Dataset Information

Identification of prognostic biomarkers for breast cancer brain metastases based on the bioinformatics analysis.

ABSTRACT:

Purpose

The prognosis of breast cancer (BC) patients who develop into brain metastases (BMs) is very poor. Thus, it is of great significance to explore the etiology of BMs in BC and identify the key genes involved in this process to improve the survival of BC patients with BMs.

Patients and methods

The gene expression data and the clinical information of BC patients were downloaded from TCGA and GEO database. Differentially expressed genes (DEGs) in TCGA-BRCA and GSE12276 were overlapped to find differentially expressed metastatic genes (DEMGs). The protein-protein interaction (PPI) network of DEMGs was constructed via STRING database. ClusterProfiler R package was applied to perform the gene ontology (GO) enrichment analysis of DEMGs. The univariate Cox regression analysis and the Kaplan-Meier (K-M) curves were plotted to screen DEMGs associated with the overall survival and the metastatic recurrence survival, which were identified as the key genes associated with the BMs in BC. The immune infiltration and the expressions of immune checkpoints for BC patients with brain relapses and BC patients with other relapses were analyzed respectively. The correlations among the expressions of key genes and the differently infiltrated immune cells or the differentially expressed immune checkpoints were calculated. The gene set enrichment analysis (GSEA) of each key gene was conducted to investigate the potential mechanisms of key genes involved in BC patients with BMs. Moreover, CTD database was used to predict the drug-gene interaction network of key genes.

Results

A total of 154 DEGs were identified in BC patients at M0 and M1 in TCGA database. A total of 667 DEGs were identified in BC patients with brain relapses and with other relapses. By overlapping these DEGs, 17 DEMGs were identified, which were enriched in the cell proliferation related biological processes and the immune related molecular functions. The univariate Cox regression analysis and the Kaplan-Meier curves revealed that CXCL9 and GPR171 were closely associated with the overall survival and the metastatic recurrence survival and were identified as key genes associated with BMs in BC. The analyses of immune infiltration and immune checkpoint expressions showed that there was a significant difference of the immune microenvironment between brain relapses and other relapses in BC. GSEA indicated that CXCL9 and GPR171 may regulate BMs in BC via the immune-related pathways.

Conclusion

Our study identified the key genes associated with BMs in BC patients and explore the underlying mechanisms involved in the etiology of BMs in BC. These findings may provide a promising approach for the treatments of BC patients with BMs.

SUBMITTER: Wu Z

PROVIDER: S-EPMC8760394 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:In the realm of clinical practice, there is currently an insufficiency of distinct biomarkers available for the detection of breast cancer. It is of utmost importance to promptly employ bioinformatics methodologies to investigate prospective biomarkers for breast cancer, with the ultimate goal of achieving early diagnosis of the disease. The initial phase of this investigation involved the identification of 2 breast cancer gene chips meeting the specified criteria within the gene expression omnibus database. Subsequently, paired data analysis was conducted on these datasets, leading to the identification of differentially expressed genes (DEGs). In addition, this study executed Gene Ontology enrichment analysis and Kyoto encyclopedia of genes and genomes pathway enrichment analysis. The subsequent stage involved the construction of a protein-protein interaction network graph using the STRING website and Cytoscape software, facilitating the calculation of Hub genes. Lastly, the UALCAN database and Kaplan-Meier survival plots were utilized to perform differential expression and survival analysis on the selected Hub genes. A total of 733 DEGs were identified from the combined analysis of 2 datasets. Among these DEGs, 441 genes were found to be downregulated, while 292 genes were upregulated. The selected DEGs underwent comprehensive analysis, including gene ontology enrichment analysis, Kyoto encyclopedia of genes and genomes pathway enrichment analysis, and establishing a protein-protein interaction network. As a result, 10 Hub genes closely associated with early diagnosis of breast cancer were identified: PDZ-binding kinase, cell cycle protein A2, cell division cycle-associated protein 8, maternal embryonic leucine zipper kinase, nucleolar and spindle-associated protein 1, BIRC5, cell cycle protein B2, hyaluronan-mediated motility receptor, mitotic arrest deficient 2-like 1, and protein regulator of cytokinesis 1. The findings of this study unveiled the significant involvement of the identified 10 Hub genes in facilitating the growth and proliferation of cancer cells, particularly cell cycle protein A2, cell division cycle-associated protein 8, maternal embryonic leucine zipper kinase, nucleolar and spindle-associated protein 1, hyaluronan-mediated motility receptor, and protein regulator of cytokinesis 1, which demonstrated a more pronounced connection with the onset and progression of breast cancer. Further analysis through differential expression and survival analysis reaffirmed their strong correlation with the incidence of breast cancer. Consequently, the investigation of these 10 pertinent Hub genes presents novel prospects for potential biomarkers and valuable insights into the early diagnosis of breast cancer.