卵巢癌术后患者Cox比例风险模型的构建与验证

Construction and validation of Cox proportional hazards model for patients with ovarian cancer after surgery

目的基于常规临床资料构建卵巢癌术后患者预后模型,初步探讨影响生存的因素,为预后评估提供参考。方法回顾性分析293例卵巢癌术后患者的临床资料及随访结局,通过Cox比例风险回归法构建预后模型,采用受试者工作特征(ROC)曲线评价模型的预测效能,基于模型分值的P_(25)和P_(75)为界点将具有完整随访结局的患者分为低、中、高风险组,Kaplan-Meier法进行生存分析以验证模型的预后分类能力。结果(1)多因素Cox回归分析显示:年龄、FIGO分期、肿瘤直径、淋巴结转移、远处转移、腹水、术前PLT与患者生存预后独立相关(P<0.05)。以上述指标构建Cox比例风险模型,计算公式:风险指数(risk index,RI)=0.452年龄(50-60)岁+0.799年龄(>60岁)+0.297FIGO分期+0.281淋巴结转移+0.774远处转移+0.473腹水+0.239肿瘤直径+0.402术前PLT(≥300×10^(9)/L)。(2)ROC分析显示:RI预测3年死亡的ROC曲线下面积(AUC)是0.71(95%CI:0.646-0.774),以RI=1.869为截断点(Youden指数最大),预测的敏感性和特异性分别0.754和0.589。RI预测5年死亡的AUC是0.753(95%CI:0.679-0.827),当RI=1.767时,敏感性为0.737,特异性为0.750。(3)生存分析显示,低风险组、中风险组和高风险组的生存曲线无交叉平行下移,三组中位生存时间分别为44(95%CI:35.433-52.567)个月、36(95%CI:29.745-42.255)个月和23(95%CI:13.171-32.829)个月,累积生存率有显著统计学差异(χ^(2)=23.762,P=0.000)。结论本研究构建的Cox比例风险模型对卵巢癌术后患者预后预测具有良好的效能,有助于卵巢癌患者预后分层和高风险人群的识别。

Objective To construct a prognostic model for patients with ovarian cancer after surgery based on routine clinical data,preliminarily explore the factors affecting survival,and provide reference for prognosis evaluation.Methods Clinical data and follow-up outcomes of 293 patients with ovarian cancer after operation were retrospectively analyzed.Cox proportional hazards regression method was used to construct a prognostic model,and receiver operating characteristic(ROC)curve was used to evaluate the predictive efficacy of the model.Based on the P25 and P75 of the model score,the patients with complete follow-up outcomes were divided into low,medium and high risk groups.The Kaplan-Meier method was used for survival analysis to verify the prognostic classification ability of the model.Results Multivariate Cox regression analysis showed that age,FIGO stage,tumor diameter,lymph node metastasis,distant metastasis,ascites and preoperative PLT were independently associated with survival prognosis(P<0.05).The Cox proportional hazards model was constructed based on the above indicators,and the formula was calculated as follows:risk index(risk index,RI)=0.452 age(50-60)years+0.799 age(>60 years)+0.297FIGO stage+0.281 lymph node metastasis+0.774 distant metastasis+0.473 ascites+0.239 tumor diameter+0.402 preoperative PLT(≥300×10^(9)/L).ROC analysis showed that the area under the ROC curve(AUC)of RI for predicting 3-year mortality was 0.71(95%CI:0.646-0.774).With RI=1.869 as the cut-off point(maximum Youden index),the sensitivity and specificity for predicting 3-year mortality were 0.754 and 0.589,respectively.The AUC of RI for predicting 5-year mortality was 0.753(95%CI:0.679-0.827).When RI=1.767,the sensitivity was 0.737,and the specificity was 0.750.Survival analysis showed that the survival curves of low risk group,medium risk group and high risk group did not cross and parallel downward shift,and the median survival time of the three groups was 44(95%CI:35.433-52.567)months,36(95%CI:29.745-42.255)months and 23(95%CI:29.745-42.255)months,respectively.13.171-32.829)months,and the cumulative survival rate was significantly different(χ^(2)=23.762,P=0.000).Conclusions The Cox proportional hazard model established in this study has good efficacy in predicting the prognosis of patients with ovarian cancer after surgery,which is helpful for the prognosis stratification of patients with ovarian cancer and the identification of high-risk groups.