Efferocytosis-Related LncRNA Signatures for Predicting Prognosis and Immune Landscape of Gastric Cancer

Abstract

Abstract Background Gastric cancer (GC) is a highly malignant form of cancer with a high level of morbidity and mortality. The detection of biomarkers is useful, but still need more evaluation criteria to guide the diagnosis and treatment in clinical practice. This study focuses on identifying long non-coding RNAs(lncRNAs) and find the relation with efferocytosis to predict prognosis and target drug. Methods We got the transcriptomic data and clinical data from TCGA platform. Risk model were identified by the least absolute shrinkage and selection operator (LASSO). Then samples were randomly divided into two groups, including training groups and test group. We analyzed the different groups with complete data on lncRNA expression and clinical information, building a risk model and verifying its feasibility. Then we conducted prognostic, pathway, and immune analyses of the risk model. We also looked at drug sensitivity to the risk model, and explored the potential function of the model. Results We constructed a risk model containing five efferocytosis-related lncRNAs (ERLs) signatures (LINC01614, AC016717.2, AC068790.7, SCAT1, and PVT1). Then the feasibility of the risk model is verified. We conducted a Cox regression analysis and constructed ROC curves to evaluate the predictive performance of risk features for overall survival (OS) in GC patients, as well as other clinical and pathological characteristics. The risk model had the highest AUC value among these characteristics. We developed a nomogram based on risk scores and clinical characteristics including age, grade, gender, and stage. We used a calibration plot to demonstrate good consistency between the nomogram and the predicted 1-year, 3-year, and 5-year survival rates. These results suggested that the risk model feature is valuable for GC patients. We then evaluated the immune response between different groups, demonstrated that individuals with a high risk score tended to have a higher state of immune infiltration. We also conducted sensitivity screening for guideline drugs, and eight drugs showed significant differences. Conclusion The 5-ERLs signatures is useful for predicting prognosis of OS, forecasting the immune response and improving treatment modalities for further clinical application in GC.