Some clinical cancer trials don't attract the wanted number of participants. And there are a number of reasons why, such as competing trials, conditions with low annual incidence, and whether they focus on older treatments rather than newer ones, according to researchers from the University of Washington and the Fred Hutchinson Cancer Research Center.
As UW's David Veenstra and his colleagues report in the Journal of the National Cancer Institute, they examined 787 phase II/III adult studies from the National Cancer Institute's Cooperative Group Program (now the National Clinical Trials Network) that attracted less than half the targeted number of participants.
Trials with low accrual, they report, tended to be less likely to study a new investigational agent or a targeted therapy and were more likely to examine multimodality, surgery, or radiation therapy. At the same time, trials with low accrual levels were also more likely to be started where there's higher competition for eligible patients and were more likely to be studying a condition with a lower annual incidence and need to enroll a larger portion of the eligible patient population.
Based on a dozen of these factors, the researchers developed a prediction model for low-accrual risk, which they then applied to a set of 46 trials. The predicted risk, the researchers report, was in line with observed accrual in those trials.
"Identifying factors that predict low accrual in trials before they are launched could help optimize trial design to improve accrual and inform trial design and portfolio prioritization efforts," Veenstra and his colleagues write in their paper
In a related editorial, Derek Raghavan from the Levine Cancer Institute says the candidate list makes intuitive sense and the work passes the "sniff test." Still, he writes that the use of such a prediction model may be a move in the wrong direction. Rather than avoiding trials that may not attract enough participants, Raghavan says efforts should focus more on reaching out to more patients.
"[T]he real issue is still the lack of patients involved in cancer trials," he says. He later notes that patients in trials generally have better outcomes than those who do not participate. "[A]s oncologists who wish the best for our patients, we should strive to improve trial enrollment, giving the associated potential for improved results," Raghavan adds.