Abstract:
|
In oncology, early phase trials with targeted therapeutics have mandated the development of novel study design strategies as the historical clinical trial design paradigm is no longer relevant. Specifically, the dose is typically determined based on the probability of severe toxicity observed during the first treatment cycle, although patients continue to receive treatment for multiple cycles. In addition, the toxicity data from multiple types and grades are summarized into a single binary outcome of dose-limiting toxicity. To overcome these limitations, a normalized total toxicity profile (nTTP), a quasi-continuous endpoint, was developed to take into account clinical multidimensionality (multiple grades and types) of toxicities. A dose finding design accounting for longitudinal repeated measures of nTTP over multiple treatment cycles, accounting for cumulative toxicity will be discussed. An extension of this design to jointly model a continuous efficacy outcome and nTTP over multiple treatment cycles will also be presented. Stage I of our design uses toxicity data to perform dose-escalation and identify a set of initially allowable (safe) doses; stage II of our design incorporates an efficacy outcome to update the set of allowable doses and randomizes the new cohort of patients to the allowable doses with emphasis towards those with higher predicted efficacy. Stage III combines all data from all treated patients to make final recommendations.
|