Presented by Horst Ruppach
Dr. Horst Ruppach is global director of viral clearance and virology at Charles River, Germany.
Ruppach’s presentation focused on two critical aspects in viral clearance studies required to demonstrate the viral inactivation or removal capacity of a biomanufacturing process. High-titer virus stock preparations are added to different process intermediates at laboratory-scale process steps to achieve a high viral load. The higher the virus load is at the start, the higher the reduction factors that can be determined. However, it is important to find a balance between a having a high viral load and keeping process performance unchanged by a virus spike. Adding virus stock solution can change a process step’s performance such that the laboratory-scale process step no longer reflects the manufacturing conditions, as suggested in guidelines on viral clearance studies.
Ruppach described an example that demonstrated the impact of virus stock preparations of different compositions on the flow rates and performance of virus filters. Keeping flow rates as close as possible to manufacturing conditions requires purified virus stock preparations and low spike ratios. But the downside is reduced viral loads and limited reduction factors, which can be demonstrated. The reduced load can be evaluated best using highly sensitive assays on product sample. The more sensitive the assay is, the more likely a high-reduction factor can be determined — even at low viral loads.
Ruppach stated that the best experimental set up for viral clearance studies should consider a reasonable and process step specific adapted virus spike with minimal impact to the process performance and application of highsensitivity assays for the relevant product fraction. This will best characterize the process step viral clearance capacity.
Use of prefilters in a virus filtration step has become an increasingly common strategy. A prefilter ensures better performance of a virus filter by removing residual impurities. More important, many proteins (e.g., antibodies) tend to aggregate and can block the filter. Ruppach detailed how this strategy is best addressed in viral clearance studies. He argued that the decoupled mode is the preferred procedure and should be applied if possible (Figure 1). In this protocol, feed is filtered through an adsorptive prefilter. Virus is spiked into that adsorptive prefiltered feed, and the resulting spiked feed is loaded onto the virus filter. The coupled mode (spiking the feed before prefiltration and running the prefilter in line with the virus filter) should be adopted only if the performance of the virus filter is affected significantly and cannot be resolved by changing spike ratio or optimizing the downscale protocol.
Ruppach described additional experiments that might be needed to determine the impact of a processspecific prefilter on viral clearance. Such approaches would be needed to differentiate removal by prefilter from removal by a virus filter. A rationale or risk assessment based on data from such additional experiments should be prepared that justify the coupled mode against the decoupled mode.