In Response: A Government perspective.

Abstract

The US Environmental Protection Agency's (USEPA) Office of Chemical Safety and Pollution Prevention conducts risk assessments for pesticides under the Federal Insecticide, Fungicide, and Rodenticide Act (FIFRA) and for both new and existing industrial chemicals under the Toxic Substances Control Act (TSCA). Although specific data requirements are associated with pesticide registrations under FIFRA 1, there are no toxicity data requirements prescribed in TSCA for either new or existing chemicals. For both pesticide and industrial chemicals, the Office of Chemical Safety and Pollution Prevention conducts risk assessments that are designed to inform risk-management decisions under FIFRA and TSCA. The level of acceptable uncertainty depends on the decision context. Each year, the USEPA receives more than 1000 new chemical notices, a TSCA requirement for all new chemicals prior to their introduction into US commerce 2. Few submissions (10–15%) have toxicity data included. Therefore, the USEPA relies heavily on data for related chemicals (analogs) and on predictive modeling (quantitative structure–activity relationships and expert systems) 3, to assess the risks of these chemicals within the TSCA–required 90 d. The USEPA is also actively assessing more than 90 industrial chemicals identified as high priority for assessment under the TSCA Work Plan 4. The TSCA Work Plan chemical risk assessments are based solely on currently existing data, tools, and models—there is no base set of data or testing required with notifications and therefore available to conduct these assessments. In addition, the USEPA's Office of Pesticide Programs is reevaluating currently registered pesticides as part of registration review to ensure that products can still be used safely and continue to meet the FIFRA standard of “no unreasonable adverse effects” on human health and the environment. This reevaluation process, which occurs every 15 yr, includes approximately 750 chemical cases for more than 1100 active ingredients. Chemical risk assessments completed as part of registration review, as well as assessments completed for new uses of existing products, are based on registrant-submitted data requirements specified in the Code of Federal Regulations at Part 158, as well as available data from peer-reviewed open literature. Test data used by the Office of Chemical Safety and Pollution Prevention in assessing pesticides and industrial chemicals are ideally derived from established test guidelines and good laboratory practices (GLP); however, data from research studies conducted using nonguideline, non-GLP studies that have been published in the scientific literature are also used routinely. For both pesticide and industrial chemical assessments, the USEPA may use a number of approaches to fill data gaps, including analogs and categories. For both of these approaches, when data are not available for the chemical being assessed, chemicals with similar structural, physicochemical, and toxicological profiles may be used to fill data gaps (i.e., properties or toxicity is read across from one chemical to another), to evaluate predictive reliability, and to assess uncertainty. Predictive tools and models are also used extensively in assessing industrial chemicals under TSCA in the absence of data, to provide a best estimate of a value or a range of values (e.g., low-end, central tendency, and high-end) for physical–chemical properties, fate, exposure, and hazards of chemicals. Such tools and models 3 enable the USEPA to quickly screen and identify chemicals of potential concern. Data for analog read-across and underlying predictive tools and models may be derived from studies conducted using established test guidelines conducted under GLP and research studies published in the open literature. A number of other testing or data challenges are encountered in assessing pesticides and industrial chemicals, including the following: difficult-to-test substances (e.g., low water solubility and high log octanol–water partition coefficient), unknown or variable composition, complex reaction products, and biological substances, volatile substances, use of solvents or carriers in testing systems, accounting for bioavailability (total organic carbon, humic acids, hardness), and accounting for hydrolysis or other degradation during testing 5. Innovative approaches and advancements in measuring the data (e.g., analytical measurement, test systems for chemical delivery, fate, toxicity) needed for assessments, particularly for new or difficult-to-test chemistries, most often emerge from research studies reported in the open literature. Hence, research studies are invaluable for resolving problems and advancing challenges associated with existing testing paradigms relied on by regulatory programs. In conducting chemical risk assessments, the Office of Chemical Safety and Pollution Prevention will consider any and all studies and data available to the USEPA (e.g., industry or government reports, journal publications), whether conducted specifically for FIFRA or TSCA submissions or for other purposes. Non-guideline and non-GLP studies are considered in pesticide and industrial chemical risk assessments. In evaluating adequacy for regulatory risk assessment, the Office of Chemical Safety and Pollution Prevention uses established guidance developed by the USEPA for pesticides 6, the USEPA and the Organisation for Economic Co-operation and Development (OECD) for evaluating high–production volume industrial chemicals 7, and USEPA guidance 8. Studies must include enough information in sufficient detail to allow USEPA analysts to evaluate the robustness and limitations of the study. Following guidelines reminds study conductors not to forget or forgo: chemical identification (e.g., precise isomer, congener, mixture composition, purity, and impurities), test systems and performance notes (e.g., species, exposure apparatus or equipment, water chemistry), analytical methods (e.g., instrumentation, standards, regents), quality assurance/quality control procedures employed and parameters recorded, study results (e.g., controls, raw or uncollated data), study design details (e.g., was unit of measure in study reported?), and statistical and modeling methods used to collate or display results. For studies conducted specifically to assess new pesticide and/or industrial chemical substances, the USEPA strongly encourages submitters to discuss measurement approaches and protocols with the USEPA ahead of time, especially if modifications to test guidelines are needed or if data from research studies or previous published studies are being submitted to support assessment of the new substance. Successful use of research studies, such as those published in the peer-reviewed literature that may not strictly follow established test guidelines, depends on robust communication of methods, test systems, study design, and conditions of the studies. Authors must be willing and able to provide original data (e.g., raw data) prior to statistical analysis. Providing such study details is increasingly possible in scientific journals that provide opportunity for including supplemental materials that will be made available online with the more concise journal publications. Journal reviewers and editors can play a role in facilitating the acceptance of such studies for regulatory application by requesting or requiring that additional study details be included in or with journal article submissions. Standardized study reporting formats exist and could be used readily by authors as a guide for reporting study details. For example, widely accepted, robust reporting templates have been developed by the OECD 9. The OECD harmonized templates are standard data reporting formats for describing salient features of studies performed to determine chemical properties or their effects on human health and the environment. These standard formats were developed flexibly; that is, they were designed specifically for summarizing data contained in any study report or publication related to particular endpoints, including studies that were not conducted according to current guidelines. More than 100 templates are available, covering physical–chemical property measurements, analytical methods, fate property measurements, and ecological and health toxicity 9. These templates include detailed, structured elements (i.e., fields, pick list items, free text prompts, and predefined tables), which are intended to prompt detailed and complete reporting; but it is recognized that not all fields will apply to all tests or studies. Notes to guide the reporter (i.e., help texts) are also available to explain the type of data that are expected in a given field. In conclusion, studies supporting regulatory decision making ideally will have been conducted under optimal conditions using established guidelines and following GLP. However, the USEPA's Office of Chemical Safety and Pollution Prevention will consider any and all available data and information in conducting risk assessments for both new and existing chemicals. In addition, we recognize that innovative approaches and advancements in measuring the data (analytical, fate, toxicity, etc.) needed for assessments, particularly for new or difficult-to-test chemistries, most often emerge from research studies reported in the open literature. Study documentation is key to determining whether a research study will be valuable for use in assessments that support regulatory decision making. Several study reporting formats currently exist and are readily accessible for use by researchers to organize and report supplemental material associated with their studies. Harmonization of reporting and broader application of robust templates for reporting research studies should be encouraged and facilitated through cooperation among researchers, journal reviewers and editors, and government authorities. The views expressed are those of the presenter and do not necessarily reflect USEPA policy or endorsement. Tala Henry Risk Assessment Division, Office of Pollution Prevention and Toxics Office of Chemical Safety and Pollution Prevention US Environmental Protection Agency Washington, DC, USA Anita Pease Environmental Fate and Effects Division, Office of Pesticide Programs Office of Chemical Safety and Pollution Prevention US Environmental Protection Agency Washington, DC, USA