Five genes involved in circular RNA-associated competitive endogenous RNA network correlates with metastasis in papillary thyroid carcinoma

This study aimed to identify potential circular RNA (circRNA), microRNA (miRNA) and mRNA biomarkers as well as their underlying regulatory mechanisms in papillary thyroid carcinoma (PTC). Three microarray datasets from the Gene Expression Omnibus database as well as expression data and clinical phenotype from The Cancer Genome Atlas (TCGA) were downloaded, followed by differential expression, functional enrichment, protein-protein interaction (PPI), and module analyses. The support vector machine (SVM)-recursive feature elimination (RFE) algorithm was used to screen the key circRNAs. Finally, the mRNA-miRNA-circRNA regulatory network and competitive endogenous RNA (ceRNA) network were constructed. The prognostic value and clinical correlations of key mRNAs were investigated using TCGA dataset, and their expression was validated using the UALCAN database. A total of 1039 mRNAs, 18 miRNAs and 137 circRNAs were differentially expressed in patients with PTC. A total of 37 key circRNAs were obtained using the SVM-RFE algorithm, whereas 46 key mRNAs were obtained from significant modules in the PPI network. A total of 11 circRNA-miRNA pairs and 40 miRNA-mRNA pairs were predicted. Based on these interaction pairs, 46 circRNA-miRNA-mRNA regulatory pairs were integrated, of which 8 regulatory pairs in line with the ceRNA hypothesis were obtained, including two circRNAs (circ_0004053 and circ_0028198), three miRNAs (miR-199a-5p, miR-199b-5p, and miR-7-5p), and five mRNAs, namely APOA2, CCL20, LPAR5, MFGE8, and TIMP1. Survival analysis showed that LPAR5 expression was associated with patient survival. APOA2 expression showed significant differences between metastatic and non-metastatic tumors, whereas CCL20, LPAR5, MFGE8 and TIMP1 showed significant differences between metastatic and non-metastatic lymph nodes. Overall, we identified several potential targets and regulatory mechanisms involved in PTC. APOA2, CCL20, LPAR5, MFGE8, and TIMP1 may be correlated with PTC metastasis.