Axillary Dissection Unnecessary in Select Small Breast Cancers

[Hopeful]

CHICAGO (EGMN) - Axillary dissection might not be necessary after sentinel node biopsy to prevent locoregional recurrence in women with T1 and T2 breast cancer and sentinel node metastasis, data from the prospective American College of Surgeons Oncology Group Z0011 trial suggest.

At a median follow-up of 6.3 years, locoregional recurrence was 2.8% in 436 women randomized to sentinel lymph node dissection (SLND) alone and 4.1% in 420 women randomized to axillary lymph node dissection (ALND) after sentinel node dissection (P = .11).

Locoregional recurrence rates remained the same when a second analysis was performed on the 388 ALND and 425 SLND-only women who received actual treatment (P = .47), principal investigator Dr. Armando E. Giuliano said at the annual meeting of the American Surgical Association. Notably, 106 (27.3%) of patients treated with ALND had additional positive nodes removed beyond the sentinel node.

"A counterintuitive conclusion is inescapable—sentinel node biopsy alone provides excellent locoregional control comparable to completion axillary lymph node dissection in node-positive women with T1 or T2 breast cancer who are treated with whole-breast irradiation and adjuvant systemic therapy," he said.

The trial was underpowered to determine overall survival, after closing early because of low accrual and event rates, said Dr. Giuliano, chief of science and medicine at the John Wayne Cancer Institute in Santa Monica, Calif.

Women from 177 institutions were eligible for the Z0011 trial if they had clinical T1 or T2 N0 breast cancer and hematoxylin and eosin stain-detected metastases in the sentinel node and had undergone lumpectomy with whole-breast irradiation. Women were ineligible if they received nodal-specific, third-field irradiation, had metastases in the sentinel node detected by immunohistochemistry, had matted nodes, or had three or more involved sentinel nodes.

Overall, 70% of participants were clinical stage T1, 83% were estrogen receptor-positive, and two-thirds were progesterone receptor-positive. A modified Bloom-Richardson score of III was reported in 29% of ALND patients and in 27.5% of SLND patients. The median age in each group was 56 years and 54 years, respectively.

In a multivariate analysis, the only independent predictors of locoregional recurrence were age 50 years or younger and a Bloom-Richardson score of III. The number of positive sentinel nodes, size of sentinel node metastasis, and number of lymph nodes removed were not associated with locoregional recurrence, Dr. Giuliano said.

Women randomized to SNLD plus ALND had a median of 17 axillary nodes removed, versus a median of just 2 sentinel nodes removed with SNLD only (P less than .001), he said. Just one positive lymph node was removed in most patients in the ALND (58%) and SLND (71.1%) groups in the intent-to-treat analysis, although 21% and 3.6% of patients, respectively, had three or more positive nodes removed.

Invited discussant Dr. Blake Cady said the Z0011 trial is reaffirms the 1970s finding from the Protocol B-04 trial that not all nodal metastases recur clinically without treatment, and that axillary dissection may not be necessary to prevent regional node recurrence in today's smaller breast cancers. Tumor size in the trial ranged from 1.6 cm-1.8 cm.

"We can now reduce to a minimum surgical morbidity after breast conservation," he said. "Axillary dissection accomplished nothing for T1, T2 clinically N0 breast cancers for regional control."

Despite the study's lack of power, the results can be extrapolated to overall survival since regional-free survival was equivalent at 8 years at 0.5% in the ALND arm and 0.9% in the SLND arm, said Dr. Cady, professor of surgery at Brown University in Providence, R.I. He noted that two extensive reviews of nodal surgery in various cancers, including breast cancer, have shown that overall survival cannot be improved over observation simply by removing an increasing number of nodes during surgery. In addition, a newer biological understanding of the organ specificity of metastatic disease suggests that lymph node-specific metastases shed nodal-specific cells that are unable to grow in vital organs and thus influence survival.

Dr. Giuliano agreed that the spread of metastatic disease might not be attributable to biologic factors, but he said that the counterintuitive locoregional findings might be explained by other factors, such as treatment of nearly the entire axilla in patients receiving opposing tangential field whole-breast irradiation and the use of adjuvant systemic chemotherapy in 96%-97% of patients. Even with target accrual, the trial could not have shown equivalency between ALND and SLND because of the low regional recurrence seen in only 0.7% of the entire cohort, Dr. Giuliano said.
Recently however, a secondary analysis from the MIRROR (Micrometastases and Isolated Tumor Cells; Relevant and Robust or Rubbish?) study of 2,680 breast cancer patients with isolated tumor cells and micrometastases of the sentinel node demonstrated more than a fourfold increase in subsequent axillary recurrence in those with micrometastases and no axillary therapy (The Oncology Report 2009; Fall:7). In the Z0011 trial, micrometastases 0.2 mm or less were present in 37.5% of ALND and 44.8% of SLND patients.

When asked in an interview about the impact of micrometastases in his trial, Dr. Giuliano said, "These findings suggest that nomograms to determine potential benefit from axillary dissection or dissection for patients with micrometastases are unnecessary despite nonrandomized database analyses suggesting otherwise."

During the discussion, an audience member asked whether the results represent a new standard of care in breast cancer and are applicable to other patients such as those with T3 cancers. Dr. Giuliano responded that any study is a sample of patients and that the results are not applicable to women treated with partial breast irradiation, those with T3 cancers since they were excluded from the trial, or those treated with mastectomy, since they do not get opposing tangential field irradiation.

Dr. Cady questioned whether clinicians should avoid axillary dissection and if the study should be repeated to achieve a larger accrual. Dr. Giuliano said Z0011 is the largest prospective, randomized trial of patients with node-positive breast cancer treated with and without axillary dissection, and need not be repeated to provide clinicians with guidance. Based on the 95% confidence interval of .31-1.39 for locoregional recurrence, he said the worst-case scenario is that sentinel lymph node biopsy alone would increase the risk of recurrence by only about 1.4%, compared with axillary dissection.

"I do think that we will have to examine the impact on survival, which we are currently analyzing, before we can totally abandon the operation, but I think that axillary dissection must be reanalyzed and probably omitted in the management of contemporary breast cancer for most patients," he concluded.

Neither Dr. Giuliano nor Dr. Cady reported any disclosures.

Hopeful