Modern Antibody-Based Therapeutics

As of late 2004, 26 modern antibody-based therapeutic agents have been approved in the European Union and the US. Some 500 such products are currently in development, ensuring that the number of approved antibody-based products will increase substantially over the coming years.

Gary Walsh

Traditionally, polyclonal antibodies have been used to induce passive immunity. These antibody preparations are made by extracting human or animal serum exposed to the antigens of interest. Antibody preparations produced by such means are heterogeneous. Antibody levels vary from bleed to bleed and, when the producer animal dies, so does the product source.

The mid 1970s development of hybridoma technology allowed production of large amounts of monospecific antibodies against virtually any antigen of interest.^1,2 This technique fuses a mouse-derived (murine), antibody-producing lymphocyte with a transformed (cancerous) myeloma cell. A proportion of the daughter hybrid cells (hybridomas) continue to produce the murine antibody and also retain the immortal character of the parent myeloma cell. The hybridoma can be stored in a viable state for years, providing an almost inexhaustible seed supply of monoclonal antibody (MAb). The most striking attribute is that all the individual antibody molecules present in a monoclonal antibody preparation are identical (See "Antibody Structure" box on page 2).

MAbs initially found wide application as in vitro diagnostic reagents in immunoassays. Starting in the 1980s, several gained regulatory approval for in vivo use — either as therapeutic or diagnostic agents.^3,4 The major target indication of modern antibody preparations is cancer, although several products aimed at additional indications also have been approved (Table 1).

RCSB Protein Data Bank, 15c8, http://www.rcsb.org/pdb

SPECIFIC SELECTIVITY

The binding of an antibody to the antigen against which it is produced is extremely specific. All human (and indeed other) cells display a range of surface antigens. Some are found on a range of cell types, while others — termed unique surface antigens (USAs) — are specific to a given cell type. Antibodies produced against unique surface antigens bind selectively to the surface of these cells. In effect these antibodies are "magic bullets" capable of selectively targeting specific cell types such as cancer cells, virally infected cells, or microbial cells at an infection site.

Antibodies generally display favorable safety profiles. Humanized antibodies (described below), in particular, are well tolerated at high doses. Furthermore, engineered antibodies can be quite versatile. Slight changes in the six to ten amino acids within the complementarity-determining region (Figure 1) can retarget the antibody, creating a new potential therapeutic. Common drug development and manufacturing approaches can be employed to bring the product to market.

Binding of an antibody to a cell surface can potentially trigger cellular destruction via immunological effector functions associated with the antibody's F_c region (for example, phagocytic destruction and complement activation). Antibody-mediated cellular destruction generally involves conjugating an antibody to a cytotoxic agent, usually a radionuclide, a toxin, or potentially an enzyme capable of converting an inactive prodrug into an active cytocidal agent (Figure 2). The antibody delivers the cytocidal agent directly to the surface of the target cell — and defines antibody-mediated cellular targeting.

Figure 2. Antibody-Based Therapeutics (a) Antibodies bind selectively to specific cell types if they are raised against a unique surface antigen (USA). (b) Effector molecules can be conjugated to the antibody. Radioactive tags are employed to either detect or destroy the target cell. (c) Toxins can also be used as cytocidal agents. (d) Enzymes (E) can selectively convert a harmless prodrug into a cytocidal agent at the cell surface.

ANTIBODY-MEDIATED CANCER DIAGNOSIS AND THERAPY

When a cell is transformed (becomes cancerous), it usually begins high-level expression of several genes that were previously either unexpressed or expressed at extremely low levels. Some of the resultant proteins are found on the cell surface and are termed tumor surface antigens (TSAs). TSAs potentially represent unique surface antigens, and antibodies raised against specific TSAs are likely to be approved for oncological applications.

Several anti-TSA antibodies conjugated to radioactive tags have been approved for the detection and treatment of various cancers (Table 1). Conjugation of radioactive tags to anti-TSA MAbs facilitates targeted tumor destruction via the ionizing effects of radioactivity. ⁵

Methods of conjugating radioactive tags to MAbs are well established. The most common radioisotopes used are β emitters. The medium-energy radioactivity they emit penetrates through several cell layers, which is suitable for the treatment of large tumors. This can be particularly advantageous in the treatment of tumors containing a proportion of TSA-negative cells.

The most popular radionuclides studied have been isotopes of iodine (¹³¹I) and yttrium (⁹⁰Y). ⁹⁰Y is particularly popular as it emits high-energy radiation, namely β^- particles, which have a mean path length of up to 5 mm and hence deep tissue penetration. ⁹⁰Y also has a relatively short half-life (64 hr) and does not emit γ radiation, which permits outpatient therapy.

Antibody Structure

Radiolabelled antibodies can also be used for in vivo diagnostic imaging (immunoscintigraphy) of cancer and other medical conditions. This requires radioisotopes emitting γ rays of suitable mean energy. Metastable technetium (^99mTc) and indium (¹¹¹In) are used most often, and a number of products have been approved.

Several trials are investigating the use of toxins conjugated to anti-TSA antibodies as targeted therapeutic agents. A variety of toxins or toxin subunits derived from plants, bacteria, and fungi have been used, including the ricin A chain and Pseudomonas exotoxin. Some antibody-TSA complexes are quickly internalized upon binding of the antibody to a cell surface. In the case of antibody-toxin conjugates this process is essential to cellular destruction. Upon their internalization, most of the antibody-toxin conjugate molecules are destroyed in the lysosomes; however, a small number survive and can subsequently induce cell death.

MAbs also find application in the detection and treatment of medical conditions other than cancer. Simulect is a chimeric antibody that specifically binds to the α-chain of the interleukin-2 (IL-2) receptor. Binding inhibits activated T-lymphocytes, which are involved in the acute rejection of transplanted tissue. Other products aimed at preventing transplant rejection include Zenapax and Orthoclone OKT3. Additional product indications include psoriasis, asthma, rheumatoid arthritis, and Crohn's disease (Table 1).

Table 1. Modern Antibody Preparations Approved in the EU and the US

ANTIBODY ENGINEERING

Many, if not most, first-generation MAb-based therapeutic products proved to be clinical disappointments. Poor efficacy was due to a number of factors. The single most important limitation was their immunogenicity. First-generation MAbs were invariably intact murine monoclonals and, as such, were highly immunogenic when administered to humans. Human anti-mouse antibodies were detected in up to 80% of patients within 14 days of MAb administration — the HAMA response. Repeat administration induced a response in most other patients, limiting the therapeutic efficacy of murine monoclonals to one or two doses.

In addition to the universal obstacle of immunogenicity, first-generation MAbs displayed complications relating to tumor penetration and target specificity. Intact antibodies are large and bulky macromolecules that often cannot effectively penetrate a solid tumor mass. This limits their therapeutic efficacy to the outermost layers of transformed cells. The specificity of antibody-mediated tumor detection or destruction obviously depends upon identifying and targeting a TSA uniquely expressed on the target cells. Absolute antigen specificity is desirable but rarely attained in practice. For example, antibodies can also target healthy tissue if it expresses a surface antigen closely related to the TSA. All too often, this can lead to significant and unacceptable side effects.

Some of these difficulties have been circumvented thanks to advances in antibody engineering. Chimeric, humanized, and human monoclonals greatly reduce or essentially eliminate immunogenic responses while, in the case of cancer, much smaller, antigen-binding antibody fragments enhance tumor penetration.

Cloning and expression of antibody or antibody-fragment genes and cDNA is now routine. Initial attempts to reduce the immunogenicity of murine monoclonals entailed producing chimeric antibodies by recombinant DNA technology. Several such products gained approval, mainly in the 1990s. They are made by splicing the gene sequences coding for the mouse-variable regions (these contain the antigen-binding complementarity determining regions) to the constant regions of a human antibody. The resultant hybrid structure is designed to retain the desired antigen-binding specificity of the parent murine monoclonal and be less immunogenic, since only a minor part of the chimeric structure is murine in origin. In practice, reduced immunogenicity is observed, and chimeric products display extended serum half-lives (200-250 hr, as compared to 30-40 hr in the case of intact murine monoclonals).

Antibody engineering also allows the production of "humanized" antibodies,⁶ in which only the gene sequences coding for the CDRs of the murine antibody replace the CDR sequences in human antibody genes. The resultant antibody is entirely human except for these short sequences and has essentially the same biochemical characteristics as native human antibodies, as confirmed in practice. These antibodies display serum half-lives virtually indistinguishable from native human antibodies (14-21 days).

Although not antibodies themselves, recombinant DNA technology has also facilitated the development of two fusion proteins that contain antibody domains fused to other proteins (marketed under the trade names Enbrel and Amevive respectively; Table 1).

ANTIBODY FRAGMENTS

There are two ways of generating antigen-binding Fab or F(ab)

₂

fragments. The older way is by direct proteolytic cleavage of intact antibodies. Genetic engineering also allows the generation of F

_V

-like fragments by linking gene sequences coding for the V

_L

and V

_H

domains with a "linker" sequence coding for a short peptide, anchoring the two domains together.

⁷

Such fragments, due to their smaller size, are capable of more readily penetrating solid tumors and should therefore show increased efficacy for treating cancers. Direct studies have confirmed their enhanced ability to penetrate tumors. Intact IgG molecules take 54 hr to travel 1 mm through a solid tumor, whereas Fab fragments can travel the same distance in 16 hr.

⁸

Despite this potential advantage, the use of antibody fragments for solid tumor therapy has been quite limited. These fragments have very short half-lives. While rapid clearance would be an advantage for in vivo diagnostic use (facilitating rapid removal of the conjugated radioactive tag), it is usually a disadvantage in a therapeutic application.

FULLY HUMAN ANTIBODIES

The development of chimeric and, in particular, humanized antibodies have mostly overcome issues of antibody antigenicity in humans. It is now possible to produce fully human antibodies via recombinant DNA technology — either via transgenic animals or via phage display technology. Research undertaken in the 1990s has led to the development of mice strains capable of producing a fully human antibody repertoire. This is achieved by inactivating endogenous murine immunoglobulin genes and effectively replacing them with human immunoglobulin loci. This approach has been commercialized by companies such as Abgenix (the XenoMouse), Medarex (HuMab mouse), and Kirin (TC mouse).

⁹

An alternative means of producing fully human antibodies entails the expression of human antibody libraries in phage, viral particles that replicate in bacteria. The phage display libraries can easily be screened for an antibody that selectively binds the antigen of interest.¹⁰ Several such antibodies are in clinical trials and are likely to be approved over the coming years.

TRANSGENIC PRODUCTION SYSTEMS

All modern antibody preparations are produced using mammalian cell lines, either hybridoma cells in the case of monoclonals or Chinese hamster ovary (CHO) or related cell lines in the case of most engineered products. Systems based on mammalian cell lines are relatively expensive. Estimates of the cost of product synthesis are in the vicinity of $300/g antibody. High production costs are slowing the adoption of antibody-based products, as therapeutic regimes often require ongoing administration of tens of milligrams of product per dose. (Comparative dosage levels of interferons, for example, would typically be in the microgram range.)

Looking to lower costs, a wide variety of engineered antibodies have been produced in various transgenic systems. These include the milk of transgenic goats, the eggs of transgenic chickens, cow blood, and a range of transgenic plants including wheat, rice, soy, and tobacco.^9,11 Although some published projections suggest that the cost of synthesis could be below $50/g, there is no approved product to support the data. Upfront developmental costs and additional technical and regulatory issues are delaying production of biopharmaceuticals in transgenic animals. Plant-based production also suffers from a number of complications, including radically altered product glycosylation patterns.

BISPECIFIC ANTIBODIES

Genetic engineering facilitates the design and production of bispecific antibodies and antibody fragments (BsAbs). As the name implies, bispecific antibodies are engineered to bind to two different, specific antigens. Such antibodies can be designed so that one antigen-binding site targets a unique surface antigen on a cancer cell while the other either binds a therapeutic agent or attracts immune effector cells to the target cell surface.

¹²

Bispecific antibodies have some disadvantages, including short half-lives and the potential to trigger serious adverse effects due to the induction of cytokine release.

ANTIBODY PRODUCTION IN VIVO

The production of a desired therapeutic antibody

in vivo

represents an elegant means of overcoming technical complications such as short half-life of injected product. A number of research groups are pursuing this strategy. One obvious practical approach is the introduction of encapsulated or otherwise protected antibody-producing cells into the body.

^13,14

Another is gene therapy, although the same technical difficulties that beset gene therapy generally also apply in this instance.¹³ Gene therapy would also allow the expression of antibodies within individual cells, allowing antibodies to target and hence inactivate (or perhaps activate) specific intracellular molecules and processes relevant to disease progression. By attachment of appropriate cellular signal sequences, such intracellular antibodies (intrabodies) can target not only the cell cytoplasm but also specific intracellular organelles.^15,16

THE FUTURE

Data from the Pharmaceutical Research and Manufacturers of America suggests its member companies had about 75 antibody-based products undergoing clinical evaluation in 2003.

¹⁷

When characterized by product type, antibody-based products therefore were second only to vaccines, of which there were 98 in development. Globally about 500 antibody products are in development.

¹⁸

Datamonitor suggests that the total market value is set to grow very substantially, reaching more than $17 billion before the end of the decade. The global market value in 2002 was estimated at $5.4 billion.

¹⁹

The scientific future will likely center around the generation and approval of increasing numbers of fully human antibodies, innovations in production — particularly in transgenic systems — and the generation and application of bispecific antibodies and intrabodies.

Overall, antibodies not only constitute a major class of approved therapeutic proteins, but they also will continue to represent one of the most significant and flexible biopharmaceutical products coming on stream for many years to come.

REFERENCES

1. Kohler G, Milstein C. Continuous culture of fused cells secreting antibody of pre-defined specificity.

Nature

1975; 256: 495-497.

2. Winter G, Milstein C. Man-made antibodies. Nature 1991; 349: 293-299.

3. Harris W. Antibody therapeutics. CRC press, Boca Raton, FL; 1997.

4. Brekke OH, Loset GA. New technologies in therapeutic antibody development. Curr. Opin. Pharmacol 2003; 3: 544-550.

5. Surface D. Expanding Radioimmuno-therapy. Radiology Today 2004 Sept. 27; 5(20): 10.

6. Harris W. Humanizing monoclonal antibodies for in vivo use. Anim. Cell Biotechnol. 1994; 6: 259-277.

7. Hudson P. Recombinant antibody fragments. Curr. Opin. Biotechnol. 1998; 9: 395-402.

8. Jain RK, Baxter LT. Mechanisms of heterogeneous distribution of monoclonal antibodies and other macromolecules in tumors: significance of elevated interstitial pressure. Cancer Res. 1988; 48: 7022-7032.

9. Lowe D. Antibody engineering for future therapeutics. BTi 2004 April/May ; 8-13.

10. Kretzschmar T, von Rruden T. Antibody discovery: phage display. Curr. Opin Biotechnol. 2002; 13: 598-602.

11. Larrick J, Thomas D. Producing proteins in transgenic plants and animals. Curr. Opin. Biotechnol. 2001; 12: 411-418.

12. Cao Y, Lam L. Bispecific antibody conjugates in therapeutics. Adv. Drug Del. Rev. 2003; 55: 171-197.

13. Sanz L, Blanco B, Alvarez-Vallina L. Antibodies and gene therapy: teaching old "magic bullets" new tricks. Trends Immunol. 2004; 25: 85-91.

14. Noel D. In vitro and in vivo secretion of cloned antibodies by genetically modifying myogenic cells. Hum. Gene Ther. 1997; 8: 1219-1229.

15. Marasco WA. Intrabodies: turning the humoral immune system outside in for intracellular immunization. Gene Ther. 1997; 4: 11-15.

16. Tse E, Rabbitts TH. Intracellular antibody-caspase-mediated cell killing: an approach for application in cancer therapy. Proc. Natl. Acad.Sci. USA. 2000; 97: 12266-12271.

17. Pharmaceutical Research and Manufacturers of America. New biotech medicines in development 2004. Washington DC. Available at www.phrma.org/newmedicines/biotech/

18. Pharmaprojects report 2004. PJB Publications. London. Available at www.pjbpubs.com/pharmaprojects/index.htm

19. Datamonitor. Therapeutic antibodies: Capitalizing on the fully human wave. 2003 London. Available at www.datamonitor.com

Gary Walsh is a BioPharm International EAB member and senior lecturer, Industrial Biochemistry Program, at University of Limerick (University of Limerick, Limerick City, Ireland, 353.61.202.664, fax 353.61.202.568, Gary.walsh@ul.ie)