April 15, 2015 (Vol. 35, No. 8)
Angelo DePalma Ph.D. Writer GEN
Single-Use, Hollow Fibers Cannot Transition from Small-Scale Processes to Large-Scale Production Fast Enough
Hollow fiber filtration has been a mainstream methodology for years. Improved efficiency and its role in continuous processing ensure that the technique’s popularity will continue to grow.
When single-use hollow fiber filtration membranes entered bioprocessing, they were adopted first for relatively small-scale processes, specifically Phase II clinical trials and earlier. Now, according to Mike Dango, product manager for filtration at GE Healthcare, customers are requesting hollow fibers for Phase III and commercial production. “End-users are pushing the scale of single-use processing beyond the point which we originally felt, as suppliers, that the economics would be viable.”
Scaling up single-use filtration systems entails a lot more than that for, say, tubing or bags. Most filtration membranes come prewetted, presterilized, and ready to use with the appropriate cleanliness specification out of the box. Scale affects the ability to develop, qualify, and validate flush schemes and sterilization protocols for filters. By contrast, many disposable fluid-management products outside of separation are manufactured dry and gamma irradiated in bulk.
“Scaling up for handling and desirable usability attributes is not as simple for filters,” Dango adds.
Several factors have influenced the evolution toward higher-capacity single-use hollow-fiber filtration membranes. Increasingly, customers recognize that the benefit of compressing time-to-market by eliminating or reducing nonvalue-added operations during bioprocessing applies at all scales. Moreover, many products entering large-scale production today “grew up” with single-use processing. It becomes easier, from an operational perspective, on the basis of successful experience with these products, to keep that practice going.
“We’re in a time now where plants are being built around the concept of single use, fast changeovers, and limited nonvalue-added activity,” Dango says. “A lot of effort has gone into producing filtration membranes and filtration lines, even at large scale, that are suited to that bioprocessing philosophy.”
Predictability of Bioprocess Scale Up
Adoption of this philosophy leads to the question of predictability on scale up. This question pertains to any scale-up project, whether it is conducted in fixed-tank or disposable process equipment. Biopharmaceutical manufacturers need assurance that what worked at 200 L will work at 2,000 L. Vendors must therefore take special care to guarantee a clear, trouble-free scale-up path.
Predictability ties in with validation, an issue on which FDA has been constantly vigilant. The FDA requires that biopharmaceutical manufacturers understand the variables that affect their process and a product’s critical quality attributes. “And they recognize that the variability within our own products may contribute to that,” Dango relates. “Customers increasingly ask that we understand that variability, and provide insight on how that might affect their processes.”
GE Healthcare has for years provided technical expertise to help customers predict the impact of scaling up. Customers now request hollow fiber products from qualification lots that represent that product line’s typical variability, for purposes of adding another layer of predictability for scale-up performance.
Put simply, this involves examining the effects of filtration membrane product quality outliers compared with the “average” hollow-fiber cartridge. The belief is that if products at the upper and lower extremes of the performance continuum work, a typical product between those two points will perform as well. Some biomanufacturers are taking this a step further by performing this qualification retrospectively, even for legacy processes.
Legacy normal-flow filtration systems such as nylon and polyvinilydine difluoride persist in bioprocessing due to validation obstacles. These products persist despite the higher flow rates, and thereby greater facility throughput and productivity, from more advanced membrane technologies such as polyethersulfone. Newer materials of construction are also less likely to give false integrity test results.
“Our industry’s understanding of what is often considered to be a relatively simple and common unit operation needs to improve,” says Nick Hutchinson, market development manager at Parker Hannifin. For example, better understanding of relationships between transmembrane pressure, capacity, throughput, and bacterial retention of filters around bioreactors “can considerable reduce the risk of contamination and increase right-first-time manufacturing.”
Fear of regulatory consequences in bioprocessing are not novel, but manufacturers often create problems where they do not exist. Troubles occur when manufacturers over-specify or provide more detail than is necessary in batch documentation. They may, for example, specify a filter of a specific media type, or go as far as to cite a part number, in which case switching out involves lengthy validation.
Manufacturers must also be aware of the point in the process where changes will occur.
“Companies often think change is a bigger hurdle than it actually is, but it does depend on the application,” Hutchinson allows. For noncritical applications, such as buffer filtration, they may be fine merely with specifying use of a “sterilizing grade filter,” and get by with a paper-based risk assessment with no validation testing whatsoever.
“Depending on the application, they might perform more testing to demonstrate equivalent performance, which would include a review of extractables and leachables. The most critical application will always be toward the end of the process, for example, during final filtration of a drug product,” Hutchinson cautions. “At that stage, switching over will be quite burdensome.”
The reason for the poor understanding of new technology deployment is historical. “People did not understand the consequences of such great detail,” Hutchinson notes. “They were trying to show regulators that they were in control of their process, and to do so they were happy to write in detailed specifications.”
“We have seen a trend over the past several years in bioprocess filtration toward continuous bioprocessing,” says David Serway, vice president of business development at Spectrum Laboratories.
Advantages of continuous bioprocessing include smaller facility footprint, lower capital investment, larger disposable needs, quicker time to market, and (potentially) higher yields.
“Membrane science is key to making this work,” Serway observes. “Perfusion technology is one key component for continuous biomanufacturing where single-use hollow fiber tangential flow filtration (TFF) is the engine that makes perfusion operate seamlessly. More and more companies are discovering the importance of TFF paired with perfusion bioreactors for continuous bioprocessing.”
The range of separations available through single-use TFF speaks for itself. Whether the desire is higher cell density, removal of metabolic waste, or antibody permeation using a microporous hollow fiber membrane, single-use TFF facilitates continuous operation with high sieving coefficients.
“Hollow fiber is one of several styles of TFF membranes that are available in a wide range of separation capabilities or molecular weight cutoff values,” Serway points out. “This means every perfusion process can be tailored to suit the application.”
Nathalie Pathier, global product manager for tangential flow filtration at Pall, views continuous concentration and single-pass TFF as enabling technologies for continuous processing. Single-pass TFF simplifies concentration by eliminating the need for a recirculation loop. Process feed enters the single-pass TFF module where separation occurs, and product at the targeted concentration leaves the retentate port.
The potential exists as well to combine this continuous concentration with chromatography. “More-concentrated processing has the benefit of enabling reduced column size and/or shorter loading times, and it works with either flow-through or affinity columns” Pathier notes. Another benefit is scalability.
According to Pathier, another trend is the movement toward automated single-use TFF, a technology that can provide increased process reliability, flexibility, and robustness—not to mention the usual benefits of single-use filtration. The benefits of presterilization, preconditioning, and the absence of cleaning or related validation steps are well known. With single-use TFF systems and modules, Pathier emphasizes, “operators can save up to half the time they spend on a TFF step through the reduction or avoidance of down times.”
For today’s highly potent or cytotoxic drugs, there is the added benefit of operator safety. The single-use TFF assembly can remain closed at the end of the process, and then it can be discarded. In many cases, however, conventional TFF cassettes require inactivation with caustic or bleach to ensure that the operator avoids exposure to hazardous drugs during disposal.
No round-up on trends in biopharmaceutical process filtration is complete without discussing the role of filtration virus safety. For Amanda Katz, product manager for virus safety at EMD Millipore, this is a topic that will never go away: “Virus contamination of biopharmaceutical processes is a persistent threat.”
Katz observes that bioprocessors control virus safety risk through careful sourcing of raw materials, virus testing, and clearance operations to remove or inactivate viruses: “Although downstream viral clearance must demonstrate robust virus removal to ensure patient safety,” she notes, “there are limited options for protecting the bioreactor, where maintaining business continuity is a priority.”
With recent publicly disclosed virus contaminations, the biopharmaceutical community is eager to improve upstream virus safety. This is evidenced by the Consortium on Adventitious Agent Contamination in Biomanufacturing, a joint effort between the Massachusetts Institute of Technology and biotech companies.
Existing upstream viral clearance technologies, or virus barriers, include high-temperature short-time processing, filtration, ultraviolet-C irradiation, and gamma irradiation. “Unfortunately each of these technologies has limitations,” Katz says. “As a result, biomanufacturers are seeking improved methodologies to prevent the introduction of virus. In response, multiple vendors are working on new virus filters specifically designed to treat chemically defined cell culture media.”
Unlike downstream operations, where viral clearance ensures patient safety, upstream virus barrier filters’ primary focus is preserving cell culture growth and productivity, and to provide a cost-effective solution for mitigating the business risk associated with a contamination.