Abstract A088: Evolution of the use of nonrandomized (NRCT) versus randomized (RCT) phase 2 trial designs preceding phase 3 registration trials for oncology drugs approved from 2004-2016

Background: In oncology drug development, prior to conduct of phase 3 registration trials, phase 2 RCTs have been conducted in many instances to provide initial evidence of a statistically significant effect on patient outcomes. However, the high antitumor activity of recent novel agents observed at early phases of development may obviate the need to demonstrate clinical activity of a drug via an RCT in hase 2. Our objective was to identify differences in the frequency with which NRCT or RCT designs were used in the phase 2 setting for oncology drugs approved from 2004-2016. Methods: We identified all hematology/oncology drugs approved by the United States Food and Drug Administration (FDA) from 2004-2016. We then searched Citeline® Trialtrove database and clinicaltrials.gov to identify the phase of trial leading to registration as well as any stand-alone phase 2 studies conducted with these agents either before or as the registration trial. For agents with phase 2 studies conducted prior to registration, we analyzed the phase 2 trial design (NRCT or RCT) and primary endpoint, and categorized by year of approval and drug type [nonimmunologic/targeted drug (NIO), immunologic drug (IO) or chemotherapeutic drug (CT)]. We then calculated percentages of RCT or NRCT by year and drug type to detect changes in frequency of trial design over time. Results: A total of 102 FDA-approved agents across 29 oncology indications were evaluated. Phases of trials leading to approval were: phase 1 (n = 2); phase 1/2 (n = 5); phase 2 (n = 35); phase 3 (n = 60). Among 77 FDA-approved agents with a stand-alone phase 2 study conducted prior to registration, drug types were: 15 CT, 50 NIO, and 12 IO. 24 agents were approved from 2004-2010, while 53 were approved from 2011-2016 (of those, 18 in 2015-2016). Of the phase 2 studies conducted prior to registration, 13 were RCT and 64 were NRCT. Primary endpoints in these phase 2 studies were: ORR (n = 57), OS or PFS (n = 8), major cytogenetic response (n = 5), TTP (n = 2), or other (n = 5). Approximately 50% of RCT (7/13) evaluated time to event endpoints such as OS, PFS, or TTP; however, for NRCT, only 5% (3/64) used a survival-time endpoint; 81% (52/64) of NRCT used ORR for primary endpoint. With regard to type of oncology drug, no specific trend in NRCT vs. RCT was observed: 86% (43/50) of NIO drugs used NRCT in phase 2; 75% (9/12) of IO drugs used NRCT in phase 2; and 80% (12/15) of CT drugs used NRCT in phase 2. For drugs approved 2004-2010, NRCT were used in phase 2 for 79% of agents (19/24 agents), and from 2011-2016, NRCT were used in phase 2 for 85 % of agents (45/53). Among NIO approved from 2004-2010, 67% of agents (10/15) used NRCT in phase 2, but from 2011-2016, 94% (33/35) used NRCT in phase 2. From 2010-2016, 82% (9/11) of IO drugs approved used NRCT in phase 2, and 82% (9/11) used ORR as a primary endpoint. Conclusions: From 2004-2016, a vast majority of FDA-approved oncology drugs used NRCT design in the phase 2 setting. The percentage of NRCT in phase 2 for approved drugs has increased in recent years, especially for approved NIO drugs. Most phase 2 NRCT for approved drugs used a primary endpoint other than survival, such as ORR. As novel oncology drugs targeting the right population enter clinical trials, demonstrating statistical improvements in survival in a phase 2 RCT is less frequently pursued prior to conduct of registration trials. Citation Format: Kelly K. Curtis, Laura Vidal Boixader, Jozsef Palatka, Liat Vidal, Nicholas Kenny, Keren R. Moss. Evolution of the use of nonrandomized (NRCT) versus randomized (RCT) phase 2 trial designs preceding phase 3 registration trials for oncology drugs approved from 2004-2016 [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2017 Oct 26-30; Philadelphia, PA. Philadelphia (PA): AACR; Mol Cancer Ther 2018;17(1 Suppl):Abstract nr A088.