Ji Xiaoyu, Sun Wei, Lv Chengzhou, Huang Jiapeng, Yu Ruonan, Dong Wenwu, Zhang Hao
Department of Thyroid Surgery, The First Hospital of China Medical University, No. 155 Nanjing North Street, Shenyang, Liaoning, 110001, China.
Endocrine. 2025 Mar;87(3):1120-1130. doi: 10.1007/s12020-024-04109-3. Epub 2024 Nov 26.
Patients with pulmonary metastases from differentiated thyroid carcinoma (DTC) have a significantly poorer prognosis compared to DTC patients without metastases. This study aimed to establish a nomogram combined with dynamic survival analysis to predict the survival probability of patients with pulmonary metastases from differentiated thyroid carcinoma and compensate for the underestimation of survival in patients with very poor prognosis.
Patient data were collected from the Surveillance, Epidemiological, and End Result (SEER) data resource from 2010 to 2019. Multivariate analysis was carried out by the Cox proportional hazards regression to construct a nomogram. Receiver operating characteristic (ROC) curves along with calibration were employed to assess the effectiveness of the model.The life table was used to estimate the conditional cancer-specific survival (CSS).
In the training set, the AUCs for the CSS nomogram were 0.728, 0.741, and 0.779, with a c-index of 0.682, indicating good predictive performance at 3, 5, and 10 years. In the validation set, the AUCs for the CSS nomogram were 0.706, 0.726, and 0.769, with a c-index of 0.696, while the AUCs for the 8th TNM staging system were 0.521, 0.555, and 0.601, with a c-index of 0.579. The overall 5-year conditional survival rate for patients increased slightly from 63.44 to 70.52%. The survival gap was greatest between patients aged <55 years and those aged ≥55 years.
We established a nomogram combined with dynamic survival analysis, which serve as promising options for prognosis estimation, to enhance personalized evaluation of survival risks and provide the basis for the development of more clinical treatment approaches.
与无转移的分化型甲状腺癌（DTC）患者相比，发生肺转移的DTC患者预后明显更差。本研究旨在建立一种列线图并结合动态生存分析，以预测分化型甲状腺癌肺转移患者的生存概率，并弥补对预后极差患者生存情况估计不足的问题。
收集2010年至2019年监测、流行病学和最终结果（SEER）数据资源中的患者数据。通过Cox比例风险回归进行多变量分析以构建列线图。采用受试者操作特征（ROC）曲线及校准来评估模型的有效性。使用生命表来估计条件癌症特异性生存（CSS）。
在训练集中，CSS列线图在3年、5年和10年时的曲线下面积（AUC）分别为0.728、0.741和0.779，一致性指数（c指数）为0.682，表明具有良好的预测性能。在验证集中，CSS列线图的AUC分别为0.706、0.726和0.769，c指数为0.696，而第8版TNM分期系统的AUC分别为0.521、0.555和0.601，c指数为0.579。患者的总体5年条件生存率从63.44%略有提高至70.52%。年龄＜55岁和≥55岁的患者之间生存差距最大。
我们建立了一种结合动态生存分析的列线图，其可作为预后评估的有前景的选择，以加强对生存风险的个性化评估，并为开发更多临床治疗方法提供依据。
