基于SEER数据库手术后原发性阑尾肿瘤患者预后列线图构建与验证

Development and validation of a nomogram to predict overall survival in patients with primary appendiceal tumors after surgery based on the SEER database

目的探讨原发性阑尾肿瘤患者手术后影响生存的独立危险因素,构建并验证列线图,帮助识别高危患者,制定个体化治疗方案。方法回顾性收集美国Surveillance,Epidemiology,and End Results(SEER)数据库2010~2015年诊断为阑尾肿瘤的患者临床资料,随机分为训练队列和验证队列。采用多因素Cox回归分析影响原发性阑尾肿瘤术后患者总生存期(OS)的独立危险因素,开发了一种新的列线图模型,并通过内部验证进行评估。结果年龄、病理分型、肿瘤分化、N分期、M分期、淋巴结清扫数量、CEA状态是影响术后阑尾肿瘤患者预后的独立危险因素(P<0.05)。该列线图训练队列的C指数为0.811(95%CI:0.797~0.825),验证队列的C指数为0.844(95%CI:0.819~0.869)。1、3、5年总生存率ROC曲线下面积(AUC)在训练队列和验证队列中分别为0.807、0.849、0.824和0.857、0.862、0.825。采用X年10次200折交叉验证,进一步验证预测模型区分不同结局事件患者的能力。校准曲线和临床决策曲线(DCA)显示具有良好的一致性和临床获益。风险分级系统将所有患者分为三组,Kaplan-Meier曲线显示不同组间OS具有良好的分层和区分能力。结论我们开发了一种新的列线图模型来预测原发性阑尾肿瘤术后OS。此外,风险分级系统有助于准确评估预后和指导治疗。

Objective Exploring the independent risk factors that affect the survival of patients with primary appendiceal tumors after surgery,constructing and validating column charts to help identify high-risk patients,and developing personalized treatment plans.Methods We retrospectively collected clinical data from patients diagnosed with appendiceal tumors in the American Surveillance,Epidemiology,and End Results(SEER)database from 2010 to 2015,and randomly divided them into a training set and a validation set.Multiple factor Cox regression analysis was used to identify independent risk factors that affect overall survival(OS)in patients with primary appendiceal tumors after surgery.A new column chart model was developed and evaluated through internal validation.Results Age,Pathology type,Tumor stage,N-stage,M-stage,lymph node dissection quantity,and CEA status were identified as independent risk factors affecting the prognosis of postoperative appendiceal tumor patients(P<0.05).The C-index of the column chart training set was 0.811(95%CI:0.797~0.825),and the C-index of the validation set was 0.844(95%CI:0.819~0.869).The area under the ROC curve(AUC)of the 1-,3-,and 5-year overall survival rates was 0.807,0.849,and 0.824 in the training set,and 0.857,0.862,and 0.825 in the validation set,respectively.Further validation of the predictive model's ability to distinguish between different outcome events was performed using X-year 10-fold 200-fold cross-validation.The calibration curve and decision curve analysis(DCA)showed good consistency and clinical benefits.The risk grading system divided all patients into three groups,and the Kaplan-Meier curve showed good stratification and discrimination ability between different groups regarding OS.Conclusions We developed a new column chart model to predict postoperative overall survival(OS)in patients with primary appendiceal tumors.In addition,the risk grading system helps accurately assess prognosis and guide treatment.