Sentinel Lymph Node Biopsy for Melanoma: American Society of Clinical Oncology and Society of Surgical Oncology Joint Clinical Practice Guideline

Abstract

Purpose

The American Society of Clinical Oncology (ASCO) and Society of Surgical Oncology (SSO) sought to provide an evidence-based guideline on the use of lymphatic mapping and sentinel lymph node (SLN) biopsy in staging patients with newly diagnosed melanoma.

Methods

A comprehensive systematic review of the literature published from January 1990 through August 2011 was completed using MEDLINE and EMBASE. Abstracts from ASCO and SSO annual meetings were included in the evidence review. An Expert Panel was convened to review the evidence and develop guideline recommendations.

Results

Seventy-three studies met full eligibility criteria. The evidence review demonstrated that SLN biopsy is an acceptable method for lymph node staging of most patients with newly diagnosed melanoma.

Recommendations

SLN biopsy is recommended for patients with intermediate-thickness melanomas (Breslow thickness, 1–4 mm) of any anatomic site; use of SLN biopsy in this population provides accurate staging. Although there are few studies focusing on patients with thick melanomas (T4; Breslow thickness, >4 mm), SLN biopsy may be recommended for staging purposes and to facilitate regional disease control. There is insufficient evidence to support routine SLN biopsy for patients with thin melanomas (T1; Breslow thickness, <1 mm), although it may be considered in selected patients with high-risk features when staging benefits outweigh risks of the procedure. Completion lymph node dissection (CLND) is recommended for all patients with a positive SLN biopsy and achieves good regional disease control. Whether CLND after a positive SLN biopsy improves survival is the subject of the ongoing Multicenter Selective Lymphadenectomy Trial II.

Copyright © 2012 American Society of Clinical Oncology and Society of Surgical Oncology. All rights reserved. No part of this document may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without written permission by the American Society of Clinical Oncology and Society of Surgical Oncology.

Appendix

Technical Considerations for Sentinel Lymph Node Biopsy

The success of a sentinel lymph node (SLN) biopsy is dependent on an interdisciplinary relationship between nuclear medicine, surgery, and pathology. Lymphoscintigraphy after injection of the radiocolloid agent is important not only for the identification of the SLN within the draining basin but also for the identification of potentially involved nodal basins. A number of vital blue dyes have been used for lymphatic mapping in conjunction with a radiocolloid agent. Identification of micrometastases is dependent on a thorough pathologic assessment, including serial sections and immunohistochemistry.

Preoperative Lymphoscintigraphy

Preoperative lymphoscintigraphy is typically performed in the nuclear medicine department preoperatively to allow for surgical planning. Lymphatic drainage from the site of a primary melanoma can be variable, especially in the head and neck or truncal regions. Drainage to multiple nodal basins may be identified, and lymphoscintigraphy should be used to guide the appropriate biopsy of all involved nodal basins and to guide the intraoperative identification of interval (in-transit) nodes, which can be the only site of nodal metastases.

A four-point intradermal injection of 0.05–1 mCi of technetium 99-labeled sulfur colloid (^99mTc-sulfur colloid) at the primary melanoma site is administered at the time of preoperative lymphoscintigraphy. Real-time images are then obtained to visualize the nodal basins. Most centers perform the lymphoscintigram on the day of surgery. There is enough sufficient residual radioactivity to detect an SLN several hours later because of the 6-hour half-life of ^99mTc-sulfur colloid.

When the primary tumor is close to the nodal basin (especially in the neck), it may be difficult to determine the discrete drainage pattern. In these cases, additional anatomic views can assist in separating the radioactivity in the nodal basin from that of the primary tumor injection.

Radio colloid agents: there is variation in the radiocolloids used across institutions. ^99mTc-sulfur colloid is used in the United States; ^99mTc-nanocolloid and ^99mTc-antimony trisulphide colloidal preparations are used in many centers outside the United States. In general, the smaller the particle size, the faster it will travel and the greater the number of nodes demonstrated. It is for this reason that many institutions in the United States filter the colloidal preparation before dispensing through a 0.22-micron filter to ensure a more consistent particle size in the injectate. Intradermal injection is preferred by most centers, because this most closely mimics the potential passage of malignant cells. Insufficient tissue tension after injection (as may be seen with a subcutaneous injection) will lead to a delay in drainage. To avoid compressing the dermal lymphatics, it is important that injected volumes are kept quite small, with volumes of approximately 0.1 mL preferred. Of note, tilmanocept (Lymphoseek; Navidea Biopharmaceuticals, Dublin, OH) is a lymphatic mapping agent that is under development and has been tested in phases II (Leong SP, Kim J, Ross M, et al.: Ann Surg Oncol 18:961–969, 2011) and III trials (Cope FO, Sondak VK, Wallace AM: J Clin Oncol 29:532s, 2011 [suppl; abstr LBA8526]).

Radiation safety aspects: both gamma cameras and gamma probes are exquisitely sensitive, such that very small amounts of radioactivity are needed to perform the procedure successfully. Doses administered range from 0.05 to 1 mCi. These doses are approximately 1/20 of the dose given for a typical ^99mTc-MDP bone scan. It has been estimated that the dose to a surgeon’s finger from a single SLN surgery is 1/30 of the yearly whole-body absorbed dose from background radiation (Alazraki N, Glass EC, Castronovo F, et al.: J Nucl Med 43:1414–1418, 2002).

Imaging: almost all centers perform gamma camera imaging before surgery in patients with melanoma after injection of radiocolloids to define involved nodal basins. This is particularly the case in distal upper and lower limb melanomas in which an epitrochlear or popliteal node may be involved, truncal melanomas in which contralateral rather than ipsilateral nodal basins are found to be involved, and head and neck melanomas in which pre-auricular, intraparotid, or suboccipital nodes may be involved before nodes in the cervical chain or supraclavicular fossa are involved.

Many centers perform dynamic imaging to determine nodes that receive direct lymphatic drainage. If dynamic imaging is not performed, there is a risk that an end-on lymphatic channel may be misidentified as a node on a single planar image. There are a variety of approaches to assist in the localization of nodes, including the use of cobalt-57 flood sources to outline the body’s anatomy, external outlining of the body’s surface using a hot source that is traced over the body’s surface, and use of hybrid low-dose single-photon emission computed tomography (SPECT-CT) imaging (Even-Sapir E, Lerman H, Lievshitz G, et al.: J Nucl Med 44:1413–1420, 2003). Many centers perform skin marking to identify nodes involved. If this is done, it should be performed in the expected operative position.

Head and neck melanomas should be evaluated with a SPECT-CT device whenever possible, because the combination of the anatomy demonstrated by the CT and SPECT images of the colloid allows very precise localization of the nodes demonstrated as well as the identification of nodes immediately adjacent to the injection site. These images can assist in the planning of the surgical incision/approach and have been shown to alter the surgical approach in between 20 and 50 % of patients compared with planar imaging (Bilde A, Von Buchwald C, Mortensen J, et al.: Acta Otolaryngol 126:1096–1103, 2006; Vermeeren L, Valdés Olmos RA, Klop WM, et al.: Head Neck 33:1–6, 2011). As with any presurgical planning, good communication between the surgeon and imaging team is essential.

Technical Details of SLN Biopsy

Intraoperative lymphatic mapping and SLN biopsy are routinely performed with both preoperative ^99mTc-sulfur colloid injection, which can be detected with a handheld gamma probe and vital blue dye. In the operating room, 1–2 mL of vital blue dye is injected intradermally at the primary tumor site. Successful delivery of the dye intradermally is important, because a subcutaneous injection into the fat may not enable adequate uptake of the radioactive tracer or dye by the cutaneous lymphatic channels. The injection of blue dye is routinely performed before sterile preparation of the patient operative sites to allow 5–10 min for the dye to reach the lymph node basin.

The commercially available vital blue dyes in the United States include isosulphan blue (Lymphazurin; Tyco Healthcare Group, Norwalk, CT) and methylene blue dye. Both blue dyes are effective for lymphatic mapping but have unique side effect profiles (Liu Y, Truini C, Ariyan S: Ann Surg Oncol 15:2412–2417, 2008; Blessing WD, Stolier AJ, Teng SC, et al.: Am J Surg 184:341–345, 2002; Simmons R, Thevarajah S, Brennan MB, et al.: Ann Surg Oncol. 10:242–247, 2003). Allergic reactions, including anaphylactic reactions, have been reported with the use of isosulphan blue. In a review of 1,835 patients injected with isosulphan blue dye for a variety of surgical procedures, 1.5 % of patients had an adverse reaction (Daley MD, Norman PH, Leak JA, et al.: J Clin Anesth 16:332–341, 2004). The majority of these patients experienced minor events (e.g, skin wheals, itching, and localized edema), but 0.75 % suffered a major anaphylactic reaction (hypotension) while under anesthesia. No deaths have been reported from any of these reactions.

Methylene blue has been associated with tissue necrosis and should be used with care at surgical sites where the majority of the blue dye will not be surgically resected (e.g, face, periorbital, wrists, or ankles). Some have diluted the blue dye to decrease risk of tissue necrosis. Small amounts of residual blue dye may persist after wide local excision (WLE) of the primary site, rarely resulting in a permanent tattoo even if the dye is unable to be totally resected. In addition, because of systemic accumulation, the blue dye will be seen in urine, stool, and lactating breasts for the first 24–36 h after injection.

The handheld gamma probe is used to identify areas of focal radiotracer uptake in the nodal basins identified on preoperative lymphoscintigraphy. A small incision is made in the nodal basin, taking into consideration the incision necessary if completion lymph node dissection is subsequently required. Surgeons trace the blue lymphatic channels or follow the path of radioactivity into the SLN. Electrocautery is used to dissect away the surrounding fatty tissue. Blue lymphatic channels and vascular structures are ligated, and care is taken to not disrupt or cauterize the capsule of the SLN.

After each SLN is removed, it is checked ex vivo to document the radioactive counts per second, and the nodal basin is rescanned with the gamma probe. In general, any lymph nodes that are blue, any lymph nodes with radioactive counts ≥10 % of the ex vivo count of the most radioactive SLN, and any palpably suspicious nodes are removed (McMasters KM, Reintgen DS, Ross MI, et al.: Ann Surg Oncol. 8:192–197, 2001). There is an average of one to three SLNs per nodal basin. If there is concern of background radiation or shine through from the primary melanoma site, WLE can be performed beforehand to decrease radiotracer activity at this site.

Concomitant WLE and sentinel lymphadenectomy are preferred. However, in patients who have undergone previous WLE, the procedure is still technically feasible (Ariyan S, Ali-Salaam P, Cheng DW, et al.: Ann Surg Oncol 14:2377–2383, 2007; Evans HL, Krag DN, Teates CD, et al.: Ann Surg Oncol 10:416–425, 2003; Kelemen PR, Essner R, Foshag LJ, et al.: J Am Coll Surg 189:247–252, 1999; Leong WL, Ghazarian DM, McCready DR: J Surg Oncol 82:143–146, 2003; McCready DR, Ghazarian DM, Hershkop MS, et al.: Can J Surg 44:432–434, 2001). In a study of 104 patients at the University of Texas MD Anderson Cancer Center (Houston, TX) who underwent sentinel lymphadenectomy after previous WLE, the SLN positivity rate was similar to that of more than 1,000 patients who had concomitant WLE and SLN removal during the same time period (Gannon CJ, Rousseau DL Jr, Ross MI, et al.: Cancer 107:2647–2652, 2006). However, because extensive resection can alter lymphatic draining and may not accurately reflect the pathologic status of the draining lymph node basin, removal of the SLN at the time of primary WLE is preferred whenever possible.

Laboratory Evaluation of SLNs

Most specimens include one to three nodes considered sentinel on the basis of their blue coloration and selective radioactivity. SLN biopsy provides a limited specimen that is susceptible to a more detailed examination than is practicable for lymphadenectomy specimens that contain multiple lymph nodes.

Maximum length, width, and thickness of SLNs are measured in millimeters. SLNs are bisected through their longest meridian to detect melanoma cells that have been delivered to the subcapsular sinus from afferent lymphatics (Cochran AJ, Wen DR, Morton DL: Am J Surg Pathol 12:612–618, 1988). The cut surfaces of both halves of the SLN are closely examined for blue dye, metastatic melanoma, and foci of melanin. Imprints for cytology, if indicated, can be made at this stage. The SLN halves, or slices 2 mm thick taken parallel to the meridian in larger SLNs, should be placed (cut face down) in cassettes and fixed in formalin for 12–24 h.

Nuclear medicine physicians and surgeons are best able to determine if a node is truly sentinel. Occasional technical problems lead to misidentification of a node as sentinel: blue dye is seldom seen when specimens arrive in the laboratory, the radioactive isotope decays rapidly from the peak emission values seen in the operating room, and few laboratories have equipment or expertise to measure tissue radioactivity.

Intraoperative Assessment of SLNs

SLNs are best evaluated by examining thin sections cut from well-fixed paraffin-embedded tissues (Morton DL, Wen DR, Foshag LJ, et al.: J Clin Oncol 11:1751–1756, 1993; Stojadinovic A, Allen PJ, Clary BM, et al.: Ann Surg 235:92–98, 2002; Scolyer RA, Thompson JF, McCarthy SW, et al.: J Am Coll Surg 201:821–823, 2005; author reply 823–824). Frozen section analysis of SLNs is not performed for melanoma because of the difficulty in reliably diagnosing microscopic metastases using immediate intraoperative pathology evaluation, and because full-face sections often require disposal of many incomplete sections with potential loss of most or all diagnostic nodal tissue. Identification of single melanoma cells, small clusters of melanoma cells, or small melanoma cells that resemble nevus cells is more difficult in frozen sections.

Evaluation of Multiple Levels of the SLN

Multiple sections are cut and stained with hematoxylin and eosin (HE) for immunohistochemistry. The number of sections to be stained and the optimal distance between them remain subject to debate. Early studies suggested that early melanoma metastases are found in a band of tissue adjacent to the longest nodal meridian (Cochran AJ, Wen DR, Morton DL: Am J Surg Pathol 12:612–618, 1988). On the basis of these early studies, examination of 10 full-face serial sections from both faces of the node has been recommended.

If tumor cells are not detected in the initial sections, additional sections may be evaluated in patients considered at high risk of nodal metastases. This approach detects melanoma in 16–20 % of SLN biopsy specimens, which is close to the incidence of metastatic nodal disease in patients with melanoma after wide excision of a primary melanoma (Morton DL, Thompson JF, Cochran AJ, et al.: N Engl J Med 355:1307–1317, 2006). The European Organisation for Research and Treatment of Cancer now requires examination of six pairs of sections cut at 50-μm intervals then stained with HE and S-100 for patients with melanoma entering clinical trials (Cook MG, Green MA, Anderson B, et al.: J Pathol 200:314–319, 2003).

SLN Tissue for Research

Accurate identification of SLN melanoma metastases is essential for optimum patient management, but it may be difficult in the presence of limited and highly localized metastases. Underdetection of SLN metastases may have potential consequences for patients. Pathologists should be cautious in providing tissue for research until the SLN has been adequately sampled and the SLN tumor status established. There is, however, a legitimate need to determine whether techniques such as real-time polymerase chain reaction truly detect small amounts of clinically relevant tumor not readily identifiable by standard histopathology.

Additional research is needed regarding the molecular and cellular events that determine SLN susceptibility to metastases to be able to reverse that susceptibility. Interleaved tissue sections—one section for histology and the next for research—provide precise histologic control for biologic investigations. Research that uses formalin-fixed paraffin-embedded tissue is readily accommodated; providing unfixed tissue is more challenging. Pathologists and investigators need to understand diagnostic tissue requirements and the regulatory limitations that govern distribution of human tissues.

Immunohistochemistry

Experienced pathologists may overlook single melanoma cells or small melanoma cell clusters in up to 12 % of SLNs based on HE examination alone. Antibodies to S-100 detect nuclear and cytoplasmic epitopes in nearly all melanomas. Although staining is relatively nonspecific, with experience, melanoma cells can be distinguished with considerable consistency.

MART-I, HMB-45, and anti-tyrosinase are antibodies that detect cytoplasmic epitopes expressed by melanocyte-derived cells, including melanoma cells. These epitopes are more specific than S-100 for melanocytic lineage, but they are not expressed by up to 25 % of melanomas, particularly metastatic melanomas (Ohsie SJ, Sarantopoulos GP, Cochran AJ, et al.: J Cutan Pathol 35:433–444, 2008). Combinations of antibodies (antibody cocktails) seem no more sensitive than S-100 and do not permit separation of melanoma cells and nevocytes on the basis of their immunophenotype. Red-colored chromogens facilitate separation of melanin-containing macrophages and melanoma cells.

It is practical to assess the immunohistochemically stained sections first, because the immunomarkers highlight small numbers of melanoma cells that are less easily seen in HE preparations. An initial low-power scan to exclude large metastases is followed by a careful examination for single tumor cells and small cell clusters within the subcapsular sinus (the common site of early metastases), the internal sinuses, and finally the parenchyma. A tumor in afferent lymphatics has the same clinical significance as an intranodal tumor. Thus, it is important to carefully examine any lymphatics that are present. Extracapsular extension by a tumor should be recorded, as should size of the largest metastatic focus and location of the metastatic tumor (Frishberg DP, Balch C, Balzer BL, et al.: Arch Pathol Lab Med 133:1560–1567, 2009).

It is important to distinguish nodal nevocytes from metastatic melanoma cells. This requires detailed cytologic evaluation as well as assessment of immunophenotype and location of cells within the nodal architecture. Melanoma cells can be distinguished on the basis of their large size, high nuclear to cytoplasmic ratio, prominent nucleoli, and atypical mitoses, whereas nodal nevocytes are generally smaller, with limited cytoplasm, and seldom show mitoses. Although both cell types may contain finely dispersed small granules of melanin, the quantity of melanin in melanoma cells usually exceeds that in nevocytes. Melanoma cells are generally positive for S-100, MART-1/Melan-A, and HMB-45, and their nuclei are reactive with Ki67/MIB1. In contrast, although nevocytes may be positive for S-100 and MART-1/Melan-A, they generally stain weakly or are negative for Ki67/MIB1 and HMB-45 (Lohmann CM, Iversen K, Jungbluth AA, et al.: Am J Surg Pathol 26:1351–1357, 2002).