About a fifth of cancer clinical trials fail to attract enough participants to produce useful data, and researchers could save time and money before a trial launch by using a recruitment-success prediction tool that measures the recruitment risk of a trial.

That’s according to a new study from the Hutchinson Institute for Cancer Outcomes Research (HICOR) at the Fred Hutchinson Cancer Research Center and the University of Washington, where researchers scrutinized data from 787 publicly funded phase II/III trials undertaken in a cooperative group setting and launched between 2000 and 2011.

Carrie Bennette, lead researcher of HICOR, said, “The two pieces of the puzzle for us here were: 1) We need to be able to design trials better, because the way they are now—with so few patients actually getting enrolled—presents a huge waste of resources from the start up; and 2) We need to design trials that are more valuable to society, where data reaches the public eye, and it can’t do that if trials get shut down due to lack of participants.”

Bennette and her associates found that 18% of the trials they studied closed due to low patient-accrual rates, or they were recruiting at less than half of the targeted participation number three or more years after launch. In the U.S., only 3 to 5% of adult cancer patients sign up for these studies, so strong recruitment already is crucial.

The researchers found several risk factors that led to low patient accrual. Those included:

• A required tissue sample or biopsy: Trials that require a tissue sample or biopsy from a person in order to determine eligibility for inclusion are less likely to have good recruitment.
• The presence of randomization: When a patient knows they may or may not get the new drug therapy as part of their participation in the trial, they are less motivated to sign up. Being randomized—a natural part of a phase III clinical trial—was off-putting to potential participants, whereas entering a phase II trial in which all patients get the investigational new drug was attractive.
• Trial competition: Increased competition for patients from ongoing trials in the same therapeutic area also puts a budding trial at risk of low recruitment.

Was any of that surprising to researchers? Not really, according to Bennette. But what was exciting about their outcomes was their ability to qualify the factors, to measure them and to place them into a mathematical model that allowed prediction of success or failure for individual trials. Although numerous studies have investigated reasons for low recruitment and retention rates for clinical trials, only a few have empirically evaluated predictors of accrual success.

The predictive model the team developed is still in draft form, but viewable in the appendix of the HICOR paper. Bennette’s currently working on converting it to a Web-based tool, and hopes to be done in about six months.

Bennette’s study was funded by grants from the federal Agency for Healthcare Research and Quality and the National Science Foundation. The findings were published in the Journal of the National Cancer Institute.

Tracy Blumenfeld, founder of RapidTrials, a company that helps sponsors get clinical trials up and running, said it’s always great to see new research on the efficiency—or lack thereof—of clinical trial operations, as it brings new focus to long-standing problems. But she added that there was little in this research that’s surprising to those in the industry, and many already have developed predictive models to help them with better recruitment and retention.

“I think to fairly assess how to overcome these issues, we have to address more than the design of the protocol,” said Blumenfeld. “We have to also look at the many problems with investigators and sites. That represents half of the chronic problem with recruitment and retention.”

Will Bennette and her associates focus on that angle next? First, they’re working on making their predictive model more user-friendly, then they will focus on a tool for estimating the value to society that individual trials bring. After that, Bennette said she’s open to looking at the site side of the equation.  

Suz Redfearn is an award-winning journalist and former senior staff writer for ClinPage.com. Her articles have appeared in numerous publications, including the Atlantic.com, the Washington Post, Slate, Salon, Politico, Men's Health, MedPage Today and Physicians Practice. Suz holds a degree in print journalism from Loyola University in New Orleans and has been a medical writer since 1990, focusing on clinical research since 2007. Email suzredfearn@gmail.com.

This article was reprinted from Volume 20, Issue 01, of CWWeekly, a leading clinical research industry newsletter providing expanded analysis on breaking news, study leads, trial results and more. Subscribe »