Diet changes may buy cancer patients time.

Increasingly, cancer researchers are testing nutritional weapons in their war on cancer and its spread. Several programs report promising results in animals merely by altering the amino acid composition of the diet. If successful in humans, these nontoxic strategies could buy patients time for other postsurgical therapies to work.

Gary G. Meadows of Washington State University in Pullman and his co-workers have at least doubled the survival time of mice injected with malignant melanoma by restricting their consumption of phenylalanine and tyrosine. They cut dietary intake of these amino acids to just 13 percent of normal -- levels that Meadows describes as "just slightly below the minimum requirement."

Compared to animals fed regular chow, mice on the amino-acid-deficient diet ate normally but grew somewhat slowly. More important, mice fed the deficient chow developed far fewer metastases -- secondary cancers spawned by the initial tumor. While black melanoma growths heavily peppered the lungs of mice eating a normal diet, the lungs of those on the deficient diet remained virtually spotless.

"Quite frankly, I never expected this," Meadows says. "You can't get this effect on metastasis with a drug." He described the findings at a recent American Institute for Cancer Research conference on diet in McLean, Va.

Even if this strategy were to work as well in humans as it has in his dozens of animal experiments, it still would not constitute a cure, Meadows notes.

"The diet isn't preventing a tumor from growing," he told SCIENCE NEWS. "It's changing tumor cells, rendering them uniformly suppressed in their ability to colonize [new tissue]."

Still, this is a major achievement, he notes, since most patients don't die from their initial cancer -- which surgeons can usually remove -- but from its metastases.

The deficient diet seems to alter tumor cells or selectively foster the production of less metastatic ones. As evidence, Meadows points to a study in which his group transplanted tumors. Even in animals eating ordinary chow, tumors derived from mice fed a deficient diet didn't establish new metastatic growths nearly as readily as transplants from animals fed normal food.

Though Meadow's team has yet to isolate the mechanism by which lowered concentrations of phenylalanine and tyrosine thwart metastasis, it's not for lack of trying. They have sought differences in how these cells might attach to new tissue, express important genes, or move in relation to the presence of certain chemical agents.

To date, the most promising lead appears in data on fibroblast growth factor -- a secretion important in angiogenesis, the proliferation of new blood vessels needed to feed a tumor. Tumor cells from animals on a deficient diet "apparently produce the growth factor but don't secrete it," Meadows says.

This dietary manipulation of metastasis "looks interesting, but I don't know how it would be implemented in humans" -- unless foods were engineered with deficiencies of these key amino acids, says metastasis researcher Hynda Kleinman at the National Institute of Dental Research in Bethesda, Md. But it is important to remember, she says, that in most patients "this would not be the only therapy."

Indeed, Meadows envisions physicians one day prescribing such a diet, or perhaps a drug designed to effect the same changes, "as an adjuvant" -- a therapy to augment chemotherapy or immuno-therapy.

At the same meeting, David M. Ota of the University of Texas' M.D. Anderson Cancer Center in Houston described another amino acid therapy to "rescue" patients from the serious side effects of chemotherapeutic agents. Working with animals, Ota and his colleagues altered the ornithine content of a diet delivered parenterally -- into the veins -- as many cancer patients must be fed.

Difluoromethylornithine, a drug to stop the growth of metastases, severely reduces platelet counts in the blood of cancer patients. By boosting the ornithine content of a parenteral diet, Ota and his co-workers blocked the platelet drop without affecting the drug's inhibition of tumor growth.