TS
Viral safety testing for biopharmaceuticals: Current and future prospects
2020 Volume 2 Issue 3 Pages 94-99
Details
2020 Volume 2 Issue 3 Pages 94-99
Biopharmaceuticals produced from animal cells and raw materials pose a risk of pathogen contamination. Thus, it is essential to pay special attention to contamination with infectious agents, including microbes, mycoplasmas, and viruses, in these products. In particular, viral contamination in cell banks, intermediates, or final products may occasionally be difficult to identify compared to contamination with other pathogens. Stringent viral tests including clearance tests have been conducted over the past ~20 years in accordance with the guideline Q5A. The safety of these products can be confirmed, to a reasonable extent, using various in vitro and in vivo viral tests, retroviral test, and clearance assessments to examine the viral removal and inactivation achieved by the purification process. However, viral detection using the current assays is not always comprehensive and focuses mainly on conceivable adventitious viruses. Thus, viral detection using high throughput sequencing (HTS) technology may improve the viral safety of biopharmaceutical products.
Viral safety is one of the major issues in the manufacture of biopharmaceuticals. Evaluation of viral safety using conventional viral tests has been conducted based on the guidelines; however, it does not detect a broad variety of virus families. Thus, to compensate for the lack of a comprehensive approach, an alternative viral assay using massive parallel sequencing technology is currently being developed.
The risk of pathogen contamination in biopharmaceuticals administered for human and viral contaminants are often more difficult to detect than other microbial contaminants. Several incidences of viral contamination manufacturing processes have been reported in the past [1]. Although no viral contamination has been detected so far in the final biopharmaceutical products, viral contamination in other types of biologics such as human blood- and plasma-derived products caused serious diseases such as human AIDS, hepatitis B or hepatitis C. Viral contamination occurring in the bioreactor lead to serious consequences for patients and manufacturers, including the complete shutdown of manufacturing facilities, leading to severe drug shortages and consequent financial losses [2, 3]. In the case of one such biopharmaceutical company that had to be completely shut down due to viral contamination, a novel Calcivirus was isolated from the contaminated bioreactor, indicating that there is a risk of unexpected viral contamination in drug manufacture using mammalian cell culture systems. So far, the viral safety of biopharmaceuticals has been tested mainly using the conventional virus assays described in the guideline [4]. While these assays have been effective, the range of viruses that may be detected using these assays is not comprehensive, and in vivo assays using small animals are required. Hence, current viral safety evaluation still needs to be partially improved. In addition, novel substrates, such as transgenic creatures, have been used for the production of biopharmaceuticals. Thus, considering the wide variety of virus families that may contaminate biopharmaceuticals, it is crucial to develop a comprehensive viral assay for safe regulation of potential contamination using high throughput sequencing (HTS).
Viral infection does not lead to common symptoms in cells. While some viruses can induce morphological changes in infected cells that can be detected relatively easily, others result in no morphological changes. Moreover, viral infections can be either productive with the release of infectious progeny particles, or latent with no virus production. The behavior of adventitious viruses depends on their interaction with various factors affecting viral replication in the target cells, such as viral receptors, cellular innate immunity, and cellular components essential for virion formation. Conventional in vitro and in vivo viral assays help detect the presence of viruses in specimens based on clinical diagnostics and were originally used to detect specific adventitious viruses assumed to be possible contaminants. These assays were later adopted to evaluate the viral safety assays described in the guidelines.
Another type of viral genome inherently lurk in the mammalian chromosome; up to 10% of the mouse genome comprises endogenous retrovirus (ERV) sequences, most of which represent the remains of ancient germ line infections [5]. It should be noted that some of these endogenous retroviruses located in the rodent chromosomes are constitutively active and produce virus-like particles in the cytoplasm and/or in the cell culture fluid [6, 7]. Transcriptional expression of endogenous retroviruses poses a risk of the generation of novel recombinant viruses to overcome the species barrier between rodents and humans [8]. Approximately 70% of biopharmaceuticals are produced using mammalian cell substrates [1], and Chinese hamster ovary (CHO) cells are the most commonly used host cells for monoclonal antibodies [9]. CHO cell derivatives are known to release retrovirus-like particles constitutively [7, 10, 11]. Although these particles are not infectious to human cells, risk mitigation should be conducted by viral clearance spiking studies for the downstream purification steps to demonstrate the capability of the purification process to remove or inactivate retroviruses using a virus similar to the endogenous retroviruses such as murine leukemia virus.
The risk of viral contamination is common for all biopharmaceuticals derived from cell lines. The viral safety of biopharmaceuticals is primarily determined based on the ICH Q5A guideline approved in 1999 by the Steering Committee of the International Council for Harmonization of Technical Requirements for Pharmaceuticals for Human Use (ICH). The scope of ICH Q5A includes therapeutic proteins produced by cell culture using the characterized cell bank as the starting material. Viral contamination events may occur mainly via three routes: (i) contaminated cell banks (ii) contaminated raw materials or manufacturing substrates (iii) contamination introduced during manipulation of cell banks or intermediates (Fig. 1). Thus, viral testing of cell banks, intermediates such as unprocessed bulk, and viral clearance tests including virus-spiking experiments in protein purification processes are crucial to maintain viral safety. The cell bank is the starting material for the manufacture of various therapeutic products, and thus, it is crucial to ensure its viral safety. Figure 1 shows the production process of recombinant proteins. From the isolation of stable producer cells to the establishment of the cell bank, there are several possibilities of introduction of viral contamination, such as: (i) virus-infected cells used for the preparation of cell banks (ii) cell lines inherently immortalized by virus infection (iii) generation of endogenous retroviruses in the cell bank (iv) cells contaminated through raw materials (v) cells contaminated by adventitious viruses during manipulation. Some cell lines are established by viral infection or transfection of the viral DNA to acquire immortalization. For example, HEK293 cells are isolated by the transfection of human fetal kidney cells with adenovirus type 5 DNA, and the 4.5 kb viral transcript can be constitutively detected in the chromosome [11]. Mammalian endogenous retroviruses are ancient sequences integrated into the germline cells that constitute approximately 10% of the murine genome and are vertically transmitted through the offspring. Endogenous retroviral genes in some rodent cells actively generate virus-like particles [12]. CHO cell derivatives constitutively generate retrovirus-like non-infectious particles in the cell culture fluid [7, 13], indicating that unprocessed bulk, cell lysed extracts, or culture fluid from CHO cells contain retrovirus-like particles as well as recombinant proteins.
Viral safety evaluation of biotechnological products derived from cell lines.
The cell bank is a starting material for the manufacture of biopharmaceuticals; therefore, the presence or absence of the virus should be carefully determined using various in vitro and in vivo tests, as well as retrovirus tests and electron microscopy (EM) studies (Table 1). In vitro tests should be carried out by inoculating the cell lysate and/or cell culture fluid into various susceptible indicator cell cultures. The choice of cells depends on the species of origin of the cell bank and should include a human and/or a non-human primate cell line such as MRC-5 and Vero cells. Vero cells have been used as indicator cells because of their high susceptibility to various viruses. However, indicator cells may not always be highly susceptible to viral infection. The minute virus of mice (MVM) belongs to the Parvovirus family without an envelope that has been known to contaminate bioreactors [1]. Vero cells were found to be >250-fold less susceptible to MVM infection than CHO cells detected by cytopathic effect (data not shown). The in vivo test also has an animal-specific range for virus infection. Cell lysate and/or cell culture fluid from the cell bank should be inoculated into animals, including suckling and adult mice, and in embryonated eggs to detect viruses that cannot grow in cell cultures. Viral testing of the unprocessed bulk using in vitro tests is important to detect the presence or absence of adventitious viruses.
| In vitro test* | Inoculation of cell lysate and/or cell culture medium to indicator cells such as Vero and MRC-5 cells (Hemagglutination/adsorption test) |
| In vivo test | Inoculation of cell lysate and/or cell culture medium to adult mice |
| Inoculation of cell lysate and/or cell culture medium to suckling mice | |
| Inoculation of cell lysate and/or cell culture medium to Guinea pig | |
| Inoculation of cell lysate and/or cell culture medium to allantoic cavity chicken embryo | |
| Inoculation of cell lysate and/or cell culture medium to yolk sac chicken embryo | |
| Retrovirus test | MINK S+L- focus assay for xenotropic murine retrovirus |
| XC plaque assay for ecotropic murine leukemia virus | |
| Observation by TEM (transmission electron microscopy) |
*In vitro test using bovine turbinate cells for bovine raw materials should be added when bovine raw materials are used.
The raw material used to produce biopharmaceuticals is also susceptible to the risk of viral contamination. Currently, the testing for adventitious agents in bovine serum is performed in compliance with the 9CFR 113 regulations (9CFR), the U.S. Department of Agriculture (USDA) or the EMA guidelines (Note for Guidance on the use of bovine serum in the manufacture of human biological medicinal products (CPMP/BWP/1793/02)). However, several viruses that are not of significant concern to the cattle and swine industries are not addressed in these regulations. Any contaminating viruses have the potential to overcome critical species barriers from animals to humans, indicating that the threat of virus contamination still exists in the manufacture of therapeutic proteins using animal cells with raw materials [14].
The ability to detect low viral concentrations statistically depends on the size of the sample. Therefore, the absence of infectious viruses in the final product is not determined solely by directly testing for their presence, but also by demonstrating that the purification process is capable of removing and/or inactivating the viruses. For example, the process of purification of recombinant antibodies has been highly improved using CHO cell derivatives, and it involves viral inactivation (low pH treatment) and removal step (virus filtration) for safety (Fig. 1). These steps are less affected by multiple conditions than other steps, such as column chromatography, and have robustness in decreasing the log reduction value of virus infectivity. Viral clearance testing should reveal that the purification process has sufficient capability to ensure the safety of the final product using different types of model viruses. ERVs inherently produced from CHO cells may be present in the cell culture fluids and get carried over into the unprocessed bulk. In CHO-K1 cells, 61 ERV-associated transcripts can be detected, three of which are translated into Gag polypeptide to form virus-like particles and released into the cell culture fluid (unpublished data). As two of the three ERVs belong to gamma-retrovirus, murine leukemia virus (MuLV) is genetically close to the ERV. MuLV is a reasonable choice to be used as a model virus in this case. The evaluation of viral clearance is based on viral infectivity rather than the copy number of the virus genome because the presence of viral nucleic acids does not directly indicate the presence of infectious viral particles.
It should be noted that viral safety testing should be carefully performed for novel substrates such as transgenic creatures. Transgenic chicken has been used for the production of biopharmaceuticals. The active ingredient, sebelipase alfa, which was approved in the US and the EU in 2015 and in Japan in 2016, is produced in the eggs of a genetically engineered chicken. Sebelipase alfa is a recombinant human lysosomal acid lipase (LAL) that replaces a genetically defective enzyme in patients causing fat to accumulate in the liver, spleen, and vasculature. To test the viral safety of the product in transgenic chicken, the set of specific viruses to be detected in egg white is different from those for drugs produced with mammalian cells, or clearance assay in the purification process (Table 2). For transgenic chickens, serological assays for influenza virus type A and avian leukosis viruses A, B, C, D, and J subtypes are required. Egg white liquor from transgenic chicken eggs is the starting material for purification of the recombinant enzyme. For in-process control, in vitro testing using Vero, MRC-5, and CEF cells, in vivo testing with the chicken embryo, electron microscopy, and PCR for several avian viruses are essential. In the viral clearance test, swine parvovirus, reovirus type 3, encephalomyocarditis virus, and influenza virus are used to confirm the sufficient log reduction values in the purification process. The viral tests are well-conceived, and model viruses are appropriate for conceivable adventitious viruses. In vitro tests are carried out for 16 viruses using a combination of MRC-5, Vero, and HeLa as indicator cell lines, and all the other viruses, except BVDV, are detected by an in vitro test [15]. These results may not prove the broad “specificity” or range of virus types that could be reliably detected with reasonably sensitive titers in in vitro tests, since the 16 viruses are well-known and most of them were adapted to replicate well in vitro. Recently, unknown viruses have been discovered and reported by HTS technology. Also coronaviruses are common in many different species of animals including camels, cattle, cats, and bats. Rarely, animal coronaviruses can infect human and then spread such as with MERS-CoV, SARS-CoV, and later with SARS-CoV-2. The COVID-19 outbreak once again proves the potential of the animal-origin source of emerging zoonotic diseases. We always keep in mind that the combination of several viral assays for cell banks can cover a part of a broad variety of whale virus species but need to improve the comprehensive viral detection approach to compensate for conventional virus evaluation as follows.
| Test article | Viral test |
|---|---|
| Monitoring test for | |
| Transgenic chicken | Serological testing: influenza virus type A, avian leukosis virus subgroup A, B, C, D, J |
| In-process control test for | |
| Egg white liquor | In vitro test (Vero cells, MRC-5 cells, CEF cells) in vivo test (chicken embryo), Electron microscopy, NAT (avian adenovirus group I, avian encephalomyelitis virus, avian orthoreovirus, avian reticuloendotheliosis virus, chicken anemia virus) |
| Clarified egg white liquor | In vitro tests (Vero cells, MRC-5 cells, CEF cells) NAT (West Nile virus, influenza virus type A) |
| Virus clearance test** | Confirmation of viral clearance in purification processModel virus: swine parvovirus, reovirus type 3, encephalomyocarditis virus, influenza virus type A, X-MuLV |
*Viral tests in this table were based on the Review reports on sebelipase alfa (https://www.pmda.go.jp/files/000219254.pdf).
**Purification process contains virus inactivation and nanofiltration step.
Conventional viral assays specified in the guidelines focus on determining the presence or absence of competent viruses. It may take several days and weeks to complete the in vitro and in vivo tests. In addition, they require skilled technicians and biosafety level 2 (BSL2) facilities to handle viruses. Currently, in vivo experiments using small animals are being replaced by alternative experiments. HTS technology has the potential to provide a solution for the issues regarding current viral safety evaluation (Fig. 2) . Transcriptome data obtained by HTS from a cell bank provide sufficient data to determine the presence or absence of adventitious and endogenous viruses. Figure 3 shows the analytical flow of virus detection using transcriptome data from the cell bank. One of the advantages of this method is that the target virus information is not needed in advance, while PCR requires a viral sequence for probe and primer sequences for assay. The PCR method has excellent sensitivity and can detect a specific viral nucleic acid in a short time, however, it cannot cope with unknown viruses and the number of virus species that can be detected is limited. The virus detection capability of in vitro test depends on the host range and viral susceptibilities of cell lines used for monitoring. On the other hand, the HTS method enables the search of whale virus families in a relatively short time, our viral databases contain approximately 160,000 viral sequences. We constructed a pipeline to detect the viral sequences present in the RNA of virus-infected cells with high sensitivity and found that viruses can be detected efficiently. The features of the HTS method are as follows: (i) the detection sensitivity fluctuates slightly depending on the conditions and the target RNA or DNA virus (ii) it is possible to detect a wide range of known viruses as well as unknown but closely related viruses (iii) a few micrograms of cellular total RNA is sufficient for analysis (iv) multiple viruses can be detected in the sequencing data (v) to reduce pseudo-positive signal background HTS data is needed (Fig. 4) . In order to develop a standard protocol for practical viral testing with HTS, international collaborative research by the Advanced Virus Detection Technologies Interest Group (AVDTIG) is in progress [16,17,18,19]. The group is composed of international regulatory, industry and academic scientists, and facilitate discussions and provide a forum for sharing data and experiences using HTS technologies.
Detection of viruses by high throughput sequencing (HTS) and current assays.
The workflow of viral detection using transcriptome data.
Detection of viral nucleic acids by high throughput sequencing (HTS).
The risk of viral contamination in cell lines and raw materials is of great importance in the production of biopharmaceuticals. Therefore, the evaluation of the viral safety of such products is essential. The conventional viral assays described in the ICH Q5A guideline have been effective in ensuring the viral safety of biopharmaceuticals for ~20 years. On the other hand, an improvement in sequencing technology enables the detection of a broad range of virus nucleic acids. The novel viral detection assay combined with in vitro assays will compensate for the disadvantages of conventional virus assays, saving time, the cost for assays, and provide a comprehensive platform for virus detection with high sensitivity.
The authors declare no COI and take full responsibility for any COI that may arise in the future.
We thank Drs. Ken Kono, Yoji Sato (NIHS) and Ryutaro Hirasawa (PMDA) for helpful discussions.
