基于多参数MRI影像结合临床病理特征预测乳腺癌肿瘤浸润淋巴细胞水平的研究  

Prediction of Tumour-Infiltrating Lymphocytes Levels in Breast Cancer Based on Multi-Parameter MRI Combining with Clinicopathological Features

作　　者：隋艺 莫蕾[1] 陈春雅 张琼琼 李雪莉 胡闻珂 余晓蒙 陈思义 唐文洁[1] 魏新华[1] 郭媛[1] SUI Yi;MO Lei;CHEN Chunya(the Second Affiliated Hospital of South China University,Guangzhou,Guangdong Province 510180,P.R.China)

机构地区：[1]广州市第一人民医院(华南理工附属第二医院)放射科,广州510180 [2]广州市第一人民医院(华南理工附属第二医院)病理科,广州510180

出　　处：《临床放射学杂志》2024年第6期917-923,共7页Journal of Clinical Radiology

基　　金：国家自然科学基金项目(编号:81901711)。

摘　　要：目的 探讨基于多参数MRI影像结合临床病理的列线图预测乳腺癌肿瘤浸润淋巴细胞(TILs)水平。方法 回顾性研究222例经手术病理证实的乳腺癌患者,评估患者MRI影像特征和临床病理特征。在光学显微镜下评估TIL比例,TIL<10%为低TIL水平,TIL≥10%为高TIL水平。基于多因素Logistic回归分析分别构建临床病理预测模型、MRI预测模型、临床病理&MRI联合预测模型,使用受试者工作特征ROC曲线下面积AUC评估模型的预测效能,使用DeLong检验对模型进行比较。选取临床病理&MRI联合预测模型构建可视化列线图,使用校准曲线对列线图进行评价。结果 共纳入222例乳腺癌患者,训练集156例,验证集66例。训练集中高TIL水平组71例(45.5%)、低TIL水平组85例(54.5%)。其中内部强化特征(P=0.004)、多灶/多中心(P=0.018)、ADC值(P=0.001)、Ki-67指数(P=0.001)和分子亚型(P=0.010)与TIL水平相关。在训练集与验证集中,临床病理&MRI联合模型性能最优,AUC值分别为0.814、0.790,使用列线图可视化该联合预测模型。结论 基于多参数MRI影像结合临床病理特征的列线图可以预测乳腺癌TIL水平,为临床提供准确、全面、高效的评估手段。Objective To explore the nomogram based on multi-parameter MRI and clinicopathological characteristics to predict Tumour-infiltrating lymphocytes(TILs)level in breast cancer.Methods A retrospective study of 222 patients with surgically and pathologically confirmed breast cancer,MRI features and clinicopathologic features of patients were evaluated.TIL<10%was considered to be low TIL levels and TIL≥10%was considered to be high TIL levels.The clinicopathological prediction model,MRI prediction model and clinicopathic-MRI combined prediction model were constructed based on multivariate logistic regression analysis.The(AUC)area under receiver operating characteristic(ROC)curve was used to evaluate the prediction efficiency of the models,and the delong test was used to compare the models.A combined clinicopathological and MRI prediction model was selected to construct a visual nomogram,and the calibration curve was used to evaluate the nomogram.Results A total of 222 patients with breast cancer were enrolled,including 156 in the training group and 66 in the validation group.In the training group,85 cases(54.5%)in the low TIL levels group and 71 cases(45.5%)in the high TIL levels group.The internal enhancement features(P=0.004),multifocal/multicentric(P=0.018),ADC value of tumor(P=0.001),Ki-67 index(P=0.001),and molecular subtype(P=0.010)were associated with TIL level.In the training set and validation set,the combined clinicopathological and MRI model had the best per-formance,with AUC values of 0.814 and 0.790,respectively.The combined prediction model was visualized using a nomogram.Conclusion The nomogram based on multi-parameter MRI combining with clinicopathological features could pre-dict TIL levels of breast cancer,which providing an accurate,comprehensive and efficient evaluation method for clinical pracice.

关 键 词：乳腺癌 磁共振成像 肿瘤浸润淋巴细胞 列线图 

分 类 号：R737.9[医药卫生—肿瘤] R445.2[医药卫生—临床医学]