Exosomal microRNA biomarkers in ovarian cancer detection: systematic review and meta-analysis

Ovarian cancer (OC) is the deadliest gynecologic malignancy, mainly due to late-stage diagnosis and the limited accuracy of biomarkers such as Carbohydrate Antigen 125 and Human Epididymis Protein 4. Exosomal microRNAs are promising non-invasive biomarkers for early OC detection. This study assessed the diagnostic accuracy of these methods through a systematic review and meta-analysis.

PubMed, EMBASE, and Web of Science were searched for studies indexed to 29 June 2025 using predefined MeSH and free-text terms. Studies were screened in Rayyan under a PICO framework. Reviewers extracted data on study design, exosomal miRNAs, sample type, detection method, and diagnostic performance. Quality was assessed using a modified QUADAS-2 tool. Meta-analysis in STATA ("MIDAS" package) calculated pooled sensitivity, specificity, and area under the summary ROC (SROC) curve. Heterogeneity was evaluated using the I² statistic, and publication bias was assessed with Deeks' funnel-plot test.

Twenty-seven studies met inclusion criteria; diagnostic data from 16 cohorts were pooled. Sensitivity and specificity were 0.82 (95 % CI: 0.76-0.86) and 0.81 (95 % CI: 0.75-0.86), with an SROC area of 0.88. No publication bias was detected.

Exosomal miRNAs show strong diagnostic accuracy for non-invasive OC detection. Standardized methods and prospective validation are required for clinical application.