Evaluating the applicability of axillary surgery omission in real-world breast cancer patients: Insights from an INSEMA-eligible cohort.

Abstract

e12580 Background: The INSEMA trial demonstrated noninferiority of omitting axillary surgery compared to sentinel lymph node biopsy (SLNB) in clinically node-negative, T1–T2 breast cancer with similar 5-year invasive disease-free survival rates and fewer complications. We explored the feasibility of applying these results to real-world populations. Methods: We conducted a retrospective analysis of 2,363 INSEMA-eligible cohort who diagnosed clinical T1-2N0M0 breast cancer at Asan medical center and underwent breast conserving surgery between 2000 and 2020, comparing them to 3,896 patients in the SLNB arm of the INSEMA trial. All patients were clinically node-negative on preoperative breast MRI. We evaluated patient demographics, tumor characteristics, and outcomes to explore the feasibility of omitting axillary surgery in a broader, real-world population. Results: The INSEMA-eligible cohort had a younger median age (50.0 years) compared with the SLNB arm of the INSEMA trial (62.0 years). The proportion of clinical T2 stage tumors was higher in the INSEMA-eligible cohort (34.3% vs. 9.6% in the INSEMA trial), as were aggressive features such as nuclear grade 3 (22.7% vs. 3.6% in the INSEMA trial) and Ki67 > 20% (30.7% vs. 13.1% in the INSEMA trial). Subtypes other than HR+/HER2- were also more common in the INSEMA-eligible cohort (15.6% vs. 4.6% in the INSEMA trial). For sentinel lymph node status, node-negative cases were slightly lower in the INSEMA-eligible cohort (81.3% vs. 85.0% in the INSEMA trial), while N1 cases were higher (13.5% vs. 11.4% in the INSEMA trial), as were N1mi cases (5.0% vs. 3.5% in the INSEMA trial). Despite these differences, primary outcomes remained comparable. Invasive locoregional relapse occurred in 2.0% of the INSEMA-eligible cohort compared with 1.4% in the SLNB arm of INSEMA trial, while distant relapse rates were 1.4% and 2.7%, respectively. Overall death was also lower in the INSEMA-eligible cohort at 0.1%, compared with 2.4% in the SLNB arm of INSEMA trial. Conclusions: Comparable outcomes between the INSEMA-eligible cohort and the INSEMA SLNB arm suggest that omitting SLNB could cautiously be considered in real-world populations. Further studies should include diverse subtypes like HER2-positive, triple-negative cancers, younger patients, and more clinical T2 cases to validate this approach.