Company-funded drug studies find harm less often

NEW YORK (Reuters Health) - When drugmakers fund studies, scientists are much less likely to find potentially harmful drug effects in the lab, new research hints.

But they do report benefits, and more so than academic investigators without industry backing.

That's the conclusion of a new government-funded report that compares studies of drugs used to stimulate red blood cell production, such as erythropoietin (EPO).

When these drugs were first licensed in the early 1990s, some researchers worried they might speed up cancer growth. That concern later turned out to be justified, and manufacturers now warn cancer patients about this side effect.

The authors of the new report, published in the Archives of Internal Medicine, searched for published studies that tested erythropoietin's effect on cancer cells.

Out of 10 studies done or backed by industry, not one concluded that the drugs might cause harm. Yet more than half of the 64 studies not paid for by pharmaceutical companies concluded exactly that.

"We found a very clear difference in the way the results were reported," said Dr. Charles L. Bennett, of the University of South Carolina in Columbia, who led the study.

The medical community has been aware for years of the impact financial conflicts may have on clinical trials involving patients. One recent study, for instance, found research was much more likely to favor a given drug when sponsored by the manufacturer (see Reuters Health story of Aug 2, 2010).

"Industry has financial incentives to tweak the science to make sure their drugs shine," Dr. Daniel Carlat, a psychiatrist in private practice outside of Boston, told Reuters Health when interviewed about that study.

But there has been much less focus on lab-based studies in which patients aren't involved.

"The basic science for a long time has been given a free pass," said Bennett. "The bigger picture is that basic science is just like clinical science."

His co-author, Dr. Sara E. Barnato of Northwestern University in Chicago, Illinois, said transparency and collaboration were essential to keep science free of bias.

"The medical community can do great things when we work together," she told Reuters Health in an e-mail. "Collaboration is key to prevent exaggerated data or 'not the entire picture' getting analyzed and translated to patient care."

Amgen, a California-based company that manufactures the erythropoietin drug Epogen, did not respond to requests for comments.

SOURCE: http://link.reuters.com/duq24p Archives of Internal Medicine, online September 13, 2010.