Abstract
Glioma is the most common central nervous system tumor and associated with poor prognosis. Identifying effective diagnostic biomarkers for glioma is particularly important in order to guide optimizing treatment. MicroRNAs (miRNAs) have drawn much attention because of their diagnostic value in diverse cancers, including glioma. We summarized studies to identify the potential diagnostic values of miRNAs in glioma patients. We included articles reporting miRNAs for differentiation of glioma patients from controls. We calculated sensitivities, specificities, and area under the curves (AUC) of individual miRNA and miRNA panels. We found that overall sensitivity, specificity, and AUC of miRNAs in diagnosis of glioma were 85% (95% confidence interval [CI]: 0.81-0.89), 90% (95% CI 0.85-0.93), and 93% (95% CI 0.91-0.95), respectively. Meta-regression analysis showed that the detection of miRNAs expression in cerebrospinal fluid (CSF) and brain tissue largely improved the diagnostic accuracy. Likewise, panels of multiple miRNAs could enhance the pooled sensitivity. Moreover, AUC of miR-21 was 0.88, with 86% sensitivity and 94% specificity. This study demonstrated that miRNAs could function as potential diagnosis markers in glioma. Detection of miRNAs in CSF and brain tissue displays high accuracy in the diagnosis of glioma.