Identification of 4 immune cells and a 5-lncRNA risk signature with prognosis for early-stage lung adenocarcinoma

Abstract

Abstract Background Lung cancer is the most common cancer and cause of cancer‐related mortality worldwide, increasing evidence indicated that there was a significant correlation between tumors and the long non‐coding RNAs (lncRNAs), as well as tumor immune infiltration, but their role in early lung adenocarcinoma (LUAD) are still unclear. Methods Gene expression data and corresponding clinical data of early-stage LUAD patients were downloaded from GEO and TCGA databases. 24 kinds of tumor-infiltrating immune cells were analyzed by quantity analysis and univariate cox regression analysis, we divided patients into two subgroups using consensus clustering, recognized the differentially expressed genes (DEGs) in the subgroups, then, established lncRNA risk signature by least absolute shrinkage and selection operator (LASSO) regression. Results A total of 718 patients were enrolled in this study, including 246 from GSE31210 dataset, 127 from GSE50081 dataset and 345 from TCGA-LUAD. We identified that Th2 cells, TFH, NK CD56dim cells and Mast cells were prognosis-related(p < 0.05), then established a 5-lncRNA risk signature (risk score = 0.374600616* LINC00857 + 0.173825706* LINC01116 + (− 0.021398903)* DRAIC + (− 0.113658256)* LINC01140 + (− 0.008403702)* XIST), and draw a nomogram showed that the signature had a well prediction accuracy and discrimination. Conclusions We identified 4 immune infiltrating cells related to the prognosis of early-stage LUAD, and established a novel 5 immune-related lncRNA signature for predicting patients’ prognosis.