Optimal utility-based design of oncology clinical development programmes

This project aims at developing a probabilistic framework for comparing sequences of clinical trial designs in oncology. Its main objective is to provide a practical method for improving efficiency of clinical development programmes, on behalf of current and future oncology patients. This project consists of three parts, focussing in on the development of probabilistic models for clinical trial endpoints relevant to early and late phase trials and on the derivation of survival statistics maximising patient utility. The probability of success and expected utility of alternative clinical development strategies will be compared using the developed models. Probabilistic dynamic programming will be leveraged to define mechanisms for updating clinical development strategies as data become available over time. The unique combination of expertise made available through this project will provide the sponsored student with an ideal set of skills relevant to develop a career in academic or industrial research and development settings.