From Wiki Journal Club
Jump to: navigation, search
Giuliano AE, et al. "Axillary dissection vs no axillary dissection in women with invasive breast cancer and sentinel node metastasis: a randomized clinical trial". The Journal of the American Medical Association. 2011. 305(6):569-575.
PubMedFull textPDF

Clinical Question

Among T1-2 breast cancer patients with a positive sentinel lymph node biopsy undergoing lumpectomy and whole-breast radiation, does axillary lymph node dissection improve survival or local control compared to nodal observation?

Bottom Line

Axillary lymph node dissection does not improve survival or local control compared to nodal observation in patients with T1-2 breast cancer and a positive sentinel lymph node biopsy undergoing lumpectomy and whole-breast radiation.

Major Points

Axillary lymph node dissection (ALND) has historically been a means of maintaining regional control of breast cancer, especially in patients with initial presentation of metastatic nodal disease. Sentinel lymph node biopsy is a commonly used method of detecting micrometastatic axillary disease at initial presentation. Due to the significant morbidity of ALND, including paresthesias, lymphedema, seroma, and infection, judicious use of the procedure is important. The randomized NSABP B-32 study demonstrated that patients with a negative sentinel lymph node biopsy (SLNB) could be spared ALND.[1] Whether ALND might be avoided in patients with clinically negative axillary nodes and fewer than 3 positive sentinel lymph nodes was unknown.

The American College of Surgeons Oncology Group (ACOSOG) Z0011 trial was a multicenter noninferiority study which enrolled and randomized 891 patients with breast cancer, T1-2 disease, clinically negative axillary nodes, 1 or 2 macrometastatic nodes on SLNB, and a plan for breast conserving therapy consisting of lumpectomy and whole-breast radiation. Adjuvant chemotherapy was not specified in the protocol, but was received by >95% of patients in both arms. Patients were randomized to either completion ALND with a minimum of 10 nodes, or to observation. The primary outcome was overall survival, using a HR <1.3 to indicate noninferiority of observation to ALND. The study was terminated before target enrollment of 1900 women because the observed mortality was lower than anticipated. At 5 years, there were no differences in terms of overall survival (92.5% observation vs. 91.8% ALND; HR 0.79), disease-free survival (83.9% vs. 82.2%), or locoregional recurrence-free survival (96.7% vs. 95.7%). Patients undergoing ALND had increased surgical morbidity as expected (25% in the observation group due to SLNB alone vs. 70% in the ALND group). Ten-year follow-up was consistent with the 5-year published results.[2]

The results of ACOSOG Z0011 support the use of nodal surveillance to maintain locoregional control as an alternative to exposing patients to the morbidity of ALND for women with tumors <5 cm who undergo lumpectomy and subsequent whole-breast radiation. The study received criticism for its under enrollment, applicability to young women, generalizability on the basis of hormone receptor status (66% were ER+/PR+), and role with regards to neoadjuvant treatment and mastectomy (both of which were exclusion criteria). Under enrollment was due to low event rate (90% survival at 5 years compared to an anticipated 80%). However increased enrollment was unlikely to lead to a meaningful detection in significance, as the majority of axillary recurrence has been shown to occur in the first few years following surgery, the overall local recurrence rates were too low to be meaningfully changed by increased enrollment (0.9% SLNB alone vs. 0.4% ALND), and rates of total regional recurrences, disease-free survival, and overall survival trended in favor of the SLNB-only group. While many questions remain with regards to the role of ALND in breast cancer management, especially regarding hormonal status, risk stratification by immunochemical profile, and neoadjuvant treatment, the ACOSOG Z0011 trial is a landmark study supporting the role of surveillance alone for women with tumors <5 cm, clinically negative axillary nodes, and 1-2 positive sentinel nodes who are treated with lumpectomy and whole-breast radiation.


NCCN Breast Cancer Guidelines (1.2018, adapted)[3]

In cases with T1-2, ALND can be avoided only if all of the following criteria are met:

  • T1 or T2 tumors
  • Clinically negative nodes
  • 1-2 positive nodes on SLNB (on the basis of H&E staining)
  • Planned breast conserving therapy
  • Planned whole-breast RT
  • No neoadjuvant chemotherapy planned


  • Multicenter, randomized, noninferiority trial
  • N=891 patients with T1 or T2 tumors and 1-2 positive nodes on SLNB undergoing lumpectomy with whole-breast radiotherapy
    • Completion ALND (n=420)
    • Observation (n=436)
  • Setting: 115 institutions affiliated with the Cancer Trials Support Unit and North Central Cancer Treatment Group
  • Enrollment: 1999-2004
  • Analysis: Intention-to-treat
  • Primary endpoint: Overall survival
  • Follow-up: 6.3 years median


Inclusion Criteria

  • Women ≥18 years
  • T1 or T2, N0, M0
  • Tumor size ≤ 5cm
  • SLNB demonstrating with nodal metastatic breast cancer per frozen section, touch preparation, or hematoxylin-eosin staining
  • SLNB within 60 days of invasive breast cancer diagnosis
  • Treatment with lumpectomy to negative margins
  • Eastern Cooperative Oncology Group/Zubrod status ≤2

Exclusion Criteria

  • Metastasis identified by immunohistchemical staining
  • Mastectomy
  • Partial or no radiation treatment
  • Palpable adenopathy or gross extranodal disease
  • ≥3 positive nodes on SLNB
  • Matted nodes
  • Neoadjuvant hormonal/chemotherapy treatment
  • No third field axillary nodal irradiation in “observation group”

Baseline Characteristics

Results are from the ALND group.

  • Age in years, median (range): 56 (24-92)
  • Clinical T stage (number, %)
    • T1: 284 (67.9%)
    • T2 134 (32.1%)
  • Tumor size in cm median (range): 1.7 (0.4-7.0)
  • Receptor status
    • ER+/PR+: 256(66.8%)
    • ER+/PR-: 61(15.9%)
    • ER-,PR+: 3(0.8%)
    • ER-,PR-: 63(16.5%)
    • Missing 37
  • Lymphovascular invasion
    • Yes 129 (40.6%)
    • No 189 (59.4%)
  • Modified Bloom Richardson Score
    • 1: 71 (22.0%)
    • 2: 158 (48.9%)
    • 3: 94 (29.1%)
  • Tumor type
    • Infiltrating ductal: 344 (82.7%)
    • Infiltrating lobular: 27 (6.5%)
    • Other: 45 (10.8%)
  • Lymph node metastases
    • 0: 4 (1.2%)
    • 1: 199 (58.0%)
    • 2: 68 (19.8%)
    • 3: 25 (7.3%)
    • ≥4 47 (13.7%)
  • Adjuvant systemic therapy: 403 (96.0%)
  • Whole breast radiation therapy: 263/296 (88.9%)


  • All patients had undergone SLNB and were planned to receive lumpectomy and whole-breast radiotherapy
  • Randomized to completion ALND or to observation
  • Adjuvant therapy was allowed at the discretion of the treating physician
  • Follow-up period 6, 12, 18, 24, 30, 36 months
  • Annual mammograms


Comparisons are observation vs. ALND groups.’’

Primary Outcomes

Overall survival at 5 years
92.5% vs. 91.8% (unadjusted HR 0.79; 90% CI 0.56-1.10; P=0.008 for noninferiority)
Models adjusting for chemotherapy, endocrine therapy, radiation treatment, age, and ER/PR receptor status did not differ in HR.

Secondary Outcomes

Disease-free survival at 5 years
83.9% vs. 82.2% (unadjusted HR=0.82; 95% CI 0.58- 1.17; P=0.14)
Models adjusting for adjuvant treatment, primary tumor size, ER/PR receptor status, tumor type, Bloom-Richardson score, and age did not substantially differ in HR.
Locoregional recurrence-free survival at 5 years
96.7% vs. 95.7% (P=0.28)

Adverse Events

Comparisons are observation vs. ALND’’

Previously reported in [4]

Adverse surgical events (wound infection, seroma, paresthesias, lymphedema)
25% vs. 70% (P≤0.001)
Wound infection
3% vs. 8% (P=0.0016)
Axillary seromas
6% vs. 14% (P=0.0001)
Axillary paresthesias at 12 months
9% vs. 39% (P<0.0001)
Subjective lymphedema at 12 months
6% vs. 19% (P<0.0001)
Objective lymphedema (arm measurements) at 12 months
6% vs. 11% (P=0.079)

Subgroup Analysis

Multivariate analysis adjusting for HR of primary and secondary outcomes are available, but did not differ substantially (no adjusted HR >1.1, with a threshold for noninferiority set at >1.3).


  • Trial stopped before target enrollment of 1900 patients
  • Unclear applicability to for patients with hormone receptor-negative tumors
  • Unclear applicability to young women (age <50), who have documented higher rates of in-breast recurrence, albeit without evidence of higher rates of nodal recurrence
  • Does not address patients undergoing neoadjuvant therapy


  • National Cancer Institute Grant U10 CA 76001 to the American College of Surgeons Oncology Group (ACOSOG)
  • No conflicts of interest reported

Further Reading

  1. Krag DN et al. Sentinel-lymph-node resection compared with conventional axillary-lymph-node dissection in clinically node-negative patients with breast cancer: overall survival findings from the NSABP B-32 randomised phase 3 trial. Lancet Oncol. 2010. 11:927-33.
  2. Giuliano AE et al. Effect of Axillary Dissection vs No Axillary Dissection on 10-Year Overall Survival Among Women With Invasive Breast Cancer and Sentinel Node Metastasis: The ACOSOG Z0011 (Alliance) Randomized Clinical Trial. JAMA 2017. 318:918-926.
  3. Gradishar WJ et al. Breast Cancer, Version 4.2017, NCCN Clinical Practice Guidelines in Oncology. J Natl Compr Canc Netw 2018. 16:310-320.
  4. Lucci A et al. Surgical complications associated with sentinel lymph node dissection (SLND) plus axillary lymph node dissection compared with SLND alone in the American College of Surgeons Oncology Group Trial Z0011. J. Clin. Oncol. 2007. 25:3657-63.