E7: Multi-Institution Review of Contralateral Axillary Metastases: Time for Reclassification as a Local-Regional Event?

Breast cancer metastases to the contralateral axillary lymph nodes (CAM) is a rare event and often portends a poor prognosis. Breast cancer guidelines consider CAM stage IV disease (M1). Retrospective studies suggest that CAM has comparable overall survival (OS) to locally advanced breast cancer (LABC) but contain heterogeneous cohorts. We sought to evaluate outcomes in a CAM cohort treated with contemporary systemic and locoregional therapy.

Methods:

We performed a retrospective multi-institution study comparing OS of patients with CAM to LABC (cN2-3 and/or pN2-3) and non-CAM M1. All CAM patients underwent axillary lymph node resection. The M1 cohort included de novo or metachronous distant metastases. OS was defined as time from CAM or metastatic diagnosis to death or last follow-up for CAM and M1, and time from primary diagnosis to death or last follow-up for LABC. Unadjusted OS was estimated with Kaplan-Meier method, and adjusted Cox proportional hazards models were used to estimate the association of group with OS.

Results:

We identified 60 CAM and 158 LABC from 9 institutions (2016-2023), and 632 M1 from a single institution. The median age was similar between groups (CAM=60 years, LABC=57, M1=57, p=0.37), and there were no statistically significant differences in hormone receptor (HR) status (p=0.36). Among CAM patients, 48.3% were HR+/HER2-, 21.7% HR-/HER2-, 10% HR+/HER2+, and 6.7% HR-/HER2+. The index cancer in CAM cases was more commonly invasive ductal carcinoma compared to LABC (81.7% vs 67.1%), as well as grade 3 (45% vs 35.4%) and triple negative (21.7% vs 13.9%). The majority of CAM were metachronous without in-breast or regional recurrence (36.7%), or synchronous with in-breast recurrence (31.7%). Treatment of LABC more frequently included chemotherapy (92.4% vs 71.7%), mastectomy (74.7% vs 51.7%), and adjuvant radiation (90.5% vs 73.3%) compared to CAM. Unadjusted OS was higher for CAM than M1 but lower than LABC (p< 0.001) (Figure). On multivariable analysis adjusting for age, race, and HR status, both CAM and M1 had inferior survival to LABC, but with a 2.14-fold difference for CAM (95% CI 1.21-3.76, p=0.009) versus a 5.98-fold difference for M1 (95% CI 3.83-9.35, p< 0.001).

Conclusions:

In this contemporary, multi-institutional study, we demonstrated that CAM patients selected for presumed curative intent treatment experienced improved OS when compared to stage IV (M1) patients. Our data add additional support for re-evaluating the current stage IV designation, and consideration of curative intent treatment in this disease.