PUB - Publikationen an der Universität Bielefeld

Background: Axillary nodal status is an important prognostic factor in early breast cancer (eBC), guiding systemic treatment and postoperative radiotherapy. As axillary surgery does not significantly affect BC mortality itself, it is considered as a staging procedure in clinically node-negative (cN0) patients (pts). The Intergroup-Sentinel-Mamma (INSEMA) trial investigated the avoidance of sentinel lymph node biopsy (SLNB) in cN0 pts (Rando1) or the omission of completion axillary lymph node dissection (cALND) in pN1a(sn) pts (Rando2). The analysis of the first randomization demonstrated non-inferiority of omitting SLNB in cN0 patients undergoing breast-conserving surgery (BCS) concerning invasive disease-free survival (iDFS), meeting the trial's primary endpoint. Here we report the analysis of the second randomization. Study Design: The INSEMA trial was conducted between 2015 and 2019 in Germany and Austria. The first randomization of this prospective trial compared no axillary surgery with SLNB in pts with invasive eBC (tumor size ≤ 5 cm; c/iN0) scheduled for BCS, including postoperative whole-breast irradiation (WBI). This randomization was carried out in a 4:1 allocation (SLNB vs. no SLNB). Pts with 1-3 macrometastases in the SLNB arm underwent a second randomization in a 1:1 ratio, to either SLNB alone or cALND. The aim was to assess whether SLNB alone is non-inferior to cALND in terms of iDFS. The analysis of Rando2 was based on the per-protocol (PP) set. Due to fewer SLNB-positive patients than expected, the iDFS analysis for the second randomization was downgraded from a co-primary to a key secondary outcome following protocol amendment #5 (December 2018). The non-inferiority margin was defined as 5-year iDFS > 76.5% (hazard ratio (HR) < 1.271) for SLNB alone, compared to an expected 5-year iDFS of 81% for the cALND arm. Results: 485 pts were recruited for Rando2 (intention-to-treat (ITT) set: N=243 with cALND vs. N=242 with SLNB alone). After excluding 99 pts (mainly due to axillary surgery performed not per randomized arm), 386 pts (cALND: N=169, SLNB alone: N=217) were included in the PP set. The median follow-up (FU) is 74.2 months. The cALND cohort is characterized by higher rates for postoperative chemotherapy (39.8% vs. 33.6%, p=0.239), conventionally fractionated WBI (87.0% vs. 75.1%, p=0.004), tumor bed boost (88.8% vs. 80.6%, p=0.035), and regional nodal irradiation (36.0% vs. 20.6%, p=0.019) compared to the SLNB alone cohort. Analysis in the PP set was unable to demonstrate non-inferiority for SLNB alone compared to cALND, with an HR of 1.6]9 (95% CI: 0.98-2.94). Estimated 5-year iDFS rates are 86.6% (81.0%-90.7%) in the SLNB alone arm and 93.8% (88.7%-96.6%) in the cALND arm (log-rank p=0.058). Estimated 5-year overall survival (OS) rates are 94.9% (90.6%-97.2%) in the SLNB alone arm and 96.2% (91.7%-98.3%) in the cALND arm (log-rank p=0.663). Among the ITT set, there was also no difference in iDFS between the arms, with an HR of 1.26 (0.80-1.99) for SLNB alone compared to cALND. Estimated 5-year iDFS rates (ITT set) are 86.0% (80.6%-90.0%) with SLNB alone and 89.3% (84.3%-92.8%) with cALND, respectively (log-rank p=0.314). Locoregional recurrences (LRR) were infrequent, with 5-year cumulative incidence rates of 1.1% vs. 0.0% (p=0.405) in the SLNB alone arm compared to cALND. The safety analysis demonstrates that patients who underwent SLNB alone benefited in terms of lymphedema rate, arm mobility, and reduced arm and shoulder pain. Conclusion: No significant differences were observed between SLNB alone vs. cALND in both subsets (PP, ITT) for iDFS, OS, and LRR. These findings after a 6-year FU are representative of cN0 pts with positive SLNB and BCS; the 10-year FU data will be presented in 2029.