兰州大学学报(医学版)

目的 探讨多模态超声对浸润性乳腺癌患者新辅助化疗（NAC）疗效的预测价值。方法 选取接受NAC的58例浸润性乳腺癌患者，分别在NAC治疗前、NAC治疗2周期后行多模态超声检查。根据病理结果，分为有效组和无效组，比较两组乳腺癌患者常规超声肿瘤切面最大径值、应变式弹性成像应变比值、剪切波弹性成像最大弹性值、超声造影的定量参数达峰时间、峰值强度等参数在NAC治疗前、后的差异。通过构建Logistic回归模型，筛选出与NAC疗效相关的独立预测参数，综合评价多模态超声对浸润性乳腺癌NAC疗效的预测价值。结果 58例患者中有效组44例，无效组14例。有效组患者的肿瘤切面最大径值、应变式弹性成像应变比值、剪切波弹性成像最大弹性值、峰值强度明显低于无效组，且差别具有统计学意义（P<0.05）。Logistic回归分析显示，峰值强度是预测NAC疗效的独立超声参数，其受试者工作特征曲线下面积（AUC）为0.966(0.927,1.000)。在NAC疗效判断上，常规超声、弹性成像与病理结果的一致性较差（Kappa值=0.320, 0.397）,AUC分别为0.640(0.527, 0.753)和0.732(0.632, 0.832)；超声造影的一致性一般（Kappa值=0.425）,AUC为0.744(0.647,0.841)。多模态超声指标的一致性较强（Kappa值=0.781）,AUC为0.919(0.853,0.985)。结论 多模态超声用于乳腺癌患者NAC的疗效判断具有较好性能，能有效提升预测NAC疗效的准确性。峰值强度是预测NAC疗效的独立参数，可作为NAC疗效的无创性预测指标。

Objective To explore the predictive value of multimodal ultrasound in assessing the efficacy of neoadjuvant chemotherapy(NAC) for patients with invasive breast cancer. Methods 58 patients with invasive breast cancer were selected who received NAC treatment and multimodal ultrasound examinations were performed on them respectively before NAC treatment and after 2 cycles of NAC treatment. According to the post-chemotherapy pathologies, patients were assigned into the positive or negative outcome groups. Then the differences in the maximum diameter value of the conventional ultrasound tumor section, the strain ratio value of strain elastography, the maximum elasticity value of shear wave elastography, and the quantitative parameters of ultrasound were compared between the two groups. In addition, the differences between before and after the independent predictive ultrasound parameters related to the efficacy of NAC were also screened out via constructing a logistic regression model to comprehensively evaluate the predictive capacity of multimodal ultrasound on the efficacy of neoadjuvant chemotherapy for invasive breast cancer. Results 44 cases were in the effective group and 14 in the ineffective group. The MD, SR, Emax and PI of patients in the former were significantly lower than those in the latter, and the difference was statistically significant(P<0.05). Logistic regression analysis showed that PI(T1) was an independent ultrasound parameter for predicting the efficacy of NAC, and plotting the receiver operating characteristic(ROC) curve for predicting the efficacy of NAC with PI showed that the area under the curve(AUC) value of PI for predicting the efficacy of NAC in breast cancer patients was 0.966(0.927, 1.000). In determining the efficacy of NAC, conventional ultrasound and elastography showed a poor agreement with pathologic findings(Kappa values = 0.320, 0.397), with AUC values of 0.640(0.527, 0.753) and 0.732(0.632, 0.832), respectively; ultrasonography showed a fair agreement(Kappa value = 0.425), with an AUC value of 0.744(0.647, 0.841); multimodal ultrasound indices showed a strong agreement with pathologic findings(Kappa value = 0.781), and the AUC value was 0.919(0.853, 0.985). Conclusion Multimodal ultrasound results showed great accordance with the pathologies in term of therapeutical outcomes among breast cancer patients treated with NAC. PI is an independent parameter for predicting the efficacy of NAC, and can be used as a noninvasive predictor of the efficacy of NAC.