Researcher admits to leaking drug data

By Maggie Fox, Health and Science Editor

WASHINGTON (Reuters) - A Texas researcher admitted he had leaked a sensitive and controversial report about the heart risks of diabetes drug Avandia to its maker, GlaxoSmithKline, the journal Nature reported on Wednesday.

Steven Haffner of the University of Texas Health Science Center at San Antonio admitted he tipped off the company days before the report was due to be released, and a member of the U.S. Congress quickly launched an investigation.

"Why I sent it is a mystery," Haffner is quoted by Nature as saying. "I don't really understand it. I wasn't feeling well. It was bad judgment."

In May, Dr. Steve Nissen of the Cleveland Clinic reported in the New England Journal of Medicine that his analysis of other studies suggested Avandia raised the risk of a heart attack by 43 percent.

The report set off a storm of controversy, round after round of regulatory meetings and lead to requirements by the U.S. Food and Drug Administration and European regulators that Avandia carry strong warnings about the risks.

Haffner was one of those asked by the journal to "peer-review" the paper, a process under which experts in the field try to poke holes in the research and ask questions.

These reviewers promise to keep the research confidential until it is published.

Nature said Haffner faxed a copy of the study 17 days before it was published to Alexander Cobitz, a Glaxo employee he knew from working on an earlier clinical trial of the drug.

CONGRESSIONAL SCRUTINY

Iowa Republican Sen. Charles Grassley, ranking member of the Senate Finance Committee, said he was asking Glaxo what it did when it received the leak.

"According to documents filed at the FDA, GSK has paid Dr. Haffner around $75,000 in consulting fees and speaking honoraria since 1999," Grassley wrote in a letter to the company.

"Did GSK contact the NEJM to report this violation of publishing ethics?" he asked.

"We did not contact the New England Journal," Glaxo spokesman Nancy Pekarek said in a telephone interview.

She said the company had already been answering questions from several sources, including the FDA, about Avandia and she said the company had done its own study that came up with similar results to Nissen's.

"The reason Dr. Haffner sent it to us in the first place is he told us he had concerns and about the methodology used in the article and he was seeking some statistical assistance," Pekarek said.