You are here:
Evaluation of toxicity equivalent calculations for use with data from in vitro aromatase inhibition assays
Citation:
Watanabe, K., W. Cheng, Dan Villeneuve, R. Conolly, E. Perkins, AND M. Mayo. Evaluation of toxicity equivalent calculations for use with data from in vitro aromatase inhibition assays. SETAC North America, Salt Lake City, UT, November 01 - 05, 2015.
Impact/Purpose:
not applicable
Description:
With growing investment in alternatives to traditional animal toxicity tests, the next generation of risk assessment must interpret new streams of data to identify hazards and protect humans and wildlife populations. If the effects of a chemical can be characterized by a battery of high throughput in vitro assays, such as those used in the US EPA ToxcastTM program, then, theoretically, biologically-based mathematical models can be developed to utilize a set of assay results to predict effects in a whole organism. Challenges arise, however, in extrapolating in vitro assay data to produce meaningful metrics that enable mathematical models to predict in vivo effects robustly. This presentation focuses on how different metrics derived from in vitro assay data affect model predictions of reproductive endpoints in the fathead minnow (Pimephales promelas). In the context of a quantitative adverse outcome pathway, we investigated how different results from two aromatase inhibition assays ultimately affect model predictions of fathead minnow fecundity. Toxicity equivalents (TEQs) were computed for known aromatase inhibitors, propiconazole and prochloraz, based upon (i) measured EC50s, and (ii) the area under the assay response curve (AUC). The TEQs were used to adjust aromatase inhibition parameters in a hypothalamic-pituitary-gonadal (HPG) axis model, and HPG axis model predictions of plasma vitellogenin concentration time series were input into an oocyte growth dynamics model. Preliminary results indicate that in addition to assay response differences, the choice of a TEQ metric has an impact upon model-predicted fecundity. Thus, in the next generation of risk assessment it will be important to understand what in vitro assays are actually measuring and to develop extrapolation methods that can harmonize results from multiple assays.