MiRNA profiling in resectable NSCLC by multiplex next-generation sequencing.

Abstract

7060 Background: Early-stage NSCLC has a relapse rate around 40% within 5 years. With the advent of microRNA (miRNA), which seems to regulate many genes critical for tumorigenesis, there is a growing interest in the characterization of miRNAome in NSCLC specimens and to correlate them with prognosis. Next generation sequencing is a useful tool to study the miRNA content of solid tumors. Here, we applied high-throughput SOLiD transcriptome sequencing to study miRNAs expression in a cohort of early-stage NSCLC patients (tumor vs normal lung). Methods: RNA was isolated from frozen lung specimens (tumor and normal lung) from resectable NSCLC patients (n=35). Samples with a RIN ≥ 7 were analyzed and enriched in the miRNA fraction. miRNAs were sequencing using a bar-code multiplex SOLiD protocol. Data normalization was carried out by rescaling all data according to their counts. Readings were mapped against mature and no-mature miRNAs using miRBase. Statistical analysis was performed with CLCbio software and considered significant when p-adj<0.005. Results: Using the SOLiD high throughput sequencing, we performed a systemic miRNA expression profiling analysis of paired samples (tumor vs normal lung). A total of 1268 miRNAs (mature and no-mature) have been detected in at least one sample. The differential expression between normal and tumor samples shown that 6 miRNAs (miR-193b, miR-182, miR-96, miR-148a, miR-299, miR-590) were upregulated and 7 (miR-145, miR-133a, miR-218, miR-125a, miR-30a, miR-126 and miR-139) were down-regulated significantly in tumor samples compared with normal lung tissues. We are performing studies using qRT-PCR in an independent cohort to further validate these findings. Conclusions: The deep sequencing technology used for differential miRNA expression is useful and novel. The use of barcoding allows multiplexing and lowers cost per sample. Several miRNAs were differentially expressed between tumor and normal tissue, but this point needs to be further validated in an independent cohort. Supported by grants TRA09-0132 (MICINN) and RD06/0020/1024 (ISCIII).