| In Phase I oncology trials, the main objective is typically to determine the maximum tolerated dose (MTD), the highest dose of a drug that does not cause unacceptable side effects. The MTD is determined by testing increasing doses on patient cohorts until the highest dose with acceptable side effects is found. To determine the MTD, many different dose escalation designs have been developed, such as the CRM, BLRM, mTPI-2, and BOIN, each of which has improved statistical properties over the traditional A+B class of designs. Yet most published Phase I trials still use a traditional 3+3 design. This dissertation proposes a robust paradigm of conducting Phase I cancer clinical trials that involves simultaneously running multiple designs to obtain a robust MTD. This paradigm will provide more confidence to clinicians and regulators in the selection of the dose for further drug development. The implementation of this paradigm with an application to the IL-21 denenicokin clinical development program is the focus of this dissertation. |
