Computational identification of putative mRNA-miRNA signatures in chemoresistant and metastatic triple-negative...

Triple-negative breast cancer (TNBC) is a heterogeneous malignancy that is recognized as having a high rate of metastasis and limited treatment options. Identifying new molecular signatures for TNBC remains urgent and challenging. In this study, we aimed to identify novel mRNA and miRNA expression profiles associated with invasive and drug-resistant TNBC using bioinformatics methods. We analyzed four datasets (GSE43502, GSE88847, GSE61723, and GSE154255) obtained from the Gene Expression Omnibus (GEO) repository. Significantly differentially expressed genes (DEGs) and miRNAs (DEMs), along with mRNA-miRNA interaction networks, were identified, visualized, and validated using R packages, enrichment analysis, protein-protein interaction (PPI) networks, and in silico validation through the UALCAN database. A total of 3242 DEGs and 95 DEMs were identified. Among these DEMs, 18 (13 downregulated and 5 upregulated) were selected as novel candidates. Target genes related to the candidate miRNAs were predicted and overlapped with DEGs. We identified 111 commonly upregulated and 129 commonly downregulated genes. Gene set enrichment analysis revealed that upregulated genes were involved in immune-related pathways, whereas downregulated genes were enriched in oncogenic signaling pathways, including MAPK and BRAF/RAF1 fusions. The mRNA-miRNA network highlighted miR-4690-5p and miR-338-3p as key regulators. In silico validation results confirmed that the BRD4 and LYN genes, as well as miR-16-2-3p, were significantly overexpressed, while the XBP1 gene and miR-5684 were markedly downregulated in TNBC samples. Overall, this study reveals computationally novel mRNA and miRNA expression profiles associated with metastatic and chemoresistant TNBC samples. These findings provide preliminary molecular insights may serve as tissue-specific and context-dependent biomarkers for future experimental investigations.