Identification of an Immune-Related LncRNA Signature in Gastric Cancer to Predict Survival and Response to Immune Checkpoint Inhibitors

Immune microenvironment in gastric cancer is closely associated with patient's prognosis. Long non-coding RNAs (lncRNAs) are emerging as key regulators of immune responses. In this study, we aimed to construct a prognostic model based on immune-related lncRNAs (IRLs) to predict the overall survival and response to immune checkpoint inhibitors (ICIs) of gastric cancer (GC) patients. The IRL signature was constructed through a bioinformatics method, and its predictive capability was validated. A stratification analysis indicates that the IRL signature can distinguish different risk patients. A nomogram based on the IRL and other clinical variables efficiently predicted the overall survival of GC patients. The landscape of tumor microenvironment and mutation status partially explain this signature's predictive capability. We found the level of cancer-associated fibroblasts, endothelial cells, M2 macrophages, and stroma cells was high in the high-risk group, while the number of CD8(+) T cells and T follicular helper cells was high in the low-risk group. Immunophenoscore (IPS) is validated for ICI response, and the IRL signature low-risk group received higher IPS, representing a more immunogenic phenotype that was more inclined to respond to ICIs. In addition, we found RNF144A-AS1 was highly expressed in GC patients and promoted the proliferation, migration, and invasive capacity of GC cells. We concluded that the IRL signature represents a novel useful model for evaluating GC survival outcomes and could be implemented to optimize the selection of patients to receive ICI treatment.

Automated summary

This study constructed a prognostic model based on immune-related lncRNAs (IRLs) for predicting overall survival and response to immune checkpoint inhibitors (ICIs) in gastric cancer patients. The IRL signature could distinguish different risk patients and a nomogram efficiently predicted overall survival. Tumor microenvironment and mutation status partially explained the predictive capability of the IRL signature.