Biological and Dose Thresholds for an Early Genomic Biomarker

Biological and Dose Thresholds for an Early Genomic Biomarker of Liver Carcinogenesis in Mice


Dr. Thomas Hill III


Oak Ridge Institute for Science and Education


Thomas Hill III, Virunya Bhat, Charles E. Wood


Oak Ridge Institute for Science and Education; National Health and Environmental Effects Research Laboratory, U.S. Environmental Protection Agency, Research Triangle Park, NC, USA; NSF International, Ann Arbor, MI, USA


Traditional data sources for cancer risk assessment are resource-intensive, retrospective, and not feasible for the vast majority of environmental chemicals. The use of quantitative short-term genomic biomarkers may streamline this process by providing protective limits for known pathways of chemical carcinogenesis. In this study, we examined biological and dose thresholds for the Cyp2b10 gene marker of constitutive androstane receptor (CAR) activation in mice. Sustained CAR activation is an established mitogenic pathway in the liver and the most common mode of action evaluated by the U.S. EPA. Liver Cyp2b10 expression at 7 days (fold-change vs. respective control group), cell proliferation at 7 days (labeling index), and tumor incidence at 18-24 months were analyzed across five established CAR-activating liver carcinogens (Cyproconazole, Nitrapyrin, Pronamide, Sulfoxaflor, & Tetraconazole) previously tested in male mice. Statistically significant dose relationships (p<0.05) were identified by Pearson’s correlation, supporting the calculation of hypothetical tumorigenic thresholds using linear regression. Benchmark dose (BMD) estimation was then used to compare Cyp2b10 and tumor incidence dose potencies within each chemical. A mean Cyp2b10 induction threshold of 69-fold (SEM ±22, range 31-136) corresponded to a significant increase in liver tumor incidence at p<0.05. Similarly, a Cyp2b10 induction threshold of 65-fold (SEM ±16, range 33-120) corresponded to the hypothetical tumorigenic threshold for liver cell proliferation within each chemical. Overall, BMD estimates for Cyp2b10 induction were 2.6 times higher (± 1.1 SD, range 0.2-6.5) than BMD estimates for tumor incidence. These findings support the idea that early genomic changes can be used to establish threshold estimates or “molecular tipping points” that are predictive of later-life outcomes. Future work in this area will inform efforts to incorporate new data types into cancer risk assessment.

This abstract does not necessarily represent the views or policies of the U.S. EPA.