Alternatives for a risk assessment on chronic noncancer effects from oral exposure to trichloroethylene.

Changes in methodologies are presently occurring for dose-response assessment in noncancer and cancer risk assessments. The benchmark dose (BMD) method is an alternative to the no-observed-adverse-effect level (NOAEL)/uncertainty factor (UF) approach for development of toxicity values. A comparison of these two methods was undertaken using trichloroethylene, an important industrial chemical and environmental contaminant. This analysis considered liver effects, kidney toxicity, and developmental defects. A range of toxicity values was obtained using the two methods from which acceptable drinking water concentrations were estimated: 1000-10,000 ppb for liver effects, 1750 ppb from kidney toxicity, and 1000-10,000 ppb from developmental defects of the eye. These values are all higher than those based upon cancer as the critical endpoint. This analysis highlighted the strengths of the BMD approach in the presence of adequate dose-response data, but it also suggested that guidance is required for addressing inadequate dose-response data. The selection of UF and critical studies were identified as areas that have a large impact upon the final dose-response values, sometimes greater than the variations arising from using the BMD rather than the NOAEL.