By Denis Kole
Virus safety is a key element in biopharmaceutical manufacturing. Inherently, animal cell culture involves risks related to virus safety. Chinese Hamster Ovary (CHO) cell lines are the most common production platform for expressing glycosylated proteins and monoclonal antibodies. Being from rodent origin, CHO cell lines may contain endogenous retroviruses or retrovirus-like particles. Additional virus safety risks come from the possibility of the cell culture being contaminated by adventitious viruses. As a consequence, the regulatory expectation for manufacturing platforms is to demonstrate the ability to remove or inactivate a wide range of viruses to minimize the impact of viruses on patient safety.
Virus inactivation (VI) is one of the key elements in providing virus safety. It targets enveloped viruses that may be present in the product feed, typically the eluate from the Protein A chromatography step. A widely accepted solution is the exposure to low pH conditions (e.g., pH 3 – 3.6) by acidifying the solution with acetic acid and maintained at that low pH for a specific length of time (typically 60 minutes).
There are various approaches for continuous VI. One approach that is being explored by the bioprocessing field relies on a plug flow contactor. In this concept the pH of the eluted monoclonal antibody is lowered in-line as it enters a tubular plug flow contactor. The length of the tubular plug flow contactor is chosen such that it provides the required residence time for the inactivation process. Challenges with the plug flow concept include scale-down validation, managing concentration gradients in the process, and managing process disturbances and upsets. In addition to this, forward processing often requires homogeneous process conditions.
Pall’s approach for continuous VI relies on repetitive batch inactivation instead. The failure mode and effects analysis (FMEA) approach taken when designing and developing Pall’s Cadence® VI system suggested that the product risks and process risks are not significantly different with this approach than in the batch equivalent. Pall’s continuous VI concept relies on the same approach as in the well-known batch VI process and as a result the critical process parameters (CPPs) are also identical. This approach eliminates many uncertainties in scale-down validation, and it provides homogeneous process conditions for forward processing. Pall’s Cadence® VI system has the potential to be used for both, low pH VI as well as solvent/detergent VI.
Virus retention filtration is the last step in providing virus safety in the downstream processing of monoclonal antibodies. It serves to remove viruses by size and is specifically designed to target small viruses. The virus filtration step is generally assumed to have no impact on any other quality attribute and hence the critical process parameters are exclusively related to the virus removal.
In a fully integrated continuous downstream processing platform, the virus retention filtration is performed at constant flow, whereas the common procedure for batch virus filtration is to run the process at constant pressure. Furthermore, unlike batch processing where the output from the previous step is pooled prior to virus filtration, it is possible that the feed to the virus filter in a continuous process has variabilities in protein concentration/pH/conductivity etc. These variabilities are a consequence of the nature of the prior step (eg. chromatography column eluate). If a surge vessel is used between these unit operations, that could be sized to provide a homogenous feed to the virus filtration step. However, if a surge vessel is not employed, the variation in protein concentration/pH/conductivity on virus removal performance needs to be risk assessed. If prior knowledge is not available, these variables may need to be included in a virus clearance study.
During the process, the virus retention filter accumulates viruses and some protein inside the tortuous flow path. Consequently, one would normally see a decay in flux during the batch virus filtration step. This flux decay is a result of a gradual increase in resistance of the filter. In continuous processing, this will translate into an increase in pressure differential across the filter. Provided that the virus validation studies are carried out such that both the minimum and maximum pressures are studied, the data will be representative for the continuous process as well. The critical process parameters for continuous virus filtration therefore would translate into the maximum volume to be processed (similar as in batch virus filtration) and the maximum back pressure that can be accepted (equivalent to the flux decay in the batch virus filtration process). These CPPs can be investigated using constant pressure and – provided that the study design covers the entire operating range of the virus filtration step – the results apply to a constant flow process.
An additional factor that needs to be considered in selecting an adequate virus filter and designing the virus filtration step is its capability of handling process interruptions. This is something that may have to be considered for every step in the continuous downstream processing platform, but it may be more critical for virus filtration. The reason for this is that not all virus retention filters tolerate process interruptions equally well. In some filters, release of the transmembrane pressure and/or process interruptions can affect the virus removal capability of the filter as a result of so-called back-diffusion.
Denis Kole is Principal Scientist, Scientific and Laboratory Services for Pall Biotech.