Cancer Guilt by Association

A new set of criteria to judge whether a chemical is likely to cause cancer in humans will get a trial run later this month when a select group of federal scientists will pick a handful of compounds to undergo a battery of tests. But the rules, which rely on a chemical's structure alone to judge its potential for causing cancer, may eventually be tested themselves--in court.

Over the last few years new insights into how chemicals cause cancer have prompted federal agencies to revise their approaches to assessing cancer risk. With such data in mind, the National Toxicology Program (NTP), a federal effort to study the most suspicious chemicals, recently revised its Biennial Report on Carcinogens (BRC)--a compendium that lists substances known or thought to cause human cancer. The new guidelines say that a substance or mixture, even if there is insufficient evidence to classify it as a carcinogen, can be proposed for listing if it "belongs to a well-defined, structurally related class of substances" whose members are listed in a previous BRC. For instance, new types of polycyclic aromatic hydrocarbons, a common industrial pollutant, would be guilty by association. "We can do more with the science now, and we need to," says an NTP spokesperson.

Scientific advisers to the NTP will meet on 18 and 19 November and, under the new guidelines, choose eight to 10 chemicals to undergo rigorous 2-year tests to determine carcinogenicity. The procedure is a double-edged sword for industries that make or use the chemicals. A BRC listing can bring on regulation, or a clean bill can free industries from worry. Any guilty verdicts from NTP that rely on the revised guidelines, says an industry analyst, "could very well test the validity of the guidelines in court."