Mengling Liu

The prenatal period is a critical window for neurodevelopment, during which maternal metabolic processes are increasingly recognized to shape later behavioral and autism-related traits. We examined associations between targeted maternal urinary metabolomic profiles across pregnancy and preschool emotional, behavioral, and autism-related outcomes in the New York University Children’s Health and Environment Study (NYU CHES), a large prospective birth cohort. Targeted metabolomics was conducted on maternal urine samples collected in early, mid, and late pregnancy. Child outcomes were assessed at a mean age of 2.3 years using the Child Behavior Checklist (CBCL) total and DSM-5–oriented autism spectrum disorder (ASD) subscale scores. We applied two-stage mixed-effects models to assess overall pregnancy metabolite levels and timepoint-specific negative binomial models to evaluate trimester-specific associations, with covariate adjustment, sex-stratification, and false discovery rate (FDR) correction. Few associations were observed two-stage models. In timepoint-specific analyses, a robust sex- and timing-specific association emerged: among males, higher third-trimester urinary lysoPC a C26:1 was associated with lower CBCL total and ASD subscale scores, surviving stringent FDR correction and remaining robust after adjustment for maternal diet quality. Additional associations involving taurine, inflammatory markers, and acylcarnitines were observed primarily in sex-stratified or diet-adjusted models but were less robust to multiple testing correction. These findings suggest that prenatal metabolic profiles, particularly lipid- and mitochondrial energy-related pathways in late gestation, may contribute to early behavioral and autism-related traits in a sex-specific manner, underscoring the importance of exposure timing and maternal metabolism in the etiology of neurodevelopmental outcomes.

BackgroundEmerging evidence suggests endocrine disrupting chemicals, including bisphenols and phthalates, may affect behavioral development in children and adolescents. Risk behavior constitutes a potentially sex hormone sensitive behavioral construct. Here, we examined longitudinal associations of phthalate and bisphenol exposure with performance-based tasks and self-reported risk-taking behaviors.MethodsWithin a population-based birth cohort in the Netherlands, urinary bisphenols and phthalate metabolite concentrations were measured in women during pregnancy (three times, n=1379) and in children (once at 6 years, n=775). At 10 years, child risk-taking behavior was assessed with the computerized experimental Columbia Card Task (CCT). At 14 years, adolescents completed a computerized self-assessment of real-life risk-taking behaviors. Linear regression and hurdle models adjusted for confounders were applied in the whole sample and stratified by sex at birth.ResultsIn girls, prenatal mono-isobutyl phthalate was associated with a higher average CCT-score, indicting more risk-taking (B per 10-fold increase in creatinine-adjusted average prenatal levels= 2.10, 95% CI: 0.68,3.51). In boys, higher prenatal and childhood mono-[(2-carboxymethyl)hexyl] phthalate levels were associated with a higher CCT-score (prenatal B=6.21, 95% CI: 1.57,10.47; childhood B=6.02, 95% CI: 1.57,10.47). No associations were observed for bisphenol exposure nor for self-reported risk-taking.ConclusionsPrenatal phthalate exposure was associated with more risk-taking in an experimental task at 10 years-of-age, with sex-specific differences. The task reflects risky decision-making, which may be a hormonally sensitive construct. Risky decision making potentially precedes real-life risk-taking, which was captured by the self-reported measure and was limited in this young sample.

ObjectivePhthalates are recognized endocrine disruptors and emerging neurotoxicants. Di-2-ethylhexyl phthalate (DEHP), one of the most widely produced phthalates, has been linked to adverse neurodevelopmental and neuropsychiatric outcomes with onset in childhood and adolescence. We hypothesized that maternal oxidative stress may play a mechanistic role in DEHP’s neurotoxicity.Materials and MethodsParticipants were drawn from the New York University Children’s Health and Environment Study (NYU CHES). Prenatal DEHP exposure and maternal lipid peroxidation were assessed using repeated creatinine-adjusted maternal urinary measurements across pregnancy. Neonatal BDNF was measured in cord serum (N=337), and internalizing and externalizing problems were assessed in early childhood using the CBCL 1½–5 (N=824). DEHP metabolites were averaged across pregnancy, and cumulative lipid peroxidation biomarkers were estimated using area under the curve (AUC) values from linear mixed-effects spline models. Partial least squares path modeling (PLS-PM) evaluated direct and indirect associations using latent constructs for DEHP exposure, lipid peroxidation, CBCL, and socioeconomic status; other covariates were modeled as single variables. Sex differences were assessed using bootstrapping and sex-stratified models, adjusting for maternal age, parity, pre-pregnancy BMI, cotinine exposure, and socioeconomic status.ResultsPrenatal DEHP exposure was positively associated with maternal lipid peroxidation in all models. Sex-stratified analyses showed that prenatal DEHP exposure was positively associated with childhood CBCL total problems in male children only, but not with neonatal BDNF in either sex. Maternal lipid peroxidation was not associated with neonatal BDNF or CBCL outcomes in either sex.ConclusionPrenatal DEHP exposure is associated with maternal oxidative stress and total behavioral problems in male children only, but oxidative stress does not mediate these relationships. Alternative upstream mechanisms may underlie both maternal oxidative stress and neurobehavioral outcomes. Given the sex-specific findings, future studies should investigate endocrine, metabolic, and epigenetic pathways to clarify DEHP neurotoxicity.

Objectives: To assess the effects of World Trade Center (WTC) exposures, obesity, and smoking on post-bronchodilator (post-BD) lung function in WTC Survivors. Methods: Data included 5,243 participants enrolled in WTC Environmental Health Center (WTC EHC) program between 2005 and 2022. WTC-related exposures included dust-cloud exposure and occupational/residential roles. Lung function included post-BD spirometry (FEV ₁ , FVC) and impulse oscillometry (R ₅ , R ₂₀ , AX). Multivariable linear and quantile regressions assessed associations with WTC exposures, BMI, and smoking, adjusting for demographics. Results: Dust-cloud exposure and Worker status were associated with elevated AX, R ₅ , and R ₂₀ , indicating small airway dysfunction. Spirometry showed minimal impact from dust exposure, though Workers had lower FEV₁ and FVC than Residents. Obesity and smoking were consistently linked to poorer lung function, with effect sizes surpassing WTC exposure. No significant interactions were found between BMI and WTC exposures. Conclusions: WTC exposures are associated with small airway dysfunction, especially in Workers. Obesity and smoking independently worsen lung function, underscoring the importance of both environmental and physiological risk factors in disaster-exposed populations. Post-BD oscillometry adds critical sensitivity in detecting injury to small airways.

Motivation: Large-scale prospective cohort studies collect longitudinal biospecimens alongside time-to-event outcomes to investigate biomarker dynamics in relation to disease risk. The nested case–control (NCC) design provides a cost-effective alternative to full cohort biomarker studies while preserving statistical efficiency. Despite advances in joint modeling for longitudinal and time-to-event outcomes, few approaches address the unique challenges posed by NCC sampling, non-normally distributed biomarkers, and competing survival outcomes. Results: Motivated by the TEDDY study, we propose “JM-NCC”, a joint modeling framework designed for NCC studies with competing events. It integrates a generalized linear mixed-effects model for potentially non-normally distributed biomarkers with a cause-specific hazard model for competing risks. Two estimation methods are developed. fJM-NCC leverages NCC sub-cohort longitudinal biomarker data and full cohort survival and clinical metadata, while wJM-NCC uses only NCC sub-cohort data. Both simulation studies and an application to TEDDY microbiome dataset demonstrate the robustness and efficiency of the proposed methods. Availability and implementation: Software is available at https://github.com/Zhaoyn-oss/JMNCC and archived on Zenodo at https://zenodo.org/records/18199759 (DOI: 10.5281/zenodo.18199759).

An important goal of environmental health research is to assess risks posed by mixtures of environmental exposures. Studies in different fields often group exposures based on their shared biological features. However, such grouping information has not been widely utilized in population-based environmental mixtures analyses due to the lack of appropriate statistical tools. Inspired by data from the National Health and Nutrition Examination Survey (NHANES), we propose a semiparametric multiple-index interaction model (MIIM) to explore the impact of three groups of persistent organic pollutants (POPs) on leukocyte telomere length (LTL). MIIM effectively addresses the challenge of high dimensionality by summarizing exposures into group-level indices, while allowing for nonlinear effects and interactions among exposures through these group indices. This formulation provides interpretable insights into both overall group effects and between-group interactions on the outcome, and allows for identification of key contributors within each group. MIIM can be applied to different types of health outcomes, including continuous, binary, and survival outcomes. We conducted Monte Carlo simulation studies to evaluate the performance of MIIM under various scenarios with high-dimensional and correlated exposure mixtures and illustrated its application to the NHANES data. By bridging biological insights with population-based epidemiological data, MIIM serves as a translational tool to explore the effects of environmental mixtures on health outcomes.

Introduction Secondhand exposure to e-cigarettes represents a potential population health risk given e-cigarette’s prevalence and their unknown health effects, particularly among vulnerable populations such as pregnant people. Objectives To explore metabolomic differences between pregnant people exposed vs. not exposed to secondhand e-cigarette aeresols, to identify possible biomarkers of exposure and metabolic pathways perturbed by e-cigarettes. Methods Exposed participants (n = 19) from the NYU Children’s Health and Environment Study were matched to unexposed participants (n = 57) at a 1:3 ratio on age, hospital of recruitment, and race/ethnicity. Early-pregnancy urine samples were analyzed via an untargeted metabolomics platform using reverse-phase liquid chromatography mass-spectrometry. Feature-exposure associations were estimated using conditional logistic regression to adjust for matching factors. A sensitivity analysis was conducted adjusting for secondhand tobacco exposure. Results Among features enriched in the exposed group were flavonoids and flavor-related compounds including homoeriodictyol and naringenin-7-O-beta-D-glucuronide, 3-acetomidocoumarin, and guaiacol pentosylglucoside; synthetic drugs such as the endocannabinoid AM1172 and the stimulant alpha-PVP; and metabolites associated with lipid metabolism, including 2,4-undecadiene-8,10-diynoic acid isobutylamide, palmitamide, glycerol trihexanoate, and tetradecyl phosphonate. Among features negatively associated with exposure were xanthines. Conclusion This study is the first untargeted metabolomics study investigating metabolomic markers of e-cigarette exposure, including secondhand exposure, in a pregnant cohort. Despite this study’s small size and exploratory nature, the results of this work suggest that flavoring components could be biomarkers for e-cigarette exposure, and that co-exposure to e-cigarettes and other drugs may be prevalent.