Dual Use Research

Journal Articles

General Resources

Facing up to flu. Nature. 2012;482(7384):131-131.

Faden RR, Karron RA. The Obligation to Prevent the Next Dual-Use Controversy. Science. 2012;335(6070):802-804.

Fidler DP, Gostin LO. The WHO pandemic influenza preparedness framework: a milestone in global governance for health. JAMA. 2011;306(2):200-201.

Fouchier RAM, Herfst S, Osterhaus ADME. Restricted Data on Influenza H5N1 Virus Transmission. Science. 2012;335(6069):662-663.

National Research Council (U.S.). Biotechnology research in an age of terrorism. Washington, DC: National Academies Press, 2004.

Public health, influenza experts agree H5N1 research critical, but extend delay: WHO to convene additional meeting to discuss next steps. World Health Organization news release. 2012.

Arturo Casadevall

Casadevall A, Relman DA. Microbial threat lists: obstacles in the quest for biosecurity? Nat. Rev. Microbiol. 2010;8(2):149-154.

Casadevall A, Shenk T. The H5N1 manuscript redaction controversy. MBio 2012;3(1).

Casadevall A, Ehrlich SA, Franz DR, Imperiale MJ, Keim PS. Biodefense research: a win-win challenge. Biosecur. Bioterror. 2008;6(4):291-292.

Imperiale MJ, Casadevall A. Bioterrorism: lessons learned since the anthrax mailings. MBio 2011;2(6):e00232-00211.

Laurie Garrett

Garrett L. Gaps between the rich and the poor. The widening differences in wealth, life expectancy, public health infrastructure and perception of threats, and the consequences for global security. EMBO Rep. 2003;4 Spec No:S15-19.

Garrett L. The path of a pandemic. Newsweek 2009;153(19-20):22-28.

Garrett L, Fidler DP. Sharing H5N1 viruses to stop a global influenza pandemic. PLoS Med. 2007;4(11):e330.

Barbara R. Jasny

Chong L, Wible B, Jasny B. Dimensions of science cooperation. Science 2011;334(6059):1066.

Jasny BR, Zahn LM. Genome-sequencing anniversary. A celebration of the genome, part I. Science 2011;331(6017):546.

Jasny BR, Chin G, Chong L, Vignieri S. Data replication & reproducibility. Again, and again, and again... Introduction. Science 2011;334(6060):1225.

W. Ian Lipkin

Gu J, Xie Z, Gao Z, et al. H5N1 infection of the respiratory tract and beyond: a molecular pathology study. Lancet 2007;370(9593):1137-1145.

Tokarz R, Kapoor V, Wu W, et al. Longitudinal molecular microbial analysis of influenza-like illness in New York City, May 2009 through May 2010. Virol. J. 2011;8:288.

Wrammert J, Koutsonanos D, Li G, et al. Broadly cross-reactive antibodies dominate the human B cell response against 2009 pandemic H1N1 influenza virus infection. J. Exp. Med. 2011;208(1):181-193.

Michael T. Osterholm

Berns KI, Casadevall A, Cohen ML, et al. Policy: Adaptations of avian flu virus are a cause for concern. Nature 2012;482(7384):153-154.

Berns KI, Casadevall A, Cohen ML, et al. Public health and biosecurity. Adaptations of avian flu virus are a cause for concern. Science 2012;335(6069):660-661.

Osterholm MT. Preparing for the next pandemic. N. Engl. J. Med. 2005;352(18):1839-1842.

Osterholm MT, Henderson DA. Life sciences at a crossroads: respiratory rransmissible H5N1. Science 2012.

Peter Palese

Fouchier RAM, García-Sastre A, Kawaoka Y, et al. Pause on avian flu transmission research. Science 2012;335(6067):400-401.

Hai R, García-Sastre A, Swayne DE, Palese P. A reassortment-incompetent live attenuated influenza virus vaccine for protection against pandemic virus strains. J. Virol. 2011;85(14):6832-6843.

Palese P, Wang TT. H5N1 influenza viruses: facts, not fear. Proc. Natl. Acad. Sci. USA 2012;109(7):2211-2213.

Varga ZT, Palese P. The influenza A virus protein PB1-F2: killing two birds with one stone? Virulence 2011;2(6):542-546.

Vincent Racaniello

Racaniello V. An update on the African polio outbreak. Interview by Kathryn Claiborn. J. Clin. Invest. 2011;121(2):460.

Racaniello V. Virology. An exit strategy for measles virus. Science 2011;334(6063):1650-1651.

Racaniello VR. Science should be in the public domain. MBio 2012;3(1).

This Week in Virology (TWiV) is a podcast about viruses, begun in September 2008 by Vincent Racaniello and Dick Despommier. The goal of the podcast to have an informal yet informative conversation about viruses which would be accessible to everyone, no matter what their science background.

Alan S. Rudolph

Rudolph AS, Coughlin JR. Developing advanced toxicology technologies for biomonitoring in national security applications. J. Appl. Toxicol. 2004;24(5):315-316.

Keim PS. Q&A: Reasons for proposed redaction of flu paper. Nature 2012;482(7384):156-157.

Sponsor

Silver Sponsor

Moderator:
W. Ian Lipkin, Columbia University

Panelists:
Arturo Casadevall, Albert Einstein College of Medicine; Member, NSABB
Laurie Garrett, Council on Foreign Relations
Barbara R. Jasny, Science
Véronique Kiermer, Nature Publishing Group
Michael T. Osterholm, University of Minnesota; Member, NSABB
Peter Palese, Mount Sinai School of Medicine
Vincent Racaniello, Columbia University
Alan S. Rudolph, Defense Threat Reduction Agency

Highlights

• No formal mechanisms currently exist by which journals can share potentially harmful research with advisory or regulatory groups.
• A special working group of the National Science Advisory Board for Biosecurity unanimously voted to recommend the exclusion of key methodological details from the published H5N1 studies.
• Experts differ on the case fatality rate of wild type H5N1, but most agree the virus would be devastating if it spread from human to human.
• Panelists debated the merits of ferret models of airborne virus transmission, the value of seroprevalence tests versus hospital diagnoses in determinations of infection rate, and the problems of using natural virus transmission to predict how leaked mutant viruses might behave.

Provoking conversation: The role of the NSABB

The discussion, moderated by Ian Lipkin, a world-renowned pathogen scientist from Columbia University, began with an explanation of the role of the NSABB in this particular controversy and more generally. In particular, panelists discussed how these studies came to the Board's notice. As Arturo Casadevall, a member of the NSABB and a researcher at Albert Einstein College of Medicine, described the organization, since 2005 it has been an advisory board mainly of scientists who report their findings to the federal government. The committee is responsible for biosecurity oversight in "dual use" research, defined by the NSABB as "biological research with legitimate scientific purpose that may be misused to pose a biologic threat to public health or national security." Barbara Jasny, an editor at Science, recalled that the journal's review team held an internal biosecurity evaluation because they were aware of the implications Fouchier's paper presented. However, like any other scientific journal, they do not have any mechanism in place to refer their concerns directly to the NSABB. In fact, it was the staffers at the National Institutes of Health (NIH)—the organization responsible for funding this research—who officially brought the paper to the NSABB's attention.

When asked to explain the process of evaluation, Michael Osterholm, another panel member from the NSABB and a professor at the University of Minnesota, clarified that there is no formal protocol because the NSABB does not have a significant history in dealing with these kinds of cases, in which the research is ready for publication. Generally, the Board prefers to evaluate research at a much earlier stage. Despite the lack of protocol, the NSABB formed a working group to discuss the manuscripts with input from infectious disease experts, the editors of the two journals involved, and the authors of the studies. After their deliberations all 23 members of the NSABB decided to recommend the exclusion of procedural details and of the identities of the specific mutations from the two papers in an effort to prevent the replication of the results (a more contagious virus) by anyone intending harm. Osterholm emphasized that the NSABB does not see itself as a final arbitrator in such issues. Instead, it aims to play an advisory role in the initial stages of discussion, prompting an international dialogue that they hope will give rise to a framework that prioritizes public safety. He firmly stated that the ideal approach should tackle the conflicts earlier on in the process and not when the research has already been done and the manuscript is ready for publication. Osterholm recognized that this process has been a learning experience for NSABB, one that will shape future research evaluation strategies.

The roots of concern: How deadly is H5N1?

The NSABB's and others' concern about publishing these studies' complete methods is not just a generic hesitation about work that makes a virus more contagious. Their objection is particularly fervent—strong enough to motivate recommending redaction—because of the belief that a more contagious H5N1 poses an unusually deadly threat. The debate over case fatality rate for H5N1 stirred contrasting arguments from Michael Osterholm and Peter Palese, from Mt. Sinai School of Medicine. A case fatality rate reflects the percentage of people formally diagnosed with a disease who die of that disease in a given time period. Palese disagreed with the World Health Organization (WHO) data, which list 583 confirmed cases with 344 deaths since 2003, a case fatality rate over 59%. He pointed out that those data only consider hospital cases and do not count any asymptomatic H5N1 individuals. Since individuals are often only brought into hospitals for treatment when their condition becomes more serious, counting only hospital diagnoses can drastically inflate death rates. Based on 10 seroprevalence studies Palese cited in a recent Proceedings of the National Academy of Sciences article, the estimated infection rates ranged from 0.2%–5.6%. Palese argued that seroprevalence studies, which test for the presence of antibodies to virus antigens, better track asymptomatic or minimally symptomatic cases than do hospital numbers. Results from these studies would dramatically reduce the publicized WHO case fatality data if taken into account. Michael Osterholm countered by claiming that the studies Palese cited do not meet the WHO's criteria of H5N1 serology and that they overestimate the number of seropositive subjects. Osterholm suggested that even if the virus is 20 times less virulent than current WHO estimates indicate, the virus would still be more destructive than the 1918 pandemic flu virus if it could spread from human to human. Arturo Casadevall agreed with Osterholm's pandemic scenario but also stressed that we must be careful in using seroprevalence data of a naturally occurring virus to predict the course of an artificially engineered virus. The latter could have far more unpredictable consequences if it leaks out to the environment and mutates spontaneously.

Affected areas with confirmed human cases of H5N1 avian influenza from 1 January to 31 December 2006, a peak in reported human infections. (Image courtesy of Public Health Mapping and GIS Communicable Diseases (CDS), World Health Organization 2007)

Realizing that there wouldn't be consensus on the validity of the WHO's mortality data and on the specifics of the H5N1 serology assays, Ian Lipkin steered the debate towards a general conversation about "dual use" research and its importance. Barbara Jasny acknowledged the harmful implications of such research, but stipulated that the value of understanding the transferability of H5N1, of analyzing virulence characteristics, and of developing therapeutic strategies in advance of a possible epidemic are benefits that must be weighed along with the risks of "dual use" research.

In addition to assessing the virus's virulence in humans, researchers seeking a measure of the risk posed by the engineered H5N1 must find a reliable predictor of its ability to pass from human to human. The suitability of ferrets—Fouchier's and Kawaoka's chosen subjects—as a model to assess mammalian transmission of the mutant H5N1 virus was also a significant point of discussion. Vincent Racaniello, a virologist from Columbia University, warned the audience that an animal model should be considered hypothetically and not as an exact standard for accurately predicting what will transpire in humans. Peter Palese echoed these beliefs, adding that the ferret model is much too sensitive to model human-to-human transmission and inconclusive results are being used improperly to shut down further scientific investigation that may provide critical insight on the genetic characteristics of influenza. Laurie Garrett, Senior Fellow for Global Health at the Council on Foreign Relations, questioned why these experiments were done on ferret models to begin with if the animals have little value for predicting virulence in humans. By contrast, Arturo Casadevall and Michael Osterholm affirmed that many flu experts from around the world, in their discussions with the NSABB, agreed that the ferret model was a suitable predictor of transmission in humans. Accordingly, these experts advised the NSABB to treat the studies' findings as a serious concern.

Managing method: The proper place of government oversight?

But should the government be the authority on the choice of a particular animal model or a specific method for "dual use" research? What, indeed, is the role of government regulation in such research? Laurie Garrett opined that the task of government is to defend the public from potential danger, and it should promote this type of influenza research if the threat is a natural pandemic. However, according to Garrett, there is a complicating factor in this situation because of the perceived threat of bioterrorism and potential accidents. In such cases, she said, the government has the responsibility to step in and regulate the activities for the sake of public safety. Thus, the definition of threat determines the extent of regulation, which is very difficult to impose on an international level, even more so than on a national level. Garrett conceded that there is little consensus on the role of international governing bodies, and the WHO's controversial international agreement to share influenza strains between countries, for the purpose of promoting research and outbreak preparedness, falls short in compelling developed countries to provide specific benefits (e.g., vaccine donations) to developing countries.

Alan Rudolph of the Defense Threat Reduction Agency provided assurance of progress made by the U.S. government in their efforts to prompt the representatives to global governing bodies to reach agreement on international policies regarding biosecurity. Michael Osterholm suggested that the level of regulation should be proportional to the level of risk, and the NSABB followed this principal during their considerations. Arturo Casadevall added that all research is "dual use" research, meaning that all research has the potential to be harmful in some way, but the NSABB has specified a distinct category called "dual use research of concern" to identify work that warrants careful review and deliberation. The current mutant influenza studies in question, fall under the category of "dual use research of concern" according to the NSABB. Rudolph explained that the NSABB has advised the government on "dual use" research and has helped increase the awareness of risks involved, which will shape policies to come. He also asserted that the agencies and institutions that sponsor the research should play a part in regulating activities.

Putting the H5N1 research in the context of the handling of other (potentially) pandemic viruses, the panelists compared the mutant H5N1 virus to the smallpox virus. Osterholm proclaimed that a smallpox virus leak, which is housed under strict bio-safety level 4 (BSL-4) conditions, would not concern him as much as a leak of influenza would. His fear of smallpox is diminished because there is a proven vaccine and an outbreak can be effectively managed. Casadevall, in support of his colleague, added that smallpox has a longer incubation time, which helps in controlling rapid transmission. Palese strongly disagreed with Osterholm on the seriousness of influenza transmission, stating that H5N1 has been attenuated over the decades, has not been definitively proven to transmit between humans, will not prove to be harmful to humans unless they are exposed to large doses, and is far less dangerous than smallpox.

Moderator:
W. Ian Lipkin, Columbia University

Panelists:
Arturo Casadevall, Albert Einstein College of Medicine; Member, NSABB
Laurie Garrett, Council on Foreign Relations
Barbara R. Jasny, Science
Véronique Kiermer, Nature Publishing Group
Michael T. Osterholm, University of Minnesota; Member, NSABB
Peter Palese, Mount Sinai School of Medicine
Vincent Racaniello, Columbia University
Alan S. Rudolph, Defense Threat Reduction Agency

Highlights

• Some argued that restricting the flow of scientific information could leave us more vulnerable to a natural or artificial pandemic virus.
• The place of regulation domestically and internationally is uncertain, but clearly more oversight is needed to secure information and material from dual use research.
• Overregulation needs to be avoided, and some scientists fear that redaction will become an institutionalized response to dual use research.
• Restricting information could prevent key studies from being reproduced by other labs—an important part of scientific validation.

Where to from here? Options for dealing with "dual use" research

One alternative to direct government enforcement is the possibility of self-regulation by scientists. Laurie Garrett described the 2004 National Academy of Sciences (NAS) publication Biotechnology Research in the Age of Terrorism, which looks at "dual use" research issues and provides recommendations to regulate the misuse of scientific knowledge. Highlighting the need for greater standardization, Garrett mentioned that the General Accountability Office (GAO), which evaluates high containment biosafety labs (BSL) around the United States, reported increased proliferation of such facilities and questionable adherence to security standards for BSL-3 or BSL-4 labs, those with the highest security ratings. To enable better self-regulation, the NAS report presents seven guidelines to identify microbial experiments of high risk, specifically those which: 1) demonstrate how to render a vaccine ineffective; 2) confer resistance to therapeutically useful antibiotics or antiviral agents; 3) enhance the virulence of a pathogen or render a nonpathogen virulent; 4) increase transmissibility of a pathogen; 5) alter the host range of a pathogen; 6) enable the evasion of diagnostic/detection modalities; 7) enable the weaponization of a biological agent or toxin. Garrett asserted the H5N1 experiments conform to certain guidelines above and pose more potential harm than benefit.

Barbara Jasny indicated that the report recommends a combination of self-governance by scientists and elaboration of existing regulations. She further clarified that the aim of such oversight is to trigger an early review of questionable research by an institutional biosafety committee and to have an international body harmonizing the efforts all over the world. Garrett voiced her concern that much progress needs to be made to empower institutional committees to regulate research activities in a timely and effective manner. Moreover, despite the national setbacks, it is far more difficult to have an international system of regulation since the governing structure is non-existent. Auturo Casadevall cautioned the audience about over regulation in science, especially in clinical research. Excessive regulation, he claimed, hinders progress and discourages researchers. He lamented that Institutional Review Boards' (IRBs') stringent protocols create an unfortunate conflict between overregulation versus academic freedom. He does not want the system to move towards that form of cumbersome regulation.

Restricting the flow of information

In this case, since the research has already been conducted, questions of method are largely moot. The debate now rests on the appropriate response to this and similar research. Weighing the potential danger against what they see as the compelling scientific and public interest in learning all we can about H5N1, the journals proposed sharing the methods with specific influenza researchers. Ian Lipkin questioned the effectiveness of sharing data with pre-selected researchers, given the serendipitous nature of science. Véronique Kiermer, executive editor of Nature, admitted that allowing only select researchers to access the scientific data will limit collaboration, which is vital to the scientific community. She also agreed that it will be challenging to determine a fair screening process to select researchers with whom the work can be shared. Interestingly, according to Jasny, the number of people who already have knowledge of key parts of the manuscript could be as many as 1000. Restriction of information that has the potential for public harm is critical, according to Osterholm, and such restriction should be proportional to the risk involved. Racaniello disagreed and said that the research community often inaccurately predicts both the risks and rewards of research programs. He raised the example of the accidental discovery of E. coli restriction enzymes, which ultimately revolutionized the field of biotechnology. This example and others show, according to Racaniello, how difficult it is to predict the authors of key experiments, and limiting information exchange could reduce the possibility of such fortuitous experiments occurring.

Panelists debated the various possible responses to research of this nature, including selective dissemination of the information, redaction of the methods from the published papers, and national and international regulation of this and future research on the subject.

On the issue of the containment of material from these experiments, Osterholm agreed that lab leaks do not happen often, but passionately supported the placement of the mutant H5N1 virus in a BSL-4 setting, in case the virus is found to be transmissible among humans. Palese countered the suggestion, adamantly declaring since in his view the current experiments do not conclusively prove H5N1 transmission in humans, the scientific community cannot live in fear. Shifting the focus from scientists, Laurie Garrett reminded the panel that it is important to include the public as a stakeholder in this debate and the WHO must avoid an agreement solely based on the perspective of scientists. The public has a stake in the containment not just of information about the studies but also of the wild type and mutant viruses themselves. She felt that there is underrepresentation of those outside the research community.

Looking to the future

Based on the NSABB's recommendation, Nature and Science have agreed to consider redacting the papers in question. Panelists were eager to discuss the merits—and the future—of this response. Kiermer believes that the H5N1 studies are still worth publishing, even without the complete methods, because the overall purpose of the papers holds value, and the involved scientists deserve credit for their work. She stated that it is important to have a proper mechanism in place that will provide necessary details to the right people, but those specifications have not been established as yet, and hopefully the moratorium on research and the delayed publication will allow for stakeholders to come to a conclusion about how to protect the public without sacrificing scientific collaboration. Barbara Jasny expressed her hope that if redaction occurred, it would be a unique event, not a sign of practices to come. She emphasized the importance of validating scientific research by the reproducibility of the work's results. Reproducibility, which relies on the scientist-to-scientist exchange of information, is a fundamental principle that should be protected, she explained. Moreover, Jasny appreciated the fact that the authors have understood the implications of their work and have chosen to work with the system, rather than to publish the full content elsewhere. Racaniello, on the other hand, expressed his disapproval of the redaction response: once a precedent has been set, he claimed, it is easily repeatable. Casadevall reassured the audience that no one wants to see redaction, but in this case it could be a temporary measure that limits a potential danger while a consensus can be reached. Garrett expressed that the WHO is in a very difficult position right now in deciding who should have access to virus strains, and the group will have to face some tough decisions ahead. Beyond the issues already discussed by the panel, Garrett brought the audience's attention to the general distrust among developing countries of Western drug companies that have made large profits on vaccines. The vaccines have been developed from isolates discovered in countries who feel they have been exploited for profit because, frequently, neither the vaccines nor the profits return to the poorer nations where the viruses were harvested. The WHO will be dealing with a politically charged climate once stakeholders reconvene in Geneva to further discuss the fate of "dual use" research in general and of mutant H5N1 studies in particular.

The panel discussion ended with a conversation about what could have been done differently to make this situation less politically charged, to render the journals', scientists', governments', and public's response to the research less crisis-driven. Palese and Racaniello concurred that the whole process started out on the wrong foot. Although the authors'—Fouchier's and and Kawaoka's—decision to publish their work through normal channels was admirable, Racaniello criticized Fouchier's characterization of the work. Fouchier's description in his initial interview in Science was "inflammatory," and probably made things worse, according to Racaniello. Silence, however, would not have been the best alternative, Garrett qualified. Kawaoka has added to the uncertainty of the situation by declining to discuss the work until recently, she claimed. Ultimately, much of the crisis could have been allayed by genuine, open, and early discussion among stakeholders from every sector, including the public, Jasny articulated. These conversations, she explained, could have continued from the founding of the NSABB in 2005, and would have, at that time, laid the groundwork for a separate body to evaluate dual use research before it was even funded. Open discussion is a crucial part of achieving understanding and reducing fear of the work and its many implications.