Jean Louis Petit (1674 to 1750), director of the French Surgical Academy, was probably the first surgeon to articulate a unified concept for breast cancer surgery (1). He emphasized the importance of an en bloc resection of the breast and axillary nodes, but his insight came too early: Although breast cancer had been considered surgically incurable for most of the 19th century, Halsted’s landmark 1894 (2) and 1907 (3) reports of his meticulous technique for “radical mastectomy” (RM)—including removal of the breast, pectoral muscles, and axillary contents—demonstrated a reduction in local recurrence (LR) and apparent cure for about one-third of patients. The concept that minimizing LR would maximize survival made RM the standard operation for the next 70 years despite later reports of more radical (extended RM) (4) or less radical (modified RM) (5) techniques. In the “Halstedian” era, the goal was to maximize cure by minimizing local failure.

In the 1970s, Fisher proposed that breast cancer was a systemic disease from the outset and that survival was largely a function of tumor biology, not extent of surgery (6). The “Fisher hypothesis” was first tested in the National Surgical Adjuvant Breast and Bowel Project (NSABP) B-04 randomized trial (1971 to 1974). At 25 years’ follow-up (7), variations in extent of axillary surgery for mastectomy patients with clinically node-positive breast cancer (RM vs. total mastectomy/radiotherapy [RT]) or clinically node-negative disease (RM vs. total mastectomy/RT vs. total mastectomy alone) yielded no differences in any category of survival. B-04 did confirm the overwhelming prognostic significance of axillary node metastasis, and ALND was incorporated into all subsequent NSABP trials for invasive breast cancer, including NSABP B-06, in which survival at 20 years (8) was comparable for mastectomy versus breast-conserving surgery. In the “Fisher” era, the primary goal of ALND was prognostication to guide systemic therapy, a secondary objective was local control, and a survival benefit was unproved.

We now know that both Halsted and Fisher were right. A remarkable series of meta-analyses from the Early Breast Cancer Trialists’ Collaborative Group (EBCTCG) confirms that breast cancer is a family of diseases with a wide spectrum of behavior (9), ranging from predominantly local (Halsted) to predominantly systemic (Fisher) phenotypes. First, the EBCTCG showed that local control and survival are related (42,000 women in 78 randomized trials) and estimated that for every four LRs prevented, one life would be saved (10). Second, they confirmed that there is no survival advantage for more radical versus less radical versions of mastectomy (3,400 women in nine trials) or for mastectomy versus breast conservation (3,100 women in seven trials) (11). Finally, they demonstrated an incremental survival benefit from the addition of systemic adjuvant therapy (chemo- or hormonal) to local treatment (8,000 to 33,000 women in six separate meta-analyses comprising 194 trials) (12). These results highlight the multidisciplinary nature of contemporary breast cancer treatment and the importance of collaboration among the disciplines of surgical, medical, and radiation oncology.

The SLN concept, first articulated by Cabanas in 1977 (13) for penile cancer and independently by Morton et al. in 1992 (14) for melanoma, is that (a) the first—or first few—regional lymph nodes draining a tumor site could predict the status of that regional node basin, (b) these nodes could be reliably mapped by injections of radiotracer and/or dye, and (c) SLN-negative patients might thereby avoid the added morbidity of regional node dissection. Krag et al. (15) in 1993 (using isotope mapping) and Giuliano et al. (16) in 1994 (using blue dye) were the first to report SLNB for breast cancer. Since then, an
extensive literature covering all aspects of the procedure and including seven randomized trials comparing SLNB with ALND has confirmed that the morbidity of SLNB is less than that of ALND, that staging accuracy is at least equivalent, and that for SLN-negative patients survival and other disease-related adverse events are comparable on long-term follow-up (17). It is now also clear that many patients with positive SLN do not require ALND; in the landmark ACOSOG Z0011 trial (18), patients with cT1–2N0 breast cancers, 1–2 hematoxylin and eosin (H&E)–positive SLN, and treated with breast conservation including whole-breast RT were randomized to SLNB alone versus SLNB plus ALND. At 10 years follow-up (19) there were no differences in any category of survival or in the rates of local, regional node or distant recurrence, even though additional positive nodes were found in 27% of ALND specimens. The International Breast Cancer Study Group (IBCSG) Trial 23–01 (20) similarly randomized patients with SLN micrometastases to SLNB versus SLNB with ALND, and at 10 years found no advantage for ALND.