Use of mouse liver tumor data in risk assessments performed by the U.S. Environmental Protection Agency.

In summary, the EPA has begun to look critically at the induction of certain types of tumors in certain species, including liver tumors in mice. The controversy over the use of such tumor data in assessing the cancer risk for humans has been going on for some time. The present agency policy is to downgrade the weight of evidence for such data under certain conditions. Review of the cancer risk assessments for the 109 chemicals that the agency has formally verified shows that a variety of chemicals yield liver tumors in mice. However, one group of substances that consistently produced such tumors was chlorinated compounds (84%). Many of these compounds not only induced liver tumors in mice but also induced liver tumors in rats and/or other types of tumors in mice and rats. However, several of the chlorinated compounds produced only mouse liver tumors. Another group of compounds that often induced liver tumors in mice was nitrogen-containing compounds (aromatic amines, hydrazines, nitrosamines). These latter substances tended to not only induce liver tumors in mice but also a variety of other tumor types in a variety of species. Mouse liver tumor data have played a major role in the classification of substances in categories B2 and C. Fifty-six percent of the chemicals in category B2 and 40% in category C were classified based at least partially on the use of mouse liver tumor data. In addition, 21 of the 29 category B2 chemicals that produced liver tumors in mice and 5 of the 8 category C chemicals are chlorinated compounds. These two results indicate the importance of chlorinated compounds to the agency, and therefore, the importance of mouse liver tumor data in agency cancer risk assessments.