When it comes to site selection, we’re looking for the wrong traits and asking the wrong questions.
In a field characterized by its cutting-edge thinking, oncology research faces ongoing struggles in finding sites that can perform in trials.
However, from AI breakthroughs to personalized vaccines, revolutionary developments are redefining oncology research right now. Yet, as we continue to propel the industry into a new era of innovation, major groundwork is needed to address the issues still holding back the development of oncology treatments. More cancer trials are entering the market than ever before and rather than adapting to that increased volume, our flawed approach to site selection perpetuates the already exacerbated underperformance of sites and maintains an exclusion of underrepresented populations.
Research results presented in ASCO’s educational program on equitable access this year revealed that oncology trials remain hugely inaccessible, reporting up to 52% of patients with commonly diagnosed cancers facing commute times of over an hour to participate in a trial. Those numbers get even worse as we look into underrepresented populations like Black patients living in rural areas.
What's the solution? Sponsors need to retire the current definition of what makes a site a “good site” and bring forth a willingness to look to lesser-known, but experienced and motivated, sites for their trials.
I spent over 15 years at large pharmaceutical companies witnessing the site selection process become more and more data driven. During that time, I led feasibility teams, owned the company SOP on country and site selection, and worked with clinical operations personnel to ensure local intelligence on standard of care and practice patterns were considered in site selection decisions. Across these roles, the common goal was finding the right sites to bring in the right patients who needed access to care, yet our system for identifying motivated, qualified researchers reduced sites to data points.
This approach wasn’t unique to my teams. Over the last two decades sponsors and CROs worldwide have increasingly relied on the use of data-driven decision-making to inform site selection, expanding the gap that lies between a patient-centric issue and a patient-centric solution. Using internal historic data, publicly available data, and feasibility responses, sponsors and CROs have become more sophisticated in utilizing this data to evaluate and select top-tier sites based on the potential to perform in startup, enrollment, and patient management.
However, even such refined data and analytics have failed to fix the problem of access. The reality is, despite robust data sets, many of those “top-tier sites” continue to underperform. Oncology trials are particularly strained with cancer patient participation rates as low as 3%. This data-driven approach excludes the sites that can combat those participation rates and solve issues of access.
The problem with both the historical and feasibility data that fuels sponsors’ site selection models is clear: historical data favors large sites with lots of industry experience. When searching for sites with the ability to perform in a particular indication, say lung cancer, the instinctual first step is to look for sites with lung cancer experience. The results list “tier one” investigators with significant experience and impressive performance in previous studies. However, the issue lies in the fact that every other sponsor team launching a lung cancer trial will get those same results.
It’s a familiar story: the trials will go to those top sites, the same set of oncology sites that the sponsors have been using for years. What follows is congestion in the small group of top sites, resulting in a failure to move the needle in broadening the trial’s reach and enrolling a more diverse population of patients. Meanwhile, thousands of community oncology sites with successful industry experience in other solid tumors, say breast, prostate and colorectal, and robust community connections that are looking to offer a lung cancer study to their patients fall into “tier three”, a category sponsors and CROs rarely select from.
Sites demoted to the third tier by these data sets are considered less experienced, less likely to bring in results, and less trustworthy. That perception isn’t based in reality but on the current system, making it impossible for their attributes and experience to be recognized.
This type of research was never easy, but with the added work for biomarker-selected patient populations, adverse event management, and numerous protocol amendments, oncology studies are becoming even harder. Nonetheless, using stringent protocol and metrics to define what a “good site” is ignores the crucial attributes that allow sites to enroll: their motivations, community connections, and patient relationships. With 74% of patients with cancer receiving treatment locally, oncology trials cannot succeed without including the community.
Progress requires sponsors to funnel their innovation into rewriting the definition of a good site. It calls for a shift away from antiquated data sets and towards a novel approach that promotes tier-three sites from the bottom of the list, recognizing the power of communities. Sponsors should look beyond the same group of large sites that stand out in a data set and start looking for the sites that would stand out to a patient. Staff profiles, languages spoken, and community engagement activities are foundational to a site’s success.
Inflated numbers and familiar site names aren’t what recruit patients, trusted relationships between the patient, their caregivers, and the local oncology team is. With emerging tech solutions, and an industry shift towards patient-centricity, defining good sites in a way that includes community researchers is within reach. I strongly believe that it’s time for patients in all communities to gain access to innovative care options.
I’ve been a part of, and seen firsthand, the innovative thinking that takes place across sponsor teams and implore sponsors to open their thinking to view the sites that serve underrepresented patients as capable, passionate researchers who deserve a role in researching oncology.
Marcy Kravet is the head of oncology strategy at Inato. She has more than 25 years of experience in clinical research. Before joining Inato in 2023, Marcy spent 15 years in large pharma overseeing clinical operations across multiple therapeutic areas.