Use of Single-Arm Trials for US Food and Drug Administration Drug Approval in Oncology, 2002-2021.

Importance: Single-arm trials have allowed for transformative therapies to be made available to patients expeditiously. However, using single-arm trials to support drug approval presents several challenges that must be carefully considered. Observations: Between January 1, 2002, and December 31, 2021, the US Food and Drug Administration granted 176 new malignant hematology and oncology indications based on single-arm trials, including 116 accelerated approvals (AAs) and 60 traditional approvals. Overall, 87 approvals (49%) were for new molecular entities or original biologics and 89 (51%) were supplemental indications. Response rate (RR) was the most common end point used to support approval in these single-arm trials (173 of 176 [98%]). Of the 116 AAs based on single-arm trials, 45 (38%) fulfilled their postmarketing requirement to verify clinical benefit, 61 (52%) are pending verification of benefit, and 10 (9%) were withdrawn from the market as of December 31, 2021. Most (56 of 61 [92%]) AAs based on single-arm trials pending verification of benefit occurred during the previous 5 years and have ongoing confirmatory trials as of December 2021. Conclusions and Relevance: Single-arm trials have been a common development strategy to support regulatory approval as early-stage expansion cohorts with promising durable RRs have become more prevalent. In the appropriate context, single-arm trials using durable RRs can allow patients expedited access to novel therapies and will continue to serve a role in advancing drug development in oncology. However, single-arm trials have a smaller noncomparative safety data set, inability to use time-to-event end points, and other limitations that require careful consideration within the context of the disease and available therapies. The randomized clinical trial remains the preferred approach in clinical investigation.