WASHINGTON (Reuters) - Breast cancer studies funded by drug companies are more likely to yield positive findings than those without pharmaceutical industry backing, according to research published on Monday.
A team led by Dr. Jeffrey Peppercorn of the University of North Carolina School of Medicine also found industry-backed studies tended to be designed differently than the others, but he said drug companies may simply be backing the least-risky alternatives.
The study was the latest to find industry-funded research producing more positive results. There are similar findings on heart, stroke and bone marrow cancer research.
Consumer advocates said the findings raised fresh questions about the integrity of company-funded research as the industry spends billions of dollars annually to bring new, potentially lucrative drugs to the market. The industry denied that studies were rigged to produce positive findings.
The researchers looked at 140 studies published in 2003, 1998 and 1993 in 10 medical journals on breast cancer therapies, nearly half of which were deemed to have had drug company involvement in the form of funding, provision of drugs or participation of a company scientist.
They focused their findings on 2003 because it was harder to determine industry involvement in the earlier studies.
In 2003, 58 percent of the studies had industry backing. Of these, 84 percent showed positive results, compared to 54 percent for other studies.
Peppercorn said there are several potential explanations.
“Those certainly include the potential that either the results are biased and negative studies are not being published, or that the results are being interpreted favorably,” Peppercorn said in a telephone interview.
Another possibility, he added, was that companies were taking “safer bets” by making sure they proceeded to costly clinical trials with drugs that had a good chance of doing well.
NOT NECESSARILY WRONG
“I wouldn’t at this point take this (study’s) result to say that the pharmaceutical industry’s doing anything wrong,” Peppercorn added.
In the 2003 studies, 66 percent of industry-backed research involved “single-arm” studies - with all patients getting the same treatment, and no control group getting another treatment to compare effectiveness - versus 33 percent of the other studies.
The company-sponsored trials also tended to target patients with advanced disease.
Peppercorn said some types of questions might not be addressed in industry-backed trials, such as evaluating the toxicity of drugs to particular groups of patients.
Alan Goldhammer, deputy vice president for regulatory affairs for the Pharmaceutical Research and Manufacturers of America trade group, said companies spend a lot of time making sure these trials yield high-quality data.
“Designing and conducting clinical trials is the business of the pharmaceutical industry,” Goldhammer said.
“The last thing anybody wants is to have questionable data, because that data will then be thrown out and wouldn’t be used by the FDA (U.S. Food and Drug Administration) in making a marketing determination.”
Dr. Sidney Wolfe, director of consumer group Public Citizen’s Health Research Group, said while the drug industry should fund biomedical research, it should have no say over its design or interpretation of results.
“Integrity gets narrowly defined as falsifying records and data. And I’m sure that kind of thing happens. But that’s not the biggest problem,” Wolfe said. The biggest problem is that studies are being designed by people with a financial interest in the outcome, Wolfe said.
Our Standards: The Thomson Reuters Trust Principles.