Filtration: New challenges met by traditional science
Filtration is found everywhere. From simple preparative lab-scale applications to quality-critical controls in large-scale GMP production. It can be used to remove life-threatening contaminants and help to safeguard patient safety, or simply to help another process to run effectively. It is often the first choice to simply separate the wanted from the unwanted, typically based on size. What is unwanted can range from macroscopic solid contaminants to bacteria, viruses and complex molecules. Each contaminant has its own impact on the effectiveness of the process or the safety or efficacy of the final drug. Filters are typically easy to use, often just fluid in - fluid out. They are therefore easy to add into a wide range of processes and procedures. But this simplicity, and the reliance and trust built over generations of use by the industry, does not mean that filters are unimpacted by the changing demands of the many processes that depend upon them.
Figure 1. Liquid filtration operations in a typical mAb downstream process. B = buffer and media filtration.
Filtration has evolved decade after decade to meet new challenges. New processes, new economic imperatives and new regulations have been supported by new filters. As the industry continues to discover and deliver ways to improve health, filtration keeps pace with its own innovation.
Intensification of filtration
Intensification means doing more with less. At the most basic level, this may be as simple as using smaller filters with higher throughputs and flowrates. Some of the most critical filters in use today can exhibit performance characteristics that are five, if not ten times better than their predecessors. But there are other impacts too. Looking beyond individual unit operations, perhaps the largest shift in this generation of manufacturing is the adoption of single-use technology. Counter to other single-use scenarios, in the context of biomanufacturing, adopting a single-use manufacturing philosophy has reduced non-value adding operations such as eliminating cleaning, set-up and batch turnaround times as well as carbon emissions. These in turn allow facilities to manufacture more, and often with reduced environmental impact. Filtration was actually one of the first technologies to support single-use processing in critical applications. Most product contact filters have always been single-use, but the stainless-steel housings and connecting pipework were not. These have been replaced with capsules integrated into single-use flow kits in most new applications.
Manually operated single-use systems have progressively been automated and now many filtration operations are either fully integrated into adjacent automated single-use operations or are supported by fully automated turnkey solutions that take the pain out of operations such as filter flushing and integrity testing. Critically all these advances work effectively to intensify operations and simultaneously control process risk.
Continual progress
Other intensification opportunities have been explored and are now routinely found in new processes. For example, cell cultures today can be more than ten-fold more productive per liter than first-generation monoclonal antibody processes. This is great for productivity and for reducing the cost of manufacture, but the increased cell density, along with potential increases in other contaminants such as host cell proteins (HCP) or host-cell DNA, has an impact on the filters that help remove it ready for subsequent purification. Filtration solutions for harvest and clarification have evolved to meet these challenges, both in terms of performance and format, such as single-use that is now the go-to choice for upstream processing.
Other manufacturing paradigms have continued to evolve but are yet to be routinely adopted. Fully continuous processes for example are rare. But processes that are intensified by development of solutions on this pathway are common. Fed batch cell culture and perfusion cell culture, especially earlier in the seed-train, have increased yields even further. Filtration here has again evolved to support. Hollow fiber cartridges designed for tangential flow filtration (TFF) have become the primary technology to support perfusion processes and gas and vent filters have evolved to support high density single-use cell cultures.
Moving downstream, traditional recirculating ultrafiltration processes using TFF for diafiltration or concentration can be replaced, or supported by recent innovations in in-line single-pass TFF (SPTFF) solutions (fig 2).
Figure 2. Cadence™ inline concentrator modules used for single-pass TFF operations.
This element of continuity in a process can reduce the shear from recirculation to safeguard quality and increase yield. The relative simplicity of SPTFF makes it easy to add into a process, so opportunities such as a pre-capture concentration step can be considered and can reduce process times and support upstream continuity. Moving further downstream, approximately one third of approved mAbs are now formulated at concentrations of greater than 100 g/L (fig 3).
Figure 3. Approved mAbs with high concentration formulation
Whether this is driven by the need to support sub-cutaneous injections, or simply decreasing storage volumes or flexibility within the drug product formulation, this target impacts the concentration step directly. Higher concentrations, sometimes exceeding 200 g/L push traditional recirculating TFF further than before and can lead to long process times, and degradation of the product if not carefully managed. The innovation in SPTFF can help achieve 200 g/L while also minimizing shear and reducing aggregation that impacts quality and yield and greater simplicity.
Virus filtration has also evolved to deliver both traditional intensification with better throughputs and to support continuous processing. Robustly removing small viruses and simultaneously transmitting large proteins is already a filtration challenge at the edge of filter performance. This virus retention performance can also be reduced by process interruptions – planned or otherwise – that allow for the back-diffusion of retained viruses and transient reductions in retention. Some modern virus filters, including the Pegasus™ Prime range are validated to show they are not impacted by such interruptions and robustly reduce the viral load with very high transmission of mAb. These performance characteristics support both traditional batch processes and the long process times common in continuous filtration without increases in process risk.
Finally bulk sterile filtration and filtration immediately prior to final filling is also not the same today as it was not so long ago. The higher average process formulations that are supported by innovation in TFF are also encountered during these quality critical operations. First, second or even third generation filters can struggle with the increased viscosity and any residual co-concentrated contaminants and aggregates. This results in larger than desired filters and increased product loses associated with filter size. For high concentration products, loses are costly and every drop counts. Modern sterilizing grade filters, such as Supor™ Prime range of filters (fig 4), can show 3 to 5-fold increases in throughput compared to the next best performing choices driving opportunities for smaller filters, simpler systems and increased yields.
Figure 4. Supor™ Prime sterilizing grade filters designed for high-throughput in high-concentration applications.
Simply put, innovation in filtration, while often not occupying the greatest mindshare has kept up with, and even enabled, innovations in drug manufacture.
More than innovation in mAb
When we think about innovation in medicine, most will cite new modalities such as gene therapies, mRNA vaccines and cell therapy. Each of these provide their own unique manufacturing challenges and this often impacts the filters used in their production. Some push the very concept of size-based separation of product from contaminant to the limit (fig 5). Lentiviral vectors used in some gene therapies for example are typically 0.12 µm in size. Compared to potential bacterial contaminants that can be as small as 0.2 µm, this similarity can lead to considerable yield loses as the vectors are retained during sterilizing grade filtration.
Figure 5. Molecule size of mAbs and novel therapeutics
In contrast, the smaller adeno-associated virus (AAV) does not suffer the same yield loses when modern filters are used. But if we risk assess the potential contaminants, it is not just about removing bacteria. Potential viral contamination originating from cell culture media or through adventitious contamination during processing needs consideration. Where mAb processes can deploy 20 nm filters towards the end of the manufacturing process to control this risk, AAV is very close in size to the panel of model viruses used to define the performance of 20 nm filters and cannot be transmitted by these filters. Nevertheless, virus filters can still be used to control a risk of contamination. Larger-pore virus filters, such as those rated as 50 nm, such as Ultipor VF DV50 virus filters, can reliably remove larger viral contaminants including retroviruses. Some can also transmit the smaller AAV with a minimal impact on yield.
This partial control for viral contaminants can be partnered with upstream controls to reduce the risk of small virus contaminants. This can include the 20 nm virus filtration of the cell culture media when this contains components of animal origin or that carry intrinsic risks of virus contamination. Filtration can be coupled with complementary controls including closed processing, enabled by single-use technology and including filtration. Together these can be accepted as a robust contamination control strategy aligned with regulatory guidance such as ICHQ5A (R2) Guideline on viral safety evaluation of biotechnology products derived from cell lines of human or animal origin. But best practice, followed by clear regulatory guidance specific to each new modality takes time to develop. Industry pathfinders work closely with filtration vendors to select, optimize or develop new filters that help overcome the new process challenges.
Filter choices can impact more than just process efficiency. Careful filter selection can be required to maintain critical quality attributes of the drug. Liposomal formulations for example have been shown as a risk factor for bacterial penetration of sterilizing grade filters. But not all filters are equally affected and making the right filter choice early in process development can help overcome these challenges and maintain sterility assurance. A poor choice can lead to significant pain and delays late in the development process when this is discovered during filter validation. Similarly, the size distribution of lipid nanoparticles (LNP) can be altered during sub-optimal filtration delivery, but again the right filter, coupled with the availability of the right expertise on hand to help with selection, can support these new therapeutic developments.
There’s always time to think sustainability
From being just a buzzword not so long ago, sustainability is now a fundamental part of process development and can often dominate process choices. The challenge is global and touches every industry, every process and every person. But it doesn’t have to wait for a radical change in technology to be actionable. An intensified process is likely to be more sustainable. We’ve already highlighted some of the benefits of smaller, higher performing filters, but the impact can snowball. From the perspective of Scope 3 emissions, those that come from suppliers, smaller filters consume less plastic when they are made. They use less water, need less packaging and weigh less when shipped. In use, as may be measured by Scope 1 and Scope 2 emissions, they consume less buffer. They can be integrated into more compact systems using less electricity, also built with less raw material. Process intensification driven by commercial goals directly delivers improvements in process sustainability.
Recent confirmation that more per- and polyfluoroalkyl substances (PFAS), sometimes referred to as forever chemicals, are damaging to health has shown that restrictions in their use and disposal are necessary. Materials containing, or made using these compounds can be found throughout the biomanufacturing industry and the impact of future restrictions has been discussed ever since the proposal for a ban was raised. Fortuitously filters are available with PFAS-free composition such as those using polyethersulfone (PES) membranes for sterilizing grade filtration and virus filtration. Conversely, sterilizing grade air filters are commonly manufactured from PFAS. Processes often need the hydrophobic characteristics of PTFE and PVDF membranes to deliver the robust performance required. The industry is actively innovating in this space however well-informed risk assessments are likely to highlight PVDF as a relatively safe choice. Existing improvements and controls in manufacture, a very low risk in use, and good controls available at the point of disposal to safeguard health may be sufficient for an exemption in this critical application within a critical industry. However, regardless of specific decisions or guidance, the industry, and the filters it needs will continue to adapt as regulations and materials technology changes.
Filter expertise: A helping hand for future success
It is easy to overlook filters. Often sub-optimal choices can go unnoticed, with any impact on a process being seen as natural and unavoidable or just not visible. That’s okay up to a point. But as processes evolve, and as new challenges are highlighted by growth in newer modalities, the impact of filter choices will only grow. Having access to expertise can accelerate the journey to the right solution and that can open the door to other improvements. Making fewer compromises and integrating filtration with other operations can drive intensification opportunities, support sustainability goals and deliver long-term process security.
So, when needed, seek a helping hand in making choices now that will yield rewards throughout the product lifecycle. Cytiva are here to help.
CY50845