基于SEER数据库的早期卵巢癌患者切除淋巴结后生存列线图模型构建  

Development and validation of nomogram to predict survival outcome in early stage ovarian cancer patients after lymph node resection

作　　者：曹铁凤 黄佳明[1] 肖丽薇 何雨诗 沈慧敏[1] CAO Tiefeng;HUANG Jiaming;XIAO Liwei;HE Yushi;SHEN Huimin(Department of Gynecology,The First Affiliated Hospital,Sun Yat-Sen University,Guangzhou 510080,China)

机构地区：[1]中山大学附属第一医院妇科,广东广州510080

出　　处：《西部医学》2024年第9期1303-1307,1312,共6页Medical Journal of West China

基　　金：国家自然科学基金-青年基金项目(82002734);广东省基础与应用基础研究基金区域联合基金-青年基金项目(2019A1515110312)。

摘　　要：目的分析并筛选系统性淋巴结切除后的早期卵巢癌的独立预后因素,并构建及验证列线图预测早期卵巢癌患者生存。方法从SEER数据库中收集2010—2017年确诊的3906例进行系统性淋巴结切除术的早期卵巢癌患者,获取相关临床信息,按照7∶3比例随机分成训练组2736例及验证组1170例。使用单因素及多因素Cox回归分析对临床信息进行筛选,确定进行系统性淋巴结切除术后的早期卵巢癌患者预后的独立危险因素。并根据独立危险因素绘制列线图构建预后预测模型。采用C-index、曲线下面积(AUC)、校准曲线对预测模型的一致性及精确性进行评价。结果患者年龄>75岁、病理分化Ⅲ或Ⅳ级、肿瘤分期高、粘液性癌是预后的独立危险因素;年龄<60岁、白种人是预后的独立保护因素。应用独立预后因素构建列线图,训练组及验证组中C-index分别为0.72(95%CI为0.69~0.74)和0.71(95%CI为0.67~0.75)。训练组及验证组中1、3、5年预后的AUC分别为0.683,0.738,0.745及0.736,0.754,0.722。校准曲线均显示出一致性良好。结论基于SEER数据库中淋巴结切除术后的早期卵巢癌患者资料确定独立预后因素,所构建列线图具有良好预测效果,有助于临床快速准确评估患者生存预后。Objective To analyze and screen independent prognostic factors for early ovarian cancer after systematic lymph node resection,and construct and validate nomogram to predict survival.Methods We collected 3906 early stage ovarian cancer patients with systematic lymph node resection diagnosed at 2010 to 2017 from the SEER database,and obtained relevant clinical information.All cases were randomly divided into a training group(2736 cases)and a validation group(1170 cases)with a 7∶3 ratio.We used univariate and multivariate Cox regression analysis to screen clinical information and identify independent risk factors for prognosis,and a nomogram was constructed based on independent risk factors.Then the consistency and accuracy of the prediction model was evaluated using C-index,area under curve(AUC),and calibration curve.Results The results of univariate and multivariate analysis showed that patients over 75 years old,grade III or IV,higher stage,and mucinous pathology were independent risk factors for prognosis.Age<60 years old and white population were independent protective factors.A nomogram was constructed by using independent prognostic factors,and the C-index values in the training group and validation group were 0.72(95%CI,0.69-0.74)and 0.71(95%CI,0.67-0.75),respectively.The AUC for 1,3,and 5 year prognosis in the training group and validation group were 0.683,0.738,0.745,and 0.736,0.754,and 0.722,respectively.The calibration curves all showed good consistency.Conclusion Based on the data of early stage ovarian cancer patients after lymph node resection in the SEER database,a nomogram perdition model was constructed.The model had good predictive effect and is helpful in clinical survival and prognosis evaluation rapidly.

关 键 词：早期卵巢癌 淋巴结切除术后 预后 SEER数据库 列线图 

分 类 号：R737.31[医药卫生—肿瘤]