Correspondence
Sentinel-Node Biopsy in Melanoma
N Engl J Med 2007; 356:418-421January 25, 2007
- Article
To the Editor:
The results of the trial by Morton et al. (Sept. 28 issue)1 involving sentinel-node biopsy in patients with melanoma do not show a survival benefit with immediate lymphadenectomy for three reasons. First, there were false positive results in the sentinel node (tiny deposits of melanoma not destined to progress to a palpable nodal recurrence, which is a surrogate measure for survival). Since there was no significant difference in survival between the study groups from the time of randomization, only 119 patients in the biopsy group would be expected to have a palpable nodal recurrence; of these patients, 26 have already been identified. Therefore, 29 patients with positive results on sentinel-node biopsy would not be expected to have a recurrence, which means that 24% of patients who had positive biopsy results underwent unnecessary lymphadenectomy.
Second, the 78 patients in the observation group who underwent delayed lymphadenectomy and the 122 patients in the biopsy group who underwent immediate lymphadenectomy were selected after randomization. Therefore, any comparison of their survival within a controlled trial is statistically invalid. Third, the authors do not include the 26 patients with false negative results — those who had a recurrence that was not detected by sentinel-node biopsy.2,3
3 ReferencesJ. Meirion Thomas, F.R.C.S.
Royal Marsden Hospital, London SW3 6JJ, United Kingdom
meirion@roseway.demon. co. uk 1
Morton DL ,Thompson JF ,Cochran AJ , et al. Sentinel-node biopsy or nodal observation in melanoma. N Engl J Med 2006;355:1307-1317[Erratum, N Engl J Med 2006;355:1944.]
Full Text | Web of Science | Medline2
Thomas JM . Caution with sentinel node biopsy in cutaneous melanoma. Br J Surg 2006;93:129-130
CrossRef | Web of Science | Medline3
Thomas JM . Time for comprehensive reporting of MSLT-1. Lancet Oncol 2006;7:9-11
CrossRef | Web of Science | Medline
To the Editor:
Morton et al. should have defined subgroups of patients in the same way in the two study groups, rather than on the basis of events, such as nodal metastasis, that were observed after randomization, since subgroup membership may be influenced by treatment. In the observation group, the investigators did not assess sentinel lymph nodes, and patients who had metastasis after randomization were defined as having “nodal metastasis,” unlike 26 identical patients in the biopsy group who had false negative results. Consequently, the analysis of nodal metastasis is invalid and results in anomalous conclusions. For example, if patients with nodal metastasis who were treated with immediate lymphadenectomy had an improvement in melanoma-specific survival and there was no difference in overall melanoma-specific survival between the two study groups, a corollary is that patients in the biopsy group who did not have nodal metastasis must have had poorer survival than those in the observation group. This implies a 50% increase in the rate of death owing to sentinel-node biopsy in this group, from approximately 6.9% to 10.4% (P=0.06 by Fisher's exact test).
Roger P. A'Hern, M.Sc.
Institute of Cancer Research, Sutton SM2 5NG, United Kingdom
roger.ahern@icr. ac. uk To the Editor:
If one assumes that the groups studied by Morton et al. are equivalent, then one would expect that the incidence of nodal disease in the two groups would be the same. However, when the 26 patients with nodal recurrence despite negative results on sentinel-node biopsy are added, the data suggest more nodal disease in the biopsy group (19.4% vs. 15.8%). This finding suggests that not all sentinel nodes are destined to become clinically relevant, which is no surprise, given the heterogeneity of the disease. If the two study groups had been equal, then clinically palpable disease would not have developed in approximately one quarter of the patients with positive sentinel nodes. Thus, comparing only the patients with positive sentinel nodes with those with clinically palpable nodes is not appropriate because the latter group included more patients with aggressive disease. Unfortunately, the subgroup analysis should not be interpreted as showing a survival benefit for patients who underwent sentinel-node biopsy followed by lymph-node dissection.
James M. Grichnik, M.D., Ph.D.
Duke University Medical Center, Durham, NC 27710
grich001@mc.duke. edu Dr. Grichnik reports receiving consulting fees and grant funding from Electro-Optical Systems and honoraria from PharmAdura. He is also a founder of and major shareholder in DigitalDerm.
To the Editor:
The decision by Morton et al. to select patients with melanomas of intermediate thickness (1.2 to 3.5 mm) as the primary study group is arbitrary. There are no naturally occurring cutoff points for the thickness of primary melanoma that delineate different biologic risks of death.1 The elimination from the study of 654 patients who had tumors with thickness measurements outside these limits seems highly irregular.
The unambiguous message of this trial is found in Figure 2B of the report. Analysis of overall advantage from the time of randomization showed no advantage of sentinel-node biopsy over observation (P=0.58). Priority can now be safely relegated to the optimal use of prognostic factors associated with the primary lesion of melanoma. These factors may yield at least as much predictive information as sentinel-node biopsy.2,3
3 ReferencesSpyros Retsas, M.D., F.R.C.P.
Cromwell Hospital, London SW5 0TU, United Kingdom
sretsas@msn.com 1
Balch CM ,Soong S-J ,Gershenwald JE , et al. Prognostic factors analysis of 17,600 melanoma patients: validation of the American Joint Committee on Cancer melanoma staging system. J Clin Oncol 2001;19:3622-3634
Web of Science | Medline2
Retsas S . Cutaneous melanoma. Lancet 2005;365:2003-2004
CrossRef | Web of Science | Medline3
Retsas S ,Henry K ,Mohammed MQ ,MacRae K . Prognostic factors of cutaneous melanoma and a new staging system proposed by the American Joint Committee on Cancer (AJCC): validation in a cohort of 1284 patients. Eur J Cancer 2002;38:511-516
CrossRef | Web of Science | Medline
To the Editor:
In the study by Morton et al., how many patients had nodal involvement other than involvement of the sentinel node when they underwent immediate complete lymphadenectomy? The authors do not mention complications associated with complete lymphadenectomy. Such data are essential, since sentinel-node biopsy has no benefit in terms of survival but only in staging of the disease, as discussed by Balch and Cascinelli in their accompanying editorial.1
1 ReferencesDan Lipsker, M.D.
Université Louis Pasteur Strasbourg, 67000 Strasbourg, France
dan.lipsker@chru-strasbourg. fr 1
Balch CM ,Cascinelli N . Sentinel-node biopsy in melanoma. N Engl J Med 2006;355:1370-1371
Full Text | Web of Science | Medline
To the Editor:
Morton et al. are to be congratulated for their landmark work on melanoma. Of concern is the accompanying editorial by Balch and Cascinelli, which states that the data “convincingly” show that sentinel-node biopsy is a standard-of-care procedure, despite the fact that Morton et al. show that sentinel-node biopsy and subsequent immediate lymphadenectomy do not confer an overall survival benefit for patients with melanoma. Although this procedure has a proven prognostic value, the long-term benefits of adjuvant therapy for patients with melanoma who are identified by means of this procedure are uncertain.1 In addition, the cost-effectiveness of the procedure2 calls into question the proposed “standard-of-care” designation.
Both Morton et al. and Balch and Cascinelli state that sentinel-node biopsy identifies patients with nodal metastases whose survival can be prolonged by lymphadenectomy. However, the authors confuse the prognostic value with the predictive value of the procedure. In fact, the study was not designed to address this specific issue, and the data do not support this contention.
2 ReferencesMatthew H. Kanzler, M.D.
Lee Levitt, M.D.
Albert Lin, M.D., M.P.H.
Santa Clara Valley Medical Center, San Jose, CA 95128
kanzlerm@yahoo.com 1
Verma S ,Quirt I ,McCready D ,Bak K ,Charette M ,Iscoe N . Systematic review of systemic adjuvant therapy for patients at high risk for recurrent melanoma. Cancer 2006;106:1431-1442
CrossRef | Web of Science | Medline2
Agnese DM ,Abdessalam SF ,Burak WE Jr ,Magro CM ,Pozderac RV ,Walker MJ . Cost-effectiveness of sentinel lymph node biopsy in thin melanomas. Surgery 2003;134:542-547
CrossRef | Web of Science | Medline
Author/Editor Response
Thomas claims that some sentinel-node micrometastases will never progress because there were fewer nodal recurrences in our observation group than in the biopsy group at the time of the interim analysis. However, sentinel-node metastases were detected immediately, whereas nodal metastases in the observation group could not be identified until they grew to a detectable size. Our recent data indicate that at 10 years, the incidence of nodal metastases in the observation group will eventually equal the total incidence of nodal metastases detected by biopsy or by biopsy with false negative results: a mean (±SE) of 20.5±2.6% in the observation group and 20.8±1.7% in the biopsy group (Figure 1Figure 1
Cumulative Incidence of Regional Node Metastasis in the Observation Group and the Biopsy Group.).Thomas states incorrectly that 26 patients in the biopsy who had false negative results were not included the analysis. As stated in our article (on page 1312 and in Figure 3B), a comparison of survival of patients in the biopsy group (including these 26 patients and 122 patients with positive biopsy results) and patients in the observation group remains significant (hazard ratio for subgroup 2 vs. subgroup 3, 0.62; P=0.02).
With regard to the comments of Lipsker, the rate of complications associated with complete lymphadenectomy in our trial has been reported.1 Additional metastases in nonsentinel nodes were found in 10 to 20% of patients, a proportion that varied according to the thickness of the primary tumor and the tumor burden.2
We direct Retsas to page 1308 of our article, which provides the basis for the stratification factors used for Breslow thickness.
Thomas, A'Hern, and Grichnik criticize our analysis of predefined subgroups of patients who had occult nodal metastases. As stated on page 1314 of our article, randomization was used to ensure balance between subgroups that could not be identified before surgery. As is shown in Table 1 of our article and Figure 1 of this letter, balance was achieved. A'Hern's calculations regarding the survival of patients without nodal metastases are incorrect; as we stated on page 1312 and as shown in Figure 3C, 5-year rates of survival were 92.9±1.3% in the biopsy group and 92.4±1.2% in the observation group (P=0.98). Our data show that as compared with observation alone, sentinel-node biopsy conferred significantly better disease-free survival (P=0.009), an end point that has qualified for regulatory approval of new cancer drugs.3
3 ReferencesDonald L. Morton, M.D.
John Wayne Cancer Institute at Saint John's Health Center, Santa Monica, CA 90404Alistair J. Cochran, M.D.
University of California at Los Angeles, Los Angeles, CA 90095John F. Thompson, M.D.
Royal Prince Alfred Hospital, Camperdown, NSW 2050, Australia1
Morton DL ,Cochran AJ ,Thompson JF , et al. Sentinel node biopsy for early-stage melanoma: accuracy and morbidity in MSLT-I, an international multicenter trial. Ann Surg 2005;242:302-311
Web of Science | Medline2
Cochran AJ ,Wen DR ,Huang RR ,Wang HJ ,Elashoff R ,Morton DL . Prediction of metastatic melanoma in nonsentinel nodes and clinical outcome based on the primary melanoma and the sentinel node. Mod Pathol 2004;17:747-755
CrossRef | Web of Science | Medline3
Johnson JR ,Williams G ,Pazdur R . End points and United States Food and Drug Administration approval of oncology drugs. J Clin Oncol 2003;21:1404-1411
CrossRef | Web of Science | Medline
Author/Editor Response
We hold to the position that sentinel-node biopsy is a “standard of care” for the staging of melanoma and for treatment planning. The standard-of-care designation applies to many approaches that assist only with diagnosis and staging, even when they do not directly lead to a survival benefit. These procedures include sentinel-node biopsy for stage I and stage II breast cancer. Certainly, the value of the information gained from sentinel-node biopsy in melanoma, which is clearly demonstrated in this randomized study, has a vital role in treatment planning and provides much more than prognostic value. Early surgical intervention of the regional lymph nodes in patients with positive biopsy results improved overall survival and provides staging information for the consideration of adjuvant biologic therapy in a setting approved by the Food and Drug Administration (FDA). Patients in the biopsy group also had prolonged disease-free survival, an end point that would be an acceptable measure for FDA approval if this study were testing a new drug. Finally, the reference cited by the correspondents regarding cost-effectiveness focuses only on sentinel-node biopsy in thin melanomas and is not germane to this trial, which involved tumors measuring 1.2 to 3.5 mm.
Charles M. Balch, M.D.
Johns Hopkins Medical Institutions, Baltimore, MD 21287Natale Cascinelli, M.D.
National Tumor Institute, 20133 Milan, Italy- Citing Articles (13)
Citing Articles
1
R. Russell-Jones. (2012) When will selective lymphadenectomy become standard of care in melanoma?. International Journal of Clinical Practiceno-no
CrossRef2
R. Botella-Estrada, E. Nagore. (2011) Estado actual del ganglio centinela en el melanoma. Actas Dermo-Sifiliográficas 102:10, 749-753
CrossRef3
F. Wright, K. Spithoff, A. Easson, C. Murray, J. Toye, D. McCready, T. Petrella. (2011) Primary Excision Margins and Sentinel Lymph Node Biopsy in Clinically Node-negative Melanoma of the Trunk or Extremities. Clinical Oncology 23:9, 572-578
CrossRef4
Gershenwald, Jeffrey E., Ross, Merrick I., . (2011) Sentinel-Lymph-Node Biopsy for Cutaneous Melanoma. New England Journal of Medicine 364:18, 1738-1745
Full Text5
I. Satzger, A. Meier, L. Hoy, B. Völker, A. Kapp, A. Hauschild, R. Gutzmer. (2011) Sentinel Node Dissection Delays Recurrence and Prolongs Melanoma-Related Survival: An Analysis of 673 Patients from a Single Center with Long-Term Follow-Up. Annals of Surgical Oncology 18:2, 514-520
CrossRef6
FLORENT GRANGE, EVE MAUBEC, CORALIE BARBE, JAMAL KASSOUMA, FABIEN VITRY, HUBERT JOHANET, FLORENCE GRANEL-BROCARD, FRANCOISE BOITIER, ANGELIQUE GIROD, BENOIT COUTURAUD, PIERRE SAEZ, SEBASTIEN ALBERT, ANNICK LE CLAINCHE, VINCENT DESCAMPS, MARIE FRANCOISE AVRIL. (2011) Sentinel Lymph Node Biopsy or Nodal Observation in Melanoma: A Prospective Study of Patient Choices. Dermatologic Surgery 37:2, 199-206
CrossRef7
Di Bei, Jianing Meng, Bi-Botti C Youan. (2010) Engineering nanomedicines for improved melanoma therapy: progress and promises. Nanomedicine 5:9, 1385-1399
CrossRef8
William G. Stebbins, Lilit Garibyan, Arthur J. Sober. (2010) Sentinel lymph node biopsy and melanoma: 2010 update. Journal of the American Academy of Dermatology 62:5, 737-748
CrossRef9
Alexander C.J. van Akkooi, Alain Spatz, Alexander M.M. Eggermont, Martin Mihm, Martin G. Cook. (2009) Expert opinion in melanoma: The sentinel node; EORTC Melanoma Group recommendations on practical methodology of the measurement of the microanatomic location of metastases and metastatic tumour burden. European Journal of Cancer 45:16, 2736-2742
CrossRef10
Joseph Meirion Thomas. (2009) Concerns relating to the conduct and statistical analysis of the Multicenter Selective Lymphadenectomy Trial (MSLT-1) in patients with melanoma. Journal of Plastic, Reconstructive & Aesthetic Surgery 62:4, 442-446
CrossRef11
James M Grichnik. (2009) Hypothesis Letter: The Reason Sentinel and Lymph Node Dissections Do Not Improve Melanoma Mortality. Journal of Investigative Dermatology 129:3, 779-781
CrossRef12
Á. Pizarro. (2008) ¿Por qué la biopsia del ganglio centinela no aumenta la supervivencia en pacientes con melanoma?. Actas Dermo-Sifiliográficas 99:5, 323-330
CrossRef13
F. Grange. (2007) Quoi de neuf en cancérologie cutanée ?. Annales de Dermatologie et de Vénéréologie 134, 8S53-8S63
CrossRef
