Construction of a Prognostic Model for Hypoxia-Related LncRNAs and Prediction of the Immune Landscape in the Digestive System Pan-Cancer

Abstract

Digestive system pan-cancer is a general term for digestive system tumors including colorectal carcinoma (CRC), esophageal carcinoma (ESCA), stomach adenocarcinoma (STAD), and liver hepatocellular carcinoma (LIHC). Since the anatomical location, function and metabolism are closely related, there may be similarities in development and progression of these tumors. Hypoxia is the consequence of an imbalance between oxygen demand and supply, and intracellular hypoxia is associated with malignant progression, treatment resistance, and poor prognosis in tumors. Therefore, an urgent and challenging task is to investigate the molecular mechanisms associated with hypoxia in digestive system pan-cancer for the prognosis and treatment of digestive tract tumors. In this study, we identified 18 hypoxia-related lncRNAs (HRlncRNAs) by co-expression analysis between hypoxia genes and lncRNAs from digestive system pan-cancer. Six HRlncRNAs were then obtained using lasso regression and multivariate cox analysis to construct a prognostic model. Next, the Akaike information criterion (AIC) values for 3-year receiver operating curve (ROC) were counted to determine the cut-off point and establish an optimal model to distinguish between high- or low-risk groups among patients with digestive system pan-cancer. To evaluate the stability of the prognosis model, we validated it in terms of survival outcomes, clinicopathological stage, tumor-infiltrating immune cells, immune checkpoint inhibitors (ICIs) and anticancer drugs sensitivity. The results suggested that high- risk group had a worse prognosis and a more positive association with tumor-infiltrating immune cells such as B cells, cancer-associated fibroblasts, endothelial cells, monocytes, macrophages and bone marrow dendritic cells in digestive system pan-cancer. Immune checkpoint inhibitors (ICIs) related biomarkers discovered that high-risk group was positively correlated with high expression of HAVCR2 in digestive system pan-cancer. The anticancer drugs sensitivity analysis showed that the high-risk group was associated with the lower half-inhibitory centration (IC50) of Imatinib in digestive system pan-cancer. In conclusion, the prognostic model of HRlncRNAs showed a promising clinical prediction value and may provide a useful reference for the diagnosis and treatment of the digestive system tumors.