基于MRI特征构建列线图预测乳腺癌Ki-67表达状态  

Nomogram to Preoperatively Predict Ki-67 Status in Breast Cancer Based on MRI Features

作　　者：刘琼[1] 张慈慈 段丽霞[1] 欧志强[1] LIU Qiong;ZHANG Ci-ci;DUAN Li-xia;OU Zhi-qiang(Department of Radiology,Guangzhou Red Cross Hospital of Jinan University,Guangzhou 510220,Guangdong Province,China)

机构地区：[1]暨南大学附属广州红十字会医院放射科,广东广州510220

出　　处：《中国CT和MRI杂志》2024年第10期79-81,共3页Chinese Journal of CT and MRI

基　　金：广东省中医局中医科研项目(20242070)。

摘　　要：目的 探讨基于MRI特征构建列线图对预测乳腺癌Ki-67表达状态的价值。方法回顾性分析2010年1月至2022年12月我院经病理证实355例乳腺癌的术前MRI特征,以Ki-67>14%为高表达组(183例),Ki-67≤14%为低表达组(172例)。采用T检验、Wilcoxon秩和检验或卡方检验比较两组间MRI特征的差异,对差异有统计学意义的指标进行Logistic回归分析获取Ki-67表达的预测因素,采用R软件建立列线图模型。结果肿瘤直径、同侧腋窝淋巴结直径、背景腺体强化、肿瘤形态、瘤周水肿、邻近血管增多征、时间信号曲线在两组之间有统计学差异(所有P<0.05),Logistic回归分析显示,肿瘤直径、瘤周水肿、血管增多征是Ki-67高表达的预测因素(所有P<0.05),列线图模型预测Ki-67的曲线下面积为0.794。结论基于MRI特征建立列线图模型可有效预测乳腺癌Ki-67表达状态,为术前评估乳腺癌Ki-67表达状态提供了有效的辅助诊断方法。Mective To explore the predictive value of a nomogra m based on MRI features for Ki-67 expression in patients with breast cancer.Methods MRI features of 355 patients with breast cancer in our hospital from January 2010 to December 2022 with pathologically confirmed diagnoses were retrospectively analyzed.The patients were classified into the Ki-67 high expression group(Ki-67> 14%,183 cases) and the low expression group(Ki-67 ≤14%,172 cases).T-test,Wilcoxon rank-sum test,or chi-square test were used to compare the differences in MRI features between the two groups,and logistic regression analysis was performed on the indicators with statistical significance to obtain independent predictive factors for Ki-67 expression.The nomogram model was established using R software.Results There were significa nt differences in tumor diameter,ipsilateral axillary lymph node diameter,background pa renchymal enhancement,tumor shape,peritumoral edema,adjacent vascular hyperplasia sign,and time intensity curve between the two groups(all P<0.05).Logistic regression analysis showed that tumor diameter,peritumoral edema,and adjacent vascular hyperplasia sign were independent risk factors for high Ki-67 expression.Based on these three risk factors,a nomogra m prediction model was established,with an area under the ROC curve(AUC) of 0.794.Conclusion The nomogram based on M RI features is helpful to predict Ki-67 expression in breast cancer,and provides an effective auxilia ry diagnostic method for evaluation of Ki-67 expression in breast cancer before surgery.

关 键 词：乳腺癌 KI-67 磁共振成像 列线图 

分 类 号：R445.2[医药卫生—影像医学与核医学]