SAR and Modeling

SAR plays a prominent role in TSCA screening of new chemicals and existing chemicals in commerce. SAR models for bioavailability, ecotoxicology and health toxicology endpoints are being used to identlify chemicals with the greatest potential for ecological or health hazard, to set testing priorities, and to provide scientific support for a testing recommendation. SAR models vary considerably in accuracy and utility for screening application depending upon the quality of available data and level of current knowledge for a toxicity endpoint. The main deficiency Of current SAR capabilities is inadequate data, and lack of knowledge of mechanisms of toxicity for many chemical classes and toxicity endpoints of potential regulatory concern. Knowledge or inference of a common mechanism of toxic action is crucial for selecting appropriate chemical analogues, guiding SAR model development, establishing model plausibility, and providing the necessary scientific rationale for model acceptance and use in prediction. This paper discusses the present role and capabilities of SAR in TSCA screening, general features and limitations of SAR, current and evolving SAR technologies, and advances most likely to lead to improvements in SAR models for use in TSCA screening. Although SAR has the clear potential to further reduce the need for testing or eliminate testing in some circumstances, the promise of SAR will not be fulfilled without proper application of these methods. This entails clear recognition of the limitations of SAR, and appreciation for the essential roles of research into mechanisms of toxicity, and strategic testing for further SAR model development and refinement. A structure-activity relationship (SAR) relates features of chemical structure to a property, effect, or biological activity associated with that chemical. The fundamental premise is that the structure of a chemical determines its physical properties and activities. The term “structureactivity relationship” has taken on a wide range of meaning over the years, from heuristic chemical associations and human expert approaches that consider primarily structural features, to formal mathematical relationships that relate specific chemical attributes to a quantitative measure of the property or activity of interest, the latter being commonly referred to as “quantitative structureactivity relationships” (QSARs). In both the pharmaceutical and chemical industries, structure-activity considerations have long been used to design chemicals with commercially desirable properties. In the environmental protection field, SAR is being used to predict adverse ecological and health effects, with applications ranging from the prediction of relevant properties, such as chemical stability, bioavailability and bioaccumulation, to the prediction of various forms of chemical toxicity. The focus of this workshop is testing and screening strategies for review of the Toxic Substances Control Act (TSCA) inventory of existing chemicals in commerce. This problem poses a significant and immediate challenge, not only in terms of the sheer numbers of chemicals that have undergone little testing or review (>10,000), but also in terms of the multiple exposure routes and ecological and health endpoints of potential concern. The foremost goal is to identify the chemicals that pose the greatest potential ecological and health risks, and to strategically allocate limited testing resources to best characterize these risks. SAR, coupled with exposure and use estimates, represents the top tier in a multiple tier screening approach for assessing chemical hazard, and provides the primary means for setting testing priorities. SAR currently plays a prominent This manuscript has been reviewed by the U. S. Environmental Protection Agency and approved for publication. Approval does not signify that the contents necessarily reflect the views and policies of the Agency, nor does mention of trade names or commercial products constitute endorsement or recommendation for use.