Tamiflu Efficacy, Safety in Doubt, Says BMJ

Troy Brown

November 14, 2012

The British Medical Journal (BMJ) has alleged that pharmaceutical giant Roche is deliberately hiding clinical trial data about the efficacy of oseltamivir (Tamiflu) in patients with influenza. The journal says global stockpiling and routine use of the drug are not supported by solid evidence and alleges that Roche concealed neurological and psychiatric adverse events associated with the neuraminidase inhibitor drug.

In an open letter from Fiona Godlee, MD, editor-in-chief of BMJ, to Professor John Bell, FRS, HonFREng, PMedSci, Regius Professor of Medicine at Oxford University in the United Kingdom and a Roche board member, published online October 29, Dr. Godlee reminds Bell of concerns that were initially voiced in 2009 about the reliability of Tamiflu research.

At that time, BMJ published an updated Cochrane review of neuraminidase inhibitors in healthy adults. That study "took the view that, since eight of the 10 [randomized controlled trials] on which effectiveness claims were based, were never published, and because the only two that had been published were funded by Roche and authored by Roche employees and Roche-paid external experts, the evidence could not be relied upon," Dr. Godlee writes.

Roche promised to make complete clinical trial data available, but Cochrane researchers and BMJ claim the company has not done so. In her letter to Roche, Dr. Godlee explained that each clinical trial report typically consists of 5 modules, but only module 1 has been provided for each trial.

"What Has Roche Got to Hide?"

In January 2012, the Cochrane Library published an updated review that included unpublished data and revealed inconsistencies and incomplete data. In her open letter to Roche, Dr. Godlee explained, "The Cochrane reviewers now know that there are at least 123 trials of Tamiflu and that the majority (60%) of patient data from Roche Phase 3 completed treatment trials remain unpublished. There are concerns on a number of fronts: the likely overstating of effectiveness and the apparent under-reporting of potentially serious adverse effects. Meanwhile, Tamiflu has just been added to the list of essential drugs alongside aspirin and beta blockers."

In an editorial published online October 29 in BMJ, Dr. Godlee asks, "What has Roche got to hide?"

However, in a media statement responding to Dr. Godlee's open letter in October 2012, Roche denies withholding clinical data.

When questioned by Medscape Medical News about whether any physician or researcher has been given all 5 modules from the 10 clinical trials requested by the Cochrane researchers, and where those data are located currently, Roche responded with a statement much like its previously published statement. Roche said it had provided the Cochrane researchers "with access to 3200 pages of very detailed information, enabling their questions to be answered."

The statement added that all completed Roche Tamiflu research results are available in peer-reviewed journals or summarized on their Web site. "More detailed clinical trial reports...are available for use by investigators on a password-protected site, enabling researchers to verify the findings of these studies," according to the company's statement.

"Roche has made full clinical study data available to health authorities around the world, according to their various requirements, for their review as part of the licensing process. It is the role of global health authorities to review detailed information on medicines when assessing benefit/risk," said the Roche response to Medscape Medical News.

"Within the scope of a signed agreement Professors Lipsitch and Hernan [from Harvard University] received the 10 studies cited in Kaiser et al," the statement to Medscape Medical News concludes.

Open Data Campaign

Hoping to prevent similar issues in the future, BMJ is instituting a new policy requiring pharmaceutical companies that sponsor clinical trials to commit to making relevant anonymized patient-level data available to physicians, scientists, and others who reasonably request it as a precondition for publication.

"We will require authors of trials of drugs and devices to commit to making their data available on reasonable request," said Dr. Godlee in an email interview with Medscape Medical News. "We cannot police this ourselves, but we will encourage researchers whose requests for access are declined to send rapid responses to the BMJ saying what they asked for and why it was declined. We will then ask the authors to account for their decision not to release the data," Dr. Godlee explained.

"This is part of a journey and by no means the last step on the road. I expect that our next step, when data deposition systems are better developed and accepted, will be to ask authors to deposit their data on submission of their article," Dr. Godlee said. "[W]e will want to extend this to trials in other areas and to other types of study design. We also hope that other journals will move in the same direction," Dr. Godlee added.

BMJ has not given up on its efforts to obtain complete clinical trial data from research on oseltamivir. In its first open data campaign, BMJ is making public all correspondence between the journal, Roche, the World Health Organization, and the Centers for Disease Control and Prevention.

"I hope very much that the BMJ's campaign will succeed in making the full data on oseltamivir available to the Cochrane reviewers, so they can give governments around the world the true picture of this drug's effectiveness and safety," Dr. Godlee said.

"Regulators, research ethics committees, and bodies like [the National Institute for Health and Clinical Excellence] who decide which drugs will be bought by governments or reimbursed by health insurers also have a key role in pushing for access and independent scrutiny of the data," Dr. Godlee explained.

Dr. Godlee has disclosed no relevant financial relationships.

BMJ. Published online October 29, 2012. Letter full text, Editorial full text, Policy editorial full text

"Reactive statement - BMJ open letter." Roche. Media statement full text

Comments

3090D553-9492-4563-8681-AD288FA52ACE

processing....