Copper and copper-binding proteins are key components of tumor progression as they play important roles in tumor invasion and migration, but their associations in gliomas remain unclear. Transcriptomic datasets of glioblastoma, low-grade glioma, and normal brain cortex were derived from the TCGA and GTEX databases. Differentially expressed genes (DEGs) of copper-binding proteins were screened and used to construct a prognostic model based on COX and LASSO regression, which was further validated by the CGGA datasets. The expressions of risk-model genes were selectively confirmed via anatomic feature-based expression analysis and immunohistochemistry. The risk score was stratified by age, gender, WHO grade, IDH1 mutation, MGMT promoter methylation, and 1p/19q codeletion status, and a nomogram was constructed and validated. A total of 21 DEGs of copper-binding proteins were identified and a six-gene risk-score model was constructed, consisting of ANG, F5, IL1A, LOXL1, LOXL2, and STEAP3, which accurately predicted 1-, 3-, and 5-year overall survival rates, with the AUC values of 0.87, 0.88, and 0.82, respectively. The high-risk group had a significantly shorter OS (p < 0.0001) and was associated with old age, wild-type IDH1, a high WHO grade, an unmethylated MGMT promoter, and 1p/19q non-codeletion and had higher levels of immune cell infiltration, cancer-immunity suppressor, and immune checkpoint gene expression as well as a higher TMB. The model based on the genes of copper-binding proteins could contribute to prognosis prediction and provide potential targets against gliomas.
Read full abstract