Construction of a TAN-associated risk score model with integrated multi-omics data analysis and clinical validation in gastric cancer

Aims: An increasing number of studies have highlighted the biological significance of neutrophil activation and polarization in tumor progression. However, the characterization of tumor-associated neutrophils (TANs) is inadequately investigated. Materials and methods: Patients' expression profiles were obtained from TCGA, GEO, and IMvigor210 databases. Six algorithms were used to assess immune cell infiltration. RNA sequencing was conducted to evaluate the differentially expressed genes between induced N1- and N2-like neutrophils. A TAN-associated risk score (TRS) model was established using a combination of weighted gene co-expression network analysis (WGCNA) and RNA-seq data and further assessed in pan-cancer. A clinical cohort of 117 GC patients was enrolled to assess the role of TANs in GC via immunohistochemistry (IHC). Key findings: A TRS signature was built with 10 TAN-related genes (TRGs) and most TRGs were highly abundant in the TANs of the GC microenvironment. The TRS model could accurately predict patients' prognosis, as well as their responses to chemotherapy and immunotherapy. The TRS was positively correlated with pro-tumor immune cells and exhibited negative relationship with anti-tumor immune cells. Additional functional analyses revealed that the signature was positively related to pro-tumor and immunosuppression pathways, such as the hypoxia pathway, across pan-cancer. Furthermore, our clinical cohort demonstrated TANs as an independent prognostic factor for GC patients. Significance: This study constructed and confirmed the value of a novel TRS model for prognostic prediction of GC and pan-cancer. Further evaluation of TRS and TANs will help strengthen the understanding of the tumor microenvironment and guide more effective therapeutic strategies.