Construction and validation of 3-genes hypoxia-related prognostic signature to predict the prognosis and therapeutic response of hepatocellular carcinoma patients

Abstract

Background Previous studies have shown that the hypoxia microenvironment significantly impacted tumor progression. However, the clinical prognostic value of hypoxia-related risk signatures and their effects on the tumor microenvironment (TME) in hepatocellular carcinoma (HCC) remains hazy. This study aimed to conduct novel hypoxia-related prognostic signatures and improve HCC prognosis and treatment. Methods Differentially expressed hypoxia-related genes (HGs) were identified with the gene set enrichment analysis (GSEA). Univariate Cox regression was utilized to generate the tumor hypoxia-related prognostic signature, which consists of 3 HGs, based on the least absolute shrinkage and selection operator (LASSO) algorithm. Then the risk score for each patient was performed. The prognostic signature’s independent prognostic usefulness was confirmed, and systematic analyses were done on the relationships between the prognostic signature and immune cell infiltration, somatic cell mutation, medication sensitivity, and putative immunological checkpoints. Results A prognostic risk model of four HGs (FDPS, SRM, and NDRG1) was constructed and validated in the training, testing, and validation datasets. To determine the model’s performance in patients with HCC, Kaplan–Meier curves and time-dependent receiver operating characteristic (ROC) curves analysis was implemented. According to immune infiltration analysis, the high-risk group had a significant infiltration of CD4+ T cells, M0 macrophages, and dendritic cells (DCs) than those of the low-risk subtype. In addition, the presence of TP53 mutations in the high-risk group was higher, in which LY317615, PF−562271, Pyrimethamine, and Sunitinib were more sensitive. The CD86, LAIR1, and LGALS9 expression were upregulated in the high-risk subtype. Conclusions The hypoxia-related risk signature is a reliable predictive model for better clinical management of HCC patients and offers clinicians a holistic viewpoint when determining the diagnosis and course of HCC treatment.