术前多模态超声对乳腺癌术后复发风险的预测价值

Predictive value of preoperative multimodal ultrasound for postoperative recurrence risk of breast cancer

目的 分析术前多模态超声特征与乳腺癌术后复发风险的相关性,并探索其预测价值。方法 选取2020年1月至2023年1月山东省菏泽市牡丹人民医院进行手术的286例(286个病灶)女性乳腺癌患者,按照7∶3的比例划分为训练集(200例)和验证集(86例)。参照《中国抗癌协会乳腺癌诊治指南与规范2021版》将训练集患者分为高风险组(68例)与低-中风险组(132例)。分析乳腺癌术后复发风险的影响因素并建立列线图预测模型。采用受试者操作特征曲线和校准曲线评估模型效能。结果 训练集与验证集一般资料比较,差异无统计学意义(P>0.05)。高风险组与低-中风险组TNM分期、病理分级、淋巴结转移、肿瘤直径、边界、高回声晕、内部回声、血流分级及弹性评分比较,差异有统计学意义(P<0.05)。肿瘤直径(OR=1.428)、边界(OR=3.425)、高回声晕(OR=3.300)、内部回声(OR=9.631)及血流分级(OR=1.725)是乳腺癌患者术后复发风险的独立危险因素(P<0.05)。据此构建列线图预测模型,训练集验证中曲线下面积(AUC)为0.832,验证集中AUC为0.800,表现良好。校正曲线一致性高。结论 基于术前多模态超声构建预测乳腺癌术后复发风险的列线图模型,具有较高实用价值。

Objective To analyze the correlation between preoperative multimodal ultrasound features and recurrence risk of breast cancer and to explore its value.Methods A total of 286 cases(286 lesions)with breast cancer patients who underwent surgery in Mudan People’s Hospital of Heze City,Shandong Province from January 2020 to January 2023 were selected.They were divided into training set(200 cases)and validation set(86 cases)according to the ratio of 7∶3.The training set was divided into high risk group(68 cases)and a low-medium risk group(132 cases)according to the Guidelines and Norms for Diagnosis and Treatment of Breast Cancer of the Chinese Anti-Cancer Association 2021 Edition.The factors influencing the risk of breast cancer recurrence after surgery were analyzed and the prediction model of nomogram was established.The performance of the model was evaluated by receiver operating characteristic curve and calibration curve.Results There was no significant difference in general data between the training set and verification set(P>0.05).TNM stage,pathological grade,lymph node metastasis,tumor diameter,boundary,hyperechoic halo,internal echo,blood flow grade,and elastic score were significantly different between high risk group and low-medium risk group(P<0.05).Tumor diameter(OR=1.428),boundary(OR=3.425),hyperechoic halo(OR=3.300),internal echo(OR=9.631),and blood flow grade(OR=1.725)were independent risk factors for postoperative recurrence of breast cancer patients(P<0.05).Based on this,a nomogram prediction model was constructed.The area under the curve(AUC)in the training set was 0.832,and the AUC in verification set was 0.800,which showed good performance.Calibration curve consistency was high.Conclusion It is of high practical value to construct a nomogram model for predicting the recurrence risk of breast cancer based on preoperative multimodal ultrasound.