基于血清CEA、SCCA联合多模态MRI构建预测宫颈癌术后复发转移的nomograms模型效能分析  

Efficacy Analysis of Nomograms Model for Predicting Postoperative Recurrence and Metastasis of Cervical Cancer Based on Serum CEA,SCCA Combined with Multimodal MRI

作　　者：李亚玲 刘研 李燕 由霈 李亚静 LI Ya-ling;LIU Yan;LI Yan;YOU Pei;LI Ya-jing(Department of Gynecology,Dingzhou People's Hospital,Dingzhou,Hebei 073000,China)

机构地区：[1]定州市人民医院妇科,河北定州073000

出　　处：《解放军医药杂志》2022年第10期31-36,共6页Medical & Pharmaceutical Journal of Chinese People’s Liberation Army

基　　金：2022年度河北省医学科学研究课题计划(20220438)。

摘　　要：目的基于血清癌胚抗原(CEA)、鳞状细胞癌抗原(SCCA)联合多模态MRI构建预测宫颈癌术后复发转移的nomograms模型,并检验其预测效能。方法选取2017年9月—2019年7月收治的宫颈癌300例,根据术后是否发生复发转移分为发生组、未发生组,比较2组一般资料、多模态MRI参数[表观扩散系数(ADC)、容积转运常数(K^(trans))、速率常数(K_(ep))、血管外细胞外间隙容积比],采用多因素Logistic回归分析探讨宫颈癌术后复发转移的影响因素,采用R语言构建并绘制宫颈癌术后复发转移的nomograms模型,采用受试者工作特征(ROC)曲线、校准曲线验证nomograms模型的预测效能。结果发生组肌层浸润深度≥1/2、盆腔淋巴结转移、脉管癌栓患者比例及CEA、SCCA水平高于未发生组(P<0.01);发生组ADC低于未发生组,K^(trans)、K_(ep)高于未发生组(P<0.01);多因素Logistic回归分析结果显示:肌层浸润深度≥1/2、盆腔淋巴结转移、脉管癌栓、CEA、SCCA、ADC、K^(trans)、K_(ep)是宫颈癌患者术后复发转移的独立危险因素(P<0.01)。绘制预测宫颈癌术后复发转移的nomograms模型图,显示C-index为0.965(95%CI:0.937,0.982),曲线下面积为0.965,且校准曲线贴近标准曲线,提示列线图模型区分度及预测效能均较好。结论肌层浸润深度、盆腔淋巴结转移、脉管癌栓、CEA、SCCA、ADC、K^(trans)、K_(ep)与宫颈癌术后复发转移有关,基于以上各因素构建的nomograms模型呈现出良好的预测效能,可为临床早期预测宫颈癌术后复发转移提供参考,从而指导临床治疗,促进远期预后的改善。Objective To construct nomograms model for predicting postoperative recurrence and metastasis of cervical cancer based on serum carcinoembryonic antigen(CEA)and squamous cell carcinoma antigen(SCCA)combined with multimodal MRI,and to test its predictive efficacy.Methods A total of 300 patients with cervical cancer admitted to our hospital from September 2017 to July 2019 were selected and divided into occurrence group and non-occurrence group according to presence of recurrence and metastasis after surgery.General data and multimodal MRI parameters[apparent diffusion coefficient(ADC),volume transfer constant(K^(trans)),rate constant(K_(ep)),extravascular extracellular volume fraction(V_(e))]of the two groups were compared,and multivariate Logistic regression analysis was used to explore the influencing factors of postoperative recurrence and metastasis of cervical cancer.The nomograms model of postoperative recurrence and metastasis of cervical cancer was constructed and drawn by R language,and the predictive value of nomograms model was verified by receiver operating characteristic(ROC)curve and calibration curve.Results The proportion of patients with muscular layer infiltration depth≥1/2,pelvic lymph node metastasis and vascular cancer thrombus,and CEA and SCCA levels were higher in the occurrence group than in the non-occurrence group(P<0.01).ADC in the occurrence group was lower than that in the non-occurrence group,while K^(trans) and K_(ep) were higher than those in the non-occurrence group(P<0.01).Multivariate Logistic regression analysis showed that muscle infiltration depth≥1/2,pelvic lymph node metastasis,vascular cancer thrombus,CEA,SCCA,ADC,K^(trans) and K_(ep) were independent risk factors for postoperative recurrence and metastasis of cervical cancer patients(P<0.01).Nomograms models of postoperative recurrence and metastasis of cervical cancer showed that C-index was 0.965(95%CI:0.937,0.982),the area under the ROC curve was 0.965;the calibration curve was close to the standard curve,indicating

关 键 词：宫颈肿瘤 癌胚抗原 鳞状细胞癌抗原 多模态MRI nomograms模型 诊断价值 

分 类 号：R737.33[医药卫生—肿瘤]