BioWorld International Correspondent

TORONTO - Antibodies are the fastest growing class of therapeutic proteins, with 11 on the market, and over 350 at different stages of clinical development.

"Sales of $2 billion to $3 billion in 2001 are projected to rise to $20 billion," Bruce Kimmel, head of protein therapeutics at Diversa Corp., of San Diego, said during a session on "Antibody Engineering: Making Smart Drugs Smarter" at the Biotechnology Industry Organization's International Biotechnology Convention & Exhibition.

But now the industry is in need of new technologies to speed discovery and optimize antibodies in development to enhance therapeutic properties, as well as improved production methods to reduce manufacturing costs, he said.

Henry Lowman, head of antibody discovery and engineering at Genentech Inc., of South San Francisco, described the company's approach to optimizing therapeutic antibodies using monovalent phage display as a means to generate new antibodies with new and different properties.

"What we want to do to make them more effective is address a series of molecular recognition events," he said.

An example is the way in which Genentech has successively refined its anti-vascular endothelial growth factor antibody, which is designed to prevent angiogenesis in tumors by blocking VEGF.

Lowman said that, by targeting regions of the antibody, the company has improved its affinity for human VEGF, and the second-generation anti-VEGF antibody, Y0317, is 10 times better at binding to human VEGF than the first-generation antibody. This is likely to deliver improvements in efficacy and cost, as demonstrated in an animal model of inhibiting wound healing in rabbits, he said. It turns out that Y0317 also is twice as efficient at binding to rabbit VEGF as the first-generation antibody.

"We obtained the same effect [on inhibition of wound healing] with the second-generation antibody, but at half the dose," Lowman said.

Y0317 is currently in clinical trials in the treatment of macular degeneration.

From this point, Genentech has gone on to generate several affinity-enhanced anti-VEGF antibodies. "Even small improvements [in affinity] are important because it reduces the effective dose," Lowman said.

Nils Lonsberg, chief scientific officer of Medarex Inc., of Princeton, N.J., told delegates an important issue facing the antibody industry was whether it should emulate the small-molecule drug discovery paradigm of lead selection, followed by lead optimization, in order to improve the properties of antibodies.

"Antibody molecules are complex enough to justify lead optimization, and historically antibody drug development has resembled the small-molecule development process," he said. "But I want to argue that a lot of these lead optimization steps can be partially circumvented by putting effort into lead selection."

One of Medarex's criteria in lead selection is affinity, and Lonsberg said the company's antibody MDX-010 CTLA-4 is a good example of how important this is. MDX-010 CTLA-4 activates antigen-specific T-cell responses, and is under development as an adjuvant for cancer vaccines. In monkey studies it enhanced antibody responses to a melanoma vaccine, and over five months of repeated dosing there was no clearance.

"This is very different from using murine antibodies, and we think one of the explanations is the way we picked the antibody to have an exquisite affinity for human CTLA-4," Lonsberg said.

Another way in which the therapeutic potential of antibodies is being enhanced is by adding functionality, and Lonsberg said he believes antibody-directed cytotoxics will soon be coming through development. The exemplar is Mylotarg, an antibody-directed radionucleotide, he said.

"Now the field is moving to small-molecule conjugates, which have an advantage because, unlike radionucleotide conjugates, you can put them in a bottle," he said.

Medarex's antibody-directed cytotoxic KW 2189, in Phase II trials, delivers the cytotoxic duocarmycin.

"We selected the linker so it can be released only when inside the cell," said Lonsberg.

Jeff Watkins, chief scientific officer of Applied Molecular Evolution Inc., of San Diego, argued that lead optimization of antibodies is important because proteins have a lot of physical characteristics that are not relevant in vivo. Using the company's technology for engineering antibodies by directed evolution, it is possible to optimize antibodies to improve safety, reduce dosing, decrease manufacturing costs and also broaden the intellectual property position, he said.

The technology involves introducing diversity into a gene of interest and using a host expression system to generate a protein library. These can then be screened for proteins that show improvements in the function of interest, said Watkins. Using the method, there is no need to generate structural information or structural models, he added.

An example of an antibody to which this technique has been applied is Synagis (palivizumab), Gaithersburg, Md.-based MedImmune Inc.'s treatment for respiratory syncytial virus, which racked up sales of $516 million in the 2000-2001 disease season. Watkins said that using AME's antibody engineering technology, the potency of the second-generation antibody (known as Numax) has been increased 25-fold.

"The technique can also be applied to optimize murine antibodies, to preserve specificity, while engineering them to minimize immunogenicity and get improved function," said Watkins.