Screening and prioritization of endocrine-disrupting chemicals in plastic toys for children based on non-target analysis and machine learning predictive model

作者：Wenwen Cai, Siyu Li, Yacai Zhang, Boyu Zhao, Keke Wang, Xiao Li, Xue Li, Jie Li, Guang‐Guo Ying, Zhiguo Cao · 发表于：Environment International · 年份：2025 · DOI：10.1016/j.envint.2025.109878 · 被引用次数：5 · 研究领域：Microplastics and Plastic Pollution、Effects and risks of endocrine disrupting chemicals、Recycling and Waste Management Techniques

The health risks associated with the exposure of infants to harmful chemicals, particularly endocrine disruptors, in plastic toys have garnered widespread attention; however, the corresponding knowledge remains limited. In this study, we employed a non-target analysis approach alongside Toxicological Priority Index (ToxPi) model based on toxicity data predicted by machine learning predictive model to systematically screen and prioritize endocrine-disrupting chemicals (EDCs) in plastic toys for very young children (n = 45). A total of 165 compounds were identified and classified into five categories: additives (30.3 %), processing aids (13.3 %), monomers and intermediates of synthetic plastics (11.5 %), non-intentionally added substances (10.9 %), and uncategorizable chemicals (33.9 %). Among these, antioxidants, plasticizers, flame retardants, and surfactants were widely detected. Emerging non-phthalate plasticizers and non-intentionally added drugs were reported for the first time in this study. In addition to the known EDCs (e.g., phthalates), the endocrine disruption prediction results indicated that antioxidants (n = 8) and antibacterial agents (n = 2) exhibited high ToxPi scores among the identified chemicals. Further, exposure risk index was calculated by incorporating both ToxPi scores and the peak intensities of the compounds identified above. Toys made from polyethylene terephthalate, silicone, acrylonitrile-butadiene-styrene, and polystyrene had higher risk index co...