Dangerous pathogens in the laboratory: from smallpox to today's SARS setbacks and tomorrow's polio-free world

Less than a year after an unprecedented international public-health effort interrupted human-to-human transmission of the coronavirus that causes severe acute respiratory syndrome (SARS-CoV), some human beings are again infected. SARS-CoV does not seem to have re-entered the human population from the exotic wild-animal markets of China that have preoccupied public-health officials worldwide, nor from some other source in nature not yet understood. Rather, the latest outbreak seems to be from a laboratory source.¹

This scenario is reminiscent of the often forgotten footnote to the smallpox eradication effort when the last human infections did not occur in Somalia, the last country with naturally occurring smallpox, but a year later in Birmingham, in the UK, originating from a laboratory with inadequate biosafety facilities.² Auspiciously, the new SARS cases are occurring as WHO's Biosafety Advisory Group prepares to examine the long-term containment of poliovirus stocks, the risks of which will rapidly increase after interruption of transmission and the ending of immunisation with oral poliovirus vaccine.³

The recent outbreak of nine cases of SARS in China, with one death, underlines again the challenges of maintaining appropriate biosafety conditions in laboratories working with dangerous pathogens. In this outbreak, two researchers at the Institute for Viral Disease Control and Prevention in Beijing developed SARS in late March and mid-April.⁴ All subsequent second-generation and third-generation cases have now been linked to close contact with one of these researchers. Investigation of the source of the outbreak, jointly by the Chinese Ministry of Health and WHO, continues to focus on this virology institute.¹

If the laboratory source is confirmed in China, this will be the third known incident of laboratory-acquired SARS-CoV infection, the first having occurred at the National University of Singapore where a postgraduate developed an illness consistent with SARS in late August, 2003.⁵ During the investigation that followed, it was concluded that the student most probably acquired the infection in the BSL-3 laboratory in which he was studying the West Nile virus 3·5 days before the onset of his illness, which is consistent with the expected incubation period of SARS. (Biosafety conditions are described as biosafety levels in four categories [BSL 1–4], with BSL-3 and BSL-4 recommended for work with pathogens that cause serious human and animal disease). It seems that transmission occurred as a result of inappropriate laboratory procedures that led to cross-contamination of the West Nile virus specimen with SARS-CoV. No other workers in the laboratory, and none of the medical staff who cared for the student while he was ill, became secondarily infected, nor did household and other contacts.

The second reported incident similarly resulted in an isolated case of SARS in early December, 2003. It occurred at the Institute of Preventive Medicine, National Defence University, Taipei, in a laboratory scientist who had been working intensely in a BSL-4 laboratory, over a long period and for long hours each day.⁶ It seems that transmission occurred after exposure to SARS-CoV from contact with droplets when cleaning the spill of a SARS-containing specimen in the laboratory's transport chamber.

Accidental transmission of a dangerous pathogen from a laboratory can occur when a susceptible and unprotected laboratory worker is exposed to the agent during laboratory procedures. These conditions were met in the smallpox laboratory in Birmingham in 1979, and in at least two of the laboratories associated with the recent cases of SARS during 2003. If the resulting human infection causes viral shedding, with exposure to susceptible workers in the laboratory, health-care system, or community, an outbreak can result. In the outbreak in Brimingham after the smallpox laboratory accident, infection spread from the initial case to a close family member and one other. In the current outbreak of SARS, chains of transmission seem to have moved from the laboratory, to a close family member, and to a hospital, from where a nurse who treated the laboratory worker then transmitted infection to five others.

Proven measures to minimise the risk of reintroducing dangerous pathogens include: limiting the number of sites where they are stored and studied to those that are absolutely necessary; protection of laboratory workers with available vaccines, protective clothing, and safe equipment; closely monitoring illnesses in laboratory workers; and adhering to standard operating procedures. Hundreds of years of combined experience in high and maximum containment laboratories have proven these biosafety measures effective if rigorously and faithfully followed—with strict national procedures to verify that appropriate conditions and procedures are maintained.

After certification of smallpox eradication, known stocks of variola virus were destroyed or transferred to one of two WHO reference laboratories where biosafety is periodically verified by the WHO Biosafety Advisory Group. During the SARS outbreak last year, many specimens were obtained from human cases of SARS and sent to many different national and international laboratories for various studies. In April, 2003, WHO provided guidelines for handling, packing, and shipping SARS specimens, and listed laboratory practices that could safely be done under BSL-2 and those that required BSL-3.⁷ These guidelines were reviewed and updated during later WHO consultations, and laboratory research activities continue at many of these sites.⁸ Unfortunately, adherence to these guidelines has now failed at two, and possibly three, different laboratories, reaffirming the importance of strong national, and possibly international, monitoring of their implementation. The predictable emergence of new dangerous pathogens in the future further highlights the need for such action.

If activities to eradicate poliomyelitis remain on target, interruption of human-to-human transmission will occur sometime during the next 18 months, and the wild poliovirus will be moved from the list of endemic infections to that of dangerous pathogens. That poliovirus reintroduction could occur was seen in 1992 when a reference strain of wild-poliovirus type I that is used in the production of inactivated poliovirus vaccine was isolated from a young child being investigated for diarrhoea.⁹ The subsequent epidemiological investigation found that the child's father was employed at a production site for inactivated poliovirus vaccine where an accident had occurred. Fortunately the child was fully immunised against poliovirus. But the child served as a healthy carrier of poliovirus to the community, although sanitation was adequate and poliovirus vaccination coverage was high enough to prevent an outbreak. A similar poliovirus reintroduction from a laboratory or production facility for poliovirus vaccine to one of the many countries that have indicated their intention to stop poliovirus immunisation after certification of global eradication could result in future outbreaks of poliomyelitis.¹⁰

Recognising the risks that would be associated with a poliovirus reintroduction, WHO and its technical partners in poliomyelitis eradication began the process of establishing a global action-plan for the long-term laboratory containment of wild polioviruses in the mid-1990s.¹¹ By 1999, international consensus had been established and the process of surveying and inventorying laboratories for wild poliovirus and infectious or potentially infectious materials began in the three WHO regions that had interrupted indigenous transmission of wild poliovirus. Through the comprehensive surveys of national laboratories that are underway or completed in 152 countries in five continents, over 160 000 facilities have been inventoried to date. About 500 facilities have reported wild-type poliovirus or potential infectious materials. At the same time, technical and engineering solutions have allowed the continued production of inactivated poliovirus vaccine from wild poliovirus under enhanced BSL-3 conditions to ensure that such a vaccine is available to the countries that choose to continue routine immunisation against the poliovirus after eradication has been certified globally and the routine use of oral poliovirus vaccine stopped.¹²

The current WHO plan for the period after the global interruption of wild-poliovirus transmission specifies BSL-3 conditions for wild-poliovirus infectious materials and BSL-2 for potentially infectious materials in non-virology laboratories, on the basis of a now outdated assumption of continued universal immunisation.¹¹ In September, 2003, an expert group recommended that because circulating vaccine-derived polioviruses would, like wild poliovirus, compromise the goal of poliomyelitis eradication, the use of oral poliovirus vaccine for routine immunisation should eventually stop.13, 14 This decision, combined with the recognition that many countries plan to forego future routine immunisation with inactivated poliovirus vaccine, has led to the understanding that Sabin poliovirus strains will also need to be stored and handled under appropriate biosafety conditions. A biosafety strategy for minimising the risk of a laboratory accident with the Sabin polio-virus is under development. In addition, consensus is being sought on the mechanisms and procedures for ensuring that the necessary stockpiles of oral poliovirus vaccine are available should poliovirus be reintroduced into human populations because of a laboratory accident.

Although an increasing number of pathogens are referred to as dangerous, in reality different pathogens present different laboratory risks. SARS-CoV seems to represent a high laboratory risk. Unlike SARS-CoV, poliovirus is not efficiently transmitted by droplets from person to person, and a vaccine is available that fully protects laboratory workers from disease and reduces the risk of infection, thereby providing additional assurances against substantial consequences should a laboratory accident occur once routine immunisation with oral poliovirus vaccine has stopped. It is also reassuring that no further accidents have occurred with the smallpox virus stored in two reference laboratories for over 20 years. Nevertheless, the recent laboratory accidents with SARS-CoV are a stark reminder that the security of public-health achievements requires greater investment to ensure that global biosafety standards for dangerous pathogens in laboratories are universally adopted, strictly adhered to, closely monitored, and rigorously enforced.