The U.S. biosecurity panel that recently lifted its objections to the publication of controversial bird flu studies has raised additional concerns about one of the papers, work conducted by a Dutch research team.

In its report to the U.S. government, posted online over the weekend, the National Science Advisory Board for Biosecurity waived its earlier objections to the study led by Yoshihiro Kawaoka at the University of Wisconsin-Madison.

But the board said the second study, by Ron Fouchier of Erasmus Medical Centre in Rotterdam, needs clarifications before it can be published and suggested additional information that was not in the study the board vetted should not be added to the final document.

"Importantly, the Board also noted that additional information that would enable the construction of an H5N1 virus that is both highly pathogenic and transmissible between mammals through the air should not be included in the manuscript," it stated.

The lingering controversy over the work began late last fall when the NSABB advised the U.S. government to ask the journals planning to publish the studies to withhold key portions so others could not repeat the work. The U.S. government followed the advice and the journals and authors reluctantly agreed, on the proviso a system be set up to allow the withheld information to be shared with public health authorities and other flu researchers on a need-to-know basis.

The studies show how H5N1 flu viruses can be mutated to the point where they can transmit through the air from ferret to ferret, mammals that are used as a proxy for humans in influenza research.

The report, which goes to U.S. Health Secretary Kathleen Sebelius, was posted online Saturday by the National Institutes of Health, the body under which the NSABB operates. The NSABB was set up in the wake of the 2001 anthrax attacks in the United States. Its mission is to advise on so-called dual-use research of concern -- legitimate research that could also pose a public health threat in the wrong hands.

At least several members of the 21-member board were caught off-guard by the report's release, saying they weren't informed and did not have a chance to approve the final wording of the document. They asked not to be identified.

It had been thought the report would only be made public after Sebelius announces whether she is going to accept the board's recommendations.

The report was released the day after publication of a leaked letter from one member of the NSABB slamming the meeting at which the final decision was reached. The news departments of the journals Science and Nature received the letter via anonymous email.

It was written by Michael Osterholm and addressed to Dr. Amy Patterson, associate director for science policy at the National Institutes of Health, and the official to whom the NSABB reports. It was copied to members of the 21-person board and to NSABB staff.

The letter was sharply critical of a meeting held March 29 and 30, called so the NSABB could reconsider modified versions of the two studies. Osterholm, a flu expert and director of the Center for Infectious Diseases Research and Policy at the University of Minnesota, said the information presented at the late March meeting was "one-sided" and designed to produce the eventual outcome.

For instance, he noted the board was told of the importance of the information for surveillance of H5N1 viruses that might be evolving towards person-to-person transmissibility in the wild. But it did not receive briefings on the state of existing surveillance efforts and the likelihood mutating viruses could be picked up in real time. Experts have suggested surveillance is too spotty and too slow to identify evolving viruses in time to take action.

"The agenda was not designed to promote a balanced reconsideration of the manuscripts," he wrote. Osterholm has refused to comment on the letter, saying only that it speaks for itself.

Reached Sunday for comment, NSABB member Dr. David Relman said the issues raised in the letter "are legitimate and deserving of serious attention."

Relman is a researcher in the departments of medicine, and microbiology and immunology at Stanford University. He and Osterholm were among six members of the board that voted against full publication of Fouchier's study.

In his letter, Osterholm warned that the decision to recommend full publication of the two studies merely "kicked the can down the road" towards what may be another imminent standoff with Fouchier, revealing that the Dutch scientist disclosed at the March meeting that he has found a mutation that makes generating a mammal-to-mammal transmissible H5N1 virus significantly easier.

"This work ... surely must be considered as a candidate for the next manuscript to be before the NSABB for review," said Osterholm, whose term on the board is coming to an end.

The NSABB's report did not specifically state that this is the addition information Fouchier should not put into his paper, which is to be published by Science.

But the report said that information on how to produce an H5N1 virus that was both highly pathogenic and transmissible through the air "could conceivably be directly misused to threaten public health or national security and additional considerations regarding communication would be necessary."

The report contained a section outlining the views of the minority members on the vote regarding Fouchier's work. Where the majority said the information in the studies couldn't be misused over the near-term, the six dissenting members said the details of Fouchier's study could be used to endanger public health or national security. They also questioned whether the information it contains will help in surveillance for H5N1 viruses that are mutating towards human-to-human spread in the wild.

"The evolutionary paths taken by naturally occurring H5N1 viruses may not be similar to those selected under these laboratory conditions," they said, adding that "excessive attention to these mutations might distract attention from others that are of greater significance in the wild."

The full report said that the board still believes the manuscripts present dual-use research concerns, but members had been persuaded the risks of not publishing were also serious.

"The Board's discussions underscored the risks associated with not sharing the information, which could jeopardize pandemic influenza preparedness efforts. Specifically, there was concern that the United States would be perceived as redacting information with potential public health benefits and that this could undermine valuable international collaborations," the report said.

In its recommendations, the report called for the U.S. to work with other governments to devise policies for oversight of dual-use research, particularly "gain-of-function" research aimed at giving a pathogen increased virulence or the ability to infect new hosts.

And it also called on the U.S. to quickly develop a mechanism whereby scientific information that needs to be kept out of the literature for security reasons can be shared on a need-to-know basis.

Currently export control laws are triggered if information must be withheld for security purposes. As this controversy has dragged on, it has become clear that those laws would have made it impossible to share the data in the Kawaoka and Fouchier papers, if they were redacted. Following the March meeting one member of the NSABB said the board would have preferred to hold to the recommendation that the studies be redacted, but had come to the conclusion that path wasn't feasible at this time.