Scientific developments in the 1990`s have important implications for the assessment of cancer risks posed by exposures to trichloroethylene (TCE). These new developments include: epidemiological studies; experimental studies of TCE carcinogenicity, metabolism and metabolite carcinogenicity; applications of new physiologically based pharmacokinetic (PBPK) models for TCE; and new pharmacodynamic data obtained for TCE and its rhetabolites. Following a review of previous assessments of TCE carcinogenicity, each of these new sets of developments is summarized. The new epidemiological data do not provide evidence of TCE carcinogenicity in humans, and the new pharmacodynamic data support the hypothesis that TCE carcinogenicity is caused by TCE-induced cytotoxicity. Based on this information, PBPK-based estimates for likely no-adverse effect levels (NOAELs) for human exposures to TCE are calculated to be 16 ppb for TCE in air respired 24 hr/day, and 210 ppb for TCE in drinking water. Cancer risks of zero are predicted for TCE exposures below these calculated NOAELs. For comparison, hypothetical cancer risks posed by lifetime ingestive and multiroute household exposures to TCE in drinking water, at the currently enforced Maximum Contaminant Level (MCL) concentration of 5 ppb are extrapolated from animal bioassay data using a conservative, linear dose-response model. These TCE-related risks are compared …