A new analysis of published studies concludes that antioxidant supplements may not provide protection against several cancers -- and could increase the risk of death.

While some experts say the report adds to the growing case against the use of antioxidant supplements, once thought to be widely beneficial in fighting disease and promoting health, others question the researchers' methods and conclusions. They say the supplements have benefits not reflected in this report.

The findings, published in the Oct. 2 issue of The Lancet, are based on an analysis of 14 randomized trials involving more than 170,000 patients taking antioxidant supplements or placebos. The trials -- about half of which the authors deemed "high quality" by virtue of their design and controls -- reported on patients who developed esophageal, gastric, colorectal, pancreatic or liver cancer.

There is "no convincing evidence that beta carotene, vitamin A, vitamin C and vitamin E or their combinations may prevent gastrointestinal cancers. . . . These antioxidant supplements may even increase mortality," study co-author Goran Bjelakovic, a professor in gastroenterology and hepatology at the University of Nis in Yugoslavia, stated in an e-mail interview.

The analysis looked at antioxidant pills, not at antioxidant vitamins found in foods. Nutrition experts remain in nearly universal agreement that antioxidants found in fresh fruit, vegetables and other whole foods have important benefits. Vitamin-rich whole foods should form the core of a healthy diet, they say.

Some experts said the report sheds new light on an existing concern. "What it does is confirm what we're already learning about antioxidants and heart disease, and that is that they don't provide the protection that we once thought they would," said David Schardt, a nutritionist at the Center for Science in the Public Interest, a Washington-based advocacy group.

Paul Limburg, director of the Mayo Clinic's gastrointestinal neoplasia clinic, said "antioxidants have demonstrated lots of appeal" in other types of research, such as studies based on self-reports of previous supplement use. But data derived from clinical trials -- studies that follow over extended periods people who take either supplements or an inert placebo -- "have been strikingly unremarkable," he said.

Still, Limburg says, the study is not definitive. "It shows, on a best-case scenario, there's no benefit. I think we still need to do further analysis" into the question of harm, Limburg said.

Other experts caution against drawing broad conclusions.

David Forman, a University of Leeds cancer epidemiologist who co-authored an editorial in the same issue of The Lancet, said he "would be very reluctant to draw any conclusions regarding mortality from the evidence that's been presented. What they've got as a result is just on the threshold of being statistically significant."

The Council for Responsible Nutrition, a trade association representing dietary supplement suppliers and manufacturers, questioned the researchers' decision to do a meta-analysis of studies that investigated a variety of antioxidants, rather than analyzing them as individual nutrients.

"They're violating one of the basic tenets [of meta-analysis] by combining things that shouldn't be combined," said John Hathcock, vice president of scientific and international affairs at the council. Some antioxidants are known to have opposing effects, so in a study population taking two supplements, the benefits of one may be masked by the negative effects of the second, he said.

In addition, participants in the studies took high doses of antioxidants, which the authors said might contribute to the negative findings.

Researchers said that the antioxidant levels that provide protective effects are not known and probably differ between individuals. Some nutrients may have benefits at lower levels but negative effects at higher ones.

Participants in the studied trials took doses considerably higher than those found in multivitamins and some antioxidant supplements. For instance, every one to two days they took between 1,500 and 15,000 micrograms of vitamin A, compared with the recommended intake (known as a Dietary Reference Intake, published by the Food and Nutrition Board of the Institute of Medicine) of 700 micrograms for most women and 900 micrograms for most men. Study participants took 30 to 600 milligrams of vitamin E, compared with a recommended intake of 15 milligrams daily for most adults. (The labeling on many bottles of over-the-counter supplements measures some pill contents in international units; some use the metric scale.)

Not all antioxidant supplements were found to be without benefit or potentially risky. Selenium appears to be an exception, Bjelakovic said, potentially leading to a reduction of gastrointestinal cancers.In four trials, selenium supplements reduced the incidence of gastrointestinal cancers, according to the study. The reductions occurred within two to four years after beginning supplementation.

Selenium has been found to offer other potential benefits, leading the National Cancer Institute (NCI) to sponsor further study into its effects. NCI is funding a clinical trial to see if selenium and vitamin E reduce the risk of prostate cancer.

Two trials of beta carotene included in the analysis drew criticism from some experts. The trials, which found that beta carotene had harmful effects, included participants who were smokers or had been exposed to asbestos and were therefore at higher risk for cardiovascular disease and lung cancer.

"No one can tell what was responsible for their increased risk," said Jeffrey Blumberg, a nutrition professor at Tufts University. "Basically this [beta carotene research] is driving the entire conclusion of this report." Blumberg sits on advisory boards for several nutritional pharmaceutical companies and receives honoraria for attending some meetings.

But other previous reports conclude that antioxidant supplements are of limited value. A meta-analysis in the July issue of the Archives of Internal Medicine looked at seven trials involving vitamin E supplementation. It found that the supplements did not help prevent cardiovascular disease and that patients' assumptions about unproven benefits may keep them from developing a healthy lifestyle.

The U. S. Preventive Services Task Force last year said there is insufficient evidence to recommend for or against the use of supplements of vitamins A, C or E; multivitamins with folic acid; or antioxidant combinations for the prevention of cancer or cardiovascular disease.

The task force also recommended against the use of beta carotene supplements, alone or in combination, for prevention of those diseases.

Howard Parnes, chief of NCI's prostate group in the division of cancer prevention, said that, regardless of the infighting about this new analysis, it's clear that consumers should focus more on eating healthy diets and less on taking supplements. With studies showing harm to some populations, and others showing no benefit, there are good reasons to be a cautious consumer of vitamin pills beyond doses found in a multivitamin.

"I really don't think people should be going out and taking large amounts of antioxidants," said Parnes, who said he doesn't take supplements but does eat a "colorful" diet rich in vegetables and fruit. "Eating a good diet is the number one way [to stay healthy]. If people for whatever reason can't or don't eat a good diet, a multivitamin is a good alternative."

*

Antioxidant supplements such as beta-carotene and vitamin A have long been touted as useful tools in warding off disease and promoting health, but newly published research suggests that the pills may actually be harmful.