Roche replies to the authors of the Cochrane Review on oseltamivir

We would like to respond to some of the issues raised with regard to the Cochrane Review on oseltamivir published in this issue of the BMJ.1

We were not provided with this article in advance to allow detailed comment, but we understand from correspondence with Dr T Jefferson of the review team that it may omit from its assessment data that show the efficacy and safety of this drug—namely, a meta-analysis of 10 studies published in 2003.2 This review indicated that oseltamivir treatment reduces lower respiratory tract infections and hospitalisations in both healthy and “at risk” adults with flu illness. The publication is one of numerous data sources supporting the efficacy and safety of the drug, including a pooled analysis by Singh et al,3 studies from Hong Kong4 and Canada5 showing a reduction in mortality, evidence of efficacy from claims database studies,6 7 8 a brief review of evidence of efficacy in cases of pandemic 2009 H1N1 flu infection,9 and the European summary of product characteristics.10

Over the past weeks we and investigators involved in certain publications evaluating the utility of Tamiflu in reducing flu related complications have been approached separately or jointly by the BMJ and Channel 4 News, requesting answers to wide ranging questions on access to study data, interpretation of data, procedures for writing publications, authorship, and other matters. Responses to all questions have been provided in full to the BMJ and Channel 4, and should leave no doubt about the high integrity of the data, publications, and interactions between Roche and independent investigators.

A key question that was raised concerned several unpublished final study reports for certain randomised clinical studies. We note that at the time the studies were conducted (1999-2003) it was not standard practice for study protocols or reports to be automatically made public as it is now. However, for transparency and to facilitate disclosure, Roche has posted study summaries, including data tables, for all the component studies online at roche-trials.com; full reports for these studies will also shortly be available by password protected access for all bona fide scientific investigation.

To reinforce the integrity of the reports and underlying data, we confirm that the reports were written to the standards of the regulatory authorities of Europe (European Medicines Agency; EMEA), the United States (Food and Drug Administration; FDA), and Japan (Ministry of Health, Labour, and Welfare; MHLW) and that they have been submitted to and accepted by these bodies for licensing purposes. These bodies have the right to inspect Roche processes and documentation, and indeed the FDA and the MHLW did so during 1999-2000.

Roche has robust procedures for both producing and sharing study reports. As well as providing reports to such regulatory bodies, Roche has, through appropriate channels, willingly shared data and reports with numerous other eligible individuals and groups, including the Cochrane review group, a research unit funded by the Medical Research Council, the authors of the cited publications and numerous others, and other research groups. We were concerned that the Cochrane review team, led by Dr T Jefferson, chose to follow-up their inquiries through a television company rather than by approaching the manufacturer directly and obtaining data under a standard agreement—a move that questioned whether the motives for inquiries were truly for clarity and scientific validation. It is unclear to us why Dr Jefferson would adopt this approach, particularly given that he was a paid ad hoc consultant to Roche working on flu and oseltamivir between 1997 and 1999. During that period he worked closely with Roche experts, many of whom are still in the company, and he would therefore not have had difficulty in contacting them directly to discuss his requirements.

Our collaborations with independent clinicians and research groups, to the requirements and standards of the regulatory authorities, are vital in our ultimate aims of facilitating access to effective treatment and improving public health. In the case of oseltamivir, as well as the documented benefits to those in high risk groups, these aims are particularly relevant in the context of the current H1N1 (2009) flu pandemic and the ongoing threat from highly pathogenic H5N1 (avian) flu infection. The role of oseltamivir in mitigating the serious threats from such flu infections has been recognised in the recommendations from the World Health Organization. We continue to pursue these aims through active and thoroughly conducted research programmes, constantly updated ethical practice, and extensive collaborations with international experts in their fields. Why Dr Jefferson and the BMJ chose to pursue their scientific enquiries through commercial television remains to be clarified.