Department of Oral and Maxillofacial Surgery, Center of Stomatology, Xiangya Hospital, Central South University, Changsha, 410008, China.
Department of Orthodontics, Center of Stomatology, Xiangya Hospital of Central South University, Changsha, 410008, China.
Curr Med Sci. 2021 Oct;41(5):953-960. doi: 10.1007/s11596-021-2435-x. Epub 2021 Oct 25.
The use of the traditional American Joint Committee on Cancer (AJCC) staging system alone has limitations in predicting the survival of gingiva squamous cell carcinoma (GSCC) patients. We aimed to establish a comprehensive prognostic nomogram with a prognostic value similar to the AJCC system.
Patients were identified from SEER database. Variables were selected by a backward stepwise selection method in a Cox regression model. A nomogram was used to predict cancer-specific survival rates for 3, 5 and 10 years in patients with GSCC. Several basic features of model validation were used to evaluate the performance of the survival model: consistency index (C-index), receiver operating characteristic (ROC) curve, calibration chart, net weight classification improvement (NRI), comprehensive discriminant improvement (IDI) and decision curve analysis (DCA).
Multivariate analyses revealed that age, race, marital status, insurance, AJCC stage, pathology grade and surgery were risk factors for survival. In particular, the C-index, the area under the ROC curve (AUC) and the calibration plots showed good performance of the nomogram. Compared to the AJCC system, NRI and IDI showed that the nomogram has improved performance. Finally, the nomogram's 3-year and 5-year and 10-year DCA curves yield net benefits higher than traditional AJCC, whether training set or a validation set.
We developed and validated the first GSCC prognosis nomogram, which has a better prognostic value than the separate AJCC staging system. Overall, the nomogram of this study is a valuable tool for clinical practice to consult patients and understand their risk for the next 3, 5 and 10 years.
单独使用传统的美国癌症联合委员会（AJCC）分期系统在预测牙龈鳞状细胞癌（GSCC）患者的生存率方面存在局限性。我们旨在建立一个综合预后列线图，其预后价值与 AJCC 系统相似。
从 SEER 数据库中识别患者。使用 Cox 回归模型中的后向逐步选择方法选择变量。使用列线图预测 GSCC 患者的癌症特异性 3、5 和 10 年生存率。使用几种模型验证的基本特征来评估生存模型的性能：一致性指数（C 指数）、接收者操作特征（ROC）曲线、校准图、净重分类改善（NRI）、综合判别改善（IDI）和决策曲线分析（DCA）。
多变量分析显示，年龄、种族、婚姻状况、保险、AJCC 分期、病理分级和手术是生存的危险因素。特别是，列线图的 C 指数、ROC 曲线下面积（AUC）和校准图显示出良好的性能。与 AJCC 系统相比，NRI 和 IDI 表明列线图的性能有所提高。最后，列线图的 3 年、5 年和 10 年 DCA 曲线产生的净收益高于传统的 AJCC，无论是训练集还是验证集。
我们开发并验证了第一个 GSCC 预后列线图，其预后价值优于单独的 AJCC 分期系统。总的来说，本研究的列线图是一个有价值的临床实践工具，可以帮助患者咨询并了解他们未来 3、5 和 10 年的风险。
