By Hannah Hoag
Under the Influence
Do drug industry–sponsored trials pose a conflict of interest?
By Hannah Hoag
It’s safe to say that without the support of the pharmaceutical industry, many new therapies would never be tested. But now some industry-sponsored breast cancer trials are raising eyebrows. New research shows that they’re more likely to report positive results than trials not supported by industry, raising concerns about conflicts of interest.
In the April 1 Cancer, Jeffrey Peppercorn, an oncologist at the University of North Carolina at Chapel Hill School of Medicine, and his colleagues evaluated pharmaceutical company involvement in 140 breast cancer clinical trials published in 1993, 1998 and 2003. They found that 67 trials, or 48 percent, reported a range of drug industry involvement in the form of co-authorship, supply of the drug or financial support.
“I noticed that more and more trials seemed to be a collaboration between pharmaceutical companies and academic centers,” says Peppercorn. “I wanted to see if it was true and what impact it might have.”
The researchers focused on studies published in the most recent year, 2003, and found that results favoring the experimental therapy were more common among trials with industry ties. Fifty-seven percent of evaluated studies from that year were industry-sponsored, and of those, 84 percent reported positive results, while 54 percent of non-industry sponsored studies reported positive results. The researchers also noted that a larger percentage of studies with company connections lacked a control group—a group of patients who, for comparison, don’t receive the investigational therapy. Studies without a control group are used in the development of new drugs but may fail to answer important questions about optimal duration of treatment, and the timing and sequence of treatments.
Although drug industry participation in clinical trials raises questions, says Peppercorn, “it doesn’t prove there is a bias.” He says his study does not establish whether companies were deliberately ignoring negative findings, or if the industry-sponsored studies were designed to provide positive results. “It could be that companies are testing better drugs,” he says. He would like to do a follow-up study that looks at degrees of pharmaceutical industry involvement and trial results.
Academic centers’ budgets fall short of the investment they need to run large, modern clinical trials, says Alan Coates, a medical oncologist at the University of Sydney in Australia, and co-chair of the scientific committee of the International Breast Cancer Study Group. “The drug industry has to be involved. No academic funding can possibly fund a trial,” he says.
Moreover, pharmaceutical company spending for the research and development of new therapies has exceeded that of the National Institutes of Health since 1992. A 2005 Harvard School of Public Health report cited statistics indicating that the pharmaceutical industry funds 70 percent of clinical trials in the U.S.
In the March 8 Nature, Coates and his colleagues advise academic researchers to aim for partnerships that limit industry’s access to the clinical trial database until trial outcomes are reached. “We are trying to prevent the situation where the drug company controls everything, holds the data, analyzes its own way and tells us the results later,” says Coates. “There appear to be cases, unfortunately, where the data were commercially unwelcome and were suppressed.”
Dwight Randle, a biochemist and the senior scientific adviser for the nonprofit organization Susan G. Komen for the Cure, says the study is interesting and that industry involvement in clinical trials should be further studied. But “the issue is not black and white; it has many shades of gray,” he says.