Between 2020 and 2024, the number of unique US sites where Phase I clinical trials for non-small cell lung cancer (NSCLC) were conducted decreased by 44% and became increasingly concentrated in the top 20 largest clinical trial sites, primarily located in large cities, according to results presented at the American Association for Cancer Research (AACR) Annual Meeting 2026, held April 17-22.
The U.S. Food and Drug Administration (FDA) has been working for decades to improve representation in clinical trials, and in the last year, the FDA’s Oncology Drug Advisory Committee (ODAC) held several meetings to address the under-enrollment of U.S. patients in clinical trials and the underrepresentation of certain populations. Brittany Avin McKelvey, Ph.D., senior director of regulatory policy at the LUNGevity Foundation, explained that the committee voted against recommending the drug for approval, in part because of questions about how relevant and representative the trial results were to the American public. To better understand the concerns raised by ODAC, she and her colleagues examined the distribution of clinical trial sites within and outside the United States.
“To fully determine the safety and effectiveness of a treatment and inform its optimal use, it is essential that clinical trial populations reflect the end-user population as accurately as possible,” McKelvey said. “Thus, by better understanding where clinical trials are conducted and how site participation changes over time, we can ensure we support policies that promote the representativeness and applicability of trial results to the American public.”
McKelvey and her team focused on a Phase I clinical trial in NSCLC. They identified all industry-sponsored phase I trials of interventional therapies published on ClinicalTrials.gov from January 2020 to December 2024. Unique site locations were determined based on addresses with at least one trial associated with a unique National Clinical Trials (NCT) number. A study was defined as having started if the start date, which indicates when the first patient was enrolled, was within the study’s defined period. A trial instance was defined as the start of a unique trial at a unique site. With open trials, there can be multiple trial instances at different sites.
Overall, during the study period, 555 trials were initiated in 47 countries for a total of 8,393 trials, with the majority in the United States (45%), China (11%), Spain (8%), South Korea (5%), France (4%), and Australia (4%).
McKelvey et al. found that the number of U.S. clinical trials increased from 819 in 2020 to 955 in 2022, but fell to 566 by the end of 2024. Similar trends were seen globally, with China (196 to 95 cases), Spain (105 to 80 cases), South Korea (80 to 66 cases), and France (48 to 47 cases) also decreasing during the study period. period. The only countries with significant increases in cases were Australia (64 to 76), Japan (32 to 49), and Brazil (9 to 29).
Additionally, the number of unique trial sites in the United States decreased from 395 to 223 during the study period. However, the number of trials at the top 20 sites with the highest trial volume remained stable over this period, with each of these sites starting a median of 7 to 11 trials per year over a 5-year period. Most of these top 20 sites were located in large cities with an average population of over 1.9 million people.
“Our analysis shows a worrying trend toward consolidating trials at top-performing sites in the United States, further supporting concerns raised at the recent FDA ODAC meeting about site saturation and a conflict with FDA’s calls to decentralize and move trials to the community,” McKelvey said. “This geographically concentrated clinical trial environment may further limit access to clinical trials.”
McKelvey noted that several factors may be contributing to this decline, including the fact that phase I trial protocols are becoming increasingly complex and the safety profile of new oncology treatments creates substantial infrastructure and expertise requirements that may be prohibitive for small centers. The increased regulatory and compliance burden associated with initiating and conducting clinical trials may be an additional barrier for small centers.
“The competitive environment for oncology trial registration can favor well-established sites with a track record, creating a self-reinforcing cycle in which high-performing sites attract more trials, develop better expertise, and become increasingly preferred by sponsors, while low-volume sites struggle to maintain the infrastructure and expertise necessary to compete in trials,” McKelvey said.
However, McKelvey noted that not all low-volume sites experienced a decline in trial instances over the study period. She explained that the lower-volume site, which is part of the National Cancer Institute Community Oncology Research Program (NCORP), was able to maintain nearly all trial instances over the observed period compared to other low-volume sites.
“This may suggest that the infrastructure support and integration of NCORP sites into research networks may buffer these community sites from factors that promote integration,” McKelvey said. “Overall, addressing this consolidation and ensuring the identification of appropriate sites for clinical research will require a coordinated effort to reduce barriers to site activation, provide infrastructure support, and, in some cases, introduce policy incentives that promote diversification of the clinical trial portfolio in the United States.”
Limitations of this study include the heterogeneity of ClinicalTrials.gov data. This made it difficult to determine whether sites with relatively similar names within the same zip code should be treated as institutions with satellite locations within the community or grouped together as a single site. Furthermore, because patient enrollment numbers are not published, the number of sites actively enrolling patients may be smaller than the number of trials examined in this study. Finally, a focus on NSCLC and phase I trials may not fully represent patterns in other cancer types or late-stage trials. McKelvey added that her team is considering research into late-stage clinical trials that would allow for broader participation by less experienced centers to compare analyses, given the known safety profile.
This research was funded by the LUNGevity Foundation through an unrestricted grant.
sauce:
American Association for Cancer Research
