A Framework for Tomorrow’s Pathogen Research

Final Report



Ravindra Gupta

Ameenah Gurib-Fakim

Shahid Jameel

David Relman

Jesse Bloom

Filippa Lentzos

February 2024

Research on biorisk management to make research with pandemic risk more safe, secure, and responsible

There are significant differences across laboratories and countries in the measures adopted to manage biorisks (Koblentz et al. 2023). Limited data exist to support whether these differences result in a measurable improvement in safety and security or whether resources are being wasted on unnecessary and costly equipment (Ritterson and Casagrande 2017). Historically, biosafety and biosecurity improvements have always added on to existing equipment, procedures, or administration because there were no data suggesting which specific improvements were particularly effective.

As data to inform biorisk management are lacking, the frequency and consequences of accidents are unknown; well-informed key decisions cannot be made in the absence of adequate evidence. Similarly, it is difficult to understand the means by which outsiders are most likely to gain access to a laboratory or how they could misuse pathogens. If robust data were available, stakeholders could identify which biorisk management measures were truly worth the investment, enabling stakeholders to spend only what is needed on safety and security and the rest on research.

Research on biorisk management is urgently needed to improve efforts at eliminating and mitigating associated risks (Palmer, Fukuyama, and Relman 2015). Such research could generate useful knowledge to (1) prevent laboratory accidents and mistakes (as the research community generates real data on which practices are safe, which are risky, and under which conditions) and (2) reduce the chance that malicious actors can access known and potential pandemic pathogens. Research data could inform changes to a protocol, policies for access control, the movement of equipment within a laboratory, the training received by key personnel, or a redesign of a risky experimental approach. Biosafety and biosecurity studies could help inform where new laboratories of various types should be built.

Naturally, proper biosecurity and biosafety precautions entail more than sound laboratory infrastructure and practices. Context matters. What works in one setting or country may not fit with practices or available resources elsewhere and running simulations, while valuable, may not consider the full spectrum of possibility regarding safety and security risks. Psychosocial and behavioral research may also shed important light on how different actors (e.g., laboratory workers) interact with laboratory infrastructure and respond to governance structures and policies.

Research to improve biosafety management

The WHO’s Laboratory Biosafety Manual informs best practices for safely handling biological agents in laboratories and covers a range of topics, including personal protective equipment, biosafety cabinets, risk assessment, decontamination, and waste management (WHO 2020). Originally published in 1983, it is now in its fourth edition. The Laboratory Biosafety Manual is viewed by the scientific and practitioner community as the gold standard for biosafety and it is in wide use.

Nonetheless, accidents occur and there is considerable benefit in better understanding the causes. For example, recently published research demonstrates how frequently snap-cap microcentrifuge tubes, which are commonly used to store and mix biological samples, splash when opened (Wyneken et al. 2023a). The frequency of splashes from these tubes no matter how they are opened suggests that laboratories should substitute these or take additional measures to reduce splashing and immediately implement these solutions to reduce risk. A laboratory simulator in which researchers are observed manipulating small volumes of fluid and running mock assays could be used to compile data on the frequency of spills, splashes, and accidents. Researchers are completing the first studies of this type (Wyneken et al. 2023b; Wyneken et al. 2023a), which may begin to answer key questions such as: How often do researchers spill? What factors (e.g., training and experience) reduce this? Importantly, this initial research has demonstrated that studies done by volunteers in mock laboratories replicate similar accident rates in real clinical laboratories that were conducting blinded studies of error. A critical finding of this research is that even experienced laboratory researchers often do not know when or where a spill occurred. This underscores the often-repeated advice of biosafety professionals to decontaminate the entire workspace after every experiment, not just after a spill.

Additionally, researchers are collecting data on the frequency, size, and pattern of contamination of the biological laboratory worker; this critical first step will guide studies on how best to reduce risk from contamination. Data generated by biosafety research can also boost compliance with safer but inconvenient practices.

More generally, basic data are lacking for how researchers in laboratories are exposed to infectious material through spills, splashes, and contamination. Unlike in other industries, in which mechanical failures alone can cause catastrophe, in biological laboratories, researchers initiate or exacerbate most accidents; for example, researchers may spill infectious material and/or respond inappropriately by violating quarantine or contaminating themselves during cleanup.

In clinical settings, accidental infections often occur when protective gear (e.g., gloves, masks, coats, etc.) is removed after sterile procedures (Mumma et al. 2018). It is suspected, but unproven, that many instances of infection or contamination in laboratory settings happen in a similar manner. Studies that document how frequently laboratory researchers contaminate their hands when taking off gloves (or breach their gloves during research) could improve practices and procedures. For examples, studies that compare the use of a single pair of gloves with the use of two overlapping pairs of gloves could demonstrate the effectiveness of one or the other strategy to either solidify or negate the use of two overlapping pairs of gloves. Similarly, studies on when respiratory protection should be worn and what type of protection is needed under different conditions could usefully guide practices and procedures. For example, a variety of routine procedures, including centrifugation and flow cytometry or cell sorting, can generate and expose laboratory workers to aerosols if the proper containment is not used or if the device is not confined to a biosafety hood. In general, if protective gear works well in most situations but not when careless or inexperienced researchers wear and/or remove it, additional investments in training and oversight/proficiency testing would be warranted.

Research to improve biosecurity management

Failures in biosecurity can occur when leadership is inadequate, oversight institutions do not have the needed expertise or proper means to assess their effectiveness, or organizational structures and risk management processes are slow to recognize consequential advances in science and technology (Palmer, Fukuyama, and Relman 2015).

Biosecurity is challenging to investigate empirically, but observational research and controlled studies can be very useful. Observational studies in training laboratories can measure the frequency of similar failures (e.g., unauthorized access, failure to report worrisome behavior, or database security glitches). Controlled studies can also measure both the rate of non-compliance with a rule and the rate at which researchers hide their non-compliance. Without a significant effort, studies could gather more biosecurity data generated in the day-to-day conduct of research or training.

A research agenda should also seek to examine the extent to which releasing information about research with potential pandemic pathogens may create so-called “information hazards (Relman 2014).” Malicious actors may misuse published information and therefore researchers, funders, and journals should consider whether information controls are appropriate.

In summary: There is an evident need to improve current efforts at eliminating and mitigating biosafety and biosecurity risks.

Researchers and their institutions, as well as funders and governments, should fund studies that will provide robust empirical evidence about the nature of biosafety and biosecurity challenges and the effectiveness of potential mitigation strategies. Such data would enable more effective risk reduction practices. Biorisk management data could inform harm–benefit studies to determine exactly how laboratories working with known and potential pandemic pathogens, including in research with pandemic risks, should be organized and managed without unnecessarily diverting funding that could be invested in the research itself. Not only will this improve biorisk management, but using new evidence to eliminate wasteful measures would make laboratories more efficient and sustainable. At the same time, biorisk management practices should be reviewed to eliminate those without evidence of added value or to replace more burdensome practices with less burdensome ones.

Next section ⇢