Researchers Questions Tamiflu's Effectiveness

Dec. 8, 2009— -- There's not enough evidence to conclude that the antiviral drug oseltamivir (Tamiflu) has any benefit for the complications of influenza in otherwise healthy patients.

That's the straightforward conclusion of an updated Cochrane review of the drug, appearing online in the British Medical Journal -- and many experts say it's not surprising.

"The currently available antiviral medications for treating influenza in otherwise healthy adults have never been shown to have much impact," said Dr. Richard Besser, ABC News senior health and medical editor.

But the path that led to that conclusion is anything but clear-cut, according to the journal's editors.

In a series of articles, including the review, the journal says that what began as a "straightforward exercise to confirm the evidence base for current policy and practice became instead a complex investigation."

The journal's editors raised a series of issues aside from the narrow clinical question, including who had access to data from Roche-funded trials, whether ghost writers were used to prepare journal manuscripts, and why some authors were not listed while -- in at least one case -- an author was cited who later said he can't recall performing the study.

A key concern was a meta-analysis of 10 studies that formed part of an earlier review -- a report whose conclusions could not be validated, in part because the drug's manufacturer, Roche, was slow in releasing the raw data, according to Fiona Godlee, the journal's editor.

"This is a long way from being the only time in which Roche and other drug companies have declined to share necessary data with legitimate authorities," Godlee said in an email.

The Cochrane group reviewed the evidence on antiflu drugs and concluded in early 2006 that -- among other things -- oseltamivir was effective in reducing the complications of flu among healthy adults.

At 150 milligrams a day, oseltamivir was effective in preventing lower respiratory tract complications in influenza cases, with an odds ratio of 0.32, the reviewers, led by Tom Jefferson, concluded at the time.

But earlier this year Jefferson and his colleagues at the Italy-based Cochrane Vaccines Field, a collaborative evaluation group, were asked by the British National Health Service to update the analysis.

At roughly the same time, they reported in BMJ, they got a letter from a Japanese pediatrician who criticized the earlier review for relying on unpublished, manufacturer-funded data contained in the 10-study meta-analysis.

The Cochrane reviewers reported that they asked Roche for the raw data in order to see if the earlier conclusions could stand, but did not get the information in time.

That meant, Jefferson and colleagues wrote, that they had to throw the whole study out, leaving only three studies — not enough to cast light on the value of the drug in preventing complications.

"Paucity of good data has undermined previous findings for oseltamivir's prevention of complications from influenza," they said in the review.

In a prepared statement, a Roche spokesman said the company "stands behind" its research.

"The Cochrane Group was wrong to exclude the data," said David Reddy, head of the company's global pandemic taskforce.

Still, many doctors say the Cochrane reviewers' findings are no surprise.

"It's generally known that the evidence for prevention of flu complications is weak," said Dr. Marvin Bittner of Creighton University in Omaha.

Dr. Amy Kaji of the University of California Los Angeles said she and her colleagues don't recommend the drug for otherwise healthy adults.

"I think we have all known this data that Tamiflu reduces symptomatology, at best, by one day," she said in an email. "The news is therefore not very new."

On the other hand, she said, in severely ill patients admitted for pneumonia and an influenza-like-illness, "there is some evidence demonstrating benefit."

Dr. Kristi Koenig of the University of California Irvine said she's seeing many otherwise healthy adults with flu-like symptoms who are worried about the H1N1 pandemic flu.

"They should understand that Tamiflu may not be indicated for them," Koenig said. "The vast majority of young healthy adults will do well with rest, fluids and ibuprofen/Tylenol, without tamiflu."

But it's a question of perspective, said Dr. William Schaffner of Vanderbilt University. "The drug is not perfect," he said in a telephone interview, but it offers at least some benefit for some people.

He said the Cochrane reviewers, while ethical and well-meaning, can sometimes be "extraordinarily prissy" about what kinds of data they'll accept.

"They're a lot more critical of some of the data than I would be," he said.

The U.S. Centers for Disease Control and Prevention doesn't comment on papers before they're published, a CDC spokesman said. But the spokesman said the agency's position has not changed from last month when Dr. Tim Uyeki of the agency's influenza division, wrote that observations suggested a benefit for the drug in reducing complications at least in some patients.

"Evidence from observational studies supports the benefit of neuraminidase inhibitors (oseltamivir or zanamivir) in reducing complications, including deaths, among hospitalized patients with 2009 pandemic influenza A (H1N1)," Uyeki wrote in the New England Journal of Medicine.

Many experts cautioned that the Cochrane review should not be taken to apply to all people with flu.

"It is important not to extend these findings to high-risk individuals or people with severe illness for whom antiviral drugs are recommended," said Besser.

"In these people, even a small benefit may be worth it," he said.