A New Era in Axillary Management for Node-Positive Women

Axillary lymph node dissection (ALND) is no longer the standard approach to the management of all breast cancer patients with metastases to the sentinel lymph nodes. Advances in mammography screening have resulted in lower nodal disease burden in women with axillary metastases, and systemic therapies — chemotherapy, endocrine therapy, and anti-HER2 therapy — play an important role in preventing the locoregional recurrence of breast cancer. Recognition of these factors has led to 2 important prospective randomized trials examining the safety of eliminating ALND for some breast cancer patients.

Practice-Changing Trials: ACOSOG Z011 and AMAROS

In the first study, the American College of Surgeons Oncology Group (ACOSOG) Z011 trial, women with clinically node-negative breast cancers less than five centimeters in size undergoing lumpectomy with whole breast irradiation who were found to have metastases in one or two sentinel nodes were randomized to ALND or no further surgery. Initial findings after a median follow-up of 6.3 years showed no survival differences and no difference in rates of nodal recurrence between groups, but side effects, including lymphedema, were significantly less common in the sentinel node biopsy group. ⁽¹⁾

This study was updated in April 2016 at the American Surgical Association annual meeting. With the median follow-up extended to 9.25 years, no differences in nodal recurrences were seen, with only two nodal recurrences in the ALND arm (0.5 percent) and five (1.1 percent) in the sentinel node–only arm. Again, no survival differences were observed.

The After Mapping of the Axilla: Radiotherapy Or Surgery? (AMAROS) study randomized a similar population with T1 and T2, clinically node-negative breast cancer, but included patients treated with mastectomy as well as breast conservation. Patients with positive sentinel nodes were randomized to ALND or sentinel node biopsy with radiation to the axilla and medial supraclavicular nodes. After a median follow-up of 6.1 years, the five-year rate of axillary recurrence was low in both groups: 0.5 percent after ALND and 1 percent after axillary radiation. There were no significant differences in disease-free survival between groups, and lymphedema was significantly less prevalent in the non-ALND group. ⁽²⁾ See Table 1 for a comparison of AMAROS and ACOSOG Z011.

Memorial Sloan Kettering Cancer Center Study: Applicability of Z011 Findings

Although the idea tested in AMAROS — that radiation of low-volume nodal disease would be an effective alternative to ALND — did not cause much controversy, the findings of the ACOSOG Z011 study generated great debate. A common criticism was that patients were a highly favorable subgroup and not representative of breast cancer patients as a whole. To test this concern, we initiated a policy of managing all patients who met ACOSOG Z011 eligibility criteria without ALND in August 2010. Patients with T1 and T2 breast cancers undergoing breast-conserving surgery and whole breast radiation therapy had ALND only if they were found to have metastases in three or more sentinel nodes or had matted nodes identified during surgery. Of 2,157 patients undergoing breast conservation between August 2010 and November 2012, 381 had sentinel node metastases. We excluded 93 patients from the study —those who had neoadjuvant therapy, conversion to mastectomy, or metastases detected only by immunohistochemistry — leaving 287 eligible patients; of these, 215 had macrometastasis. ALND was avoided in 84 percent of this initial group. ⁽⁴⁾ We have since expanded our series to 646 consecutive patients treated through July 2015, and 83 percent received sentinel node biopsy alone. Another concern regarding the application of the ACOSOG Z011 results has been the use of this approach in women with more-aggressive cancer types, since most of the patients included in ACOSOG Z011 were postmenopausal with estrogen receptor (ER) positive cancers. We examined whether women with high-risk cancers — defined as triple negative, HER2 positive, or occurring in women younger than 50 years of age — were more likely to require ALND than women lacking these features. As summarized in Table 2, ALND use did not differ between groups. In patients who required ALND, the number of positive nodes was not greater in the high-risk group, indicating that being at high risk for distant metastases does not necessarily mean being at high risk for a heavy axillary nodal tumor burden. With a median follow-up of 27 months (range 2 to 63 months), there have been no isolated axillary recurrences in the 614 patients managed without ALND in the MSK series.

*High risk: Triple negative, HER2 positive, age < 50 years

Implications for Clinical Practice

The long-term results of ACOSOG Z011 confirm the safety of this approach, and our MSK study confirms its widespread applicability. Implementation has necessitated some changes in clinical practice. For example, we do not use axillary ultrasound to screen clinically node-negative women. It was not a part of the ACOSOG Z011 eligibility criteria, and when we examined whether the identification of abnormal nodes on axillary ultrasound identified a population requiring ALND, only 30 percent of those with abnormal nodes visualized had metastases in three or more sentinel nodes (see Table 3) and required ALND. ⁽⁵⁾ In a study of 153 patients with positive axillary node needle biopsies, 49 percent had metastases in only one or two sentinel nodes and did not require ALND. ⁽⁶⁾ Based on these findings, axillary ultrasound and needle biopsy in clinically node-negative women represent unnecessary procedures that do not change the treatment plan and increase the cost of care.

Since the finding of a single positive node is no longer enough to justify the routine use of ALND, we have also eliminated the use of intraoperative frozen sections of sentinel nodes in Z011-eligible patients. Nodal irradiation, as in the AMAROS trial, is considered for patients with microscopic extracapsular extension, lymphovascular invasion, and a heavier total tumor burden as manifested by larger tumor size and involvement of two sentinel nodes. This tailored approach to each patient’s individual tumor burden has reduced the morbidity of surgery for a large number of women and is a significant advance in the local therapy of breast cancer.