Yuan Jie, Li Zhirong, Tu Likuan, Cao Yijia, Li Qing, Li Fan
Department of General Surgery, University-Town Hospital of Chongqing Medical University, Chongqing, China.
Department of Surgery and Anesthesiology, University-Town Hospital of Chongqing Medical University, Chongqing, China.
Gland Surg. 2025 Aug 31;14(8):1497-1509. doi: 10.21037/gs-2025-111. Epub 2025 Aug 26.
Locally advanced differentiated thyroid cancer (LADTC) refers to a severe stage of differentiated thyroid cancer (DTC) with a relatively poor prognosis. This study aimed to construct a dynamic nomogram and risk stratification system to predict cancer-specific survival (CSS) in patients with LADTC.
A total of 4,856 patients diagnosed with LADTC from 2004 to 2020 were included from the Surveillance, Epidemiology, and End Results database. Least absolute shrinkage and selection operator (LASSO) and Cox regression analyses were utilized to identify variables and construct the dynamic nomogram. The performance of the nomogram was assessed using the concordance index (C-index), time-dependent receiver operating characteristic (ROC) curve, area under the curve (AUC), and calibration plot, while decision curve analysis (DCA) was conducted to evaluate clinical benefits. The improvement of the nomogram in comparison to the American Joint Committee on Cancer (AJCC) staging system was evaluated using the C-index, net reclassification index (NRI), and integrated discrimination improvement (IDI). A risk stratification system was established according to the total score of each patient in the nomogram.
Eight variables were identified to construct the nomogram. The C-index, time-dependent ROC curve, AUC, calibration plot, and DCA demonstrated the strong performance and clinical benefits of the nomogram. The C-index, NRI, and IDI indicated that the nomogram outperformed the AJCC staging system in prognostic prediction. The risk stratification system demonstrated the favorable ability to categorize patients with LADTC.
A dynamic nomogram and risk stratification system were constructed and validated to assist clinicians in evaluating prognostic risk and devising personalized treatment strategies for patients with LADTC.
局部晚期分化型甲状腺癌（LADTC）是分化型甲状腺癌（DTC）的一个严重阶段，预后相对较差。本研究旨在构建一个动态列线图和风险分层系统，以预测LADTC患者的癌症特异性生存（CSS）。
从监测、流行病学和最终结果数据库中纳入了2004年至2020年期间诊断为LADTC的4856例患者。采用最小绝对收缩和选择算子（LASSO）和Cox回归分析来识别变量并构建动态列线图。使用一致性指数（C指数）、时间依赖性受试者工作特征（ROC）曲线、曲线下面积（AUC）和校准图评估列线图的性能，同时进行决策曲线分析（DCA）以评估临床益处。使用C指数、净重新分类指数（NRI）和综合鉴别改善（IDI）评估列线图相对于美国癌症联合委员会（AJCC）分期系统的改进。根据列线图中每位患者的总分建立风险分层系统。
确定了8个变量来构建列线图。C指数、时间依赖性ROC曲线、AUC、校准图和DCA显示了列线图的强大性能和临床益处。C指数、NRI和IDI表明列线图在预后预测方面优于AJCC分期系统。风险分层系统显示出对LADTC患者进行分类的良好能力。
构建并验证了一个动态列线图和风险分层系统，以协助临床医生评估LADTC患者的预后风险并制定个性化治疗策略。
