By David N. Leff

Everybody in the world gets cancer every day.

It stands to statistical reason that with billions of cells in the body incessantly dividing and multiplying, a few are bound to jump the reservation and go into malignant mode — multiplying out of control.

But fortunately, there is a control, which kills off these rogue cells before they can proliferate into full-blown tumors. This immunologic surveillance system patrols the body to spot errant cells, and deploys cytotoxic lymphocytes — killer T cells — to nip those precancerous upstarts in the bud. That, at least, is one theory.

For those killer cells to shoot down the wannabe tumor cells, they have first to recognize them, fix them in their sights, as it were. This fire-control presentation of the target to the gunner is the job of HLA class 1 molecules — human leukocyte antigens. They deftly pick up peptide fragments of the tumor's own antigens, and bring them to the attention of the surveillance patrol.

That's how it should work in the best of all possible worlds. But in the real word, cancer risk factors such as flawed genes, viruses, cigarettes, toxic chemicals, and radiation allow the odd proto-tumor to escape or evade surveillance, grow large, and spread throughout the body.

Metastatic melanoma is among the deadliest of such runaway cancers. But oncologists striving to identify tumor-associated antigenic targets find melanoma a useful research model. For one thing, it's so difficult to treat; for another, it provides clear-cut responses to immunotherapeutic maneuvers.

For well over a decade, the chief of surgery at the National Cancer Institute, in Bethesda, Md., Steven Rosenberg, has been trying to develop the equivalent of enhanced immune surveillance to treat patients in the last stages of metastic melanoma.

In 1985, he and his colleagues found that interleukin-2 (IL-2), a cytokine that boosts the growth of T lymphocytes, could make melanotic tumors shrink. Three years later they discovered another kind of T cell, which infiltrates tumors and destroys them from within.

"Four or five years ago," Rosenberg told BioWorld Today, "we cloned the genes for the tumor antigens by which lymphocytes recognize melanomas. When we give them to patients, they can cause regression."

Rosenberg reports his latest melanoma study, a Phase II trial, in the March 1998 issue of Nature Medicine, under the title: "Immunologic and therapeutic evaluation of a synthetic peptide vaccine for the treatment of patients with metastatic melanoma."

Synthesized Peptide Hiked Immunogenicity

Among the handful of known melanoma tumor-associated antigens, the co-authors chose one, glycoprotein 100, as the immunogenic target of their vaccine. They identified a nine-amino-acid sequence on the gp100 peptide, and modified it chemically. By synthesizing it in the lab to make it more potent, they increased its ability to bind to HLA molecules, then injected it into 31 patients.

"When we did that," Rosenberg recounted, "we could see something we've never been able to see before. I don't think anybody else has seen it either. That is, a very potent ability to raise in those patients lymphocytes that can recognize their tumors.

"That happened 10 out of 11 times," he continued, "and was very reproducible. Then, when we added high-dose IL-2 to those 31 patients we had treated with this synthetic vaccine, the response rate was 42 percent — 13 of 31 patients — whose tumors shrunk by 50 percent or more at all sites." This compares with 17 percent to whom he gave IL-2 alone in an earlier trial.

"Basically," Rosenberg explained, "the vaccine stimulates the generation of these immune cells. Then the IL-2 can activate and expand them."

The study, he reported in Nature Medicine, ended the middle of 1997. Since then, the longest duration of response to the vaccine-plus-IL-2 combination reached six or seven months. Rosenberg added that "several patients continue to have good responses to the treatment."

His team has now moved from the single-antigen vaccine to immunizing patients with four different peptides from three melanoma tumor antigens. "We're trying to prevent the tumor from evading or escaping, much the same strategy as the triple-drug HIV therapy. The tumor can mutate to avoid one antigen," Rosenberg pointed out, "but it's very unlikely to change to avoid all four at the same time."

Recruitment to this new trial is in progress, with a dozen patients already enrolled. "We'll be treating about 50 this next year," Rosenberg observed. But he also made the point that analyzing the relative roles of IL-2 and the peptide vaccine will require randomized clinical trials. Such studies, treating large numbers of patients, are now in the planning stage at several university centers in the U.S., he said.

"The present trial," he commented, "shows that it is possible to stimulate a vigorous immune response even when patients have a heavy tumor burden. Work is in progress," Rosenberg concluded, "to develop similar strategies to treat other cancers, such as ovarian, breast and prostate."

A group of European immunotherapists at the Universities of Zurich, Switzerland, and Heidelberg, Germany, report a different melanoma vaccine in this same March issue of Nature Medicine. It relies on dendritic cells, which are particularly effective at presenting tumor antigens to killer lymphocytes. Their paper bears the title: "Vaccination of melanoma patients with peptide- or tumor-lysate-pulsed dendritic cells." The vaccine achieved tumor regression in 31 percent — five out of 16 — of their trial patients, with two experiencing complete abolition of tumors, a response that has lasted more than one year. *