Like ships at sea that fly flags of identity, a growing tumor unfurlsantigenic molecules on its cell surfaces, to signal its host's immunedefenses that it's foreign.

Those defenses, primarily T lymphocytes activated by the tumorantigens, order action stations and prepare to do battle with themalignant enemy. Then, after a brief engagement, the T cells soundretreat, while the cancer keeps on growing.

This surrender by the immune system's front-line troops, the T cells,leaves it up to surgeons, radiologists, chemotherapists andimmunotherapists to fight what is usually a losing engagement withthe malignancy. Just what it is that turns the embattled T cells off inthe heat of the fray is only now beginning to be understood _ andacted upon.

One of the actors is post-doctoral immunologist Dana Leach, in thelaboratory of James Allison, who directs the Cancer ResearchLaboratory at the University of California's Berkeley campus. Leachand Allison are co-authors of a paper in the current Science, datedMarch 22, 1996, titled: "Enhancement of antitumor immunity byCTLA-4 blockade."

"CTLA-4 ," Leach explained to BioWorld Today, "stands for`cytolytic T-lymphocyte antigen-4.' This molecule pops up on thesurface of activated T cells," he continued, "and decides to turn themback off again, if they don't continue to see the appropriate signalsthat the tumor cells are non-self, and therefore fair game fordestruction."

Leach and Allison countermand that molecule's surrender message,and unspike the T cells' guns again, with a single monoclonalantibody targeted at CTLA-4. In mice, their antibody not onlyabolished growing human colon carcinomas but immunized theanimals against a new challenge by the same cancer.

It Takes Two Co-Stimuli To Galvanize T Cells

When a tumor cell sends out its "I'm foreign" signal, the T celldoesn't get the message directly. First, the antigen must be"presented" to it via the major histocompatibility complex (MHC), acluster of genes encoding molecules on specialized cells that bringthe non-self antigen to the attention of the immune defenses. But thatMHC stimulation of the T cells isn't enough to activate them.

They must also be co-stimulated by the interaction of another T-cellsurface molecule, CD28, with the lymphocyte's receptor, B7, alsoperched on that antigen-presenting cell.

It takes both of these co-stimulatory sets of partners to keep T cellsfiring their guns at the tumor cell. But it only takes one homologue(molecular look-alike) of CD28 _ namely, CTLA-4 _ to silencethose T cells.

The California immunologists reasoned that blocking CTLA-4 wouldremove inhibitory signals in the costimulatory pathway, resulting inenhanced rejection of tumor cells.

Accordingly, they seeded 4 million colon carcinoma cells into groupsof mice, then injected them with antibodies to CTLA-4. All of theanimals rejected their tumors. In contrast, control mice, treated withsaline solution after inoculation with cancer cells, developedprogressively growing tumors, and had to be euthanized at 35 days.

Ten weeks after their initial tumor injections, five mice who hadreceived CTLA-4 antibody and shrugged off their tumors got a freshdose of 4 million malignant cells. All but one proved immune to thisreturn engagement, whereas all controls developed massive, rapidlygrowing tumor burdens, requiring euthanasia, again at 35 days.

The odd mouse out among the immunized cohort did develop a veryslow-growing tumor, and was joined by two other animals 42 daysafter the challenge.

"Our antibody," Leach observed, "did indeed target the CTLA-4molecule, and prevented its normal function in vivo. This effect," headded, "is unique to our lab. And one of its beauties is that it's soeasy. It just requires the injection of an antibody."

With such striking murine results, can human trials be far behind?

Clinical Trials In `Not-Too-Distant Future'

"We are now negotiating with M.D. clinicians who work with cancerpatients, and are very interested in pursuing this," Leach said. "Theywould like to begin Phase I trials in the not-too-distant future."

So the California lab is now working on a human version of themouse CTLA-4 monoclonal antibody. "We have to isolate lots ofhuman CTLA-4, inoculate hamsters with it, then start screeningmonoclonal hybridomas. This takes a while; it's a well-establishedbut time-consuming process."

Leach went on: "We're just basic immunologists, working with themouse as a model. We don't do clinical science here, but we certainlyhave contacts with people who do. They tell us: `When it gets to thepoint of humanizing an antibody, that's a piece of cake; we can doit.'"

The Berkeley team has to resolve one uncertainty before human trialscan commence, namely long-term side effects they may beunwittingly perpetrating in the immune system.

"Our concern is," Leach spelled out, "if in preventing the function ofa switch that normally turns things off, are we going to start to turnthings on that aren't supposed to be on? We might be turning on asevere generalized autoimmune reaction that could be very serious,especially to a patient in a health-compromised situation."

Eventually though, Leach concluded, improvements in his lab's invivo vaccine plus the ex vivo cell-culturing techniques that othercenters are pursuing may well bring about a new human cancerimmunotherapy.

He foresees "combinations of these things whether five years or 10years from now, I don't know," and added: "But 10 years in thehistory of science is nothing. In the life of an individual cancerpatient it's obviously a lifetime." n

-- David N. Leff Science Editor

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