevalence of tobacco use, tobacco-related mortality, and life-years gained.
• With respect to cosmetics, an approach of combining predictive computer modeling and in vitro test methods has been developed in order to minimize animal testing when evaluating skin sensitization potential of cosmetic ingredients.