基于SEER数据库Gleason评分7分前列腺癌临床预测模型的建立及验证

Development and validation of clinical prediction model for prostate cancer patients with Gleason score 7 based on SEER database

目的建立并验证针对Gleason评分(Gleason score,GS)7分前列腺癌(PCa)患者生存预后的临床预测模型。方法对SEER数据库中2010-2016年间初次诊断为GS 7分PCa患者的信息进行回顾性检索,并被随机分配到训练集(n=58405)和验证集(n=24858)。采用R Studio软件对所有变量进行单因素COX回归分析,并将P<0.05的变量纳入多因素COX回归分析中,使用R Studio软件将多因素COX回归分析中各项独立预后危险因素构建成Nomogram模型,采用C指数(C-index)和绘制Nomogram校准曲线来验证模型的准确性和判别能力。结果成功建立了基于年龄、种族、原发部位及肿瘤分期(TNM)、手术方式、是否放疗、婚姻状态等预后因素的Nomogram,C-index为0.735(95%CI 0.733~0.737),校准曲线与理想曲线一致性较高,经内部验证后具有较高的准确性与适用性。结论本研究通过SEER数据库确定了前列腺癌患者预后的独立危险因素,并且建立了个体化预测PCa患者生存预后的Nomogram。

Objective To establish and verify the clinical prediction model for prostate cancer patients with Gleason score 7.Methods A retrospective search was conducted on the information of prostate cancer patients with Gleason 7 diagnosed for the first time in the SEER database from 2010 to 2016.They were randomly assigned to the training set(n=58405)and the verification set(n=24858).Univariate Cox regression analysis was performed on all variables with R Studio software,and the variables with P<0.05 were included in the multivariate Cox regression analysis.The independent prognostic risk factors in multivariate COX regression analysis were constructed into Nomogram model by R Studio software.The accuracy and the discriminant ability of the model were verified by C-index and Nomogram calibration curve.Results Nomogram was established based on age,race,primary location,tumor stage,operation,radiotherapy,marital status and other prognostic factors.The C-index of the Nomogram for predicting the overall survival was 0.735(95%CI 0.733-0.737).The calibration curve also demonstrated an excellent agreement between prediction and observation for the overall survival probability of the patients with prostate cancer.Conclusion The study based on the SEER database identifies independent risk factors for the prognosis of prostate cancer,and the Nomogram which can individually predict the prognosis of prostate cancer is successfully established.