Helicobacter pylori Infection-Related Long Non-Coding RNA Signatures Predict the Prognostic Status for Gastric Cancer Patients

Background Helicobacter pylori (H. pylori) is a type I biological carcinogen, which may cause about 75% of the total incidence of gastric cancer worldwide. H. pylori infection can induce and activate the cancer-promoting signaling pathway and affect the occurrence and outcome of gastric cancer through controlling the regulatory functions of long non-coding RNAs (lncRNAs). However, we have no understanding of the prognostic worth of lncRNAs for gastric cancer patients infected with H. pylori. Method We screened differentially expressed lncRNAs using DESeq2 method among TCGA database. And we built the H. pylori infection-related lncRNAs regulatory patterns. Then, we constructed H. pylori infection-based lncRNAs prognostic signatures for gastric cancer patients together with H. pylori infection, via uni-variable and multi-variable COX regression analyses. Based on receiver operator characteristic curve (ROC) analysis, we evaluated the prediction effectiveness for this model. Results We identified 115 H. pylori infection-related genes were differentially expressed among H. pylori-infected gastric cancer tissues versus gastric cancer tissues. Functional enrichment analysis implies that H. pylori infection might interfere with the immune-related pathways among gastric cancer tissues. Then, we built H. pylori infection-related dys-regulated lncRNA regulatory networks. We also identified 13 differentially expressed lncRNAs were associated with prognosis for gastric cancer patients together with H. pylori infection. Kaplan-Meier analysis demonstrated that the lncRNA signatures were correlated with the poor prognosis. What is more, the AUC of the lncRNA signatures was 0.712. Also, this prognostic prediction model was superior to the traditional clinical characters. Conclusion We successfully constructed a H. pylori-related lncRNA risk signature and nomogram associated with H. pylori-infected gastric cancer patients prognosis, and the signature and nomogram can predict the prognosis of these patients.