Screening Chemical Alternatives with Limited Ecotoxicity Data To Support Safe and Sustainable by Design Innovation

Abstract

Enabling effective early-stage screening of chemicals is essential for advancing safety and sustainability, as highlighted by the European Commission’s Safe and Sustainable by Design framework. However, such screening is often constrained by the limitedavailability of ecotoxicity data. Here, we investigate how chemical screening can be performed under data-limitedconditions. We analyzed 132 chemicals with extensive chronic ecotoxicity data and derived Environmental Risk Assessment (ERA) and Life Cycle Assessment (LCA) metricsEnvironmental Threshold Concentrations (ETCs) and Ecotoxicity Effect Factors (EEFs). Eight data-reduction scenarios were compared with a data-rich reference, varying (i) acute versus chronic endpoints, (ii) taxonomic coverage (single species vs three trophic levels: algae-fish-crustaceans), and (iii) calculation method. Across ERA and LCA, the scenario including three trophic levels and chronic data showed the lowest uncertainty and best agreement with the reference. Single-species scenarios introduced substantial uncertainty, particularly for specifically acting chemicals. Replacing chronic with acute data increased uncertainty only moderately. For ERA, ETCs from the lowest toxicity value aligned more closely with the reference than ETCs from species sensitivity distributions. For LCA, the calculation method had limited influence on EEFs. Our approach can support transparent decision-making for the safe and sustainable design of chemicals, where large uncertainties are unavoidable.