Neutrophil Extracellularr Trap-Associated Genes Define Molecular Subtypes and Immune Microenvironment in Pancreatic Ductal Adenocarcinoma

Background: Pancreatic ductal adenocarcinoma (PDAC) is an aggressive cancer with poor progno-sis. Neutrophil extracellular traps (NETs) influence the tumor microenvironment and promote PDAC progression, yet their diagnostic relevance and roles in molecular subtyping remain insufficiently defined. Methods: Six GEO transcriptomic datasets were integrated, normalized, and batch-corrected to identify differentially expressed genes (DEGs). Candidate NETs-related genes were obtained by intersecting DEGs, key WGCNA module genes, and NETs genes curated from GeneCards and the literature. A diagnostic model was built using 113 combinations of machine-learning algorithms and validated in independent GEO datasets and the TCGA-PAAD cohort. Molecular subtypes were identified with ConsensusClusterPlus and characterized using ESTIMATE, CIBERSORT, TIDE, and immune checkpoint analyses. Single-cell RNA-seq data from 15 PDAC samples were used to explore gene localization and functional pathways. Results: We identified 749 DEGs and one PDAC-associated WGCNA module, leading to 22 can-didate genes and eight core NETs-related genes (CARD11, CXCL10, IL1A, ITGA2, LAMC2, MYC, SLC2A1, XDH). The Lasso+glmBoost model showed strong diagnostic accuracy across all datasets (AUC 0.919–0.998). Two molecular subtypes (C1 and C2) were defined: C2 exhibited stronger immune infiltration and better survival, while C1 showed elevated TIDE scores, more regulatory T cells, and enhanced immune evasion. Single-cell analyses revealed predominant expression of NETs core genes in ductal cells, enriched in epithelial–mesenchymal transition (EMT) and pro-tumorigenic pathways. Conclusions: This study identifies eight NETs-related diagnostic genes and proposes a robust diagnostic model and molecular subtyping system that capture PDAC heterogeneity and provide insights for precision immunotherapy.