Background: Sepsis, a life-threatening condition resulting from a dysregulated host response to infection, frequently progresses to acute lung injury (ALI), a severe complication associated with high mortality due to the lack of reliable early-stage biomarkers. Neutrophil extracellular traps (NETs) and the associated inflammatory cascade aggravate disease progression via positive feedback mechanisms. This study aimed to investigate the heterogeneity of NETs- and inflammation-related genes (NIRGs) in sepsis-induced ALI (siALI), construct a diagnostic model, and characterize the immune microenvironment (IME) and its association with immune cell infiltration (ICI).

Methods: Publicly available transcriptomic datasets from the Gene Expression Omnibus (GEO) (GSE66890, GSE32707, GSE10474), GeneCards and relevant literature were used to identify differentially expressed Neutrophil extracellular traps and Inflammatory-Related Differentially Expressed Genes (NIRDEGs), which were subsequently validated. Perturbed biological processes and immune-cell dynamics were examined via Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG), and Gene Set Enrichment Analysis (GSEA), as well as immune cell deconvolution using CIBERSORT. Multiple machine learning algorithms were employed to construct a diagnosis model, which was evaluated using receiver operating characteristic (ROC) curve analysis, calibration plots, decision curve analysis (DCA), and external validation. Regulatory networks (mRNA-transcription factor, mRNA-miRNA, and mRNA-drug) were constructed using ChIPBase, StarBase 3.0, and the Comparative Toxicogenomics Database (CTD).

Results: Eight genes, S100 Calcium Binding Protein A12 (S100A12), Proteinase 3 (PRTN3), Toll-like receptor 2 (TLR2), triggering receptors expressed on myeloid cells-1 (TREM1), Serum and glucocorticoid-induced kinanse-1 (SGK1), Phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit gamma (PIK3CG), Fibrinogen-like protein 2 (FGL2), and Toll-like receptor 8 (TLR8), were identified as candidate diagnostic biomarkers and therapeutic targets for siALI. High-risk patients showed enrichment in thrombopoietin (TPO), receptor for advanced glycation end-products (RAGE), Toll-like receptor (TLR), and NOD-like receptor (NLR) signaling pathways. Distinct patterns of ICI effectively distinguished sepsis from siALI, and the identified genes were related to immune regulation.

Conclusions: This study elucidates the molecular architecture and immune microenvironment of siALI, offering a robust foundation for early diagnosis and personalized therapeutic strategies.