Why Aren’t Advanced Cancer Patients Participating in Clinical Trials?

Participation in clinical trials among patients with advanced cancers has modestly increased over time, but remains low, with a minority of physicians and practices accounting for most enrollments, a retrospective cohort study found.

Among over 205,000 patients with advanced or metastatic cancers, 5.2% participated in a clinical trial from 2011 through 2024, increasing from 3.9% in 2011 to 6.8% in 2024, reported Changchuan Jiang, MD, MPH, of the University of Texas Southwestern Medical Center in Dallas, and colleagues.

Yet, “this growth was underpinned by profound and widening variation at physician and practice levels,” they wrote in JAMA Network Open.

Of the 1,664 physicians included in this study, 23.4% accounted for 83.6% of all enrollments, while 19.6% of practices accounted for 80.5% of all enrolled patients. About 30% of physicians enrolled no patients.

Physician-level (12.3%) and practice-level (15.1%) factors explained a greater share of variation in trial participation compared with patient-level characteristics (11.4%), with physician- and practice-level contributions increasing over time.

Access to these trials “is governed by a minority of highly engaged super enroller oncologists or practices,” Jiang and team observed. “This finding signifies that a culture of clinical trial enrollment at the practice level is as vital as specialized skill and workflow, necessitating physician- and practice-level interventions to collectively increase patient participation in therapeutic clinical trials.”

In an accompanying commentary, Raymond Osarogiagbon, MBBS, of the Baptist Cancer Center in Memphis, Tennessee, noted that “despite stated willingness, the vast majority of adults who receive cancer treatment in the U.S. do not participate in a clinical trial.”

“Overcoming this stark reality presents one of the greatest prevailing opportunities to improve population-level oncology outcomes,” he added.

This study used data from the Flatiron Health Research Database and included 205,743 patients with advanced or metastatic cancers diagnosed and treated between January 2011 and December 2024, by 1,664 physicians across 107 trial-available U.S. oncology practices. Mean age at advanced cancer diagnosis was 67.5, and 52.2% were men.

The most common diagnoses included non-small cell lung cancer (29.5%), colorectal cancer (13.5%), and breast cancer (13.1%); 75.2% of patients had a baseline Eastern Cooperative Oncology Group performance status score of 0 to 1.

Patients treated by more subspecialized physicians (1-2 or 3-4 cancer groups vs 5 or more cancer groups) had significantly higher odds of trial participation (OR 4.11, 95% CI 2.53-6.68, and OR 2.38, 95% CI 1.37-4.14), as did those treated by physicians with greater prior trial enrollment activity, with a clear dose-response association (1-2 enrollments: OR 1.38, 95% CI 1.23-1.54; ≥3 enrollments: OR 3.59, 95% CI 2.99-4.31).

In addition, practices with high research engagement also had higher odds of trial participation (OR 2.68, 95% CI 2.12-3.39), while practices serving patients with the most socioeconomic disadvantages had lower odds of participation (OR 0.42, 95% CI 0.26-0.69).

At the patient level, older age, lower area-level socioeconomic status, and poorer ECOG scores were associated with decreased odds of enrollment. Black patients had significantly lower odds of participating in trials compared with white patients (OR 0.74, 95% CI 0.62-0.88).

Seventeen practices with no observed trial enrollment were excluded from the study population. Their patients were more likely to be Hispanic and reside in low socioeconomic status areas.

“In other words, the places where such patients received their cancer treatments were unlikely to have clinical trials,” noted Osarogiagbon in his commentary. “This illustrates possibly the most straightforward opportunity for enrolling more diverse patients into clinical trials: building clinical trial infrastructure and opening clinical trials at facilities where the population of interest already receives their care.”

Jiang and colleagues acknowledged that their study had several limitations, including the fact that the database they used represents community oncology settings and may not generalize to academic or other settings.