Integrative Network Toxicology and Molecular Docking Elucidate the Molecular Mechanisms and Immune Implications of Diethyl Terephthalate in Bladder Cancer

Objective Diethyl terephthalate (DET) is a diester small molecule that may originate as a by-product or a degradation/migration component from polyethylene terephthalate (PET)-related materials during their lifecycle processes such as synthesis, processing, use, and recycling. However, the potential molecular links between DET exposure and the development and progression of bladder cancer (BLCA) remain inadequately elucidated. This study aims to integrate network toxicology and bioinformatics strategies to identify potential DET targets and their intersection with BLCA-related gene networks, further construct a prognostic model to screen key genes, assess correlations with the immune microenvironment, and provide computational evidence at the structural level through molecular docking, thereby offering candidate clues for subsequent mechanistic research and experimental validation. Methods Potential DET targets were predicted by inputting its SMILES (CCOC(= O)C1 = CC = C(C = C1)C(= O)OCC) into the SwissTargetPrediction database; BLCA-related gene sets were retrieved from OMIM, TTD, and GeneCards databases and intersected with DET targets to obtain candidate common targets, which were then visualized using Cytoscape; GO and KEGG enrichment analyses were performed with significance determined by a Benjamini–Hochberg corrected FDR < 0.05. Using expression and follow-up data from the TCGA-BLCA cohort, univariate Cox, LASSO, and multivariate Cox regressions were employed to screen prognostic key genes and construct a risk score model. Model performance was evaluated using Kaplan–Meier survival analysis and time-dependent ROC analysis, and a nomogram along with calibration curves were built by integrating clinical variables to assess predictive consistency. CIBERSORT was used to estimate the infiltration proportions of 22 immune cell types and analyze the correlation between key gene expression and immune infiltration. Finally, receptor/ligand structures were obtained from PDB and PubChem, and molecular docking was performed using AutoDock Vina with PyMOL for visualizing interaction patterns. Results A total of 115 potential DET targets were predicted; intersection with BLCA-related gene sets yielded 86 candidate common targets. Enrichment analysis showed that common targets were significantly enriched in molecular function terms such as serine/threonine kinase activity and phosphatidylinositol kinase activity, and in tumor-related signaling pathways such as cAMP and ErbB pathways. Enrichment in terms like "Chemical carcinogenesis - ROS" suggested potential associations with oxidative stress and chemical carcinogenesis processes. Prognostic modeling identified six key genes (NQO1, AKR1B1, ADK, GAK, PDE5A, MAPK10). The constructed risk score model enabled high/low-risk stratification (Kaplan–Meier: P < 0.001) and demonstrated moderate discriminative ability for 1-, 3-, and 5-year survival (AUC = 0.675, 0.658, 0.651). Immune infiltration analysis indicated significant correlations between the expression of key genes and the infiltration proportions of CD8⁺ T cells, dendritic cells, mast cells, NK cells, and macrophage polarization-related subsets. Molecular docking showed that DET could achieve binding conformations with binding energies lower than − 5 kcal/mol for all six key target proteins, with the lowest binding energy observed for AKR1B1 (− 7.9 kcal/mol). Conclusion This study proposes, at a systems level, a potential association between the DET-related target network and key BLCA signaling regulation, prognostic risk stratification, and immune microenvironment characteristics, providing a candidate framework of key genes and pathways. Given that the analysis is primarily based on public cohorts and computational predictions, further assessment of causal likelihood and translational potential requires integration with population exposure measurements, in vitro/in vivo functional experiments, and external validation using independent cohorts.