A Novel Neutrophil Extracellular Traps Signature for Overall Survival Prediction and Tumor Microenvironment Identification in Gastric Cancer.

Abstract

Neutrophil extracellular traps (NETs) released by neutrophils are crucial for cancer development, metastasis, and can indicate gastric cancer (GC) patients' prognosis. This study reveals the relevance of NETs-related genes to GC through transcriptome analysis. We obtained transcriptome sequencing data of GC from UCSC Xena and screened prognostic NETs-related genes by GEPIA2 database. The signature for NETs was subsequently created using the LASSO-Cox regression. The clinical value of model was further explored using the nomogram and was externally validated by the GEO database. After that, we employed GO, KEGG, and GSEA enrichment analyses to evaluate the bio-functional enrichment and related pathways. Additionally, ESTIMATE, MCP counter, and ssGSEA scores were used to investigate the immunological microenvironment of GC patients. Finally, in the external cohort, neutrophil elastase (NE)-DNA complexes were measured by ELISA, and the prognostic value of NE-DNA in GC was investigated using Cox analysis. Seven NETs-associated genes (PDE4B, CD93, CTSG, IL6, ELANE, KCNJ15, and CRISPLD2) were filtered to establish the signature and participated in building the nomogram. In comparison to the high-risk group, the overall survival (OS) was much longer in the low-risk group (P=0.005). The validation cohort demonstrated the acceptable predictive ability of the nomogram. The signature was enriched in biological features such as extracellular matrix organization, epithelial-mesenchymal transition and inflammatory response. Moreover, there were substantial differences in immune cell infiltration across the different risk groups (p<0.001), especially the high-risk group having more immune cells that are engaged in the antigen presentation process and associated functions. Finally, in the external cohort, NE-DNA levels were shown to be an independent factor affecting OS prognosis (p=0.006). Overall, this research identified a novel signature based on seven NETs-associated genes to predict prognosis and identify tumor microenvironment of GC. And high NE-DNA level may be a critical factor in the poor OS associated with NETs.