Quantification of the Uncertainties in Extrapolating From In Vitro Androgen Receptor Antagonism to In Vivo Hershberger Assay Endpoints and Adverse Reproductive Development in Male Rats

Abstract

Abstract Multiple molecular initiating events exist that disrupt male sexual differentiation in utero including androgen receptor (AR) antagonism and inhibition of synthesis, and metabolism of fetal testosterone. Disruption of androgen signaling by AR antagonists in utero reduces anogenital distance (AGD) and induces malformations in F1 male rat offspring. We are developing a quantitative network of adverse outcome pathways that includes multiple molecular initiating events and key events linking anti-AR activities to permanent reproductive abnormalities. Here, our objective was to determine how accurately the EC50s for AR antagonism in vitro or ED50s for reduced tissue growth in the Hershberger assay (HA) (key events in the adverse outcome pathway) predict the ED50s for reduced AGD in male rats exposed in utero to AR antagonists. This effort included in-house data and published studies from the last 60 years on AR antagonism in vitro and in vivo effects in the HA and on AGD after in utero exposure. In total, more than 250 studies were selected and included in the analysis with data from about 60 potentially antiandrogenic chemicals. The ability to predict ED50s for key events and adverse developmental effects from the in vitro EC50s displays considerable uncertainty with R2 values for HA and AGD of < 6%. In contrast, there is considerably less uncertainty in extrapolating from the ED50s in the HA to the ED50s for AGD (R2 value of about 85%). In summary, the current results suggest that the key events measured in the HA can be extrapolated with reasonable certainty to predict the ED50s for the adverse in utero effects of antiandrogenic chemicals on male rat offspring.