Chen, Endong1; Zheng, Zhouci1; Zhang, Wei1; Jin, Yixiang1; Jin, Wenxu1; Wang, Qingxuan1; Cai, Yefeng1; Li, Quan1; Wang, Ouchen1; Zhang, Xiaohua1
1 Department of Oncology, The First Affiliated Hospital of Wenzhou Medical University,
Wenzhou, Zhejiang Province, China
Background/Purpose: Papillary thyroid carcinoma (PTC) is the most common type of thyroid cancer. In 2014, The Cancer Genome Atlas (TCGA) described the genomic landscape of 496 PTCs and identified two major clusters defined as BRAF V600E-like and RAS-like. The discovery of abnormal expression of BRAFV600E-RAS Score (BRS) in PTCs has been remain unclear. It’s important to develop a model for prediction of lymph node (LN) metastasis in patients with these two clusters PTCs.
Methods: We downloaded a 71-gene signature of BRS of RNA-Seq profiles and clinicopathological features of 373 PTC tumors who had the lymph node status records from TCGA. Multivariable logistic regression analysis was used to develop the predicting model, we incorporated the genetic signature and independent preoperative clinical factors.
Results: The genetic signature, which consisted of 3 nonzero coefficient genes (CREB5, MET and MLEC), was signi?cantly associated with LN status. The 3 selected genes model yielded a C-index of 0.743 (95% CI, 0.695 to 0.787). Addition of clinical characteristics to the model revealed incremental prognostic value. Predictors contained in the personal prediction model included the genetic signature, age, and gender. The model showed a better discrimination, with a C-index of 0.778 (95% CI, 0.732 to 0.819; P = 0.0059).
Discussion & Conclusion: This study displays a novel model that integrates the gene expression profiles and clinical factors, which may be conveniently used to facilitate the preoperative FNAB current molecular-based approaches prediction of LN metastasis in patients with PTCs.