Evolving UF/DF Capabilities

Ultrafiltration/diafiltration is a core downstream bioprocessing technology that continues to evolve. “Batch ultrafiltration/diafiltration (UF/DF) is a very economical, high yield, and robust separation process based on size exclusion that finds application for a wide range of biotherapeutics,” notes Andrew Bulpin, head of process solutions for MilliporeSigma. UF involves separation of components on the basis of molecular weight/size. It is a pressure-driven process in which soluble macromolecules (e.g., proteins as products) are retained while small molecular-weight particles (e.g., salts, amino acids, mono- or disaccharides), and fluid/water pass through the membrane as waste, according to Mayank Dadhwal, filtration global product manager at GE Healthcare Life Sciences. DF is used to exchange buffer solutions.

Batch UF/DF is used in nearly every biotherapeutic process for final formulation, notes Bulpin. DF is used to exchange the buffer to one that is preferred for storage stability and ease of administration; UF is employed to concentrate the therapeutic to its final formulation concentration. For high-concentration products intended for administration via subcutaneous injection, UF/DF studies are often conducted to understand the relevant viscosity limitation and determine the maximum concentration that can be achieved, according to Marc Jenke, senior product manager for crossflow filtration systems at Sartorius Stedim Biotech.

UF/DF is also used in some cases for buffer exchange and bioprocess fluid concentration prior to or in between different chromatography steps during downstream processing. Vaccine purification using batch UF/DF allows retention of the cell, virus, or glycoprotein vaccine by the UF membrane, while smaller impurities pass through the membrane. In addition, batch UF/DF is used to remove unreacted reagents from therapeutic conjugates such as antibody-drug conjugates (ADCs), PEGylated proteins, carbohydrate conjugates, and others.

For instance, biopharmaceutical contract development and manufacturing organization CMC Biologics typically uses UF/DF for final buffer exchange and concentration, according to Magnus Schoeder, director of downstream process development with the company. On occasion, UF/DF is used to concentrate clarified harvest material or as an intermediate buffer exchange step prior to chromatography.

Based on TFF

The key to UF/DF is the use of tangential flow filtration (TFF), also referred to as crossflow filtration. Unlike in traditional filtration where the entire fluid stream passes directly through the filter, in TFF part of the feed stream passes across the membrane surface, creating back transport of retained solutes from the filter surface. The result is prevention of the build-up of solids and clogging of the membrane. “TFF commonly reaches steady-state flow to allow extended operation with high throughputs,” says Bulpin.

UF/DF systems generally consist of a retentate tank to store in-process therapeutic, a membrane assembly, a feed pump to provide tangential flow, and pressure to drive the solvent through the membrane. Equipment for diafiltrate buffer addition, process monitoring and control, product recovery, integrity testing, cleaning, etc., are also needed to provide cGMP batch records, ensure consistent product, achieve high process yields, and permit reuse for economical operation, according to Bulpin.

Crossflow filtration systems are the gold standard for performance of trials to determine the ideal final formulation of high-viscosity products, according to Jenke. “These systems are easy to use and are available for both development and commercial production, allowing straightforward transfer of processes from one department to the other. In addition, there are established service and training concepts, operators know exactly what to do, and release of final product is facilitated because quality personnel can properly interpret the data,” he explains.

TFF systems for UF/DF can be either stainless-steel or single-use and are typically configured with hollow fibers or cassettes. “As the biopharmaceutical market expands, the need for greater facility utilization increases because multiple drugs must be released to the market with very short turnaround times. UF/DF system design is therefore crucial to facility fit for both multi-product and single-product facilities,” observes Dadhwal. It can be difficult to design a one-size-fits-all UF/DF system, however, given that a wide range of process titers and drug substance concentrations are processed in multiproduct facilities.

Employing two stainless-steel UF/DF systems to achieve high drug substance concentrations can significantly increase both cost and process complexity, he adds. “Automated single-use TFF systems with disposable flowpaths and reduced hold-up volumes that can be used with highly concentrated fluids not only increase facility flexibility, they help minimize the risk for cross-contamination and allow faster changeover times due to reduced need for cleaning and greater process consistency due to reduction in the opportunities for operator error,” Dadhwal says.

CMC Biologics uses plate and frame systems with various pumps, load cells, online detectors, and control loops, either in the form of conventional clean-in-place/steam-in-place capable skids or single-use setups with disposable product flowpaths, according to Schoeder.

Technologies other than crossflow filtration are used in some cases for UF/DF processes. At the bench scale, for instance, size-exclusion chromatography (SEC) is commonly used to achieve size-based separations, according to Bulpin. It is impractical for larger-scale processes, however, because it requires dilute protein concentrations and low flow rates. Bulpin adds that crystallization and freeze-drying have been applied for concentration, but they tend to be more expensive, less widely applicable, and require extensive development. Large-scale countercurrent dialysis is possible for very sensitive molecules, according to Schoeder, but has limitations around scalability and is not frequently used.

One alternative that is gaining attention for high-concentration formulations is single-pass TFF (SPTFF). It can relieve the working volume constraints of conventional systems, according to Dadhwal, but is an emerging concept. As a result, conventional TFF systems are still the most accepted solutions for UF/DF applications.

Challenges of changing feed streams

Changes in upstream production processes have led to higher titers and increased batch sizes for UF/DF processes. Final formulations with higher concentration targets are now preferred because they allow for ease of administration. These changes have led to the need for larger UF/DF systems and closer attention paid to scale-up issues such as mixing, pumping damage, and holdup volumes, according to Bulpin. “Higher final concentrations have also led to a greater understanding of solute partitioning such as Donnan exclusion, where charged excipients interact with higher concentrations of charged therapeutic products to alter the final buffer after diafiltration,” he notes.

Growing interest in more potent molecules such as ADCs has also had an impact. “For these biomolecules, avoiding exposure to operators and minimizing liquid waste are important,” Dadhwal comments. In addition, crossflow filtration solutions for bioconjugates must also be compatible with various chemicals, including solvents, according to Jenke. “Development of UF/DF production systems with materials that provide the necessary separation/concentration performance while also being compatible with various chemicals has been a focus at Sartorius Stedim Biotech,” he remarks.

Overall, there has been increasing interest in implementation of single-use TFF systems for UF/DF. Similarly, Dadhwal notes that cell- and gene-therapies, which are entering the commercialization stage, benefit from single-use UF/DF systems because they are produced on a much smaller scale than blockbuster monoclonal antibodies. These therapies also may have requirements such as sterility of the liquid path due to the size of the virus or yield concerns when viruses are sterile filtered. “In both cases, completely closed operation can be achieved with single-use systems,” he says.

At CMC Biologics, one challenge has been the need to balance product quality, processing time, facility fit, and product recovery for high-concentration (200 g/L and higher) formulations for subcutaneous applications. New challenges, according to Schoeder, include higher viscosity solutions with less-efficient mass-transfer and mixing challenges, designing manufacturing systems that allow for processing of high initial volumes/high fluxes and yield low volumes/low fluxes at the end, with subsequent low dead-volume product recovery.

Solutions for evolving needs

These challenges have led to advances in TFF technology for UF/DF applications. In particular, the introduction of increasingly sophisticated single-use solutions has had a significant impact, according to Dadhwal. “Novel TFF membranes with modified channel geometries and screen designs for improved handling of high-viscosity scenarios have been appreciated at CMC Biologics,” says Schoeder. These new modules optimize the design and manufacture of the feed screen channel to enable high back-transport of solutes from the membrane while also reducing the pressure drop along the feed channel, according to Bulpin.

Schoeder adds that the introduction of variable path-length, flow-through detectors for online monitoring of concentration throughout the entire process has also been important for improving UF/DF performance.

Bulpin sees SPTFF as an important advance that is enabling continuous concentration of larger batch sizes. “While batch systems require repeated passes of the biotherapeutic solution through filter assemblies to achieve a target final concentration, SPTFF operates existing filters in a different configuration to achieve target concentrations continuously in one pass. They can also be employed at various stages of a process to eliminate tank bottlenecks and improve the efficiencies of other unit operations,” he explains.

The rapid growth of highly potent ADCs has also challenged UF/DF system manufacturers to develop self-contained, holderless filter capsules that can be operated in a closed system to minimize operator exposure, according to Bulpin. “These pre-sterilized capsules are convenient for plug-and-play operation to improve the throughput in pilot plants and clinical manufacturing operations,” he notes.  

Sartorius Stedim Biotech has focused the development of multi-parallel, small-scale crossflow filtration systems with the potential to decrease the hurdles when going from upstream to downstream processing, according to Jenke. The company has a new high-throughput TFF system for parallel screening that can be used for the development of both upstream and downstream (e.g., UF/DF) processes that it will be bringing to market in the near future. The automated system consists of a 10-m² membrane with a 5-mL recirculating volume and up to 16 crossflow channels, allowing the use of small process volumes and speeding up R&D efforts. “Each channel represents a fully equipped and automated crossflow solution. This new system is ideal for process optimization, screening of agents with a relevance for upstream and downstream development, target selection, and determination of optimal buffer type and pH and conductivity values, and more,” Jenke asserts. “It also enables critical experiments to be performed earlier in product development.”

New manufacturing paradigms

Continuous processing is of great interest because it enables removal of intermediate tanks, increases productivity, and improves product quality, according to Bulpin. “Continuous processing is the new evolving concept in biomanufacturing, so shaping UF/DF systems to adapt to the continuous manufacturing scenario is the new challenge to be addressed,” asserts Dadhwal. He notes that automation of UF/DF processes is the answer, both at the R&D stage for optimized processes developed using a quality-by-design approach and at the pilot plant and commercial production stages combined with single-use technology for facilitation of continuous operations and effective process monitoring and data acquisition.

Specific examples of equipment that can enable continuous UF/DF processes include sterile UF capsules, which according to Bulpin, aid continuous UF by facilitating easy assembly of closed systems for extended bioburden-free processing times.

Beyond continuous processing, a key ongoing trend in the biopharmaceutical industry is the need to increase operational efficiency and lower the cost of manufacturing. One approach, according to Dadhwal, is to perform processes in controlled, non-classified spaces. “Taking this approach results in the need for more closed operations. In these instances, single-use technology provides an advantage with the possibility to use aseptic connectors and easily avoid bioburden issues with gamma-irradiated tubing assemblies,” he observes.

In addition to more single-use alternatives for UF/DF, Schoeder expects going forward that there will be increasing demand for more SPTFF solutions, including single-pass diafiltration membranes. Other areas with needed improvement include enhanced pilot- and production-scale UF/DF skids and improved automated process development and characterization systems for high-concentration formulation products.

There is also an ongoing need for new UF/DF filters with different pore size ratings, new materials of construction, new designs, and new ways of operating to achieve target separations for new applications, according to Bulpin. New processes can also require tighter filter specifications.

The use of design-of-experiment approaches and data analysis will play an important role in development of downstream manufacturing processes in the near future, according to Jenke. “There is only limited data currently available for UF/DF processes, despite the fact that it can have an impact on molecular stability and overall process efficiency. We see significant opportunity for future developments to improve the manufacturability of this essential downstream process by providing technology that is readily scalable and generates data that are readily analyzed,” he explains.

Article Details

BioPharm International
Volume 31, Number 11
January 2018
Pages: 24–27

Citation

When referring to this article, please cite it as C. Challener, “Evolving UF/DF Capabilities," BioPharm International 31 (1) 2018.