This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Starz, H. Articles by Balda, B.-R. Search for Related Content PubMed PubMed Citation Articles by Starz, H. Articles by Balda, B.-R.

Sentinel Lymphonodectomy and S-Classification: A Successful Strategy for Better Prediction and Improvement of Outcome of Melanoma

From the Department of Dermatology and Allergology, Klinikum Augsburg, Augsburg, Germany.

Correspondence: Address correspondence and reprint requests to: Hans Starz, MD, Department of Dermatology and Allergology, Klinikum Augsburg, Stenglinstrasse 2, D-86156 Augsburg, Germany; Fax: 49-821-4003336; E-mail: hstarz{at}web.de

ABSTRACT

The most successful strategies in the management of melanoma have always been based on early diagnosis and timely surgical removal. Sentinel lymphonodectomy (SLNE) is the most reliable technique for the detection of melanoma micrometastases in regional lymph nodes. The micromorphometric S-classification, a routinely determinable surrogate of tumor burden in the sentinel lymph node (SLN), has high prognostic relevance. SIII metastases, defined by a depth of invasion (d) greater than 1 mm below the capsular level, imply a risk of more than 50% for the presence of nonsentinel lymph node metastases in the same basin and for the emergence of distant metastases within 5 years of follow-up. Corresponding risks with SI metastases (d 0.3 mm) and SII metastases (0.3 mm < d 1 mm) do not exceed 15%. The survival curve for patients with SIII metastases approaches that of patients in the pre-SLNE era who underwent delayed lymph node dissection for subsequently detected nodal macrometastases. The survival of patients with initially removed SI and SII metastases is much better, similar to that of patients with S0 metastases. This explains the significant survival benefit of SLN-guided surgery in the entire population of patients with melanomas thicker than 0.75 mm, although the outcome of the subgroup without nodal metastases is not influenced by SLNE.

Key Words: Benefit • Melanoma • Metastasis • Morphometry • Prognosis • Sentinel lymph node

In 1863, Virchow¹ reported that the first metastasis of a malignant solid neoplasm is frequently found in the lymph node that is the initial target of the lymphatics originating in the tumor area. This observation was a major argument against the predominant theory of cancer beginning as a systemic disease. Virchow taught that solid cancers start as local lesions, which tend to form regional lymph node metastases before becoming a systemic disease. The sentinel lymphonodectomy (SLNE) method reported by Morton et al.² in 1992 permits the early detection of regional lymph node metastases in patients with melanomas, with only minimal surgically induced morbidity. Further technical improvements have since increased the diagnostic reliability of this method and have led to the internationally accepted standards summarized as the Augsburg Consensus in 2000.³

In many centers around the world, SLNE has been incorporated into primary melanoma treatment. The diagnosis of SLN micrometastases is usually followed by a completion lymph node dissection (CLND) in the respective basin. Other hospitals hold to the traditional wait-and-see strategy, where lymph node dissections are performed only when a nodal macrometastasis becomes clinically evident. These conflicting alternatives are being prospectively compared in a large randomized multicenter trial (Multicenter Selective Lymphadenectomy Trial, NCI 29,605), with special attention on their impact on overall survival.⁴ Conclusive results from this trial, however, cannot be expected earlier than 2006. At our department, SLNE selectively supplemented by CLND was introduced in 1994 and has been used routinely since 1995.^5–7 A comparison of outcomes since 1995 with those of the wait-and-see era (until 1993) was undertaken to answer three questions. First, does SLN-guided surgery benefit melanoma patients with respect to metastases and overall survival? Second, which patients are most likely to benefit? Third, which new independent predictor for further nodal and distant metastases as well as for overall survival can be routinely extracted from the SLN evaluation?

PATIENTS AND METHODS

Since 1995, 522 patients with melanomas thicker than .75 mm and without preoperatively detected metastases have undergone gamma-probe-guided SLNE in addition to wide local excision of the primary tumor at our department. Of the 522 patients, 324 were treated between 1995 and 2000. This SLNE cohort of 324 patients was compared with a pre-SLNE cohort of 274 consecutive melanoma patients (same distribution of clinical stages) who underwent wide local excision alone between 1987 and 1993. If regional nodal metastases became evident during follow-up, then delayed lymph node dissection (DLND) was performed. Ulceration of the primary melanomas had not been systematically recorded in the past and therefore could not be taken into consideration.

Evaluation of SLNs and Non-SLNs
Immediately following SLNE, SLNs were fixed in 4.5% formalin, usually for 24 hours. They were then cut into 1-mm slices parallel to their longest axis. Each slice provided at least three paraffin sections, one for H&E staining and two for immunohistochemistry with anti-S100 and HMB45. If melanoma cells were microscopically identified in the SLN, their maximum distance (d) from the interior margin of the lymph node capsule was measured. Invasion deeper than 1 mm below the capsular level defines classification SIII, 0.31 to 1 mm defines classification SII, and subcapsular presence of tumor cells not deeper than .3 mm defines classification SI. S0 means absence of microscopically identifiable tumor cells in the SLN. The S-classification is separately determined for all lymph node basins containing SLNs. If metastases exist in more than one SLN of a single lymph node basin, the maximum d value of all these SLNs defines the S-classification. This simplified version of the original S-classification⁸ can be used for various methods of SLN sampling, including the one preferred by Cochran.⁹ Whereas the original S-classification uses Arabic numerals (S0, S1, S2, S3), the simplified version uses Roman numerals (S0, SI, SII, SIII).

When CLND was performed, the entire CLND specimen was fixed in formalin and meticulously cut into thin tissue slices to permit detection of even small lymph nodes (average number, 29). All nodes were evaluated in the same way as SLNs, except that immunolabeling was usually restricted to anti-S100.

Follow-Up Examinations and Further Treatment
CLND was strongly recommended in all 39 cases with SIII metastases. Only three of these patients refused CLND, in contrast to 13 of 43 patients with SII metastases and 20 of 37 patients with SI metastases. Because an SI or SII designation implies a relatively low likelihood of nonsentinel lymph node metastases, patients’ acceptance of CLND (informed consent) was less likely in SI and SII cases than in SIII cases. Radiation and adjuvant treatment such as chemotherapies or immunotherapies were not recommended for patients with melanoma-positive SLNs.

Over a follow-up period of at least 10 years, each melanoma patient was encouraged to visit during special consultation hours at our department, according to a fixed schedule. In addition to anamnesis and clinical investigation, radiologic and sonographic examinations were routinely performed. Furthermore, all patients were advised to inform us about any suspicious observations as quickly as possible. If a suspected metastasis could not be excluded by computed tomography, magnetic resonance imaging, and/or positron emission tomography, then a biopsy was performed for definite histological diagnosis. If the suspected focus was not accessible for biopsy, then noninvasive reinvestigations were performed within short time intervals. If metastases were diagnosed, their surgical removal was always considered, except when risks due to surgery prevailed. In these cases, chemotherapy with dacarbazine or vindesine was administered.

Statistical Methods
A binary logistic regression was applied to identify the relevant predictors for the presence versus absence of nonsentinel lymph node metastases in CLND specimens. Overall survival and survival without distant metastases were studied with use of Kaplan-Meier curves and univariate log rank tests. Multivariate analyses considering various potential predictors of distant metastasis and overall survival were performed with the Cox regression technique. The starting point for all time-dependent ana-lyses was the initial melanoma treatment.

RESULTS

There were no significant differences between the SLNE and the pre-SLNE cohorts with respect to gender, age, site of the primary melanoma, and Breslow tumor thickness (Table 1).

View larger version (50K): [in this window] [in a new window]   FIG. 1. Distribution of sentinel lymph node (SLN) S-classification according to Breslow thickness of the primary melanoma in 324 SLNE patients. When several SLNs were in different lymph node regions, the highest S-classification was used.

Distant metastases were diagnosed in 64 of the 274 pre-SLNE patients but in only 26 of the 324 SLNE patients. This discrepancy may be due partially to a longer median follow-up in the pre-SLNE cohort (95 vs. 45.5 months), which would also explain the longer median time before diagnosis of distant metastasis in that cohort (36 vs. 25 months). Nonetheless, the divergence of the respective Kaplan-Meier curves (Fig. 2A) was highly significant (P = .0057 by log rank test). The SLNE criterion remained significant (P = .026) when other potential predictors of distant metastases were considered in a multivariate Cox regression (Table 3).

View larger version (13K): [in this window] [in a new window]   FIG. 2. Survival without distant metastasis (A) and overall survival (B) in all SLNE patients (n = 324) and all pre-SLNE patients (n = 274).

Subgroup Analysis
Node-positive patients were those who had melanoma-positive SLNs or subsequently developed regional nodal metastases. These patients represented 20.8% of the pre-SLNE cohort and 21.6% of the SLNE cohort. Distant metastases developed in 37 of the 57 node-positive pre-SLNE patients vs. 15 of the 70 node-positive SLNE patients (SLNE ± CLND) (Table 1). Although median follow-up periods must be considered (59 months for pre-SLNE vs. 35 months for SLNE subgroups), the Kaplan-Meier curves for survival without distant metastases (Fig. 3A) were significantly divergent (P = .0048 by log rank test). Delayed divergence and thus a reduced level of significance (P = .0419 by log rank test) were observed for the respective overall survival curves (Fig. 3B).

View larger version (14K): [in this window] [in a new window]   FIG. 3. Survival without distant metastases (A) and overall survival (B) of the 70 SLNE patients and the 57 pre-SLNE patients who had histologically diagnosed metastases in regional lymph nodes. The 70 SLNE individuals comprise 65 patients with melanoma-positive SLNs and 5 patients with nodal metastases diagnosed during follow-up.

When primary melanomas were intermediate in thickness (0.76 to 4 mm), the rate of distant metastasis was 51 of 237 in the pre-SLNE cohort (median follow-up, 119 months) and 16 of 295 in the SLNE cohort (median follow-up, 45 months). Kaplan-Meier curves for survival without distant metastases (Fig 4A) and overall survival (Fig 4B) were significantly better in the SLNE cohort than in the pre-SLNE cohort (P = .0006 and .0076, respectively, by log rank test).

View larger version (13K): [in this window] [in a new window]   FIG. 4. Survival without distant metastases (A) and overall survival (B) of the 295 SLNE patients and the 237 pre-SLNE patients who had intermediate-thickness melanomas (0.76 to 4.00 mm, according to Breslow).

S-Classification as Predictor of Distant Metastasis and Survival
Table 1 (top) shows the distribution of S-classifications in the SLNE cohort. The S-classification was a highly significant predictor for distant metastasis (Fig. 5A) and overall survival (Fig. 5B). The related P values in the univariate log rank tests were each <.0001. The main reason for this high significance was the unfavorable outcome of SIII metastases in comparison with S0, SI, or SII cases. In Cox regressions with gender, age, tumor site, and tumor thickness as covariates, S-classification remained a significant independent predictor for distant metastasis (P = .014) and overall survival (P = .009) (Table 4).

View larger version (15K): [in this window] [in a new window]   FIG. 5. Survival without distant metastases (A) and overall survival (B) of all SLNE patients (n = 324), grouped by S-classification.

The average prognosis of melanoma is much better today than it was 30 years ago,¹⁰ mainly because of earlier diagnosis. Today, 95% of the cutaneous melanomas documented in the Tumor Center Munich are diagnosed in clinical stages T1–4 N0 M0. While the outcome for patients with melanomas no thicker than 0.75 mm is excellent following wide local excision only, about 20% of those with thicker melanomas develop regional lymph node metastases. Our results and other publications indicate that most of these metastases can be detected in microstages by SLNE and by thorough histological and immunohistochemical evaluation of the SLNs.^2,3,7–9 In the present study, SLN micrometastases were found in 65 of 324 melanoma patients (20%). Later, evidence of regional nodal metastasis—despite S0 classification of the SLNs—was observed in only five of 259 melanoma patients (2%).

Nodal macrometastasis is a strong predictor for distant metastases, and only about 30% of these patients are long-term survivors. The same applies for patients with SIII metastases in their SLNs. More than 50% of these individuals have additional metastases in nonsentinel lymph nodes, and more than 50% develop distant metastases within 5 years. If, however, SLNs with only SI or SII microinvolvement have been removed, the risk of distant dissemination or death from melanoma is not significantly higher than that for patients without SLN metastases (S0). In these cases, the early diagnosis of regional lymph node involvement and its surgical treatment by SLNE and selectively added CLND are obviously beneficial and explain the outcome advantage of SLNE versus the wait-and-see concept. In analogy, Cascinelli et al.¹¹ described a significant survival benefit for early removal of nodal melanoma metastases by elective lymph node dissection versus delayed removal after clinical manifestation of nodal involvement. As expected, SLNE had no significant impact on outcome when S0 patients were compared with pre-SLNE patients who never developed regional nodal metastases.

Despite the obvious link between lymph node metastases and subsequent distant dissemination,¹² the primary melanoma must not be neglected as a potential source of hematogenous metastases. For example, the 5-year distant disease-free survival and overall survival of our patients with T4 primaries and S0 SLNs were both less than 60%, not significantly better than survival rates for patients with T4 primaries and SIII SLNs (data not shown in detail). Systemic metastasis originating directly from the primary melanoma is inaccessible to SLNE and may explain the absence of a statistically relevant survival benefit for SLN-guided surgery in patients with melanomas thicker than 4 mm. Such a benefit, however, appears to be significant for patients with intermediate-thickness melanomas (.76 to 4 mm). It parallels similar experiences with elective lymph node dissection in the past.¹³

All these findings together strongly support the hypothesis that SLNE is more than a highly sensitive staging tool. It is key for the early elimination of a major source of systemic metastasis.¹² The most efficient principles of melanoma therapy—early diagnosis and timely surgical removal¹⁰—are thus successfully extended to the regional lymph nodes. Conversely, locoregional interventions will usually fail when systemic microdissemination is already underway. This complication, however, appears to be relatively rare if neither the primary melanoma nor the SLN metastases have reached advanced stages.

Although historical comparisons need cautious interpretations and verification by randomized prospective multicenter trials, the results of the present retrospective study are in perfect agreement with the above-mentioned hypotheses. A further retrospective study from five German centers has recently confirmed our findings.¹⁴ Definite clarification is expected from the Multicenter Selective Lymphadenectomy Trial (NCI 29,605) in 2006.⁴ In the meantime, SLNE selectively supplemented by CLND is probably superior to the wait-and-see strategy for primary management of melanoma patients.

FOOTNOTES

Sentinel lymphonodectomy (SLNE) is the most reliable technique for the detection of melanoma micrometastases in regional lymph nodes. The micromorphometric S-classification, a routinely determinable surrogate of tumor burden in the sentinel lymph node (SLN), has high prognostic relevance.

Received for publication December 1, 2003. Accepted for publication December 10, 2003.

REFERENCES

1. Virchow R. Dritte Vorlesung. Allgemeine Physiologie der Geschwülste. In: Virchow R. Die krankhaften Geschwülste. Berlin: Hirschwald-Verlag, 1863: 33–56.
2. Morton DL, Wen DR, Wong JH, et al. Technical details of intraoperative lymphatic mapping for early stage melanoma. Arch Surg 1992; 127: 392–9.[Abstract]
3. Cochran AJ, Balda BR, Starz H, et al. The Augsburg Consensus. Techniques of lymphatic mapping, sentinel lymphadenectomy, and completion lymphadenectomy in cutaneous malignancies. Cancer 2000; 89: 236–41.[CrossRef][Medline]
4. Morton DL, Thompson JF, Essner R, et al. Validation of the accuracy of intraoperative lymphatic mapping and sentinel lymphadenectomy for early-stage melanoma: a multicenter trial. Multicenter Selective Lymphadenectomy Trial Group. Ann Surg 1999; 230: 453–63.[CrossRef][Medline]
5. Bachter D, Balda BR, Vogt H, et al. Die "sentinel" Lymphonodektomie mittels Szintillationsdetektor. Eine neue Strategie in der Behandlung maligner Melanome. Hautarzt 1996; 47: 754–8.[Medline]
6. Bachter D, Balda BR, Vogt H, et al. Primary therapy of malignant melanomas: sentinel lymphadenectomy. Int J Dermatol 1998; 37: 278–82.[CrossRef][Medline]
7. Starz H, De Donno A, Balda BR. The Augsburg experience: histological aspects and patient outcomes. Ann Surg Oncol 2001; 8 (9 Suppl): 48S–51S.
8. Starz H, Balda BR, Krämer KU, et al. A micromorphometry-based concept for routine classification of sentinel lymph node metastases and its clinical relevance for patients with melanoma. Cancer 2001; 91: 2110–21.[CrossRef][Medline]
9. Cochran AJ, Starz H. Pathologic aspects. In: Cody HS III. Sentinel Lymph Node Biopsy. London: Martin Dunitz Ltd, 2002: 79–90.
10. Volkenandt M, Schmidt M, Konz B, et al. Klinisch-epidemiologische Daten von Patienten mit malignen Melanomen aus dem Bereich des Tumorzentrums München von 1977 bis 1997. Hautarzt 1999; 50: 470–8.[Medline]
11. Cascinelli N, Morabito A, Santinami M, et al. Immediate or delayed dissection of regional nodes in patients with melanoma of the trunk: a randomised trial. WHO Melanoma Programme. Lancet 1998; 351: 793–6.[CrossRef][Medline]
12. Sleeman JP. The lymph node as a bridgehead in the metastatic dissemination of tumors. Rec Res Cancer Res 2000; 157: 55–81.[Medline]
13. Balch CM, Soong S, Ross MI, et al. Long-term results of a multi-institutional randomized trial comparing prognostic factors and surgical results for intermediate thickness melanomas (1.0 to 4.0 mm). Intergroup Melanoma Surgical Trial. Ann Surg Oncol 2000; 7: 87–97.[Abstract]
14. Kretschmer L, Hilgers R, Möhrle M, et al. Patients with lymphatic metastasis from cutaneous malignant melanoma benefit from sentinel lymphonodectomy and early excision of their nodal disease. Eur J Cancer 2004; 40: 212–8.

This article has been cited by other articles:

				K. Y. Bilimoria, C. M. Balch, D. J. Bentrem, M. S. Talamonti, C. Y. Ko, J. R. Lange, D. P. Winchester, and J. D. Wayne Complete Lymph Node Dissection for Sentinel Node-Positive Melanoma: Assessment of Practice Patterns in the United States Ann. Surg. Oncol., June 1, 2008; 15(6): 1566 - 1576. [Abstract] [Full Text] [PDF]

				I. Satzger, B. Volker, A. Meier, A. Kapp, and R. Gutzmer Criteria in Sentinel Lymph Nodes of Melanoma Patients that Predict Involvement of Nonsentinel Lymph Nodes Ann. Surg. Oncol., June 1, 2008; 15(6): 1723 - 1732. [Abstract] [Full Text] [PDF]

				C. R. Rossi, G. L. De Salvo, E. Bonandini, S. Mocellin, M. Foletto, S. Pasquali, P. Pilati, M. Lise, D. Nitti, E. Rizzo, et al. Factors Predictive of Nonsentinel Lymph Node Involvement and Clinical Outcome in Melanoma Patients With Metastatic Sentinel Lymph Node Ann. Surg. Oncol., April 1, 2008; 15(4): 1202 - 1210. [Abstract] [Full Text] [PDF]

				A. M. Fink, H. Lass, H. Hartleb, W. Jurecka, H. Salzer, and A. Steiner S-Classification of Sentinel Lymph Node Predicts Axillary Nonsentinel Lymph Node Status in Patients with Breast Cancer Ann. Surg. Oncol., March 1, 2008; 15(3): 848 - 853. [Abstract] [Full Text] [PDF]

				I. Koskivuo, L. Talve, P. Vihinen, M. Maki, T. Vahlberg, and E. Suominen Sentinel Lymph Node Biopsy in Cutaneous Melanoma: A Case-Control Study Ann. Surg. Oncol., December 1, 2007; 14(12): 3566 - 3574. [Abstract] [Full Text] [PDF]

				R. Scolyer, R Murali, J. Gershenwald, A. Cochran, and J. Thompson Clinical relevance of melanoma micrometastases in sentinel nodes: too early to tell Ann. Onc., April 1, 2007; 18(4): 806 - 808. [Full Text] [PDF]

				A. Govindarajan, D. M. Ghazarian, D. R. McCready, and W. L. Leong Histological Features of Melanoma Sentinel Lymph Node Metastases Associated with Status of the Completion Lymphadenectomy and Rate of Subsequent Relapse Ann. Surg. Oncol., February 1, 2007; 14(2): 906 - 912. [Abstract] [Full Text] [PDF]

				A. van Akkooi, J. de Wilt, C Verhoef, P. Schmitz, A. van Geel, A. Eggermont, and M Kliffen Clinical relevance of melanoma micrometastases (<0.1 mm) in sentinel nodes: are these nodes to be considered negative? Ann. Onc., October 1, 2006; 17(10): 1578 - 1585. [Abstract] [Full Text] [PDF]

				N. Cascinelli, E. Bombardieri, R. Bufalino, T. Camerini, A. Carbone, C. Clemente, L. Lenisa, L. Mascheroni, A. Maurichi, E. Pennacchioli, et al. Sentinel and Nonsentinel Node Status in Stage IB and II Melanoma Patients: Two-Step Prognostic Indicators of Survival J. Clin. Oncol., September 20, 2006; 24(27): 4464 - 4471. [Abstract] [Full Text] [PDF]

				L. Kretschmer, I. Beckmann, K.-M. Thoms, C. Mitteldorf, H. P. Bertsch, and C. Neumann Factors Predicting the Risk of In-Transit Recurrence After Sentinel Lymphonodectomy in Patients With Cutaneous Malignant Melanoma Ann. Surg. Oncol., August 1, 2006; 13(8): 1105 - 1112. [Abstract] [Full Text] [PDF]

				R. P. Abratt, W. H.H. Reece, N. H. Enas, and R. Stephens Randomized phase II studies and their ethics. J. Clin. Oncol., December 20, 2005; 23(36): 9443 - 9443. [Full Text] [PDF]

				J. M. Thomas Time to Re-Evaluate Sentinel Node Biopsy in Melanoma Post-Multicenter Selective Lymphadenectomy Trial J. Clin. Oncol., December 20, 2005; 23(36): 9443 - 9444. [Full Text] [PDF]

				K. M. McMasters Editorial: What Good is Sentinel Lymph Node Biopsy for Melanoma if it does not Improve Survival? Ann. Surg. Oncol., September 1, 2004; 11(9): 810 - 812. [Full Text] [PDF]

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Starz, H. Articles by Balda, B.-R. Search for Related Content PubMed PubMed Citation Articles by Starz, H. Articles by Balda, B.-R.