Proponents of transgenic technology point to process economics as the most compelling argument for its use. Part 1 of this series, introducing the technology and major players in the transgenics arena, appeared in BioPharm's March 1999 issue.
Transgenic technology offers a costeffective route to the production of therapeutic and nutraceutical proteins, an area of increasing importance to the global pharmaceutical and health care industries. This article focusing on process economics is the second in a series exploring the technology and its commercial development and potential.
The production of human proteins and peptides in the milk of transgenic livestock has several benefits, but its greatest advantage over other methods may lie in process economics. Alternative production methods - such as fermentation, mammalian cell culture or, in the case of blood proteins, purification from plasma have their own strengths but often cannot produce raw material or finished product on the large scale as cost effectively as transgenic livestock can.
Fermentation. Human therapeutic protein production is often complicated by required, complex posttranslational modifications that allow the protein to be biologically active, Bacterial fermentation, although a relatively cheap and well-established production method, cannot produce such modifications, which may include, for example, the addition of carbohydrates. Yeast-based systems are capable of some such processing but are still not ideal for production of fully bioactive mammalian proteins.
Cel culture. One answer to those shortfalls, and the closest parallel to fermentation-based systems, is the use of mammalian cell culture. Posttranslational modification is carried out in cell cultures, and bioactive proteins can then be harvested. However, as will be seen, the costs involved in setting up and maintaining a large-scale mammalian cell culture facility are relatively high.
Transgenics. With transgenic livestock, proteins expressed in the mammary gland can be modified appropriately for full activity. For example, PPL Therapeutics has successfully expressed the blood-clotting protein fibrinogen in the milk of transgenic sheep (Figure 1). Fibrinogen is a complex plasma protein (360 kDa) made up of two sets of three different polypeptide chains, and thus it requires assembly and modification for full bioactivity. It is synthesized naturally in liver cells, where the six chains are assembled and linked by 29 inter- and intrachain disulfide bonds. Fibrinogen produced in the milk of transgenic animals has been shown to be correctly assembled and to interact with thrombin to form clots (1). The assembly, from a multigene construct, indicates the potential of transgenic technology to produce complex human proteins. With high-level expression of fibrinogen (5 g/L) in sheep, a flock was established to produce material for clinical trials. Transgenic technology contrasts with various mammalian cell culture systems, which have also produced apparently functional human fibrinogen but at expression levels 1,000-fold lower than the transgenic system (1).
Blood proteins. Donated plasma is an obvious source of human blood proteins, providing them in a fully modified and active form. Of course there are other problems. Safety concerns include the possible transmission of human pathogens such as HIV or prions. Eliminating such contaminants, where possible, adds significantly to the cost of plasma-derived proteins. The seriousness of the safety issue is reflected in the May 1997 recommendation by the United Kingdom's Committee on Safety of Medicines to phase out the use of UK-derived plasma for fractionation because of the theoretical risk of new-variant Creutzfeld-Jacob's Disease transmission (2). In the United States, the Surgeon General's advisory committee on blood safety and availability has recommended that the Department of Health and Human Services (HHS) accelerate the transition from plasma derivatives to recombinant analogues where medically appropriate, and in August FDA banned blood donations by anyone who'd been in the United Kingdom between 1980 and 1996 (3,4). Furthermore, it was recommended that HHS support research into alternatives to plasma-based technologies for conditions such as alpha-1-antitrypsin (AAT) deficiency and von Willerbrand's disease.
Comparative Production Costs
Two key economic considerations behind choosing a method for producing a protein or peptide are the cost of raw product and the capital investment needed to obtain that product. Figure 2 indicates the costs of raw product from cell culture, plasma, yeast, and transgenics. The most marked difference is between transgenics and mammalian cell culture: The costs of raw product are approximately 10 times lower for transgenics - $120/liter compared with $10/liter (5). Production by yeast fermentation, although less than half the cost of cell culture, is still more expensive than transgenics. The cost of raw product from plasma is also high, but it is important to remember that more than one product might be purified from that raw material.
The disparity in raw product cost is linked closely to the capital investment necessary to establish production facilities for the various technologies (Figure 3). As for raw products, the costs associated with establishing a transgenic production facility are lower than for cell culture, yeast fermentation, or plasma facilities. The differences can be illustrated by looking at specific examples.
In 1996, Novo Nordisk spent $100 million to build a 120,000-ft^sup 2^ facility in Clayton, NC, for the production of human insulin. At a cost of $59 million in 1997, Biogen completed a new facility, also in North Carolina, for the production of Avonex. In that same year, Covance dedicated its multiuse facility for more than $50 million. PPL's fully integrated production facility in Scotland cost about $17 million and includes sheep housing, a milking parlor, and 15,000 ft^sub 2^ of GMP-- qualified purification suites. It is producing more than 1 kg of clinical-grade AAT per week (over 50 kg a year). A mammalian cell culture facility capable of producing the same quantity of finished product could be up to 10 times more expensive to establish.
Costs of Transgenic Production
If the transgenic route is chosen, a number of factors determine product cost. They include the scale of production and the purification process required, and they influence decisions made in the choice of species and development time.
Scale of production. Transgenic technology offers flexibility of scale for protein production, depending mainly on the species selected. Pilot studies determine the effectiveness of gene constructs (the human gene of interest combined with a promoter gene that directs expression in the mammary glands) and establish that a protein indeed can be expressed in milk. Those studies are carried out in small animals, normally mice or rabbits. That work obviously is less costly than working with large animals. For certain products (such as Pharming's alpha-- glucosidase treatment for Pompe's disease) transgenic rabbit production is sufficient for commercial development because the amount of product needed is relatively small. That Pharming product has entered clinical testing in humans with material derived from rabbits. Genzyme Transgenics also has used small-animal production systems, and it intends to enter the clinic with a malaria vaccine produced in transgenic mice (6).
When larger quantities of product are needed, Figure 4 shows how costs per kilogram reduce as the scale of production increases. Table I provides more detail: the distribution of costs for producing a transgenic product at 10-kg and 1,000-kg levels. At the 10-kg level, the greatest costs come from the purificiation and processing plant and from labor. At the 1,000-kg level, the relative expenses of labor and plant are reduced. Thus, as the production scale increases, transgenics becomes cheaper because the cost of milk production is low, and there is no need to invest in a large infrastructure (large stirred tanks, air handling systems, and so on) to manufacture the raw product.
Expression levels. A key determinant in the cost of a transgenics program is the expression level of the protein in the animal's milk. That depends on various factors including the site at which the introduced gene is inserted into the genome. Figure 5 shows relative costs per gram across a range of expression levels from 0.01 g/L to 100 g/L. For commercial programs needing up to 100 kg of product, an expression level of 1 gIL is desirable. At that level, the extra development time needed to increase expression from 1 gIL to 10 g/L would save only 19% of the cost, so it probably would not be justified. Expression levels in excess of lg/L are routinely achieved for transgenics processes, whereas those in mammalian cell culture are generally less than 0.5 gIL.
The volume of milk produced by transgenic mammals is another key factor influencing process economics. For many products, sheep or goats can be economically and therefore commercially viable. When extremely large quantities of product are required (tons), the additional time needed to generate transgenic cattle may be justified. Expression levels in cattle have been good: 2.4 gIL for a transgenic cow carrying the gene for human alpha-lactalbumin, with each producing up to 10,000 liters of milk per year. One cow could by herself produce 24 kg of recombinant protein in a year. A small herd of cattle (100) could produce over two tons. In cell culture applications, the rule is that larger bioreactors provide for lower overall costs per gram. The same is true for transgenics.
Purification yied. A final cost driver in transgenic protein production is the yield of the purification process. Figure 6 shows that improving the purification yield from 10 to 90 percent only halves the actual cost of the product. With low-cost raw materials, the cost of transgenic products is primarily incurred by downstream processing. To reduce the cost of transgenic-derived products, the use of cheaper process steps can be considered, even at the expense of yield. Process yield in purification is not as critical when the cost of raw product manufacture is low. That is certainly not the case for mammalian cell culture processes, where overall process yield can be a dominant issue in deciding whether to commercialize.
Transgenic AAT Production
The process economics of transgenic production can be illustrated by looking briefly at our company's lead product, AAT. The protein, which has now completed Phase II clinical trials for the treatment of cystic fibrosis, has been produced at high levels in the milk of transgenic sheep (7,8). The production flock stably expresses the protein at around 12 g/L. That clearly exceeds the 1 g/L rule of thumb and sets the basis for economic production of the protein by transgenic means.
Production of human therapeutic and nutraceutic proteins by transgenic livestock offers some advantages over alternative methods. Like mammalian cell culture, the technique can produce complex proteins that require posttranslational modification and cannot be produced in yeast or bacterial cultures. However, the costs involved in transgenic protein production are significantly lower than those for mammalian cell culture. As has been described in this article, those costs can be reduced further by optimizing expression levels, lactation volumes, and purification yields.
[Reference]
References
[Reference]
(1) D. Prunkard et al., "High-Level Expression of Recombinant Human Fibrinogen in the Milk of Transgenic Mice," Nat. Biotechnol. 14 (7) 867-871 (1996).
[Reference]
(2) U.K. Department of Health, "Committee on Safety of Medicines Complete Review of Blood Products," 13 May 1998 (www.doh.gov.uk/cjd/blood.htm). (3) Meeting of the Advisory Committee on Blood Safety and Availability, Surgeon General's
[Reference]
Office, 27-28 August 1998, Washington, DC. (4) A. Manning, "Mad-Cow Fear Leads to Blood
Donation Ban," USA Today, 18 August 1999. (5) N. Rudolph, "Technologies and Economics for Protein Production in Transgenic Animal Milk," Gen. Eng. News, 17(18), p. 16 (1997). (6) M. Young, "Genzyme Transgenics" presentation at the Transgenic and Cloning: Commercial Opportunities conference, 2426 June 1998, Washington DC.
(7) G. Wright et al., "High Level Expression of Active Human Alpha-1-Antitrypsin in the Milk of Transgenic Sheep," Bio/Technology 9, 830-834 (1991).
[Reference]
(8) A. Carver et al., "Transgenic Livestock as Bioreactors: Stable Expression of Human Alpha-1-Antitrypsin by a Flock of Sheep," Bio/Technology 11,1263-1270 (1993). BP
[Author Affiliation]
Paul Rohicht is the U.S. business development manager for PPL Therapeutics Inc., 1700 Kraft Drive, Suite 2400, Blacksburg, VA 24060, (610) 682-0444, fax (610) 682-0557, prohricht@ ppl-therapeutics.com.
Комментариев нет:
Отправить комментарий