Pharmaceutical Majors Build Positions in Monoclonal Antibodies

November 2, 2006
Patricia Van Arnum

Patricia Van Arnum was executive editor of Pharmaceutical Technology.

Pharmaceutical Technology, Pharmaceutical Technology-11-02-2006, Volume 30, Issue 11

AstraZeneca's purchase of Cambridge Antibody Technology and Merck's acquisitions of GlycoFi and Abmaxis are the latest efforts by pharmaceutical majors to build critical mass in biologics capabilities.

Strong growth is projected for biologic-based drugs, and the major pharmaceutical companies are moving forward to add biologics production capabilities and product pipelines. Several companies have announced deals in 2006 to take advantage of growing opportunities in this field.

There are 418 biotechnology products either in human clinical trials or under review by the US Food and Drug Administration (Rockville, MD) estimates the Pharmaceutical Research and Manufacturers of America (Washington, DC). Monoclonal antibodies account for 38% or 160 of these products and recombinant proteins and hormones account for 10% or 43 products (1).

Big Pharma's sales of biologics products are expected to increase by a compound annual growth rate (CAGR) of 13% through 2004–2010. Monoclonal antibodies will lead growth with a CAGR of 20.8% compared with growth of only 7.3% for therapeutic proteins, according to estimates by Datamonitor PLC (London) (2). Figure 1 outlines growth patterns for these product types.

With prospects strong, several pharmaceutical companies have acquired companies specializing in monoclonal antibodies or biopharmaceutical technology production. One of the largest deals in 2006 was AstraZeneca's (London) acquisition of Cambridge Antibody Technology (Cambridge, UK). Other deals include Merck & Co. Inc.'s (Whitehouse Station, NJ) $400-million acquisition of GlycoFi, Inc. (Lebanon, NH), a company specializing in yeast glycoengineering, and its $80-million acquisition of Abmaxis, Inc. (Santa Clara, CA).

Figure 1

In addition, several technology providers have signed pacts this year with pharmaceutical and biotechnology majors for technology to improve biopharmaceutical production. Wyeth (Madison, NJ) signed a pact with Cardinal Health (Dublin, Ohio) to evaluate developing cell lines using Cardinal Health's "GPEx" (Gene Product Expression) cell-line engineering technology for biopharmaceutical production. Novartis (Basel, Switzerland) signed a pact with Crucell NV (Leiden, Netherlands) for the selection of biopharmaceutical production cell lines. Schering-Plough Corporation (Kenilworth, NJ) partnered with Xoma, Inc. (Berkeley, CA) for monoclonal antibody discovery and development. And, Pfizer, Inc. (New York, NY) and Bristol-Myers Squibb Company (New York, NY) are already partnered with Medarex, Inc. (Princeton, NJ), which specializes in monoclonal antibody development.

AstraZeneca seeks to build biologics

Analysts point to the importance of these collaborations. "We have come to think of Big Pharma M&A activity as being driven by the need to increase critical mass and strip out costs in an attempt to improve the bottom line. But these strategic acquisitions are quite different," says a recent analysis by Wood Mackenzie (Edinburgh, Scotland).

"AstraZeneca is actively seeking Cambridge Antibody Technology's technology platform and expertise in phage display because of its potential for driving top-line growth," says the analysis. The move is likened to "Roche's approach with Genentech and Antisoma where Roche recognized that the true value lay in the culture of these companies and the expertise of their staff. It decided that an arms-length relationship was the best way to retain those strengths."

Antisoma (London, UK), a biopharmaceutical company specializing in anticancer treatments, signed a pact in 2002 with Roche (Basel, Switzerland) potentially valued up to $500 million under which Roche agreed to develop drug candidates from Antisoma's pipeline.

AstraZeneca's move to acquire Cambridge Antibody Technology is part of an overall strategy by CEO David Brennan to enhance the company's biologics portfolio. By 2010, the company has the goal of having biologic therapeutic agents account for as much as 25% of its drug candidates for full-scale development.

Company Web sites

Cambridge Antibody Technology has five human antibody drug candidates under development and seven product licenses at various stages of clinical development. Its licensed products include "Humira" (adalimubab) marketed by Abbott Laboratories (Abbott Park, IL). Cambridge Antibody Technology has a proprietary technology for isolating human monoclonal antibodies using phase display and ribosome display systems, a technology that it uses to assemble libraries of monoclonal antibodies.

The company also produces in-house development quantities of its monoclonal antibodies. It is partnered with the contract manufacturing organization Lonza Biologics (Basel, Switzerland) for the manufacture and supply of clinical-grade antibody drugs to Cambridge Antibody Technology.

Merck builds technology platform

By acquiring GlycoFi, its larger acquisition in biologics in 2006, Merck strengthened its position in yeast glycoengineering, a method that can be used to produce monoclonal antibodies and other therapeutic protein agents. Glycoengineering allows proteins such as monoclonal antibodies to be made with prespecified and defined human carbohydrate side chains. GlycoFi is working to establish the scale-up and recombinant production of proteins in yeast as an alternative to production in mammalian cell cultures, with the aim of producing specific designer glycoproteins at large scale using engineered yeasts. GlycoFi estimates that glycoproteins comprise about 70% of all approved therapeutic proteins.

Earlier this year, GlycoFi and Dartmouth College (Andover, NH) reported the first production of monoclonal antibodies with human sugar structures in yeast (3). The research showed that antibodies with human sugar structures (glycosylation) can be produced in glycoengineered yeast cell lines. GlycoFi says by controlling the sugar structures of antibodies, their therapeutic potency can be improved. The same approach offers the potential to optimize other glycosylation-dependent properties, such as solubility, half-life, or tissue distribution. The approach may also be used to improve the production and scale-up economics of antibody manufacturing.

"Mammalian cell cultures currently used for most therapeutic protein production produce a mixture of glycoforms and typically do not allow for the control of glycosylation," said Tillman Gerngross, chief scientific officer of GlycoFi, and professor of bioengineering at Dartmouth College, in commenting on the research. "We have spent the last five years engineering yeast cell lines that perform human glycosylation, which now allows us to glycosylate proteins with unprecedented control and uniformity."

GlycoFi explains that monoclonal antibodies often achieve their therapeutic benefit through two binding events: the binding of the variable domain of the antibody to a specific marker protein, such as the CD20 receptor on the surface of cancer cells, followed by the recruitment of immune system "effector" cells that bind the constant domain of the antibody and destroy the cancer cell to which the antibody is bound. Research has shown that this process, known as antibody-dependent cell cytotoxicity (ADCC), is sensitive to the composition of sugars or glycans in the antibody's constant region, explains the company. In addition, in the absence of these sugars, the antibodies can bind to antigens but do not elicit ADCC.

In the study, the researchers used several glycoengineered yeast cell lines to produce a library of glycoforms of the anti-CD20 antibody rituximab and compared their receptor-binding properties to the mammalian-derived commercial counterpart, "Rituxan" (rituximab) (3). Rituxan is commercialized in the United States by Genentech, Inc. (South San Francisco, CA) and Biogen Idec (Cambridge, MA).

GlycoFi noted that the polypeptide backbones of each of the antibody variants produced in the GlycoFi yeast remained identical, and only the glycosylation structures of each antibody were altered. Comparisons of the antibody variants with Rituxan showed that antibody binding varied with changes in the glycosylation structure and that certain antibody glycoforms showed significantly increased antibody-mediated cell killing compared with Rituxan, says the company.

"By controlling the sugar structures on antibodies we have shown that the antibodies' ability to kill cancer cells can be significantly improved and that therapeutic proteins can be optimized by controlling their sugar structures," said Huijuan Li, associate director of analytical development at GlycoFi, in commenting on the research and the lead author of the study. She noted that though the current effort focuses on antibodies, the approach can be applied to any therapeutic glycoprotein.

With Abmaxis, its second biologics acquisition in 2006, Merck gained an antibody engineering technology platform based on in-silico immunization. This technology enables structure-centric computational design followed by experimental selection of optimized human or humanized monoclonal antibodies. The starting point for the process can be an antigen target, or a murine, animal, or human antibody. Merck purchased Abmaxis following a collaboration established in late 2004. Under that pact, Abmaxis successfully re-engineered a Merck human monoclonal antibody and improved antibody affinity more than 70-fold while retaining its specificity, according to Merck.

Novartis teams with Crucell

Outside of acquisitions, the pharmaceutical majors are teaming with specialized technology providers for biopharmaceutical production. In September 2006, Novartis signed a pact with Crucell under which Novartis will apply Crucell's "Star" technology for the selection of biopharmaceutical production cell lines and to adapt such procedures to specific parental cell lines. The program will be performed with different monoclonal antibodies.

The Star technology contains genetic elements, called Star elements, which enable stable and high-yield gene expression important to recombinant antibody and protein production in mammalian cells. The Star technology was discovered by Arie Otte, who founded ChromaGenics BV, a spinoff from the University of Amsterdam. Crucell acquired Chromagenics in 2004.

Specifically, the technology focuses on counteracting epigenetic gene repression. Epigenetic gene-regulation mechanisms play an important role in maintaining the gene expression of a cell, which is important for the expression of a gene of interest such as an antibody or therapeutic protein. The expression of the gene of interest, however, is often lost because of epigenetic gene-regulation mechanisms. Crucell has identified DNA elements in the human genome that are able to counteract epigenetic gene repression. These are the Star elements. By flanking a gene of interest with Star elements, stable mammalian cell lines can be generated that show a higher expression of the gene than without the Star elements. This allows higher yields of the protein of interest, explains the company.

The Star technology may be used to produce antibodies and proteins on mammalian cell lines such as Chinese hamster ovary cell lines.

Crucell also signed a pact with UCB (Belgium, Brussels) for its Star technology. It further is partnered with several biopharmaceutical companies, including Medarex, Genzyme Corporation (Cambridge, MA), Xoma, and Millennium Pharmaceuticals, Inc. (Cambridge, MA).

Crucell also provides its "PER.C6" human cell technologies and is partnered with DSM (Heerlen, Netherlands) to use Crucell's PER.C6 human cell line as a manufacturing platform for monoclonal antibodies and recombinant proteins. Crucell and DSM are planning to open a new research and development center dedicated to the PER.C6 technology in Cambridge, Massachusetts by the end of 2006. This PER.C6 technology platform consists of cell-line generation technology, tailored cell-culture media, fermentation processes, equipment design, scale-up, technology transfer, and regulatory support. The R&D center also will have dedicated space for PER.C6 users.

Wyeth and Cardinal team

In September, Wyeth entered into a feasibility agreement with Cardinal to determine cell lines using the GPEx technology. Cardinal will use GPEx to engineer cell lines that express two undisclosed biopharmaceuticals to determine if GPEx is a feasible augmentation to Wyeth's current production methods.

The GPEx technology enables genetic engineering of stable mammalian cell lines used to produce recombinant proteins and monoclonal antibodies

Earlier this year, Cardinal also made a pact with Johnson & Johnson Company's (New Brunswick, NJ) subsidiary Centocor (Malvern, PA) to use GPEx to engineer cell lines expressing undisclosed Centocor monoclonal antibodies. Cardinal also is partnered with Trubion Pharmaceuticals, Inc. (Seattle, WA) to use CPEx for Trubion's "SMIP" (small modular immunopharmaceutical) drug candidates. These biologics are single-chain polypeptides that are engineered for full binding and activity function of a monoclonal antibody, but are roughly one-third to one-half the size of conventional therapeutic monoclonal antibodies.

Xoma advances position

Another approach is Xoma's bacterial cell expression (BCE) technology, which is used to discover, screen, develop, and manufacture recombinant proteins, including therapeutic antibodies and antibody fragments by using genetically engineered bacteria for recombinant expression of target proteins. Lucentis (ranibizumab), an antibody fragment to treat neovascular (wet) age-related macular degeneration, is the first drug approved using Xoma's BCE method. The drug is marketed by Novartis and Genentech and received FDA approval earlier this year.

A second Xoma technology ("Human Engineering") is a humanization method for modifying nonhuman antibodies to make them suitable for medical purposes in humans. The technology is based on conserved structure-function relationships among antibodies and defines which amino acid residues in a nonhuman antibody variable region are candidates for substitution.

References

1. Pharmaceutical Research and Manufacturers of America, 2006 Report: Medicines in Development-Biotechnology (PhRMA, Washington, DC, August 2006), http://www.phrma.org/files/Biotech%202006.pdf.

2. S.King, "The Evolving Pharmaceutical Value Chain: Forecasting Growth for Small and Large Molecules" Pharm. Technol. 30 (10), "Technology Outlook: APIs, Intermediates, and Formulation" supplement, s6–s10 (2006).

3. T. Gerngross et al., "Optimization of Humanized IgGs in Glycoengineered Pichia Pastoris," Nature Biotechnology24 (2), 210–215 (2006).