CHICAGO (Reuters) - Virtually all of the experts who wrote favorably about GlaxoSmithKline Plc’s troubled diabetes drug Avandia had financial ties to drug makers, a finding that shows the need for reform of such relationships, U.S. researchers said on Thursday.
A team at the Mayo Clinic in Rochester, Minnesota, pored through more than 200 scientific studies and commentaries that offered positive opinions about the drug after a May 2007 study suggested Avandia significantly increased the risk of heart attacks.
They found that 94 percent of the authors who defended the drug, known generically as rosiglitazone, had ties to drug companies, and nearly half had financial ties that presented a conflict of interest.
“It was almost three to four times more likely that somebody who had a relationship with a pharmaceutical company had a favorable opinion about the medication,” Dr Victor Montori of the Mayo Clinic, whose study appears in the British Medical Journal, said in a telephone interview.
Montori looked at the financial relationships of researchers who had ties with pharmaceutical companies in general, and particularly those who wrote about rosiglitazone or the rival drug pioglitazone, made by Takeda Pharmaceutical Co Ltd under the brand name Actos.
“If you were to look at the proportion of people with favorable opinions, 94 percent of them had a relationship with a pharmaceutical company,” Montori said.
“If you were to look at the folks with an unfavorable opinion, 28 percent had a relationship with a pharmaceutical company,” he said.
FAILURE TO REPORT
The team checked both disclosures in scientific papers that mentioned Avandia and other papers by the same authors.
About a quarter of those did not report that relationship in the paper that was about Avandia.
“If the public was directly looking at financial relationships in the studies of interest, that information will not be present in a fourth of those studies,” Montori said.
The Mayo team’s findings “underscore the need for further changes in disclosure procedures in order for the scientific record to be trusted,” the researchers wrote.
Glaxo spokeswoman Mary Anne Rhyne defended the drug.
“Of the 202 publications reviewed by the authors, only 10 were original scientific research,” she said in a statement.
“Many of the articles reviewed were opinion pieces -- editorials, commentaries or letters. It is important to note that the authors’ conclusions do not impugn the validity of the scientific data,” she said.
Montori said the results were essentially the same if the researcher had a relationship with the maker of Avandia or the competing drug Actos, although they looked harder at research on Avandia.
“The reason is, it seems, there is a large overlap in the investigators who write opinions about these drugs that are funded by both companies,” he said.
Montori said it is not clear whether having financial ties to a drug company influences a researcher’s opinion, or whether drug companies seek out investigators who already have a good opinion of their drug and fund their studies.
What is clear, he said, is that drug company ties do appear to influence scientific opinion about the safety of drugs, and this issue may play a role in the debate about a drug’s safety.
Rhyne said Glaxo plans to disclose research payments made to healthcare professionals and their institutions starting in the first quarter of 2011. Those will cover payments made to U.S. researchers in 2010, then will be extended to researchers and institutions outside the United States.
Last month two U.S. senators released a report and internal documents from Food and Drug Administration safety reviewers who recommended pulling Avandia from the U.S. market.
The FDA is reviewing data on Avandia and plans to hold a public meeting in July to discuss the risks and benefits of the drug. Meanwhile, the agency says doctors and patients should continue to use Avandia as directed.
Glaxo last month said the scientific evidence did not establish that Avandia increased heart attack risks.
Sales of Avandia, which topped $3 billion in 2006, fell to $1.2 billion in 2009.
Editing by Maggie Fox and Eric Beech
Our Standards: The Thomson Reuters Trust Principles.