Identification of tumor antigens and immunogenic cell death-related subtypes for the improvement of immunotherapy of breast cancer

Abstract

The current immunotherapy strategy for breast cancer is limited. Tumor neoantigens have been proven to be a promising biomarker and potential target of immunotherapy in a variety of tumors. However, their effectiveness for breast cancer remains unclear. Immunogenic cell death (ICD) is a regulated form of cell death that can reshape the tumor immune microenvironment and activate adaptive immune responses. To this end, we screened potential antigens that could be used both for the development of immunotherapy and differentiating the patient-specific immune responses based on ICD-related risk signatures, in order to formulate an accurate scheme for breast cancer immunotherapy. We retrieved the gene expression profiles of the breast invasive cancer cohort and their corresponding clinical control data from The Cancer Genome Atlas. The Gene Expression Profiling Interactive Analysis (GEPIA) database was used to evaluate tumor antigen expression, the cBioPortal program was used to identify genetic variations, and the TIMER website was used to estimate the immune infiltration signatures. The risk score predictive model based on the ICD-related genes was constructed using the least absolute shrinkage and selection operator (LASSO) Cox regression algorithm, and the cohort was divided into low- and high-risk score groups. Two tumor antigens, namely, CCNE1 and PLK1, were associated with poor prognosis and infiltration of antigen-presenting cells. Furthermore, the ICD-related risk signature could significantly predict survival outcomes. The risk groups based on the ICD-related signature predictive model showed diverse immune infiltration and molecular and clinical features. The high-risk group was associated with low immune cell infiltration, immune score, expression of immune checkpoints, and human leukocyte antigen genes but high levels of CCNE1 and PLK1 and poor survival outcome. In conclusion, CCNE1 and PLK1 were identified as potential antigens in breast cancer. The ICD-related prognostic model distinguished immune response heterogeneity and predicted prognosis. Patients with high ICD-related risk scores were suitable to receive combination treatments based on CCNE1 or PLK1 and immune checkpoint inhibitors. In the future, these results will help us develop more accurate treatment schemes for patients with breast cancer.