Risk averse is a turn of phrase often leveled at biopharmaceutical companies. Risk aversion is an important characteristic of organization behavior in biopharmaceutical manufacturing, particularly, in relation to ensuring product quality and patient safety. These things can never be compromised.
But at what point does risk aversion stifle innovation? Is it possible that internal quality department functions are unnecessarily cautious even when the weight of scientific analysis suggests that a given approach should be accepted?
An example currently being debated is related to closed systems and whether these systems can be established with a sufficient degree of confidence that classified environmental controls around bioprocesses can be removed. Properly executed risk analysis shows that closed systems can be established and operated with acceptably low probability of contamination. Using closed systems opens up many new possibilities for how facilities are designed and operated and may also present lower risk to the operation and, ultimately, the product.
Improved Technology Pushes Change
The history of processing drug substance follows precedence where facility designs based on replicating sterile operations like vial filling gained regulatory acceptance and then all subsequent facilities took the safe route and did the same. However, this approach is being challenged as technology improves and bioprocessing expertise increases. Using a pre-existing facility as a template for the design of a new facility can mean poor design practices are retained and innovation is suppressed.
The rise of single-use elements provides solutions to many of the problems previously holding back end-to-end creation of clean and closed systems. Cascades of classified and controlled space, gowning, air locks, steam generation, and environment monitoring add cost but no actual value. This means non-value adding waste (NVA) is built into the business and product costs are higher than necessary.
Despite advances, there is resistance to fully embracing the benefits of operating facilities as closed. It means that the most engineers and operators have been able to get is a “half-way house” of reduced environmental controls, fewer air changes per hour, and simplified facility designs with fewer silos and less segregation. The reason for this is uncertainty; and it is not the technology that is particularly in question. The risk of regulatory rejection is more of a concern.
In the past, regulatory response was perhaps more predictable as the troika of US, European Union, and Japan health authorities dominated the landscape. Now there are many more agencies to satisfy, and their perspectives are more difficult to foresee.
Few companies want to be the first to expand the boundaries with new approaches particularly if they have yet to be tested with industry regulators. However, implementation examples are starting to be seen. In one existing plant, for example, a thorough risk analysis was done from a closed-system perspective and risk mitigation actions implemented. Room classifications were lowered, some to controlled-not-classified (CNC) status. In another example, European regulators approved commercial supply from a facility with mainly single-use technology operating in CNC space. In another
example, an operation for producing clinical lots is in mainly CNC areas. Some companies have taken the justifiable step to have operators in street clothes with overshoes. Not an intuitively agreeable step to take in a conservative industry.
A New Approach
If bold attitudes are to be considered, then there is another change of mindset being debated. The train of thought goes as follows; one scenario to envisage for a bioprocessing operation is where there has been accidental ingress of contamination into the process stream and it remains undetected. A typical failure mode and effects analysis would tell us that a low detection score combined with a high severity score would require careful consideration and risk mitigation. If online detection monitoring tools are available, then logic dictates one would want to detect the contamination as soon as possible. If this idea is taken one step further again, then if there is a breach, it needs to happen heavily and quickly. Does this mean we are better off if the space around a bioprocess has more contaminating organisms than those with fewer? In other words, the real benefit of locating a bioprocess in a suite of lower classification is that it helps in forensic engineering and scientific identification of a breach of an otherwise closed system.
When biopharmaceutical companies are told by regulatory agencies that they need to reduce the incidence of process contamination from the environment, their inability to adequately close their process systems forces them to improve the qulity of the environment. However, does this reduce the risk? Does this reduce the probability and possiblty the severity of the contamination, or does it simply reduce the level to a point where detection because difficult? The resulting phenomenon is one of inconsistent or irreproducible mishaps.
It is unlikely this last approach will become reality any time soon. There is clearly a paradigm shift underway, however, and bioprocessing plants will increasingly be operated in non-classified space. This will become best practice in two to three years and accepted and common practice in five to eight years.
There are many activities where the pharmaceutical industry is not risk averse. In fact when it comes to investing money in new drug development, companies can be proud of the history of finding winners from a sea of risky projects. With a careful eye on the odds, the technical operations side of the business should move forward with new approaches with greater confidence.
Simon Chalk is director of the BioPhorum Operations Group, firstname.lastname@example.org.