Identification of a novel immune-related gene signature by single-cell and bulk sequencing for the prediction of the immune landscape and prognosis of breast cancer

Abstract

Background: As a common cause of cancer-related deaths in women, BRCA (breast cancer) shows complexity and requires precise biomarkers and treatment methods. This study delves into the molecular makeup of BRCA, focusing on immune profiles, molecular subtypes, gene expression and single-cell analysis.

Methods: XCell was used to assess immune infiltration based on TCGA (the Cancer Genome Atlas) data and the clustering analysis was made. Differentially expressed genes were examined in distinct clusters, and the WGCNA (weighted correlation network analysis) was made to establish co-expression networks. The prognostic models were developed by Cox and LASSO-Cox regression. The clustering analysis, GSEA (Gene set enrichment analysis), GSVA (gene set variation analysis) and communication analysis of the single-cell dataset GSE161529 were performed to investigate the functional relevance. Real-time polymerase chain reaction (RT-PCR) was employed for evaluating gene expression.

Results: The results revealed significant differences in immune cell infiltration between two clusters (C1 and C2). C2 had poorer survival outcomes, which was associated with higher expression of immune checkpoints PD1 and PD-L1. The gene modules identified via WGCNA were correlated with the immune-based subtypes. Then, a prognostic model comprising seven genes (ACSL1, ABCB5, XG, ADH4, OPN4, NPR3, NLGN1) was used to divide patients into high- and low-risk subgroups. The high-risk group had worse prognosis and higher scores of TIDE (Tumor Immune Dysfunction and Exclusion). The single-cell analysis depicted the immune landscape. Macrophages and endothelial cells exhibited higher AUCell scores. In cellular communication analysis, notably significant ligand-receptor interactions of HLA-DRA-> CD4 and TNFSF13B-> HLA-DPB1 were observed. The proportion of endothelial cells was correlated with risk scores. Finally, RT-PCR results illustrated the expression of seven genes in BRCA specimens.

Conclusion: The integrative analysis provides new insights into molecular complexities of BRCA. Immune profiles and gene signatures hold potential for improving stratification of BRCA patients and guiding the development of personalized immunotherapy strategies.

Keywords: Bioinformatics; Breast cancer; Immune infiltration; Molecular subtyping; Prognostic model; Single-cell analysis.