|
Abstract:
|
On Phase 1 design for oncology drugs, additional patients may be enrolled in an effort to increase the safety information at the presumed maximum tolerated dose (MTD) from the dose escalation portion. In this study, we set up statistical rules to determine the number of patients to be enrolled based on two desirable properties: neither too toxic nor nontoxic. In addition, the escalation/de-escalation/MTD declaration rules for the expanded dose cohort (i.e., n > 6) are established after the result from each patient is available. Both frequency method and Bayesian method can be incorporated into the design. Simulation studies are performed to further examine the appropriateness of these two methods.
|
