How Should We Proceed with ypN+ Breast Cancer Following Positive Sentinel Lymph Node Biopsy after Neoadjuvant Chemotherapy?

Abstract

The appropriate treatment of the axilla in patients who receive neoadjuvant chemotherapy and subsequently have a positive sentinel lymph node (SLN) is unclear (ypN+). The Alliance A011202 trial is comparing axillary lymph node dissection (ALND) with nodal radiation therapy (RT) to comprehensive RT alone. Our study sought to provide guidance as to how to proceed in the interim in patients with SLN positive disease, ypN+, who do not undergo ALND. From 2001 to present, we retrospectively reviewed 74 women with clinical node positive or negative breast cancer who underwent neoadjuvant chemotherapy and on SLN biopsy or ALND were found to have ypN+ disease. 27 out of 74 women underwent comprehensive RT alone, instead of an ALND. Of the 47 women who underwent ALND, 4 declined RT, 3 were treated with 2-field technique and the remaining 40 women received 3-field RT, limiting dose to the dissected axilla. Doses to the axilla and supraclavicular regions ranged from 45-50.4 Gy, delivered in 23-28 fractions. Patients were followed routinely with exam and imaging. Dates and sites of local, regional, and distant failure were recorded. The effects of risk factors were examined using Cox proportional hazards models to estimate the hazard of recurrence at any given time and supplemental Fisher’s exact testing. The median follow up was 46 months (range 0-189), with a median follow up of 44.5 months for RT and 46 months for ALND. Overall local, regional and distant failure rates were found to be 6.8%, 9.5% and 14.9%, respectively. When looking at individual risk factors for recurrence, we found that the hazard of recurrence at any given time for patients with lymphovascular invasion (LVI) was 3.58 (95% CI: 1.30-9.86) times higher compared to patients who did not have LVI (p =.01). No other factors were found to be significant such as ECE, grade, hormone status or margins. When comparing patients treated with ALND and nodal RT vs comprehensive RT, we found a protective trend associated with RT, however the results did not reach significance (p=.18). Local and regional first recurrent rates were 0% and 3.7% with comprehensive RT, compared to 10.6% and 12.8% with ALND and nodal RT. 2 patients who recurred in the ALND group showed simultaneous local and regional failure, making a locoregional failure rate of 23.4%. Distant failure rates were similar at 14.8% with RT and 14.9% with ALND. Lymphedema rates of 17.4% were seen with RT compared to 26.5% with ALND (p=.39). Based on our findings, in patients who do not undergo ALND, a subset of women may benefit from comprehensive RT as it appears to be an effective means of local and regional disease control in patients with ypN+ disease. We encourage participation in Alliance A011202 to obtain randomized data in this cohort. We have an ongoing analysis to determine if the ALND group had higher risk features than the comprehensive RT group, which may have predisposed them to higher rates of failure.