Aim: To identify the diagnostic accuracy of differential expressed circulating miRNAs in gliomas and their biological function in tumour progression
Background: Glioma is the most common malignant central nervous system tumour with rapid progression and metastasis, which require we identify some diagnostic tools for gliomas during the early stages. MicroRNAs (miRNAs) demonstrated their accurate predictive capacity in cancers and neurological diseases including gliomas. Despite that altered miRNAs are correlated with diagnosis and prognosis with gliomas from healthy controls, there is still a controversial issue.
Methods: PubMed, Medline and Cochrane database were searched for collecting studies which demonstrated single miRNA or miRNA panel derived from circulating blood were potential diagnostic biomarker for gliomas. The pooled diagnostic parameters were calculated by random effect models and the overall diagnostic performance of alter miRNAs was illustrated by the summary receiver operator characteristic (SROC) curves. The pooled sensitivity, specificity, positive likelihood ratio (PLR) and negative likelihood ratio (NLR) from each study were calculated by mixed model.
Results: Eleven studies involving 27 miRNAs from serum or plasma met our criteria and were included in this meta-analysis. Sixteen of 27 miRNAs were up regulated in the serum or plasma compared with the healthy controls, whilst 11 miRNAs were down expressed. The pooled PLR, NLR and DOR were 6.95 (95%CI, 4.82-10.00), 0.17 (95%CI, 0.10-0.21) and 47.41 (95%CI, 25.13-89.43), respectively. The pooled sensitivity, specificity and area under the curve (AUC) were 0.87 (95%CI, 0.82-0.91), 0.87 (95%CI, 0.82-0.91) and 0.93 (95%CI, 0.90-0.95), respectively, which demonstrated circulating miRNAs are capable of distinguishing gliomas from healthy controls.
Conclusions: Collectively, this meta analysis demonstrated circulating (serum or plasma) miRNAs are promising diagnostic biomarkers for gliomas.