Identification of novel diagnostic biomarkers for thyroid carcinoma

// Xiliang Wang 1, 2 , Qing Zhang 1 , Zhiming Cai 1 , Yifan Dai 3 and Lisha Mou 1 1 Shenzhen Xenotransplantation Medical Engineering Research and Development Center, Institute of Translational Medicine, Shenzhen Second People's Hospital, First Affiliated Hospital of Shenzhen University, Shenzhen 518035, China 2 Department of Biochemistry in Zhongshan School of Medicine, Sun Yat-Sen University, Guangzhou 510080, China 3 Jiangsu Key Laboratory of Xenotransplantation, Nanjing Medical University, Nanjing 210029, China Correspondence to: Lisha Mou, email: lishamou@gmail.com Keywords: thyroid carcinoma; bioinformatics; dysregulation network; biomarker Received: June 27, 2017      Accepted: November 19, 2017      Published: December 04, 2017 ABSTRACT Thyroid carcinoma (THCA) is the most universal endocrine malignancy worldwide. Unfortunately, a limited number of large-scale analyses have been performed to identify biomarkers for THCA. Here, we conducted a meta-analysis using 505 THCA patients and 59 normal controls from The Cancer Genome Atlas. After identifying differentially expressed long non-coding RNA (lncRNA) and protein coding genes (PCG), we found vast difference in various lncRNA-PCG co-expressed pairs in THCA. A dysregulation network with scale-free topology was constructed. Four molecules (LA16c-380H5.2, RP11-203J24.8, MLF1 and SDC4) could potentially serve as diagnostic biomarkers of THCA with high sensitivity and specificity. We further represent a diagnostic panel with expression cutoff values. Our results demonstrate the potential application of those four molecules as novel independent biomarkers for THCA diagnosis.