May 1, 2006 (Vol. 26, No. 9)
Overcoming the Technical, Regulatory, and Financial Problems Plaguing the Industry
The production of therapeutic proteins in transgenic animals and plants seemed like a slam-dunk about a decade ago. The potential benefits of transgenics when compared with traditional cell culture or fermentation are well known. Nevertheless, no human therapeutic produced in a transgenic animal, plant, or fungus has yet been approved.
Although a few companies have survived and even fewer are thriving, numerous regulatory, technical, and financial problems have beset the transgenics industry.
Larry Grill, Ph.D., a founder of and now consultant for Large Scale Biology (LSBC; www.lsbc.com),says the large pharmaceutical and biotech companies have little incentive to invest in transgenic technology. “People complain about the high cost of drugs, but that dissatisfaction is less of a concern for the pharmaceutical companies than the fear that a new process could potentially lead to new lawsuits against them.“
An LSBC project, production of the natural protease inhibitor aprotinin, solved a major problem with the proteins manufacture. Aprotinin is extracted from cow lungs. Concerns about the use of animal-derived products as human drugs, for example the potential of infection with the bovine spongiform encephalopathy, has made the supply of aprotinin unreliable. LSBC produced the material in plants, thus avoiding any potential for contamination from mammalian infectious agents. Their product was identical to that produced in cows and came through at higher purity. “The problem was the protein had to go through full clinical trials because the impurity profile was different from the Bayer product,“ says Dr. Grill.
Another great transgenics idea, the edible pharmaceutical, is barely limping along. “The biggest problem with edible medicinal food plants is consistency,“ says Frank C. Gwazdauskas, Ph.D., a professor in the department of dairy science at VirginiaTech. “And with transgenic animals, unless you can get stable transmission of the target gene from generation to generation, expression levels can be problematic. A lot of sorting and segregation takes place after the first generation.“
Regulatory issues are far from settled. On February 22, the EMEA refused the marketing application for ATryn (antithrombin alfa), GTC Biotherapeutics(www.gtcbiotherapeutics.com) infusion drug for surgical patients with congenital antithrombin deficiency. ATryn is produced in goat milk (GTC also uses cows to express recombinant proteins).
After excluding data from pregnant patients, EMEAs Committee for Medical Products for Human Use determined that not enough patients were enrolled in a key clinical study to support approval, according to a GTC press release. GTC and development/marketing partner Leo Pharma have re-submitted their application.
The decision not to approve ATryn had nothing to do with its production platform, says Harry Meade, Ph.D., CSO. Rather, EMEA cited a slight change in downstream purification between the submitted purification technique and that used for the Phase III study, as well as the need for more patients.
The regulatory mantra for transgenics has long been that nobody wants to be first, lest they incur undue regulatory wrath. “Everyone is waiting for the first company to get through the gate,“ says Dr. Meade. GTC, on the other hand, prides itself in its pioneering regulatory position. Other GTC pipeline products include recombinant human serum albumin, a malaria vaccine, and a CD137 antibody to solid tumors.
GTC has survived the treacherous road to commercializing products from transgenic animals by dedicating itself to upstream expression and production and focusing on difficult-to-make products. For example, ATryn is difficult to express in cell culture, but relatively easy to produce in milk. More than one transgenics company has gone out of business because of large-scale purification hang ups, according Dr. Meade.
GTC entered agreements with traditional biotech companies for production of certain therapeutic proteins in cow or goat milk. None of these products has reached the clinic since the sponsors decided to stick with cell culture. The research agreements have, however, helped GTC stay afloat in a business that burns through cash. The companys decision to outsource purification has freed up capital for research, development, and ongoing operations.
Biolex Therapeutics (www.biolex.com) works on producing approvable transgenic therapeutic proteins in plants. Its LEX System manufacturing platform uses permanently transfected lemna plants, or duckweed, to express traditional biotherapeutics, such as Mabs and interferons.
Biolex lead product, an injectible, controlled-release alpha interferon, is under development with delivery firm OctoPlus (www.octoplus.nl). The product, Locteron, has completed Phase I human testing and is administered every other week as opposed to once a week for traditional interferon. Biolex is manufacturing Phase II material and expects to enter that phase by the end of this year.
Biolex recently doubled the manufacturing capacity at its Pittsboro, NC, facility, a project that took about half a year. According to Glen Williams, senior vp of operations, this capacity should serve Biolex through Phase III.
Compared with cell culture or fermentation, Biolex lemna expression system is a model of simplicity. Large cell culture facilities take years to build, cost hundreds of millions of dollars, and require exquisitely precise timing since drug approval is never a sure thing. “The LEX System allows manufacturers to install and scale up capacity quickly at far lower capital expense,“ says Williams. Moreover, downstream processing is simplified due to more predictable feedstock and the total absence of pathogenic viruses or microbial toxins.
Since its inception, Biolex focused on development of difficult-to-manufacture proteins rather than on promulgating novel transgenic technology for its own sake. This strategy attracted the attention from such biomanufacturers as Centocor, Medarex, and MedImmune for lemna-based manufacturing of protein products. The deal with Centocor covers up to 10 protein therapeutics. Medarex has been so pleased with progress that it has expanded the collaboration. In November, 2005, Biolex teamed with Kringle Pharma to produce that companys NK4 hepatocyte growth factor fragment for treating cancer.
Man Cannot Live on Science Alone
Every transgenics company has wonderful science, but successful pharmaceutical companies do not live by science alone. To persevere companies must focus on products, not technology. And it pays to operate in large, hot therapeutic areas like diabetes.
In November of last year, SemBioSys Genetics (www.sembiosys.com)announced proof-of-concept for a plant-produced insulin, using the companys Stratosome expression system. A study, published in the January 2006 edition of Plant Biotechnology Journal, described insulin production in the seeds of the plant Arabidopsis thaliana.
According to CEO, Andrew Baum, plant-based production could reduce capital costs for an insulin facility by 70%, and cost-of-goods by more than 40% compared with current production methods. The next step, says Baum, is to express the peptide in safflower, the SemBioSys commercial production system of choice.
Stratosome expresses proteins in plant seeds, attaching them to oily molecules, naturally occurring in seeds. Proteins are stable in that form more or less indefinitely. Purification is simplified by the separation of the oily structures from homogenized seeds. Similar oil-protein structures may be used to deliver protein drugs by mouth.
Greenovations (www.greenovation.com) moss bioreactor is reminiscent of the Biolex Lex expression system, but there are many differences. Instead of intact plants the expression system is cultured, suspended cells of the well-characterized Physcomitrella patens moss. Physcomitrella can be made to glycosylate proteins in human-like fashion, which reduces concerns about immunogenicity. In addition, the cultures are easily scaled to several thousand liters. Since proteins are excreted into the medium, downstream purification is easier than with mammalian cell cultures.
According to the company, ion exchange is typically the first step. And, since it is a continuous system, high yields are possible —up to 30 mg/L per day. In October, 2005, the company introduced a polyethylene glycol-mediated transient gene expression system for gene expression in moss with or without the use of viral or agrobacterial vectors.
Transgenic plant biotechnology relies on either stable genetic transformation or transient infection by viral vectors. Although transient expression is much faster, it has been limited by viruses low-infectivity and their small transgene-carrying capability.
Last year, Icon Genetics(www.icongenetics.com) published a paper in Nature Biotechnology on a novel transient gene expression system for plants. Icons method uses Agrobacterium as the infective agent for viral replicon delivery, which introduces the gene of interest in all mature plant leaves simultaneously, leading to high gene expression levels. According to the company, indefinitely scalable transfection method is rapid, achieves high yields, and is capable of human-like post-translational glycosylations. The companys first-generation platform achieves up to five g per kg of biomass after only one to two weeks of cultivation. Bayer acquired Icon in January.
Chicken or the Egg?
Origen Therapeutics (www.origentherapeutics.com) produces polyclonal antibodies in the whites of chicken eggs. The company believes that polyclonals have certain advantages over monoclonals, particularly for infectious diseases where drug resistance is rampant.
The company can also produce monoclonals by targeting gene expression in specific tissues. For example Mabs are expressed in the tubular gland cells of the oviduct and deposited in egg whites. A paper in Nature Biotechnology last year demonstrated that egg-derived proteins compare favorably with those produced in CHO cells.
Surprisingly, this expression platform is transient. “Embryonic stem cells in chickens are different from those in mice,“ explains Bob Kay, Ph.D., CEO. In mice the progenitor cells contribute to both somatic and germ cell makeup, but in chickens the germ cells are unaffected. “Chimeric birds cannot be used to generate permanent cell lines.“
Current yields are in the 3-5-mg per egg range, which Dr. Kay expects will reach 50-100 mg once the expression system is optimized. At these levels, a quite manageable 5,000-bird flock can produce about 100 kg of protein per year. Downstream processing begins with sterile egg processing machinery that cracks open and separates the eggs, followed by traditional protein separation techniques.
Origen is not alone in the quest to produce therapeutic proteins in chicken eggs. Viragen (www.viragen.com) announced in January that it expressed significant quantities of interferon beta-1a in the whites of eggs laid by transgenic hens. Viragen collaborated on the project with the Roslin Institute and Oxford BioMedica.
What the Future Holds
For years the industrial enzyme industry pointed its finger at therapeutic biotech companies and asked why the latters products are so expensive. Indeed, biomanufacturers have a lot to learn from companies like Dyadic (www.dyadic-group.com) that produce enzymes and proteins for pennies per gram. According to chief scientist Glenn Nedwin, Ph.D., “If you cant make multiple grams per liter with industrial enzymes, you dont have a product.“
Dyadic uses filamentous fungi to make industrial enzymes and says it has the capability to produce Mabs in this host. The company is eager to demonstrate its protein-producing technology for other therapeutic proteins and intermediates as well.
Fungi are in some aspects the ideal host for bioproteins. Like E. coli, they are inexpensive to grow and scale up, but unlike bacteria they glycosylate, a definite plus for humanized Mabs. Transfection is somewhat more difficult with fungi than with bacteria, but Dyadic has optimized a fungus to accept foreign genes and churn out proteins with relative ease.
Most industrial enzymes are expressed at between five and 20 grams per liter of culture volume. Some, like amyloglucosidase, are produced at close to 100 g/L, a volumetric efficiency that should be the envy of anyone who has ever used large-scale mammalian cell culture.
Robert F. Kennedy once said, “I look at the world and wonder, why not?“ All the elements of success are in place, including breathtaking science, investment dollars, and some of the smartest minds in the business. One begins to wonder, however, whether transgenics time has not perhaps already run out.
“Adoption of new technologies is always slow,“ says Baum. “One always tends to underestimate how difficult it will be and how long it will take.“