Value of miR-21 levels as potential biomarkers in the early diagnosis of hepatocellular carcinoma：a meta-analysis.

Background: Many studies have reported that miR-21 levels are different between hepatocellular carcinoma (HCC) patients and healthy controls, which could be used as a potential diagnostic biomarker for HCC. However, the diagnostic value of miR-21 for HCC varied greatly in previous studies. Therefore, this meta-analysis aims to provide higher grade evidence to investigate the diagnostic value of miR-21 for HCC. Methods: The databases of PubMed, Embase, Web of Science, and Chinese databases (CNKI and VIP) were searched. The indices of miR-21 in the diagnosis of HCC were pooled using bivariate random-effect models. QUADAS-2 was used to evaluate the quality of included studies. All statistical analyses were performed by STATA (12.0) software. Results: Totally, 1589 subjects from 14 publications were included in this study. The pooled sensitivity, specificity, positive likelihood ratios (PLR), negative likelihood ratios (NLR), and area under the curve (AUC) were 0.83 (0.77-0.88), 0.80 (0.74-0.85), 4.12 (3.04-5.57), 0.21 (0.15-0.30), and 0.88 (0.85-0.91), respectively. Subgroup analysis showed that the AUC was higher in Non-China subgroup, qRT-PCR subgroup, and plasma subgroup than that in China subgroup, ddPCR subgroup, and serum subgroup, respectively. However, the AUC was not significantly different between the healthy control subgroup and chronic hepatitis control subgroup. Significant heterogeneity was found in this meta-analysis, while no evident publication bias was identified. Conclusions: miR-21 is a valuable biomarker for the early diagnosis of HCC.