SAN FRANCISCO – Regional recurrence is rare in breast cancer patients with a negative sentinel lymph node biopsy and does not appear to be associated with reduced overall survival, based on the latest results of a clinical trial including more than 5,500 patients.
"Regional recurrences were very low and remained so over the follow-up period," Dr. Kelly K. Hunt said at the annual meeting of the American Surgical Association. The 3-year cumulative incidence rate for regional recurrence was 0.003, and the rate at 5 years was 0.005. In comparison, the 3-year and 5-year rates for local recurrence were 0.013 and 0.024, respectively; the 3-year and 5-year rates for distant recurrence were 0.017 and 0.028, respectively.
Dr. Kelly K. Hunt
Sentinel lymph node (SLN) dissection has been shown to improve staging, compared with axillary lymph node dissection (ALND) alone, and reduced morbidity and improved quality of life have been associated with SLN dissection. Thus, SLN dissection has replaced ALND in clinical practice for staging of regional lymph nodes in women with clinically node-negative, early-stage breast cancer. However, studies have shown false-negative rates ranging from 5% to 17%, noted Dr. Hunt of the department of surgical oncology at the MD Anderson Cancer Center in Houston.
She and her fellow researchers analyzed data from the American College of Surgeons Oncology Group (ACOSOG) Z0010 trial, a prospective multicenter trial designed to evaluate the prognostic implications of SLN and bone marrow micrometastases in women with early-stage breast cancer.
Eligible patients had biopsy-proven T1 or T2 breast cancer with clinically negative lymph nodes and were candidates for lumpectomy and whole breast irradiation. A total of 5,539 patients were enrolled from 126 participating sites. A total of 5,200 patients were fully evaluable, and a sentinel node was identified in 5,119.
Of those, 3,904 patients had negative lymph nodes based on standard H&E staining. Of the patients with negative SLNs on H&E staining, there were 3,326 who had sufficient samples for immunohistochemical staining for cytokeratin. Ten percent of these patients had positive nodes on immunohistochemistry (IHC).
In outcomes published last year, women with H&E positive sentinel nodes had decreased survival compared with those with H&E negative sentinel nodes (JAMA 2011;306:385-93 [doi:10.1001/jama.2011.1034]).
Patients were largely postmenopausal women with ductal breast cancers. They had mostly T1 tumors, many of which were hormone receptor–positive, low to intermediate grade. The majority of patients (98%) had breast-conserving surgery, and 68% patients received hormonal therapy as part of adjuvant treatment. Most patients (92%) finished adjuvant radiation, and almost half (44%) also received systemic chemotherapy. The median follow-up was 8.4 years. There were 127 local failures, 20 regional recurrences, and 134 distant recurrences. There were 317 deaths.
"Within this small group of locoregional recurrences, we did look at the differences in the patients who had IHC-positive vs. IHC-negative sentinel nodes and we did not see any difference between those two groups," Dr. Hunt said. As a result, they combined the two groups for subsequent analyses. "We did not see a difference in recurrence rates based on IHC status, despite the fact that we would suspect that some of these patients have residual disease in the axilla based on published false-negative rates from multicenter trials."
They also examined clinical factors, pathologic factors, and treatment factors that may have affected local and regional recurrences on univariable and multivariable analyses. In both models, younger age and negative hormone receptor status predicted both a greater rate of local failures and locoregional recurrences. With respect to distant recurrence, the important factors were increasing tumor size, the presence of lymphovascular invasion, and higher tumor grade.
Older age, larger tumor size, higher tumor grade, and the presence of a local recurrence predicted reduced overall survival. The hazard rate for local recurrence alone on multivariable analysis was 6.73 (P less than .0001).
"This study reassures anyone practicing in this field ... that those early clinical judgments we all made that drove early adoption of sentinel lymph node axillary staging and early, occasionally angst-ridden decisions to forgo axillary dissection in patients with early-stage breast cancer and minimal axillary nodal disease ... that those choices were correct," said invited discussant Dr. Barbara L. Bass, the John F. and Carolyn Bookout Distinguished Endowed Chair in the department of surgery at the Methodist Hospital in Houston and a professor of surgery at Cornell University in New York.
The regional recurrence rates are particularly interesting, she noted. "It practically makes you think that the axilla – our current bedrock for decision making and staging – is in fact a hostile soil for tumor cells, compared to the breast and distant sites." She asked why the axilla was practically immune from recurrence. "Clearly, those lymph nodes in that space were still there, but tumor does not go or grow there."