Title: “Information theory in dose-finding trials”

Speaker: Prof Thomas Jaki, Lancaster University

Abstract: In this talk we consider novel developments in information theory and their application to dose-finding. Focus will be given on a novel criterion for the allocation of patients in Phase I dose-escalation clinical trials aiming to find the maximum tolerated dose (MTD).

Conventionally, using a model-based approach the next patient is allocated to the dose with the toxicity estimate closest (in terms of the absolute or squared distance) to the maximum acceptable toxicity. This approach, however, ignores the uncertainty in point estimates and ethical concerns of assigning a lot of patients to overly toxic doses. On the basis of information theory, we propose a criterion which accounts for both of these issues. The criterion requires a specification of one additional parameter only which has a simple and intuitive interpretation. We incorporate the proposed criterion into the one-parameter Bayesian continual reassessment method (CRM) and show, using simulations, that it results in the same proportion of correct selections on average as the original design, but in fewer mean number of toxic responses. We then extend the ideas to utilize similar concept for other complex phase I settings such as dose-schedule trials.