Delta超声组学对早期乳腺癌新辅助治疗疗效的预测价值  

Predictive Value of Delta Ultrasound in the Efficacy of Neoadjuvant Therapy for Breast Cancer

作　　者：钟晓情 曾乔 刘义钢 付志勇[1] 符蓓 袁新春[2] ZHONG Xiaoqing;ZENG Qiao;LIU Yigang(Jiangxi Cancer Hospital,Nanchang,330029)

机构地区：[1]江西省肿瘤医院,330029 [2]南昌大学第一附属医院,330000

出　　处：《实用癌症杂志》2024年第11期1805-1810,1818,共7页The Practical Journal of Cancer

基　　金：江西省卫生健康委科技计划(编号:202410387)。

摘　　要：目的探讨Delta超声组学对乳腺癌新辅助治疗(NAT)后疗效的预测价值。方法将140例乳腺癌新辅助治疗患者按照软件中的随机种子数以7∶3比例分成训练集(98例)及验证集(42例)。基于肿瘤NAT治疗前、NAT治疗2个疗程后最大径相对退缩值构建传统影像模型;基于治疗前与2个疗程后的超声图像提取和筛选相应影像组学特征并计算其Delta,构建Delta超声组学模型并计算其影像组学得分(Radscore);基于有意义的临床病理指标结合Radscore构建综合模型,采用受试者操作特征曲线(ROC)评估模型预测乳腺癌NAT后病理完全缓解(pCR)的效能,比较它们的曲线下面积(AUC)、准确度、灵敏度、特异度。使用临床校正曲线与临床决策曲线对模型进行评估与对比。结果综合模型预测pCR的效能最佳,其AUC值在训练集与验证集分别达0.92、0.85;其次为Delta超声组学模型,AUC分别为0.80、0.82;传统影像模型效能最低,AUC分别为0.67、0.72。临床校准曲线显示综合模型预测值与实际值一致性较高,临床决策曲线显示综合模型优于其他两种模型。结论Delta超声组学是早期预测乳腺癌NAT疗效的标记物。基于Delta超声组学结合临床病理指标构建的综合模型可作为早期准确预测NAT疗效的评估手段。Objective To explore the predictive value of Delta ultrasound in the efficacy of neoadjuvant therapy for breast cancer.Methods The clinical and ultrasonic image data of 140 patients with breast cancer who received NAT were retrospectively collected,and were divided into training sets(98 cases)and verification sets(42 cases)according to the ratio of 7∶3 of the number of random seeds in the software.Build traditional imaging evaluation models,Delta Ultrasound Radiomics models,and comprehensive models.Evaluated the diagnostic performance of different models on pCR in the training and validation sets using ROC curves,and compared their area under the curve(AUC),accuracy,sensitivity,and specificity.Using comprehensive discrimination improvement index to compare the diagnostic performance of the model.Results The comprehensive model had the best performance in predicting pCR,with AUC values of 0.92 and 0.85 in the training and validation sets,respectively;Next was the Delta ultrasound imaging omics model,with AUC of 0.80 and 0.82,respectively;The traditional image model had the lowest performance,with AUC of 0.67 and 0.72,respectively.The clinical calibration curve showed that the predicted values of the comprehensive model were consistent with the actual values,while the clinical decision curve showed that the comprehensive model was superior to the other 2 models.Conclusion Delta ultrasound imaging omics is a marker for early prediction of neoadjuvant therapy for breast cancer.The comprehensive model based on Delta ultrasound imaging omics combined with clinicopathological indicators can be used as an early and accurate evaluation method to predict the efficacy of neoadjuvant therapy for breast cancer.

关 键 词：乳腺癌 Delta超声组学 新辅助治疗 

分 类 号：R737.9[医药卫生—肿瘤]