by Stephen Braigen
By now you have likely heard some whisperings of controversy in the influenza world. If not, I’ll attempt to give a brief summary, though this situation, like many controversies in science, is not a simple tale.
Last fall, Dr. Yoshihiro Kawaoka and Dr. Ron Fouchier submitted manuscripts to Nature and Science, respectively, detailing their work on transmissibility in mammals of a particularly virulent strain of influenza (hereafter referred to as H5N1). The ability to jump from one mammalian host to another is a trait that H5N1 in the wild has not perfected, much to our relief. These works outlined the generation and passage of mutant H5N1 strains through a mammalian host (ferrets, in this case). The exact methodology is neither available, nor is it particularly relevant to this commentary, but the takeaway is that the strains were generated in part by the directed manipulation of their genomes, and that in the end, both groups generated viruses that were transmissible in ferrets.
This transmissibility generated concerns of dual use potential – the prospect of using research for malicious purposes, as a bioweapon, for example, and consequently the papers were brought to the attention of the National Science Advisory Board for Biosecurity (NSABB). After a lengthy period of discussion, the group recommended that the papers be published, albeit with specific details of both groups’ methodologies redacted in press, but available to individuals who are deemed to be deserving of the full techniques. Both Nature and Science tentatively agreed to this recommendation, on the condition that such a mechanism for release of the complete paper to appropriate parties was constructed in a timely fashion.
This recommendation evoked concern over setting precedent for censoring research, notably amongst prominent virologists Peter Palese and Vincent Racaniello. Numerous arguments have been made on the internet, in academic journals and perhaps most heatedly in a panel discussion at the New York Academy of Science in which both Palese and Racaniello had the opportunity to discuss the issue with members of the NSABB – Michael Osterholm, director of the Center for Infectious Disease Research and Policy (CIDRAP), and Arturo Casadevall, the Leo and Julia Forchheimer Professor and Chair of the Department of Microbiology and Immunology here at the Albert Einstein College of Medicine.
The argument by critics of the NSABB is two-pronged; first off, that transmissibility in a ferret does not equivocate to transmissibility in humans, and as such, these mutant strains do not pose a biosecurity threat. Second, that H5N1 itself may not be as dangerous as it seems. This contention revolves around the World Health Organization’s oft-cited 59% case fatality rate. The issue, some say, is that the true number of people who develop infections may not be reflected in this rate, as there may be individuals who develop limited symptoms or none at all, and consequently would not be detected as having the disease. This argument is supported by a number of studies detecting H5N1 antibodies in individuals with no clinical history of the illness.
On top of all this, at a lecture at the ASM Biodefense meeting in Washington on February 29, 2012, Dr. Fouchier revealed that his mutant strain is not nearly as virulent as the public and scientists following the story were led to believe. The ferrets who were exposed to the mutant virus by aerosol transmission from infected ferrets did not display any apparent disease. The only situation which led to high rates of infection was through delivery of large quantities of the virus to the lower respiratory tract – in other words, an unlikely method of transmission. In effect, Fouchier’s group generated an attenuated strain of this virus, which could be a great tool for developing treatments and vaccines against the pathogenic strain of H5N1.
Naturally, there are counters to the arguments – on February 24th, the open access journal mBio published an opinion piece by Osterholm and Nicholas Kelly, another member of the CIDRAP staff, thoroughly breaking down the idea that we may be underestimating the number of people contracting H5N1 and not getting sick from it.
While Osterholm may make a good argument, what is being discussed no longer concerns the H5N1 research in question. I am neither the first nor the most authoritative person to point out that a redaction will do little to prevent the specific methodology from being accessible. This debate now centers around setting precedent on redacting science and as a consequence limiting the dissemination of information.
The argument presented at the NYAS panel discussion and elsewhere is that access to the materials and methods of these manuscripts must be limited not for fear of terrorists, morally bankrupt state entities or other nefarious groups, but because of curious DIY biologists, amateurs, and hobbyists. Osterholm drew the analogy to early computer hackers, breaking into secure systems just to prove that they can. His concern is that someone – either in a laboratory setting or in a garage with makeshift equipment – will generate these strains to prove that they can. Dr. Casadevall made a similar comparison to amateur bomb-making.
Here’s where the problems lie. Cutting away all the “might be virulent” and “H5N1 isn’t as bad as we think” arguments, at the heart of the matter lie two papers, which outline two distinct ways to generate influenza strains transmissible amongst mammals. As far as we can tell without the full papers in our hands, neither work details a particularly novel method of doing so – the redacted information will only go so far as limiting the specific pieces of DNA that were changed in Fouchier’s initial experiments. In addition, this work has been presented at conferences, in full, to a large number of people. In short, this work would not be hard for anybody with a bit of background in virology to reproduce, even without the specific mutations.
In addition, the full genomes of various deadly pathogens, such as the 1918 H1N1 flu and smallpox, have been published. But I have another example, which juxtaposes the current concerns of hobbyists and amateurs accidentally making a highly virulent pathogen. Bacillis anthracis is a common soil bacterium, which forms spores when deprived of nutrients. These spores are hardy, small and most importantly in the context of biosafety, can be aerosolized. When aerosolized spores are inhaled, they set up camp in the lungs, actively growing and producing the deadly anthrax toxin, leading to an infection that is more often than not fatal.
While it would certainly not be simple to do, a determined hobbyist or DIY biologist could generate such spores and with some elbow grease, develop an aerosol delivery system. Such work would not require the genetic manipulation that these H5N1 mutants did. And yet, there is no proposal to regulate to methodology required to facilitate such activities. Even more damning to the suggestion to limit such information is that no DIY biologist, hobbyist, or terrorist has done this, for any reason.
Of course we should have boards like the NSABB and processes, as difficult as they may be, to determine the risks of certain research, as well as systems to safeguard against accidental or deliberate release of dangerous biological agents, but the research performed by Drs. Kawaoka and Fouchier is not the point at which we should start limiting access. Not because either side of the discussion is right or wrong about the dangers presented by these mutants, but because limiting access will only serve to set precedent for policies regarding the censorship of science. The last thing we need is a new way for policymakers and politicians to quell the dissemination of information.