Since invention in the 1970s of monoclonal antibodies, which mostoften are produced from mice and were once billed as "magicbullets" for cancer, drug developers have tried to overcome theproblem of human rejection of the murine-derived therapeuticproteins.

The human immune system, recognizing the mouse antibodies asforeign, attacks them and destroys their effectiveness as a repetitive,long-term treatment for such disorders as cancer and autoimmunediseases.

Protein Design Labs Inc., of Mountain View, Calif., which developeda "humanized" monoclonal antibody that counters rejectionproblems, has received the first broad patent on the technology fromthe European Patent Office. The company has had the same patentapplication on file in the U.S. since 1988.

"This technology is so fundamental," said Peter Dworkin, spokesmanfor Protein Design Labs, "that it would apply to most humanizedmouse monoclonal antibodies."

He estimated 20 such antibodies from various companies are underclinical development in Europe and may be covered by ProteinDesign Labs' SMART antibody patent protection.

Protein Design Labs has licensed the technology to somepharmaceutical companies under collaborative agreements.

It is working with Roche Holdings Ltd., of Basel, Switzerland, on thedevelopment of Zenapax, a monoclonal antibody that binds to theinterleukin-2 receptor for prevention of kidney transplant rejection.Two Phase III trials of Zenapax are ongoing.

Another SMART antibody is in Phase II/III trials in the U.S. formyeloid leukemia and Phase I studies in Japan for the sameindication. The SMART M195 antibody is being developed incollaboration with Kanebo Ltd., of Osaka, Japan, which licensedAsian rights to the drug.

In describing the significance of the patent, David Stone, an analystwith Cowen & Co. in Boston, said monoclonal antibodies with thebest chance of working are either fully human antibodies orhumanized mouse antibodies. The former can be developed forinfectious diseases, but murine-derived antibodies are required fortargeting antigens associated with cancer and autoimmune disorders.

Using transgenic mice genetically altered with human immunesystems is a method of generating all types of monoclonal antibodiesfor human use, but the technology is in the early stages ofdevelopment.

The problem of human immune system rejection of murinemonoclonal antibodies _ called human anti-mouse antibody(HAMA) response _ has been a major stumbling block in thedevelopment of magic bullets.

Only two monoclonal antibodies have made it to market in the drugs. Malvern, Pa.-based Centocor Inc.'s ReoPro is an antibodyfragment used as a platelet inhibitor to prevent blood clots and is notdesigned to stay in the body for long periods. The second is NewBrunswick, N.J.-based Johnson & Johnson's Orthoclone OKT3,which is used to prevent acute kidney transplant rejection.

Monoclonal antibodies were called magic bullets because they arecreated to zero in on a specific antigen. The antibody's variabledomain binds to the target and another region, called the constantdomain, interacts with the immune system.

To battle HAMA, chimeric monoclonal antibodies were developed inwhich the constant domain from a human antibody is combined withthe mouse antibody's variable domain, but there still is too much ofthe mouse.

The next step was adding more human characteristics to the variabledomain to create an antibody with even fewer murine parts thanchimeric antibodies.

The monoclonal antibody's variable domain consists of acomplementarity determining region (CDR), which does the actualbinding to the antigen, and the framework region which maintains atight fit. At first, the mouse CDR was combined with a frameworkfrom a human antibody, but the binding often was too loose.

Protein Design Labs then identified the specific mouse frameworkamino acids that are necessary for preserving the CDR's high bindingaffinity. Those were kept and the rest of the murine framework aminoacids were replaced with human amino acids, creating a monoclonalantibody that is as close to human as possible.

The company also developed computer-assisted design techniques toidentify the specific murine framework amino acids that must beretained. Without the software technology, the only way to find thoseessential amino acids is through costly trial and errorexperimentation.

Dworkin said Protein Design Labs contends its patent covershumanized monoclonal antibodies regardless of the method used toachieve the appropriate framework combination of human and mouseamino acids.

Because the patent is so broad, Stone said it likely will be challenged."But we are optimistic it will be upheld because Protein Design Labswas the first to recognize the need for the framework changes andthey figured out how to do it," he said.

Dworkin said other companies have filed patents in Europe and theyhave nine months to file an opposition. A challenge could take yearsto resolve.

However, Protein Design, Dworkin said, already has beenapproached by pharmaceutical companies inquiring about a license inEurope.

In the U.S., patent submissions are confidential and disputes do notsurface until after a patent is issued.

Protein Design Labs' stock (NASDAQ:PDLI) closed Friday at $24,down $1.50. n

-- Charles Craig

(c) 1997 American Health Consultants. All rights reserved.