December 1, 2011 (Vol. 31, No. 21)

Angelo DePalma Ph.D. Writer GEN

Improvements in This Critical Process Tend to Be More Evolutionary than Revolutionary

Filters have been a part of bioprocessing since the beginning. Yet “broad optimization potential” still exists for established processes and in filter selection for new processes, according to Holger Bromm, marketing director for filtration technologies at Sartorius Stedim Biotech.

“Our product development in the last few years has focused on specific applications such as media and buffer prep, and filtration of downstream intermediates, which are core applications within bioprocesses.” Even within these categories filterability can differ significantly, for example, supplemented versus chemically defined media, or mAb product pools just before and just after capture/elute.

For these distinct applications, Sartorius Stedim has introduced the Sartopore® 2 XL product line, which is based on Sartopore 2 products but carries prefilter membranes optimized for specific applications. Two such products have been developed for media and downstream applications and one each for sterilizing-grade filtration and mycoplasma. To further optimize performance and reduce filtration cost per liter the company increased the effective filtration area by 30%.

As bioprocess streams become more complex, prefiltration becomes increasingly necessary. “Contaminants need to be removed reliably to protect more expensive sterilizing-grade and mycoplasma-retentive filters,” says Bromm. He notes that, while polyether sulfone (PES) membranes are now well accepted due to their excellent flow and throughput characteristics, PES has not been widely employed for prefiltration. Sartorius Stedim has addressed this deficiency through its Sartoguard PES membranes, which, according to Bromm, “transfer the benefits of PES membranes to prefiltration and help reduce costs further.”


Sartopure® GF Plus adsorptive depth filters are designed for removal of contaminants like colloids, lipids, protein aggregates, and particles from biopharmaceutical fluids. They are used for protection of membrane filters, chromatography columns, and ultrafiltration systems in pharmaceutical and biotechnological production processes.[Sartorius Stedim Biotech]

Replacing Unit Operations

Rising volumetric productivity is causing bioprocessors to downsize from very large stainless steel reactors to vessels in the 1 kL to 2 kL range. As a result, some firms are switching from centrifugation to disposable depth filtration at the harvest step, according to Jon Petrone, vp of technical services at Pall. Centrifugation is a capital- and labor-intensive operation that serves very large batches well, where the cost of using multiple depth filters would be high. But at 1–2 kL a batch may be harvested with one disposable depth filter, without the need to acquire, inventory, and store capital equipment.

Filtration also plays a role in concentrating proteins post-harvest, and later on in the formulation of the drug product. “We’ve seen some processes lower volumes by a factor of 10 or 20 at this stage, further reducing downstream volumetric demands,” Petrone notes.

Viscosity becomes a serious product recovery issue at these concentrations, which can reach 80 g/L or more. Significant product can remain behind when it drains out after conventional tangential flow filtration (TFF). Flushing out product requires adding buffer, which dilutes the process stream. Pall’s solution is its Cascade single-pass TFF, which provides direct flow-through concentrations with no recirculation of product required. The company claims protein recoveries of greater than 98% at concentration factors greater than 20 times.

Single-pass TFF works similarly well just prior to final formulation, where drug substance concentrations may reach a syrupy 250 g/L. “With single-pass TFF the only dilution occurs from chasing out the holdup volume,” Petrone says. “It is no longer necessary to sacrifice product recovery for high concentrations.”


Pall’s Stax™ depth filter system, which is composed of a vertically configured stack of single-use depth filter capsules, reportedly offers flexibility and ease of use.

Disposable Process Steps

End users increasingly demand integration of disposable process equipment, and vendors are responding.

Designing integrated systems requires a deep understanding of the entire process not just filtration. “Manufacturers are experiencing some growing pains in these areas,” says Jason Walker, strategic business leader for life sciences at 3M.

“But as suppliers we must figure out how to deliver on all these needs,” he adds. “Otherwise we’re not bringing much value to the customer.”

Despite its acquisition of CUNO in 2005, and retiring the CUNO brand in 2009, 3M may still not be familiar as a filtration company. The new 3M purification business sells a full line of depth, sterilizing, and single-use filters, and has branched out to membrane adsorbers and disposable bioprocess containers.

Customization and timeliness is not limited to clinical- and production-scale manufacturing. Walker says that his development-stage customers, who optimize processes earlier than ever, expect the same performance from benchtop and pilot-scale equipment as from filters designed for larger processes. 3M recently launched a range of intermediate scale-up devices for depth filtration to meet this need.


3M’s Zeta Plus encapsulated depth filters and bioprocess containers provide single-use solutions for cell harvest.

Mark Carroll, biosafety product R&D group manager at EMD Millipore, agrees that integration has moved to the fore in bioprocess filtration. EMD’s Mobius® and Mobius FlexReady product lines are examples of customization and integration of tubing, filtration, and connectors. Seamless (both physical and metaphorical) integration is an evolutionary point that is much desired and anticipated, but not quite available for all products and systems today.

“We’re seeing a lot of activity around materials of construction for both filters and systems as well, particularly for gamma compatibility” Carroll says. Gamma irradiation predominates pre-sterilization technologies and drives not just materials but filter and system cleanliness, ease of use, and how easily filters interface with larger process assemblies.

“If you look at some assembled systems today you have to scratch your head. They’re confusing, difficult to use, and every one of those connections is a possible source of contamination or error.”

Carroll foresees the day, in the not-too-distant future, when capsule-less filters and the process systems that run them are indistinguishable, and a lot more user-friendly. “That will require a higher degree of cleanliness right out of the box.”

EMD has already taken steps to assure that membranes and clarification media are cleaner and require lower levels of preservatives, and thereby, do not require flushing with large volumes of buffer.

Filter integration typically occurs through either direct connection of a filter to a single-use bag system or by aseptic onside connection of a filter transfer set. In both cases the assemblies have been presterilized by gamma irradiation. “However, some of our customers have requested that we integrate filters that will not be gamma-irradiated,” notes Bromm.

These products are part of established processes. For these applications Sartorius Stedim offers filter transfer sets using tubing and aseptic connectors that have been presterilized by autoclaving and may be connected onside to other presterilized single-use components.


EMD Millipore’s Mobius FlexReady solution for trace contaminant removal integrates hardware with single-use flowpath and single-use ChromaSorb™ membrane adsorber binds negative impurities from host cell protein, DNA, endotoxins, and viruses at salt concentrations above those of traditional resins and membranes, the firm reports.

Continuous Improvement

Improvements in bioprocess filtration tend to be more evolutionary than revolutionary. Changes, such as replacing centrifugation with depth filtration, are adopted over time. Advances in sterilizing-grade filters, membrane adsorbers, and virus filtration are similarly incremental.

Sterilizing-grade filtration based on size exclusion is well established and understood, thanks in part to cooperative workshops conducted by the U.S. FDA and the Parenteral Drug Association. Because of the maturity of the technology, innovation in sterilizing filters will be incremental, says Vincent Pizzi, strategic marketing manager at GE Healthcare Life Sciences (www.gelifesciences.com).

“Improvements will come in the form of higher flow rates, novel materials of construction, and pore morphologies.” But vendors that introduce new products with alternative polymers would do well to keep a close eye on leachables and extractables, as impurity profiles come under increasing scrutiny.

Vendors have been experimenting with filter pore morphologies for years, with the objective of improving throughput while reducing clogging. Parker-Hannifin (www.parker.com/pro?cessfiltration) introduced an hourglass-shaped design in 1991 consisting of three bonded layers with wide, narrow, and wide pores. Others have copied this design, while Sartopore from Sartorius and Pall’s Supor® membrane filters employ more modern tapered V- or funnel-shaped pores, with the larger side facing upstream.

Membrane adsorbers have taken a long time to “gain traction” in bioprocessing, Pizzi explains. The membranes have lower capacity compared with resins and are generally not regenerated. Yet, Pizzi expects further utilization of membrane adsorbers for monoclonal antibody processes, particularly beyond the development stage. Vaccine manufacturing is another application area actively engaged in evaluating the use of membrane adsorbers as a polishing step to remove host cell proteins and DNA impurities.

“Some companies are considering crossing into those large-scale manufacturing areas with the selective use of membranes by exploiting the adsorbers’ high throughput and flow rates combined with low buffer consumption. The main thrust here is leaning out processes—substituting a chromatography column, when applicable, with a simple, single-use membrane cartridge that provides a robust flow-through process step in the same manner as a column chromatography step.”

Pizzi notes that virus filtration “is on everyone’s radar screen since it’s the most expensive filtration step.” Cost of viral clearance filtration and validation are obvious areas that need improvement. Alternatives to filtration, like irradiation, UV, heat, or chemical inactivation are unlikely to be adopted widely beyond a few niche applications. “All these techniques have the potential to affect the protein, and in bioprocessors must demonstrate through validation the ability to clear the inactivated virus particles.”

Costs, Sustainability, Support, Supply

All the vendors interviewed for this article mentioned cost containment as a driver for new filter and integrated systems development. Nobody likes to spend more money than they have to, but for high-priced biologic drugs the degree to which cost of goods actually matters is unclear. On the other hand streamlining processes (to simplify operations and improve quality) is an undisputed need. That’s an area where vendors with strong materials and engineering expertise can make a difference. “We are constantly brainstorming with customers on how to leverage our materials science expertise to help from the perspective of cost and process simplification,” says Walker.

Several vendors provide calculations demonstrating that disposable purification equipment saves time and money in the long run. Replacing an anion-exchange resin step with a membrane eliminates resin/column preparation and cleaning time, and reduces buffer volumes significantly. Membranes are now available bearing cation exchange, HIC, and affinity chemistries as well. Yet, many bioprocessors do not appreciate the economics.

EMD Millipore’s take on the cost-of-goods issue is to produce filtration products that are more efficient, not necessarily less expensive. “Customers are looking for efficiencies, and as an industry we must provide them,” says Carroll.

Today’s biotech companies, he explains, carry very low inventories, conduct production campaigns under short lead times, and expect a vendor to deliver customized systems practically “just-in-time.” As expected, timely delivery and customization work at odds. Working with the growing number of overseas biotech companies and contract manufacturers adds yet two other factors, distance and divergent regulations, to the equation.

Top companies continue to address sustainability concerns with single-use process equipment. Carroll admits that on the surface it seems counter-intuitive to connect single-use plastic products with environmentalism. “However, considering capital and energy requirements of the building and operating facilities based on multi-use equipment reveals that single-use approaches can be much more sustainable.” EMD recently instituted a filter take-back program, through which it turns waste plastic into new products.

The biopharmaceutical business has become globalized, and this has created issues of supply and support, according to Nick Hutchinson, market development manager for pharma and biotech at Parker Hannifin. “We find that customers value rapid responsiveness to technical inquiries, but also support for filter selection, sizing, and trouble-shooting.” At the same time, lean manufacturing initiatives are creating awareness for holding inventories low.

Parker Hannifin specializes in PTFE (Teflon) membranes for gas filtration, PES for liquids, and polypropylene fibers for prefiltration. PTFE is highly hydrophobic, but Hutchinson notes that his company has rendered the material hydrophilic through a chemical modification that he could not disclose. He has “reason to believe, based on success in ophthalmic markets,” that this material could be adapted to bioprocess filtration as well.

“Globalization has special significance for biosimilars, where cost of goods is likely to be a stronger driver than for originator molecules.”

Previous articleNearly Two-to-One Majority Sees FDA or EMA Approval of Gene Therapy by 2013
Next articleRoche’s Cobas EGFR Mutation Test Granted CE Mark