Recent events have drawn attention to the hypothesis that some xenobiotics in the environment may elicit toxicities in humans by modulating endocrine pathways. From the perspective of regulatory toxicology, pursuit of this hypothesis w ill prove difficult, because current risk assessment m ethods do not readily apply to substances with very high potencies, reversibility, transgener-ational effects, and subtle biological outcomes. Such xenobiotics typically persist in the body and bioaccumulate in the food chain. Yet the exposures are important only during critical periods of vulnerability, and no sustained bio-markers of these important exposures currently exist. This article describes some recent efforts by federal agencies to pursue the hypothesis, including research planning and screening of potential endocrine modulating xenobiotics and risk assessment of dioxin conflicts with its policy for substances that cause thyroid follicular carcinomas.

Induction of cancer by inorganic arsenic occurs inconsistently between species and between routes of exposure, and it exhibits different dose-response relationships between different target organs. Inhaled or ingested arsenic causes cancer in humans but not in other species. Inhaled arsenic primarily induces lung cancer, whereas ingested arsenic induces cancer at multiple sites, including the skin and various other organs. Cancer potency appears to vary by route of exposure (ingestion or inhalation) and by organ site, and increases markedly at higher exposures in some instances. To understand what might explain these inconsistencies, we reviewed several hypotheses about the mechanism of cancer induction by arsenic. Arsenic disposition does not provide satisfactory explanations. Induction of cell proliferation by arsenic is a mechanism of carcinogenesis that is biologically plausible and compatible with differential effects for species or differential dose rates for organ sites. The presence of other carcinogens, or risk modifiers, at levels that correlate with arsenic in drinking water supplies, may be a factor in all three inconsistencies: interspecies specificity, organ sensitivity to route of administration, and organ sensitivity to dose rate.