Ironically, the patients who can benefit the most from CLND in terms of regional nodal basin disease control are the patients who are least likely to experience a survival benefit.
The practice of routine complete lymph node dissection (CLND) after removing a positive sentinel lymph node (SLN) in patients with melanoma, known as “selective lymphadenectomy,” has been embraced as the standard of care. However, many surgeons have questioned the strength of the evidence supporting this obligatory transition to CLND. Along with the important goal of reducing treatment-related morbidity without compromising oncologic outcomes, attention has been focused on determining whether the practice of CLND is worth the endured surgical morbidity.
The long-term results of the Multicenter Selective Lymphadenectomy Trial (MSLT)-1, combined with results from other studies, offer convincing evidence that SLN biopsy provides relevant staging/prognostic information. Furthermore, improved melanoma-specific survival, enhanced regional lymph node basin disease control, and reduced surgical morbidity have been achieved with SLN biopsy for the node-positive subset of patients (who represent approximately 20% of patients who undergo biopsy), compared with treating patients when they develop clinically palpable nodal disease. These data support the proof of concept that treating regional lymph node disease early is of clinical value. However, since all of the patients with a positive SLN in the MSLT-1 trial underwent CLND, it is difficult to determine the fractional benefit of CLND beyond what is achieved with removal of the SLNs. Theoretically, the only patients who can derive benefit from CLND are those who also harbor additional microscopic disease in non-SLNs, representing approximately 10% to 20% of SLN-positive patients. Several studies have demonstrated that patients with microscopic non–SLN involvement have a particularly unfavorable prognosis, suggesting that early treatment of nodal disease beyond SLNs has little impact on melanoma-specific survival. Many have come to the conclusion that most, if not all, of the survival benefits observed in the MSLT-1 trial are attributable to the removal of SLNs alone. This notion has a fair amount of support, as an analysis of a Surveillance, Epidemiology, and End Results database demonstrated that only 50% of patients with a positive SLN actually undergo CLND, reflecting the level of clinical equipoise that exists within the at-large surgical community. Publications of nonrandomized studies comparing the outcomes of SLN-positive patients who underwent CLND with those who had only an SLN biopsy suggest that CLND provides no significant survival advantage.[5,6] These studies are retrospective, have imbalances in prognostic factors and significant surgeon and patient bias, and include only short-term follow-up in the group without CLND. Therefore, these results should not support a change in practice to one that routinely omits CLND, but illustrates the current level of equipoise and the importance of the long-term outcomes of randomized trials.
Two prospective randomized trials in the SLN-positive patient population have been completed, comparing observation with nodal basin ultrasound vs CLND. The MSLT-2 trial completed accrual 2 years ago (ClinicalTrials.gov identifier: NCT00297895). Unfortunately, a prevailing feeling is that a significant amount of selection bias existed in the accrual process by both surgeons and patients in favor of patients who were predicted to have a low risk of harboring additional positive nodes. Combined with the routine use of ultrasound surveillance, many feel that this scenario likely reduces the potential impact of the CLND treatment arm. If the long-term survival difference is insignificant, proponents of CLND will argue that the trial was underpowered to answer the question. Opponents will argue that these results accurately reflect the most common patient population we treat and that a rational, safe, and less morbid alternative treatment strategy for this low-risk group is ultrasound surveillance.
The German Dermatologic Cooperative Oncology Group (DeCOG)-SLT trial studied a similar but smaller group of patients. The preliminary results of this trial were recently published in Lancet Oncology after a median follow-up of 3 years. The demographics of this trial clearly favored the group of patients with a low risk for non–SLN involvement, as almost 70% were in the < 1-mm SLN tumor burden subgroup. Unfortunately, the report does not provide complete data on the percentage of patients in the CLND treatment arm who had additional positive nodes, but an incomplete analysis detected a frequency rate of 18%. Regardless, no difference in melanoma-specific survival was observed in the overall cohort or in either ad hoc nodal burden subset. Therefore, at this point, there is no evidence that CLND will impact survival outcome.
This should not be a surprise to anyone, given that the only patients who can potentially derive benefit-the patients with non–SLN involvement-represent only 18% of the patients who were randomized. This is a similar scenario as in the MSLT-1 trial, where the only patients who could derive benefit were the 20% with a positive SLN. Unfortunately, the DeCOG trial is particularly underpowered because only 270 patients were randomized in each arm, and with an 18% non–SLN involvement rate, assessment of impact of CLND on natural history would be based on 40 patients in each arm at most-hardly a robust analysis even if it were to be performed. The MSLT-2 trial is larger, and therefore will have more non–SLN-positive patients for such a subset analysis, which will compare outcomes of patients randomized to CLND with those who were randomized to observation and underwent a lymphadenectomy to treat the development of clinically palpable nodal disease. This analysis would be of great interest.
Although it is logical to presume that early treatment of any microscopic nodal disease (sentinel or nonsentinel) should be equally effective in preventing nodal disease progression, and should result in the same favorable survival impact as observed in MSLT-1, the biology of patients with non–SLN metastases appears to be more unfavorable, and less likely to derive benefit with surgery. Hence, why bad biology trumps surgery.
Advocates of CLND argue that the potential for improved survival is not the only reason for recommending the procedure. Accurate staging (prognostic assessment), adjuvant trial eligibility, and regional nodal disease control are also important goals in the comprehensive management of melanoma patients with early nodal metastases.[8,9]
Certainly the number of positive nodes and the presence of non–SLN involvement affect the predicted risk for distant disease relapse. Such risk assessment is critical when determining the risk/benefit ratio of receiving systemic adjuvant therapy, particularly given the associated high toxicity profiles of the three currently approved therapies (high-dose interferon, pegylated interferon, and high-dose ipilimumab). Unfortunately, consistent overall survival benefits have not been demonstrated with either of the interferon regimens, and long-term overall survival outcomes for adjuvant ipilimumab are not yet available. Furthermore, the routine application of CLND will upstage less than 20% of the patients, and if the primary tumor is ulcerated or if three SLNs are involved, an even smaller percentage of patients will benefit from the additional information obtained. Overall, it seems like the routine practice of CLND for all SLN-positive patients is a high price to pay in terms of cost and surgical morbidity, considering the limited number of patients who can benefit from the additional staging information.
Another offered benefit of CLND is eligibility for the currently accruing high-priority adjuvant trial randomizing patients with nodal metastases to the current standard adjuvant therapy vs anti–programmed death 1 therapy. This trial mandates CLND for trial participation. While I feel this is an unfortunate mandate, as the critical endpoints should be distant relapse–free survival and overall survival, neither of which would likely be impacted by the small increase in regional nodal basin failures, it is a problem that will hopefully disappear once the trial has completed accrual.
Durable regional nodal basin disease control is more likely to be achieved when the regional lymphadenectomy is performed to treat microscopic nodal disease as opposed to clinically apparent nodal metastases. Furthermore, recent studies have shown that the short- and long-term surgical morbidity of the regional dissection is lower when treating SLN metastases vs treating patients who have palpable nodal disease. Long-term morbidity is particularly high if adjuvant nodal basin irradiation is applied. Although the evidence to support these assertions is fairly robust, the morbidity of CLND in the best of settings and hands is still significant. Again, why subject every patient to the cost and morbidity of an operation when only a select few can actually derive benefit? It would seem to make sense that a more selective approach, based on predictors for the presence of non–SLN involvement, would be a rational strategy. Furthermore, the clinical scenario of actually losing regional disease control and its associated significant morbidity is really a fear propagated by the CLND supporters. Such clinical scenarios are extremely rare occurrences and should not happen in patients who are followed closely with a careful physical examination selectively supplemented by ultrasound. Such an approach would likely identify clinical nodal involvement early and with less burden of disease, which in turn will reduce the need for adjuvant basin irradiation and avoid its associated treatment morbidities.
Little high-level evidence exists in support of the routine use of CLND in patients with positive SLNs, particularly if melanoma-specific survival is the only desired endpoint. This adopted “standard of care” is based in surgical dogma rather than surgical science. While we need to wait for the long-term results of the previously described randomized trials, it is reasonable to offer patients a selective approach based on their underlying comorbidities, predicted risk of non–SLN involvement and surgical morbidity, and the mutually agreed upon goals for treatment. Ironically, the patients who can benefit the most from CLND in terms of regional nodal basin disease control are the patients who are least likely to experience a survival benefit.
Financial Disclosure:The author has no significant financial interest in or other relationship with the manufacturer of any product or provider of any service mentioned in this article.
1. Morton DL, Thompson JF, Cochran AJ, et al. Final trial report of sentinel-node biopsy versus nodal observation in melanoma. N Engl J Med. 2014;370:599-609.
2. Balch CM, Gershenwald JE, Soong SJ, Thompson JF. Update on the melanoma staging system: the importance of sentinel node staging and primary tumor mitotic rate. J Surg Oncol. 2011;104:379-85.
3. Bamboat ZM, Konstantinidis IT, Kuk D, et al. Observation after a positive sentinel lymph node biopsy in patients with melanoma. Ann Surg Oncol. 2014;21:3117-23.
4. Bilimoria KY, Balch CM, Bentrem DJ, et al. Complete lymph node dissection for sentinel node-positive melanoma: assessment of practice patterns in the United States. Ann Surg Oncol. 2008;15:1566-76.
5. Wong SL, Morton DL, Thompson JF, et al. Melanoma patients with positive sentinel nodes who did not undergo completion lymphadenectomy: a multi-institutional study. Ann Surg Oncol. 2006;13:809-16.
6. Melstrom LG, Taylor E, Kuk D, et al. International multi-institutional management and outcome of melanoma patients with positive sentinel lymph nodes in more than one nodal basin. Ann Surg Oncol. 2014;21:4324-9.
7. Leiter U, Stadler R, Mauch C, et al. Complete lymph node dissection versus no dissection in patients with sentinel lymph node biopsy positive melanoma (DeCOG-SLT): a multicentre, randomised, phase 3 trial. Lancet Oncol. 2016;17:757-67.
8. Ariyan C, Brady MS, GÃ¶nen M, et al. Positive nonsentinel node status predicts mortality in patients with cutaneous melanoma. Ann Surg Oncol. 2009;16:186-90.
9. Brown RE, Ross MI, Edwards MJ, et al. The prognostic significance of nonsentinel lymph node metastasis in melanoma. Ann Surg Oncol. 2010;17:3330-5.
10. Kunte C, Geimer T, Baumert J, et al. Analysis of predictive factors for the outcome of complete lymph node dissection in melanoma patients with metastatic sentinel lymph nodes. J Am Acad Dermatol. 2011;64:655-62.
11. Guggenheim M, Dummer R, Jung FJ, et al. The influence of sentinel lymph node tumour burden on additional lymph node involvement and disease-free survival in cutaneous melanoma-a retrospective analysis of 392 cases. Br J Cancer. 2008;98:1922-8.
12. Gershenwald JE, Andtbacka RH, Prieto VG, et al. Microscopic tumor burden in sentinel lymph nodes predicts synchronous nonsentinel lymph node involvement in patients with melanoma. J Clin Oncol. 2008;26:4296-303.
13. Kretschmer L, Thoms KM, Peeters S, et al. Postoperative morbidity of lymph node excision for cutaneous melanoma-sentinel lymphonodectomy versus complete regional lymph node dissection. Melanoma Res. 2008;18:16-21.