Assessing how in vitro assay types predict in vivo toxicology data

ABSTRACT

In vivo animal bioassays are increasingly being supplemented with in vitro assays to serve as the new standard for chemical toxicity tests. Despite this shift, investigators face challenges related to increased reliance on in vitro data. The aim of this study was to deploy a streamlined method to assess the ability of in vitro data to predict similar results as in vivo data by correlating chemical toxicity rankings obtained using Benchmark Doses and Benchmark Dose Lower Limits (BMD(L)s) derived from in vivo and in vitro assays. In vitro and in vivo assay characteristics were assessed for their impact on the predictive ability of in vitro data. Minimum best-fit BMD(L)s were calculated for chemicals using Environmental Protection Agency’s (EPA’s) Benchmark Dose Software (BMDS). Forty-one chemicals met the inclusion criteria of this study. Relative chemical toxicity rankings were assessed through Kappa statistics, Pearson correlations, and/or Ordinary Least Squares (OLS) regressions. Results illustrated likely ability of in vitro data to predict similar results as short-term in vivo data. Further, rankings derived from in vitro cytotoxicity assays, unlike stress response assays, significantly correlated with rankings derived from short-term in vivo assays. These results support the use of in vitro data as a prioritization tool within toxicity testing.

KEYWORDS:

• In vitro
• in vivo
• benchmark dose
• benchmark dose lower limit
• correlation

Acknowledgments

The views expressed in this paper are those of the authors and do not necessarily reflect the statements, opinions, views, conclusions, or polices of The George Washington University.

Declaration of Interest Statement

No potential competing interest was reported by the authors.