Alternative models for low dose-response analysis of biochemical and immunological endpoints for...

As part of the current reassessment of dioxin, the empirical relationships between administered tetrachlorodibenzo-p-dioxin and selected immune and biochemical endpoints were investigated. A dose-response analysis from the published literature was performed using Linear, Sigmoid-Emax and Power Law functions. The results of this analysis indicate that the use of a wide dose range may bias the interpretation of low-dose phenomenon. The Power Law function was applied exclusively to low-dose subsets enabling estimation of dose response in the low-dose range. Subsequently, high-dose data were fit using Power Law subset analysis. This approach resulted in a change in slope value from low- to high-dose subsets for thymic atrophy, immune suppression, benzo(a)pyrene hydroxylase activity, and ethoxyresorufin-o-deethylase activity. This change suggests that there is a high probability that there is a tissue concentration along the dose-response continuum which results in biological activity from low to high dose. This analysis also demonstrates that the Power Law functional fit to the low-dose data differs quantitatively from the fit to the high-dose data for several noncancer endpoints. Therefore, by using the Power Law function a more accurate and biologically relevant assessment of risk can be produced for non-cancer endpoints.