含DCE-MRI、DWI定量参数及血清CEA水平的列线图模型对直肠癌术后复发转移风险的预测价值  

Predictive value of a nomogram based on DCE-MRI and DWI quantitative parameters and serum CEA level for risk of postoperative recurrence/metastasis of rectal cancer

作　　者：董伍真 倪浩亮 蔡成[1] Wu-Zhen Dong;Hao-Liang Ni;Cheng Cai(Jinhua Central Hospital,Jinhua 321000,Zhejiang Province,China)

机构地区：[1]金华市中心医院,浙江省金华市321000

出　　处：《世界华人消化杂志》2023年第18期773-781,共9页World Chinese Journal of Digestology

摘　　要：背景直肠癌预后较差,早期预测直肠癌根治性手术后复发转移对改善预后具有重要意义.本研究同时整合多模态磁共振成像(magnetic resonance imaging,MRI)技术相关参数、肿瘤标志物、临床病理特征等多种影响因素,并建立列线图模型,为临床制定干预措施提供依据.目的探讨基于磁共振动态增强(dynamic contrast enhanced MRI,DCE-MRI)、磁共振扩散加权成像(diffusion-weighted imaging,DWI)定量参数及血清癌胚抗原(carcinoembryonic antigen,CEA)水平构建列线图模型,分析该模型对直肠癌术后复发转移风险的预测价值,以指导临床制定干预措施.方法选取2019-03-01/2022-02-28于本院行腹腔镜辅助下直肠癌根治术的120例患者直肠癌为研究对象,根据患者术后1年内复发转移情况分为复发转移组(n=29),无复发转移组(n=91).对比两组3.0T MRI多模态成像技术相关参数[表观扩散系数(apparent diffusion coefficient,ADC)、血管通透性常数(Transfer rate constant,Ktrans)、血液回流常数(Rate constant,Kep)、血管外细胞外间隙容积百分比(extravascular extracellular space volume fraction,Ve)].单因素、Lasso模型筛选术后复发转移相关预测因子,采用Logistic回归分析术后复发转移的影响因素.基于影响因素构建列线图模型,评价该模型对术后复发转移的预测价值,校准曲线、决策曲线分析(decision curve analysis,DCA)验证该模型校准度、临床效用性.结果复发转移组ADC低于无复发转移组,Ktrans、Kep高于无复发转移组(P<0.05);梗阻、分化程度、临床分期、淋巴结转移、术后CEA及ADC、Ktrans、Kep为术后复发转移的独立影响因素(P<0.05);列线图模型预测术后复发转移的AUC高于ADC、Ktrans、Kep联合预测(P<0.05),且具有良好校准度、临床效用性.结论基于DCE-MRI、DWI定量参数及血清CEA水平构建列线图模型对直肠癌术后复发转移具有一定预测价值,临床可针对高危因素制定干预措施BACKGROUND The prognosis of rectal cancer is poor,and early prediction of recurrence and metastasis after radical surgery is of great significance for improving its prognosis.This study integrated multiple influencing factors such as multimodal magnetic resonance imaging(MRI)parameters,tumor markers,and clinicopathological features to develop a nomogram to provide a basis for the development of clinical intervention measures for this malignancy.AIM To develop a nomogram based on dynamic contrast enhanced MRI(DCE-MRI)and diffusion-weighted imaging(DWI)quantitative parameters and serum carcinoembryonic antigen(CEA)level,and to analyze the predictive value of this model for the risk of postoperative recurrence and metastasis of rectal cancer,so as to guide the development of clinical intervention measures for this malignancy.METHODS A total of 120 patients who underwent laparoscopic-assisted radical resection of rectal cancer at our hospital from March 1,2019 to February 28,2022 were selected as research subjects.According to the presence of recurrence/metastasis within 1 year after surgery,the patients were divided into a recurrence/metastasis group(n=29)and a no recurrence/metastasis group(n=91).The relevant parameters[apparent diffusion coefficient(ADC),transfer rate constant(Ktrans),blood return constant(Kep),and extravascular extracellular space volume fraction(Ve)]of multimodal MRI imaging techniques were compared between the two groups to analyze their predictive value for postoperative recurrence/metastasis.Univariate analysis with Lasso model screening for predictive factors related to postoperative recurrence/metastasis was performed,and logistic regression analysis was used to analyze the influencing factors of postoperative recurrence/metastasis.A nomogram was developed based on the influencing factors identified,and the predictive value of the model for postoperative recurrence/metastasis was assessed.Calibration curve and decision curve analysis(DCA)were used to verify the calibration degree and clinic

关 键 词：直肠癌 MRI 多模态成像技术 复发 转移 列线图 预测 

分 类 号：R735.34[医药卫生—肿瘤]