RESUMO
Melanoma is a destructive skin disease with few therapeutic options in the developed stage and therefore there is a critical need for reliable biomarkers for early diagnosis. In this context, microRNAs could play an important role as diagnostic biomarkers. Three datasets with accession numbers GSE31568, GSE61741 and GSE20994 were downloaded from the Gene Expression Omnibus (GEO) database. MATLAB software was used to analyze differentially expressed miRNAs between cutaneous melanoma plasma samples and normal plasma samples (control). Plasma levels of miR-193b, miR-146b-3p and miR-483-3p were evaluated by the RT-PCR method. Furthermore, linear regression followed by receiver operating characteristic analyses was performed to estimate whether selected plasma miRNAs were able to distinguish between cases and controls. Finally, the data were analyzed by unpaired Mann-Whitney U test using Graph pad prism 8 computer software. Specifically, miR-193b and miR-146b-3p were downregulated in the plasma of melanoma patients compared with control groups which were decreased 5 × [Formula: see text]-fold in miR-193b and 58-fold in miR-146b-3p, while miR-483-3p was upregulated 3.5-fold. After receiver operating characteristic (ROC) curve analysis, miR-193b with the most area under the curve (AUC: 1.00, 95% confidence interval 1.00-1.00, p < 0.0001) had the best discriminatory power, and miR-146b-3p had the large area under the curve (AUC: 0.96, 95% confidence interval 0.96-1.00, p < 0.0001) and consequently the high discriminatory power. Between these three miRNAs, miR-193b and miR-146b-3p had a high capacity to distinguish between melanoma patients and control groups that are appropriate to be applied in melanoma diagnosis as an early and noninvasive method.
