Can cohort studies detect any human cancer excess that may result from exposure to dioxin? Maybe.

We use the available information about concentrations of dioxin in highly exposed humans along with estimates of the half-life of dioxin and durations of exposure to calculate lifetime average daily doses (LADDs) of dioxin in those populations. We compare those LADDs to the LADDs of dioxin used in rodent carcinogen tests, and we also compare the tissue levels of dioxin in exposed humans and the test rodents. Using the maximum likelihood estimate of the Environmental Protection Agency's (EPA) calculated potency factor for the carcinogenicity of dioxin, we calculate the number of cancers that dioxin might be expected to cause in the exposed human populations. If dioxin causes an increase in total cancers or of a common cancer such as lung cancer, the calculated expected increases in cancer are below the limit of detectability in epidemiologic studies. If, on the other hand, dioxin causes a tumor with incidence comparable to that of soft tissue sarcomas, the calculated expected increases should be detectable in dioxin-exposed chemical workers and in the Seveso population. Our calculations lead to the conclusion that the failure of epidemiology to produce convincing evidence of any chronic human health effects from dioxin probably results from the relatively low exposures experienced by humans. From our analyses, there is no reason to allude to differential sensitivities between test animals and humans to explain the lack of convincing evidence for dioxin causing human cancers.