基于阴性淋巴结数目的胃印戒细胞癌预后评估模型的建立与验证

目的 探讨阴性淋巴结数目（NLNC）对胃印戒细胞癌（GSRC）患者预后的影响及构建G S R C 患者的预后预测模型。方法 基于SEER数据库收集GSRC患者2101例，随机分为建模组和验证组，检验临床病理特征与GSRC预后的关系。多因素Cox比例风险回归模型分析影响总生存的独立危险因素并建立预后预测模型。一致性指数（C?index）、校准曲线、净分类指数（NRI）、综合判别指数（IDI）和临床决策曲线（DCA）对列线图进行准确性和临床适用性评估。结果 所有患者按照7:3比例划分，建模组1473例，验证组628例。NLNC>10是GSRC患者预后的保护因素（HR=0.578, 95%CI: 0.504~0.662），根据多因素Cox比例风险回归模型筛选的变量建立Nomogram图，建模组和验证组的C-index分别为0.737（95%CI: 0.720~0.753）和0.724（95%CI: 0.699~0.749），区分度良好，校准曲线显示模型的一致性较高。NRI=17.77%，连续NRI=36.34%，IDI=4.2%，表明该模型较传统模型是正向收益，DCA决策曲线远离基准线表明模型临床适用性好。结论 NLNC增加是GSRC患者预后的有利因素。本研究建立的列线图相对准确，可预测GSRC患者的预后。

Development and Validation of Prognostic Nomogram Based on Negative Lymph Node Count for Patients with Gastric Signet Ring Cell Carcinoma#br#

Objective To explore the influence of negative lymph node count (NLNC) on the prognosis of patients with gastric signet ring cell carcinoma (GSRC) and develop a prognostic nomogram based on NLNC. Methods On the basis of the SEER database, 2 101 patients diagnosed with GSRC were collected and randomly divided into the modeling group and validation group to test the relationship between clinicopathological characteristics and the prognosis of GSRC. The multivariate Cox proportional hazard regression model was used to analyze the independent risk factors affecting overall survival and establish a prognostic prediction model. The consistency index (C-index), calibration curve, net reclassification index (NRI), integrated discrimination improvement (IDI), and decision curve analysis (DCA) were used to evaluate the accuracy and clinical applicability of the nomogram. Results All patients were divided according to the ratio of 7:3, with 1 473 in the modeling group and 628 in the validation group. NLNC>10 (HR=0.578, 95%CI: 0.504-0.662, P<0.001) was a protective factor for the prognosis of patients with GSRC, and the nomogram model was established based on multivariate Cox proportional hazards model. The C-index values of the nomogram were 0.737 (95%CI: 0.720-0.753) and 0.724 (95%CI: 0.699-0.749) in the modeling and validation groups, respectively, showing good discrimination. The calibration curves showed high consistency of the model. NRI=17.77%, continuous NRI=36.34%, and IDI=4.2% indicated that the model had positive returns compared with the traditional model. The DCA was far from the baseline, indicating that the model had good clinical applicability. Conclusion The increase in NLNC is a favorable factor for the prognosis of patients with GSRC, and a relatively accurate nomogram was established to predict the prognosis of patients with GSRC and help clinicians conduct individualized prognostic evaluations.