Interpreting Single-Arm Trials in Rare Driver Populations

Matthew A. Gubens, MD, MS, Julia K. Rotow, MD, and Estelamari Rodriguez, MD, MPH, discuss how to interpret the single-arm trials supporting ROS1 tyrosine kinase inhibitors in ROS1-positive non-small cell lung cancer (NSCLC). Dr. Gubens notes that agents gain approval in rapid succession without head-to-head randomized data, so clinicians rely on cross-trial comparisons with their inherent limitations. He calls durability the top consideration in patient conversations, followed by tolerability when efficacy appears similar. Dr. Rotow explains that the rarity of ROS1 fusions makes large randomized phase 3 trials impractical, since mature data could take 6 to 10 years. She relies on single-arm studies while remaining cautious about differences in enrolled populations and follow-up maturity. Dr. Rodriguez notes that the largest datasets include about 180 patients and some intracranial data come from only 10 patients; however, response rates for newer agents approach 80%. She views intracranial activity and landmark analyses at 12 and 18 months as key signals of durable benefit.