By Angelo De Palma, Ph.D., Contributing EditorBiotechnology companies sink or swim based on cell culture manufacturing capacity: too little means product delays, too much and accountants pull out their hair over “excess capacity” and “underperforming assets.” With cell culture the manufacturing method of choice for biotech’s fastest-growing product segment—monoclonal antibodies (MAbs)—one could say that as cell culture goes, so goes the industry.Warnings of imminent, severe cell culture capacity shortages picked up steam after Immunex’s troubles, in the late 1990s, with its Enbrel (etanercept) MAb rheumatoid arthritis drug.Enbrel was so well-received following its 1998 approval that Immunex could not meet demand. The company announced plans to build a dedicated manufacturing facility in Rhode Island and, later, to contract out a chunk of Enbrel production to Amgen (Thousand Oaks, Calif.). Three years later partner became parent, as Amgen acquired Immunex for $16 billion.Enbrel was one of biotech’s great success stories, not the unmitigated disaster portrayed by commentators and editors. “Immunex predicted that Enbrel would be merely a successful drug, not a fantastically successful drug,” observes Nick Shackley, vice president of business development at Cambrex Corp. (East Rutherford, N.J.), a biotech contract manufacturing organization (CMO). Nevertheless, the stampede of ominous capacity predictions was on and continues to this day.Predictions of shortages were not only dire, but claimed four-significant-figure certitude. An investment banker quoted in a January 2002 Chemical and Engineering News article foretold of a “best case” scenario (depending on one’s point of view, since it required several pipeline products to fail) that would result in the liberation of just 12,150 liters of capacity worldwide.Those awaiting another “Enbrel” were disappointed. What’s puzzling is that five years later the consensus on biotech capacity is that there is no consensus on whether the “crunch” occurred at all, came and went, was postponed, or was simply a figment of someone’s imagination.Some experts believe that manufacturing capacity was never in danger of running out and will not run out any time soon. Others feel capacity shortages have merely been deferred to a later time, when new product approvals and more exuberant equities markets will conspire to rekindle capacity concerns.The optimists“Whether the capacity crunch exists or not depends on which company you talk to and when you talk to them,” admits Ken Tindall, Ph.D., senior vice president for science and business development at the North Carolina Biotechnology Center (Research Triangle Park, N.C.).Data from Bioplan Associates (Rockville, Md.) suggest a huge perceptual gap among bioprocess experts regarding future capacity shortages (see chart). Forty-four percent of survey respondents believed they would experience capacity shortages by 2008; 28% disagreed and 28% were “neutral” or had no opinion.Earlier concerns over capacity shortages relied on overly-optimistic estimates of demand and approval rates, says Janice Reichert, Ph.D., senior fellow at The Tufts Center for the Study of Drug Development (Boston, Mass.). Could current capacity worries be based on the same misconceptions?“I’ve heard ridiculous claims for the number of MAbs that would be approved in the next five years, something like 100,” notes Dr. Reichert. “These estimates are ridiculous given the number of pipeline products and historic success rates. You can’t expect better than about a 25% approval success rate, which means that no more than two or three MAb products are likely to be approved in any given year for the foreseeable future.” Just one MAb, Biogen Idec’s (Cambridge, Mass.) Antegren, is currently under regulatory review.Roger Lias, Ph.D., vice president of business development at KBI BioPharma (Durham, N.C.), believes that product-approval optimists and volumetric-capacity pessimists feed off each other. “The analyst and banking communities did us a huge disservice with their doom and gloom scenarios,” he notes. “Most took an overly simplistic view of biotech’s capacity needs based on the successes of a few monoclonal antibodies, approval rates, and what were up to then normal process yields. They matched capacity they were aware of with products they knew about. They relied too heavily on biotech’s marketing and launch date projections. It was a very naÃ¯ve math game.” Doom-and-gloomers also ignored big biotech’s underutilized capacity and willingness to share it, not to mention improvements in process yield and efficiency that allow processors to get more protein from smaller volumes.That’s not to say that no biotech sponsor will ever experience difficulty finding bioreactor space. Lack of planning, unexpected demand, multiple products competing for facilities, and mishaps could all contribute to shortages, but optimistic observers don’t see these factors affecting the industry globally.“Predicting capacity needs is immensely complex and it’s still difficult for individual companies to predict what they will need in five or ten years,” Dr. Lias admits. “We’ll probably see periods of relative abundance and shortage—a seesaw effect. One successful high-dose product can tip the scales towards shortage, but as long as there is reasonable balance everything will be OK.”Nick Shackley of Cambrex believes capacity issues are something of an embarrassment for many manufacturers, especially those that responded to shortfall predictions but later put off expansion plans. “The crunch that never happened left a lot of people with egg on their faces.”The pessimistsAlthough severe cell culture shortages never materialized, many capacity-crunch Jeremiahs are sticking to their positions, namely that capacity fundamentals have not changed.Howard Levine, Ph.D., presidetn of BioProcess Technology Consultants (Acton, Mass.), thinks that while biotech is safe for now, danger lurks. "Capacity and supply/demand will be fairly well matched through 2007, but unless something changes we will experience a capacity shortfall in the 2008 to 2010 time frame," he warns (see chart).Based on volumetric capacity, the top three firms (Amgen, Biogen Idec, and Genentech) possess about 60% of the worldwide captive cell culture capacity, according to Cambrex’s Shackley. As much as 25% of capacity is held by other “innovator” companies. That leaves only about 15% of global capacity to the CMOs, a buffer too small, in the view of Finnegan and others, to absorb modest increases in demand.Uncertainties related to product approval and high capital investments still operate, however, especially for firms that lack captive capacity. “You don’t really know if you will need large-scale manufacturing until fairly late in clinical trials,” notes Dr. Tindall. “By the time an approval comes through, building or expansion may not be an option without incurring serious delays.”
|PREDICTING CAPACITY NEEDS: A BETTER WAY?
Biotech capacity planning is a game of incomplete knowledge based on events that may or may not occur years down the road. Bio-drug developers tend to think of multi-hundred-million-dollar investments and four- to six-year timelines for facility building as an all-or-nothing proposition. Thus every capacity decision becomes a capacity crunch unto itself, irrespective of industrywide excess or scarcity.
By hedging their bets on a product’s success manufacturers may have their cake and eat it, too. As part of its pharmaceutical advisory practice, 2Value Consulting Group (New York, N.Y.) offers strategic capacity planning using a “real options” approach.
Real options is based on the Black-Scholes formula, a complex algorithm that won Robert Merton and Myron Scholes the Nobel Prize in economics in 1997. Originally applied to financial options markets, real options allows pharm/biopharm developers to exploit drug approval uncertainty to their advantage, says Uriel Kusiatin, a principal at 2Value. “Black-Scholes tells us that uncertainty is not bad, but good,” Kusiatin explains. “It allows you to mitigate downside risk while taking advantage of opportunities from the upside perspective.”
As implemented by 2Value consultants, Real Options Analysis breaks drug development and capacity acquisition into manageable, parallel time points or milestones. For example, development might be parsed simply into one preclinical and three clinical phases. At each stage, depending on the calculated risk of success (the subject of perhaps another algorithm), developers might take progressive steps towards either reserving space at a CMO or building a new facility.
Developers could acquire the land and begin speaking with architects and engineers early in Phase I, construct a shell as the product enters Phase II, and so on. This way, at every point the investment will be commensurate with the progress the product has made through the pipeline.
|Increasing Capacity: Betting on Technology
Regardless of one’s stand on the capacity debate, the sky hasn’t fallen and it is unlikely to fall any time soon. Future concerns for handling unexpected demand or unusually high approval rates may be rendered moot by the ongoing march of biomanufacturing technology, says Shawn R. Smith, business segment director for cell culture at GIBCO, a subsidiary of Invitrogen Corp. (Carlsbad, Calif.).
Higher-yielding expression systems, longer-producing cells, media optimization and straightforward process improvements are combining to improve volumetric productivity. Most of this work occurs behind the scenes, at universities and research-based manufacturers, CMOs, vendors and small biotech firms.
Process improvements range from paradigm-breaking (transgenics) to slow-but-steady (process analytics). Among these, cell line engineering and media optimization may be the most accessible near-term bets.
Cell line engineering optimizes protein production versus time through genetic manipulation, simple evolution, or both. Cellular engineering’s goals are higher protein productivity, longer useful cell life, and production of more human-like, less-allergenic products.
Examples include Lonza Biologics’ (Rochester, N.Y.) glutamine synthetase (GS) system, which bioengineers hybridoma cell lines (the expression system of choice for MAbs) and Chinese hamster ovary (CHO) cells. Lonza outlicenses GS and employs it to differentiate itself from other CMOs. Another cell line innovation is the PER.C6 cell line from Crucell (Leiden, the Netherlands), used to generate therapeutic adenovirus, vaccine-producing viruses and MAbs.
Media optimization lacks the glamour of cell line engineering but is essential considering that almost every productive cell line differs physiologically from its parent line, and therefore requires a slightly different environment for optimal growth.
A less-than-obvious goal of media optimization is creating new formats that reduce process steps by simplifying or decreasing the number of individual additives during media prep. For example, Invitrogen's Advanced Granulation Technology delivers complex specialty media in dry format as a single powder. Conventional media rely on a basal formulation plus periodic supplement infusions.
“Most media suppliers conduct optimization free or at minimal cost to end users, with the idea of securing long-term supply positions for the resultant optimized formulation,” says GIBCO’s Smith.
|Smaller Volumes: Answering the Capacity Planning Headache
It’s no secret that biomanufacturing is still rather inefficient compared with other process industries, even those that employ fermentation. Rather than building bigger tanks and constructing ever-larger facilities to overcome inherent inefficiencies, tomorrow’s bioprocessors will utilize existing capacity more effectively.
By combining novel microbial/cellular expression systems, cell line engineering and media optimization, biotech has already made significant progress. Today’s state-of-the-art expression systems generate up to two grams of protein per liter of process volume — a twentyfold increase over 1990 levels.
On the pharmacology side, improved bioavailability and protein stability could cut dosages for MAb products as much as tenfold. Examples include proteins modified with polyethylene glycol (PEG) residues, encapsulated in vesicles, or delivered through implantable or ingested biodegradable materials. Amgen (Thousand Oaks, Calif.) has already developed a version of its G-CSF white blood cell booster based on PEGylation.
Significant progress has been made at the unit operation level as well. KBI BioPharma’s (Durham, N.C.) centrifugal bioreactor (CBR) technology applies advanced fluidized bed methodology to immobilize and maintain cells at extremely high density, in a low-shear environment, without the use of membranes or other solid supports. Fluidized cell beds are held between opposing centrifugal and hydrodynamic forces for long periods, facilitating highly efficient continuous processing. According to Roger Lias, Ph.D., vice president of business development of KBI Biopharma, CBR could reduce manufacturing-related capital operating costs significantly and thus tilt the capacity see-saw, at least for a while, in biotech’s favor.
“CBR has recently moved beyond proof-of-principle studies and into applications development,” comments Dr. Lias. “We believe that by offering a more consistent product than what is obtained from conventional stirred tank bioreactors, CBR will provide downstream efficiencies as well.”
For now, KBI BioPharma is focusing on demonstrating high productivity for MAbs and recombinant products and plans to enter research collaborations to move its technology forward.