Axillary Surgery in Breast Cancer — Primary Results of the INSEMA Trial

Authors: Reimer T, Stachs A, Veselinovic K, Kühn T, Heil J, Polata S, Marmé F, Müller T, Hildebrandt G, Krug D, Ataseven B, Reitsamer R, Ruth S, Denkert C, Bekes I, Zahm D-M, Thill M, Golatta M, Holtschmidt J, Knauer M, Nekljudova V, Loibl S, Gerber B

Affiliation: Multicenter trial across Germany and Austria; the German Breast Group.

Journal: The New England Journal of Medicine, March 2025

PMID: 39665649.

Key takeaways

• Omitting sentinel lymph node biopsy in clinically node-negative patients undergoing breast-conserving therapy was noninferior for 5-year invasive disease–free survival in a mostly low-risk, HR-positive, HER2-negative population.

• Axillary recurrence was higher without axillary surgery, but absolute risk remained low at 5 years.

• Long-term lymphedema and arm morbidity were meaningfully lower when axillary surgery was omitted.

• External validity is narrow: most patients were older and had small, luminal tumors; higher-risk biology was underrepresented.

Background

Sentinel lymph node biopsy is standard for axillary staging in early breast cancer, but even limited axillary surgery can cause lymphedema, pain, and long-term functional deficits. INSEMA examines whether staging can be safely omitted in carefully selected patients undergoing breast-conserving therapy with whole-breast irradiation.

Objective

Determine whether omission of axillary surgery is noninferior to sentinel lymph node biopsy for invasive disease–free survival in clinically node-negative T1–T2 patients treated with breast-conserving surgery and whole-breast irradiation.

Methods

• Design: Prospective, randomized noninferiority trial conducted at 151 centers in Germany and Austria; Therapeutic Level I.

• Sample size: 5,154 clinically node-negative patients undergoing breast-conserving surgery randomized 1:4 to omission vs SLNB.

• Key demographics: Median age 62; 90% clinical T1; 79% pathologic T1; HR-positive/HER2-negative about 95%.

• Inclusion criteria: Women 18 years or older; clinically node-negative by exam and imaging; T1–T2 (≤5 cm); planned upfront breast-conserving surgery.

• Axillary imaging rule: Suspicious lymph nodes on preop axillary ultrasound triggered needle/core biopsy. Patients were eligible for randomization only if biopsy was negative; biopsy-proven nodal disease was excluded.

• Interventions & comparators: Randomized 1:4 to omit axillary surgery vs sentinel lymph node biopsy.

• Radiation protocol: Whole-breast irradiation for all; axilla not specifically targeted; Regional nodal irradiation was not performed except for 4 or more positive nodes in the surgery arm; partial breast irradiation was excluded.

• Primary endpoint: Invasive disease–free survival in the per-protocol population.

• Noninferiority margin: Required 5-year iDFS at least 85% in omission arm and upper 95% CI for HR less than 1.271.

• Statistics: Cox proportional hazards; noninferiority via CI for HR; multivariable adjustment for stratification factors (age, tumor size, grade).

Results

• Primary outcome, per-protocol: 5-year iDFS 91.9% omission vs 91.7% sentinel; HR 0.91 (95% CI, 0.73–1.14), meeting noninferiority.

• Event count and power: Planned event-driven analysis for 851 events, but analysis performed with 525 events due to low event rate, reducing power versus plan.

• Overall survival, per-protocol: 5-year OS 98.2% omission vs 96.9% sentinel.

• Axillary recurrence: 1.0% omission vs 0.3% sentinel at 5 years.

• Other first events, per-protocol: death 1.4% omission vs 2.4% sentinel; distant relapse 2.7% both; invasive ipsilateral breast recurrence 0.8% vs 1.1%.

• Safety and function at last follow-up: lower persistent lymphedema (1.8% vs 5.7%), mobility restriction (2.0% vs 3.5%), and pain with movement (2.0% vs 4.2%) with omission.

Conclusion

In clinically node-negative patients undergoing breast-conserving therapy, omission of surgical axillary staging was noninferior to sentinel lymph node biopsy for invasive disease–free survival after about 6 years, with less long-term arm morbidity.

Strengths

• Large randomized noninferiority design with mature median follow-up of 73.6 months.

• Radiotherapy quality controls and explicit constraints against deliberate axillary coverage improve interpretability of axillary omission.

• Captures patient-important endpoints: persistent lymphedema and arm symptoms.

Limitations

• Population is overwhelmingly low-risk: older, small tumors, luminal biology; higher-risk subtypes were underrepresented, limiting generalizability.

• Primary analysis occurred with 525 vs planned 851 events, decreasing statistical power and precision for uncommon but important failures like axillary relapse.

• Safety depends on preoperative axillary ultrasound and biopsy standards that vary widely across practice.

• Breast-conserving surgical technique was not granularly reported; the paper does not specify oncoplastic approaches or volume-replacement methods, limiting reconstruction-specific interpretation.

Clinical relevance

This trial supports de-escalating axillary surgery in carefully selected clinically node-negative patients having lumpectomy and whole-breast irradiation, with a clear reduction in long-term lymphedema and arm symptoms. It does not answer how to manage patients where nodal status would change systemic therapy or regional nodal irradiation decisions, because those higher-risk scenarios were uncommon in the enrolled cohort.

Additional considerations

• Whole-breast irradiation delivers incidental axillary dose; tangent design variability in real-world practice could change axillary control when surgery is omitted.

• The paper is surgically non-granular for the breast: it does not detail the reconstructive approach (i.e. oncoplastic mastopexy) which may further alter lymphatic drainage.

• The event shortfall reduces certainty about rare but consequential failures, and limits subgroup inference.