Development Pathways For ATMPs: Virus Safety Challenges And Regulatory Perspective

This article was created from the author's presentation at the 2018 Bioprocessing Asia conference, with Cytiva as principal sponsor. The BioProcessing Asia Conference series was created to provide a platform to advance the contribution of bioprocessing sciences towards the development and manufacture of affordable biopharmaceutical products in Asia.

Advanced therapy medicinal products (ATMPs) are medical products for human use, often used in regenerative medicine. They include gene therapy, somatic cell therapy, tissue engineered, and regenerative medicine products. In the EU, ATMPs are regulated under a consolidated framework for advanced therapies as pharmaceutical products.¹ Within the EU, control of safety, consistency, reproducibility, and uniformity are key aspects of compliance for ATMPs. Annex 2 of the EU guidelines for medicinal products for human veterinary use has been updated to include GMP regulations for ATMPs. In the U.S., cellular and gene therapy products are regulated by the Center for Biologics Evaluation and Research.² Yet, in all regulated countries, it is clear that if animal or human cells are being used to develop an ATMP, there must be measures in place to ensure minimum risk when it comes to virus contamination.

Companies in the commercial sector with experience developing biopharmaceuticals understand these virus safety issues and have experience preventing them. However, those in the non-commercial sector without that background, such as universities, medical centers, and government, where many of these products are being developed, often do not have the in-depth experience when it comes to GMP compliance.³ And because doctors face situations where they need to make real-time decisions during the preparation of materials for ATMPs, there is an increased risk of viral contamination in these settings. For these reasons, it is important to understand the contamination risks and the solutions available to control them.

Virus Safety For ATMPs

Given the complex nature of ATMPs, ensuring the safety and quality of such products requires a rigorous, science-based approach with careful consideration of:

• sourcing of tissues and materials
• testing of product intermediates and/or reagents to ensure safety
• design of the manufacturing process of all components to minimize risk.

Many of these principles are captured in the ICH Q5A guidance.⁴ Viruses come in many shapes and sizes, and they infect all kingdoms—plants, vertebrates, and invertebrates. The goal of virus safety is to ensure that, if these contaminants were to be present, they do not find their way into clinical products, or, at a minimum, will not be clinically relevant to the patient who receives the ATMP. Past virus contamination events in cell-manufactured biologics demonstrate the need to be prepared for unusual sources of potential contamination.⁵

For example, Genentech published its experience in the 1990s with multiple contamination events of the Minute virus of mice (MVM) in its CHO fermenters.⁶ Although the source of the contamination was never clearly identified, MVM is a virus of the family Parvoviridae affecting mice, so the assumption was that the contamination was caused by rodents either in Genentech’s facility or at its suppliers’ facilities. Mice infected with MVM carry the virus in high titers in multiple tissues (as high as 10⁷ /milliliter), and it is excreted in their urine.⁷ Theoretical calculations indicate there were likely no more than one to three infectious particles per liter of medium, showing it does not require large amounts of a virus to cause a contamination issue. Without any current test methods available that can detect viruses at levels so low, the issue becomes even bigger when considering that Parvoviruses are among the most resistant viruses and survive for extended periods in the environment.

In another example, two contamination events with different Vesivirus 2117 (V2117) strains were detected in 2008 at separate Genzyme facilities on different continents (North America and Europe) with no identified shared components in use.^8,9 Later, a contamination event in 2009 at the same U.S. facility showed a virus 99 percent identical at the sequence level as the virus from 2008 was still contaminating cultures. Phylogenetic sequence analysis of the different strains of V2117 that have been observed in CHO cultures (the original contamination was reported in 2003 from bioreactors that were contaminated in the late 1990s) clearly demonstrate the U.S. and Belgium virus contaminants to be from different virus strains (Figure 1). Intriguingly, another Vesivirus isolated from a dog with gastroenteritis clustered along with these CHO-derived Vesivirus contaminants (submitted to GenBank in 2011; Acc. #JN204722), raising the possibility that these contaminants might be of canine origin.

Figure 1: Phylogenetic tree of Vesivirus 2117 isolates present in the GenBank sequence database in 2011 (prepared using the UPGMA method using the virus capsid protein alignment ¹⁰). The percentage of replicate trees in which the associated taxa clustered together in the bootstrap test are shown next to the branches (500 replicates). The tree is drawn to scale, with branch lengths in the same units as those of the evolutionary distances used to infer the phylogenetic tree. (Note: In recent papers, additional Vesivirus 2117-like contaminants isolated from dogs with gastroenteritis were identified and are not included in this tree.^{11, 12})

There are several theoretical possibilities of where contamination events, such as those observed with MVM and V2117, could originate. One is that the viruses are arboviruses, which are transmitted by insects. Biopharmaceutical manufacturers have reported receiving powdered cell culture media that, even after quality control testing, were contaminated with particulates that were later proven to be insect body parts. Although most arboviruses are enveloped viruses that cannot survive long periods of time, there are exceptions (e.g., Bluetongue). Another possible source is infected workers at a facility. In a GMP recombinant pharmaceutical facility, many manufacturers have requirements that if, for example, an employee lives on a farm, they must implement daily procedures to minimize the risk of viruses being transmitted from the farm into the facility. Employees could also unknowingly carry viruses transmitted from family pets, a risk not currently controlled by most manufacturers (nor, for example, their suppliers of cell culture media or media supplements).

In the end, controlling such risks is challenging and requires barrier technology for media prior to entering the bioreactor to ensure that low levels of contaminating virus do not present a significant risk. It further highlights the importance of ongoing risk assessments similar to those already conducted in the blood plasma industry, which could limit the risk of virus contamination events.

ATMPs And Possible Virus Safety Strategies

The safety tripod for virus, or product, safety is a careful balance among sourcing, testing, and virus clearance. This may require characterization of master cell banks as well as a selection of the raw materials and defining sourcing policies. There may also be in-process testing, using either infectivity or Polymerase Chain Reaction (PCR) assays. Finally, there must be validation of the manufacturing process (wherever possible), where a virus can be removed/inactivated through, e.g., chromatography, virus filtration, or detergent treatment.

A drawback of in-process testing using infectivity or PCR assays is that they have a defined limit of detection that is dependent on the type of detection system being used, the virus in question, and the volume of sample being tested. In addition, a negative result does not mean the sample is negative; it could just mean the virus is below the level of detection for the assay. For most manufacturing processes where virus clearance is not implemented, this level of risk is unacceptable, making it imperative that either a better process for virus detection is available or that steps are implemented to control this risk more carefully (e.g., via virus clearance).

Table 1 provides an example residual risk calculation that can be used for any component entering a bioreactor and demonstrates the high level of residual risk for any component where virus clearance has not been applied. Only a few infectious virus particles are required to initiate an infection in a bioreactor scenario, which will then spread through the whole culture.

Table 1: Example residual risk calculation for medium components used in a bioreactor.

The importance of implementing robust virus clearance in order to effectively control virus risk has also been demonstrated for the blood plasma industry, where risk reduction by careful donor selection and subsequent testing each demonstrated a risk reduction by approximately 100-fold (i.e., two log₁₀ depending on the virus type).¹³ Significantly higher levels of risk reduction, as much as 10 logs where you have two robust steps present (i.e., two by five 5 log₁₀ reduction steps) can be achieved compared to virus inactivation or removal. Testing is still important to quantify the amount of virus entering the beginning of the process and to ensure the process is not overloaded beyond the validated capacity to clear the virus. For Intravenous immunoglobulins derived from blood plasma, the introduction of robust solvent detergent treatments prevented the transmission of hepatitis C virus through those products.¹⁴

How can industry then best control virus safety risks with ATMPs? For the upstream process, there is already quality control testing of the starting materials (e.g., cell banks), but cell culture medium and other components may be used with only limited bacterial testing (e.g., only sterility and mycoplasma). The contamination events with MVM and V2117 demonstrate a need to apply risk control measures including implementing barrier technologies to prevent contaminants in the media or media additives from entering the cells.

For cell-based therapies, where no clearance of contaminating virus can be achieved downstream, the focus for controlling risk shifts to controlling both the cells as well as the medium used to grow the cells. First, there must be full documentation of the history of the cells. If they are taken from a patient at the bedside and reprocessed in culture before being injected back into the patient, documentation will be available. If, though, non-allogenic cells are being used, there must be knowledge about the kind of patient from which they were collected and how were they stored and processed. Many virus infections have no clinical symptoms or result in no cytopathic effect in infected cells, so it is sometimes impossible to know if donor cells are infected with a virus, as they may appear perfectly healthy. There may also be latent viruses that could be dormant during collection but reactivated later. In addition, poorly controlled collection or handling procedures can introduce contaminants.

Next, most somatic or tissue-engineered cells require cell culture either to expand the number of available cells for therapy or to differentiate the cells. This often requires the use of bovine serum or human- and/or bovine-derived supplements, which present a risk as purified growth factors are often not tested or manufactured to GMP standards.¹⁵ Even chemically defined medium can be contaminated (see above discussion with MVM and V2117). Due to these potential risks that need to be considered when preparing ATMP products, barrier technology may be an effective risk control measure.

Implementation Of Barrier Technology

The implementation of barrier technologies presents the best control measure for reducing the risk of introducing adventitious agents into a cell culture system. These technologies include high-temperature/short-time, small virus filtration, heat, UV treatment, gamma radiation, and other membrane technologies. Any of these could provide additional protection against low-level virus contamination. Hollow fiber anion exchange chromatography using QyuSpeed D (QSD) is an example of using a membrane-based technology to improve the virus and prion safety of pooled human platelet lysate (HPL), which has become a standard supplement for ex vivo cell culture for clinical protocols.¹⁶ The risk of pathogen contamination of HPL increases with the platelet pool size, so researchers used hollow fiber anion exchange chromatography to remove resistant and untested blood-borne pathogens without affecting the capacity of HPL-supplemented growth media to support ex-vivo cell expansion.¹⁴

Next-generation sequencing (NGS) could also potentially be used as a screening tool to look for virus contamination in biological products. It has shown tremendous progress in application in recent years, but it presents several issues in implementation.¹⁷ Its sensitivity has been called into question as well as its detection of non-relevant virus specific sequences. These are sequences that may look like virus-specific sequences but are simply remnants of a previous virus infection or incomplete virus sequences. These junk sequences must be filtered out so that, when comparing a cell line to the sequence database, the data returned is only what is relevant to the risk assessment. The other aspect is standardization of the technology and the bioinformatics. From a regulatory perspective, this is still at an early stage, and if it is going to be used for the release of a biopharmaceutical product, there must be some standardization of those procedures. The FDA, along with the biotech industry, has established an interest group to evaluate and further develop NGS technologies.¹⁷ While it may take some time to mature, NGS could be a suitable way to complete a rapid characterization of a cell line to provide some assurance it is safe before it goes into the patient.

Overall, ATMPs represent a class of medicines with a higher virus risk profile that need to be effective controlled. The most effective key to minimizing the potential for virus contamination is risk management. Understand where the risks originate from, and when a risk is identified, implement steps to reduce that risk by implementing effective sourcing and a risk-based approach to QC testing. The potential for risk reduction at this stage should not be underestimated.

1. REGULATION (EC) No 1394/2007 OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL on advanced therapy medicinal products and amending Directive 2001/83/EC and Regulation (EC) No 726/2004, EC, Editor. 13 November 2007.
2. US FDA: Regulatory Considerations for Human Cells, Tissues, and Cellular and TissueBased Products: Minimal Manipulation and Homologous Use, D.o.H.a.H. Services, 2017.
3. Hanna, E., et al., Advanced therapy medicinal products: current and future perspectives. J Mark Access Health Policy, 2016. 4.
4. ICH, Q5A Viral Safety Evaluation of Biotechnology Products Derived From Cell Lines of Human or Animal Origin. 1998.
5. Garnick, R.L., Experience with viral contamination in cell culture. Dev Biol Stand, 1996. 88: p. 49-56.
6. Garnick, R.L., Raw materials as a source of contamination in large-scale cell culture. Dev Biol Stand, 1998. 93: p. 21-9.
7. Besselsen, D.G., et al., Identification of novel murine parvovirus strains by epidemiological analysis of naturally infected mice. J Gen Virol, 2006. 87(Pt 6): p. 1543-56.
8. Oehmig, A., et al., Identification of a calicivirus isolate of unknown origin. J Gen Virol, 2003. 84(Pt 10): p. 2837-45.
9. Qiu, Y., et al., Identification and quantitation of Vesivirus 2117 particles in bioreactor fluids from infected Chinese hamster ovary cell cultures. Biotechnol Bioeng, 2013. 110(5): p. 1342-53.
10. Nei, M., F. Tajima, and Y. Tateno, Accuracy of estimated phylogenetic trees from molecular data. II. Gene frequency data. J Mol Evol, 1983. 19(2): p. 153-70.
11. Martella, V., et al., Detection and Full-Length Genome Characterization of Novel Canine Vesiviruses. Emerg Infect Dis, 2015. 21(8): p. 1433-6.
12. Renshaw, R.W., et al., Characterization of a Vesivirus Associated with an Outbreak of Acute Hemorrhagic Gastroenteritis in Domestic Dogs. J Clin Microbiol, 2018. 56(5).
13. Waytes, A.T., et al., A safer plasma supply from remunerated donors--"The Immuno/Community Bio-Resources experiment". Dev Biol (Basel), 2000. 102: p. 37-51.
14. Burnouf-Radosevich, M., [Viral safety of intravenous immunoglobulins G for therapeutic use]. Transfus Clin Biol, 1995. 2(3): p. 167-79.
15. CPMP, Note for Guidance on the use of bovine serum in the manufacture of the human biological medicinal products, B.W. Party, Editor. 2003, CPMP.
16. Kao, Y.C., et al., Removal process of prion and parvovirus from human platelet lysates used as clinical-grade supplement for ex vivo cell expansion. Cytotherapy, 2016. 18(7): p. 911-24.
17. Khan, A.S., et al., Advanced Virus Detection Technologies Interest Group (AVDTIG): Efforts on High Throughput Sequencing (HTS) for Virus Detection. PDA J Pharm Sci Technol, 2016. 70(6): p. 591-595.

Andy Bailey specialised in Virology serving for 9 years at the MRC Virology Unit in Glasgow. In 1995, he moved to the industry sector, initially as Director of Virus Validation services with Q-One Biotech Ltd, and later at the Global Pathogen Safety group of Baxter Healthcare in Vienna, Austria. Over the last 24 years Andy has been actively involved in the virus and prion safety field, presenting at numerous regulatory agencies either in support of products or as an invited speaker at expert workshops. In 2005 he founded ViruSure with the goal of providing a high quality science based testing service to the biopharmaceutical industry.