Stephen F. Carroll

Over the past 10 years, many new tools added to the drug discovery toolbox have accelerated the rate at which new targets and potential products are identified. Included among those tools are major advances in genomics, proteomics, microarray technologies, and high-throughput screening. Despite those advances, however, the necessary steps for converting new discoveries into promising therapeutics have changed very little (Figure 1).

For example, once a new drug target is identified, its role in an appropriate disease process needs to be validated. Then a therapeutic molecule must be identified and shown to be effective, and techniques for manufacturing and characterizing it are developed. Although some efforts can be conducted in parallel, in most instances the rate-limiting and most costly steps of the process involve the development efforts: those occurring after product validation and continuing through clinical testing. In essence, development is a funnel through which each new product must pass (Figure 2).

Once a product candidate has been identified and the decision made to move it into development, many organizations find their own infrastructure to be resource constrained. Even large pharmaceutical and biotechnology companies, many of which already have development and manufacturing capabilities, often have more projects than their internal development resources can accommodate. Similarly, smaller companies may not (yet) have a development infrastructure (manufacturing facilities, experienced development, clinical, regulatory, quality control or other staff and cap) to move their own products forward aggressively.

Traditionally, three approaches have been used to resolve resource limitations in a development program: building new internal capabilities; contracting the work to service vendors (contract research organizations, CROs, and contract manufacturing organizations, CMOs); or partnering with a larger pharmaceutical company. Although each choice has certain advantages, they share major disadvantages related to time and money.

Building a development infrastructure requires several years (and potentially several hundred million dollars) and must be planned in advance if the new plant is to be available for an up-and-coming project. Similarly, contract service providers and other pharmaceutical companies may have the available resources, but with them your project can become one of many — and you could lose control of its timing. The timing of each project is a key factor in development because the earlier a product can reach the market (or the earlier a decision can be made to terminate a program), the earlier the opportunity for revenue (or dollars saved). Clearly, the advantages of more quickly launching a new therapeutic cannot be over-emphasized. Shaving six months off the development timeline not only brings the new therapeutic to patients more quickly, but it also brings in revenues six months earlier.

During the past few years, a fourth development option has been added to the list above that addresses many disadvantages. This new option, the formation of a development partnership, is a collaboration between two companies: one with a promising new product candidate and another with an established development infrastructure. In this type of partnership, each company has an equity stake in the product’s success in a development program usually comanaged by both. Similarly, program costs can be shared between the two companies, which helps to reduce expenses and minimize risk. Some key advantages of development partnerships include increased speed and flexibility, joint control of the development program, shared expenses, reduced risk, and access to another’s intellectual property and/or know-how.

That last point is an important one because critical intellectual property and know-how can greatly accelerate product development and help minimize royalty stacking in competitive technological areas. For example, if the product candidate is an antibody, licenses to a number of different patent portfolios will often be needed in order to commercialize. When the resulting royalties are all tallied up, the company can end up with double-digit royalty obligations just for the technology licenses alone.

Profile of a Development Partner:

XOMA has a long history of collaborative product development dating back to the early days of biotechnology (Figure 3).

Founded in 1980, XOMA began manufacturing and clinically testing antibody-based products as early as 1983 ¹. Since that time, the company has continued to expand its development infrastructure and expertise to include other recombinant proteins², ³. It has manufactured and tested 11 different protein products (including nine monoclonal antibodies) in clinical trials involving 20 different indications. As integral components of those programs, XOMA established collaborations for the development of therapeutic products with five different pharmaceutical and biotechnology companies since 1986 (examples of which will be covered in greater detail below).

Over its 22-year history, XOMA has established a fully integrated development infrastructure that ranges from early target validation and product identification through manufacturing, clinical trials, and regulatory filings. Included within that infrastructure are capabilities such as those listed in the sidebar on the previous page. Figure 4 illustrates these skills graphically and highlights certain aspects of XOMA’s know-how and intellectual property.

XOMA has been a pioneer in developing new technologies that aid in product development, largely for its internal programs and particularly regarding antibodies. These efforts have led to the development of proprietary expression vectors and an efficient expression system that allows production of recombinant proteins from bacteria. The company wasfirst to demonstrate that fully functional antibodies and antibody fragments can be secreted by bacteria ⁴, and it has developed an antibody humanization technology known as “Human Engineering” ⁵. That technique is a simple and cost-effective approach to reducing the immunogenicity of nonhuman antibodies while preserving (or even enhancing) antibody affinity. These technologies are not only being used to facilitate internal and collaborative project development, but they are also being licensed to other companies for their own use. Along those lines, XOMA has established cross-licenses with several groups that allow it access to additional development tools such as phage display (current cross-licenses include Cambridge Antibody Technologies, Dyax, MorphoSys, and Biosite) and certain production technologies for recombinant proteins (cross-licenses include Genentech, Celltech, and Enzon).

When combined, these tools extend XOMA’s capabilities, provide potential benefits for its partners, and create a bridge between discovery-stage projects and those ready to enter into full development.

CASE STUDIES

Here are summarized several projects that highlight aspects of XOMA’s development partnerships.

Genentech Collaboration: In 1996, Genentech and XOMA formed a collaborative development program for Raptiva, a humanized monoclonal antibody directed against CD-11a, the alpha subunit of the LFA-1 integrin. Interestingly, unlike in most other collaborations between smaller and larger companies, the molecule originated at Genentech and was to be developed by XOMA. When the collaboration was initiated, the antibody had already been humanized, a number of preclinical studies had been completed, and a cell bank had been prepared. Therefore, a master cell bank (MCB) and a working cell bank (WCB, made from the MCB and used to inoculate the fermentors) still needed to be made. Other internal projects at Genentech, however, prevented the molecule from entering development there. As a consequence, it became XOMA’s responsibility to scale up the manufacturing process, produce material for nonclinical and clinical studies, complete the efficacy and toxicology studies required to file an Investigational New Drug application (IND), and to conduct clinical trials through phase II. And for these efforts, XOMA obtained significant ownership of the product.

As outlined in Figure 5, XOMA received the MCB in April 1996 and within a month had initiated fermentation of the product in a 2,750-L fermentor. Two months later, we initiated a six-month toxicology study, the early results of which allowed us to file an IND a month later. Thus, within a period of four months after receipt of the MCB, we had scaled production to 2,750 L, purified sufficient material for toxicology and initial clinical studies, developed a formulation suitable for intravenous administration, completed the necessary nonclinical and toxicology studies, and filed the IND. Clinical studies were subsequently conducted in patients with psoriasis, first as a series of phase I and phase II studies using the intravenous formulation, and completed roughly 28 months after filing of the IND. During that same period, Genentech developed a lyophilized formulation of the antibody suitable for subcutaneous administration, which XOMA then tested in psoriasis patients.

Once the phase II studies were completed and activity in humans demonstrated, Genentech took on the responsibility of conducting the phase III trials in psoriasis. In addition, manufacturing was successfully transferred from XOMA to Genentech at that point. Lyophilized product was used for the phase III trials and has also been used in subsequent phase I/II studies for new indications. A biologics license application (BLA) was submitted in December 2002 for the use of Raptiva in patients with moderate to severe psoriasis. Additionally, XOMA continues to run clinical trials exploring other indications for Rapitva.

It is evident from the above description that the timeline for the Raptiva project was quite aggressive. This type of speed clearly resulted from close interactions between two focused companies, a process that was managed by a joint core team. With representatives from both XOMA and Genentech, that core team was able to review and address issues as they arose and determine which organization was best able to handle each necessary task.

Internal Case Study (ING1): A second example relates to an internal program with an antibody called ING1. This antibody recognizes the human epithelial cell adhesion molecule (Ep-CAM), a 40-kDa glycoprotein surface antigen present on many adenocarcinomas (breast, colorectal, prostate, and so on). As such, it has the potential to be a useful treatment for a variety of cancers ⁶. In preparation for entering the clinic, the antibody was human engineered, a technology that alters selected amino acid residues within the antibody variable regions to make the antibody appear more human to the immune system ⁵. Because this is a structural approach unrelated to complementarity-determining region (CDR) grafting, it does not reduce antibody affinity and is applicable to antibodies from any source (e.g. mouse, rat, rabbit). Figure 6 is a molecular model of the human engineered ING1 antibody.

The development timeline for ING1 is shown in Figure 7. Beginning with DNA encoding ING1, cell line development was initiated with transfections of Chinese hamster ovary (CHO) cells, screening of those transfectants, and selection of appropriate clones. The clones were then used to prepare a MCB and the MWCB used to inoculate fermentors for the production of clinical material. Once all the nonclinical efficacy and toxicology studies had been completed, an IND was filed for Phase I studies in cancer patients. Overall, the time needed to go from DNA to IND was just over one year.

The current status of this program is that phase I studies with the intravenous formulation have been completed. The incidence of an immune response to ING1 is both low and similar to that reported for humanized or human antibodies ⁷. We are currently completing a phase I study of a subcutaneous formulation that we anticipate could greatly simplify patient use of the medication. These data demonstrate both that our human engineering technology can be used to produce antibodies thatare poorly immunogenic in humans and that the methodology represents a simple alternative to other humanized or human antibody techniques.

Millennium Collaboration: In December 2001, XOMA entered into a collaboration with Millennium Pharmaceuticals Inc. to codevelop two protein therapeutics for cardiovascular disease. One is a humanized monoclonal antibody and the other is a fusion protein combining two human cell surface receptors. As in the Genentech collaboration, it is XOMA’s responsibility to develop the molecules through phase II studies — and for those efforts, XOMA obtains significant ownership of the products. Work on the projects began at XOMA in January 2002 with receipt of the appropriate DNAs, which were cloned into our proprietary mammalian cell expression vectors and then transfected into CHO cells, leading to cell line selection (Figure 8).

Four months later, clones producing high levels of protein were selected for the preparation of MCBs and MWCBs. We then developed methods of purification andformulation, initiated production in fermentors, and made product available for completing the nonclinical efficacy and toxicology studies. Overall, the time from receipt of DNA to filing for clinical trials with the first product was 14 months.

ACCELERATING DEVELOPMENT

In the world of new therapeutic development, partnerships have become an important component of getting new drugs to the market. In most instances, the question is not so much “whether” to partner but rather “when.” For example, a review of biotech products approved in 2002 ⁸ indicates that only three of fifteen were from biotechnology companies that were able to take the product to the market on their own. Partnering to leverage the strengths of two (or more) companies in an effort to accelerate new product development can provide significant benefits not only to both organizations, but also to patient populations in need of new therapeutics. Based on our experiences, the key components of managing a partnership include those listed in the “Benefits and Management” sidebar.

Several approaches exist that allow biotechnology companies to deal with a lack of available internal resources to complete the development of new therapeutics. Among them, development partnerships provide some advantages that may be appealing to both large and small organizations.

1. Spitler LE, et al. Therapy of Patients with Malignant Melanoma Using a Monoclonal Antimelanoma Antibody–Ricin A Chain Immunotoxin. Cancer Res. 1987; 47(6); 1,717–1,723.

2. Von der Mohlen, MA, et al. Inhibition of Endotoxin-Induced Cytokine Release and Neutrophil Activation in Humans by Use of Recombinant Bactericidal/Permeability-Increasing Protein. J. Infect. Dis. 1995; 172(1); 144–151.

3. Better, M, et al. T Cell Targeted Immunofusion Proteins from Escherichia Coli. J. Biol. Chem. 1995; 270(25); 14,951–14,957.

4. Better, M, et al. Escherichia Coli Secretion of an Active Chimeric AntibodyFragment. Science 1988; 240(4,855); 1,041–1,043.

5. Studnicka, GM, et al. Human-Engineered Monoclonal Antibodies Retain Full Specific Binding Activity by Preserving Non-CDR Complementarity-Modulating Residues. Protein Eng. 1994; 7(6); 805–814.

6. Ammons, WS, et al. In Vitro and In Vivo Pharmacology and Pharmacokinetics of a Human-Engineered Monoclonal Antibody to Epithelial Cell Adhesion Molecules. Neoplasia 2003; 5(2); 146–154.

7. Better, M, et all. Recombinant Human Engineered ING-1 Monoclonal Antibody, ING-1 (heMAb), Exhibits Minimal Immuno-genicity in Patients with Advanced Adeno-carcinoma. Proceedings ASCO 2002; 21; 20a.

8. Van Brunt, J. Biotechs Scored 23 Product Approvals in 2002, Including Nine New Therapies. Signals Online Magazine 2003; 3 February (www.signalsmag.com).

Stephen F. Carroll, PhD, is vice president of scientific and product development at XOMA (US) LLC, 2910 Seventh Street, Berkeley, CA 94710, 510-204-7536, carroll@xoma.com, www.xoma.com. Prior to his current position, he was vice president of preclinical research at XOMA.

To make any comments on this article, or to ask a question of the author, please contact the publisher. If you would like to submit an article please subscribe to our PL Intelligence service.

The opinions expressed in the articles published in this section do not necessarily reflect those of Pharmalicensing or UTEK Corporation. No actions including proposals to or agreements with other companies should be taken by any reader without obtaining specific business or legal advice. Neither the publisher nor the authors accept any liability for any actions or activities undertaken by any reader or other third party as a consequence of these articles or for any errors or omissions therein.