Not so sweet: cancers in rats that consumed aspartame.

A large, new test in rats suggests that the artificial sweetener aspartame may be a carcinogen. But scientists not affiliated with the research express doubts about the study's validity and point to earlier trials that produced the opposite result.

Aspartame, sold under the brand names Equal and NutraSweet, is used in thousands of products, including diet soft drinks and sugarfree gum. The acceptable daily intake set by the Food and Drug Administration is 50 milligrams per kilogram body weight (mg/kg) per day, the equivalent of about 20 cans of diet soda.

Eight past studies looked at whether the sweetener causes cancer in lab animals, says veterinary pathologist James Swenberg of the University of North Carolina at Chapel Hill. "There's one equivocal study, and the rest are clearly negative for this endpoint [cancer]," he says.

But medical oncologist Morando Soffritti says that the earlier trials were small, funded by aspartame's manufacturer, or never published in detail in the scientific literature. Soffritti led the new study, which appears in the March Environmental-Health Perspectives. He works at a nonprofit organization, the European Ramazzini Foundation of Oncology and Environmental Sciences in Bologna, Italy.

Soffritti and his colleagues examined 1,800 rats, making their study larger than the past aspartame-carcinogenicity tests. The researchers divided the animals into seven groups and gave each group drinking water with either no aspartame or a daily dose of aspartame between 4 mg/kg and 5,000 mg/kg.

The experiment continued until the animals died, which took an average of 2 years. Pathologists dissected each animal to detect any cancers.

The researchers found more cancers, particularly more lymphomas and leukemias, in animals exposed to aspartame than in unexposed animals. The number of cancers detected increased in proportion to the amount of the sweetener that the animal had ingested, the team reports.

That dose-response relationship is a strong sign that aspartame causes cancer, Soffritti says. "We have shown that aspartame is a carcinogenic agent," he concludes.

Because the study is so large and includes a wide range of doses, it has "a fair amount of power," says toxicologist John Bucher of the National Toxicology Program in Research Triangle Park, N.C. The study might detect a subtle carcinogenic effect that had been invisible in previous, smaller studies, he says.

However, the relationship between aspartame dose and cancer incidence reported by the Ramazzini researchers was "not overly strong," Bucher adds. "It's not like [what] one would get with a frank, strong carcinogen."

Lois Swirsky Gold, who directs the Carcinogenic Potency Database at the University of California, Berkeley, says that she isn't convinced that the data reflect an increase in tumors with dose.

Furthermore, says Swenberg, the Ramazzini researchers bucked the convention that laboratory animals be sacrificed at a predetermined point to ensure that their tissues get preserved immediately upon death. The team presumably ended up with less-than-well-preserved tissues, in which identification of cancers would be difficult, he says.

He and Gold both recommend that an independent group of pathologists reevaluate the tissues identified as tumors.

The FDA says that it has requested that the Ramazzini researchers share all their raw data.

STATS

2.2 million

Number of adults in the U.S. with schizophrenia