Yoshihiro Kawaoka can now publish his flu study. Credit: Jeff Miller/University of Wisconsin-Madison

The dust is beginning to settle on the months-long controversy over two studies in which the H5N1 avian influenza virus was modified to be transmissible between mammals. But scientists and authorities still need to address the lack of international oversight for studies in which pathogens are deliberately made more dangerous, speakers emphasized at a two-day meeting held last week at the Royal Society in London. The meeting brought together scientists, research funders and experts in security, bioethics and foreign policy, just days after the US National Science Advisory Board for Biosecurity (NSABB) revised its earlier stance and recommended publication of the two studies.

In December 2011, the board recommended redaction of experimental details of the studies, on the basis of concerns about bioterrorism and the increased likelihood of accidental release of the viruses. But after considering revised versions of the manuscripts on 29–30 March, the NSABB voted unanimously in favour of full publication of the paper submitted to Nature by Yoshihiro Kawaoka of the University of Wisconsin–Madison, and 12–6 for publication of the content (although not the specific wording) of the paper submitted to Science by Ron Fouchier of the Erasmus Medical Center in Rotterdam, the Netherlands.

Kawaoka presented his findings for the first time at the Royal Society meeting (see Naturehttp://doi.org/hsn;2012), but Fouchier gave only a summary of his, saying that a more detailed description was prohibited by Dutch export-control laws, which require a permit to disseminate samples of, and information about, certain dangerous pathogens.

In December, the NSABB said that the information in the papers could help H5N1 surveillance efforts and so should be made available to experts on a need-to-know basis. A major factor in the board’s change of heart was that subsequent international discussions concluded that there was no practical way to selectively share the data, and that national export controls may restrict distribution anyway, says Michael Imperiale, a virologist at the University of Michigan in Ann Arbor and a member of the NSABB.

This left the board with the stark choice of publishing either all or none of the research, with publication becoming “the only way for public-health authorities to know what to look for”, says Imperiale, who this time voted in favour of publishing both papers.

Some on the board were also influenced by clarifications to both papers — particularly those suggesting that Fouchier’s viruses were not as pathogenic as they had initially seemed — and by presentations from Kawaoka and Fouchier indicating that some combinations of the mutations generated in their laboratories had already been seen in the wild. But Imperiale says that his vote was unaffected by the revisions. “I don’t think the risks have changed; the authors have changed the host range and transmission properties of a deadly virus,” he says. “I don’t think the benefits have changed either.”

Another factor in the NSABB’s decision was the announcement on 29 March of a new US policy requiring that all publicly funded research on certain pathogens be assessed from the outset by the funding agency for the risk that it may be misused (see Naturehttp://doi.org/hsp;2012). Experts have generally welcomed the guidelines, which they say should, if properly applied, also help to avert future repeats of the H5N1 controversy, in which the NSABB learned of the papers only shortly before their planned publication.

Arthur Caplan, a bioethicist at the Uni­versity of Pennsylvania in Philadelphia, was one of several speakers at the Royal Society meeting to argue that it would be a “mistake” to think that the issues raised by the papers are now fully resolved. Besides bioterrorism, a major concern is that publishing them will result in worldwide proliferation of similar research, possibly in labs that may not have well developed biosafety cultures and training. Both Fouchier and Kawaoka worked in facilities rated at ‘biosafety level 3 enhanced’, but to expect that all such research would be done carefully every­where is “utter malarkey”, Caplan told the meeting.

The World Health Organization has recommended that work to make H5N1 viruses more transmissible in mammals, which flu researchers voluntarily halted in January, remain suspended until the relevant authorities have assessed the safety conditions for such research (see Nature 482, 447–448; 2012). US and Dutch authorities are expected to release their verdicts within weeks, which Kawaoka said “would be the time to lift the moratorium”. But others warned that doing so before broader debate, including planned hearings by US lawmakers, would be premature and could be perceived as arrogant.

The challenge for any oversight system will be to avoid discouraging important science while ensuring that work is limited to labs with appropriate safety standards, says one scientist, who criticizes the reluctance of many flu researchers to admit that such studies carry risks. “You’d have to have your head in the sand not to accept that there are some risks here.”