Clinical use of microRNAs as potential non-invasive biomarkers for detecting non-small cell lung cancer: a meta-analysis.

Abstract

Emerging studies have revealed that microRNA (miRNA) in body fluid may serve as a potential biomarker to detect non-small cell lung cancer (NSCLC). However, the diagnostic accuracy of miRNA for NSCLC detection is still under debate because there is inconsistency in previous studies. Hence, we conducted this meta-analysis to comprehensively evaluate the diagnostic performance of miRNA. A systematic literature search was performed to retrieve relevant articles in PubMed and other databases, and STATA 12.0 (StataCorp, College Station, TX, USA) was used to calculate the pooled parameters. A total of 28 articles involving 2121 NSCLC patients and 1582 healthy controls were included in this meta-analysis. The overall pooled sensitivity and specificity of miRNA were 0.75 and 0.79, respectively. The pooled positive likelihood ratio was 3.6, negative likelihood ratio was 0.32 and diagnostic odds ratio was 12. The area under the curve (AUC) was 0.84. Subgroup and meta-regression analyses established that miRNA assays were more accurate in Caucasian populations (AUC of 0.86, sensitivity of 0.79 and specificity of 0.82, respectively) than in Asian populations (AUC, sensitivity and specificity of 0.83, 0.72 and 0.80, respectively). In addition, the multiple miRNA assays (AUC of 0.89, sensitivity of 0.83 and specificity of 0.82, respectively) showed a higher accuracy than single miRNA assays (AUC, sensitivity and specificity of 0.81, 0.77 and 0.71, respectively) in NSCLC detection. Subgroup analyses based on specimen types suggested that blood-based miRNA (AUC of 0.86, sensitivity of 0.78 and specificity of 0.80, respectively) may have a higher diagnostic accuracy as biomarkers than sputum-based miRNA (AUC of 0.81, sensitivity of 0.69 and specificity of 0.80, respectively). In conclusion, miRNA may serve as a potential biomarker in NSCLS detection, especially the multiple miRNA from blood, with a relatively high diagnostic accuracy.