Abstract
Understanding the potential environmental impact of cosmetics and personal care products (PCPs) and successfully developing eco-friendly formulations requires knowledge of the aquatic toxicity of ingredients. However, for the cosmetics industry, aquatic toxicity assessments are challenged by the large number of ingredients, restrictions on animal testing, and lack of chronic toxicity data. Therefore, alternative approaches are required to fill these gaps. This study aimed to evaluate the potential applicability of acute-to-chronic ratios (ACRs) and in silico tools (VEGA and OECD QSAR Toolbox) to estimate chronic aquatic toxicity values (no observed effect concentrations [NOECs] for algae, daphnids, and fish) for a set of 41 ingredients of cosmetics and PCPs with well-defined molecular structures. Extrapolations and predictions were compared with experimental chronic toxicity data. We observed that ACR-based extrapolations were strongly correlated with experimental NOECs, particularly when using the lowest ACRs, with algae showing the strongest correlation. Mode of action (MoA)-based ACRs did not improve the prediction of chronic values when compared with general ACRs. Extrapolated values were not consistently conservative, with both over and underpredictions observed. The OECD QSAR Toolbox yielded more acceptable chronic toxicity predictions than VEGA, yet both exhibited weaker correlations with experimental data than the ACR-based approach. Predictive performance of the in silico models improved when analyzing the subgroup of MoA 1 substances for algae. Overall, our findings support the use of ACRs for filling chronic toxicity data gaps in screening-level assessments within environmental hazard-based tools for developing eco-friendly cosmetics and PCPs, whereas the freely available in silico tools require further refinement to improve performance in this context.
Funding Information
  • Conselho Nacional de Desenvolvimento Científico e Tecnológico (421823/2022-4)
  • Conselho Nacional de Desenvolvimento Científico e Tecnológico (311120/2023-7)
  • Grupo Boticário

This publication has 41 references indexed in Scilit: