August 1, 2006 (Vol. 26, No. 14)
Looking at the Needs and Challenges and Considering the Range of Possible Alternatives
The capacity of large-scale antibody production facilities to meet the worldwide needs of the pharmaceutical industry remains a key issue.
Wolfgang Berthold, Ph.D., CTO at Biogen Idec (www.biogen.com), was one of the speakers to address the question at IBC’s “Antibody Development and Production” meeting in Carlsbad, CA. This is a complex issue, dependent on interacting variables, including the nature of the drugs, their life cycles, the steps from crude to finished product, availability of alternative high-output systems, and regulatory constraints.
Moreover, purification technologies are dependent on the nature and source of the crude material. Plant, animal, bacterial, and fungal material all present different challenges for purification specialists. A particle-free feedstock is essential, and reaching this stage may require mechanical pretreatment and innovative precipitation methods. Plant material may contain phenolic compounds, and viruses are a risk in animal-derived material.
Financial considerations, which are notoriously hard to predict, include insurance reimbursement for the product and the capital investment required for new plant construction. Given the complexity of the problem, it is not surprising that there are disagreements among analysts as to the optimum strategy for achieving these products and how the challenges should be met.
Various expression systems have experienced large increases in productivity in recent years as a result of improvements in media performance and genetic engineering and selection of cell lines. As cellular output is enhanced, increases in titers will lower demand for bioreactor capacity. However, Dr. Berthold’s investigations show that only one-third of biopharmaceutical production cost derives from the biosynthetic system employed.
Other factors, including the total amount of product needed and the ceiling price of the product, may determine the desirability of alternative systems. Time-to-market may also be a powerful factor in determining which expression system to employ.
So, manufacturers in coming years will find themselves faced with difficult and critical decisions without solid and objective resources for selecting antibody production alternatives.
Cell Culture Platform Process Improvement
The vagaries of mammalian cell lines and their response to a range of cultivation strategies was addressed by Diana Hoganson, Ph.D., research scientist at Applied Molecular Evolution (www.amevolution.com).
A well-recognized strategy for generating high-level antibody producers is the use of the dihydrofolate reductase (DHFR) gene as a selecting tool. When an antibody gene is transposed next to the DHFR gene, and the cells are placed in the presence of an inhibitor of this enzyme, a dramatic selective pressure is placed on the cell line. The result is that the antibody gene is multiplied within the cells, producing variants with increased levels of antibody synthesis.
However, when Dr. Hoganson evaluated independent isolates derived from the same strain of CHO cells, she found they responded in different ways to the culture media. This is not surprising, since cultured mammalian cells are notoriously unstable and rapidly evolve in different directions when subjected to the amplification and selection process.
Thus a labor-intensive process of optimization is required to bring the cells to a level of peak performance. Rather than reiterate process development exercises, each new project is optimized starting from the most recent process. Over time, this approach allows for shortening of time lines, reduction in development costs, greater process flexibility, and possible improvement in product quality.
Among the challenges Dr. Hoganson faced were maintaining oxygen levels and controlling carbon dioxide, osmolality and pH, all of which are critical for optimum performance.
In one case, Dr. Hogansan evaluated variations in nutrient content and observed that media with lower nutritional content actually allowed higher titers than a richer media, suggesting that some component(s) in the media may be inhibitory. In this case, a commercial medium was then used to scale up to a 100-L GMP production run.
Dr. Hoganson also evaluated the optimized process using Wave Biotech LLC’s (www.wavebiotech.com) Wave 200 disposable biobags, which performed equivalently to the permanent stirred tank bioreactor. Given the convenience of the disposable cultivation units, this observation constitutes a positive endorsement of the Wave technology, she notes.
“We find that platform development is critical to the success of our team, allowing us to meet yield, timeline, and product-quality requirements. It is a never-ending task, requiring constant upgrading,” Dr. Hoganson stated. “However, even with a viable protocol in place, project-specific improvements continue.”
Matt Olsen, an applications scientist at Stedim Biosystems (www.stedim.com), discussed his company’s products for controlled freezing and thawing of biological solutions in disposable single-use containers. Although freezing may be essential for storage and transport of biological solutions, the process can wreak havoc on highly fragile aqueous protein solutions.
As water crystals form during freezing, the solute concentration increases in the remaining portion of the solution. The process is compounded for large volumes as solute molecules flee from the growing mass of ice, moving toward the interior from the outside by both diffusion away from the ice front and convection within the solution due to both temperature and density gradients.
This process is exacerbated if freezing proceeds slowly, a common feature of uncontrolled processes using bottles or standard bioprocess bags placed in laboratory freezers. The overall result is solute concentration changes occurring differentially throughout the frozen volume.
Stedim Biosystems markets a line of controlled freeze-and-thaw equipment, referred to as Celsius, that maintains the intended formulation conditions through out the solution volume during the freezing process. Controlled freezing is presented as a method capable of minimizing this macroscopic freeze concentration in production volume equipment.
The effects of freezing on the distribution of protein and solutes present in the freezing containers were examined. When compared with conventional freezing in a carboy, the Celsius system was far superior, freezing solution homogeneously and 10 times as rapidly, according to Olsen, who noted that the Celsius system also improves solution-storage conditions including homogeneity for frozen storage of biological solutions with possible ramifications for product quality.
“Developing fully human antibodies is a big challenge and a continuous learning process,” according to James Thomas, Ph.D., vp for process and analytical sciences at Amgen (www.amgen.com). His team has a systematic approach to the production of biologicals, focusing on antibodies and antibody-like molecules.
With a large number of different therapeutic antibody development projects under way, Amgen seeks a broad encompassing approach to manufacture and analysis. Dr. Thomas and his colleagues are exploiting molecular similarities to achieve high quality and efficiency in production technologies. "While every antibody is somewhat different due to differences in primary sequence and post-translational modifications, the similarity of these molecules presents a compelling case," he asserted.
Although approximately 80% of antibody molecules within a given IgG subclass are homologous, the different amino acid residues responsible for antibody specificity can result in profoundly different physical and chemical properties. In the case of Fc fusion molecules, the energetics of folding can be dramatically affected by the peptide or protein used as a fusion partner. The challenges have generated skepticism as to whether it would be possible to build a generic downstream platform.
“Working through these problems has required the concerted effort of a large team,” Dr. Thomas stated. “The fact that we’re involved in a number of antibody projects means that we can leverage our resources in a continuous learning environment; working through platforms to achieve better and better solutions.”
According to Datamonitor, the market for therapeutic antibodies was projected to be well over $30 billion by 2010. Such a phenomenal expansion should leave plenty of room in the industry for alternative expression systems.
Alternative Expression Systems
While FDA-approved antibodies are produced exclusively in mammalian cells (mainly modified CHO cell lines grown in bioreactors), there has been no dearth of proposed alternatives. Early in the history of therapeutic antibody development it was thought that E. coli would prove to be the organism of choice, but the problem of inclusion bodies and the inability of bacteria to carry out secondary modifications made other alternatives more attractive.
Such alternatives include the yeast Pichia, transgenic plants and animals, algae, and filamentous fungi. However at this point, only baker’s yeast, Saccharomyces cerevisiae, has passed the threshold of FDA approval for biologicals, according to Nalini Motwani, Ph.D., founder and president ApoLife (www.apolife.com). Products already on the market include hepatitis B vaccine, human insulin, and other recombinant proteins.
Dr. Motwani discussed her company’s efforts to generate antibodies in yeast, which is well within the realm of GRAS (generally regarded as safe) biological systems. Currently, ApoLife uses a twin cassette system to produce antibodies, which employs two promoters for activating light-chain and heavy-chain genes in the same plasmid. The plug-n-play vectors are easily manipulated, and the resulting functional IgG secreted from stable yeast cell lines can be used to screen antibodies for discovery and lead selection.
“Our technology and methods constitute yeast expression system for full IgG secretion,” according to Dr. Motwani. “The intimate knowledge of the genetics of this organism allows rapid strain development for small- or large-scale customers.”
Moreover, the rapid screening of antibodies and elimination of unproductive approaches translates into rapid turnaround and shortened time for discovery. ApoLife carries out cloning and expression in yeast and small-scale production. Upscaled production for clinical trials is accomplished through outsourcing partners. The ApoLife technology can also be linked to phage-display systems and to fusion proteins for rapid antibody production.
The main barrier to the adoption of alternative expressions systems by the biotechnology industry has been an unwillingness to abandon mammalian cells, which have a long history of approval in the antibody production field, for a platform that will be evaluated with extreme concern by the U.S. and European regulatory agencies. The benefits of Saccharomyces should serve as an impetus for companies to take the risks associated with an unproven but promising approach.
Antibody Purification Needs
“Companies continue to strive for improvements in yield and quality of antibody products, while also seeking faster process implementation and scale-up,” according to Fred Mann, Ph.D., technical marketing manager of the bioprocess division at Millipore (www.millipore.com). “With these needs in mind, we have focused on improving our purification technology through the development of disposable modules for clarification and prefiltration of biological solutions.”
Millipore recently carried out an industry-wide survey of managers in the field of antibody process development and manufacturing to assess their performance indicators and how these could be improved. The survey showed that most companies are focused on cycle time, deviations, and manufacturing cost. The most prevalent reason listed for deviations were procedural issues, and the solution initiated most frequently by the companies surveyed was improvements in error proofing.
Another area investigated in the survey was operational improvements that managers felt would best help them to meet their manufacturing goals. Innovations in cell culture and fermentation were the most frequently cited, supporting the belief that even though yields per cell have exploded in recent years, there is still much room for further increases. Next in order of priority came process chromatography followed by clarification.
For several years Millipore has offered Millstak+HC filtration media units, a disc-shaped technology comprising filtering systems for the clarification of cell culture harvest fluids. However these filters have some disadvantages, including bulkiness and handling difficulties.
To meet the challenges posed by Millstak+HC, Millipore has introduced its Pod platform as an integrated media housing unit consisting of lightweight, self-contained, disposable modules. The system, with its disposable feed ports and fittings, delivers improved flexibility and ease of use. It can be configured to handle 10,000- to 20,000-L batch volumes with greatly reduced buffer and washing requirements.
Because of these improvements, overall costs of operation can be considerably less than those obtained with traditional lenticular filters. For example, taking water and buffer cost savings into account, the Pod system was shown to be 13% less expensive compared to the lenticular system, according to Millipore.
Disposables are becoming more popular in the area of cell cultivation. So, the move into the downstream processing area is a natural extension of this convenient and labor-saving technology.
In addition to its products for media clarification, Millipore is also focused on the purification of antibodies. Its lead product, ProSep-vA Ultra, is a higher-performance chromatography media for Mab capture. Protein A is the most widely used antibody purification molecule, and in the last few years, variant forms of the molecule have been introduced that allow for improved purification.
“While there are other antibody purification molecules, such as Protein L and Protein G, available, Protein A continues to be the most versatile and adaptable,” Dr. Mann said. “Our ProSep controlled pore glass-based Protein A resin is also designed to aid process development and implementation by providing high capacity and throughput coupled with robust scalability.”
“Recent advances in the upstream production of antibodies have increased the time-to-market and cost-efficiency pressures on the downstream processes,” added Hans J. Johansson, Ph.D., senior scientist, R&D, at GE Healthcare Bio-Sciences (www.amershamhealth-us.com).
To address these needs, new media with improved capacity, higher productivity, and longer life span are highly favored. GE Healthcare recently launched new affinity and ion-exchange media for Mab purification. These include MabSelect SuRe affinity chromatography media, CaptoS, a cation exchange chromatography product, and Capto Q, an anion exchange media.
MabSelect SuRe is a genetically engineered form of protein A with much improved properties of stability, higher capacity, and extended life time, according to Dr. Johannson. The modifications included a number of amino acid substitutions that have improved the alkali stability of the product without altering its other characteristics.
Capto S, a strong cation exchanger with a sulfoethyl ligand, is a particle of 90-181m average size that uses dextran as a size extender. Capto Q, a strong anion exchanger with a quartenary amine ligand, is a similar 90-181m particle with a dextran size extender. The new offerings provide higher capacity, improved flow, and alkaline protease stability, all features that generate significant economic impact. In fact, Dr. Johannson’s calculation demonstrate a 40% reduction in running costs of the purification technology as opposed to earlier products.
The industry faces a number of unique questions in the near future, according to Dr. Berthold. The vast amount of material required for antibody-based pharmaceuticals is unprecedented in the history of the drug industry. Whereas the total need for a hormonal drug may have run to kilograms per year, antibodies need hundreds of tons per year. From an industrial and economic standpoint, they will be treated as commodities, and the cost of goods becomes a much more significant component of the overall financial equation.
These gargantuan demands place pressures on the living systems that generate biological molecules, as companies push for higher output. Dr. Berthold believes that there is a biophysical frontier of solubility that has been reached for synthetic rates for mammalian cells, constituting a plateau that cannot be breached. And even if alternative systems, such as yeast, could produce more grams per liter, which is in doubt, one is still faced with the challenges of secondary modifications that all nonmammalian alternatives are saddled with.
The extremely dense nature of these solutions, however, forces new challenges in downstream purification and formulation that comprises two-thirds of the cost of goods.
While none of these issues are insurmountable, all place great demands on companies to explore more innovative resolutions for the scientific and economic demands.
Additionally, the vast market for antibodies is not likely to abate within the foreseeable future. Antibodies are honed by the forces of evolution to be superior regulators of cellular activity and their long half life in the circulation makes them unique therapeutic molecules. Despite the high development costs, antibodies’ exceptional properties place them in the forefront of new drug development and they will continue to figure in the treatment of cancer and other complex human ailments for years to come.