Who has not already found a hair in a mushroom fricassee at the restaurant?
It spoils our appetite and makes us feel like never going back to that restaurant. For the hair, the solution is quite easy, a hat... but what about the things that we cannot see like viruses.
In 2010, two porcine viruses (Porcine Circovirus 1 and 2) were detected in two oral vaccines against rotavirus gastroenteritis (disease) (Victoria et al., 2010). The commercialization of those two vaccines was immediately put on hold by the authorities. After a risk analysis and different modifications of the vaccine manufacturing process, the authorities re-allowed the commercialization of the vaccine. Why? Because the risk of complication from swine viruses was found to be less dangerous than the rotavirus infections that kill 527,000 people annually among children aged <5 years. The pharmaceutical companies and the authorities are now working together to limit the risks associated with unknown viruses. The scientists estimate that there are nearly 1032 individual viruses on Earth capable of infecting humans.
Following a product’s release on the market, there are several follow-up measures to monitor the quality of medicine. A new part of this monitoring is the viral risk assessment. The ICH Q9 guidance provides principles and examples of tools for quality risk management. The safety of the medicine for the patient is the priority but risk evaluation should be based on scientific knowledge. The risk assessment needs to be controlled, communicated, and reviewed periodically. The collaboration between stakeholders such as pharmaceutical companies, authorities, academic and scientific institutions is crucial in this process as we are looking for something nobody knows about. This process is typically divided into three parts, the risk assessment, the risk control, and the risk review. A model for quality risk assessment is outlined in Figure 1 (extract from ICH Q9).
Figure 1. Overview of a quality risk assessment process
The first part of this assessment is the identification of hazards potentially leading to a viral risk of contamination. These aspects include manufacturing, cell lines, media, factories, suppliers, operators, etc. Who would have thought that a mouse virus would end up in a vaccine eluate without a murine element being used in the manufacture of that eluate? In 2011, Moody et al., detect the Mouse Minute Virus (MMV) in the Chinese Hamster Ovary (CHO) cell line, the same cell used to produce biological drugs. How did this happen? In a primary supplier’s warehouse, a mouse contaminated by the MMV virus urinated on a sugar bag used to produce CHO growing media. The quality of the risk assessment is principally based on risk identification.
Based on risk identification a list of potential viruses that can be present in the medicine was edited. The second step of the risk assessment or risk analysis step is the estimation of the risk associated with the identified hazards. With the collaboration of scientists, some viruses can be easily removed from the assessment. For example, if the raw material comes from New Zealand and the virus has only been described in one region of Africa, the risk of finding the virus is almost 0. Also, some components of the product may cause the virus to be inactivated. Increasing bulk temperature for inactivated proteins present in the bulk may also cause the inactivation of some viruses.
Finally, after various analyses, a quantitative estimation of risk or a qualitative description of a range of risk can be edited. Different method like FMEA (Failure Mode Effects Analysis) model is used to perform a quantitative estimation. For each remaining virus, a risk score is attributed and it is used to define the priority and the type of actions that need to be put in place.
This step is about decision-making: Do we accept a potential risk? The process can be changed; a new QC testing, like a PCR, can be introduced. But PCR assays are virus-specific and not useful for board detection of novel viruses. Moreover, a modification in the process can induce new risks. To avoid the risk of prion contamination (which can induce the Creutzfelt-Jakob disease) with the bovine serum, a porcine serum can be used with the risk of introducing the PCV virus in the process. The ICH Q5A gives recommendations to pharmaceutical companies to study the different steps in the vaccine process allowing to eliminate the virus, like storage conditions or inactivation. But this is not possible with all vaccines, like live attenuated vaccines, where the vaccination method is based on the replication of the viruses present in the vaccine (Measles, Mumps, and Rubella vaccination strategy). According to the ICH Q9, taking risks is necessary but those need to be controlled and re-evaluated periodically with scientific work and new technologies.
The results and the output of the risk assessment need to be appropriately communicated and documented as required by GMP. Unfortunately in this example, there was no official communication between authorities and pharmaceutical companies. Why? Because such a risk might frighten an uninformed audience and cause more concern than necessary. The anti-vaccine movements could use this risk analysis against the vaccination program. It is important to keep in mind that the people dealing with these matters at the regulatory level are scientists who can not only mitigate the risk but also give clues to limit the risk. The main role of the authorities is to ensure the safety of the vaccine and not the financial interest of pharmaceutical companies.
Viral Risk & New Technology
Why not use new technologies to detect viral signatures in the products? Next-generation sequencing (NGS) is non-Sanger based high-throughput DNA sequencing, using the concept of sequencing millions of DNA fragments at the same time (see figure 2 _ Bunnik and Le Roch, 2013). You can sequence more than 45 human genomes in a single day while the older method took 13 years for one human genome. There are several different methods for NGS and each method differs by the chemistry technology used or the size of the DNA fragment they sequence. Despite the potential of this method, the scientific community is facing a new challenge. Indeed, NGS can detect all the nucleic acids present in a sample but does not distinguish a virus from a sequence that looks like a virus. Also, given its sensitivity, the method could detect minute amounts of DNA that could come from the preparation or reagents used for the experiment. It is important to remember that this technology was used to detect the PCV virus in the vaccine that launched all the discussion about viral safety (Victoria et al, 2010).
Figure 2. Comparison between Sanger sequencing and NGS technologies. Sanger sequencing is limited to determining the order of one fragment of DNA per reaction, up to a maximum length of
In a nutshell, the authorities know how difficult it can be to evaluate something we do not know and they know that no company is willing to find a hair on the plate it proposes.
Blog by Celine Powis de Tenbossche
Check out our COVID19-blog: Facts vs Fiction
And our COVID19 blog: What are the different types of vaccines in development?