Abstract
CONFERENCE PROCEEDING Proceedings of the PDA/FDA Adventitious Viruses in Biologics: Detection and Mitigation Strategies Workshop in Bethesda, MD, USA; December 1–3, 2010
Guest Editors: Arifa Khan (Bethesda, MD), Patricia Hughes (Bethesda, MD) and Michael Wiebe (San Francisco, CA)
The production of biotechnology products using mammalian cell lines offers an inherent risk of viral contamination because of the scale of the process and the complexity of the materials employed. The testing of production cell lines, raw materials, and test execution at appropriate stages of production all combined with viral inactivation or removal strategies ensures that an infectious agent is absent from the purified final product. Perhaps because of these efforts, biotechnology products have not been linked to a negative clinical consequence. However, manufacturing viral contaminations still do occur and may have a great potential negative impact to our patients by disrupting the drug product supply chain. In this paper, additional end-to-end complementary viral safety program considerations are suggested beyond the traditional viral testing and inactivation/removal strategies. These additional points of consideration should be thought of as augmenting the above approaches to further provide a reasonable measure of mitigating the risk of viral contaminations within the biopharmaceutical manufacturing facility.
The scope of this paper is on biologics produced in mammalian cells with an emphasis on viral contaminations involving Chinese hamster ovary cell production, although for the examples given as lessons learned with previous industry contaminations, vaccine production issues have been included as a general reference.
I. Consider That Virus Contaminations Will Occur—Lessons Learned
Biopharmaceutical production is a very complex and often a large-scale process. Many different culture media components are used in large amounts to support a cellular substrate that is permissive for viral infection and propagation. At least in part, viral contamination can be spread theoretically from a single infected cell by a single infectious entity because the process involves expansion of cell lines into master and working cell banks followed by growth into a cellular population for large-scale bioreactor applications. Depending on the design of the manufacturing process, the bioreactor cell culture fluid is harvested from days to many weeks post-bioreactor start-up followed by purification activities that lead to the final drug product.
Because of the basic design of the manufacturing process of biologics, one could easily suggest that there is an inherent risk for introducing a viral or other adventitious agent at any step along the path to the final drug product. To promote the absence of an infectious agent in the purified product, the biopharmaceutical industry follows regulatory guidances that outline strategies for viral testing along with viral removal and inactivation activities (1–7). Together such strategies have offered a good level of assurance to prevent a negative clinical consequence to patients treated with biologic products. However, even with the testing and removal/inactivation activities in place, viral contaminations do occur and can indeed affect patients by affecting their drug product supply chain (Figure 1).
Illustration of a key point that production of a biological should be considered to have an inherent risk of viral contamination due to the size and complexity of the production activities. Impact to patients from a contamination event can still be realized even if the testing and viral inactivation/removal activities are in place to protect the final drug product.
Although the frequency of viral contamination events may be low in relationship to the number of bioreactor runs (i.e., 1 contamination event in over 1000 cell culture starts), the impact can be most significant in terms of cost, patient, and agency trust, and as stated above the patients' supply of the product. Figure 2 shows the published industry contamination events since 1998, including the Genentech events in the 1993 and 1994 contaminations with minute mice virus (MMV), which will be described below. One point worthy to mention is that Figure 2 shows only the published contamination events. Other events most likely have occurred but have not been reported and information not shared with the rest of the industry. However, knowledge shared between biopharmaceutical firms after such contamination events as to root cause and lessons learned has increased over the past few years. However, one cannot help but notice that the renewed interest in the topic of viral contaminations appears to correlate with the recent published contaminations. Knowledge sharing between firms and with the regulatory agencies along with constant vigilance should be considered beneficial to the industry and ultimately to patients.
Published industry viral contamination events of biologics and vaccine manufacturing facilities. The virus identified, the infected cell substrate, and the year of the event is noted. Figure courtesy of Michael Wiebe, Ph.D., Quantum Consulting, LLC.
I.A. A Few Common Viral Contamination Themes
By no means is the following information a comprehensive account of observations from all published viral contaminations. However, there are a few common themes that have been reported within various forums that deserve mention:
The same species of virus may be seen in multiple manufacturing facility geographic locations and associated with different manufacturing processes.
In examples of multiple viral contaminations between campaigns within the same bioreactor and process, there may not be a common production media raw material lot number used between the incidences.
Most often, there is no definitive entry point identified for the contaminant. Raw materials such as various manufacturing components like fetal bovine serum appear to be the most likely contamination source. However, environmental factors and human introduction of a contaminant are often cited as the most probable cause as well.
Routine testing for adventitious agents may be negative during the contaminated run.
Positive viral detection results have been found to be conflicting between different contracted testing laboratories during routine testing and during investigations.
Electron microscopy and polymerase chain reaction (PCR) test methods are common methods to verify the contaminant during an investigation.
The contamination event may have initially been detected only by cell culture performance indicators. The indicators include lower cell viability, decreased oxygen consumption, or other cell culture performance measures that were labeled as suspicious.
Firms have been unaware that some of the critical cell culture media components such as serum were not treated to mitigate the risk of adventitious agents, or if aware did not assess the risk of using such material.
Those accountable for a manufacturing process have not always been familiar with all components in the cell culture media formulation, such as how the component was manufactured or raw materials used to manufacture the component.
There have not always been established contamination response action plans or even in-production action plans within controlled documents to account for out-of-trend or suspicious culture observations. Also lacking may be sufficient in-house expertise along with a defined decision-making process for halting a production run.
I.B. Cell Culture Contaminations Cannot Be Linked to Any One Source
As mentioned above, a viral contamination root cause, origin of the infectious agent, or entry point is often not clearly discerned as an investigation outcome. In general, a viral contaminant may be derived from the cell lines themselves or be introduced during the production processes (Figure 3). The complexity in attempting to identify a definitive source of an infectious agent starts at realizing that all mammalian cells are susceptible to viral infection. Therefore, even at the earliest stage in the handling of the presumptive production cell bank one can introduce virus through contaminated reagents or infection from the tissue source of the cells. The added complexity comes from the possibility of latent virus infection and endogenous retrovirovirus that may be passed on through multiple cell bank lineages. Likewise, during the manufacturing process itself, there may be facility airflow risks, operator introduction of an infectious agent, use of contaminated gases, or use of contaminated production culture media raw materials.
Illustration showing the complexity in identifying the source of a viral contamination in a biologics product.
Although often raw materials are suspected, identifying the entry point to the raw material itself is one investigation path that may prove to be a difficult task. For example, many questions around the raw material supply chain may be relevant to the actual contamination issue. For example, one has to consider the origin of the material, whether animal, non-animal, or perhaps the component was derived from a synthetic process. Manufacturing controls, cleaning, potential for co-mingling of infected starting raw materials, warehouse controls that involve pest control programs, staff interactions throughout the entire supply chain, and point-of-use activities in the biologics manufacturing facility are all relevant considerations. More will be discussed below about raw materials used in the manufacturing process with regard to animal- versus non-animal-derived media components.
I.C. Genentech MMV Viral Contamination—Lessons Learned
Genentech's experience with two large-scale manufacturing contaminations in 1993 and 1994 by MMV provides a “lessons learned” example, as the two events had significantly different outcomes (8). Comparing the two events that occurred between many successful bioreactor runs led to the development and implementation of viral barrier approaches that are currently in use. However, like that described above, no definitive root cause or source of the MMV was identified for the contamination.
As to the two MMV contamination events, one should note a few differences in the event detection and outcome and particularly the evolution of the viral barrier considerations and executed approaches (Figure 4). The 1993 event was considered to be “late” detection, or rather too late for preventing the spread of contamination to downstream processes. Virus was first indicated in the general viral screening assay via the hemadsorption portion of the test several days post-sample inoculation, followed by further testing to verify MMV as the contaminant. The time required for completing these testing activities was key to allowing the spread of the contamination to multiple bioreactors and downstream production activities. The overall impact of this event was extensive decontamination of the facility along with significant loss of product.
Summary of the Genentech MMV contamination events in 1993 and 1994. The figure points out the significant different outcomes between the two contamination events along with the evolution of the viral barrier considerations and executed approaches.
The outcome of the 1993 contamination, however, also yielded the development of an MMV PCR test as an early warning, in-process hold-step in combination with a cell-based MMV infectivity assay at the end of the culture. Further considerations also involved review and enhancement of a pest control program on the Genentech campus and specific attention to raw material vendor audits with emphasis on their own pest control programs. Following the 1993 event, work was started to investigate production-related raw material viral reduction methods, like the application of heat treatments.
The 1994 MMV contamination event demonstrated a significantly different outcome in that the contamination was isolated to a single bioreactor with minimal loss of product. Early detection utilizing the MMV PCR method prior to harvest was a significant factor in this outcome. However, what should be noted is this event led to the development and implementation of the viral barriers including high temperature/short time (HTST), autoclaving, and filtration as viral reduction approaches for raw material feed streams. In addition, serum was removed from the working cell bank culture and production processes. Augmenting the work that was done in the 1993 contamination, the MMV detection assays were applied to small-scale production activities and an emphasis was placed on personnel and equipment segregation and traffic flows within and around the facility.
II. Consider the Risks of Your Raw Materials
The focus of this section will be on cell culture production media raw materials. However, the overall message can be applied to the entire production process. That is, think of knowledge of your raw materials as a viral barrier, in that one should attempt to identify those viral risks associated with the materials. Cell culture media used in the manufacturing process are typically composed of any combination of purified salts, amino acids, vitamins, and water. These media may be supplemented with additives such as animal serum (fetal bovine or calf serum), which provide a source of nutrients, lipids, trace elements, and growth factors. Replacing the use of serum with protein hydrolysates, also called peptones, has been employed as a means to optimize culture media for productivity, maximize consistent product quality from run to run, and lower the cost of production. Peptones are a cocktail of amino acids and polypeptides obtained by enzymatic or acidic digestion of proteins of a given origin (i.e., meat from a variety of animal sources) as well as from non-animal sources like yeast, soy, cotton seed, rice, and so on. As the replacement for serum, peptones are the nutritional cell supplement to stimulate cell growth and protein expression during biopharmaceutical manufacturing.
Animal-derived peptone was one of the first supplements used to replace the use of serum. However, such animal sources imply an increased risk of virus, and/or prion contamination. Animal and public health threats of transmissible spongiform encephalopathies (TSEs), such as bovine spongiform encephalopathy (BSE; mad cow disease) found in cattle and the human derivative of this neurodegenerative disease (variant Creutzfeldt-Jakob disease) helped to initiate a move to remove animal-derived components from the entire production process. From the biopharmaceutical manufacturers' point of view, switching to non-animal-sourced raw materials such as those of plant raw materials is warranted because there appears to be an obvious reduction in the above noted safety risks. Furthermore, plant-derived peptones have been shown to have the ability to promote equivalent cell growth and production compared to similar animal-derived, peptone-supplemented cultures.
However, caution should be afforded when weighing just how much the safety risk has been reduced when switching to plant-based cell culture materials. Although the TSE risk may be reduced when using plant-derived materials, the potential presence of virus should not be ignored when considering that the plant-derived raw materials are derived from an agricultural resource—the farm.
When switching from animal-derived production raw materials to plant-derived materials, one should consider the following when assessing viral risks:
There are no regulations that govern farming practices as it relates to plant-derived materials for the biopharmaceutical industry.
Raw materials derived from plants are usually expected to be fit for human consumption.
Raw material vendors usually use growers who follow applicable federal, state, or local laws with regard to good farming practices such as those provided by the USDA and FDA.
Potential for contamination of crops with animal materials (i.e., rodents, insects, and soil) should be considered.
Manure may be used as fertilizer during the agricultural process, and some countries may use human waste for that purpose.
Therefore, animal-based contaminants should be considered when using plant-derived materials along with the notion that expanding the source of your raw materials may increase the potential for a greater variety of potential viral concerns. Knowledge of the raw material end-to-end supply chain should always be considered.
III. Consider the Limitations of Cell-Based Viral Detection
Although viral testing is a requirement and the basic design of the methods are outlined in several regulatory documents, one should understand those methods' limitations. When considering the cell-based viral detection methods, those limitations are based on the most basic principles of cell biology and cell culture practices. Figure 5 summarizes the types of viral detection assays that are generally employed for biopharmaceutical manufacturing. These include assays that are based on in vitro and in vivo methods, electron microscopy, and PCR. This repertoire of methods serves the purpose of demonstrating the presence or absence of virus in raw materials, manufacturing cell banks, and unprocessed bulk harvest fluid for lot release testing (5). Figure 6 illustrates the various points at which these testing activities take place within appropriate stages of the manufacturing process.
General summary of the types of viral detection methods and their uses.
Illustration of the points at which various testing activities are executed throughout the manufacturing process at appropriate stages. The absence of virus would be demonstrated in raw materials and bulk harvest cell culture fluid. A representative sample of the unprocessed bulk prior to further processing represents the most suitable point to detect adventitious viral contaminants. PCR for rodent parvovirus is shown as the early-warning hold-step described in section I.C. of this report.
In vitro viral screening assays have over the past several years been the focus of attention for several manufacturing firms since there have been reported false positives (9), false negatives, and conflicting final reportable results demonstrated between different testing laboratories on the same sample aliquot (10). The in vitro viral screening method is based on the principle that a virus will replicate in a suitable host cell, demonstrating its presence through morphological changes observed in the infected cells and/or by hemadsorption of erythrocytes onto the cell surfaces or hemagglutination. Suitable indicator cells are used based on their ability to become infected by target viruses. For the general viral screening assay, three different indicator cells are generally employed: a human diploid cell line (MRC-5), a primate cell (Vero), and a cell line that represents the same species as the cell substrate used in the manufacturing process (1). This design affords the opportunity to detect a wide spectrum of viruses infectious to humans and to the manufacturing cell substrate. For other specific viruses such as MMV, a cell line demonstrated to be sensitive to this virus (324K cells) is employed. For the 9CFR testing of raw materials, bovine or porcine viruses are screened using species-appropriate indicator cells.
The actual execution of the in vitro screening methods involves the inoculation of the test article (i.e., unprocessed bulk fluid from the bioreactor) onto the indicator cells for a defined period of time. The cells are subsequently incubated for 2 to 4 weeks and observed under the microscope for evidence of infection throughout the duration of the assay as described above. The detection of some viruses may not be realized or inconsistent because of two contributing limiting factors: (1) the nature of viral infection itself and (2) the nature of the methods' dependency of the cells to become infected. Namely:
Not all viruses induce microscopically noticeable morphological changes in the indicator cells (cytopathic effect) and/or hemadsorption.
The panel of indicator cell lines used in the methods will show different susceptibilities of infection for a particular virus due to the mechanism of infection.
Infectivity potential of the indicator cells may be affected by basic cell culture practices and the test article itself.
Because the in vitro methods rely on the microscopic detection of infection, there is an inherent insensitivity and variability that may come about by how the cells are maintained and by direct impact from the sample matrix (Figure 7). Good cell culture practices should be considered essential for maintaining the indicator cell lines with specific attention to consistency in adhering to passage limits, subculture procedures, seeding densities, and confluency at the time of sample inoculation. Such factors when varied between test sessions may influence cell proliferation and/or other intracellular mechanisms that may negatively affect infection and subsequent further propagation of the infection. Ultimately the result is a failure to microscopically discern signs of a cytopathic effect. In our firm's experience, failure of outside contract laboratories to detect reference-spiked qualification samples was due to such non-robust cell culture practices between test sessions.
Graphical illustration of the inherent insensitivity and variability that can come about by cell culture conditions and an impact from the sample matrix. These impacting factors can be reduced by controlled culture conditions and validation of the method for the intended sample matrix.
Likewise, when the above cell culture conditions are not robustly controlled, morphological indications of a viral-induced cytopathic effect may be realized, resulting in a false-positive indication (9). False-negative and false-positive results can be minimized through adhering to cell culture procedures outlined in quality control documents along with validating the method for the intended sample matrix. In all, viral infectivity assays are a regulatory requirement and an essential tool for demonstrating the presence or absence of an infectious agent. However, based on the noted limitations, testing activities should not be the primary element of one's overall viral safety program.
IV. Summary: Consider All Parts of Your Viral Barrier Strategy
A biopharmaceutical viral safety program has additional considerations beyond the viral inactivation/removal and testing activities of raw materials, cell lines, and the testing at appropriate stages during production (Figure 8). Industry experience and our own contamination events have shown us that additional activities, including controlled sourcing and understanding the risks of your raw materials whether animal-derived or non-animal-derived, applying viral inactivation steps (heat) to raw material feed streams, and recognizing the limitations of viral detection methods, adds an important, holistic strategy to facility viral safety. Such a broader view of complementary considerations adds to the assurance of an uninterrupted supply of drug to patients (Figure 9).
Illustration of the considerations focused on testing and viral inactivation/removal as a primary approach.
Illustration of the considerations focused on testing and viral inactivation/removal as a primary approach.
- © PDA, Inc. 2011
Jump to section
Related Articles
Cited By...
- No citing articles found.