Cuproptosis/ferroptosis-related gene signature is correlated with immune infiltration and predict the prognosis for patients with breast cancer
Background: Breast invasive carcinoma (BRCA) is a malignant tumor with high morbidity and mortality, and the prognosis is still unsatisfactory. Both ferroptosis and cuproptosis are apoptosis-independent cell deaths caused by the imbalance of corresponding metal components in cells and can affect the proliferation rate of cancer cells. The aim in this study was to develop a prognostic model of cuproptosis/ferroptosis-related genes (CFRGs) to predict survival in BRCA patients. Methods: Transcriptomic and clinical data for breast cancer patients were obtained from The Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) databases. Cuproptosis and ferroptosis scores were determined for the BRCA samples from the TCGA cohort using Gene Set Variation Analysis (GSVA), followed by weighted gene coexpression network analysis (WGCNA) to screen out the CFRGs. The intersection of the differentially expressed genes grouped by high and low was determined using X-tile. Univariate Cox regression and least absolute shrinkage and selection operator (LASSO) were used in the TGCA cohort to identify the CFRG-related signature. In addition, the relationship between risk scores and immune infiltration levels was investigated using various algorithms, and model genes were analyzed in terms of single-cell sequencing. Finally, the expression of the signature genes was validated with quantitative real-time PCR (qRT‒PCR) and immunohistochemistry (IHC). Results: A total of 5 CFRGs (ANKRD52, HOXC10, KNOP1, SGPP1, TRIM45) were identified and were used to construct proportional hazards regression models. The high-risk groups in the training and validation sets had significantly worse survival rates. Tumor mutational burden (TMB) was positively correlated with the risk score. Conversely, Tumor Immune Dysfunction and Exclusion (TIDE) and tumor purity were inversely associated with risk scores. In addition, the infiltration degree of antitumor immune cells and the expression of immune checkpoints were lower in the high-risk group. In addition, risk scores and mTOR, Hif-1, ErbB, MAPK, PI3K/AKT, TGF-β and other pathway signals were correlated with progression. Conclusion: We can accurately predict the survival of patients through the constructed CFRG-related prognostic model. In addition, we can also predict patient immunotherapy and immune cell infiltration.