Creating Therapeutic Proteins from Bioengineered Systems

We’re not talking about green in the sense Shakespeare meant. To apply a contemporary meaning to the phrase, think of a field of lettuce as a protein-production lab – and the implications of that. It’s an idea that merits support.

Abstract

If scientific and social obstacles can be overcome, the use of plants and animals as living production facilities to effect protein expression could be the key to reducing R&D costs – and ultimately, the cost of therapies that will benefit millions of people.

At least 50 recombinant proteins, vaccines, or monoclonal antibodies are in phase 3 trials (Tufts 2003). Considering that research and development spending has increased 12-fold during the last three decades, the current mantra for research and development is “faster, better, cheaper.” Most of these efforts are directed toward reducing preclinical costs and duration of clinical trials, and improving clinical success rates.

In part, the high cost of biologics is tied to the complex nature of their manufacturing process. Cell-culture facilities are not cheap to build or to operate. Thus, the production of antibodies and other therapeutic pharmaceutical proteins in whole animals and nonmammalian transgenic platforms¹ can be a powerful production toolbox. In this emerging science, nature serves as a host for protein expression. So-called transgenic plants and, to a lesser extent, animals and insects become, in essence, little pharmaceutical factories. The potential for cost-saving is enormous.

Richard V. McCloskey, MD

R&D COSTS

Most people making payment and coverage decisions focus on the cost of goods. Industry, then, must focus on the cost of ingredients and the time spent to make the goods.

According to the annual report of the Tufts Center for the Study of Drug Development, manufacturing capacity “…will improve as new facilities come online and more efficient methods of production are adopted” (Tufts 2003). The assumptions that underlie this statement are monumental. Both manufacturing costs and the time required in the manufacturing process must be reduced, or attempts at reducing the time it takes to obtain regulatory approval will be blunted.

Based on the assumption that a facility can produce 2 million liters of cell culture fluid containing an antibody a year, the per-facility cost is about $800 million. Cost alone thus provides a sufficient incentive for generating alternative (and, by implication, less expensive) ways to make these products. Notably, if the acquisition and administration of biologic agents are unaffordable to patients — and if the healthcare system refuses to assume the costs — there can be no research, development, or therapy.

Although many bioengineered platforms in plants, animals, and insects are in early stages of R&D, these areas must be explored with the specific intention of reducing healthcare costs. There are signs of progress: antithrombin III from transgenic goats has reached regulatory submission, gastric lipase from corn is in clinical trials, and ovarian cancer antibody from chloroplasts and some vaccines are close to clinical trials.

CHALLENGES OF SUPPLY AND DEMAND

We must be able to meet patient demand for desirable products. Bioengineered systems merit both support and development. No system of production should be excluded from research and development without careful consideration. Evaluation of a given product necessitates developing several platforms simultaneously, because the more data that are available to determine the value of the product, the better. Those in the transgenic industry need to produce data for multiple potential products in multiple systems to compete with the vigorous efforts that are underway to reduce the costs of goods in mammalian production systems.

About 14 percent of therapeutic agents in clinical trials are biotechnology agents. The biologics market is poised to grow explosively, especially relative to monoclonal antibodies. Nevertheless, the manufacturing literature is replete with alarms warning of an existing shortfall in production capacity — a deficit that will only grow worse over the next five years. The paradox is that heightened need for a product contributes to an increasingly severe production bottleneck.

Infliximab provides a case study. If infliximab is administered six times a year to 10,000 rheumatoid arthritis patients weighing 80 kilograms (176 pounds) at a dosing regimen of 5 milligrams per kilogram, 24 kilograms of antibody would have to be produced each year.

In a worst-case scenario, an efficacious biologic could not be brought to market in the time frame permitted by the clinical program; 20 to 50 products now in development might end up in this cul de sac (Lias 2001). In some cases, patient registries have become necessary to distribute limited supplies of a product during a period of manufacturing expansion.

STRATEGIES FOR TRANSGENIC USE AND COST REDUCTION

Transgenic plants offer the potential for lower production costs, reduced capital expenditures, and greater manufacturing output. Although the science is complicated, the concept is simple: protein expression is achieved when a human antibody-producing gene, such as an influenza antibody, is permanently incorporated into a plant genome. In time, the plant reproduces that protein.

Table 1 highlights the reasons biotech companies are drawn to developing human therapeutics through plant transgenic systems.

TABLE 1.

Potential advantages of transgenic plant systems

No human pathogens

No mammalian contaminants

Expandable platforms

Low overhead

Speed

Efficient use of capital

Low biomass

Stability

Human or humanized antibodies (or their components) for the treatment of chronic diseases (as opposed to acute events) are most amenable to these systems, though some one-time treatments (e.g., treatment for drug overdose or infectious disease) are as well.

A particularly good target for reducing production costs would be long-term therapies requiring tens of kilos of antibody per 10,000 patients per year. Examples of this include: therapies with a dosing regimen of 1 or more mg/kg; products for which the route of application requires excessive protein to cover the target adequately; and products that are administered to hundreds of thousands of patients.

An example of the savings that can be attained (and ultimately passed on to patients and payers) can be seen in the figure. Use of transgenic plants to produce antibodies have resulted in 4- to 7-fold savings over producing them in bioreactors.² The difference between plant and lab costs is due largely to the lack of purification expenses (incurred in the removal of mammalian contaminants) in transgenic plants. Plant bacteria are removed in the downstream process.

In traditional platforms for biopharmaceutical development, production and purification tend to consume most of the capital costs. Estimates vary from 20 to 75 percent of total capital costs for production and from 25 to 80 percent for purification. Transgenic plants, many believe, permit a large reduction in purification expenses, although some are skeptical of this analysis.

To assess the value of transgenic systems appropriately, additional data are needed regarding several aspects of these systems, such as scalability. Additional examples of issues that warrant further study are presented in Table 2.

TABLE 2.

Data needed to assess the value of transgenic systems

Protein structure

Glycosylation

Apples-to-apples cost comparisons

Stewardship attitudes: public, regulatory, and food industry

SCALE-UP POTENTIAL

Compared to animal transgenics, plant-based systems have the ability to scale-up production to large volumes. The upshot is that mass production — a concept not necessarily associated with the biotech industry — can help to reduce costs.

For many transgenic products, biologic activity and potential immunogenicity have yet to be determined, as opposed to proteins produced by a “good” mammalian cell line or recombinant methods.

Transient gene expression may be a shorter path to making a product, but the required plant virus complicates the process. A permanently transformed plant makes the same product all the time. Method selection probably would vary from one desired product to another.³

Plant chloroplast platforms may avoid transgene dissemination via pollen while increasing secretory protein as much as 300-fold that of nuclear transgenic approaches (Staub 2000). Many other strategies have been applied to minimize protein loss or degradation (Sharp 2001) The duckweed (lemnaceae genera) system does not use open fields, and it has the advantage of being a contained procedure with similarities to current mammalian technologies. Platforms using duckweed, moss, and algae use light and air as carbon and energy sources, and they use simple salt solutions for growth media.

RELEVANCE TO HUMANS

There is progress toward the important goal of getting active and passive immunity data in humans. Trials are variable in duration leading to some delays in data analysis, but there is a growing body of information relative to humans (Rishi 2001).

Human antibodies, or parts thereof, can be produced in plants as biologically active immune substances. These “plantibodies” can be administered when immediate protection from an infectious disease is needed — for example, when antibodies are needed for protection from rabies, varicella, or diphtheria. These products are in critically short supply, primarily because they are derived from human plasma.

Alternatively, an active immune response can be induced — for example, to protect against rabies or hepatitis. An edible vaccine would have tremendous advantages in parts of the world where distribution of conventional vaccines is impractical.

WHERE TO FROM HERE?

Unquestionably, the prospect of reducing capital outlays propels interest in making proteins and antibodies through transgenic plant systems. Lower costs translate to affordable products for more patients.

There are barriers to overcome before this scenario becomes more practical, however. For some platforms, data do not exist or cannot be translated easily for comparison with current real costs. Many costs, such as those for quality assurance and quality control, may be inversely related to batch size. The simpler the purification scheme, the greater the savings — again, compared to real-time production for a real-time product.

The prospect of growing the raw material in the field, for instance, is financially appealing compared to the cost of a new sterile fermentation suite. Embedded in the thinking about transgenic plants is the notion of capitalizing on the savings as scale increases, though there are not yet enough data to substantiate this.

In the United States, the regulatory pathway concerning transgenic platforms is a work in progress. (The European Union prohibits genetic modification of plants.) This is complicated by a commonly held negative perception that genetically modified plants are primarily for the benefit of the manufacturers or industries that are involved in food production. Such misunderstanding makes it difficult to predict legal and social developments.

The advantages of using transgenic plants to develop biotech agents must be presented responsibly and weighed carefully against the associated risks. Most of these issues can be addressed only when a product is available for testing. We are about a year away from seeing responses to these issues, which will undoubtedly shape the course of these projects.

Cost-effectiveness, cost-benefit, and cost-savings data are important to payers, as is the cost of goods. The imperative must be to reduce the cost of the biological product itself. We have a relatively good idea what it costs to produce new mammalian cell-culture facilities. Reducing the cost below current levels will open the benefits of these potent, life-improving treatments to the chronically ill.

FIGURE.

Comparative capital costs

Production of 300 kilograms per year of antibody in various platforms, in millions of dollars

SOURCE: WATLER