News & AnalysisBiosecurity

Will Flu Papers Lead to New Research Oversight?

See allHide authors and affiliations

Science  06 Jan 2012:
Vol. 335, Issue 6064, pp. 20-22
DOI: 10.1126/science.335.6064.20
Game changer?

Efforts to make H5N1 (left) transmissible in mammals rekindle “dual use” concerns.


In the fall of 2001, just weeks after the trauma of 11 September, letters laced with powdered anthrax caused death and panic in the United States. Ever since, biological scientists have debated whether, one day, the need to keep sensitive information from aspiring bioterrorists would force them to impose new limits on the academic openness they had long taken for granted.

A decade later, that day appears to have arrived. Just before Christmas, the U.S. government announced that a biosecurity advisory board had asked two research teams to strike key details from papers in press at Science and Nature. The studies describe how researchers made the deadly H5N1 avian influenza more transmissible between mammals—possibly providing a blueprint on how to set off a flu pandemic. The researchers and the journals agreed, but only if the U.S. government comes up with a system that allows “responsible” scientists to see the deleted information, which public health experts say could be crucial to monitoring H5N1 outbreaks and developing drugs and vaccines.

The unprecedented decision by the National Science Advisory Board for Biosecurity (NSABB) has sparked fierce criticism and strong support. And the episode has implications far beyond a couple of paragraphs in a pair of papers. Already, senior U.S. officials are scrambling to develop new, tougher oversight procedures for evaluating and possibly blocking potentially risky “dual use” studies before they begin—reviews that critics say the flu experiments should have received. And officials are struggling to devise a workable system for sharing the redacted details with some scientists but not others. “There is no perfect solution” to that problem, says Anthony Fauci, director of the U.S. National Institute of Allergy and Infectious Diseases (NIAID) in Bethesda, Maryland, which funded the studies. “There is not even a good solution.”

“This is an Asilomar moment.”



The World Health Organization (WHO), meanwhile, fears that the fallout could damage a 2011 global agreement to share influenza samples that was years in the making. And NSABB chair Paul Keim, a microbial geneticist at Northern Arizona University in Flagstaff, would like to see a voluntary moratorium on publicizing similar studies pending international talks on how to proceed. “This is an Asilomar moment,” Keim says, referring to a landmark 1975 meeting in Asilomar, California, where, after halting their research, scientists drew up safety guidelines for working with recombinant DNA technology.

The studies—led by Ron Fouchier of Erasmus MC in Rotterdam, the Netherlands, and Yoshihiro Kawaoka of the University of Wisconsin, Madison, and the University of Tokyo's Institute of Medical Science—were designed to see if changes in H5N1's genetic makeup might make it more capable of jumping from human to human. Kawaoka received an NIAID grant in 2006, while Fouchier's work was done under a subcontract for Adolfo García-Sastre of Mount Sinai School of Medicine in New York City, who runs an NIAID-funded influenza center.

The studies had not raised eyebrows before. Their dual-use aspects “didn't hit the radar screen” of the scientists who reviewed the proposals for NIAID, Fauci says, in part because “similar types of research looking at alteration of transmissibility have been going on forever.” University biosafety committees in the United States and the Netherlands also green-lighted the work; such panels typically focus on lab safety, not dual-use aspects.

Little is known about the content of Kawaoka's study, under review by Nature. But Fouchier discussed his work, under review at Science, at a September meeting in Malta. His lab initially tried making the virus more transmissible in ferrets—virologists' preferred animal model—by mutating key genes. That didn't work, so researchers tried an old method called “passaging”: transferring the virus from ferret to ferret to prod it to adapt. After 10 iterations, Fouchier's team had a virus that transmitted well thanks to five mutations.

When Fouchier's paper arrived at Science, “it was obvious” that it needed special review, says Editor-in-Chief Bruce Alberts (who did not discuss the paper's content with Science's reporters). The journal quickly recruited outside specialists, including biosecurity experts who serve on NSABB.

NSABB itself was first alerted to the studies by NIAID staff late last summer and received copies of the papers in mid-October. The problems were immediately obvious, Keim says. A working group of eight voting and a dozen ex officio members—including White House and NIAID staffers—started discussing the papers in teleconferences. “It was wham-wham-wham, three meetings the first week,” Keim recalls. In November, the working group made its recommendations to the full 25-member NSABB. Several more “very, very serious discussions” followed, says NSABB member Michael Imperiale, a virologist at the University of Michigan, Ann Arbor. “We realized that the stakes were very high for the life sciences.”

“There is no perfect solution. There is not even a good solution.”



NSABB had been there before. In 2005, it had approved a controversial paper about the resurrection of the H1N1 flu strain that caused the 1918 pandemic (Science, 7 October 2005, p. 77). “That was a very dangerous virus, too,” Keim says, but for various reasons, the H5N1 work seemed even more problematic. “Slowly, you get to the line where something shouldn't be communicated,” he says. “These papers exceeded that line in our minds.” In the end, the board was unanimous: Key details had to go.

On 30 November, NSABB members conveyed that message to the scientists and journals. The board, however, has no regulatory authority; it only makes recommendations. And NSABB's advice was so general that “it took us a while to figure out exactly what they wanted,” Alberts says. By late December, all sides could announce a deal: The journals would publish an incomplete description of the studies—if the government delivered a “transparent plan” for sharing the details.

Kawaoka and Fouchier went along reluctantly. And many of their colleagues say the deal is antithetical to the scientific spirit and unnecessary. “I think it's an overreaction,” García-Sastre says. Only a small group of highly skilled virologists could make use of the mutations, he says, and the well-known passaging method is “not an ‘Aha’ idea.” Peter Palese, another flu scientist at Mount Sinai, isn't convinced that ferret studies can even predict a flu strain's behavior in humans. Moreover, highly pathogenic H5N1 has been around for a long time, he writes in an e-mail: “If it would be so easy to make highly transmissible H5N1 viruses, nature would have done it already.”

But a senior flu scientist recruited by NSABB as a special adviser says the panel did the right thing. Robert Webster, 79, of St. Jude Children's Research Hospital in Memphis, Tennessee, is the unofficial dean of flu science and an expert on H5N1. He believes the studies were justifiable and publishing their main message is essential—which is that H5N1 is a global human threat and “not just a problem for chickens.” At the same time, “I don't think science should be putting out recipes” for bioterrorists, Webster says.

Now, developing a system to distribute the hidden details is just one of the “very difficult tasks” facing U.S. officials, Fauci says. Fouchier has estimated that perhaps 1000 researchers at 100 institutes should be allowed to see his work, for example. But the government has few legal means of restricting the flow of information other than outright classification, Fauci notes, and “NIH does not do classified research.” Alberts, however, is optimistic that officials will find some “credible method,” perhaps a secure Web site or marked paper copies that could be distributed via embassies.

It won't work forever, Imperiale predicts. “This information will eventually get out one way or another. We aren't deluding ourselves.” Even slowing the flow, however, will provide valuable time to develop countermeasures, Imperiale says.

Fauci is also part of a government-wide group that is fiercely haggling over new procedures for spotting “dual-use research of concern” before it starts. NIAID is already reviewing funded projects, he says, to make sure there are no more “surprises.” But what to do about future proposals isn't clear.

One likely option is expanding a questionnaire-based screening system already in place at a few major universities and for government researchers within the U.S. National Institutes of Health (NIH) and the Centers for Disease Control and Prevention. It asks researchers to check a box if their study might “enhance the harmful consequences of a biological agent or toxin,” such as by increasing transmissibility. The form is then reviewed by an Institutional Biosafety Committee (IBC)—bodies set up after Asilomar to oversee DNA research—which can recommend changes.

Many researchers and university officials worry that most IBCs—there are some 700 in the United States—don't have the necessary expertise and that broadly worded questionnaires could hinder research. Studies, however, suggest that few projects would get flagged. When NIH screened 3444 intramural DNA research projects from 2009, for instance, it found just two that merited special review. “There is no excuse for not doing screening,” argues Richard Ebright, a biologist at Rutgers University in Piscataway, New Jersey, and the Howard Hughes Medical Institute. The two flu studies, he says, “would have been clearly identified as problematic much earlier if this process had been followed.”

Clamping down in the United States will have little effect elsewhere. Some biosecurity advocates want a global panel to review studies that could result in a pathogen with pandemic potential; it could be similar to a WHO smallpox research panel that can reject proposed experiments. At the least, an international debate should occur, says Keim, who believes that in the meantime, scientists should neither publish the results of H5N1 transmissibility studies nor present them at meetings. That idea is getting mixed reviews. “I don't think that's in the best interest of science,” says virologist Andrew Pekosz of Johns Hopkins University in Baltimore, Maryland. A moratorium could “handicap” young researchers building their careers, he notes.

WHO officials, meanwhile, said in a 30 December statement that they are “deeply concerned about the potential negative consequences” of the controversy on global efforts to share information about H5N1. After the virus started spreading rapidly in 2003, some nations complained about WHO's system for sharing viral samples and research benefits: Indonesia even suspended participation. To ease tensions, WHO brokered a new agreement, released in May 2011. Redacting the flu papers could violate that agreement—a particularly sensitive issue because WHO says it provided the viral samples used by Fouchier and Kawaoka.

Meanwhile, some scientists worry that the focus on bioterrorism has diverted attention from a more likely risk: a lab accident that could unintentionally release a newly created viral strain. Sloppy lab practices in Asia, they note, triggered three minioutbreaks of the virus that causes SARS. Escape of an engineered H5N1 virus, Ebright says, would “really raise the question of whether there are some studies that aren't worth the risk.”

View Abstract

Stay Connected to Science

Navigate This Article