原文

“This new report does a wonderful job of assimilating current knowledge about using antivirals against influenza, and highlights real difficulties in developing and using such antiviral drugs. Influenza is a virus that can vary widely in the severity of disease it causes, depending on the strain of virus and the person infected. Researchers and pharmaceutical companies alike who are aiming towards developing new antivirals for influenza should bear these difficulties in mind for the future.

“The report concludes that, although the neuraminidase inhibitor drugs like Tamiflu and Relenza are effective, using them to treat people with mild influenza did not bring a huge benefit and sometimes caused vomiting as a side effect. However the report importantly emphasises that the drugs are much more important for treating severe influenza, and that early during an outbreak, before the severity is known, using them is prudent. In addition the report stresses that clinical judgement should guide treatments on an individual patient basis.

“Perhaps one of the key issues the public and government want this report to address is whether an antiviral stockpile was or is justified. The report makes it very clear that influenza severity is difficult to predict ahead of time. Although the 2009 pandemic was thankfully mild, future influenza outbreaks may be more severe and in that case one would expect more people to benefit from using these drugs both in terms of lives saved and getting the country back to work. Whether in the end this is ‘cost effective’ can be modelled to some extent, but it is difficult to put a value on the panic that can ensue in the early weeks of a new outbreak of a virus that has the potential to cause severe disease.

“One of the primary outputs of the report is the call by the report’s authors for more trials using these drugs in hospital settings for severe cases.  Whilst such trials are an important part of evidence-based medicine,  the report itself does emphasise that retrospective analysis can also be used to inform medical practice and acknowledges the persuasive data obtained from hospitalised patients in 2009 showing that using NAIs saved lives.  On the other hand the report laments the fact that we don’t have evidence from random controlled trials conducted on severe flu cases and strongly recommends conducting such trials during future outbreaks. It will be important going forward to be clear about how such trials will be conducted, for example, how does one allocate a placebo group in this situation? If a pregnant woman comes to hospital with flu, can we justify using her to test the effect of not giving Tamiflu, bearing in mind the retrospective analysis from 2009? These difficult situations are not that dissimilar from those faced recently in West Africa thinking about how to properly conduct trials with agents against Ebola, another deadly viral disease. Whilst new styles of clinical trial are being designed to address these concerns, it is important that patients at high risk are not denied this licensed drug.

原文

“This report is excellent and provides a dispassionate view of these drugs and their benefits which have limitations. It makes clear that in some circumstances, such as when a potentially significant fraction of key workers may be ill and not able to work, even temporarily, that a small benefit may firstly reduce the time off work but most importantly, reduce the height of a peak in the numbers absent from work on any particular day. This latter point is not made clearly in the report though it is alluded to. It can have a major economic benefit in possibly preventing a cascade of problems (beyond a “tipping point’) when a major epidemic or pandemic occurs.

“The downside is that strains of flu may become rapidly resistant to it, and this seems to have happened already.

“The benefits of the drugs in the most severely ill or vulnerable are uncertain and the proposal to have good randomised trials in such groups is to be welcomed.

“The wise course may be to reserve such drugs for key workers and, if the randomised trials proposed by the report show benefits in the high risk groups, given to those at high risk and not those at low risk.

“Media-induced panics that led to widespread use of the drugs in the previous pandemic are harmful; perhaps we will have learnt a lesson. This report is absolutely rigorous in its evaluation of our current knowledge.”