Diagnostic value of microRNA-21 in the diagnosis of lung cancer: evidence from a meta-analysis involving 11 studies.

Molecular biomarkers that can be detected in easily accessible body fluids have been proposed as non-invasive, cost-effective, and useful tools for cancer diagnosis. Recently, extensive research has explored the involvement of the aberrant expression of microRNA-21 (miRNA-21, miR-21) in lung cancer. Inconsistent results, however, have prevented its widespread use in diagnosis. In light of this situation, our meta-analysis aimed to systematically determine whether aberrant miR-21 expression can distinguish patients with lung cancer from cancer-free controls with a high level of diagnostic accuracy. A comprehensive literature search for relevant studies published before December 23, 2013 was conducted in the MEDLINE, EMBASE, the Cochrane Library, and three Chinese databases. The pooled sensitivity, specificity and other parameters were used to assess the overall performance of miR-21-based assays. Statistical analysis was conducted using the STATA 11.0 software. Eleven research articles involving 676 patients with lung cancer and 529 healthy controls were considered eligible for inclusion in the present meta-analysis. The following summary parameters were calculated from all the included studies: sensitivity of 0.66 (95 % confidence interval [CI]: 0.57-0.74), specificity of 0.82 (95 % CI: 0.74-0.88), positive likelihood ratio (PLR) of 3.70 (95 % CI: 2.50-5.60), negative likelihood ratio (NLR) of 0.42 (95 % CI: 0.32-0.54); diagnostic odds ratio (DOR) of 9.00 (95 % CI: 5.00-16.00), and area under the curve (AUC) of 0.81 (95 % CI: 0.77-0.84). In addition, we added two pre-specified covariates (ethnicity and specimen types) to the bivariate model to assess their impact on the diagnostic value of miR-21 for lung cancer. Similar results were also observed in subgroup analyses, indicating a relatively low level of accuracy. The current meta-analysis indicates that a single miR-21 may not be sufficient to identify lung cancer and that more miRNAs should be used to detect lung carcinoma.