How should we manage the new threats posed by biomanufacturing?
03 September 2020
Over the last five decades, there have been thousands of incidents in biological laboratories in which pathogen containment has failed1. Those incidents have caused over 150 fatalities since 1970.
(Click here to view article in digital edition)
These days, most of the world is suffering a pandemic caused by SARS-CoV-2, a virus closely related to SARS-CoV, which was involved in recent loss of containment events (Singapore 2003, Taiwan 2003, Beijing 2004) in laboratory settings.2 In addition to the lives lost in the current pandemic, its economic impact alone is forecast by the International Monetary Fund (IMF) to exceed 6% of the Gross World Product, in excess of five trillion US dollars.3
There has been some speculation in Western media about SARS-COV-2 accidentally leaking out of the Wuhan Institute of Virology (Wuhan, China). These allegations have been repeatedly denied by Chinese officials and refuted by the scientific community,4 and we will absolutely not endorse them here. We must acknowledge, though, that:
a) Repeated biosafety incidents have proven that an outbreak can be caused by an uncontained infectious agent. Safety measures in laboratories keep improving but, as we all know too well, there is no infallible safeguard.
b) The 2020 pandemic has proven that a local infectious episode can propagate to the entire world and end up having devastating consequences.
Therefore, and for the sake of argument, let us consider a Pandemic Induced by a Loss of Containment of a Biological Agent (PILOCBA) as a potential new class of industrial incident.
To put things in perspective, we have revisited a previous work, presented in 2016.5 At that time, DEKRA reported that the potential distribution function provided an illuminating means of understanding industrial incidents. To put it in very simple terms, we discovered that for every ten incidents with x fatalities, there is approximately one incident with ten times x fatalities. We have carried out the same type of analysis here, taking the reported cost of the incident as our stochastic variable. Furthermore, we have adopted the IMF’s estimate as the potential cost of a PILOCBA. Figure 1 shows our findings.
If we consider a PILOCBA as an industrial incident, we can see immediately its cost is substantially more than incidents observed in the past (the Fukushima Nuclear Power Plant incident is another outlier).
Figure 1 – Distribution of costs associated with industrial incidents
It bears remembering, however, that humankind has been using micro-organisms to manufacture goods for millennia, producing bread, wine, beer and cheese to name a few common examples. Recently, we have even developed methodologies to genetically modify micro-organisms and “teach” them to make what we need. The pharmaceutical industry pioneered this development, but it has been extended to other applications, such as:
• Bioremediation: the use of living organisms to clean up hazardous chemical spills underground or in the sea.
• Food manufacturing: in meat replacement products and some other specialty items.
• Biodesulphurisation (BDS): a non-invasive approach to removing sulphur from fuels, used in the chemical and petrochemical industries.
• Microbiologically Induced Calcium Carbonate Precipitation (MICP): for repairing cracks, preventing corrosion in concrete and other cementation applications.
Indeed, we increasingly use micro-organisms to manufacture almost any product we need.
On the other hand, a loss of containment of hazardous biological agents can have disastrous consequences. It is easy to connect the dots: is it possible to have an industrial accident involving biological agents? As it is not impossible, we believe it is a matter of when, not if it will occur.
What can be done?
It is not, by far, the first time that humankind has faced hazards of its own making, from the control of fire by early humans to aeronautics, chemistry and the nuclear industry. In every case we have succeeded in harnessing those hazards until they become acceptable. This time will not be different.
Figure 2 – Evolution of process safety management
At DEKRA we strongly support recycling some of the tools and practices from process safety to deal with the new hazards and risks posed by biomanufacturing. The Center for Chemical Process Safety defines process safety as:
“…a disciplined framework for managing the integrity of operating systems and processes handling hazardous substances by applying good design principles, engineering, and operating practices. It deals with the prevention and control of incidents that have the potential to release hazardous materials or energy. Such incidents can cause toxic effects, fire, or explosion and could ultimately result in serious injuries, property damage, lost production, and environmental impact.”6
In order to include the risks associated with a loss of containment of hazardous biological agents, we simply need to replace “hazardous substances” with “hazardous substances and biological agents.”
The approach to managing process safety has evolved over several decades, passing through the stages shown in figure 2.
At a very early stage, prior to the establishment of the Center for Chemical Process Safety (CCPS), “doing things right” was considered sufficient to safely conduct chemical processes. Much effort was therefore dedicated to developing standards, regulations, guidelines, best practices, etc. Certain dramatic events, such as Bhopal (December 3, 1984) and San Juan Ixhuatepec (November 19, 1984) disproved this line of thinking, and led the American Institute of Chemical Engineers to create the CCPS and task it with the development of appropriate tools to manage chemical hazards. It seems that the biohazard world is still, at least in part, at this early stage, with “Biosafety in Microbiological and Biomedical Laboratories” being a de-facto standard of good practice.7
We could summarise the “safe distance” approach as keeping hazardous activities far away from the general population, so that if something goes wrong, people will not be harmed. This criterion was, and still is, rather popular in industrial regulations. Its main advantage is its simplicity: anybody can measure distance! Its drawbacks, however, are numerous. It is not easy to envision how this type of approach could be used to effectively manage biohazards; at most, we can think of the “social distancing” practices imposed to control the COVID-19 pandemic.
One of the problems with the distance approach is the difficulty of determining the appropriate distance: too little, and hazards are not mitigated, too much, and the result is suboptimal land use. To solve this problem, a new approach unfolded based on calculations of the consequences of hypothetical incidents and by which vulnerable populations are kept farther away than the calculated distances. This type of approach is theoretically also possible for biohazards, but it is arguable whether it would lead to any meaningful consequences. For, instance, what happens if the calculated consequences are as catastrophic as a PILOCBA?
Table 1 – Workstreams and CCPS elements
Finally, the CCPS concluded that the right approach was risk-based: the risk of any given event is the expected value of the damage it causes, or more simply put, the product of its likelihood and the resulting damage. Prevention efforts are then commensurate with the magnitude of the risk. It seems very natural to extrapolate this principle to biohazards: we certainly need to put more effort in preventing a pandemic than a minor outbreak. As a matter of fact, “Biosafety in Microbiological and Biomedical Laboratories” already has an implicit risk approach, calibrating safety requirements according to the infectiousness of the micro-organism being handled.
Furthermore, the CCPS identified the essential elements for world-class process safety performance. These are shown in Table 1, as grouped in DEKRA’s Organizational Process Safety solution scheme into seven workstreams.
Every one of the elements and workstreams seems fully applicable to biohazards, as long as its framework is properly identified. Let us consider, for instance, “compliance with standards”: certainly, biomanufacturers will need to keep track of applicable trade standards and regulations and comply with them. Another example, “hazard identification and risk analysis”: manufacturers ought to be able to identify the potential hazards of any new process and assess its risks. All the elements are, in fact, valid in a biomanufacturing context, and, indeed, exhaustive.
We can conclude, therefore, that the basic framework for managing risks set forth by the CCPS will remain valid with the addition of two further capacities:
• The development of new tools and methodologies or the adaptation of existing ones. Perhaps we can think of applying techniques such as HAZOP or quantitative risk analysis to a biomanufacturing facility, with at least some fine-tuning. We will also need to develop consequence modelling in line with the new types of hazards. The PSM structure will need to be revamped to accommodate this entirely new class of hazards.
• A whole new cohort of process safety experts with knowledge of biological processes to complement the expertise we already have in terms of chemical processes. Needless to say, appropriate competence development programs will need to be created.
Putting human ingenuity to work
New bioengineering technologies are increasingly applied to manufacture diverse types of goods, from food to pharmaceuticals, while other new applications use genetically modified micro-organisms to perform a range of tasks, such as remediating chemical or oil spills. While the possibilities are exciting, we plainly see that exposing humans to new micro-organisms, including engineered ones, can turn into a catastrophe with unprecedented consequences.
Fortunately, the human capacity to innovate and adapt can also make us safer. Process safety, for example, developed as the result of humans working to identify, assess and manage risks caused by hazardous materials, from flammable dusts (sugar, flour etc.) to those linked with dangerous industrial chemical reactions and new chemicals.
At DEKRA, we believe that the process safety framework is sufficiently robust and flexible to accommodate the new risks. What is still needed are new or adapted tools as well as additional experts with the appropriate background to adequately confront biomanufacturing risks. Relying on human ingenuity, experience and know-how, we can meet new challenges and conquer risks head-on.
1. Centers for Disease Control and Prevention. Biosafety in Microbiological and Biomedical Laboratories (5th edition). National Institutes of Health, 2009.
2. Della-Porta, T. “Laboratory accidents and breaches in biosafety – they do occur!”. Microbiology Australia, May 2008.
3. International Monetary Fund (April 2020). World Economic Outlook. Chapter 1. The great lockdown.
5. A. Trujillo (2016). “Industrial Accidents: are more Serious Events than Bhopal Possible?” Chemical Engineering Transactions. Vol. 48.
Dr. Arturo Trujillo, DEKRA Process Safety Consulting
6. CCPS (April 2020). https://www.aiche.org/ccps/process-safety-faqs#What%20is%20Process%20Safety
7. Amazingly, the first edition, titled “Biosafety in Microbiological and Biomedical Laboratories,” is dated 1984, the very same year of the Bhopal and San Juan Ixhuatepec disasters.
About the author:
Dr. Arturo Trujillo is Global Director of DEKRA Process Safety Consulting. His main areas of expertise are diverse types of process hazard analysis (HAZOP, What-if, HAZID), consequence analysis and quantitative risk analysis. He has facilitated more than 200 HAZOPs over the last 25 years, especially in the oil & gas, energy, chemicals and pharmaceutical industries.
Contact Details and Archive...