RESUMO
Lung cancer, predominantly by non-small cell lung cancer (NSCLC), is the leading cause of cancer-related deaths over the world. Late diagnosis is one of important reasons for high mortality rate in lung cancer. Current diagnostic approaches have disadvantages such as low accuracy, high cost, invasive procedure, etc. MicroRNAs were previously proposed as promising novel biomarkers in cancer screening. In this study, we evaluated the predictive power of four candidate miRNAs in NSCLC detection. Our study involved 152 NSCLC patients and 300 healthy controls. Blood samples were obtained from the total 452 subjects. After miRNA extraction from serum, the expression of miRNAs in cases and controls were quantified by qRT-PCR and normalized to the level of U6 small RNA. Statistical analyses were performed to compare miRNA levels between cases and controls. Stratified analyses were employed to compare miRNA levels in NSCLC patients with different clinical characteristics. Serum miR-148a, miR-148b, and miR-152 were significantly downregulated in NSCLC patients. However, overexpression of serum miR-21 was observed in NSCLC patients. The combination of four candidate miRNAs exhibited the highest predictive accuracy in NSCLC screening compared with individual miRNAs (AUC = 0.97). Low level of miRNA-148/152 members may associate with advanced stage, large tumor size, malignant cell differentiation, and metastasis. High expression of miR-21 was possibly correlated with large size tumor and advanced cancer stage. Our results showed the dysregulation of miR-148/152 family and miR-21 in NSCLC patients. Hence, the four candidate miRNAs have great potential to serve as promising novel biomarkers in NSCLC screening. Further large-scale studies are needed to validate our results.