Downstream processing has been considered a “bottleneck” in the manufacture of protein biotherapeutics ever since cell culture engineers began dramatically improving production efficiencies around the turn of the century. And as single-use technologies have grown in importance and acceptance, offering more solutions every year, their biggest challenges too have been in the separation, purification, and processing that follows product expression in cell culture. Many of the technologies familiar to process engineers — e.g., centrifugation and chromatography — present technical and economic problems.
In a recent white paper (1), BioPlan Associates reported that respondents to its 13th annual survey of biomanufacturers reported downstream processing as requiring the most technological improvement of any part of bioprocessing. But productivity is improving. For the coming year, less than half (45%) of respondents expect capacity problems — compared with 88% of the industry in 2005.
Where would user companies like to see suppliers direct their development efforts? Nearly three quarters of respondents (especially contract manufacturers) cited downstream continuous bioprocessing and disposables for purification. Our featured report in May will address the former more specifically — but in general, although interest is high, not many relevant products are available yet. In 2016, bioprocessing facility budgets were up 5% for downstream technologies. “Essentially, nearly all respondents want improvements in downstream technologies,” the report claims (1).
Separation and purification technologies are slowly catching up to upstream processing, however, and vendors are filling the gaps in their offerings. Filtration continues to advance, of course, with some options even encroaching on the adsorption mechanics of chromatography. And now even drug-product filling operations can choose single-use options.
Harvest and Clarification
Modern bioproduction technologies have given us expression titers measured in grams per liter of culture — whereas before the turn of the century, milligrams per liter were common. High-density culturing is one reason, but other advancements include more complicated media and feeds, culture strategies, and optimization efforts. Some of those improvements upstream can cause trouble for harvest, clarification, and beyond.
The first phase of downstream processing typically includes centrifugation or primary filtration steps followed by secondary filtration before purification involving chromatography (2).
Harvesting here is just the same as in agriculture: collecting the material produced by your hard-working life forms (in this case, animal cells or microbes). Along with the expressed protein of interest come host-cell proteins that are interesting only as contaminants; nucleic acids; leftover nutrients, supplements, and byproducts; secondary metabolites; and water. Clarification follows to prepare this messy product stream for downstream chromatography and purification. With solids making up 3–5% of the culture volume, the clarification process alone typically takes the yield of active protein down by 10–15% (2). The most demanding processes involve high turbidities and increased particle/contaminant loads as well as high densities and titers (3).
Filtration: Complications related to downstream processing are ironic in one sense: Disposability in bioprocessing pretty much began with filter cartridges, although upstream/production single-use applications surpassed downstream technologies along the way. Filters remain the most popular single-use technologies, with 91% of companies currently using disposables citing filter cartridges in the latest BioPlan Associates survey, with robust growth continuing (1). Use of depth filters is are widespread (82%), although only 5% of respondents report using disposable tangential flow filtration (TFF) devices.
Another up-and-coming filtration technology is “body-feed filtration,” which incorporates filter-aids such as diatomaceous earth (DE) to increase filter capacity — of particular use with the highly concentrated feed streams that can come from high–cell-density cultures. Adsorptive depth filtration (ADF) incorporates DE into the filter itself (3). Depth filters have particular utility in clarification, which increases with the help of filter aids, flocculating agents that settle impurities out of a harvest solution, or protective prefilters. Sartorius Stedim and MilliporeSigma are well-known proponents of such approaches (2–4). See the next article in this featured report for more discussion.
Filters play many roles in downstream processing — beyond harvest and clarification to virus reduction, buffer exchange, volume reduction, and final sterile filtration just to name a few. Filtration systems are highly automatable, as well (see the box, below), which is increasingly needed in modern biomanufacturing.
|How Automation Can Help|
|Multiple small-scale bioreactors in the 1-L to 10-L size range are widely used in process development and early stage material supply laboratories. The objective here is to get purified protein from a platform process so it can be sent for analysis. At this stage, users don’t need an optimized process; optimization will come later. For initial clarification, they need a simple plug-and-play system based on depth filters — with, for example, glass-fiber media — followed by a sterilizing-grade filter (with a polyethersulphone membrane). These systems require minimal flushing and need to be stable for gamma irradiation.
Biological production processes are inherently variable, so process engineers either use oversized filters or accept some occasional product losses to premature filter clogging. Here is where automation can help. By putting a single-use pressure sensor in line between your bioreactor and clarification filter — and another between your clarification and sterilizing-grade devices — you can monitor pressure build-up and respond as needed.
Parker domnick hunter’s SciLog range of automated normal-flow filtration (NFF) systems include a rate/pressure (R/P Stat) feature that takes automation a step further. In R/P Stat mode, a process runs at a constant flow rate while pressure is monitored. If pressure reaches a preprogrammed limit, then pump speed is reduced to maintain it below that limit while allowing the process to continue at a slower flow rate. This presents an alternative to requiring an operator to stand over the system and make manual interventions as necessary. With such automation, filter capacity can be increased by up to 30% while operators are freed up from continual system monitoring. This ensures full product recovery regardless of feedstream quality.
—Guy Matthews (market development manager at Parker domnick hunter)
Centrifugation has been problematic for conversion to single use, but some suppliers do offer solutions. Most notable are kSep systems (now part of Sartorius Stedim Biotech) and Carr Centritech UniFuge systems from PneumaticScaleAngelus. The former are based on fluidized-bed centrifuge technology originally developed by KBI Biopharma; the latter comprise more traditional technology with irradiated and disposable product-contact surfaces.
The Sartorius technology can be used both for harvesting cells as product or discarding them as by-products. Balanced centrifugal and fluid-flow forces retains particles as a concentrated fluidized bed under a continuous flow of media or buffer. Some companies are applying it toward continuous processing. The Carr system offers continuous operation as well. Both are highly automatable, although they face limitations in scalability and process monitoring (5).
Other Options: Harvest clarification methods such as feed pretreatment involve single-use technology only in that they require tubing to move harvested material and treatments to and from a mixing system (which may or may not be disposable). Acids and salts can cause solutes to precipitate out, but they also can denature proteins; cationic polymers bind contaminants together into cloudy flocs that can be filtered out, but the polymers themselves become contaminants that must be removed later. If such methods are used, they are likely to be combined with the above technologies, whether single-use or multiuse forms thereof.
When you ask about single-use chromatography, the answer usually comes in the form of prepacked columns (e.g., ReadyToProcess brand from GE Healthcare, OPUS columns from Repligen, and Chromabolt and Mobius FlexReady brands from MilliporeSigma). They aren’t strictly single-use in nature, though: Most resins, gels, and other chemistry-based separation media are too expensive to use only once. Instead, they are washed and equilibrated for repeated use with the same product stream, then discarded along with their polymer columns once a batch is complete. In addition to the usual benefits related to cleaning and cleaning validation, however, this approach saves users the time, cost, and fussiness of column packing — giving them consistent results from expert suppliers instead.
Column volumes currently available (e.g., ≤20 L) require several cycles to purify a 1,000-L batch. And that takes time, thus adding cost. The alternative of adding more columns also costs more — lots more — unless you’re talking about continuous multicolumn processes such as that described in this special section by authors from CMC Biologics and Pall Life Sciences. The key with disposable chromatography is to balance cost of goods (CoG) of materials and time/labor. The more expensive the medium (e.g., protein A affinity resins), the more sense it makes to use traditional multiuse technology. So too with frequent harvesting and media with long lifetimes. However, mixed-mode sorbents and sequential chromatography are improving performance while reducing costs. Meanwhile, smaller columns are becoming more popular thanks to high-capacity, high-flow resins and smaller production batches (e.g., from high-density cultures).
A new column-free chromatography technology is drawing publicly stated interest from companies such as Medimmune (Astrazeneca) and Regeneron: Continuous Countercurrent Tangential Chromatography (CCTC) from ChromaTan Corporation. A slurry of resin sequentially binds product as it flows through a series of mixers and hollow-fiber membranes, where it is also washed, eluted, and stripped in a continuous process. The “countercurrent” refers to buffers flowing through in the opposite direction, both lessening the amount of buffer used and improving resin use efficiency. Like the Pall process highlighted elsewhere in this report, CCTC has potential for continuous processing — and more on this small company’s progress is coming in our May featured report.
Alternatives to chromatography resins — in columns or otherwise — are available from membrane suppliers. Functional filtration and membrane adsorbers typically are limited in dynamic binding capacity (DBC) compared with column/resin technology, but they handle significantly higher flow rates.
Nearly one in five respondents to BioPlan Associates’ 13th annual survey cited chromatography columns as currently causing them significant or severe capacity constraints (1). Membrane adsorbers, however, have yet to take over a significant portion of the market. But their adoption is growing, with first use in respondents’ facilities reportedly up 13% for 2016 and 10-year market growth of 31%. The concept is not entirely new technology, but recent introductions — e.g., salt-tolerant devices and new ligand technologies — offer improvements in robustness and efficiency. The article from Renaud Jacquemart and James G. Stout in this special insert provides more discussion.
Formulation, Fill and Finish
Mixing and storage systems are another single-use technology making inroads with biomanufacturers, at least in part because of their many uses. Their annual adoption rate was up 16% for 2016, and their 10-year growth has been about 50% (1). Downstream applications include viral safety (e.g., detergent treatment), storage of process intermediates, and product formulation, to name a few.
Fill and Finish: BioPlan Associates identified single-use technology as the key trend in biopharmaceutical fill and finish operations, with nearly two-thirds of their respondents ranking it number one (1). Over a third of respondents plan to implement new fill–finish technologies at their facilities in the next two years.
Major suppliers such as MilliporeSigma and Pall have introduced solutions to meet this need, and companies such as Biotest, Disposable-Lab, and Merck have implemented them (6–8). The basic idea has been to adapt fluid-path technology with metal filling needles, combining them with bag containment and pumping systems. As you’ll see in the interview at the end of this featured report, facility design and engineering firm NNE Pharmaplan is a major proponent of these ideas.
Pumps are important throughout downstream bioprocessing, of course. And although polymer tubing and connectors are established single-use components for fluid handling, pumps thus far have proven to be more of a challenge (9). Solutions are in the works to address these needs, and companies such as PSG Dover already have put forth some options. It offers a line of positive displacement diaphragm pumps (Quattroflow) with product-wetted plastic chambers that can be replaced. Other options include a rotary pump from Quantex Arc and a disposable pump-head system for Masterflex peristaltic pumps from Cole Parmer.
Challenges Yet to Be Overcome
Finally, another gap that remains to be filled completely for downstream processing relates to sensing and sampling. Most such solutions that have been made available so far are meant for upstream production applications. PendoTech offers pressure sensors, however, as well as those for ultraviolet absorbance and conductivity (10). Sartorius Stedim Biotech has incorporated such options into its own product offerings (11).
But what might be the biggest challenge facing downstream processors who want to use disposable systems and components isn’t so much technical as it is business related. What users really want from their single-use technology providers is standardization of designs that would allow them to mix and match components to put together systems that work best for their own processes. Bioprocessors say that this would improve adoption and implementation of disposables; suppliers are reluctant to share with their competitors. Not long ago, in fact, if you brought up the question of single-use standardization at an industry conference, you might get a lot of laughs but not much real discussion. However, companies on both sides are taking the idea more seriously now. And the “alphabet soup” of organizations concerned with single-use technology are helping to make it happen (12). Standards could reduce the risk of process failures and allow suppliers to stick with proven features while focusing their attention on needed innovations.
1 Top 15 Trends in Biopharmaceutical Manufacturing, 2016. BioPlan Associates, Inc.: Rockville, MD, 2016.
2 LeMerdy S. Evolving Clarification Strategies to Meet New Challenges. BioProcess Int. October 2014: insert.
3 Minow B, et al. High–Cell-Density Clarification By Single-Use Diatomaceous Earth Filtration. BioProcess Int. 12(4) 2014: S36–S46.
4 Schreffler J, et al. Characterization of Postcapture Impurity Removal Across an Adsorptive Depth Filter. BioProcess Int. 13(3) 2015: 36–45.
5 Pattasseril J, et al. Downstream Technology Landscape for Large-Scale Therapeutic Cell Processing. BioProcess Int. 11(3) 2013: S38–S47.
6 Camposano D, Mills A, Piton C A Single-Use, Clinical-Scale Filling System: From Design to Delivery. BioProcess Int. 14(6) 2016: 50–59.
7 Gross R, et al. Establishing Single-Use Assemblies on Filling Equipment. BioProcess Int. 12(4) 2014: S48–S54.
8 Zambaux J-P, Barry J. Development of a Single-Use Filling Needle. BioProcess Int. 12(5) 2014: 46–53.
9 Wittkoff W, Prasad R. Single-Use Pumps Take Center Stage. BioProcess Int. 11(4) 2013: S18–S23.
10 Annarelli D. Novel Single-Use Sensors for Biopharmaceutical Applications. BioProcess Int. 12(7) 2014: 58–59.
11 Weichert H, et al. Integrated Optical Single-Use Sensors: Moving Toward a True Single-Use Factory for Biologics and Vaccine Production. BioProcess Int. 12(8) 2014: S20–S24. 1
2 Vogel JD, Eustis M. The Single-Use Watering Hole: Where Innovation Needs Collaboration, Harmonization, and Standardization. BioProcess Int. 13(1) 2015: insert.
Bird P, Hutchinson N. Automation of a Single-Use Final Bulk Filtration Step: Enhancing Operational Flexibility and Facilitating Compliant, Right-First-Time Manufacturing. BioProcess Int. 13(3) 2015: S40–S43, S52.
Blomberg M. The New Hybrid: Single-Use Systems Enabled by Process Automation. BioProcess Int. 13(3) 2015: S34–S39.
Grier S, Yakabu S. Prepacked Chromatography Columns: Evaluation for Use in Pilot and Large-Scale Bioprocessing. BioProcess Int. 14(4) 2016: 48–53.
McGlaughlin MS. An Emerging Answer to the Downstream Bottleneck. BioProcess Int. 10(5) 2012: S58–S61.
Metzger M, et al. Evaluating Adsorptive Filtration As a Unit Operation for Virus Removal. BioProcess Int. 13(2) 2015: 36–44.
Mok Y, et al. Best Practices for Critical Sterile Filter Operation: A Case Study. BioProcess Int. 14(5) 2016: 28–33.
O’Brien TP, et al. Large-Scale, Single-Use Depth Filtration Systems. BioProcess Int. 10(5) 2012: S50–S57.
Quinlan A. Advances in Chromatography Automation. BioProcess Int. 13(1) 2015: 16–17.
Cheryl Scott is cofounder and senior technical editor of BioProcess International, PO Box 70, Dexter, OR 97431; 1-646-957-8879; firstname.lastname@example.org.