Developing a modern approach to assess ecological risk from pesticides without unnecessary vertebrate animal testing

David Dreier; Christian Picard; Kent Kabler; Natalia Ryan; Haitian Lu; Odette Alexander-Watkins; John Abbott; Richard Currie; Douglas Wolf; Tharacad Ramanarayanan

Just Accepted

This article has been peer reviewed and accepted for publication. It is in production and has not been edited, so may differ from the final published form.

Developing a modern approach to assess ecological risk from pesticides without unnecessary vertebrate animal testing

David Dreier , Christian Picard, Kent Kabler, Natalia Ryan, Haitian Lu, Odette Alexander-Watkins, John Abbott, Richard Currie, Douglas Wolf , Tharacad Ramanarayanan

Abstract

Rationale. Recent directives to reduce animal testing have implications for ecological risk assessment, as several vertebrate tests are used to support these assessments. Therefore, a modern approach was devised to address these key knowledge needs without the use of chemical-specific vertebrate testing. Methodology. An ecological risk assessment for a novel ACCase inhibitor herbicide was conducted using alternative lines of evidence. For fish, chemical toxicity distributions were constructed to quantify the probability of effects, and these distributions were compared with exposure estimates for a representative use in soybeans. The effect distributions were further refined based on invertebrate toxicity and partitioning behavior. For birds and mammals, a joint probability curve was constructed by integrating chemical toxicity distributions and Kenaga exposure distributions. Results. The lines of evidence presented in this predictive risk assessment suggest the intended use of a new ACCase inhibitor is unlikely to affect fish, birds, or mammals. Exposure was unlikely to exceed effect estimates, regardless of whether they were derived based on chemical-read across, invertebrate toxicity, or partitioning behavior. Discussion. Key knowledge needs for ecological risk assessment can be informed by lines of evidence that do not require animal testing. The present study demonstrates such an approach by comparing predicted exposure and effects, which are expected to be protective. This predictive approach can be extended to other active ingredients and chemical classes, as well as other taxonomic groups of interest. Future research should aim to integrate new approach methods in a predictive risk assessment framework.

EN23105 Accepted 08 May 2024