EPA’s ToxCast™ Research Program
The objective of the ToxCast™ research program developed by The National Center for Computational Toxicology (NCCT) of EPA’s Office and Research and Development is to develop cost-effective innovative approaches to rapidly screen and prioritize many chemicals for further toxicological testing.
Using data from state-of-the-art high throughput screening (HTS) bioassays developed by the pharmaceutical industry, ToxCast™ is building computational models to better predict the potential human toxicity of chemicals. These hazard predictions should provide the Agency's regulatory programs with science-based information that will be helpful in setting priorities for more targeted toxicological evaluations. This will help the Agency focus toxicity testing on those chemicals and effects with the greatest potential for causing harm.
On this page:
- ToxCast™ Phases
- ToxRef Database and Model Development
- ToxCast™ and Pesticide Risk Assessment
- More Information
ToxCast™ is a multi-year effort that is divided into three distinct phases:
- Phase I: Proof of Concept - 300 chemicals of known toxicity tested in over 400 different HTS bioassays. The results will be used to create a predictive bioactivity signature for each toxic effect;
- Phase II: Confirmation and Expansion - generating HTS data on at least 300 additional chemicals for comparison with Phase I results;
- Phase III: Expansion to thousands of environmental chemicals and pesticides for which little toxicological information is available. Results of HTS bioassays will be used to predict the toxicity of these chemicals.
The ToxCast™ program is committed to transparency and public release of all data and analysis tools. This approach is intended to build confidence in the outputs for regulators and stakeholders. Updates on the program are available through monthly “chemical prioritization community-of-practice” conference calls. Additional goals of the program are to capture equivalent information on the exposure potential of chemicals in the environment so that priority-setting can be based on potential for both hazard and exposure.
An important aspect of the ToxCast™ research program is the development of a database containing animal toxicity studies called ToxRefDB, which is a joint effort between NCCT and EPA’s Office of Pesticide Programs (OPP). Pesticide manufacturers are required to submit standard animal toxicity studies on active ingredients in pesticides to license a pesticide product. The Agency reviews and evaluates these required toxicity studies in a “Data Evaluation Record” (DER). ToxRef Database (DB) contains information from these DERs in a standardized and searchable format.
The ToxRefDB is critical to developing the ToxCast™ predictive approach because it contains data on chemicals of known toxicity. The known toxic effect of each chemical will be used to interpret the results from the Phase I and II HTS bioassays. This will provide a library of HTS bioactivity signatures for toxic effects which will be used to develop computational models to predict toxicity. The ToxCast™ Web page will include the information in ToxRefDB and data from the HTS bioassays for all chemicals involved in ToxCast™.
OPP’s DERs are not the only source of information being used in the ToxCast™ program to support the development of predictive models. Comparable toxicity data from other toxicity databases (e.g., National Toxicology Program) are also being captured in the ACToR (Aggregated Computational Toxicology Resource) database.
ToxCast™ will be building computational models to better predict the toxicity potential of environmental chemicals. The Office of Prevention, Pesticides and Toxic Substances anticipates taking advantage of these models to prioritize in vivo animal testing of products that have limited toxicity data available such as:
- antimicrobial pesticides
- inert ingredients in pesticide products
- manufacturing process impurities
- metabolites and environmental degradates of concern
- new and existing industrial chemicals
Filling these gaps in our knowledge of the toxicity of chemicals will improve our ability to protect human health and the environment.
The ToxCast™ research program has included approximately 300 conventional pesticide active ingredients in Phase I of the program. Pesticides were chosen because many animal toxicity tests have been done in the past, yielding considerable data that can be used as the basis for developing computational models to predict toxicity. All of the pesticides included in the ToxCast™ research program have been previously approved for registration and have met the safety standard for registration.
For a pesticide to be registered, its intended use must not cause “unreasonable adverse effects on the environment.” Unreasonable adverse effects on the environment include unacceptable risk to humans and the environment and large economic effects. Social and environmental costs and benefits of the use of any pesticide are also taken into account. EPA requires a manufacturer to provide scientific data from more than 100 different studies concerning potential hazard and exposure as the basis for assessing the safety of a pesticide.
It should be noted that the animal toxicity data included in ToxRefDB does not represent all of the toxicity information that EPA has on ToxCast™ pesticides. At this time, ToxRefDB includes:
- rat chronic and cancer assays,
- mouse cancer assays,
- rat 90-day toxicity studies,
- rat multigenerational reproduction studies, and
- rodent and rabbit prenatal developmental studies.
For registration purposes, in addition to these studies, EPA has information on other mammalian toxicity effects (e.g., neurotoxicity, metabolism), aquatic life, wildlife toxicity and plant toxicity studies, and use patterns, environmental fate and persistence, and pesticide residue levels.
ToxRefDB contains only hazard information and does not represent all the information needed for a complete risk assessment for pesticides. Determining risk requires information on both pesticide hazard and exposure. For example, liver toxicity may be reported in a rat study, but if the effective dose is 100 to 1000 times higher than what humans will encounter, then that pesticide and use is not a risk for liver toxicity in humans.
Currently, ToxCast™ is still under development and when the ToxCast™ model and data have been adequately peer reviewed and undergone public comment, OPP will consider this new screening tool for priority-setting.
- ToxCast™ Program [Dix et al. 2007. The ToxCast™ program for prioritizing toxicity testing of environmental chemicals. Toxicol Sci. 95(1):5-12]
- ACToR (Aggregated Computational Toxicology Resource)
- National Center for Computational Toxicology
- Chemical Prioritization Community of Practice