You are here:
Study Design and Initial Results for EPA’s Non-Targeted Analysis Collaborative Trial (ENTACT) (SETAC 11/04/2018)
Citation:
Ulrich, E., J. Sobus, J. Grossman, A. Chao, R. Singh, A. McEachran, K. Mansouri, S. Newton, Chris Grulke, A. Richard, AND A. Williams. Study Design and Initial Results for EPA’s Non-Targeted Analysis Collaborative Trial (ENTACT) (SETAC 11/04/2018). 2018 SETAC North America, Sacramento, CA, November 04 - 08, 2018.
Impact/Purpose:
EPA’s Non-Targeted Analysis Collaborative Research Trial (ENTACT) is being conducted to evaluate the state-of-the-science for suspect screening (SSA) and non-targeted analysis (NTA) approaches. The goal of the work is to demonstrate how the study design and data improves NTA methods.
Description:
Non-targeted analysis (NTA) is a relatively new technique that strives to identify as many compounds present in a sample as possible, providing mostly qualitative results. NTA is useful to fill knowledge gaps about the many environmental chemicals posing exposure potential. However, most NTA approaches are in their infancy, with variable quality control and poor understanding of the domain of applicability. EPA’s Non-Targeted Analysis Collaborative Trial (ENTACT) is designed to address these challenges through a controlled study involving multiple laboratories using different approaches on identical known, but blinded, samples. This presentation will cover the study design for ENTACT, specific results from the EPA analyses, and preliminary results from other participating laboratories (30 total). ENTACT samples include 10 synthetic mixtures containing 95-365 chemicals each, 3 exposure-related media extracts (unaltered and fortified with a mixture), and single chemical standards. Participating laboratories were initially blinded to the contents of the samples. After submission of initial results, participants are unblinded to fully evaluate their performance and allow adjustments to their workflow to improve results. About half the participants have submitted results using a variety of chromatography and mass spectrometry (MS) platforms. EPA analyses with liquid chromatography (LC) time-of-flight MS detected 60% of spiked substances in the mixtures, with true positive rates reaching a maximum of 68% after unblinding. Reproducibility rates averaged 75% for compounds repeated in multiple mixtures. Across a broader range of participants, method comparison between gas chromatography (GC) and LC showed that of 1,269 unique substances in the mixtures, 195 were not detected by either method; 809 were detected by GC (378 only GC), and 801 were detected by LC (265 only LC). Statistical separation of the GConly, LC-only, and non-detected substance sets (Kruskal-Wallace non-parametric test, P <0.05) based on 8 physico-chemical properties (BP, MP, Koa, Koc, Kow, Henry’s Law, VP, and water solubility) predicted from the OPEn quantitative structure-activity Relationship Application (OPERA) suggests differences in chemical space coverage among the 3 sets. ENTACT is the first study of its kind to objectively evaluate the performance of NTA methods on a common set of chemicals, providing a benchmark for current methods, and revealing areas for improvement and development.
