Exploratory Analysis of Selected MicroRNAs as Tissue-Based Candidate Biomarkers in Ovarian Cancer

The aim of the study was to evaluate the expression and exploratory diagnostic potential of selected microRNAs (miRNAs) in ovarian cancer tissue compared with benign ovarian lesions. Tissue samples from 24 patients with epithelial ovarian cancer and 22 patients with benign ovarian lesions were analyzed by quantitative real-time polymerase chain reaction. Ten candidate miRNAs were assessed; six met the predefined detection criteria and were retained for analysis. Group differences were tested with non-parametric methods and false discovery rate (FDR) correction. Diagnostic performance was evaluated by receiver operating characteristic (ROC) analysis, and multivariable logistic regression adjusted for age and menopausal status. Survival analyses were exploratory. miR-381-3p was downregulated, whereas miR-582-5p and miR-425-5p were upregulated in ovarian cancer compared with benign ovarian lesions; all three differences remained significant after FDR correction. miR-425-5p showed the highest individual diagnostic accuracy (AUC = 0.854) and remained associated with ovarian cancer after adjustment for age and menopausal status (OR = 7.84; 95% CI = 2.05-29.97; p = 0.003). This association was also retained in Firth penalized logistic regression. An exploratory three-miRNA panel achieved an apparent AUC of 0.858 but did not significantly outperform miR-425-5p alone (DeLong p = 0.596); its optimism-corrected bootstrap AUC was 0.822. No significant associations with survival were identified. miR-425-5p was the strongest candidate biomarker in this exploratory tissue-based study. Independent validation is required before clinical application.