You are here:
Using the Concordance of In Vitro and In Vivo Data to Evaluate Extrapolation Assumptions
Citation:
Honda, G., R. Pearce, L. Pham, R. Setzer, B. Wetmore, N. Sipes, J. Gilbert, B. Franz, R. Thomas, AND J. Wambaugh. Using the Concordance of In Vitro and In Vivo Data to Evaluate Extrapolation Assumptions. PLOS ONE . Public Library of Science, San Francisco, CA, 14(5):1-33, (2019). https://doi.org/10.1371/journal.pone.0217564
Impact/Purpose:
There are many chemicals in commercial use with little to no information regarding their potential toxicity. To address this issue and avoid the high cost of animal studies, new approach methodologies are being developed that use isolated cells in culture to identify the biological processes and pathways that may be disrupted by chemicals. The computational methods evaluated in this work help link the concentration at which an effect is observed on the cellular level to an administered dose to a person or animal. This work demonstrates that transforming in vitro concentrations to administered equivalent doses increases the concordance between in vitro bioactivity and in vivo responses. We evaluate different IVIVE assumptions by comparing the correlation between in vitro bioactive concentrations and predicted in vivo internal concentrations. Since there is no single assay that can be expected to perfectly predict a given endpoint, we examine the of amount of correlation between all available assay-endpoint combinations and compare the summarized results.
Description:
• In vitro-in vivo extrapolation (IVIVE) enables the evaluation of potential chemical toxicity via high-throughput screening (HTS) assays as an alternative to resource intensive animal studies • We report new rat-specific high throughput toxicokinetic (HTTK) data for 65 chemicals which we analyze jointly with other rat specific HTTK data from the literature. We use these data to parameterize a high throughput physiologically-based toxicokinetic (HT-PBTK) model to then predict plasma and tissue concentrations associated with points of departure (PODs) for various toxicological endpoints observed in vivo in rat. • We evaluate different IVIVE assumptions by comparing the correlation between in vitro bioactive concentrations and predicted in vivo internal concentrations. Since there is no single assay that can be expected to perfectly predict a given endpoint, we examine the of amount of correlation between all available assay-endpoint combinations and compare the summarized results. • We demonstrate that PBTK reveals the correlation between in vitro screening assays and in vivo toxicity. However, given the limitations of the data sets available for direct evaluation of IVIVE, a unique set of "optimal" model assumptions could not be identified.
URLs/Downloads:
DOI: Using the Concordance of In Vitro and In Vivo Data to Evaluate Extrapolation Assumptions