Identification and validation of prognostic genes associated with T-cell exhaustion and macrophage polarization in breast cancer

TL;DR

This study identifies 7 genes linked to T-cell exhaustion and macrophage polarization in breast cancer, offering new insights for potential treatments.

Contribution

A novel risk model using 7 prognostic genes was developed to predict survival and treatment response in breast cancer.

Findings

Seven prognostic genes (PGK1, BTG2, TANK, CFB, EIF4E3, TNFRSF18, BATF) were identified with high survival prediction accuracy.

Significant differences in immune infiltration, immunotherapy markers, and drug sensitivity were found between risk groups.

A lncRNA-miRNA-mRNA regulatory network was constructed, revealing molecular interactions in breast cancer.

Abstract

The most frequent malignant tumor in women is breast cancer (BRCA). It has been discovered that T-cell exhaustion and macrophages play significant roles in BRCA. It was necessary to explore prognostic genes associated with T-cell exhaustion and macrophage polarization in BRCA.

The following data were included: 35 macrophage polarization-related genes (MPRGs), 683 T-cell exhaustion-related genes (TEXRGs), GSE20685, as well as TCGA-BRCA. Initially, candidate genes were identified through crossing differentially expressed genes (DEGs) obtained by differential expression analysis, key module genes associated with MPRGs, as well as TEXRGs. Next, 101 combinations of 10 machine learning algorithms and univariate Cox analysis were utilized to screen for prognostic genes. Concurrently, a risk model was built for validation in TCGA-BRCA and GSE20685. Next, we conducted immune infiltration, immunotherapy, mutation analysis, molecular regulatory network, as well as drug sensitivity between the two risk groups. Ultimately, we did the reverse transcription-quantitative polymerase chain reaction (RT-qPCR).

According to random survival forest (RSF) algorithm (the best combination with the greatest C-index of 0.799), 7 prognostic genes were selected, which are PGK1, BTG2, TANK, CFB, EIF4E3, TNFRSF18, and BATF. After that, we created a risk model, and in the low-risk samples, there was a relatively high survival rate. Next, between two risk parts, the 7 differential immune cells were found. There was a significant difference in 25 immunological checkpoint (ICI) genes between the two risk parts. Next, a lncRNAs-miRNA-mRNA network with 65 nodes and 70 edges was built. Additionally, 84 medications were shown to differ significantly between the two risk groups. Finally, the expression of BTG2, TANK, and EIF4E3 was verified by RT-PCR, which was consistent with the bioinformatics analysis.

The 7 prognostic genes (PGK1, BTG2, TANK, CFB, EIF4E3, TNFRSF18, and BATF) were screened, providing new insights into potential treatments for BRCA.

Figures