The sentinel node (SN) concept as currently applied to breast cancer and melanoma is predated by the idea that a single lymph node can reflect the tumor status of an entire lymphatic basin. Famous examples include the Virchow node (left supraclavicular node to which gastric cancer spreads), the Sister Mary Joseph node (an umbilical lymph node that represents metastatic intra-abdominal spread), the Delphian node of the thyroid, and the Cloquet node of the groin.1 The concept of the SN technique as first described by Cabanas in 1977 for use in squamous cell carcinoma of the penis was based on detailed penile lymphangiographic studies that demonstrated consistent drainage of the penile lymphatics into a node located near the saphenous/femoral vein junction.2 When this so-called SN was negative for tumor, metastasis to other ilioinguinal lymph nodes did not occur. Cabanas therefore postulated that the status of the SN could be used to decide whether or not regional lymphatic clearance was necessary. Although multiple studies have since found that a fixed-location SN is an unreliable indicator of nodal status in penile cancer, this work paved the way for mapping the SN in patients with solid cancers that drain via the lymphatics.
Sentinel node biopsy (SNB) for melanoma was first described in 1992 by Morton and colleagues.3 They defined the SN as the first draining node of a tumor. If nodal spread has occurred, it will target the SN before other lymph nodes. Therefore, if the SN is tumor negative, the other nodes should be negative as well. These investigators showed that when a vital blue dye is injected around the site of the primary melanoma, the SN is the first blue-staining node in the lymphatic basin and therefore the first nodal site of lymphatic drainage from a primary cutaneous melanoma. After identification and removal of the SN, Morton et al performed a completion lymph node dissection for that nodal basin.3 They found that the pathologic status of the SN was a highly accurate predictor of the pathologic status of the entire nodal basin. These findings suggested that melanoma could be accurately staged with procedures that were far less extensive than complete nodal dissections.
The status of the axilla is the most important prognostic factor in breast cancer. Before the development of SNB, an axillary lymph node dissection (ALND) was required to stage the axilla. This procedure can be associated with significant morbidity, including nerve damage or lymphedema. In an effort to spare patients from these potential complications, attempts were made to develop a less invasive technique for identification of positive nodes in the axilla. Giuliano and colleagues4 successfully adapted SNB for breast cancer and began a pilot study in 1991. This study was reported in 1994 after 174 lymphatic mapping procedures were performed using a vital dye injected at the primary breast cancer site. SNs were identified in 114 (65.5%) of 174 procedures and accurately predicted axillary nodal status in 109 (95.6%) of 114 cases. On the basis of these results, they concluded that intraoperative lymphatic mapping could accurately identify the SN and that this technique could enhance staging accuracy while potentially avoiding the need for ALND. SNB has also been performed using radiolabeled colloid and/or blue dye and is now considered the standard of care for staging of the axilla in breast cancer.
SNB can successfully stage the axilla after any type of diagnostic biopsy and regardless of the size and location of the primary tumor. SNB can be used in males with breast cancer,5 in elderly and/or obese patients, and in patients with multicentric breast cancer.
There are few absolute contraindications to SNB for patients with clinically node-negative breast cancer. These include allergy to tracers used for lymphatic mapping, dye in pregnancy, and inflammatory breast cancer (Table 62-1).
Use of blue dye for lymphatic mapping is contraindicated by pregnancy because the ability of the dye to enter the fetal circulation and its potential effects on the fetus are not completely understood. However, radiolabeled colloids can be safely used in pregnant patients.6 Some centers do offer lymphoscintigraphy and SNB for these women; however, most surgeons prefer routine (elective) ALND if the patient is pregnant.7 Alternatively, if a patient is close to term, it is sometimes possible to wait until after the patient delivers and then perform SNB.
In patients with inflammatory breast cancer, the lymphatic channels are occluded and may not follow a normal drainage pattern due to the typically extensive lymphatic involvement. There are currently insufficient data to support performing SNB in patients with inflammatory breast cancer, and it should not be performed.
A few controversies regarding the proper use of SNB still exist. These areas of controversy will be discussed below (Table 62-2).
The incidence of ductal carcinoma in situ (DCIS) has increased since introduction of routine screening mammography; DCIS is now found in 15% to 20% of all screening mammograms.8,9 Even though the neoplastic cells in DCIS have not extended through the basement membrane, 5% to 15% of patients with DCIS will have tumor-positive SNs, and 10% to 20% will subsequently be found to have invasive cancer.10-13 However, not all patients with DCIS need SNB. We recommend SNB for patients whose DCIS is large (larger than 4 cm), palpable, extensive, high-grade, or requires mastectomy.
Although ALND is the standard of care in patients with clinically palpable axillary disease, approximately 30% of patients with clinically positive axillae will have histologically tumor-negative axillary nodes.14,15 Preoperative axillary ultrasound and ultrasound-guided fine-needle aspiration (FNA) are useful in this situation.16 Patients with positive FNA results can undergo ALND, whereas patients with negative FNA results can undergo SNB, with ALND if an SN contains tumor. However, all clinically palpable or suspicious nodes should be removed at the time of surgery, regardless of uptake of the mapping agent.
It was previously thought that prior axillary surgery disrupted the normal lymphatic drainage pattern, thereby removing SNB as an option for these patients. Recent studies have shown that reoperative SNB is possible and is dependent on the number of lymph nodes removed.17 It will become progressively more common to see patients with local recurrence after prior axillary surgery. Although we do not believe that prior axillary surgery is an absolute contraindication to SNB, the possibility of lymphatic disruption mandates use of lymphoscintigraphy in addition to dye-directed lymphatic mapping.
Prophylactic mastectomy is an accepted form of treatment for women who have a strong family history of breast cancer or are carriers of BRCA1 and BRCA2 genetic mutations. The reported risk of occult breast cancer in a mastectomy specimen is approximately 5%.18 Once the breast is removed, the ability to perform an SNB is lost. Therefore, if an invasive cancer is identified in the mastectomy specimen, axillary staging can only be accomplished with an ALND. Performing SNB at the time of prophylactic mastectomy spares the patient an additional operation for ALND if invasive cancer is later identified in the mastectomy specimen. A study by Dupont et al19 evaluated patients who underwent prophylactic mastectomy for either lobular carcinoma in situ or BRCA1/2 genetic mutations: 2 of 57 patients had metastatic nodal disease in the absence of carcinoma in the breast and then underwent a complete axillary node dissection. Additionally, 2 patients had invasive breast cancer with negative SNs and were spared an ALND. Therefore, 7% (4 of 57) of patients in this study had a change in their surgical management as a result of these findings.19 SNB should be offered to all women undergoing prophylactic mastectomy who have an increased risk for invasive breast carcinoma, although it is often not helpful.
Neoadjuvant chemotherapy (NAC) can downstage large or locally advanced breast cancers, thereby allowing breast-conserving surgery. The exact timing and optimal use of SNB in this setting continues to be a matter of debate. Those in favor of SNB before NAC argue that axillary staging is critical to determine which patients will subsequently need an ALND or axillary radiation.7 It is thought that NAC results in excessive fibrosis of the tumor-involved lymphatics and that obstruction of these lymphatic channels with cellular debris or tumor emboli may lead to inaccurate lymphatic mapping.20 A retrospective study of the National Surgical Adjuvant Breast and Bowel Project (NSABP) B-27 NAC trial demonstrated that the overall rate of SN identification was 84.8%. The SN identification rate was 90% with a radiocolloid mapping agent, 77% with blue dye, and 88% with both agents. The false-negative rate (FNR) was 14% with blue dye, 5% with radiocolloid, and 9.3% with both agents.21 A recent meta-analysis by Xing et al22 found that the SN identification rate after NAC was 90% and the FNR was 12%. These results are inferior to those usually seen for SNB and suggest that SNB performed after NAC is not always reliable. An American Society of Clinical Oncology (ASCO) panel from 2005 has concluded that there are insufficient data to suggest appropriate timing of SNB for patients receiving NAC.23 Furthermore, there is no evidence that a tumor-negative SNB specimen after NAC eliminates the need to treat the axilla. At our institution, we commonly perform SNB before NAC because we believe this more accurately stages the axilla, and those patients who are SN negative can avoid axillary treatment.
Preoperative lymphoscintigraphy can identify the lymphatic drainage basin and approximate the location of the SN(s). It is very helpful in patients with melanoma because skin lesions, particularly those on the trunk or the head/neck, may have multiple possible drainage patterns. However, preoperative lymphoscintigraphy is not routine for SN localization in patients with breast cancer. Although it can identify internal mammary SNs, the clinical importance of nonaxillary SNs remains controversial.24 Borgstein et al25 reported that the use of an intraoperative gamma probe was superior to preoperative lymphoscintigraphy in identifying radioactive nodes. Intraoperative use of a hand-held gamma probe is sufficient for identification of radioactive SNs, but preoperative lymphoscintigraphy may be performed, especially if extra-axillary SNs are sought.
The technique of SNB has been described using radiolabeled colloid, blue dye, or a combination of the two. Krag et al26 first described the use of a radioisotope and a handheld gamma probe to identify the SN in 1993. The SN was identified in 18 of 22 patients in this study, for an identification rate of 82%. In 1994, Giuliano et al4 reported his experience using 1% isosulfan blue in 172 patients with T1 to T3 breast cancers. This group reported an identification rate of 65.5% (114 of 174), and the nodal status was accurately predicted in 109 of 114 cases (95.6%). Sentinel nodes identified in the last 87 procedures were 100% predictive. Albertini et al27 in 1996 reported an identification rate of 92% using a combination of both of these techniques. A recent meta-analysis evaluating these 3 techniques reviewed 69 studies and over 8000 patients who underwent SNB. This study found that the success rate for SN identification was 89.2% with radioisotope alone, 83.1% with blue dye alone, and 91.9% when both techniques were used. The FNRs were 8.8%, 10.9%, and 7.0%, respectively.28 This suggests that the combination technique is the most accurate and reliable in inexperienced hands. Finally, in 1999, Morrow et al29 reported the only randomized trial comparing blue dye with radiolabeled colloid. Patients with clinical T1 or T2 tumors and negative axilla were randomized to SNB with blue dye (n = 50) or blue dye plus radiolabeled colloid (n = 42). The SN predicted the status of the axilla in 96% of cases. There was no difference in the SN identification rate between the 2 groups (88% for blue dye alone, 86% for blue dye plus radiolabeled colloid). These authors concluded that there was no advantage to using blue dye plus radiolabeled colloid over blue dye alone, even for surgeons learning the technique. At our institution, we use blue dye alone, unless the patient has had prior axillary surgery, has a contraindication to dye (eg, allergy or pregnancy), or we seek extra-axillary SNs. However, the technique used should be determined by the training and experience of the surgeon.
In the United States, technetium 99m sulfur colloid and isosulfan blue dye (Lymphazurin 1%, United States Surgical, a division of Tyco Healthcare Group LP, Norwalk, CT) are the most widely used agents for lymphatic mapping.20 Isosulfan blue dye has the disadvantage of being associated with a 1% to 3% incidence of allergic and anaphylactic reactions such as urticaria, rash, blue hives, pruritus, and hypotension.30,31 The American College of Surgeons Oncology Group Z0010 study, a prospective multicenter trial designed to evaluate the prognostic significance of micrometastases in SNs, reported anaphylaxis in 0.1% of patients (5 of 4975) in whom isosulfan blue dye was used.32 Methylene blue is equivalent to isosulfan blue, is less expensive, and has a lower risk of allergic reactions. However, it can cause skin and nipple necrosis when injected intradermally and as such, it must be used in a 1:2 dilution.33,34
When initially described by Giuliano et al,4 blue dye was injected in a peritumoral fashion, or if the primary tumor had been excised, it was injected into the wall of the biopsy cavity and surrounding breast parenchyma. In 1997, Veronesi et al35 reported that subdermal injection of radioisotope resulted in an identification rate of 98.2% and an FNR of 4.7%. It is now known that because the breast and its overlying skin have the same axillary lymphatic drainage pattern, the mapping agent can be injected at intradermal, subdermal, periareolar, or subareolar sites.36 Subareolar injection is expeditious and accurate for multicentric and unicentric disease.37 However, subareolar and dermal injections of blue dye can cause persistent discoloration of the breast that can last for several months,20 and this method cannot evaluate for nonaxillary nodes. Several studies have reported high identification and concordance rates for subareolar injection of blue dye and/or radioisotope.38-40 A study by Rodier et al41 in 2007 using both blue dye and radioisotope found that periareolar injection was equivalent to peritumoral injection in SN detection. Multiple other studies have supported the finding that identification rates using subareolar and peritumoral injection are similar.39,42,43 Therefore, surgeons should use the technique that is most familiar to them.