Yang Yan, Wang Da-Song, Yang Lei, Huang Yun-Hui, He Yu, Chen Mao-Shan, Wang Zheng-Yan, Fan Li, Yang Hong-Wei
Department of Breast and Thyroid Surgery, Suining Central Hospital, Suining, People's Republic of China.
Cancer Biomark. 2025 Feb;42(2):18758592241311195. doi: 10.1177/18758592241311195. Epub 2025 Apr 2.
BackgroundPrecise recognition of neck lymph node metastasis (LNM) is essential for choosing the suitable scope of operation for papillary thyroid cancer(PTC) patients.ObjectiveThe purpose of our study was to establish an effective nomogram integrating both gene biomarkers and clinicopathologic features for preoperatively predicting LNM in PTC patients.MethodsWe gathered clinical information and gene expression data for PTC samples from The Cancer Genome Atlas database (TCGA). WGCNA and differential analysis were applied to identify LNM-related differentially expressed genes in PTC patients. We developed a risk score based on the 3-gene signature predicting LNM using the LASSO regression analysis. Furthermore, multivariate logistic regression analysis was performed to establish a nomogram. We evaluated the discriminative ability of the nomogram by calculating the area under the ROC curve. Besides, we applied the decision curve analyses and calibration curve to assess the nomogram's actual benefits and accuracy.ResultsSignificant predictors of LNM in PTC patients were eventually screened to develop a nomogram, which included age, histological type, focus type, T stage, and risk score calculated based on IQGAP2, BTBD11 and MT1G expression levels. The AUC value of the nomogram for training and validation set was 0.802 (95% CI 0.750-0.855) and 0.718 (95% CI 0.624-0.811). Moreover, the nomogram has outstanding calibration and actual clinical patient benefits.ConclusionsWe identified a nomogram based on the 3-gene signature and clinical characteristics that effectively predicted LNM in PTC patients, which offers guidance for the preoperative assessment the appropriate scope of operation in PTC patients.
背景
准确识别颈部淋巴结转移（LNM）对于选择合适的甲状腺乳头状癌（PTC）患者手术范围至关重要。
目的
我们研究的目的是建立一种有效的列线图，整合基因生物标志物和临床病理特征，用于术前预测PTC患者的LNM。
方法
我们从癌症基因组图谱数据库（TCGA）收集了PTC样本的临床信息和基因表达数据。应用加权基因共表达网络分析（WGCNA）和差异分析来识别PTC患者中与LNM相关的差异表达基因。我们使用套索回归分析基于预测LNM的三基因特征开发了一个风险评分。此外，进行多因素逻辑回归分析以建立列线图。我们通过计算ROC曲线下面积评估列线图的鉴别能力。此外，我们应用决策曲线分析和校准曲线来评估列线图的实际益处和准确性。
结果
最终筛选出PTC患者LNM的显著预测因素以建立列线图，其中包括年龄、组织学类型、病灶类型、T分期以及基于IQGAP2、BTBD11和MT1G表达水平计算的风险评分。训练集和验证集列线图的AUC值分别为0.802（95%CI 0.750 - 0.855）和0.718（95%CI 0.624 - 0.811）。此外，列线图具有出色的校准和实际临床患者益处。
结论
我们基于三基因特征和临床特征识别出一种列线图，可有效预测PTC患者的LNM，为术前评估PTC患者合适的手术范围提供指导。
