The Halsted radical mastectomy included an en bloc resection of axillary contents. Now, more than a century later, there is little controversy about the use of breast-conserving surgical treatments in place of mastectomy, while the routine use of axillary dissection, or at least axillary sampling, has persisted despite the fact that it is the greatest source of morbidity from breast surgery. Is it time to abandon this procedure all together, at least in some patients? Halsted and his successors removed and/or irradiated lymph nodes because they thought that breast cancer always spreads regionally before metastasizing distantly, and that all regional disease had to be removed to improve survival. The National Surgical Adjuvant Breast and Bowel Project (NSABP) demonstrated in trial B04 that this is not the case. 1 One thousand seventy-nine patients without clinical evidence of axillary adenopathy were randomly assigned to simple mastectomy, a radical mastectomy with lymph node dissection, or simple mastectomy with radiation therapy to the chest wall and axilla. Sixty-eight (19%) of the patients randomly assigned to receive no surgery or radiotherapy to the axilla eventually developed positive axillary nodes. This was about half of the rate expected from the incidence of positive nodes in the radical mastectomy arm of the study, and proved that disease left behind can be kept in check or eradicated by the body’s own defense mechanisms. It was possible to remove the axillary recurrence in all but one of the 68 patients, thus removing the argument that axillary recurrences cannot be managed surgically. Most importantly, with 25 years of follow-up, no significant survival differences have emerged. There have been no randomized trials of similar or larger size addressing the role of axillary surgery without axillary radiotherapy subsequent to NSABP B04. In this issue of the Journal of Clinical Oncology, the International Breast Cancer Study Group (IBCSG) 2 reports the result of a trial in which older patients (defined as those 60 years of age) who were to be treated with tamoxifen regardless of nodal status were randomly assigned to receive clearance of the axillary lymph nodes or no surgical intervention in the axilla after either mastectomy or lumpectomy. Among those randomly assigned to axillary clearance, 28% were node-positive. Radiotherapy was not recommended for patients treated with mastectomy; 60% of those treated with breast-conserving therapy (33% of all patients in the study) had radiotherapy. There was significantly more restriction in arm movement and arm pain during the first few months after treatment among those randomly assigned to axillary clearance, but these differences were small and not significant after longer followup, and there were no differences in overall quality of life parameters. There were no differences in disease-free or overall survival, but there were only 473 patients in the study and the median follow-up was just 6.6 years. This study was originally designed to be much larger, but because of poor accrual the study goals were amended to focus on quality-of-life issues. It seems plausible that the poor accrual reflects a lack of uncertainty by the patients and physicians involved in this trial about the value of axillary node dissection in this patient population. Because of the small size of this study, champions of axillary