Weighted Gene Co-expression Network Analysis Identifies TBC1D10C as A New Prognostic Biomarker for Breast Cancer

Abstract

Background: Immune checkpoint inhibitors are a promising therapeutic strategy for breast cancer (BRCA) patients. The tumor microenvironment (TME) can downregulate the immune response to cancer therapy. Our study aimed to find a TME-related biomarker to identify patients who might respond to immunotherapy. Method: We downloaded raw data from several databases including TCGA and MDACC to identify TME hub genes associated with overall survival (OS) and the progression-free interval (PFI) by WGCNA. Correlations between hub genes and either tumor-infiltrating immune cells or immune checkpoints were conducted by ssGSEA.Result: TME-related green and black modules were selected by WGCNA to further screen hub genes. Random forest, univariate and multivariate Cox regressions were applied to screen hub genes (MYO1G, TBC1D10C, SELPLG, and LRRC15) and construct a nomogram to predict survival of BRCA patients. The C-index for the nomogram was 0.713. A DCA of the predictive model revealed that the net benefit of nomogram was significantly higher than others and the calibration curve demonstrated a good performance by the nomogram. Only TBC1D10C was correlated with both OS and the PFI (both p-values<0.05). TBC1D10C also had a high positive association with tumor-infiltrating immune cells and common immune checkpoints (PD-1, CTLA-4, and TIGIT).Conclusion: We constructed a TME-related gene signature model to predict the survival probability of BRCA patients. We also identified a hub gene, TBC1D10C, which was correlated with both OS and the PFI and had a high positive association with tumor-infiltrating immune cells and common immune checkpoints. TBC1D10C may be a new biomarker to select patients who may benefit from immunotherapy.