April 1, 2014 (Vol. 34, No. 7)
Since bioprocesses consist of discrete, successive unit operations, relatively slow steps could cause holdups or bottlenecks. This fact, in light of upstream efficiency improvements, has given rise to the notion of “capacity mismatch” between cell culture and purification—the consequence, if you will, of too much of a good thing.
Data presented in GEN and elsewhere by Eric Langer, president of consulting firm BioPlan Associates, suggest that the problem is variable. As of 2011, more than half as many companies reported experiencing “serious bottlenecks” (11.8%) as no bottlenecks at all (20.5%). Given the continuum of responses and the diversity of processes and products, a rate of nearly one in eight processes facing upstream/downstream capacity mismatches is a serious issue.
Or is it?
“It depends on what one means by bottleneck,” says Mani Krishnan, director of biopharm regulatory and compliance strategy at EMD Millipore.
No one argues that downstream productivity has kept pace with upstream titers that have risen 3-, 5-, or 10-fold during the last dozen years.
“If biomanufacturing were a continuous process, the bottlenecks would indeed occur when upstream efficiencies increase and downstream efficiencies do not,” Krishnan continues. But bioprocessing is not continuous. “It is generally a batch process. In the best case, it is semicontinuous, so a 10-fold improvement upstream with no concomitant improvement downstream need not always create bottlenecks during purification.”
Distinctions between batch and continuous processing explain why downstream operations usually find time to “catch up” with bioreactor production. Instead of facing a daily influx of raw product, purification specialists have two weeks or longer—depending on the duration of the culture—to purify a batch. Not that most proteins stay in hold for anywhere near that long. Even when multiple bioreactors operate concurrently, purification has four or five days to purify material.
“Most bioprocessors don’t view this, under normal circumstances, as a bottleneck,” Krishnan concludes. Furthermore, the emergence of platform technologies for monoclonal antibodies have streamlined downstream processing and made it more predictable.
If anything, the perception of downstream bottlenecks results from a previous under-utilization of downstream capacity. When processes entered purification every two or three weeks, and purification took four or five days, downstream capacity would be under-utilized to a degree that would give fits to accountants in other industries. Langer has noted that capacity utilization for cell culture-based biomanufacturing is less than 70% and for microbial production barely 60%, whereas U.S. Federal Reserve data suggest that capacity utilization averaged across all industries is 80%.
Krishnan explains: “Biopharm companies needed a certain capacity to purify protein in a timely manner, but often the equipment was idle. Today, with bioreactors coming downstream every four or five days (or even faster) at high-throughput companies, downstream utilization is as high as 95%. This is especially true for highly streamlined operations at very large biotech companies.”
Krishnan admits that downstream capacity constraints could cause problems for high-volume production of unusually labile molecules, but upstream efficiency alone is insufficient to cause bottlenecks.
Having to balance between upstream and downstream capacity in a highly regulated industry is not totally unexpected. “To be fair, biomanufacturing is not a very mature industry,” Krishnan says. Moreover, until recently, its cost structure left plenty of room for inefficiencies. That is changing, as the twin uncertainties of cost pressures and government intervention in the marketplace provide new incentive to streamline and innovate.
“We’re often guilty of simplistic thinking when it comes to cause and effect in this industry,” Krishnan adds. “Industry has found ways of handling capacity mismatches. Otherwise we’d have seen a huge increase in capacity building during the last five years, which has not occurred. In fact, companies have mothballed large facilities due to over-capacity. Some companies with very high utilization do indeed feel the capacity crunch, and they are in fact building new facilities, but that is not the norm. However, the drivers for the new facilities go beyond increasing titers in the bioreactors.”
Technology Saves the Day
According to Uwe Gottschalk, Ph.D., vp of purification technologies at Sartorius Stedim Biotech, Langer’s data may be representative, if not accurate. “The answers will be all over the place. Some companies will have experienced no issues whatsoever. Others will say they have, but the bottlenecks have resolved. And some will tell you they have no idea how to overcome downstream bottlenecks. It depends on their business model, size, and product.”
Trouble-free companies tend to be either large firms or startups. The former generate tons of monoclonal antibodies and maintain an engineering staff that efficiently handles process bottlenecks. The latter are most likely to adopt cutting-edge technologies that obviate manufacturing bottlenecks. “In addition to investing the most in new technologies, these companies are adaptive and flexible,” Dr. Gottschalk says.
And eventually, as the blockbuster era winds down, as markets become more fragmented, and as batches return to manageable volumes, such firms will dominate biotech manufacturing. When these companies construct new facilities, Dr. Gottschalk explains, they will mostly likely operate with significantly smaller volumes and with more versatile downstream capabilities than today’s large antibody processes.
It is through these smaller, nimbler firms that downstream innovation will enter large-scale bioprocessing, either through acquisition or emulation. According to Dr. Gottschalk, next-paradigm facilities are already “mushrooming” all over the globe. “In these plants, the facilities and technology are decoupled.” Moreover, downstream capabilities will be versatile enough to meet challenges brought about by significant changes in titer.
“What remains of remaining blockbusters will be legacy products, which will generate legacy titers of one to two grams per liter.” Thus, at very high manufacturing volumes, upstream-downstream mismatches will revert to levels that for years have been considered manageable. Dr. Gottschalk likes to describe bioreactors as the “heartbeat” of a facility, with their output determining how resources are allocated elsewhere.
In the end, technology will save the day, but that has not always been apparent. “Established companies, with deep pockets, were once complacent about new purification technologies. But now they’re also jumping onto the bandwagon, particularly vaccine companies, to meet demands for decentralized production,” Dr. Gottschalk remarks. “They can’t achieve that from traditional facilities.”