^{1}

^{2}

^{1}

^{2}

Edited by: Kathleen Vogel, Cornell University, USA

Reviewed by: Kathleen Vogel, Cornell University, USA; Simon Wain-Hobson, Institut Pasteur, France

This article was submitted to Infectious Diseases, a section of the journal Frontiers in Public Health.

This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

In letters to the journals Science and Nature (

Now is the time to address the next critical question: what is the likelihood that one of these viruses will escape from a lab and seed the very pandemic the researchers claim they are trying to prevent? As we shall estimate, that probability could be as high as 27%, a risk too dangerous to live with.

First, from the calculations in two in-depth pandemic risk analyses (_{0}, in standard epidemiology notation), varying the details of commutes to and from work on public transportation, and whether infected acquaintances are quarantined before spreading infection. The Merler (2013) study, based on a computer-generated population grid of size and varying density of the Netherlands, supports our concern over a lab escape not being detected until it is too late: “there is a non-negligible probability (5–15%), strongly dependent on reproduction number and probability of developing clinical symptoms, that the escape event is not detected at all.”

Different methodologies were used in the Klotz (2014) and Merler (2013) risk analyses. Additional analyses are needed using other methodologies, such as the mathematical model employed for SARS (

Given such a dire predicted outcome by the existing studies, the critical question is: what is the probability that a worker acquires an undetected infection in the lab in the first place? To answer this question, we reproduce here one part of the Klotz (2014) analysis: the probability of an escape through an LAI from at least one of the many labs expected to be involved in this research enterprise.

A 2013 Centers for Disease Control report is a significant source of recent data on LAIs (

Thus, the probability of escape for a single year, _{1}, can only be calculated as 4 LAIs/2,044 lab years = 0.002 or 0.2% per lab per year. This is clearly an underestimate since BSL-2 and BSL-4 labs contribute to the denominator. (The denominator used here, 2,004, equals the number of BSL-2 plus number of BSL-3 plus number of BSL-4 labs. But the denominator in our calculation should be just the number of BSL-3 labs, so the denominator is overestimated and the percent escape is then underestimated. Although requested, the CDC has not supplied us with the number of BSL-3 labs for us to do the exact calculation.) This basic probability is consistent with that for SARS escapes in Asia through LAIs (

To illustrate potential risk, the probability of no escape from a single lab in a single year is (1 − _{1}), so

Given the Science and Nature articles listed above (

We noted above that the probability _{1} = 0.2% is conservative, estimated from the CDC data alone. The first Department of Homeland Security risk assessment for the planned National Bio- and Agro-Defense Facility in Manhattan, Kansas estimated a significantly higher escape risk, over 70% likelihood for the 50-year life of the facility (_{1} = 2.4% per year. The National Research Council (

With this higher number, which we take as a worst-case scenario, the likelihood of at least one escape from 10 labs in 10 years becomes 91%, almost a certainty. It follows that, if the likelihood of one LAI leading to a pandemic is 30% in the worst-case scenario, the likelihood of an LAI-caused pandemic resulting from this whole research enterprise could be as high as 30 × 91% = 27%, a likelihood that is too dangerous to live with, as we noted. While this represents a worst-case scenario, it is not improbable.

Recent self-reported mistakes at the CDC (

Our concern is shared by many virologists and epidemiologists. A recent letter to the President of the European Commission (

The risk of a man-made pandemic from a lab escape is not hypothetical. Lab escapes of high-consequence pathogens resulting in transmission beyond lab personnel have occurred (

Do benefits outweigh risks? Those who support PPP experiments either believe the probability of PPP escape is infinitesimal or the benefits in preventing a pandemic are great enough to justify the risk. In making decisions for what lines of research will lead to new knowledge, experts must rely on intuition honed by years of research in a particular field. In the case of this PPP research, in our opinion it would take extraordinary benefits and significant reduction of risk via extraordinary biosafety measures to correct such a massive overbalance of highly uncertain benefits to too-likely risks (Wain-Hobson, 2013).

Whatever number we are gambling with, it is clearly far too high a risk to human lives. This Asian bird flu virus research to develop strains transmissible via aerosols among mammals, and perhaps some other PPP research as well, should for the present be banned. We must emphasize that we have been considering only a very small subset of pathogen research. Most pathogen research should proceed unimpeded by unnecessary regulations.

Special precautions in BSL-4 laboratories for work with PPPs should be adopted (

Training a full-time technical staff for work with PPPs. Experiments could be directed by scientists outside the laboratory using modern audio-video technology.

Requiring the staff to follow up extended work shifts with periods of quarantine before they leave the containment area to assure that no PPP escapes from the containment area through an LAI.

Restricting these PPP laboratories to remote locations, where an aerosol escape or other containment failure would pose the least risk of infecting an outside community.

We label BSL-4 laboratories with the special precautions, BSL-4+. While PPP experiments would be carried out primarily under BSL-4+ containment, BSL-3 containment with the special precautions might suffice for some work.

Given the global threat, the international community should insist on discussions leading to an international agreement that would require the strictest oversight to conduct this particular research anywhere. To place responsibility with the international community where it belongs and to provide maximum transparency, policy makers should require that international inspectors have access to facilities at any time on short notice.

As it stands, there is no proactive oversight nor regulations for this PPP research, so any and all of the world’s nations can carry out this dangerous work without regard to consequences. But consequences would be shared by all of us. In the meantime, insurance companies who routinely provide insurance for biological research should consider excluding such risky research from coverage.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.