Background: The diagnostic value of circulating microRNAs (miRNAs) detection in prostate cancer (PC) patients is currently under debate. Thus, we performed this meta-analysis of published literature to systematically evaluate the diagnostic potential of circulating miRNAs in PC.
Methods: Eligible studies were searched in PubMed, Embase and Chinese National Knowledge Infrastructure databases. Sensitivity and specificity were pooled using a random-effects model and were used to plot the summary receiver operator characteristic (SROC) curve. All analyses were performed using the Stata 13.0 software and Meta-Disc 1.4 for windows.
Results: A total of ten articles were included for the meta-analysis according to the inclusion criteria. The pooled results based on all included studies showed circulating miRNAs have a relatively good diagnostic performance, with a sensitivity of 0.74, a specificity of 0.71 and an area under SROC curve (AUC) of 0.77 in indiscriminating PC from controls. Furthermore, meta-regression and subgroup analyses indicate that multiple circulating miRNAs detection displayed a better diagnostic performance than single one, with an AUC rising from 0.75 to 0.81. Additional interesting finding was that Caucasian-based circulating miRNAs assays could reach a higher accuracy compared with non-Caucasian-based one for PC with the p value of 0.0378.
Conclusion: Our results confirmed the potential use of circulating miRNAs in the early diagnosis of PC, especially the combination of multiple circulating miRNAs. However, large-scale prospective studies are still needed to further validate our findings.
Keywords: Circulating microRNA; Diagnosis; Meta-analysis; Prostate cancer.