Predictors of positive axillary non-sentinel lymph nodes in breast cancer patients with positive sentinel lymph node biopsy after neoadjuvant systemic therapy

Aim: To assess the rate of positive non-sentinel lymph nodes (non-SLNs) after neoadjuvant systemic therapy (NAST) in breast cancer (BC) following positive sentinel lymph node biopsy (SLNB). Materials and methods: From institutional database, 265 consecutive patients receiving NAST for cT1-3, any N, M0 BC between 2001 and 2018 were identified. Patients presented clinically negative axilla before surgery and were candidate for SLNB. Following metastatic SLNB, completion axillary lymph node dissection (AxLND) was performed. Non-SLNs rate was investigated using multivariate (MV) logistic regression models. The distribution of non-SLNs across the axilla was observed. Results: Positive non-SLNs were found in 62.3% of cases and showed no correlation with SLN metastasis size. At MV, statistically significant variables associated with non-SLNs were older age (p = 0.025), clinically positive lymph nodes (p = 0.002), SLN extracapsular extension (ECE, p = 0.001), and higher ratio of positive SLNs/total SLNs (p = 0.016). ECE and higher nodal ratio were independent predictors of III axillary level positivity. By categorizing patients in intermediate-and high-risk groups using the study variables, positive non-SLNs were found in the range of 23-56% across the three axillary levels, rates which did not support radiotherapy volume deescalation. The III axillary level lower involvement (6.3%) was better identified with the RAPCHEM trial criteria based on the ypN status after AxLND. Conclusions: Involved non-SLNs rate following positive SLNB after NAST is nearly double the rate observed after primary surgery, supporting some intervention on the axilla. If AxLND is limited to I and II level, the involvement of the III level up to 31% of the cases seems to require some additional treatment, while the omission in selected cases needs further investigation. (c) 2021 Elsevier B.V. All rights reserved. Radiotherapy and Oncology 163 (2021) 128-135