Systematic Analysis to Identify the MIR99AHG-has-miR-21-5p- CeRNA Regulatory Network as Potential Biomarkers in Lung Cancer.

Lung cancer (LC) remains an extremely lethal disease worldwide, and effective prognostic biomarkers are at top priority. With the rapid development of high-throughput sequencing and bioinformatic analysis methods, the quest to characterize cancer transcriptomes continues to move forward. However, the integrated systematic analysis of lncRNA-miRNA-mRNA regulatory network in LC is lacking. In this study, we collected samples of cancer tissues and adjacent normal tissues from patients with lung cancer and conducted transcriptome and small RNA sequencing to identify differentially expressed genes (DEGs), miRNAs (DEMs), and lncRNAs (DELs). The regulatory roles of miRNAs in LC were explained by functional analysis on DEM-targeted genes. The lncRNA-miRNA pairs, miRNA-mRNA pairs, and lncRNA-mRNA pairs were identified and combined to construct the interplay of lncRNA-miRNA-mRNA. We evaluated the prognostic value of selected lncRNA-miRNA-mRNA by Kaplan-Meier analysis. Finally, we analyzed the expression levels of selected DEM, DELs, and DEGs in lung cancer patients and healthy people to verify our findings. A total of 1492 DEGs, 12 DEMs, and 604 DELs were identified in LC patients. Based on the bioinformatic analysis and the regulatory mechanism of ceRNAs, 3 lncRNAs (GATA2-AS1, LINC00632, MIR99AHG), 1 miRNA (hsa-miR-21-5p) and 5 targeted genes (, , , and ) were figured out first. Through further Kaplan-Meier analysis screening the prognostic value, we finally found the hub subnetwork (MIR99AHG-hsa-miR-21-5p-) by collating lncRNA-miRNA pairs, miRNA-mRNA pairs and lncRNA-mRNA pairs. As the key of ceRNA regulatory network, the expression of miRNA-21-5p in lung cancer patients was significantly higher than that in healthy people ( < 0.01), and its high expression was significantly associated with poor prognosis ( = 0.0025). Our study successfully constructed a MIR99AHG-hsa-miR-21-5p- mutually regulatory network, suggesting the potential efficient biomarkers in LC.