Polychlorinated alkanes (PCAs) are widely used in products as adhesives. While the environmental and human health risks of PCAs are increasingly recognized, little attention has been paid to their presence in food-contact materials such as fruit stickers. This study represents the first investigation into PCA contamination in fruit stickers and their potential contribution to human dietary exposure. The ∑PCAs-C_10-21 concentrations ranged between 36 and 150 ng/sticker, with the highest contribution from PCAs-C_18-21. Specifically, Cl_1-2 congeners contributed most to ∑PCAs, resulting in low chlorine content in the fruit stickers, i.e. 33% ± 5%, 25% ± 3%, and 16% ± 2%, for PCAs-C_10-13, PCAs-C_14-17, and PCAs-C_18-21, respectively, in contrast to technical PCA mixtures, which are typically dominated by Cl_4-8 congeners. Preliminary tests suggested potential transfer of PCAs from stickers to fruits, with an estimated transfer rate of 23%-36% of ∑PCAs. The average human intake of PCAs via fruit sticker residues was estimated at 4.5 ng/day, with a worst-case scenario of 83 ng/day. The estimated daily intakes were 1-3 orders of magnitude lower than the exposure from other pathways, such as the ingestion of dust.

Consumption of drinking water is a major and actionable human exposure pathway for per- and polyfluoroalkyl substances (PFAS), making it a focus of regulatory efforts. Population-based clinical guidelines recently established serum PFAS levels associated with health risk, but translating these into corresponding external drinking water exposure concentrations remains a data gap. Toxicokinetic (TK) models that relate drinking water PFAS concentrations to population serum levels provide a framework to link clinical guidelines with environmental exposure. In this study, we expand and evaluate a probabilistic, one-compartment TK model to estimate population serum PFAS concentrations for six regulated PFAS. We implement the model to estimate drinking water PFAS concentrations that maintain serum concentrations in sensitive populations below a clinical action level recommended by the National Academies of Sciences, Engineering and Medicine. Model revisions include incorporating population variability distributions for exposure factors, implementing continuous modeling from birth to account for transgenerational transfer, improving computational efficiency, and adding two additional PFAS. Distributions of modifiable parameters are combined using Monte Carlo simulations to predict the distribution of serum PFAS concentrations across defined populations. Model predictions reproduced empirical serum concentrations for populations of infants, children, and adults within two-fold error of observed central tendencies and upper percentiles for six PFAS. Drinking water concentrations that maintain serum concentrations below the National Academies of Sciences, Engineering, and Medicine clinical action level for breastfed infants range from 19 to 162 ng/L for the six PFAS, individually. These drinking water concentrations can inform regulatory, policy, and clinical decision-making, although they are not appropriate as health-protective drinking water standards.

Traditional epidemiological analyses may be limited when addressing examining the environmental etiology of autism spectrum disorder (ASD), including assumptions of linearity, difficulties in handling high-dimensional exposure data, and challenges in detecting gene-environment or exposure-exposure interactions. The emergence of artificial intelligence (AI) - ranging from conventional machine learning to deep learning and large language models (LLMs) - has provided new tools for modeling the complex associations between environmental exposures and health outcomes. Machine learning models have shown strong predictive performance in individual-level air pollution exposure assessment and has enabled the detection of untargeted chemical substances from large-scale mass spectrometry data. In this review, we examine AI applications in four key areas of ASD environmental etiology research: environmental exposure assessment, ASD phenotyping, exposure-outcome association analysis, and risk prediction. Our findings suggest that although LLMs are emerging in environmental toxicology research, ASD environmental etiology studies remain predominantly based on conventional machine-learning methods, with deep learning and LLMs being less applied in this field. Further progress will require large-scale prospective cohort studies, adoption of advanced AI methods, unified data infrastructure, and interation across disciplines.

Although extreme temperatures are recognized as an environmental threat, their contribution to the global burden of childhood cancer remains poorly quantified. We aimed to estimate childhood cancer incidence and mortality rates associated with heat and cold exposures across 204 countries and territories from 1990 to 2021. Daily ambient temperature series from the U.S. National Oceanic and Atmospheric Administration meteorological stations were used to calculate location-specific cumulative heat exposure, cumulative cold exposure, heat waves and cold waves. National estimates of childhood cancer burden among individuals aged 0-19 years in 204 countries and territories were extracted from the Global Burden of Disease 2021 repository. Linear mixed-effects models were fitted to quantify exposure-outcome associations. Subtype-specific analyses were performed for seven solid tumors. A marked global increase in heat-wave frequency has occurred since 1990, particularly in Eastern Europe, Africa, and South Asia. Each additional heat-wave event was significantly associated with 0.32 additional cases per 100,000 population and 19.79 additional deaths per 10,000,000 population. The association with incidence appeared stronger in females than in males. The association of extreme temperatures was predominantly observed in low and low-middle socio-demographic index nations and countries with < 300 million inhabitants, although the heat-wave-associated effect estimate was larger in nations with > 300 million residents. Malignant neoplasms of bone and articular cartilage showed the strongest associations with heat waves. The influence of temperature on the association with childhood cancer burden has strengthened, exhibiting an overall upward trajectory over the study period. This first global analysis demonstrates that extreme temperatures, particularly heat, impose a substantial burden on childhood cancer.

Whether prenatal metal mixtures exert sex-dependent effects on anogenital distance (AGD) - a biomarker of prenatal androgen action - remains unknown. To address this gap, we conducted a prospective cohort study of 248 mother-infant pairs, measuring 10 metals in first-trimester maternal plasma and AGD at age 2 years. To evaluate the associations and identify the most important metal contributors, we combined single-metal linear regression with three complementary advanced computational approaches (extreme gradient boosting, Bayesian additive regression trees, Bayesian kernel machine regression). We further applied propensity score stratification and overlap weighting to strengthen causal inference. The analyses revealed sex-specific patterns: in males, cadmium (Cd), thallium (Tl), and arsenic (As) were inversely associated with anopenile distance (AGD-AP), with Cd showing the strongest effect (-6.96%, P_FDR = 0.013), followed by As (-2.83%, P_FDR = 0.033) and Tl (-2.16%, P_FDR = 0.040). In females, nickel (Ni) was suggestively positively associated with anoclitoral distance (AGD-AC) (6.94%, P = 0.014), though not significant after false discovery rate correction. Machine learning identified Tl and Cd as the top predictors in males, and As in females. These associations were supported by propensity score-based sensitivity analyses, particularly for male Tl and Cd. No significant associations were found for AGD-AS/AGD-AF in either sex. These findings demonstrate that prenatal metal exposure exerts sex-specific opposing effects on AGD, with Cd and Tl in males showing the most robust inverse associations. Our results highlight the value of integrating machine learning with traditional epidemiological approaches and underscore the importance of sex-stratified analyses in mixture exposure analysis.

Toxicokinetics (TK) characterizes the absorption, distribution, metabolism, and excretion (ADME) of chemicals in the body and connects exposure with internal dose and potential toxicity. Because in vivo and in vitro TK assays are costly and low-throughput, and demand for emerging chemicals is growing, artificial intelligence and machine learning (AI/ML) offer alternative tools to predict ADME profiles for data-poor chemicals. This review examines recent developments in AI-based TK parameterization with relevance to environmental exposure assessment. First, we outline the essential elements of AI-based TK parameterization paradigms, including data sources, molecular representations, and commonly used ML algorithms. Second, the recent applications of AI/ML models across four ADME processes are introduced. Third, we discuss current limitations and future perspectives, including model reliability, interpretability, data availability for environmental chemicals, and integrating AI-predicted TK parameters into mechanistic models, such as physiologically based pharmacokinetic (PBPK) models and other downstream modeling frameworks. Although many existing studies rely on pharmaceutical or mixed chemical datasets, these approaches could support TK parameterization and exposure assessment for environmental chemicals. Further progress will depend on improved data quality, broader environmental chemical coverage, and more rigorous model validation and uncertainty evaluation.

Non-persistent pesticides have been associated with altered sex-hormone profiles, but evidence among postmenopausal women remains limited. This cross-sectional study examined associations of individual urinary pesticide biomarkers and their mixture with circulating sex hormones among 593 postmenopausal women from the 2013-2016 National Health and Nutrition Examination Survey (NHANES). Survey-weighted linear regression evaluated individual biomarkers in a directed acyclic graph (DAG)-informed primary model and two progressively expanded sensitivity models incorporating lifestyle, clinical, and dietary covariates. Bayesian kernel machine regression (BKMR) evaluated overall mixture associations, biomarker-specific relative importance, and exploratory bivariate exposure-response patterns. In the primary weighted models, 3,5,6-trichloro-2-pyridinol (TCPY) and para-nitrophenol (PNP) were inversely associated with total testosterone (TT) and the free androgen index (FAI), and the directions were generally consistent across expanded models. No clear associations with estradiol (E2) or sex hormone-binding globulin (SHBG) were observed in the primary models. In BKMR, higher joint biomarker percentiles were associated with lower TT and FAI. PNP had the highest posterior inclusion probability (PIP) for TT (0.921), whereas PNP (0.607) and TCPY (0.523) had the highest PIPs for the FAI. These cross-sectional findings suggest inverse associations of the pesticide-biomarker mixture with androgen-related hormones, with PNP and TCPY showing the highest outcome-specific PIPs within the fitted BKMR models; however, causality cannot be established.