Construction of a miRNA Panel for Differentiating Lung Adenocarcinoma Brain Metastases and Glioblastoma

Brain metastases (BM) are the most common type of intracranial malignant tumor and are associated with high mortality. More than 50% of BM cases originate from lung cancer, and lung adenocarcinoma (LUAD) is most commonly associated with the development of BM (25%). The differential diagnosis of solitary BM and glioblastoma (GBM)-one of the most aggressive and fatal malignant brain tumors-remains a considerable challenge. Given the major role of microRNAs (miRNAs) in regulating gene expression, their clinical potential as biomarkers for tumor diagnosis and prognosis offers significant promise.

Next-generation RNA Sequencing (RNA-seq) was used to assess the miRNA expression profiles of 6 LUAD-BM, 6 GBM, and 6 control (non-tumoral brain tissue samples) human brain tissue samples. miRNAs exhibiting the most significant differential expression in LUAD-BM patients in comparison to both control subjects and GBM patients were selected for validation through RT-qPCR.

The analysis of RNA-seq data revealed the presence of 229 differentially expressed miRNAs in the comparison between LUAD-BM and control samples and 46 in the comparison between LU-AD-BM and GBM samples. Eight miRNAs were selected for further analysis, four of which were upregulated and four downregulated, based on the significant differences in their expression levels observed between the LUAD-BM samples and the other two groups, as confirmed with the Mann-Whitney U test. Functional enrichment analysis was also conducted based on a miRNA-centered target analysis performed using the miRNet tool. To assess the diagnostic potential of these differentially expressed miRNAs, we performed a receiver operating characteristic (ROC) curve analysis.

A panel of eight miRNAs was identified in human brain tissue samples, exhibiting high accuracy in distinguishing LUAD-BM from both GBM and control samples.