Abstract
Genetic variants such as copy number variation (CNV), microsatellite instability (MSI), and tumor mutation burden (TMB) have been reported to associate with the immune microenvironment and prognosis of patients with breast cancer. In this study, we performed an integrated analysis of CNV, MSI, and TMB data obtained from The Cancer Genome Atlas, thereby generating two genetic variants-related subgroups. We characterized the differences between the two subgroups in terms of prognosis, MSI burden, TMB, CNV, mutation landscape, and immune landscape. We found that cluster 2 was marked by a worse prognosis and lower TMB. According to these groupings, we identified 130 differentially expressed genes, which were subjected to univariate and least absolute shrinkage and selection operator-penalized multivariate modeling. Consequently, we constructed an 11-gene signature risk model called the genomic variation-related prognostic risk model (GVRM). Using ROC analysis and a calibration plot, we estimated the prognostic prediction of this GVRM. We confirmed the predictive efficiency of this GVRM by validating it in another independent International Cancer Genome Consortium cohort. Our results conclude that an 11-gene signature developed by integrated analysis of CNV, MSI, and TMB has a high potential to predict breast cancer prognosis, which provided a strong rationale for further investigating molecular mechanisms and guiding clinical decision-making in breast cancer.