Tian Dandan, Li Xiaoqin, Jia Zhongzhi
Department of Ultrasound, The Affiliated Changzhou No. 2 People's Hospital of Nanjing Medical University, Changzhou, Jiangsu, 213164, People's Republic of China.
Department of Interventional Vascular, The Affiliated Changzhou No. 2 People's Hospital of Nanjing Medical University, Changzhou, Jiangsu, 213164, People's Republic of China.
Cancer Manag Res. 2024 Nov 11;16:1571-1585. doi: 10.2147/CMAR.S485708. eCollection 2024.
To analyze the risk factors of cervical lymph node metastasis (LNM) of thyroid papillary carcinoma (PTC) and construct the prediction model.
Clinical data of 1105 patients with pathologically confirmed PTC in our hospital from February 2019 to May 2024 were retrospectively analyzed, and randomly divided into a training set and validation set according to the proportion of 7:3. With cervical central LNM (CLNM) and lateral LNM (LLNM) as outcome variables respectively, ultrasound characteristics were analyzed and C-TIRADS scores were performed Combined with the general situation of the patient, preoperative serum thyroglobulin (Tg) level, BRAFV600E (hereinafter referred to as BRAF) gene mutation and other characteristics of the patient, analysis was conducted to determine the independent risk factors for cervical CLNM and LLNM of PTC, and establish Nomogram prediction models. The test data set is used to validate the model. The area under the ROC curve (AUC) and the decision curve analysis (DCA) were used to evaluate the prediction efficiency of the model.
The analysis shows that male, age < 55 years old, tumor diameter ≥ 1 cm, capsular invasion, positive serum thyroglobulin (Tg), BRAF gene mutation type and C-TIRADS score are independent risk factors for cervical CLNM in PTC (P < 0.05). Tumor diameter ≥ 1 cm, capsular invasion, tumor located at the upper pole and presence of CLNM are independent risk factors for LLNM in PTC. Based on the above risk factors, Nomogram prediction models for CLNM and LLNM are constructed respectively. The AUC of the CLNM prediction model is 91.5%. LLNM model is 96.1%.
Ultrasound indicators, C-TIRADS score combined with BRAF gene status, Tg and clinical indicators of patients have important value in predicting cervical CLNM and LLNM in PTC. The Nomogram prediction models constructed based on the above indicators can effectively predict the risk of LNM in PTC.
分析甲状腺乳头状癌（PTC）颈部淋巴结转移（LNM）的危险因素并构建预测模型。
回顾性分析2019年2月至2024年5月我院1105例经病理确诊的PTC患者的临床资料，按照7:3的比例随机分为训练集和验证集。分别以颈部中央区LNM（CLNM）和侧方LNM（LLNM）为结局变量，分析超声特征并进行C-TIRADS评分，结合患者的一般情况、术前血清甲状腺球蛋白（Tg）水平、BRAFV600E（以下简称BRAF）基因突变等患者特征，分析确定PTC颈部CLNM和LLNM的独立危险因素，并建立列线图预测模型。用测试数据集对模型进行验证。采用ROC曲线下面积（AUC）和决策曲线分析（DCA）评估模型的预测效能。
分析显示，男性、年龄<55岁、肿瘤直径≥1 cm（厘米）、包膜侵犯、血清甲状腺球蛋白（Tg）阳性、BRAF基因突变类型及C-TIRADS评分是PTC颈部CLNM的独立危险因素（P<0.05）。肿瘤直径≥1 cm、包膜侵犯、肿瘤位于上极及存在CLNM是PTC中LLNM的独立危险因素。基于上述危险因素，分别构建CLNM和LLNM的列线图预测模型。CLNM预测模型的AUC为91.5%。LLNM模型为96.1%。
超声指标、C-TIRADS评分结合BRAF基因状态、Tg及患者临床指标对预测PTC颈部CLNM和LLNM具有重要价值。基于上述指标构建的列线图预测模型可有效预测PTC中LNM的风险。
