Comprehensive analysis to construct a novel immune-related prognostic panel in aging-related gastric cancer based on the lncRNA-miRNA-mRNA ceRNA network

Introduction: Gastric cancer (GC) is the fifth frequent malignancy and is responsible for the third leading cause of cancer-related deaths. Gastric cancer is an aging-related disease, with incidence and mortality rates increasing with aging. The development of GC is affected by lncRNAs, miRNAs, and mRNAs at the transcriptional and posttranscriptional levels. This study aimed to establish a prognostic panel for GC based on competing endogenous RNA (ceRNA) networks. Methods: RNA sequences were obtained from the TCGA database. Different expressions of RNAs were scrutinized with the EdgeR package. The ceRNA network was built using the starBase database and the Cytoscape. The prognostic panel was constituted with the LASSO algorithm. We developed a nomogram comprising clinical characteristic and risk score. The receiver operating characteristic (ROC) was used to evaluate the accuracy of the nomogram prediction. Hub RNAs expressions were detected by qPCR, immunohistochemistry and western blot respectively. Clinical relevance and survival analyses were analyzed. The relationship between RNAs and immune infiltrations, as well as immune checkpoints, was analyzed and evaluated using the CIBERSORT, TIMER and TISIDB databases. Results: Four DElncRNAs, 21 DEmiRNAs and 45 DEmRNAs were included in the ceRNA network. A 3-element panel (comprising lncRNA PVT1, hsa-miR-130a-3p and RECK) with poor overall survival (OS) was established and qPCR was applied to validate the expressions of hub RNAs. Hub RNAs were firmly associated with T, M, and N stage. The CIBERSORT database showed that the high lassoScore group exhibited a significantly high ratio of resting memory CD4(+) T cells, M2 macrophages and a significantly low ratio of activated memory CD4(+) T cells and M1 macrophages. According to the TIMER database, this panel was linked to immune infiltrations and immune cell gene markers. TISIDB database indicated that RECK was positively correlated with immune checkpoints (including CD160, CD244, PDCD1, and TGFBR1). Discussion: A novel triple prognostic panel of GC constructed based on the ceRNA network was associated with clinical prognostic, clinicopathological features, immune infiltrations, immune checkpoints and immune gene markers. This panel might provide potential therapeutic targets for GC and more experimental verification research is needed.