Nomogram Based on US and Clinicopathologic Characteristics: Axillary Nodal Evaluation Following Neoadjuvant Chemotherapy in Patients With Node-Positive Breast Cancer

In order to prevent surgical over-treatment of axilla in patients with the conversion of lymph node (LN) following neoadjuvant chemotherapy (NAC), it is crucial to have accurate axilla staging procedures. This multi-center study is designed to develop a convenient method to predict the axillary response to NAC in breast cancer patients. A total of 1019 patients were randomly assigned to the training and validation groups at a ratio of 7:3. US characteristics of both primary tumors and axillary LNs independently serve as predictors for the axillary response to NAC in breast cancer patients. In the validation cohort, the discrimination of US model (AUC, 0.76) was superior to clinicopathologic model (AUC, 0.68); the combined model (AUC, 0.85) demonstrates strong discriminatory power in predicting nodal pCR. US could indeed play a valuable role in identifying more nonresponders to NAC in axillary LNs. The nomogram, constructed with readily available clinicopathologic features and US characteristics, exhibited a FNR of 16.67% in all patients and 10.53% in patients with triple negative breast cancer. This nomogram might potentially serve as a valuable visual tool to aid clinicians in making informed treatment decisions and optimizing patient care for patients with node-postive breast cancer receiving NAC, especially for the patients with triple negative breast cancer. Background: To develop a convenient modality to predict axillary response to neoadjuvant chemotherapy (NAC) in breast cancer patients. Materials and Methods: In this multi-center study, a total of 1019 breast cancer patients with biopsy-proven positive lymph node (LN) receiving NAC were randomly assigned to the training and validation groups at a ratio of 7:3. Clinicopathologic and ultrasound (US) characteristics of both primary tumors and LNs were used to develop corresponding prediction models, and a nomogram integrating clinicopathologic and US predictors was generated to predict the axillary response to NAC. Results: Axillary pathological complete response (pCR) was achieved in 47.79% of the patients. The expression of estrogen receptor, human epidermal growth factor receptor -2, Ki-67 score, and clinical nodal stage were independent predictors for nodal response to NAC. Location and radiological response of primary tumors, cortical thickness and shape of LNs on US were also significantly associated with nodal pCR. In the validation cohort, the discrimination of US model (area under the curve [AUC], 0.76) was superior to clinicopathologic model (AUC, 0.68); the combined model (AUC, 0.85) demonstrates strong discriminatory power in predicting nodal pCR. Calibration curves of the nomogram based on the combined model demonstrated that substantial agreement can be observed between the predictions and observations. This nomogram showed a false-negative rates of 16.67% in all patients and 10.53% in patients with triple negative breast cancer. Conclusion: Nomogram incorporating routine clinicopathologic and US characteristics can predict nodal pCR and represents a tool to aid in treatment decisions for the axilla after NAC in breast cancer patients.