January 18, 2012

expert reaction to Cochrane review of Tamiflu and Relenza efficacy

The Cochrane review looked at the efficacy of the two leading flu vaccinations.

Dr Jake Dunning, Clinical Research Fellow, Imperial College London, said:

“Are these new data from trials of swine flu or seasonal flu? Does that matter?

“The latest Cochrane review fails to appreciate the differences between the virology and clinical aspects of pandemic H1N1 2009 influenza and the ‘traditional’ seasonal strains and therefore, the conclusions drawn may not be relevant to the current influenza landscape. Apart from the differences in virus structure, pandemic H1N1 2009 influenza virus caused a very different pattern of illness in a very different age group, even when compared to other seasonal H1N1 strains. The authors argue that because regulatory bodies licensed the use of antivirals for pandemic H1N1 2009 based on evidence from studies of pre-pandemic seasonal influenza, then it was appropriate to assess similar data in the latest Cochrane review. Regulators had little choice, having to make rapid decisions using the best available evidence at the onset of a pandemic. The alternative would have been to await data from randomised controlled trials conducted during the outbreak, studies which are almost impossible to conduct in this context and any data generated would probably have arrived too late to be of use. Such an approach certainly wouldn’t work if there was a bird flu pandemic.”

What is new about these data? How do they compare with previous Cochrane Reviews and what we already know in this area?

“We do need greater transparency and sharing of trial data in all aspects of clinical trials. Those charged with withholding of data should be given a chance to respond to the accusations. Where possible, study design and end-points in future trials also need to be appropriate to clinical need and regulatory requirements. For example, antivirals such as Tamiflu only benefit patients who are actually infected with influenza virus; many other viruses cause flu-like illness, so if you have one of those and take a flu antiviral, chances are you won’t improve, but you may well become more aware of any side-effects that occur on top of your non-improving symptoms. While it seems sensible to test antiviral effectiveness only in patients with confirmed flu, most patients in the community cannot be tested for influenza virus and in this context, doctors would offer antivirals as ‘best-guess’ therapy. Attempting to bring study design and the ‘real world’ closer together is not easy, however, and compromise may be required when interpreting and applying the findings of studies.

“The earlier Cochrane review of antivirals, published as part of a wider campaign and in the midst of a pandemic, was potentially damaging to public health efforts and the conclusions drawn were not particularly relevant to the clinical problems we were seeing in patients infected with the new strain of influenza. This lack of relevance to pandemic H1N1 2009 was suggested by the authors at the time, but the message seemed to get lost. Cochrane reviews are highly influential, so I fear that the headlines generated by the latest review will also lead clinicians and patients to believe that a current lack of acceptable data from seasonal flu studies means that antivirals have no real role in the treatment of flu today, which is nonsense.”

Are there any implications for the current flu season and should people ask for tamiflu/relenza if they become unwell?

“In the UK, a huge number of hospital and intensive care beds were occupied by patients with pandemic H1N1 2009 last winter and most of these patients received antivirals as part of a package of care. The vast majority did not receive early antivirals in the community, however, despite many being in recognised at-risk groups. We have had a quiet flu season so far this year, but history suggests that a pandemic strain will continue to cause peaks of severe illness, especially in younger adults and children, for five to ten years following its emergence. Unfortunately, the latest Cochrane review fails to address the problems we are likely to face in the current, post-pandemic phase and may act as a distraction.

“What we really need at this time is a degree of pragmatism, accepting the limitations of available data, but not dismissing the reported beneficial effects of antivirals in the treatment of pandemic H1N1 2009. The Department of Health followed this approach and concluded, in its recent and comprehensive review, that antivirals do benefit those with severe flu and those who are known to be at risk of developing complications of flu. This is an important message for patients and clinicians.”

‘Neuraminidase inhibitors for preventing and treating influenza in healthy adults and children’ by Tom Jefferson et al., published in The Cochrane Library on Wednesday 18th January 2012.