Experimental Validation of miR-4443, miR-572, and miR-150-5p in Serum and Tissue of Breast Cancer Patients as a Potential Diagnostic Biomarker: A Study Based on Bioinformatics Prediction.

Breast cancer is the most common invasive cancer diagnosed in females and is also the main cause of cancer-related deaths leading to more than 500,000 deaths annually. The present study aims to identify a promising panel of microRNAs (miRNAs) using bioinformatics analysis, and to clinically validate their utility for diagnosing breast cancer patients with high accuracy in a clinical setting. First, in the in silico phase of our study, using bioinformatics analysis and the data available in the GEO database, miRNAs that were increased in the interstitial fluid of the tumor tissues (differentially expressed miRNAs), were screened and their related target genes were selected. Multimir package of R software was utilized to determine the target genes of the differentially expressed miRNAs (DEMs). The biological functions of discovered genes were analyzed using Gene Ontology (GO) analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis. In order to determine the molecular mechanisms behind important signaling pathways and cellular functions, the protein-protein interaction network was built using STRING and Cytoscape software. After that, in the laboratory phase, the expression level of three candidate miRNAs on the serum samples of 26 breast cancer patients and 26 control, as well as 14 tumor tissue samples and 14 adjacent normal tissue samples, has been investigated by Real-time PCR method. Then sensitivity and specificity of candidate miRNAs were evaluated through the ROC curve analysis. After in silico analysis, we revealed that three miRNAs including miR-4443, miR-572, and miR-150-5p were highly increased in the interstitial fluid of breast cancer patients compared to breast cancer tissues. Moreover, our results revealed that the expression level of miR-4443, miR-572, and miR-150-5p were significantly decreased in the serum of breast cancer patients compare to normal controls. Also, the expression level of miR-4443 and miR-150-5p was significantly decreased in the tumor tissue compared to the adjacent non-tumor tissue. Also, ROC curve analysis showed that these three miRNAs have high sensitivity and specificity for the diagnosis of breast cancer patients. Data analysis was conducted with GraphPad Prism software. Our findings suggest the potential utility of measuring tumor-derived miRNAs in serum as an important approach for the blood-based detection of breast cancer patients. It appears that miR-4443, miR-572, and miR-150-5p may serve as promising diagnostic biomarkers with high sensitivity and specificity. However, it's important to note that further research will be needed to definitively establish the use of these miRNAs as potential biomarkers in clinical practice.