Abstract
BACKGROUND: Non-small cell lung cancer (NSCLC) accounts for a great number of all lung cancer cases. Hypoxia, one of the hallmarks in solid cancer, is closely involved in cancer cell progression and migration. This study aimed to develop a molecular subtyping system based on hypoxia-related genes and construct a prognostic model for NSCLC patients. METHODS: Unsupervised consensus clustering was used to classify molecular subtypes. Mutation and immune analyses were conducted to compare differences among the molecular subtypes. Univariate Cox regression, least absolute shrinkage and selection operator (LASSO) analysis, and step Akaike information criterion (stepAIC) were performed to screen prognostic genes. RESULTS: Two molecular subtypes (C1 and C2) were identified based on hypoxia-related genes and showed significant differences in survival, enriched pathways, tumor microenvironment (TME), and sensitivity to immunotherapy and chemotherapy. Interestingly, C1 subtype had better survival and response to targeted therapies. Oncogenic pathways, such as hypoxia, epithelial mesenchymal transition (EMT), NOTCH signaling, and p53 signaling pathways were more enriched in C2 subtype. A 6-gene prognostic model with robust ability was developed to classify NSCLC patients into high-risk and low-risk groups. CONCLUSION: The novel molecular subtypes could assist personalized therapies to select suitable patients. The six prognostic genes may be novel targets for further understanding mechanisms of NSCLC development associated with hypoxia and exploiting novel targeted therapies.