Roche offers researchers access to all Tamiflu trialsBMJ 2013; 346 doi: http://dx.doi.org/10.1136/bmj.f2157 (Published 04 April 2013) Cite this as: BMJ 2013;346:f2157
More than three years after the Cochrane Collaboration first asked Roche for the full clinical study reports for its influenza drug oseltamivir (Tamiflu), the Swiss company has offered the collaboration access to “all 74 Roche sponsored trials.”
Don MacLean, life cycle leader for Tamiflu at Roche, emailed the Cochrane researchers on 2 April to propose providing data in a staggered approach over the next few months.
“In line with European Union law, each CSR [clinical study report] will be edited by Roche to ensure patient confidentiality and to protect legitimate commercial interests,” he wrote.
A full phase III clinical study report typically consists of 2000 to 3000 pages, and redaction would be a “large undertaking,” he added.
The Cochrane group has cautiously welcomed the move, pointing out that Roche has previously promised access to data. MacLean’s email follows GlaxoSmithKline’s decision to release 30 clinical study reports on its influenza drug zanamivir (Relenza) to the same Cochrane group. The group is concerned, however, that data redaction and other problems may make analysis and interpretation impossible.
Sile Lane, director of campaigns at Sense about Science, said that by acknowledging that there were 74 trials and agreeing to publish the results, Roche had recognised that arguments in favour of secrecy no longer held.
“This should be a good moment for them to sign up to the commitment set out at AllTrials for all trials relating to treatments in current use to be registered, and the results reported. It shouldn’t have taken the researchers years of persistence and publicity to get [access to] these Tamiflu results. Roche has an opportunity to tell the public and research community that it won’t happen like this again,” she said.
The announcement comes after years of wrangling for access to what the Cochrane group believes to be the full dataset. Instead of offering access to 74 trials, in December 2009 Roche gave the researchers access to one part of 10 Tamiflu trials (each trial report comes in four to five parts).
Carl Heneghan, one of the Cochrane reviewers, said that when the collaboration asked for more data in 2011, it was told by MacLean that Roche’s view was that “you [the Cochrane group] have all the detail you need to undertake a review and so we have decided not to supply any more detailed information.”
Heneghan commented: “The very fact that the 74 studies are now being released undermines these original statements: that we had all the detail we needed and that it wasn’t necessary and it has become increasingly clear that this is not the case. It then follows that regulators were in the same situation as we were: lacking data to come to firm conclusions.”
Then in March 2013, Roche said that the company had appointed various third parties to review the data on Tamiflu, identify any gaps, and decide on an analysis plan. It had set up a group for this purpose and it invited the Cochrane Collaboration to participate.
The so called Multiparty Group for Advice on Science (MUGAS) will be organised by the European Scientific Working Group on Influenza, funded by an unrestricted grant from Roche. This group is due to meet on 18 June.
However, the Cochrane Collaboration has asked for further clarification before deciding whether it will accept the invitation. The researchers are concerned that three of the four key MUGAS partners are scientific advisers or consultants to Roche.1
A previous BMJ investigation has also highlighted that the European Scientific Working Group on Influenza has a remit to lobby politicians; to “develop a policy for antiviral stockpiling”; and to impress on government representatives that “use of antivirals is beneficial and safe.”2
Tom Jefferson, one of the Cochrane reviewers, said: “The roles of public health bodies, researchers, governments, experts, and the media in promoting oseltamivir need to come under public scrutiny given the highly incomplete and biased evidence base used to make decisions. The last three years have shown that industry key opinion leaders have contributed in preventing access to the full trial dataset and obscuring accountability for decisions made.”
However, Albert Osterhaus, one of the key members of MUGAS and a Roche scientific adviser,3 told the group in an email that their concerns were misplaced. He said that some Cochrane groups openly declare their collaboration with industry on their websites “knowing that transparent cooperation with industry is a powerful tool to move things forward.
“The MUGAS partners suggest that we stop wasting precious energy to this pointless discussion about scientific integrity and instead engage in a constructive collaboration to address urgent health issues,” he added.
Cite this as: BMJ 2013;346:f2157