Identification of serum miRNAs by nano-quantum dots microarray as diagnostic biomarkers for early detection of non-small cell lung cancer.

Circulating microRNAs (miRNAs) are potential noninvasive biomarkers for cancer detection. We used preoperative serum samples from non-small cell lung cancer (NSCLC) patients and healthy controls to investigate whether serum levels of candidate miRNAs could be used as diagnostic biomarkers in patients with resectable NSCLC and whether they were associated with clinicopathologic characteristics. We initially detected expression of 12 miRNAs using quantitative reverse transcriptase polymerase chain reaction (qRT-PCR) in preoperative serum samples of 94 NSCLC patients and 58 healthy controls. We further validated our results using the fluorescence quantum dots liquid bead array for differentially expressed miRNAs in serum samples of 70 NSCLC patients and 54 healthy controls. Receiver operating characteristic (ROC) analysis was performed to select the best diagnostic miRNA cutoff value. A predictive model of miRNAs for NSCLC was derived by multivariate logistic regression. We found that five serum miRNAs (miR-16-5p, miR-17b-5p, miR-19-3p, miR-20a-5p, and miR-92-3p) were significantly downregulated in NSCLC, while miR-15b-5p was significantly upregulated (p < 0.05). Multivariate logistic regression analysis revealed that miR-15b-5p, miR-16-5p, and miR-20a-5p expression were independent diagnostic factors for the identification of patients with NSCLC after adjustment for patient's age and sex. In addition, the expression of serum miR-106-5p was higher in stage I than in stages IIa-IIIb, and no significant association was observed between expression of miRNAs and other variables including pathological type, tumor size, and lymph nodes status. Six serum miRNAs could potentially serve as noninvasive diagnostic biomarkers for resectable NSCLC. The predictive model combining miR-15b-5p, miR-16-5p, and miR-20a-5p was the best diagnostic approach.