Photo: GSK Marburg

Safety Testing of Recombinant Protein Products Derived from Cell Lines of Human and Animal Origin

(last update: April 2014)

It was in 1982 that the World Health Organisation (WHO) introduced safety testing guidelines for the use of continuous cell lines employed as substrates for the production of inactivated viral vaccines. Since then, progress in the biomedical sciences and the exploitation of biotechnology has led to the development of new biological medicinal products at an unprecedented rate. In addition to the WHO the US Food & Drug Administration (FDA), Centre for Drugs Evaluation and Research (CDER, for recombinant proteins) and Centre for Biologics Evaluation and Research (CBER for vaccines, gene and cell therapies and blood products), the European Medicines Agency (EMA) and the Japanese Pharmaceutical and Medical Device Agency (PMDA) have prepared guidelines for the manufacture and quality control of biotechnology derived medicinal products. The regulation of biotechnology products, is a risk based approach taking into account risk-benefit assessments, sound scientific principles and the experience with these products in the last three decades. The safety testing guidelines issued by all these regulatory agencies have been complemented by guidelines produced during the International Conference on Harmonisation (ICH) process. They form the scientific and regulatory basis for ensuring the safety of new biological entities. In addition to guidelines for final medicinal products the EMA has issued a guideline for the viral safety evaluation of biotechnology investigational medicinal products (IMPs) with the aim of providing a harmonised approach to safety requirements within the EU for clinical trial materials.

Major concern: viral contamination

The major concern when using mammalian cell lines for production of a biotechnology product, is the risk of microbial (bacteria, fungi and mycoplasma) and viral contamination. Such contamination could have serious clinical consequences and can arise from the contamination of the cell substrate or by the introduction of adventitious agents during production.

To date however, biotechnology products derived from cell lines have not been implicated in the transmission of infectious agents. To ensure the continued safety of these products, the adoption of a sound testing program is essential. The strategy that has evolved since 1982 consists of three complimentary approaches:

  • selecting and testing cell lines and raw materials (including cell culture media and animal derived raw materials), for the absence of undesirable microbes and viruses which may be infectious and/or pathogenic for humans;
  • testing the product at appropriate stages during production for the absence of contaminating infectious viruses and microbial agents
  • assessing the capacity of the manufacturing process to inactivate or remove infectious viruses.

Confidence that infectious agents are absent from the final purified product, in most instances, cannot be derived solely from direct testing due to the inherent limitations of assays, and due to the fact that detection of low levels of contamination depends, for statistical reasons, on the size of the sample being tested. Testing the capacity of the purification scheme to remove or inactivate known or unknown viruses is therefore a critical element in the overall safety testing strategy.

Potential sources of viral contamination

Viral contamination can originate from the two following main sources.

  1. Contamination of the Cell Substrate. Viruses could be introduced into the cell substrate from several sources, including derivation from infected animals (e.g. rodent viruses in murine hybridomas and Chinese hamster cells) and the use of contaminated cell culture reagents, such as bovine viruses from animal sera or porcine viruses from trypsin. Rodent parvoviruses such as Minute Virus of Mice (MMV) may be introduced through the inclusion of contaminated components (glucose and amino acids) in the culture media. Endogenous viruses, such as retroviruses, pose a particular problem in assessing the safety of cell substrates, as they may be transmitted in the germ line, since the viral genome persists within the cell. These endogenous retroviruses may be expressed without deleterious effects on the cells, and could be infectious to human cells.
  2. Adventitious viruses introduced during the manufacturing process. The likely sources of contamination include the use of contaminated cell culture media, a breakdown in GMP allowing operator or other external contamination, the use of a contaminated excipient during formulation or other reagents used in the process, such as a monoclonal antibody affinity chromatography column.

The importance of the sourcing and the qualification of all raw materials used in the manufacturing process has become a key area for the biopharmaceutical industry, not just to exclude viral contamination, but also to address the concerns of other contaminants such as transmissible spongiform encephalopathy agents, such as BSE. Cell culture media which exclude all animal components are now widely used in the industry to reduce the potential contamination problems.

Master Cell Bank

Production of recombinant proteins starts with the derivation of a stable Master Cell Bank (MCB) that is produced under GMP conditions.
As all production lots will originate from MCB cells the testing should be very exhaustive and will include the following assays.

  1. Assays for cell line identity: isoenzyme analysis, DNA fingerprinting and karyology (for less well characterised cell lines)
  2. Assays to show absence of bacteria, fungi and mycoplasma: sterility assay; mycoplasma assay by culture or by PCR. For cell lines from species susceptible to mycobacterium infection an assay to detect mycobacterium may be appropriate.
  3. Assays for adventitious viruses using a) detector cell lines with cytopathic effect, haemadsorbtion and haemagglutination endpoints; b) embryonated eggs, suckling and adult mice with morbidity or mortality endpoints and transmission electron microscopy. Recently comparison studies have questioned the utility of continuing to use in vivo infectivity assays to detect adventitious contaminants.
  4. Transmission electron microscopy (TEM) on cell profiles to detect any endogenous retrovirus particles or adventitious agents.
  5. For human cells PCR assays to detect human viruses not readily detected by tissue culture such as the provirus of human retroviruses (HIV, HTLV), latent human herpes viruses (CMV, EBV, HHV 6,7 and 8), and human hepatitis viruses (hepatitis A, B and C). For rodent cell lines murine and hamster antibody product (MAP, HAP) to detect a wide range of rodent viruses.
  6. Infectivity and PCR assays to detect bovine and porcine viral contaminants that could have derived from the use of animal derived components during the history of the cell line. Even if the MCB has been derived to grow in animal origin free media it is recommended to screen for the absence of bovine and porcine viruses since the original parental cells would have been grown in animal protein containing media.
  7. Infectivity assays to detect retroviruses. If these infectivity assays and TEM are negative the MCB should be screened using a PCR based reverse transcriptase assay.
  8. Recent regulatory guidelines suggest manufacturers should also consider using new technologies to detect microbial and viral contaminants that are based on nucleic acid arrays or multiple parallel (deep) sequencing.
Working Cell Bank

Normally, the WCB originates after only a small number of passages from a well characterised MCB, so only limited testing is required:
Identity: isoenzymes or DNA fingerprinting. Karyology where appropriate.
Sterility and mycoplasma. Mycobacterium if appropriate.
In vitro assay to detect adventitious viruses.

Cells at the Limit of In Vitro Cell Age Used for Production (CAL) or End of Production Cells (EOPC)

Cells at the limit of in vitro cell age used for production are cells at the highest population doubling level (PDL) that will be claimed in a Marketing Authorisation (MA). End of Production cells (EOPC) are the cells at the end of the production process at the particular scale used to produce the IMP. The PDL of the EOPC will change during scale up during clinical development.
Either CAL or EOPC should be evaluated once for those endogenous viruses which may not have been detected in the MCB and WCB. Also by conducting adventitious virus assays on cells after maximum opportunity for viral amplification, shows that the production process is not prone to such contamination.
The EMA in its guidance on viral safety evaluation for IMPs suggests that for well characterised cell lines such CHO, BHK, NS0 and Sp2/0 that CAL testing does not need to be performed if testing is performed on the unpurified bulk harvest. At commercial scale production CAL cells should be examined prior to MA.
When CAL or EOPC are tested the following assays should be used:
Identity: isoenzymes or DNA fingerprinting. Karyology where appropriate.
Sterility and mycoplasma. Mycobacterium if appropriate.
In vitro and in vivo assays to detect adventitious viruses and Transmission Electron Microscopy.
Retrovirus infectivity assays including co-cultivation with rodent and human cells.
Detection of specific viruses of the species of the production cell line such as PCR assays for human viruses. For well characterised rodent cells, to save on the use of animals the MAP and HAP assays are not normally repeated. However an infectivity or PCR assay to detect MMV contamination should be included.

Bulk Harvest

A representative sample of the unprocessed bulk, removed from the bioreactor prior to further processing, represents one of the most suitable levels to detect adventitious viral contaminants with a high probability of detection. The testing that should be considered includes:
Sterility and mycoplasma
In vitro assay for adventitious viruses. If the cell line is a human cell line or not well characterised then an in vivo assay for adventitious viruses should be performed.
TEM on at least 3 bulk harvests to determine the virus load being expressed by the cell line.
If the cell line is susceptible to infection with MMV then an infectivity assay or PCR for MMV should be included.

Purified Bulk

The testing of the purified bulk will vary, depending upon the results of the cell bank and unprocessed bulk testing, however, the key assays to perform are:
Microbiological contaminants: Sterility, endotoxin
Assays should be included for process related contaminants such as Residual host cell DNA, host cell protein and bovine serum albumin if bovine serum has been used in production.

Final Filed Product

The tests to perform include:
Microbiological contaminants: Sterility
Endotoxin: LAL or Rabbit Pyrogen
Toxicity: Abnormal Toxicity or General Safety

Viral Clearance Studies

These studies are conducted to assess the capacity of the purification process to remove or inactivate viruses or other contaminants. The type and extent of viral clearance studies to be conducted, will depend on the product and stage of downstream process development and should be considered on a case by case basis. The results of the cell bank testing, in process testing, the nature of the culture medium used, the product and its application and the ability of the process to inactivate or remove viruses, are all factors which influence the design of the study. For products produced using well characterised cells such as CHO the viral clearance study would be performed using murine leukaemia virus (MLV) as a model for endogenous retroviruses and MMV as a relevant small stable virus. A minimum of 2 robust steps such as low pH inactivation or virus removing filters should be evaluated with 2 independent spiking runs for step with each virus.
When the final downstream process has been fixed a VC study using 4 model/relevant viruses should be performed and robustness of the individual steps should be evaluated.

Regulatory Guidelines

Please consult the Interesting Links pages and visit the websites of US FDA, EMA, Japanese PMDA and ICH for documents.


Martin Wisher, April 2014

©2017 ACTIP

Log in with your credentials

Forgot your details?