Thoracic Radiation Oncology Clinical Trial Accrual and Reasons for Nonenrollment: Results of a Large, Prospective, Multiyear Analysis.

ABSTRACT

PURPOSE: Clinical trials are considered the gold standard in evidence-based medicine, yet few patients with cancer ultimately enroll. Here we examine patients screened for thoracic radiation oncology clinical trials to better understand enrollment trends. METHODS AND MATERIALS A prospective database tracking screening and enrollment for patients referred for thoracic radiation oncology consultation at our institution from 2016 to 2019 was evaluated. Proportional enrollment rates, patient and disease characteristics, self-reported socioeconomic factors, and reasons for ineligibility or nonenrollment across 17 radiation therapy trials were compared. RESULTS: Enrollment data on 2372 patients were available for analysis. Of these patients, 40.0% (949) were deemed "not eligible" (NE) for any trial or were unwilling to be further screened. Reasons for ineligibility included stage (44%), histology (13%), radiation therapy not indicated (12%), patient decision (7%), and enrollment in a competing medical or surgical oncology trial (5%). The remaining 60.0% (1423) were "potentially eligible" (PE) for one or more trials. Most had non-small cell lung cancer (71%) or esophageal cancer (16%), and there were significantly fewer stage IV PE (29%) versus NE (49%) patients (P < .0001). Of 2372 patients, 281 (11.9%) enrolled. Notable reasons for nonenrollment were inclusion and exclusion criteria (58%), patients declining enrollment (14%), and physician decision (5%). The proportion of white patients was higher in the PE versus NE group (82.5% vs 75.8%; P < .001). Additionally, white race (87.9% vs 81.2%; P = .008), English language preference (96.4% vs 92.9%; P = .032), and non-Hispanic/Latino ethnicity (94.0% vs 90.1%; P = .042) were significantly different in enrolled versus nonenrolled PE patients. CONCLUSIONS: Only 12% of patients screened for radiation therapy trials ultimately enrolled, and more than two-thirds had no trial available or were found ineligible. In addition, 19% of potential eligible patients did not enroll because the patient or physician declined. Future trials may benefit from pragmatic designs with more inclusive enrollment criteria and multidisciplinary engagement of referring providers.