Abstract RF2-07: Nodal disease burden in patients with clinically node-positive breast cancer undergoing tailored axillary surgery with or without axillary dissection in the neoadjuvant and upfront surgery setting: pre-planned TAXIS study (OPBC-03, SAKK 23/16, IBCSG 57-18, ABCSG-53, GBG 101)

Abstract Introduction: The safety of omitting axillary dissection (ALND) in patients with clinically node-positive breast cancer (cN+ BC) may depend on the nodal disease burden left behind in the axilla. This study aimed to evaluate nodal disease volume, quantify nodal understaging without ALND, and identify factors associated with additional nodal disease at ALND in these patients. Methods: The international phase-III TAXIS trial (NCT03513614) randomized patients with cN+ stage II-III BC to ALND or axillary radiotherapy (ART) following tailored axillary surgery (TAS). Nodal disease was detected by imaging or palpation at initial diagnosis. TAS removed sentinel, biopsied, and palpably suspicious nodes. Patients had upfront surgery or residual nodal disease after neoadjuvant chemotherapy (NACT). 1500 patients were randomized from 08/2018 to 08/2025. Results: Of 1418 patients with available data, 712 (50.2%) underwent ALND. Nodal disease was detected by imaging in 735 patients (51.8%), and by palpation in 683 (48.2%). Clinical nodal stage was cN1 in 1232 (86.9%) and cN2/3 in 186 (13.1%). Tumors were HR-positive and HER2-negative in 1140 patients (80.4%), HER2-positive in 160 (11.3%), and triple negative in 96 (6.8%). 552 patients (38.9%) underwent NACT. TAS removed a median of 5 nodes (interquartile range [IQR] 3-7), 2 (IQR 1-4) of which were positive. After TAS, ALND removed a median of 11 additional nodes (IQR 8-16), one (IQR 0-4) of which was positive. Among 712 patients undergoing ALND, additional positive nodes after TAS were removed in 430 (60.4%), AJCC pN upstaging occurred in 226 (31.7%), and 336 (47.2%) had (y)pN2/3 stage. Nodal burden by use of NACT is shown in the table. On multivariable logistic regression the number of positive nodes on TAS was associated with higher odds of having additional positive nodes on ALND in the NACT (odds ratio [OR] 1.57, 95% confidence interval [CI] 1.21-2.03, p<0.001) and upfront surgery group (OR 1.39, 95%CI 1.23-1.57, p<0.001). Furthermore, the odds of having additional positive nodes on ALND were higher in macrometastatic vs. isolated tumor cells/micrometastatic nodal disease on TAS after NACT (OR 2.24, 95%CI 1.10-4.55, p=0.026), and in palpable vs. imaging-detected nodal disease at upfront surgery (OR 1.66, 95%CI 1.05-2.63, p=0.032). Conclusion: Nodal disease burden is high in patients included in the TAXIS trial. Among patients who underwent ALND, almost half had (y)pN2/3 disease, and 60.4% had additional positive nodes removed that were missed by TAS. Nodal disease volume on TAS was associated with higher odds of having additional positive nodes on ALND. Interim analysis raised no safety concerns in the TAXIS study, and long-term follow-up will determine if ART is oncologically non-inferior to ALND. Citation Format: W. P. Weber, C. Tausch, S. Hayoz, Z. Matrai, G. Xepapadakis, C. Simonson, V. Bjelic-Radisic, G. T. Lam, G. Montagna, M. Gnant, L. H. Rosenberger, E. de Bree, R. Satler, M. Fehr, C. Leo, L. Lelievre, S. Bucher, S. Schmid, R. Exner, K. Reisenberger, U. Beckmann, S. Muenst, G. Henke, D. R. Zwahlen, T. Ruhstaller, K. Ribi, C. Urban, A. Crown, J. E. Lee, J. Boileau, A. D. Williams, Y. Jonghan, M. L. DiNome, A. Poultsidi, E. Gonzales, S. M. Wong, A. Schulz, M. Nealeigh, S. G. Ahn, A. M. Botty van den Bruele, B. J. Chae, A. Mueller, D. Hagen, J. M. Ryu, A. Savolt, C. Kurzeder, J. Heil, D. Egle, M. Heidinger, M. Knauer. Nodal disease burden in patients with clinically node-positive breast cancer undergoing tailored axillary surgery with or without axillary dissection in the neoadjuvant and upfront surgery setting: pre-planned TAXIS study (OPBC-03, SAKK 23/16, IBCSG 57-18, ABCSG-53, GBG 101) [abstract]. In: Proceedings of the San Antonio Breast Cancer Symposium 2025; 2025 Dec 9-12; San Antonio, TX. Philadelphia (PA): AACR; Clin Cancer Res 2026;32(4 Suppl):Abstract nr RF2-07.