AUTHOR=Li Zhuoxuan , Tong Yiwei , Chen Xiaosong , Shen Kunwei 

TITLE=Accuracy of ultrasonographic changes during neoadjuvant chemotherapy to predict axillary lymph node response in clinical node-positive breast cancer patients

JOURNAL=Frontiers in Oncology

VOLUME=Volume 12 - 2022

YEAR=2022

URL=https://www.frontiersin.org/journals/oncology/articles/10.3389/fonc.2022.845823

DOI=10.3389/fonc.2022.845823

ISSN=2234-943X

ABSTRACT=Purpose: To evaluate whether changes of ultrasound features during neoadjuvant chemotherapy (NAC) could predict axillary node response in clinical node-positive breast cancer patients.
Methods: Patients with biopsy-proven node-positive disease receiving NAC from February 2009 and March 2021 were included. Ultrasound (US) images were obtained using a 5-12-MHz linear array transducer before NAC, after 2 cycles, and at completion of NAC. Long and short diameter, cortical thickness, vascularity and hilum status of the metastatic node were retrospectively reviewed according to Breast imaging-reporting and data system (BI-RADS). The included population were randomly divided into a training set and a validation set at a 2:1 ratio using simple random sampling method. Factors associated with node response were identified through univariate and multivariate analyses. A nomogram combining clinical and changes of ultrasonographic (US) features was developed and validated. The receiver operating characteristic (ROC), calibration plots were applied to evaluate nomogram performance and discrimination.
Results: A total of 296 breast cancer patients were included, 108 (36.5%) of whom achieved axillary pathologic complete response (pCR) and 188 (63.5%) had residual nodal disease. Multivariate regression indicated that independent predictors of node pCR contain ultrasound features in addition to clinical features, clinical features including neoadjuvant targeted therapy and clinical response, ultrasound features after NAC including cortical thickness, hilum status and reduction in short diameter ≥50%. Nomogram combining clinical features and US features showed better diagnostic performance compared to clinical only model in the training cohort (AUC: 0.799 vs. 0.699, P=0.001) and the validation cohort (AUC: 0.764 vs. 0.638, P=0.027). 
Conclusions: Ultrasound changes during NAC could improve the accuracy to predict node response after NAC in clinical node-positive breast cancer patients.