Alternative testing strategies for predicting developmental toxicity of antifungal compound

Determination of safe human exposure levels of chemicals in toxicological risk assessments largely relies on animal toxicity data. In these toxicity studies, the highest number of animals are used for reproductive and developmental toxicity testing. Because of economic and ethical reasons, there is large interest in the development of in vitro and/or in silico test systems as alternatives for the animal studies. The aim of the present thesis was to evaluate the applicability of combined in vitro approaches taking toxicokinetic and toxicodynamic aspects into account, as well as of an integrated in vitro and in silico approach for prediction of developmental toxicity using a series of antifungal compounds as the model compounds. Transplacental transfer of compounds is highly likely to play an important role in developmental toxicity, so we developed and validated an in vitro placental barrier model using BeWo b30 cells to predict placental transfer. Then we investigated the applicability of the ES-D3 cell differentiation assay combined with the in vitro BeWo transport model to predict the relative in vivo developmental toxicity potencies of two sets of selected antifungal compounds. The data obtained show that the combined in vitro approach provided a correct prediction for the relative in vivo developmental toxicity, whereas the ES-D3 cell differentiation assay as stand-alone did not. In order to detect specific structural alterations induced by chemicals<strong>,</strong> we investigated the applicability of the ex ovo assay of chicken embryos to predict the specific alterations induced by the antifungal compounds. Data revealed that the ex ovo assay of chicken embryos is able to assess the teratogenic potential of antifungal compounds, and, when combined with the in vitro BeWo transport model, is able to better predict relative in vivo prenatal developmental toxicity potencies. Subsequently, we translated in vitro concentration–response data of the antifungal compound tebuconazole, obtained in the ES-D3 cell differentiation assay and the ex ovo assay of chicken embryos, into predicted in vivo dose–response data using physiologically based kinetic (PBK) modelling-facilitated reverse dosimetry. The results show that the BMD10 values from predicted dose–response data from both assays are in concordance with BMD10 values derived from in vivo data (within 5-fold difference). This revealed that PBK modeling is a promising tool to predict in vivo dose-response curves based on the results of in vitro toxicity assays, and may therefore be used to set a point of departure for deriving safe exposure limits in risk assessment. It is concluded the combined in vitro approaches and the integrated in vitro-in silico approaches appear to be promising for the screening and prioritization of chemicals and to provide reference values, such as BMD10 values, without using animals, therefore contributing to the 3R principle of animal testing.