Abstract
Background
Brain metastasis (BMs) is commonly observed in patients with non-small cell lung cancer (NSCLC) and is associated with poor prognosis. Specific tissue microRNAs have emerged as clinical biomarkers for the diagnosis of NSCLC, and their profiles in cerebrospinal fluid (CSF) may be ideal biomarkers for the early detection of BMs in NSCLC patients.
Methods
Based on previous studies, we selected miR-155, miR-21, and miR-492 as specific microRNA cluster for further investigation. We detected and compared the concentration of this miRNA cluster in the serum and CSF of the patients with NSCLC with or without BMs. A receiver operating characteristic curve was used to evaluate the role of microRNAs in predicting BMs in NSCLC patients. The cutoff values were set as the criteria for abnormal expression for qualitative analysis, and patients with all or one of the miRNAs in this cluster with abnormal expression were classified as positive cases.
Results
The serum and CSF concentrations of miR-155, miR-21, and miR-492 were significantly higher in NSCLC patients than in controls. NSCLC patients with BMs had higher concentrations of miR-155, miR-492, and miR-21 in the CSF than without BMs. The miR-155, miR-492, and miR-21 in the CSF were used for predicting BMs from NSCLC, and their areas under the curve (AUC) were 0.911, 0.863, and 0.872, respectively. The concentrations of miR-155, miR-492, and miR-21 in the CSF closely correlated with those in the serum of NSCLC. According to the qualitative analysis, the percentages of specific microRNA clusters in patients with NSCLC with BMs were significantly greater than in patients with NSCLC without BMs.
Conclusions
Our findings indicate that the concentrations of specific microRNA clusters, including miR-155, miR-21, and miR-492, increased in the CSF of patients with NSCLC with BMs, suggesting the potential value in predicting BMs in NSCLC patients.