In the wake of disclosures that top government labs mishandled anthrax, smallpox and avian flu, US health authorities are considering the once unthinkable: cutting the burgeoning number of labs working with the planet's most dangerous microbes.
When the US Centers for Disease Control and Prevention last week unveiled a report documenting multiple safety breaches at its labs, its director for the first time suggested the country turn back the rapid-fire proliferation of such research units, which have tripled in little more than a decade to at least 1,500.
"One of the things that we want to do is reduce the number of laboratories that work with dangerous agents to the absolute minimum necessary," said CDC Director Dr Thomas Frieden.
"Reduce the number of people who have access to those laboratories to the absolute minimum necessary. Reduce the number of dangerous pathogens we work with."
His remarks may vindicate the views of a small group of biosafety and biosecurity experts who see that as the only way to protect dangerous viruses and bacteria from both lab accidents and thefts.
They point to an alarming rise in the number of incidents of lost or escaped microbes from such labs in recent years and see the CDC cases as proof that even the best facilities are vulnerable.
Following through on this idea would require a wholesale shift in US biodefense policy, which spans preparedness for disease outbreaks and for the use of biological agents in terror attacks.
While the CDC and US Department of Agriculture are responsible for registering labs that work with "select agents" - microbes and poisons that could be used as bioweapons - they cannot rescind that approval unless there has been a clear violation of the rules for handling those microbes.
"Just as with domestic spying by the National Security Agency, drone attacks and a long list of other things, the White House seems to feel it must maintain the policies of the last administration or risk being called weak on homeland security," said molecular biologist Richard Ebright of Rutgers University.
Testing deadly powders
According to a 2013 report by the Government Accountability Office, the investigative arm of Congress, 415 labs had registered with the CDC or the USDA in 2004 to work with select agents. By 2010, the number had grown to 1,495, the GAO found.
That was intentional.
In 2001, anthrax stolen from a federal bioweapons lab killed five people and sickened 17 more. At the time, only two US labs were capable of identifying anthrax in samples of mysterious powders, which "were flowing in by the thousands," said epidemiologist D.A. Henderson, a distinguished scholar at the University of Pittsburgh Medical Center.
Henderson was tapped by the federal government to vastly increase the number of such labs, both to detect suspected pathogens like anthrax and to conduct biodefense research, such as developing vaccines.
"We had more white powder coming out of more places than you can possibly imagine," he said. "The number of powdered donuts that got subjected to testing, I'd hate to think."
As a result, "there was a rush to get more BSL-3 and BSL-4 facilities," he said, referring to the highest levels of biosafety. "Universities were anxious to build them," since the work brought millions of dollars in funding as well as prestige.
A decade later, the country had spent $19 billion on biodefense research. But there has been no national assessment of how many such labs are needed for security, the GAO found.
"Increasing the number of (such) laboratories," it concluded, "increases the aggregate national risk" because of the chances of intentional or accidental escape.
It also increases the number of individuals with federal approval to work with select agents. With the additional spending, the number of people with access to bioweapons agents also "increased by a factor of 20 to 40," said Ebright.
According to a 2012 report by CDC scientists, there were 16 incidents of lost or escaped microbes from select-agent labs in 2004, meaning everything from misplaced samples to an infected researcher walking out the door harboring a virus. That rose to 128 in 2008 and 269 in 2010.
"It is almost exactly two per week and accelerating," said epidemiologist Marc Lipsitch of Harvard School of Public Health, and suggests that staff training, physical measures, and other elements of biosafety are failing more often rather than less.
How many labs?
The CDC's anthrax breach, plus its mishandling of a highly pathogenic flu virus also revealed last week, show that even the most respected labs can violate protocols in potentially dangerous ways. It therefore makes no sense for the United States to have a dozen BSL-4 labs, which work with pathogens that are easily transmitted by air between people (anthrax is not contagious and so is handled in BSL-3 labs) and cause severe or fatal diseases for which there are no vaccines or treatments, said biologist Lynn Klotz, a member of the Scientists' Working Group on Chemical and Biological Weapons.
"At minimum, these labs should be in a remote, rural area," said Klotz. "That way, if there is a mechanical failure and something gets out, there is much less risk of harm." He and others say several high-security labs should be candidates for closing.
After years of litigation, Boston University received permission from city, state, and federal authorities this year to open a BSL-4 lab in the city's densely populated South End, but critics consider it a prime example of a facility whose potential contribution to research is swamped by the risks of exposing a large urban area to an escaped pathogen. Other labs targeted by both scientists and community members include the National Bio- and Agrodefense Facility in Manhattan, Kansas, partly because it is located in a tornado zone and will work with pathogens that could devastate the cattle industry.
Some experts argue that having fewer labs is not the solution, and that improving adherence to safety protocols makes more sense. To many, however, it is simple mathematics: the fewer labs working with nature's most dangerous microbes, the lower the probability of an escape.
"Reducing the number of labs may help better police the remaining ones," said Michael Osterholm of the University of Minnesota, a member of the National Scientific Advisory Board for Biosecurity, which advises the government.
The first chance to gauge any political support for that option may come this week, when CDC's Frieden testifies before a House subcommittee about the anthrax and flu releases. "I'm sure there will be many changes to come," said biosafety consultant Debra Sharpe. "I just hope they will be well thought out and not knee-jerk fixes for political expediency."