Wang Xue, Zhao Yu, Hu Yumin, Zhou Yongjin, Ye Xinjian, Liu Kun, Bai Guanghui, Guo Anna, Du Meimei, Jiang Lezhen, Wang Jinhong, Yan Zhihan
Department of Radiology, The Second Affiliated Hospital and Yuying Children's Hospital of Wenzhou Medical University, Wenzhou 325027, China.
Department of Gynecology and Obstetrics, The Second Affiliated Hospital and Yuying Children's Hospital of Wenzhou Medical University, Wenzhou 325027, China.
Oncotarget. 2017 Jul 11;8(28):46390-46397. doi: 10.18632/oncotarget.18553.
Previous researchers obtained various apparent diffusion coefficient (ADC) cutoff values to differentiate endometrial carcinoma from benign mimickers with 1.5T magnetic resonance imaging (MRI). Few studies have used 3T MRI or validated the effectiveness of these cutoff ADC values prospectively. This study was designed in two stages to obtain a cutoff ADC value at 3T MRI and to validate prospectively the role of the ADC value. First, we conducted a retrospective study of 60 patients to evaluate the diagnostic value of ADC by obtain a theoretical cutoff ADC value for differentiating between benign and malignant endometrial lesions. Student's t test revealed that ADC values for stage I endometrial carcinomas were significantly lower than those for benign lesions. The area under the curve value of the receiver operating characteristic curve was 0.993, and the cutoff ADC value was 0.98 × 10-3 mm2/s. The sensitivity, specificity, and overall accuracy of diagnosing stage I endometrial carcinoma were 100%, 97.1%, and 98.3%, respectively. Second, we conducted a prospective study of 26 patients to validate the use of the cutoff ADC value obtained in the study's first stage. The sensitivity, specificity, and overall accuracy for differentiating malignant from benign endometrial lesions based on the cutoff ADC value obtained earlier were as follows: radiologist 1 attained 86.67%, 100.0%, and 92.31%, respectively; radiologist 2 attained 86.67%, 91.0%, and 88.5%, respectively. Our results suggest that ADC values could be a potential biomarker for use as a quantitative and qualitative tool for differentiating between early-stage endometrial carcinomas and benign mimickers.
以往的研究人员通过1.5T磁共振成像（MRI）获得了各种表观扩散系数（ADC）临界值，以鉴别子宫内膜癌与良性类似病变。很少有研究使用3T MRI或前瞻性地验证这些临界ADC值的有效性。本研究分两个阶段进行，以获得3T MRI时的临界ADC值，并前瞻性地验证ADC值的作用。首先，我们对60例患者进行了回顾性研究，通过获得区分良性和恶性子宫内膜病变的理论临界ADC值来评估ADC的诊断价值。学生t检验显示，I期子宫内膜癌的ADC值显著低于良性病变。受试者操作特征曲线的曲线下面积值为0.993，临界ADC值为0.98×10-3 mm2/s。诊断I期子宫内膜癌的敏感性、特异性和总体准确率分别为100%、97.1%和98.3%。其次，我们对26例患者进行了前瞻性研究，以验证在研究第一阶段获得的临界ADC值的应用。根据早期获得的临界ADC值区分恶性和良性子宫内膜病变的敏感性、特异性和总体准确率如下：放射科医生1分别达到86.67%、100.0%和92.31%；放射科医生2分别达到86.67%、91.0%和88.5%。我们的结果表明，ADC值可能是一种潜在的生物标志物，可作为区分早期子宫内膜癌和良性类似病变的定量和定性工具。
