The Tg rasH2 transgenic mouse has been developed as an alternative to the lifetime mouse bioassay to predict the carcinogeni c potential of chemicals. Unlike the p53 / mouse, the Tg rasH2 mouse is sensitive to both genotoxic and nongenotoxi c carcinogens . The Tg rasH2 mouse, of cially designated CB6F1-TgN (RasH2), contains multiple copies of the human c-Ha-ras oncogen e and promoter within its genome. These mice develop spontaneou s and chemically induced neoplasms earlier in life and in greater numbers than wild-type mice, re ecting their enhanced sensitivity to neoplasia. The most common spontaneous neoplasms in control Tg rasH2 mice 8 to 9 months of age are lung adenomas and carcinomas (7.4% incidence), splenic hemangioma s and hemangiosarcoma s (5.4%), forestomach squamous cell papillomas and carcinomas (2.4%), and skin neoplasms (1.2%). Simulations have demonstrated that 20 to 25 mice/sex/treatment group are required to provide the assay with adequate statistical power. Four of 6 known or suspected human carcinogens tested in Tg rasH2 mice were positive in this assay. For 19 nonmutageni c agents testing positive in conventiona l rodent bioassays, 7 chemicals were positive, 10 chemicals were negative, and 2 were equivocal. None of the 10 nonmutageni c rodent carcinogen s that were negative in the Tg rasH2 mouse model are considered to be human carcinogens . All nonmutageni c chemicals that were negative in the conventiona l rodent bioassays were also negative in the Tg rasH2 model. Results for 15 of 18 mutageni c chemicals tested in Tg rasH2 mice agreed with the results of conventiona l rodent bioassays, and 3 results were equivocal. The Tg rasH2 mouse model appears to predict known or suspected human carcinogen s as well as the traditional mouse bioassay, but with fewer positive results for nongenotoxi c compounds that are not considered human carcinogens . The Tg rasH2 mouse model is the most thoroughly tested in vivo alternative to the lifetime mouse bioassay for nongenotoxi c compounds administered by oral or parenteral routes. The U.S. FDA Carcinogenicit y Assessment Committee has determined that the Tg rasH2 model has been adequately evaluated for consideration for carcinogenicity testing of pharmaceutica l candidates and its use could contribute to the weight of evidence for carcinogenicity assessment. The FDA will consider proposals to replace lifetime mouse carcinogenicity studies with 6-month Tg rasH2 mouse studies to support pharmaceutical registration on a case-by-cas e basis.