Recent developments in adaptive designs for Phase I/II dose-finding studies.

Cancer dose-finding trials aim at assessing the maximum tolerated dose (MTD) of a new treatment or combination. Owing to ethical constraint, they are based on adaptive designs, sequentially allocating patients to increased doses on the basis of previous responses. More recently, the concept of MTD has been extended to the largest concept of most successful or most desirable dose, based on efficacy criterion under toxicity restrictions. The aim of this paper is to present three main approaches proposed to estimate such a dose, in the setting of Phase I/II trials. Two case-studies allow to illustrate these new approaches.