Predictive value of microRNAs as novel biomarkers in detection of lymphoma.

MicroRNAs (miRNAs) have attracted many attentions in lymphoma diagnostic research. The inconsistence of diagnostic performance in these existed literatures leading us to conduct this meta-analysis. In order to have a scientific and reliable study, all related articles were screened from Medline, Embase, CNKI and other databases. The sensitivity and specificity of each involved research were used to plot the summary receiver operator characteristic (SROC) curve and calculate the area under the curve (AUC). The QUADAS-2 tool was applied to estimate the quality of included studies. In addition, Deeks' funnel plot asymmetry test was performed to estimate publication bias. Overall, 14 studies from 6 articles were included to evaluate the whole test performance. The overall pooled results were as follows: sensitivity was 0.91 (95% CI: 0.83-0.95), specificity was 0.84 (95% CI: 0.75-0.90), the AUC was 0.93 (95% CI: 0.91-0.95), positive likelihood ratio-PLR was 5.5 (95% CI: 3.5-8.8), negative likelihood ratio-NLR was 0.11 (95% CI: 0.06-0.21), and diagnostic odds ratio-DOR was 50 (95% CI: 19-128). In summary, results from meta-analysis showed that miRNAs analysis might significantly increase the diagnostic accuracy of lymphoma. Further massive prospective studies still needed to validate our conclusion before clinical application.