Carlson, Robert H.
In a study of 7,776 clinical cancer trials undertaken between 2005 and 2011, approximately 25 percent failed to complete, were prematurely terminated, or withdrawn, according to data presented at the Genitourinary Cancers Symposium (Abstract 288).
Some of the studies were stopped early because of lack of funding, others due to sponsor cancellation, and of course, trials were and should be stopped for toxicity/adverse events or definitive results at interim analysis. But, the leading reason for cancellations, at approximately 10 percent of all trials, was failure to accrue sufficient numbers of patients.
So said the study's senior author, Matthew Galsky, MD, Director of Genitourinary Medical Oncology and Associate Medical Director of the Cancer Clinical Trials Office at Icahn School of Medicine at Mount Sinai School of Medicine in New York City, speaking in a teleconference before the start of the meeting, which is cosponsored by the American Society of Clinical Oncology, the American Society for Radiation Oncology, and the Society of Urologic Oncology.
Among all the discontinuations, failure to accrue accounted for 37.9 percent. This reflects trial participation by the oft-cited figure of only three to five percent participation of patients in all adult clinical trials, a rate that Galsky noted has remained steady for decades. “Clinical trials initiated that close without enrolling the intended number of patients represent the extreme example of inefficiency in the system,” he said.
MATTHEW D. GALSKY, MD
The researchers—first author was Kristian D. Stensland, a fourth-year medical student—found the 7,776 adult cancer clinical trials at ClinicalTrials.gov. During the time period, approximately 11 percent of cancellations were initiated by the sponsor; 10 percent were due to results at interim analysis; eight percent were due to toxicity or adverse events; six percent for lack of funding; 3.5 percent because the trial was considered to be no longer needed; 2.4 percent because an agent was not available; and two percent due to the departure of the principal investigator.
Galsky said many of the industry-sponsored trial cancellations were business decisions—in some cases because of slow accrual but also because of competing trials. “For example,” he said, “we spent a year designing and opening a GU trial of a combination of standard chemotherapy plus a new drug that looked exciting at the time. About a month before that trial was slated to begin enrolling, the company reported the results of their own Phase III trial in lung cancer that showed no efficacy, and the company stopped our trial.”
As medical science moves forward, some trials become unnecessary, as the agent being tested is replaced and is no longer the standard of care.
Trials sponsored by industry were about two times more likely to fail, the researchers said, as were single-institution trials.
Galsky said issues of efficacy or toxicities are certainly good reasons to stop trials, and that those can still generate information that contributes to medical science.
In a trend in the other direction, trials were about one-third more likely to be completed if they were conducted outside the United States with or without a U.S. component. “The literature shows that sponsors are increasingly going overseas to conduct trials, because trials can get done a little more efficiently, can open faster, and involve less red tape and regulatory burden,” he said. “Couple that with the fact that patients elsewhere have fewer standard treatment options than in the U.S., and that leads to more efficient enrollment in [foreign] clinical trials.”
Those are not all good reasons to discontinue, though, he said. “We certainly want at least enough oversight to make sure clinical trials are done to maximize patient safety. But we've probably gone a little too far in the U.S. in the other direction in regulatory burdens on trial conduct.”
Broader Discussion of How to Improve Efficiencies
The moderator of the teleconference, Charles J. Ryan, MD, Professor of Clinical Medicine and Urology at the Helen Diller Family Comprehensive Cancer Center at the University of California, San Francisco and Leader of the Genitourinary Medical Oncology Program, said the study casts light on one of the major frustrations the entire cancer community shares in designing and implementing clinical trials: “I hope that this will start a broader discussion of how to improve efficiencies,” he said.
Galsky said the study was prompted because “the GU literature is inundated with clinical trials that terminated prematurely, particularly in bladder cancer.” But in fact the study data showed that trials in GU cancers are no more likely than those for other cancers to be terminated early.
As noted in the Crowdsourcing article in this issue (page 1 and next page), he and his colleagues have now embarked on a novel approach to increase accrual to cancer clinical trials, integrating new information technology approaches to try and increase the burden of participation and break down geographic barriers to participation.