Within a matter of weeks in early 2003, severe acute respiratory syndrome (SARS) spread from the Guangdong province of China to rapidly infect individuals in some 37 countries around the world. The brief epidemic infected more than 8,000 people and killed nearly 800, almost 10% of those infected. Fortunately, timely public-health actions, such as isolation of victims, stopped SARS before it became a world-wide epidemic. It is gone from nature, at least for now. We dodged a world-wide epidemic.
But SARS lives on, imprisoned in BSL3 and BSL4 laboratories around the world, and has already escaped more than once through infected lab workers. If we experience another SARS epidemic, many scientists feel that it will have started from a laboratory escape.
An escape of a highly contagious pathogen from a lab in a city is more likely to seed an epidemic; and this time, we may not be able to dodge it. What if a lab researcher is infected with a highly contagious deadly disease that is transmitted by casual contact, a victim's cough or from contaminated surfaces? Besides SARS, the 1918 pandemic flu also comes to mind. The Boston University National Emerging Infectious Disease Laboratory likely will research SARS and the 1918 pandemic flu.
Under the reasonable assumption that employees tend to live near where they work, the probability of an epidemic would be far greater from an infected employee living and working in or near Boston. For instance, the number of casual contacts with strangers will be sizeable for an infected researcher taking public transportation, a likely way to commute. Transmission of infection to others, called secondary infections, would be almost impossible to trace. For a laboratory located in the suburbs or rural area, most employees would drive, so their daily casual contacts with strangers would be fewer, and there is at least some chance of tracing others exposed.
Tetra Tech, a world-wide consulting firm, was hired by the National Institutes of Health to carry out yet another risk analysis for the BU NEIDL. The analysis is in progress. This is BU's third attempt at a believable risk analysis for their lab. It is worthwhile to review the history of risk assessments for the BU lab.
Attempt number one: This 2004 risk analysis considered only one accident scenario and only one pathogen, a small anthrax spill in the laboratory. A number of lawsuits brought by residents of Roxbury, the largely African-American Boston neighborhood and the site of the NEIDL, challenged the certification of the risk analysis by the Massachusetts' Executive Office of Environmental Affairs (EOEA). In 2006, a Superior Court judge deemed the certification "arbitrary and capricious" and ordered the BSL4 laboratory not to open until an acceptable risk analysis was carried out.
Attempt number two: The NIH then set out on its own risk analysis, which it unveiled in 2007. The analysis was comprised of fifteen pdf files-with dozens of photographs, drawings, graphs and statistics-that gave the impression of a precise and effort-laden risk analysis. But it takes only a quick scan of the many files to realize that the whole analysis was set up to give the answer BU wanted, namely that the inner city Roxbury location for the lab was acceptable, not only acceptable but the safest location for the laboratory.
Here is how NIH reached this suspect conclusion: As ordered by the Superior Court, they did consider alternative sites in Massachusetts besides densely populated Roxbury, namely Tyngsborough (a BU-owned suburban site) and Peterborough (a BU-owned rural site). The deadly and contagious viruses chosen for the analysis were ebola, sabia, monkeypox and Rift Valley fever. These choices addressed another criticism of the first risk analysis that no contagious pathogens were considered. (Anthrax is not contagious.) So far, all fine and good.
These are all "exotic" viruses (to use the NIH terminology) but represent no present or likely future public-health threat to the United States. The only scenario that NIH chose to analyze was a single researcher infected at work with one of the four exotic viruses, and then bringing his/her infection home. The viruses are only mildly contagious, so intimate contact is required for transmission. Symptoms appear quickly so patients can be diagnosed and precautions taken, unlike the HIV/AIDS virus that can silently infect large numbers of victims without anyone being aware.
Since the viruses would likely infect only family members, health care providers and others in intimate contact with the initial victim, the number of secondary infections would be similar no matter where the victim lives or works, so population density at the victim's home or laboratory workplace would not much matter. NIH asked only the question that would give them the answer they wanted. Location is nothing, to reverse the common real-estate mantra.
But how does urban Roxbury become the safest site? The answer lies in Rift Valley fever virus, which is transmitted by mosquitos from cattle and other farm animals to humans. Even though many of Boston's streets were laid out by cattle in colonial times, there are no cattle there now. So Rift Valley fever virus would infect more people in suburban or rural settings where farm animals live. Urban Roxbury then becomes the safest place for the NEIDL compared to suburban and urban locations.
Between the first and second risk-analysis attempts, Massachusetts changed governors from Mitt Romney to Deval Patrick and the management of EOEA changed as well. The new EOEA folks made a smart move. They realized that they did not have the background to understand the issues, so they asked the National Academy of Sciences to appoint a National Research Council (NRC) committee of experts to critique the risk analysis. The NRC committee delivered their detailed critique in late 2007, concluding that the NIH analysis was not sound and credible, that the worst case scenarios had not been adequately identified, and that the information underlying the alternative site analysis was insufficient or inappropriate. The critique also questioned the infectious agents selected.
Now back to the in-progress Tetra Tech risk analysis. Tetra Tech presented its preliminary results to local residents at the Roxbury Community College in October 2010. Among its findings was that a secondary infection of SARS to someone outside the lab from a lab researcher would occur once in 10,000 years in a worst-case scenario, and likely only once in over a million years. Tetra Tech looked at only two scenarios, a centrifuge accident and a massive earthquake that would level the laboratory. They did not look at the risk of a SARS-infected lab worker, unaware he/she was infected, transmitting the infection to someone outside the laboratory.
The NRC committee commenting on the Tetra Tech preliminary work concluded "at this point in time it cannot endorse the illustrative analyses presented as scientifically and technically sound or likely to lead to a thorough analysis of the public health concerns previously raised by the NRC." The committee also noted "Consideration of the available case studies (such as the SARS case described below) suggests the possibility that transfer of a pathogen outside the laboratory by an infected worker is an important class of risk events."
There have been at least three SARS escapes from laboratories through infected lab workers. The incident from the NRC 2010 document quoted below warns of the danger of a future SARS escape in a densely populated area:
"In China, SARS/CoV was grown in a BSL-3 laboratory by a worker who apparently had worn inappropriate personal protective equipment (PPE) and then treated the sample to inactivate the virus before removing it to a BSL-1 laboratory for further work on the open bench. The worker failed to verify the complete inactivation of the virus and subsequently became ill and was admitted to a fever hospital. The laboratory was not notified of this development and the worker later returned to the laboratory. A second worker who handled the "inactivated" sample also became ill. A graduate student who observed the laboratory procedure later traveled by train to her home several hundred miles away. After returning to the laboratory she became ill and once again traveled to her home by train where her mother, a physician, admitted her to a hospital and treated her. The student was asked if she worked with SARS/CoV (she said no because her research involved another virus). It was not until the mother became ill and died that SARS/CoV was identified. Other laboratory workers also became ill and other hospital personnel died. This case study illustrates several important points: people make mistakes (improper PPE); not everyone follows procedures (failure to test sample for inactivity); people may die if not properly diagnosed and treated."1
Another message from this story is that research on deadly, highly contagious pathogens should be conducted only in BSL4 laboratories in isolated locations where extra precautions in addition to location are available, never in a populated area since SARS has also escaped from a BSL4 laboratory.
How many bites of the apple will Boston University have before they realize that a BSL4 Laboratory in densely populated Boston is a bad and dangerous idea? When will the city or state step up and say "No! No BSL4 lab in Boston," following Cambridge's lead?
Lynn C. Klotz is co-author of Breeding Bio Insecurity: How U.S. Biodefense Is Exporting Fear, Globalizing Risk, and Making Us All Less Secure. He is working with scientists and Roxbury residents to propose an alternative vision for the Boston University labs.