Advanced Cancer Trial Participation Varies Substantially Across Physicians and Practices

A cohort study published in JAMA Network Open examined clinical trial participation among 205,743 adults with advanced or metastatic cancer diagnosed and treated during 2011–2024. Flatiron Health Research Database records covered 1,664 physicians at 107 US oncology practices with trials available. Participation was defined as receipt of a clinical study drug; researchers used regression models to assess patient-, physician-, and practice-level variation and associated characteristics.

Overall, 10,785 patients (5.2%) participated, with annual participation increasing from 3.9% in 2011 to 6.8% in 2024. Just 389 physicians (23.4%) accounted for 83.6% of enrollments, while 21 practices (19.6%) accounted for 80.5%. Nearly 30% of physicians enrolled no patients. Physician-level and practice-level factors explained 12.3% and 15.1% of variation, respectively, compared with 11.4% for observable patient characteristics.

Physician subspecialization, previous trial enrollment activity, and practice research engagement were associated with higher participation. Enrollment activity in both preceding years was associated with an odds ratio of 1.58 (95% CI, 1.43–1.76), and practice research engagement with an odds ratio of 2.68 (95% CI, 2.12–3.39).

The findings highlight research infrastructure and physician engagement as potential targets for improving accrual. However, the observational design cannot establish causality, and restriction to trial-available practices limits generalizability. The study-drug endpoint also does not capture participation in trials without investigational drug treatment.