Dose-time response in mouse skin tumor induction by 7, 12-dimethylbenz[a]anthracene and 12-O-tetradecanoyl-phorbol-13-acetate.

The question was addressed whether the dose-response relationship derived from a carcinogenicity study can be used for mechanistic interpretation and to what extent the shape of the curve is dependent on the duration of the bioassay and the time of analysis. The mouse skin tumor model was used. It allows recording of the time of tumor appearance without interim sacrifice. Groups of 16 female NMRI mice were treated twice weekly by dermal administration with combinations of (a) 2.5 nmol 12-O-tetradecanoyl-phorbol-13-acetate (TPA) plus 0, 0.3, 1, 3, or 10 nmol 7,12-dimethylbenz[a]anthracene (DMBA) or with (b) 2.5 nmol DMBA plus 0, 0.1, 0.3, 1, 3, or 10 nmol TPA. The appearance of the first papilloma was recorded for each animal and the cumulative incidence data were analyzed in two ways: (i) With the usual dose-prevalence representation at a fixed time point, the dose response for TPA was sigmoidal, while it was linear-superlinear for DMBA. This was observed at all time points, indicating that the dose-response information may be used for a distinction between DNA-reactive and tumor-promoting mechanisms of action. (ii) When time-to-tumor and loss of tumor-free lifetime was analyzed as a function of dose, there was again a marked difference between DMBA and TPA. The tumor-free lifetime increased with each step of dose reduction but the slope was about four times larger for TPA compared with that for DMBA. Further reduction of the TPA dose could result in a situation in which the natural life span sets a limit to an observable effect. Under the conditions of this bioassay for mouse skin papilloma induction by a combination treatment with DMBA plus TPA, the findings support the idea of a no-effect low-dose threshold for the tumor-promoting agent.