Li Wenhan, Tang Yao, Li Jianhui, Han Wei, Wang Danfang, Wang Yongheng
Department of Surgical Oncology, Shaanxi Provincial People's Hospital.
Department of Surgical Oncology,The Third Affiliated Hospital, School of Medicine, Xi'an Jiaotong University.
Medicine (Baltimore). 2021 May 7;100(18):e25811. doi: 10.1097/MD.0000000000025811.
This study aims to explore the prognostic variables for elderly papillary thyroid microcarcinoma (PTMC) patients as well as create a nomogram that could predict the occurrence of cervical lymph node metastasis (CLNM) on the basis of a large population database with high quality.A total of 5165 PTMC patients from Surveillance, Epidemiology, and End Results database database were enrolled in the study. In the meantime, we retrospectively collected 205 PTMC patients who underwent thyroidectomy in our medical center as an external control to test the accuracy of the model. The independent predictors of survival were identified by multivariate Cox regression analysis. Risk factors were selected as nomogram parameters to develop a model to predict CLNM. The C-index and calibration plots were used to evaluate CLNM model discrimination. The predictive nomogram was further validated in the external validation set.76.8% of the enrolled patients underwent thyroidectomy. Overall survival and cancer-specific survival were significantly better in patients who underwent surgery than in those who did not (P < .001). Sex, tumor size, and extent of tumor were included in a multivariable logistic regression model to predict lymph node metastasis. The nomogram had good discrimination with a C-index of 0.71. The calibration curves showed perfect agreement between nomogram predictions and actual observations.Elderly PTMC patients who received a surgical approach without radiotherapy showed survival advantage than those with other treatment strategies. Moreover, a nomogram model was established to predict the risk of CLNM, which will help clinicians in making treatment decisions.
本研究旨在探索老年甲状腺微小乳头状癌（PTMC）患者的预后变量，并基于高质量的大样本人群数据库创建一个能预测颈部淋巴结转移（CLNM）发生的列线图。本研究纳入了监测、流行病学和最终结果数据库中的5165例PTMC患者。同时，我们回顾性收集了在我们医疗中心接受甲状腺切除术的205例PTMC患者作为外部对照，以检验模型的准确性。通过多因素Cox回归分析确定生存的独立预测因素。选择危险因素作为列线图参数来建立预测CLNM的模型。使用C指数和校准图来评估CLNM模型的区分度。预测列线图在外部验证集中进一步得到验证。76.8%的入组患者接受了甲状腺切除术。接受手术的患者的总生存期和癌症特异性生存期显著优于未接受手术的患者（P<0.001）。将性别、肿瘤大小和肿瘤范围纳入多变量逻辑回归模型以预测淋巴结转移。该列线图具有良好的区分度，C指数为0.71。校准曲线显示列线图预测与实际观察结果完全一致。接受手术且未接受放疗的老年PTMC患者比采用其他治疗策略的患者具有生存优势。此外，建立了一个预测CLNM风险的列线图模型，这将有助于临床医生做出治疗决策。
