Abstract
OBJECTIVE: This study aimed to assess the practicality and reliability of utilizing microRNAs (miRNAs) as a potential screening and diagnosing tool for non-small cell lung cancers (NSCLCs) in Northern Thailand. METHODS: Small RNA sequencing and a literature review was performed to obtain a list of serum miRNA candidates. Serum levels of these selected miRNA candidates were measured in patients with NSCLC and healthy volunteers by real-time RT-PCR and receiver operating characteristic curve (ROC) were used to assess diagnostic performance. RESULTS: Sequencing data revealed 148 known miRNAs and 230 novel putative miRNAs in serum samples; 19 serum miRNAs were significantly downregulated and 242 were upregulated. Seven miRNAs selected according to sequencing data and 11 miRNAs according to previous reports were evaluated in training cohort (45 lung cancer patients, 26 controls) and 6 miRNAs were found differentially expressed (p < 0.05, Mann Whitney U test) and associated (p < 0.05, Chi-square test) with NSCLC development. Further analysis and verification identified an optimal combination of 4 miRNAs composed of hsa-miR23a, hsa-miR26b, hsa-miR4488 and novel-130 to provide the optimal AUC of 0.901±0.034. Detection of serum miRNA by real-time RT-PCR showed good reproducibility with the coefficient of variation (CV) ≤ 4%. The optimal screening miRNAs panel was primarily identified through sequencing data of local patient population, thus indicating that the etiology of NSCLCs may differ from one population to other and thus require a unique panel of miRNAs for their identification. CONCLUSION: Circulating miRNA is a feasible screening tool for NSCLCs. Nevertheless, populations with different lung cancer etiology may need to identify their own most suitable miRNA panel.