Object: Ovarian cancer is the primary cause of cancer-associated deaths among gynaecological malignancies. Increasing evidence suggests that microRNAs may be potential biomarkers for the diagnosis and prognosis of cancer. In this study, we conducted a systematic review and meta-analysis to summarize the global research and to evaluate the overall diagnostic accuracy of miRNAs in detecting ovarian cancer.
Methods: A systematic literature search was conducted for relevant studies through July 20, 2017, in English databases (CENTRAL, MEDLINE, and EMBASE), the Grey reference database and Chinese databases. Statistical analysis was conducted using OpenMetaAnalyst, STATA 14.0 and RevMan 5.3. Pooled sensitivity, specificity, and other parameters were used to assess the overall miRNA assay performance using a bivariate random-effects model (BRM). Meta-regression and subgroup analyses were performed to dissect the heterogeneity. Sensitivity analysis was performed to assess the robustness of our analysis, and the publication bias of the selected studies was assessed using Deeks' funnel plot asymmetry test.
Results: Thirteen articles described 33 studies, including 1081 patients with ovarian cancer and 518 controls. The pooled results were as follows: sensitivity, 0.89 (95% CI: 0.84-0.93); specificity, 0.64 (95% CI: 0.56-0.72); positive likelihood ratio, 2.18 (95% CI: 1.89-2.51); negative likelihood ratio, 0.15 (95% CI: 0.11-0.22); and diagnostic odds ratio (DOR), 13.21 (95% CI: 9.00-19.38). We conducted subgroup analyses based on ethnicity, research design, and miRNA profiling and found that multiple miRNA panels were more accurate in detecting ovarian cancer, with a combined DOR of 30.06 (95% CI: 8.58-105.37).
Conclusion: Per the meta-analysis, circulating miRNAs may be novel and non-invasive biomarkers for detecting ovarian cancer, particularly multiple miRNA panels, which have potential diagnostic value as screening tools in clinical practice.
Keywords: Diagnosis; Meta-analysis; Multiple miRNA panels; Ovarian cancer; microRNAs.