|
Abstract:
|
Alternatives to the well-known 3+3 dose escalation design for phase I oncology trials abound. Many have proven to be safer and to more accurately identify the maximum tolerated dose (MTD). The modified toxicity probability interval (mTPI) design has become established as one of these reliable alternatives to 3+3. mTPI is an adaptive design that is guided by the posterior inference for a simple Bayesian model. Additionally, mTPI's simple design and implementation make it favorable to those who shy away from other model based designs that require computationally intensive calculations during patient enrollment. The MTD is determined using isotonic regression based on toxicity data for all cohorts. While the benefits and applications of mTPI have been described in literature for single agent dose escalation trials, very little has been done towards evaluating mTPI in combination, dual dose escalation trial designs. In this presentation, methods are proposed for the implementation of dose escalation and MTD determination when two investigational products are involved. Incorporated are recommendations for capping sample size per cohort. The methods are evaluated through simulation.
|
