Construction and verification of a hypoxia-related nine-gene prognostic model in uveal melanoma based on integrated single-cell and bulk RNA sequencing analyses

Abstract

Uveal melanoma (UM) is the most common primary intraocular tumor with high metastasis and poor prognosis among adults. Hypoxia participates in the metastasis process in various types of cancers. It is reported that the increased expression of hypoxia inducible factor 1 alpha subunit (HIF1A), a hypoxia-related molecule, is associated with worse prognoses of UM patients. Based on the integrated analysis of single-cell sequencing (scRNA-seq) dataset from Gene Expression Omnibus (GEO) and bulk RNA-seq dataset from the Cancer Genome Atlas (TCGA), we found hypoxia was the key feature in UM progression and identified 47 common hypoxia-related differentially expressed genes (DEGs) for the following research. Univariate cox analysis and LASSO-Cox regression analysis were performed to establish a nine-gene prognostic model. According to this model, UM patients could be divided into high- and low-risk groups, with a significant difference in overall survival and progression free survival between the two groups (P < 0.001). The accuracy of the predictive model was also verified on two other independent datasets. In addition, Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analyses revealed that these hypoxia-related DEGs were enriched in immune and cancer related pathways. The proportion of immune infiltration and the expression of immune biomarkers were different between high- and low-risk UM patients, providing potential targets for UM immunotherapy. Hence, our hypoxia-related nine-gene model could efficiently predict the prognosis and guide personalized therapies for UM patients.