基于CT⁃PCI与临床指标构建预测晚期上皮性卵巢癌减瘤术结局的列线图模型  

Construction of Nomogram Model to Predict the Outcome of Debulking Surgery for Advanced Epithelial Ovarian Cancer Based on CT Peritoneal Cancer Index and Clinical Indicators

作　　者：赵玲[1,2] 刘历 王海燕 方元 李咏梅 ZHAO Ling;LIU Li;WANG Haiyan;FANG Yuan;LI Yongmei(Department of Radiology,The First Affiliated Hospital of Chongqing Medical University;Department of Radiology,The People’s Hospital of Yubei District of Chongqing)

机构地区：[1]重庆医科大学附属第一医院放射科 [2]重庆市渝北区人民医院放射科

出　　处：《中国医学计算机成像杂志》2023年第3期283-289,共7页Chinese Computed Medical Imaging

摘　　要：目的:基于术前CT腹膜癌指数(CT⁃PCI)、血清人附睾蛋白4(HE4)、淋巴细胞与单核细胞比值(LMR)等指标构建列线图预测模型,探讨其在晚期上皮性卵巢癌初次减瘤术结局中的预测价值。方法:根据初次减瘤术结果将患者分为满意减瘤术组和不满意减瘤术组,采用受试者工作特征(ROC)曲线分析评估术前CT⁃PCI、血清糖类抗原125(CA125)、HE4及LMR等指标单独对减瘤术结局的预测价值,使用二元logistic回归进行多因素分析,并构建预测晚期上皮性卵巢癌不满意减瘤术的列线图模型。结果:⑴不满意减瘤术组的术前CT⁃PCI、血清CA125、HE4、年龄≥60岁比率及美国麻醉医师协会(ASA)3~4级比率显著高于满意减瘤术组,LMR显著低于满意减瘤组(P<0.05)。⑵CT⁃PCI、血清CA125、HE4与LMR预测减瘤术结局的最佳临界值分别为7.5分、530 U/mL、311 pmol/L及3.0,ROC曲线下面积(AUC)分别为0.785、0.685、0.711及0.689,灵敏度分别为74.6%、79.7%、74.6%及74.6%,特异度分别为70.6%、55.9%、64.7%及61.8%。⑶在多因素分析中,年龄≥60岁、CT⁃PCI、HE4及LMR是不满意减瘤术的独立影响因素,基于四者构建的列线图模型的最佳临界值为167分,预测不满意减瘤术的AUC为0.878,灵敏度及特异度分别为78.0%、85.3%。结论:术前CT⁃PCI、血清HE4、LMR及高龄是晚期上皮性卵巢癌减瘤术结局的独立影响因素,基于四者构建的列线图模型对不满意减瘤术有较高的预测价值。Purpose:To construct a nomogram prediction model based on preoperative CT peritoneal cancer index(CT-PCI),serum human epididymis protein 4(HE4),lymphocyte to monocyte ratio(LMR)and other indicators,and to explore its predictive value in the outcome of primary debulking surgery for advanced epithelial ovarian cancer.Methods:According to the results of primary debulking surgery,patients were divided into optimal debulking surgery group and suboptimal debulking surgery group.Receiver operating characteristic(ROC)curve analysis was used to evaluate the predictive value of preoperative CT-PCI,serum carbohydrate antigen(CA125),HE4 and LMR on the outcome of debulking surgery.Multivariate logistic regression was used to analyze and construct a nomogram model for predicting suboptimal debulking surgery in advanced epithelial ovarian cancer.Results:⑴The preoperative CT-PCI,serum CA125,HE4,the percentages of age≥60 years and American Society of Anesthesiologists(ASA)3-4 in the suboptimal debulking surgery group were significantly higher than those in the optimal debulking surgery group,and the LMR was significantly lower than that in the optimal debulking surgery group(P<0.05).⑵The optimal cut-off values of CT-PCI,serum CA125,HE4 and LMR for predicting the outcome of suboptimal debulking surgery were 7.5 points,530 U/mL,311 pmol/L and 3.0,respectively.The areas under ROC curve(AUCs)were 0.785,0.685,0.711 and 0.689,respectively.The sensitivities were 74.6%,79.7%,74.6%and 74.6%,and the specificities were 70.6%,55.9%,64.7%and 61.8%,respectively.⑶In multivariate analysis,age≥60 years old,CT-PCI,HE4 and LMR were the independent influencing factors of suboptimal debulking surgery.The optimal cut-off value of the nomogram model constructed based on the four factors was 167 points.The AUC of predicting the outcome of suboptimal debulking surgery was 0.878,and the sensitivity and specificity were 78.0%and 85.3%respectively.Conclusion:Preoperative CT-PCI,serum HE4,LMR and old age are independent influencing factors for the

关 键 词：上皮性卵巢癌 进展期 肿瘤细胞减灭术 计算机体层成像 腹膜癌指数 列线图模型 

分 类 号：R445.3[医药卫生—影像医学与核医学]