You are here:
AOP-driven Predictive Models for Carcinogenicity: an exercise in interoperable data application.
Citation:
Hill, T., C. Corton, AND C. Wood. AOP-driven Predictive Models for Carcinogenicity: an exercise in interoperable data application. OpenTox Euro 2017, Basel, Switzerland, SWITZERLAND, November 21 - 23, 2017.
Impact/Purpose:
Dr. Hill was selected as a platform speaker by Douglas Connect, GmbH. (the conference sponsor) for OpenTox Euro 2017, which will take place at the Hotel Stücki and Technology Park Basel, November 21-23, 2017, in Basel, Switzerland. OpenTox is an international organization whose goal is development of an interoperable toxicology data framework that can be used worldwide to create novel applications for predictive toxicology. Their work actively supports the development, validation and reproducibility of in silico models in addition to integration of novel data streams and strategies such as the Adverse Outcome Pathway (AOP) framework into environmental and human health assessment. OpenTox Euro 2017 will address topics of international interest in the field such as defined approaches for regulatory application, classification of experimental designs, integration of software platforms and data sets in big data projects, and the computational modeling of exposure and transport. Integration of independent data sets within an AOP framework is critical to Dr. Hill’s research for the US EPA Chemical Safety for Sustainability (CSS) National Research Program (https://www.epa.gov/chemical-research) under the mentorship of Dr. Charles Wood. In combination with computational modeling using novel data streams, these concepts are also foundational to the success of Dr. Hill’s new projects on quantitative AOPs for cancer under the mentorship of Dr. Rory Conolly. This speaking invitation to present his research and encourage both data sharing and networking within this audience of peers will enhance Dr.Hill's ability to conduct assigned projects, and provide professional development and career development commensurate with the goals and objectives as outlined in his ORISE project description. In addition to dissemination of scientific findings at the international level, any networking and data sharing connections initiated by Dr. Hill will enhance the future collaborative potential and research opportunities for the AOPDD project, the CSS program, and other scientific and regulatory activities within the US EPA.
Description:
Traditional methods and data sources for risk assessment are resource-intensive, retrospective, and not a feasible approach to address the tremendous regulatory burden of unclassified chemicals. As a result, the adverse outcome pathway (AOP) concept was developed to facilitate a more quantitative and prospective appraisal of the biology leading to a toxicological (adverse) outcome. A central premise of this framework is that key events are requisite but not sufficient at a qualitative level to produce an adverse outcome. Progression beyond a key event (KE) depends on the degree or amount of disruption needed to propagate the key event relationship (KER) towards the adverse outcome. This idea has led to increased interest in quantitative biologic thresholds, or “molecular tipping points,” as a basis for adversity determinations using short-term data. This concept is critical for pathway-based risk assessment because it would prescribe chemical-agnostic hazard levels that enable predictive evaluations within a specific AOP construct. When evaluated against AOP-derived biological thresholds for known pathways of chemical carcinogenesis, quantitative genomic data from short-term studies have the potential to reduce the chemical testing burden, expedite the assessment process and redirect limited resources to chemicals of immediate concern. This presentation will describe the development of AOP-driven biologic thresholds from archival data (ToxCastTM, ToxRefDB, Pharmacopia, TG-GATES) as potential screening tools using a rodent liver cancer model. Operational examples will be provided for both an AOP single-gene marker and an AOP-driven genomic fingerprinting system that can identify putative carcinogens as well as classify their pathway association across six discrete AOP for rodent liver cancer. These examples will demonstrate evolutionary new approaches in predictive toxicology as well as the obligate requirement for interoperable data applications to effectively safeguard our global environmental health from a legion of unclassified chemicals. This abstract does not necessarily represent the views or policies of the U.S. EPA.