Tools to enable animal to human translation: Assessing the value of disease models

The value of animal research has been increasingly debated owing to high failure rates in drug development. A potential contributor to this situation is the lack of predictivity of animal models of disease. Until recently, most initiatives focused on well-known problems related to the poor reproducibility and reporting standards of animal studies. Only now, more methodologies are becoming available to assess how translatable the data from animal models of disease are. In this thesis, we present the Framework to Identify Models of Disease (FIMD) to evaluate the similarities between animal models of disease and patients. FIMD delineates a straightforward procedure to validate, assess and compare disease models. We describe the development, validation and application of FIMD in detail across two indications: type 2 diabetes (pilot study) and Duchenne Muscular Dystrophy (DMD) (complete validation). We complement and validate the pilot study results with a systematic review and meta-analysis of animal studies of type 2 diabetes, investigating animal model predictivity at the therapeutic level. Next, we use the IB-derisk tool to integrate animal pharmacokinetic and pharmacodynamic data from Investigator’s Brochures with the corresponding human data from clinical trials, evaluating the predictive value of disease models at a pharmacological level. Finally, we promote a discussion on the application, implementation and impact of FIMD, systematic reviews and meta-analysis, and the IB-derisk tool. We elaborate on the role of relevant stakeholders and possible actions to improve the quality of animal research in drug development. When combined, these methods provide a holistic, systematic and evidence-based manner to select animal models of disease based on their ability to predict the human response.