Structural Identifiability and Indistinguishability Analysis as Tools for Quantitative and Systems Pharmacology to Support the 3Rs

点击翻译按钮获取中文摘要

Animals are extensively used in toxicology testing to assess safety and risk of the process examined on humans and the environment. It is estimated that over 1 million animals are used across Europe each year in toxicology tests with around one fifth of all animals used in the UK involved in such tests. The mainstream use of valid, robust mathematical models that can be used with greater fidelity and confidence than those currently used in practice to accurately simulate the outcomes of such tests would therefore clearly have huge impacts on the number of animals required in the future, supporting the replacement and reduction of the use of animals in toxicology testing experiments and providing step changes in animal welfare. This proposal aims to overcome certain shortcomings that exist in the modelling of toxicological processes by introducing and applying established and novel systems modelling tools and techniques namely structural identifiability and indistinguishability analyses. These tools have proven ability to improve the fidelity of models in terms of confidence in parameter estimation and optimising experiment design. These techniques will be applied to exemplar problems in toxicology that will provide a more robust generic analytical framework to improve parameter estimation, sensitivity, robustness, reliability and validity of these models. In turn this step change in modelling will enhance experiment design, improve risk and safety assessment and thus help support the replacement, reduction and refinement of animal use. This research will be supported by industry (AstraZeneca) and will be applied to models for the improvement of both human and animal safety. This work will provide new tools and seed the emergence of a new field, namely Quantitative and Systems Toxicology.