Citation:

Isaacs, K., T. Wall, K. Paul-Friedman, J. Franzosa, H. Goeden, A. Williams, K. Dionisio, J. Lambert, M. Linnenbrink, A. Singh, J. Wambaugh, A. Bogdan, AND C. Greene. Screening for drinking water contaminants of concern using an automated exposure-focused workflow. Journal of Exposure Science and Environmental Epidemiology . Nature Publishing Group, London, Uk, 34:136-147, (2024). https://doi.org/10.1038/s41370-023-00552-y

Impact/Purpose:

The US Environmental Protection Agency’s Center for Computational Toxicology and Exposure (CCTE) and the Minnesota Department of Health (MDH) are collaborating to use new chemical data generated from scientific approaches such as read-across, QSAR, high-throughput toxicology screening, and computational modeling of exposure and toxicokinetics to prioritize chemicals for further evaluation and inform risk assessments of emerging contaminants of concern. This peer-reviewed manuscript describes the development of an automated workflow for scoring and prioritizing chemicals based on their exposure potential via drinking water, using data and models from EPA's ExpoCast project.

Description:

Background: The number of chemicals present in the environment exceeds the capacity of government bodies to characterize risk. Therefore, data-informed and reproducible processes are needed for identifying chemicals for further assessment. The Minnesota Department of Health (MDH), under its Contaminants of Emerging Concern (CEC) initiative, uses a standardized process to screen potential drinking water contaminants based on toxicity and exposure potential. Objective: Recently, MDH partnered with the U.S. Environmental Protection Agency (EPA) Office of Research and Development (ORD) to accelerate the screening process via development of an automated workflow accessing relevant exposure data, including exposure new approach methodologies (NAMs) from ORD’s ExpoCast project. Methods: The workflow incorporated information from 27 data sources related to persistence and fate, release potential, water occurrence, and exposure potential, making use of ORD tools for harmonization of chemical names and identifiers. The workflow also incorporated data and criteria specific to Minnesota and MDH’s regulatory authority. The collected data were used to score chemicals using quantitative algorithms developed by MDH. The workflow was applied to 1867 case study chemicals, including 82 chemicals that were previously manually evaluated by MDH. Results: Evaluation of the automated and manual results for these 82 chemicals indicated reasonable agreement between the scores although agreement depended on data availability; automated scores were lower than manual scores for chemicals with fewer available data. Case study chemicals with high exposure scores included disinfection by-products, pharmaceuticals, consumer product chemicals, per- and polyfluoroalkyl substances, pesticides, and metals. Scores were integrated with in vitro bioactivity data to assess the feasibility of using NAMs for further risk prioritization. Significance: This workflow will allow MDH to accelerate exposure screening and expand the number of chemicals examined, freeing resources for in-depth assessments. The workflow will be useful in screening large libraries of chemicals for candidates for the CEC program.

URLs/Downloads:

Record Details: