Serum miR-181-5p, miR-214-3p, and miR-223-3p expression in epithelial ovarian cancer.

e17539 Background: Detection of epithelial ovarian cancer (EOC) remains challenging because of nonspecific symptoms and the limited performance of current serum biomarkers. Circulating microRNAs (miRNAs) are stable in blood and represent promising minimally invasive biomarker candidates. This study evaluated the diagnostic and prognostic relevance of serum miR-181-5p, miR-214-3p, and miR-223-3p in patients with EOC compared with healthy controls. Methods: In this single-center case–control study, pretreatment serum samples were obtained from 77 patients with histologically confirmed EOC and 74 healthy female controls. Serum miRNA expression was quantified by quantitative real-time PCR and normalized to U6 RNA. Differences between groups were analyzed using nonparametric tests and analysis of covariance adjusting for age and body mass index. Diagnostic performance was assessed by receiver operating characteristic (ROC) analysis. Associations with clinicopathologic features, platinum sensitivity, progression-free survival, and overall survival were evaluated using logistic regression and Cox proportional hazards models. Results: Serum miR-181-5p and miR-223-3p levels were significantly reduced in EOC compared with controls (both p<0.001), whereas miR-214-3p showed no significant difference (p=0.505). After adjustment for age and BMI, reductions remained significant for miR-181-5p (p<0.001) and miR-223-3p (p=0.004); miR-214-3p remained nonsignificant (p=0.052). ROC analysis showed moderate diagnostic accuracy for miR-181-5p (AUC 0.704; 95% CI 0.614–0.794) and miR-223-3p (AUC 0.732; 95% CI 0.644–0.820); miR-214-3p had limited utility (AUC 0.573; p=0.233). A combined model modestly improved discrimination (AUC 0.764; 95% CI 0.68–0.84). No significant associations were observed between any miRNA and FIGO stage, grade, histology, platinum resistance, PFS, or OS (all p > 0.05). Conclusions: Serum miR-181-5p and miR-223-3p are downregulated in epithelial ovarian cancer and demonstrate moderate diagnostic performance but lack prognostic or predictive value at diagnosis. These findings support their potential contribution to future multimodal biomarker strategies rather than their use as standalone markers. Diagnostic performance of circulating miRNAs for discriminating epithelial ovarian cancer from healthy controls. miRNA AUC (95% CI) p-value miR-181-5p 0.704 (0.614–0.794) <0.001 miR-223-3p 0.732 (0.644–0.820) <0.001 miR-214-3p 0.573 (0.453–0.693) 0.233 Combined model (miR-181-5p + miR-223-3p) 0.764 (0.68–0.84) 0.001 AUC, area under the curve; CI, confidence interval.