Ⅰ期子宫内膜癌预后简易评分系统的建立
Establishment of a simple prognostic scoring system for stage Ⅰ endometrial carcinoma
目的 构建Ⅰ期子宫内膜癌预后简易评分系统为临床提供参考工具。 方法 回顾性收集2014年至2021年于重庆医科大学附属第一医院接受了手术治疗的Ⅰ期子宫内膜癌患者的病例资料,通过单因素及多因素Cox回归分析筛选与患者复发及死亡相关的影响因子。基于多因素分析结果,构建简易评分系统,利用受试者工作特征(receiver operating characteristic,ROC)曲线及其曲线下面积(area under the curve,AUC)比较评分系统和单一预测因子对患者预后的预测准确度。利用X-title软件确定评分系统的截值点并对患者进行风险分组,比较不同分组患者预后分布情况。 结果 多因素分析结果提示,患者年龄(复发:Wald χ2 =4.482,P=0.034;死亡:Wald χ2 =8.401,P=0.004)、病理分型(复发:Wald χ2 =6.692,P=0.010;死亡:Wald χ2 =4.222,P=0.040)、肌层浸润情况(复发:Wald χ2 =6.372,P=0.034;死亡:Waldχ2 =7.183,P=0.007)、淋巴脉管间隙侵犯(lymphovascular space invasion,LVSI)(复发:Wald χ2 =13.396,P<0.001;死亡:Wald χ2 =8.449,P=0.004)、雌激素受体(estrogen receptor,ER)表达(复发:Wald χ2 =7.679,P=0.006;死亡:Wald χ2 =8.532,P=0.003)及P53表达(复发:Wald χ2 =7.886,P=0.005;死亡:Wald χ2 =7.174,P=0.007)均是患者复发及死亡的影响因素。基于以上6个因素成功构建评估患者预后的简易评分系统,ROC曲线显示评分系统的AUC值大于单一预测因子。X-title软件确定了评分系统的截值点并自动将患者分为低风险组(0~1分)、中风险组(2~3分)及高风险组(4~6分),3组患者之间总体预后水平差异有统计学意义(P<0.001)。 结论 本研究构建的简易评分系统对Ⅰ期子宫内膜癌患者预后评估有一定价值。
Objective To establish a simple prognostic scoring system for stage Ⅰ endometrial carcinoma,and to provide a reference tool for clinical practice. Methods A retrospective analysis was performed for the case data of the patients with stage Ⅰ endometrial carcinoma who received surgical treatment in The First Affiliated Hospital of Chongqing Medical University from 2014 to 2021,and univariate and multivariate Cox regression analyses were used to identify the influencing factors for recurrence and death. A simple scoring system was established based on the results of the multivariate analysis,and the receiver operating characteristic(ROC)curve and the area under the ROC curve(AUC)were used to compare the accuracy of the scoring system versus each single predictive factor in predicting the prognosis of patients. X-title software was used to determine the cut-off value of the scoring system and divide the patients into different risk groups,and then the prognosis distribution of patients was compared between groups. Results The multivariate analysis showed that age(recurrence:Wald χ2 =4.482,P=0.034;death:Wald χ2=8.401,P=0.004),pathological type(recurrence:Wald χ2 =6.692,P=0.010;death:Wald χ2 =4.222,P=0.040),myometrial invasion(recurrence:Wald χ2 =6.372,P=0.034;death:Wald χ2 =7.183,P=0.007),lymphovascular space invasion(recurrence:Wald χ2 =13.396,P<0.001;death:Wald χ2 =8.449,P=0.004),the expression of estrogen receptor(recurrence:Wald χ2 =7.679,P=0.006;death:Wald χ2 =8.532,P=0.003),and P53 expression(recurrence:Wald χ2 =7.886,P=0.005;death:Wald χ2 =7.174,P=0.007)were influencing factors for recurrence and death. A simple scoring system was successfully established for the prognosis of patients based on the above six factors,and the ROC curve analysis showed that the scoring system had a larger AUC than each predictive factor alone. The cut-off value of the scoring system was determined by X-title software,and the patients were divided into low-risk group(0-1 points),intermediate-risk group(2-3 points),and high-risk group(4-6 points); there was a significant difference in overall prognosis between the three groups(P<0.001). Conclusion The simple scoring system established in this study has a certain value in assessing the prognosis of patients with stage I endometrial carcinoma.
stage I / endometrial carcinoma / prognosis / scoring system
/
〈 |
|
〉 |