Exploration of Different Hypoxia Patterns and Construction of a Hypoxia-Related Gene Prognostic Index in Colorectal Cancer

INTRODUCTION: Immune checkpoint inhibitor (ICI) therapy has been proven to be a highly efficacious treatment for colorectal adenocarcinoma (COAD). However, it is still unclear how to identify those who might benefit the most from ICI therapy. Hypoxia facilitates the progression of the tumor from different aspects, including proliferation, metabolism, angiogenesis, and migration, and improves resistance to ICI. Therefore, it is essential to conduct a comprehensive understanding of the influences of hypoxia in COAD and identify a biomarker for predicting the benefit of ICI. METHODS: An unsupervised consensus clustering algorithm was used to identify distinct hypoxia-related patterns for COAD patients from TCGA and the GEO cohorts. The ssGSEA algorithm was then used to explore the different biological processes, KEGG pathways, and immune characteristics among distinct hypoxia-related clusters. Some hypoxia-related hub genes were then selected by weighted gene coexpression network analysis (WGCNA). Subsequently, univariate Cox regression analysis, multivariate Cox regression analysis, and least absolute shrinkage and selection operator (LASSO) regression were utilized to construct a hypoxia-related gene prognostic index (HRGPI). Finally, validation was also conducted for HRGPI in prognostic value, distinguishing hypoxia-related characteristics and benefits of ICI. RESULTS: We identified four hypoxia-related clusters and found that different hypoxia response patterns induced different prognoses significantly. Again, we found different hypoxia response patterns presented distinct characteristics of biological processes, signaling pathways, and immune features. Severe hypoxia conditions promoted activation of some cancer-related signaling pathways, including Wnt, Notch, ECM-related pathways, and remodeled the tumor microenvironment of COAD, tending to present as an immune-excluded phenotype. Subsequently, we selected nine genes (ANO1, HOXC6, SLC2A4, VIP, CD1A, STC2, OLFM2, ATP6V1B1, HMCN2) to construct our HRGPI, which has shown an excellent prognostic value. Finally, we found that HRGPI has an advantage in distinguishing immune and molecular characteristics of hypoxia response patterns, and it could also be an excellent predictive indicator for clinical response to ICI therapy. CONCLUSION: Different hypoxia response patterns activate different signaling pathways, presenting distinct biological processes and immune features. HRGPI is an independent prognostic factor for COAD patients, and it could also be used as an excellent predictive indicator for clinical response to ICI therapy.