Utilising QSARs to predict fish acute toxicity of pesticide metabolites

In 2013 the European Food Safety Authority (EFSA) published a guidance document which outlined opportunities where non-laboratory testing methods, for example in silico models, may be considered to address plant protection product (PPP) data requirements, specifically those related to pesticide metabolites [1]. We recognised this as an opportunity to establish the utility of quantitative structural activity relationship models (QSARs) in predicting acute fish toxicity of pesticide metabolites, with the aim of increasing industry and regulatory engagement with this type of approach.

In partnership with scientists from Dow AgroSciences, Syngenta, and BASF, we carried out retrospective data analysis comparing experimental fish acute toxicity data for 150 pesticide metabolites within a publicly available database with values predicted using QSARs. The publication detailing this work was published in Regulatory Toxicology and Pharmacology: The utility of QSARs in predicting acute fish toxicity of pesticide metabolites: A retrospective validation approach.

The analysis demonstrated that QSARs have potential to be utilised in the future to predict whether pesticide metabolites are acutely toxic to fish. The application of such a model by the agrochemical industry could ultimately lead to a decrease in the number of fish acute toxicity studies that are carried out to meet regulatory requirements, saving up to 1,800 fish per year across Europe.