Early Diagnostic Value of Circulating MiRNA-21 in Lung Cancer: A Meta-Analysis

To evaluate the early diagnostic value of circulating miRNA-21 in diagnosis of lung cancer, databases such as Wan Fang, VIP, PubMed, and Elsevier were systematically searched from 2005 to 2013 to collect relevant references in which the diagnostic value had been evaluated. The statistics were consolidated and the qualities of the studies were classified. The data were analyzed using Meta Disc1.4 software. The diagnostic value of circulating miRNA-21 in lung cancer was assessed by pooling sensitivity, specificity, the likelihood ratio, and the Summary Receiver Operating Characteristic (SROC) curve. Publication biases of the studies involved were analyzed using Stata 11.0 software. A total of 143 papers were collected of which 8 were included, which contained 600 cases and 440 controls. A heterogeneity test proved the existence of homogeneity in this study. Upon analysis using random effects models, the weighted sensitivity was 0.68, the specificity 0.77, the positive likelihood ratio 2.84, the negative likelihood ratio 0.40, and the SROC Area Under the Curve (AUC) was 0.8133. Further analysis by subgroup showed that the 5 indicators mentioned above were 0.72, 0.84, 4.50, 0.27, and 0.8987, respectively, for the serum group and 0.63, 0.70, 1.95, 0.53, and 0.7318, respectively, for the plasma group. We conclude that circulating miRNA-21 can be regarded a valuable reference in diagnosis of lung cancer. This research showed that in lung cancer the early diagnostic value of miRNA-21 in serum was better than that in plasma.