Axillary dissection versus no axillary dissection in patients with sentinel-node micrometastases (IBCSG 23-01): a phase 3 randomised controlled trial

Abstract

Background: For patients with breast cancer and metastases in the sentinel nodes, axillary dissection has been standard treatment. However, for patients with limited sentinel-node involvement, axillary dissection might be overtreatment. We designed IBCSG trial 23-01 to determine whether no axillary dissection was non-inferior to axillary dissection in patients with one or more micrometastatic (≤2 mm) sentinel nodes and tumour of maximum 5 cm.

Methods: In this multicentre, randomised, non-inferiority, phase 3 trial, patients were eligible if they had clinically non-palpable axillary lymph node(s) and a primary tumour of 5 cm or less and who, after sentinel-node biopsy, had one or more micrometastatic (≤2 mm) sentinel lymph nodes with no extracapsular extension. Patients were randomly assigned (in a 1:1 ratio) to either undergo axillary dissection or not to undergo axillary dissection. Randomisation was stratified by centre and menopausal status. Treatment assignment was not masked. The primary endpoint was disease-free survival. Non-inferiority was defined as a hazard ratio (HR) of less than 1·25 for no axillary dissection versus axillary dissection. The analysis was by intention to treat. Per protocol, disease and survival information continues to be collected yearly. This trial is registered with ClinicalTrials.gov, NCT00072293.

Findings: Between April 1, 2001, and Feb 28, 2010, 465 patients were randomly assigned to axillary dissection and 469 to no axillary dissection. After the exclusion of three patients, 464 patients were in the axillary dissection group and 467 patients were in the no axillary dissection group. After a median follow-up of 5·0 (IQR 3·6-7·3) years, we recorded 69 disease-free survival events in the axillary dissection group and 55 events in the no axillary dissection group. Breast-cancer-related events were recorded in 48 patients in the axillary dissection group and 47 in the no axillary dissection group (ten local recurrences in the axillary dissection group and eight in the no axillary dissection group; three and nine contralateral breast cancers; one and five [corrected] regional recurrences; and 34 and 25 distant relapses). Other non-breast cancer events were recorded in 21 patients in the axillary dissection group and eight in the no axillary dissection group (20 and six second non-breast malignancies; and one and two deaths not due to a cancer event). 5-year disease-free survival was 87·8% (95% CI 84·4-91·2) in the group without axillary dissection and 84·4% (80·7-88·1) in the group with axillary dissection (log-rank p=0·16; HR for no axillary dissection vs axillary dissection was 0·78, 95% CI 0·55-1·11, non-inferiority p=0·0042). Patients with reported long-term surgical events (grade 3-4) included one sensory neuropathy (grade 3), three lymphoedema (two grade 3 and one grade 4), and three motor neuropathy (grade 3), all in the group that underwent axillary dissection, and one grade 3 motor neuropathy in the group without axillary dissection. One serious adverse event was reported, a postoperative infection in the axilla in the group with axillary dissection.

Interpretation: Axillary dissection could be avoided in patients with early breast cancer and limited sentinel-node involvement, thus eliminating complications of axillary surgery with no adverse effect on survival.

Funding: None.

Conflict of interest statement

CONFLICTS OF INTEREST

None of the authors have any conflicts of interest.

Copyright © 2013 Elsevier Ltd. All rights reserved.

Figures

Figure 1

CONSORT diagram showing the 931 patients in the analytic cohort of IBCSG Trial 23- 01 for the intention-to-treat (ITT) analysis. The trial included a registration step prior to surgery with eligibility based on clinical features. Only patients with eligible pathological features determined at primary breast surgery (one or more positive sentinel nodes and largest tumour lesion size ≤ 5 cm) were eligible for randomisation.

Figure 2

Comparison of axillary dissection (AD, solid line) to no axillary dissection (No AD, dashed line) for disease-free survival (A), cumulative incidence of breast cancer events (B), and overall survival (C) in the intention-to-treat population of 931 patients.

Figure 3

Hazard ratios and confidence intervals comparing axillary dissection (AD) with no axillary dissection (No AD) among subgroups of the intention-to-treat population of 931 patients. Each subgroup hazard ratio is shown as a black square with the size of the square being inversely proportional to the variance of the corresponding log-hazard-ratio estimate (i.e., larger squares indicate lower variability in the estimate). The hazard ratio for all patients is shown as a diamond. The horizontal axis is displayed on a logarithmic scale.