Unraveling the immunogenic cell death pathways in gastric adenocarcinoma: A multi-omics study.

ABSTRACT

BACKGROUND: Gastric cancer (GC) is a prevalent malignant tumor of the gastrointestinal (GI) system. However, the lack of reliable biomarkers has made its diagnosis, prognosis, and treatment challenging. Immunogenic cell death (ICD) is a type of programmed cell death that is strongly related to the immune system. However, its function in GC requires further investigation. METHOD: We used multi-omics and multi-angle approaches to comprehensively explore the prognostic features of ICD in patients with stomach adenocarcinoma (STAD). At the single-cell level, we screened genes associated with ICD at the transcriptome level, selected prognostic genes related to ICD using weighted gene co-expression network analysis (WGCNA) and machine learning, and constructed a prognostic model. In addition, we constructed nomograms that incorporated pertinent clinical features and provided effective tools for prognostic prediction in clinical settings. We also investigated the sensitivity of the risk subgroups to both immunotherapy and drugs. Finally, in addition to quantitative real-time polymerase chain reaction, immunofluorescence was used to validate the expression of ICD-linked genes. RESULTS: Based on single-cell and transcriptome WGCNA analyses, we identified 34 ICD-related genes, of which 11 were related to prognosis. We established a prognostic model using the least absolute shrinkage and selection operator (LASSO) algorithm and identified dissimilarities in overall survival (OS) and progression-free survival (PFS) in risk subgroups. The nomograms associated with the ICD-related signature (ICDRS) demonstrated a good predictive value for clinical applications. Moreover, we detected changes in the tumor microenvironment (TME), including biological functions, mutation landscapes, and immune cell infiltration, between the high- and low-risk groups. CONCLUSION: We constructed an ICD-related prognostic model that incorporated features related to cell death. This model can serve as a useful tool for predicting the prognosis of GC, targeted prevention, and personalized medicine.