Development and Validation of a Novel Hypoxia Score for Predicting Prognosis and Immune Microenvironment in Rectal Cancer

Abstract

Hypoxia plays a major role in various tumor types. However, few studies have concentrated on the prognostic model of hypoxia-related genes in rectal cancer and the effect of hypoxia on neutrophil-mediated immunosuppression. We performed Kaplan–Meier analysis, random survival forest analysis, and Cox regression analysis on 342 hypoxia-related genes, constructed hypoxia score in the Gene Expression Omnibus (GEO) cohort, and verified them in the Cancer Genome Atlas (TCGA) cohort. Then the patients were divided into two groups according to the risk level. The overall survival rate of the high-risk (HRisk) group was significantly higher than that of the low-risk (LRisk) group (GEO, p < 0.001; TCGA, p = 0.016). Through receiver operating characteristic and decision curve analysis, the nomogram based on hypoxia score has excellent prediction ability. Functional enrichment analysis showed that hypoxia, metastasis, inflammation, immunity, and other related pathways were enriched. The HRisk group was associated with lower tumor purity, higher immune and stromal score, higher neutrophils, and lower activated memory CD4 + T cells. More importantly, the checkpoint of neutrophil-mediated immunosuppression increased in the HRisk group. In conclusion, a hypoxia score based on 5 hypoxia-related genes can be used to predict the prognosis of rectal cancer and ANLN with a cancer-suppressing effect and SRPX (Sushi Repeat Containing Protein X-Linked) with a cancer-promoting effect may be potential therapeutic targets for rectal cancer.