A study conducted through a consortium of biotech companies and the Massachusetts Institute of Technology’s Center for Biomedical Innovation provides valuable evaluation of past contamination events and best practices for reducing risk.

Biologics manufacturing has a long history of safety. However, the use of cell culture to produce recombinant proteins does carry some risk of contamination either through bacterial or viral contaminants. While these events have been rare, they come at a very steep price when they do happen. The cost of a contamination can easily reach tens of millions of dollars and can cause drug shortages that negatively impact patients. For the biopharmaceutical company, it can also mean the loss of competitive advantage; a reduction in valuation and extensive remediation with facility shut downs. Thus reducing contamination risk is a high priority and requires regular study to improve prevention and detection methods.

Due in part to the low number of incidents and the lack of coordinated efforts to share contamination event information, it has been difficult to evaluate previous contamination events and to learn from them.

This was addressed in a recent journal article titled, “Viral Contamination in Biologic Manufacture and Implications for Emerging Therapies”, (Nature Biotechnology April 2020). The paper was made possible through a consortium of biotech companies and the Massachusetts Institute of Technology’s Center for Biomedical Innovation. Their goal was to collect data on contamination events industry wide and to report on the most common viral contaminants, source of contamination, cell lines affected, corrective actions and the impact of these events. The result is a compelling look at drivers of previous contaminations, contamination sources for different cell lines, and the impact on cell and gene therapy manufacturing.

I was fortunate to be able to interview study author, Paul Barone, PhD, Center for Biomedical Innovation, Massachusetts Institute of Technology, about the key takeaways from the study and how these can be applied to manufacturing today.

Study Scope

The Consortium on Adventitious Agent Contamination in Biomanufacturing (CAACB), a biopharmaceutical industry consortium including more than 20 biotechnology companies housed at the Massachusetts Institute of Technology’s Center for Biomedical Innovation, collected comprehensive data on virus contaminations in cell culture operations from CAACB member companies. These data were consolidated with information from published reports of virus contamination events.

The scope of the project was limited to virus contaminations in mammalian cell culture manufacturing. The project did not include bacterial or yeast fermentation, plasma fractionation or egg-based production of vaccines and covered manufacturing from the pilot to commercial scales, including both current Good Manufacturing Practice (cGMP) and non-cGMP operations.

Findings

Current Viral Contamination Risk Reduction Methods

For the purpose of this study, authors chose to focus on viral contamination because it is more difficult to detect and carries the risk of transmitting human pathogens. The study authors described current best practices for reducing the risk of viral contamination – “selection of appropriate starting and raw materials with low risk of containing adventitious virus, testing of cell banks and in-process materials to ensure they are free from detectable viruses, and incorporation of steps to remove and inactivate potential undetected adventitious and endogenous viral contaminations during purification of the product.”

In our discussion, Dr. Barone pointed out that this approach has worked well in recombinant protein production with only 26 virus contaminations over the past 36 years and 18 reported directly as a result of the study. However, nearly 50% of the companies that participated in the study had suffered a contamination event and these companies, by some estimates, represent 75% of global mammalian cell manufacturing capacity. The takeaway is that while virus contaminations are rare, every company is at risk. This is how viral contamination should be approached with respect to prevention and detection.

Different Risk Profiles for Cell Lines

Of the viral contamination events, 67% were in CHO cell lines and 33% were found in human or primate lines. As the authors note, this is not surprising as CHO cells are the most common cell line used in biologics manufacturing.

Dr. Barone explained that four different viruses were detected in the CHO cell line contaminations, none human in origin, and only one that posed a pathogenic risk to humans (Cache Valley virus). Eleven of the twelve CHO cell contaminations came from raw material or media components. This was very different than the viral contaminations in the human or primate cell lines. In these lines, there were five different viruses represented that did not overlap with the viruses found in CHO cell lines. Virus contamination was found to come from operators or the cell lines themselves. Four of the five viruses were known to be pathogenic in humans compared to only one of the CHO viruses. Thus viral contamination in human and primate lines represents a greater safety risk to patients because of the pathogenic risk to humans.

Dr. Barone reiterated that these two types of cell lines, CHO or human/primate, require different approaches for the prevention and detection of viral contamination. For instance in CHO cells, the primary introduction risk is in raw materials and media components, so enhanced risk reduction for these components should be implemented.

In cell and gene therapy, this may not be as easy since the contamination is primarily introduced through operators and the cell lines themselves. Risk reduction for these lines requires that cell lines be tested and well characterized where possible. Aseptic procedures should be used at all times with closed system operations employed as much as possible.

Animal Derived Raw Materials Present a Greater Risk

The study found that animal derived raw materials, especially serum, carry a higher risk of being contaminated with virus and are being replaced in the industry where possible. Three of the four viruses that contaminated the CHO cell lines were suspected or positively identified to have come from serum. However, Dr. Barone explains that removal of serum does not eliminate the risk of contamination. For instance, in one of the contamination events with Minute virus of mice (MVM), also known as Mouse minute virus (MMV), the process contained no animal derived raw materials.

While there is a paradigm shift toward animal free raw materials, the move is slow in cell and gene therapy applications. Dr. Barone explained that this is understandable because with these cell lines the biggest hurdle is understanding the key nutrients and amino acids needed to adequately replace serum.

If animal derived materials can’t be eliminated from the culture, Dr. Barone recommends treating the material to remove or inactivate the virus. Methods like gamma irradiation, nanofiltration, flash pasteurization (HTST), and UV-C irradiation can significantly reduce viral risk. In the study, none of the four manufacturers that implemented HTST heat treatment in media experienced a contamination event afterwards.

Raw Material Testing

While raw materials were the suspected source of 11 of the 18 contaminations, testing raw materials in advance did not always detect the presence of virus as testing only identified virus in three events. To clarify, Dr. Barone did state that the study did not ask companies how many times a positive raw material test prevented a contamination event. In other words, data was not collected on the instances when raw material testing succeeded, only when it failed.

Dr. Barone went on to say that testing raw materials alone is not enough to prevent contamination, but should be used as part of a broader strategy. He described that raw materials should be evaluated for risk based on how likely they are to contain a virus.

Serum has a long history of contamination, so it should be identified as a high-risk material. For serum, using gamma irradiation or other treatments could reduce risk. For other materials, filtration may be the best option. The point is that each raw material should be evaluated and a strategy for risk mitigation employed.

While testing is not foolproof, Dr. Barone does think that testing is an important part of a larger mitigation plan. Study data showed that viruses or their potential presence in raw materials, especially where the virus may be heterogeneously distributed, could be increased to a detectable level by concentrating or increasing the viral titer.

Dr. Barone also discussed the importance of in process testing particularly if a rapid testing method can be used. The study found that a rapid test implemented two days before bioreactor harvest was shown to identify viral contamination in several cases, which prevented the contamination from being moved forward in the process. This stopped a larger, more costly contamination from occurring. He went on to say that with CHO culture, there are a limited number of viruses likely to be a risk for contamination – MVM representing the biggest risk. Because of the small number of possible viral contaminants, companies can select a few that are high risk for rapid PCR testing. Time is often the biggest challenge for in process testing so using a targeted rapid PCR testing method could be used to identify viral contamination very quickly.

Implications for Cell and Gene Therapy Manufacturing

Dr. Barone explained that cell and gene therapies present unique challenges for viral contamination.

Gene Therapy

For gene therapies, a human cell line, primarily HEK, is used. The production process also involves adding a helper virus that has to be removed as part of the purification process. Viral clearance is used to some degree in both AAV and Lentivirus processes, which confers a level of viral safety. In addition, the cell lines used are well characterized and should be free of viral contamination.

Allogeneic Cell Therapies

In allogeneic cell therapies, the cell lines used are well characterized and cell lines are deemed low risk. There are more open operations in these processes that increase risk and these cultures often include higher risk, animal derived raw materials. All currently used viral clearance methods are off the table with cell therapies. In fact, Dr. Barone said that study authors could not even envision a technology capable of viral clearance that would remove both infected cells and the virus without also killing the cells. Thus, a focus on prevention of viral contamination is the best approach. This means good aseptic techniques, transfer to automated processes and removal of operator handling as much as possible. In addition, treating raw materials to remove or inactivate any virus prior to use in culture is important. Eliminating serum would also be a good prevention technique, although this may not be possible with all processes.

Autologous Cell Therapies

Dr. Barone explains that autologous cell therapies are the most challenging because the cells come from different patients with different disease states and different levels of disease progression. This means that the cells are much less healthy than well-characterized allogeneic cell lines. It also means that there is cell variability across the cells, which impacts the cells’ nutritional needs and makes eliminating animal derived raw materials especially difficult. Furthermore, because the turnaround time on autologous therapies needs to be very fast, tests on cells to ensure that they are free from viruses cannot generally be performed prior to beginning cell therapy manufacturing. As with allogeneic cell therapy manufacturing, the best mitigation strategy is prevention.

Summary

The study provides a welcome and previously absent comprehensive review of viral contamination events in biomanufacturing. While the industry has identified robust methods of viral clearance for recombinant protein biologics, cell therapies don’t have this option. This is particularly concerning with emerging viruses, like SARS-CoV-2, where information about the ability of the virus to infect biomanufacturing cell lines is not known at the outset. Finally, the takeaway that Dr. Barone stressed is that reducing viral contamination risk requires a comprehensive plan. There is no one size fits all approach. Effective risk mitigation requires evaluation of risk at each stage of the process and the tactics available to mitigate the risk.