A spirited—at times, bordering on nasty—public forum on recent H5N1 avian flu research highlighted the diverse opinions within the scientific community on just how to deal with dual-use research, which is officially defined as research that can be used for good or ill purposes. The experimental work discussed at the Feb. 2 forum involved directed mutations of the H5N1 virus such that aerosol transmission between mammals (ferrets) was achieved for the first time, a mutation that has not occurred in nature (C&EN, Feb. 6, page 6).
The forum, hosted by the New York Academy of Sciences (NYAS), capped an intense week of news about the yet-to-be published research. Just days earlier the National Science Advisory Board for Biosecurity (NSABB) recommended that work done by independent groups in the U.S. and the Netherlands and reported in two different papers be published only in greatly redacted form—that is, with the experimental methods and results excised from the publications. The on-hold papers have already been peer-reviewed and accepted by the journals and .
The event also followed a call for a 60-day moratorium agreed to by one of the two principal investigators involved, Yoshihiro Kawaoka of the University of Tokyo and University of Wisconsin, Madison (Nature, DOI: 10.1038/nature10884). The other PI is Ron A. M. Fouchier of Erasmus Medical Center, in the Netherlands, who is said to be in agreement. Neither researcher attended the NYAS forum.
The H5N1 research brings to light a bigger question about the broad definition of dual-use research and the absence of any agreed-upon protocols to deal with it. This ambiguity, NYAS panelists noted, potentially raises concerns about a multitude of federally funded research projects across scientific disciplines.
“The focus is on H5N1, but there are broader issues at play—this is just an example,” said NYAS panel moderator Ian W. Lipkin, a professor of epidemiology, neurology, and pathology at Columbia University. He pointed out that there is a need in the scientific community to focus on the “intellectual, emotional, and ethical” concerns surrounding dual-use research from any discipline.
The forum attracted a packed auditorium of students, academics, journalists, and members of the general public at NYAS headquarters in lower Manhattan, just a few blocks from the new 9/11 Memorial and the rising towers of the city’s new World Trade Center. The irony of holding this dual-use discussion so near Ground Zero was noted by forum participants.
The panelists included prominent influenza researchers in the U.S., NSABB members, a former science journalist who has written extensively about pandemic health threats, and editors from Science and Nature who were intimately involved with the handling of the papers in question.
Lipkin in his opening remarks commented on the many stakeholders in influenza research—the public, scientists, pharmaceutical companies, and people in the developing world where such diseases are often endemic. “Scientific investigation of pathogenic diseases is essential,” he said, and H5N1 is no different. “If we don’t make a plan ourselves,” he said, regarding the dual-use characteristics, “research critical to the public interest could be curtailed.”
NSABB Acting Chair Paul Keim, a microbiology professor at Northern Arizona University who attended the forum via webcast, and other members of the board said the H5N1 papers were brought to their attention in October 2011 from various sources, including program officers at the National Institutes of Health. Editors from Science and Nature insist that they never referred either of the papers to NSABB—and never would. Science Deputy Editor Barbara R. Jasny, who was an NYAS forum panelist, said that her journal did convene its own body of advisers when the difficult ethical issues regarding publication of the research became apparent.
Michael T. Osterhelm, NSABB member and director of the Center for Infectious Disease Research & Policy at the University of Minnesota, emphasized that the 23 members of the advisory board were unanimous in their recommendation that only greatly redacted versions of the Fouchier and Kawaoka papers be published. He was also a forum panelist.
“I would hesitate to say that we are establishing a protocol,” Osterhelm said. “We are not the final arbiter.” NSABB members, he said, are recommending a step back because once the papers’ research methodologies and results become widely available, “you can’t unring the bell.” Anyone seeking the information, for any purpose, could readily find it, he pointed out.
And while research journals and federal agencies are under no obligation to follow NSABB’s recommendations, the advisory board and its decisions do carry weight in the scientific community. The board was created in light of a recommendation in the 2004 National Research Council report, “Biotechnology Research in an Age of Terrorism.”
NSABB members came to the table to discuss the H5N1 papers with differing opinions, recalled member Arturo Casadevall, a professor of microbiology at Albert Einstein College of Medicine. Initially, he said, he was opposed to the recommendation for publishing only redacted versions of the research papers. But “the process is deliberative and people can change their minds,” he said. “I changed mine.”
But other NYAS panel members are alarmed by the NSABB recommendation, which they see as an effort to shut down productive lines of research and censor experimental results.
“We need to specifically address the situation of H5N1,” said Peter Palese, chair of the department of microbiology at Mount Sinai School of Medicine. The Fouchier and Kawaoka research groups, he said, “are passaging an H5N1 virus in ferrets. This is perfectly legitimate,” and many other groups have done similar work, he noted, referring to the experimental method in which viral adaptation is achieved by repeated passage in cells or in naive animal hosts. He said the experiments are necessary because they help researchers identify the very mutations they should be monitoring in wild strains of the virus that could signal onset of a human pandemic.
Palese went on to question the case fatality rate of H5N1 infection in humans, which is estimated to be in the range of 50 to 80%. He explained that this rate is calculated only from documented cases of human H5N1 infection as defined by World Health Organization (WHO) guidelines. The overwhelming majority of infections, including those that do not result in death or that create asymptomatic H5N1 carriers, are not likely to be reported to WHO, he said. These undocumented infections occur in poor, rural areas where medical care, including access to advanced laboratory diagnostics, is not likely. Because the number of nonlethal cases is being underreported, Palese said, “the case fatality rate is much lower than what WHO is putting out.”
The questioning of the case fatality rate by Palese led to a testy exchange with NSABB member Osterhelm, who accused Palese of putting forward false information. “The study you continue to cite is the worst one of all,” Osterhelm said, referring to a January Proceedings of the National Academy of Sciences article cowritten by Palese and Mount Sinai School of Medicine colleague Taia T. Wang, titled “H5N1 influenza Viruses: Facts, Not Fear” (DOI: 10.1073/pnas.1121297109).
Casadevall chimed in on the exchange and said, “The [fatality] numbers are unbelievable no matter any way you look at it. When these things [the experimental viruses] get out and combine with existing strains, the situation becomes completely unpredictable.”
Palese continued to draw fire when he questioned the use of ferrets in the experiments. “People are equating transmission in ferrets with ease of transmission in humans,” he said. The animals are indeed a model, he continued, but the implications for human disease are unknowable at this time. “In fact, the ferret is much too sensitive” to make good predictions about human infection, Palese said. “An animal model is just that—a model.”
“But then what would be a good model, sir?” fellow panelist Laurie Garret, a senior fellow at the Council on Foreign Relations and a former journalist who has written extensively on emerging diseases, shouted across the table.
A visibly shaken and angry Palese responded: “These are not conclusive experiments, they are not compelling, and we are using them to shut down a lot of science.”
Other panelists also raised questions about how to advance important but risky research. Jasny tried to get at a root question: “What is it that causes virus to jump from species to species—to become more virulent?” Getting the answer to this and other important questions in virology requires dual-use research, she said. “Any kind of research can have risk associated with it—it doesn’t mean we ignore the risk. How do we move ahead?”
Panel member Alan S. Rudolph, director of the joint science and technology office for the government’s Chemical & Biological Technologies Directorate, noted concerns about including international perspectives on the research—some of which did occur outside the U.S.
“The problem extends well beyond our borders,” Rudolph said. He noted that quite a lot of activity on the H5N1 experiments is starting to happen abroad, including a closed-door meeting at WHO headquarters in Geneva scheduled for Feb. 16 and 17. No agenda has been posted, but the invitation-only conference is likely to touch on data sharing and other issues, including the possibility of distributing the mutated experimental viruses to other researchers.
Garrett noted that international views on how to handle this research may be very different than the consensus in the U.S. She said it is important to note that, when people discuss what is ultimately a health care issue, the U.S. is often criticized in the international community. The U.S. insistence on intellectual property protections for prescription drugs, she said, is often seen as an uncaring obstacle to health care for people in the developing world.
As for H5N1, Garrett pointed out a newspaper editorial in Indonesia that hails Fouchier as a hero for achieving breakthroughs on H5N1 research that could lead to new protections for people in a country with potentially huge reservoirs of the virus in domesticated birds.
NSABB’s recommendation to redact information prior to publication of the papers, however, comes a bit late, several panelists noted. This is because the Fouchier and Kawaoka papers—including methodology and results—have already been discussed at a meeting in Malta last year. What’s more, the eyes of many experts in influenza research have seen the papers during peer review. Some panelists estimated that about 1,000 people around the world have read the papers or have direct knowledge of the work, leading to further speculation about the value of at least some limited censorship of the data now.
If the papers do appear in Science and Nature in redacted form, which seems all but certain, the panelists agreed that sharing the full data sets and deciding which researchers are qualified to receive them will be fraught with difficulty.
“We need to maintain the integrity of the scientific process,” said panelist Véronique Kiermer, executive editor for Nature and the Nature journals. She said the roles of serendipity, interdisciplinary attention, and other factors in scientific research cannot be underestimated for sparking “things that can be field-changing.” In any system to limit distribution of the redacted data sets, “we need to be very careful about how to establish these criteria and decide who these people will be,” she said.
So far Nature has made no public statement about its plan for the H5N1 paper under its control. Science Editor-in-Chief Bruce M. Alberts has stated his support for the NSABB recommendation but has also suggested that the journal wait for a credible plan for data sharing from the U.S. government before publishing the paper it controls.
As the forum concluded, the panelists remained split in their opinions. All seemed to agree that biology was at another crossroads, especially with the growing body of research in synthetic biology, and that responsibility for the safety and welfare of the general public should be of paramount concern.
“But we can’t be wrong,” said Osterhelm of the H5N1 research. “There’s no going back. We have no choice but to act on this one.”
Palese asked perhaps the ultimate question about H5N1 and any dual-use research: “Where do you stop being afraid?”