IVIM-DWI联合DCE-MRI对乳腺癌Nottingham指数的预测价值

Prediction of the Nottingham prognostic index in breast cancer based on intravoxel incoherent motion diffusion weighted imaging and dynamic contrast enhancement imaging

目的:探讨体素内不相干运动扩散加权成像(IVIM-DWI)联合动态增强磁共振成像(DCE-MRI)对乳腺癌Nottingham指数的预测价值。方法:搜集2019年10月-2022年6月在本院经病理证实的166例乳腺癌患者的临床和影像资料。所有患者有完整的免疫组化检查资料,包括雌激素受体(ER)、孕激素受体(PR)、人类表皮生长因子受体-2(HER-2)和细胞增殖抗原标记物(Ki-67);同时获取Nottingham指数(NPI)值,根据NPI值,将患者分为2组:预后良好组(NPI<3.4为)和预后不良组(NPI≥3.4)。所有病例术前行IVIM-DWI及DCE-MRI检查,分别获得病灶的真实扩散系数(D)、灌注相关扩散系数(D^(*))、灌注分数(f)以及容量转移常数(Ktrans)、血管外细胞外间隙容积比(Ve)和速率常数(Kep)共6项定量参数值。比较ER阴性组与阳性组、PR阴性组与阳性组、HER-2阴性组与阳性组、Ki-67高表达组与低表达组、预后良好组与预后不良组之间6项定量参数值的差异。将NPI不同表达状态之间有统计学差异的IVIM-DWI及DCE-MRI定量参数分别纳入二元Logistic回归分析,获得对预后有鉴别诊断价值的独立影响因素,并分别构建IVIM-DWI、DCE-MRI及IVIM-DWI+DCE-MRI预测模型。采用受试者工作特征曲线(ROC)及曲线下面积(AUC)判断各模型对NPI的诊断效能,采用Delong检验比较各模型AUC的差异。结果:ER阴性组的D^(*)、f和Ktrans值显著高于阳性组(P<0.05);PR阴性组的D^(*)值显著高于阳性组(P<0.05);HER-2阳性组的Ktrans和Ve值大于阴性组(P<0.05);Ki-67高表达组的D^(*)及Ktrans、Kep值大于低表达组,而D值小于低表达组(P<0.05);预后不良组的D^(*)及Ktrans、Kep值大于预后良好组,而D值小于预后良好组(P<0.05)。D、Ktrans及Kep为鉴别NPI表达水平的独立影响因素,其中D值的AUC较大,为0.852。将上述独立影响因素建模,其中IVIM-DWI模型中仅有D值,而DCE-MRI模型为Ktrans+Kep;将IVIM-DWI和DCE-MRI模型联合后,AUC为0.888,诊断效能高于单一模型(Z=1.963,P=0.049;Z=3.418,P=0.006),相应的诊断敏感度为87.5%、特异度为87.3%、符合率为82.5%。结论:IVIM-DWI联合DCE-MRI可用于乳腺癌NPI的预测,有利于患者的个体化治疗。

Objective:To investigate the predictive value of intravoxel incoherent motion diffusion weighted imaging(IVIM-DWI)and dynamic contrast-enhanced magnetic resonance imaging(DCE-MRI)for the Nottingham prognostic index in breast cancer.Methods:A total of 166 patients with breast cancer confirmed by pathology were collected from October 2019 to June 2022.All patients had complete immunohistochemical examination information,including estrogen receptor(ER),progesterone receptor(PR),human epidermal growth factor receptor-2(HER-2),and cell proliferation antigen marker(Ki-67).Based on the Nottingham prognostic index(NPI)value,NPI score<3.4 was considered good prognosis,and NPI score≥3.4 was considered poor prognosis.IVIM-DWI and DCE-MRI scanning were performed in all patients.True diffusion coefficient(D),perfusion-related diffusion coefficient(D^(*)),perfusion fraction(f),volume transfer constant(K trans),extravascular extracellular space distribute volume per unit tissue volume(V e)and rate constant(K ep)were analyzed.The quantitative parameters of MRI were compared between ER negative group and positive group,PR negative group and positive group,HER-2 negative group and positive group,Ki-67 high expression group and low expression group,good prognosis group(NPI<3.4)and poor prognosis group(NPI≥3.4).The quantitative parameters of IVIM-DWI and DCE-MRI with statistical differences were placed into the binary Logistic regression equation to obtain independent influencing factors with differential diagnosis value for NPI.The IVIM-DWI,DCE-MRI and IVIM-DWI+DCE-MRI models were constructed based on the above independent factors.The receiver operating characteristic curve(ROC)and area under curve(AUC)to were used to evaluate the differential diagnostic efficacy of various models for NPI,and Delong test was used to compare the differences in AUCs.Results:D^(*),f and K trans values in the ER negative group were higher than that of positive group(P<0.05).The HER-2 positive group had higher K trans and V e values than that of the negative group(P<0.05).D^(*)and K trans and K ep values were higher in the Ki-67 high expression group than those of the lower expression group,while D was smaller than that of the low expression group(P<0.05).D^(*)and K trans and K ep values in high-grade Nottingham group were higher than those of the low-grade Nottingham group and D was less than the low-grade Nottingham group(P<0.05).D,K trans and K ep were independent factors to identify the Nottingham index.The D value showed a larger AUC of 0.852.Combining the IVIM-DWI and DCE-MRI models,the AUC was 0.888,and the diagnostic efficacy was higher than that of single model(Z=1.963,P=0.049;Z=3.418,P=0.006),the sensitivity was 87.5%,the specificity was 87.3%,and the accuracy was 82.5%.Conclusion:IVIM-DWI combined with DCE-MRI can be used to predict NPI of breast cancer,which is beneficial to individulized treatment of patients.