Evaluating the role of phase I expansion cohorts in oncologic drug development.

Importance Use of expansion cohorts (EC) in phase I trials is increasing. However, the utility of phase I EC in aiding drug development is unclear. We sought to determine factors associated with the inclusion of EC in phase I studies and the impact of EC on subsequent clinical development. Methods We performed a systematic review of all phase I trials published in the Journal of Clinical Oncology between June 2004 and May 2014. Presence of an EC, number of participants, funding source, class of agent, tumor type, and maximum tolerated dose (MTD)/recommended phase 2 dose (RP2D) were identified. Subsequent conduct of phase II studies and FDA approval of the study agent was also assessed. Results We identified 252 phase I studies. An EC was included in 105 studies. Average accrual on EC studies was 47 compared to 31 in studies without EC (p < 0.0001). There was no impact of time on the inclusion of EC. Only 4 % of phase I studies with an EC provided sample size justification. Source of funding had the only significant association with inclusion of EC. Addition of a phase I EC did not impact the phase I MTD/RP2D, subsequent phase II trial, or FDA approval. Conclusion The importance of including an EC in phase I trials is subject to ongoing debate. Our results demonstrated little benefit to including EC in phase I studies. These findings support that innovative design strategies are needed to optimize the utility of EC in phase I studies.