The role of elective lymph node dissection in the treatment of patients with early-stage melanoma remains controversial. Some surgeons advocate the routine use of elective node dissection in patients with intermediate-thickness primary tumors, but the cost, morbidity, and low yield of tumor-positive lymph nodes associated with this approach make it less appealing than wide excision and observation. Multiple retrospective studies suggest a survival advantage as high as 25% for patients undergoing elective node dissection in the setting of clinically negative nodes, as opposed to delayed node dissection for clinically evident nodal metastases. Although two randomized prospective studies failed to demonstrate a survival advantage in patients undergoing elective node dissection, as compared with those having wide excision alone, both studies were criticized for their design [1,2].
The role of elective lymph node dissection in the treatment ofpatients with early-stage melanoma remains controversial. Somesurgeons advocate the routine use of elective node dissectionin patients with intermediate-thickness primary tumors, but thecost, morbidity, and low yield of tumor-positive lymph nodes associatedwith this approach make it less appealing than wide excision andobservation. Multiple retrospective studies suggest a survivaladvantage as high as 25% for patients undergoing elective nodedissection in the setting of clinically negative nodes, as opposedto delayed node dissection for clinically evident nodal metastases.Although two randomized prospective studies failed to demonstratea survival advantage in patients undergoing elective node dissection,as compared with those having wide excision alone, both studieswere criticized for their design [1,2].
In 1992, Morton and associates described their initial experiencewith a minimally invasive technique to determine the presenceof occult regional lymph node metastases . This technique,intraoperative lymphatic mapping and selective lymph node dissection,was devised as an alternative to elective node dissection in patientswith intermediate-thickness melanoma. The technique employs avital blue dye to map the regional lymph nodes and identify occultlymph node metastases in the primary ("sentinel") node(s).Only those patients with occult tumor cells in the sentinel node(s)are subjected to a complete regional lymph node dissection; patientswith tumor-free sentinel nodes are spared the cost and morbidityof complete node dissection.
Intraoperative lymphatic mapping is based on the hypothesis thateach skin site drains via the dermal lymphatics to the adjacentlymph node basin and to one or two particular sentinel nodes.Thus, metastases to the regional nodes spread in a direct sequentialfashion from the primary site. The presence of skip metastasesis rare. Only when the sentinel node(s) contain(s) tumor is therea risk of disease in the other regional lymph nodes. When thesentinel node(s) is(are) free of tumor, the remaining nodes inthe same lymph node basin are unlikely to contain metastases.
Steps in the Sentinel Node Technique
The sentinel node technique is performed in three steps, beginningwith preoperative lymphoscintigraphy to identify the regionalbasin and sentinel node(s) at risk for metastases. We performlymphoscintigraphy using a variety of radiopharmaceuticals, includingtechnetium-labeled human serum albumin (HSA), or sulfur or albumincolloid . In their review, North and Spellman recommend preoperativelymphoscintigraphy only in those patients who have primary melanomason the torso or at sites that may be expected to have ambiguousdrainage patterns. However, at the John Wayne Cancer Institutewe use preoperative lympho-scintigraphy in all patients in orderto determine the regional lymph node basin at risk and to mapthe site of the sentinel node(s). It is important to realize thatthe location of the sentinel node(s) often cannot be determinedby strict anatomic guidelines. Moreover, the sentinel node oftenis not the lymph node closest to the primary lesion.
The second step in the sentinel node procedure is intraoperativelymphatic mapping. We use isosulfan blue dye (patent blue V isno longer available in the United States), injecting approximately0.5 to 1 mL intradermally adjacent to the primary lesion or aroundthe biopsy site. The injection can be repeated every 15 to 20minutes if the sentinel node is not found. While larger volumesof dye can be used, as noted by North and Spellman, it is importantnot to use excess dye when the sentinel node is anatomically closeto the primary, as permeation of dye through soft tissue can obscuremapping of the lymphatics. It is equally important that the dyenot be injected into the excisional biopsy cavity. We have notedstasis of the dye when injections are made in this manner.
Following injection of the blue dye at the primary site, an incisionis made overlying the regional lymph node basin and the sentinelnode(s) is(are) located and dissected. We have found that preoperativelymphoscintigraphy is useful for determining not only the siteof the sentinel node(s) but also the approximate time it takesthe blue dye to move from the primary site to the sentinel node(s).The blue-stained lymphatics are followed from the edge of thewound and traced to the primary draining (sentinel) node(s). Eachsentinel node is isolated from the adjacent tissue and excised.
The third step in the sentinel node technique is pathologic examinationof the sentinel node(s). Although Morton's group originally employedfrozen-section analysis using both hematoxylin and eosin (H &E) staining and immunohistochemical techniques, we have foundapproximately a 5% incidence of false-negative sentinel nodeswith frozen-section analysis. Therefore, we now examine permanentsections, first by conventional techniques with bivalving of thesentinel node(s) and examination of both faces. We then use immunohistochemicalstaining when H & E staining is negative.
Newer molecular biology techniques, such as reverse transcriptase-polymerasechain reaction (RT-PCR), can be used to recognize specific melanomagene sequences and may further enhance the sensitivity of thesentinel node technique by detecting submicroscopic metastases. However, RT-PCR is costly and labor-intensive and requiresa laboratory that is specialized in handling minute samples ofmessenger RNA (mRNA). Even under the best conditions, mRNA contaminationis still possible. Moreover, the significance of a sentinel nodethat is RT-PCR-positive and the rate of false-positive resultswith this technique are unknown.
Results With the Technique
In 1992, Morton and associates reported their initial experiencewith intraoperative lymphatic mapping and selective sentinel lymphadenectomy[3,6]. The sentinel node was detected in 82% of the first 237regional lymph node basins dissected. By performing complete nodedissections, Morton and colleagues were able to determine an accuraterate of false-negative sentinel nodes; less than 1% of nonsentinelnodes were the exclusive site of metastases--a false-negativerate of less than 1% for this technique.
Since Morton's original description of the sentinel node technique,a number of investigators have reported their experience withthis procedure [7,8]. These studies have verified its high degreeof accuracy for the detection of sentinel nodes, low incidenceof false-negative sentinel nodes, and low rates of regional recurrence.
To improve the success rates for locating the sentinel node(s),several investigators have modified the mapping technique throughthe use of radiopharmaceuticals and a hand-held gamma probe. Asmentioned by North and Spellman, a study of 106 melanoma patientsreported by Albertini and associates from the University of SouthFlorida demonstrated that the gamma probe helped find sentinelnodes that were not demonstrated by blue dye alone . A totalof 16 patients had probe-identified sentinel nodes that did notstain blue; in 2 cases, these nonstained sentinel nodes containedmetastases.
We have reviewed our experience in 32 patients who underwent amapping technique in which we combined the blue dye with technetium-labeledHSA . The concordance between preoperative lymphoscintigraphyand intraoperative findings was 100%. In addition, the gamma probehelped identify additional sentinel nodes that did not absorbthe dye well. However, we did not find any additional tumor-containingnodes with this technique.
An Alternative to Traditional Management
We believe that intraoperative lymphatic mapping with selectivelymph node dissection represents an alternative approach to thetraditional management of patients with intermediate- thicknessprimary melanoma. Although our initial experience demonstrateda steep learning curve for mastering this technique, other investigatorshave been able to quickly gain experience and achieve a high degreeof accuracy by using the original guidelines that we proposed.
More-recent studies using the hand-held gamma probe suggest thatthis device may enhance the surgeon's ability to find the sentinelnode. We typically use technetium-labeled HSA in combination withthe blue dye or inject HSA immediately prior to dye injection.This technique is particularly useful when one is first learningthe sentinel node procedure and in basins such as the axilla whereblue-stained nodes are often difficult to locate. However, radio-pharmaceuticalscan also produce ambiguous results; we see great variation incount ratios between blue-stained sentinel nodes and nonstainednodes. Even though sentinel node(s) may have a higher ratio ofradioactivity, the presence of high counts in adjacent nonsentinelnodes may be misleading.
Ongoing Randomized Trial
A major international randomized trial, the Multicenter SelectiveLymphadenectomy Trial, is now in progress to determine whetherthere is a survival advantage when selective sentinel lymphadenectomyis added to wide excision for patients with intermediate- thickness(1- to 4-mm) primary melanoma. Only those patients who have atumor-containing sentinel node undergo complete node dissection.A secondary aim of this study is to determine whether the sentinelnode technique improves the staging of patients with cutaneousmelanoma. Patients with regional lymph node metastases are encouragedto receive interferon alfa-2b (Intron A).
The results of two randomized prospective studies examining therole of elective lymph node dissection will soon be available.Even if these trials fail to demonstrate a survival advantagefor patients undergoing elective node dissection in addition towide excision, sentinel lymphadenectomy will likely become thestandard of care in patients with intermediate-thickness melanoma,in order to improve the staging of their disease and identifycandidates for adjuvant therapy.
1. Veronesi U, Adamus J, Bandiera DC, et al: Inefficacy of immediatenode dissection in stage I melanoma of the limbs. N Engl J Med297: 627-630, 1977.
2. Sim FH, Taylor WF, Pritchard DJ, et al: Lymphadenectomy inthe management of stage I malignant melanoma: a prospective randomizedstudy. Mayo Clin Proc 61: 697-705, 1986.
3. Morton DL, Wen DR, Wong JH, et al: Technical details of intraoperativelymphatic mapping for early stage melanoma. Arch Surg 127: 392-399,1992.
4. Glass EC, Essner R, Giuliano A, Morton DL: Comparative efficacyof three lymphoscintigraphic agents. Proc Soc Nucl Med 36: 199,1995.
5. Wang X, Heller R, Van Voorhis N, et al: Detection of submicroscopiclymph node metastases with polymerase chain reaction in patientswith malignant melanoma. Ann Surg 220: 768-774, 1994.
6. Morton DL, Wen DR, Essner R, et al: Intraoperative lymphaticmapping and selective cervical lymphadenectomy for early-stagemelanomas of the head and neck. J Clin Oncol 11: 1751-1756, 1993.
7. Reintgen D, Cruse CW, Wells K: The orderly progression of melanomanodal metastases. Ann Surg 220: 759-767, 1994.
8. Thompson J, McCarthy W, Robinson E, et al: Sentinel lymph nodebiopsy in 102 patients with clinical stage I melanoma undergoingelective node dissection. Presented at the Society of SurgicalOncology, Houston, Texas, March 1994.
9. Albertini JJ, Cruse CW, Rapaport D, et al: Intraoperative radiolymphoscintigraphyimproves sentinel lymph node identification for patients withmelanoma. Ann Surg 223: 217-224, 1996.
10. Essner R, Glass EC, Morton DL: Preoperative and intraoperativelymphatic mapping employing technetium-99m human serum albuminfor early stage melanoma. Proc Soc Nucl Med 5: 223, 1995.