H5N1: Bungling dual-use governance

Recent months have seen an increasingly confusing debate about new research on the adaptability and transmissibility of avian influenza A/H5N1, which was undertaken by groups in the Netherlands and the United States. Both studies were funded by the US National Institutes of Health (NIH), and research results were sent for publication to Science and Nature, respectively. But the studies were dual-use, meaning they included “biological research with legitimate scientific purpose, the results of which may be misused to pose a biologic threat to public health and/or national security,” according to NIH. As a result, before publication, both papers were sent to the National Science Advisory Board for Biosecurity (NSABB) for review in October 2011. NSABB engaged in extensive consultations with the principal investigators of both research groups: Ron Fouchier at Erasmus University in Rotterdam, the Netherlands, and Yoshihiro Kawaoka at the University of Wisconsin, Madison. After careful consideration of the benefits and risks involved in publishing the research, the board concluded that the experiments fell under the seven categories of dual-use research “of concern” — mainly experiments that could lead to bioterrorism and must therefore be scrutinized by the “scientific and medical community” — as identified by the Fink Committee of the National Academies in 2004. The board found that the risks of publication outweighed the benefits. For the first time ever, the NSABB recommended the publication of the papers in redacted form, without a detailed methods section. In principle, both the US government and the two journals agreed to follow the board’s advice, and, in theory, this could have been the end of the story. The fact that it is not the end of the story is disturbing for at least three reasons.

1. Weird science. Clearly, in the two unpublished papers submitted to Science and Nature, the arguments and the methods used to arrive at the conclusions — that H5N1 is highly adaptable, transmissible, and deadly — must have been strong, coherent, and unambiguous; otherwise, the papers would not have been accepted for publication by these top journals. However, after the board recommendation to publish only redacted versions of the papers by the teams led by Fouchier and Kawaoka, the scientific narrative began to change and can currently at best be described as a moving target. A January 25 article in the US journal PNAS (Proceedings of the National Academy of Sciences) claims that the case fatality rate for H5N1 has been misrepresented by the World Health Organization (WHO) and is “likely orders of magnitude too high” and “fear needs to be put to rest.” In plain speak, according to the PNAS article, H5N1 is apparently a much less dangerous virus than everybody up to now thought and therefore is unsuitable as a bioterrorism agent. However, as John Treanor, an expert on influenza vaccines at the University of Rochester, puts it: If that is actually the case and “H5 infections are no more serious than the regular flu,” it raises a question — “Why are you spending all this money studying it?” So far, those trying to play down risks associated with H5N1 have not provided an answer to this question.

Meanwhile, an article in the American Society for Microbiology’s mBio journal asserts that ferrets are not good models for transmissibility among humans. Thus, even though transmissibility among ferrets has been demonstrated in the experiments conducted by Fouchier and Kawaoka, this does not mean that, if humans were exposed to the genetically modified strains, a pandemic would result. Yet this argument is also puzzling: Fouchier and Kawaoka are respected scientists; obviously they thought that H5N1 studies in ferrets would provide some reliable information that could be extrapolated to a human population. Otherwise, there was no point in using the ferret model to study the virus in the first place (unless one is deeply concerned with the plight of ferrets). Even if one accepts that the H5N1 case fatality rate is overestimated, that ferrets are poor models of study, and that scientific uncertainty in this case exists, it would seem that the very same uncertainty also means that a pandemic cannot be excluded. As Lisa Murillo from Los Alamos National Laboratory argues, “The consequences of underestimating these viruses’ pathogenicity or propensity for transmission among humans are simply too great.”

2. Lack of US governance. Much, if not all, of this debate around the backpedaling of some scientists in the flu virology community could have been avoided if there had been a dual-use governance or oversight system in place that would have provided clear guidance for measures to follow-up the NSABB recommendations — or even a system in place to review experiments before they are funded, as my colleague Laura Kahn has argued. Unfortunately, such a system is not in place in the United States — or elsewhere, for that matter. Which is why the probing thoughts and questions posed by US Representative Jim Sensenbrenner in a March 1 letter to White House science adviser John Holdren are a very welcome addition to the dual-use governance debate. Instead of arguing about whether the science is good or not, Sensenbrenner was more concerned about protocol when dual-use experiments are conducted. After reading Sensenbrenner’s letter, one not only has to ask — as he does — what, if any, systems are in place to conduct dual-use assessments before the research is underway, but also whether such a system should be located within NIH. Obviously, NIH brings a particular institutional culture to the dual-use issue, one that emphasizes the beneficial aspects of virology in the case at hand and in life science research more broadly. This institutional approach has also been visible in NIH’s attempt to internationalize the H5N1 debate.

3. Hijacked global governance. This leads to the international dimension of dual-use governance and the role of the World Health Organization, which convened a technical consultation meeting in Geneva, Switzerland, this February. Anthony Fauci, director of the National Institute of Allergy and Infectious Diseases (NIAID), recently said that NIH had been concerned about perceptions of a “US-centric process” and therefore recommended and supported the WHO meeting. Unfortunately, the WHO event in Geneva was a gathering “mostly of public health experts and flu scientists, with only token representation at most other relevant disciplines.” Given the composition of the technical consultation meeting, it does not come as a surprise that the key findings support scientists in their aim to be able to continue pursuing and publishing their H5N1-related research. According to Fauci, who participated in the Geneva meeting, a consensus emerged that “being able to pursue openly this type of research by the public health and scientific community outweighs the issues of a terrorist getting enough information to do something nefarious.” So this ad hoc group that was lacking in biosecurity expertise arrived at a conclusion that is diametrically opposed to the NSABB recommendation, which was arrived at by following established, albeit limited, governance procedures. Why the group didn’t seek out such biosecurity expertise remains a mystery.

Equally baffling is the absence of any reference to the Biological Weapons Convention (BWC) in this international discourse. This is all the more surprising as BWC states parties — of which the United States is a rather prominent one — have since 1991 in successive BWC review conferences agreed that “experimentation involving open-air release of pathogens or toxins harmful to humans, animals and plants that have no justification for prophylactic, protective or other peaceful purposes is inconsistent with the undertakings contained in Article I.” This understanding was most recently reconfirmed during the Seventh BWC Review Conference in December 2011. It seems that all contributors to the dual-use debate — scientists across the planet, including those from NIAID, NIH, and WHO — have been blissfully unaware of this potential restriction on transmissibility research. So unaware, in fact, that nobody has even sought confirmation of whether the H5N1 research does or does not violate BWC obligations.

Perhaps Representative Sensenbrenner’s letter will inspire the White House to put an end to this messy debate and establish a clear oversight system based on international consensus. The H5N1 case, in which US-funded research was carried out in the Netherlands, clearly demonstrates that the days of national regulations being sufficient are over. Such consensus should be initially sought among the core group of BWC states parties who have contributed to the science and technology assessments for past BWC review conferences. In the United States and abroad, the question that must be addressed is whether the institutions funding viral research — like NIH — can also be at the center of a dual-use oversight system that seeks to prevent misuse in the life sciences. More broadly, states urgently need to link up their science and biosecurity policies. As Los Alamos’s Murillo highlights: “Fundamentally, the way biological research is done needs to change. … And if increased security hampers scientific research and progress in influenza virus or other pathogen research, so be it. Significantly worse things could happen.” And as the WHO technical consultation meeting in Geneva has shown, increased security is a lot less likely if one puts a public health fox in charge of the biosecurity henhouse.