| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Article |
1 Department of Surgery/Plastic and Reconstructive Surgery, Indiana University School of Medicine, Indiana University–Purdue University, Indianapolis, Indiana 46202
2 Department of Medicine/Biostatistics, Indiana University School of Medicine, Indiana University–Purdue University, Indianapolis, Indiana 46202
Correspondence: Address correspondence and reprint requests to: Jeffrey D. Wagner, MD, Suite 570, 8040 Clearvista Parkway, Indianapolis, IN 46256, USA; E-mail: jdwagner{at}insightbb.com.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1 mm). Methods: This was a retrospective study of patients who underwent SLNB for clinically localized melanoma at Indiana University Medical Center between 1994 and 2003. SLNB results and traditional melanoma prognostic indicators were studied in univariate log-rank tests.
Results: One hundred eighty-four patients with melanomas 1 mm thick underwent SLNB. SLNB was tumor positive in 12 patients (6.5%). Univariate analysis of SLNB results revealed that Breslow thickness, Clark level of invasion, and mitotic index were associated with SLNB status. Tumor positivity was observed at different rates in tumor thickness subsets: <.75 mm, 2.3%; and .75 to 1.0 mm, 10.2% (P = .0372). Disease-free survival and overall survival were significantly associated with SLNB results in melanomas 1 mm (log-rank test: P < .0001 and P = .0125, respectively) at a median follow-up of 26.3 months.
Conclusions: SLNB histology in melanomas 1.0 mm deep is a significant predictor of outcome. SLNB should be considered for selected patients with melanomas .75 to 1.0 mm.
Key Words: Melanoma • Metastases • Sentinel lymph node • Prognosis • Tumor thickness • Breslow level
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1 mm).1 Although generally considered prognostically favorable, with 5-year survival rates reported as high as 95%,2 some thin melanomas do retain the ability to metastasize, with an estimated risk of 3% to 5%, depending on the length of follow-up.3–6 Recurrence rates for subsets of thin melanomas have been reported as high as 18% in patients treated by wide local excision alone.7 Breslow tumor thickness is generally considered the most useful prognostic factor in patients with early-stage melanoma. In the most recent version of the American Joint Committee on Cancer staging system, the threshold for T1/T2 melanoma was changed from .75 to 1 mm.2 Ulceration and Clark level of invasion were also incorporated into the staging system for thin melanomas.2 Additional prognostic factors, including age >50 years,5 histological tumor regression,4,8 vertical growth phase,9,10 and the anatomical site of the primary tumor,7 have been reported to correlate with outcome in this disease subset.
Sentinel lymph node biopsy (SLNB) is currently routinely recommended for melanomas >1 mm thick and in selected patients with thin melanomas (1 mm).11 Gershenwald et al.12 confirmed that tumor thickness and ulceration were important prognostic discriminators in patients with a negative SLNB. However, a positive SLNB has been shown to be the best predictor of recurrence and survival in patients with clinically node-negative cutaneous melanoma.12,13
Selection of patients for SLNB is an area of debate. The incidence of a positive SLNB in melanomas <1 mm thick has been reported to be 0% to nearly 8%.12–15 SLNB has generally been considered to have a low yield in thin melanomas. However, with the minimal morbidity of SLNB and the availability of adjuvant therapy for stage III melanoma patients, others have recommended more liberal indications for SLNB in thin melanoma patients believed to be at an increased risk for metastases.7
The indications for SLNB have evolved since they were first described by Morton et al.16 in 1992. At Indiana University, SLNB was originally offered to patients with clinically localized, intermediate-thickness melanoma (1–4 mm). As experience with the procedure grew, indications were liberalized to include selected patients with deep melanomas and those with high-risk thin melanomas (1 mm). Specifically, selection criteria were broadened to include unknown tumor thickness (related to biopsy technique), ulceration, a high tumor mitotic rate, and sometimes truncal location, regression, and Clark level IV or V.11 The objective of this study was to determine the incidence of sentinel lymph node (SLN) positivity, the predictors of SLNB positivity, and the prognostic importance of SLNB in patients with melanomas 1 mm thick.
|
PATIENTS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1 mm) at Indiana University Medical Center and affiliated hospitals between 1994 and 2003. Patient demographics, primary tumor thickness, Clark level of invasion, location of primary tumor, presence of tumor ulceration, presence of regression, mitotic index, growth phase, and the pathology of the examined nodes were recorded. Disease status and the dates of last clinical contact were determined.
Methods
SLNB was performed by using a combined technique of vital blue dye and intraoperative radiolocalization we have previously described.11 Preoperative lymphoscintigraphy was used to identify the basin(s) at risk for lymph node disease. After the induction of anesthesia, .5 to 2.0 mL of isosulfan blue dye (Lymphazurin; Zenith Parentherals, Rosemont, IL) was injected intradermally around the site of the primary melanoma. Nodal basins identified by lymphoscintigraphy were explored through limited incisions guided by the use of a handheld gamma probe (C-Track; Care Wise Medical Products, Morgan Hill, CA). All blue nodes were identified and removed as SLNs. The most radioactive SLN was then compared with the residual lymph node basin radioactivity, and, if necessary, additional radioactive nodes were removed until the ratio of the most radioactive ex vivo SLN to that of the residual basin was at least 10:1. If histological analysis revealed metastatic melanoma in the removed SLN(s), then a complete regional lymph node dissection was performed.
At the discretion of the surgeon, grossly suspicious SLNs were sometimes submitted for frozen-section analysis. Nonsuspicious and frozen section–negative SLN(s) were fixed in formalin, submitted for 1-mm step sections, and analyzed by hematoxylin and eosin stains. SLNs negative for metastases were then subjected to further analysis with S-100 and/or HMB-45 immunostains. Non-SLNs were analyzed in routine fashion after formalin fixation, and one to three sections from each node were reserved for hematoxylin and eosin staining.
Statistical Analysis
Two-sample t-tests were performed to investigate the possible differences in age, Clark level, and Breslow thickness between SLNB-positive and SLNB-negative patients. Fisher’s exact tests were used to test for an association between SLNB-positive and SLNB-negative patients with respect to ulceration, regression, and growth phase.
A univariate log-rank test (for categorical variables) or Cox regression (for continuous variables) was used to investigate the effects of age, sex, Breslow thickness, Clark level, location, mitotic activity, presence of ulceration, and SLN status on disease-free survival and overall survival for all patients and for SLNB-positive and -negative subsets (age was examined as a categorical variable; 50 vs. >50 years). Multivariate analysis was not performed because the model would not converge as a result of the small number of events for disease-free survival and overall survival.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1 mm (range, .19–1.0 mm) underwent SLNB between 1994 and 2003. SLNB was tumor positive in 12 patients (6.5%). Tumor positivity was observed at different rates in tumor thickness subsets: <.75 mm, 2.3%; and .75 to 1.0 mm, 10.2% (Table 1
). In melanomas thinner than .5 mm, no positive SLN was found (n = 24).
|
1 mm deep are listed in Table 2. Univariate analysis of SLNB results by clinical variables with melanoma 1 mm is shown in Table 3. Breslow thickness, Clark level of invasion, and mitotic index were statistically significant in disease subsets 1 mm in predicting the SLNB result. Age, histological ulceration, regression, and growth phase were not associated with SLNB status in this disease subset. In the disease subset .75 to 1.0 mm thick, only mitotic index was predictive of the SLNB result (P = .0007; Mann-Whitney test; Table 4).
|
|
|
1 mm with a positive SLNB were alive with no evidence of disease, and 1patient (8%) had died of disease. Of those SLN-negative patients with melanomas 1 mm, 166 (98%) were alive with no evidence of disease, and 1 (.6%) had died of disease. Five patients were lost to follow-up. Subset analysis showed that disease-free survival and overall survival were both significantly associated with SLNB results in patients with melanomas 1 mm (log-rank test; P < .0001 and P = .0125, respectively; Fig. 1).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1 mm—i.e., which lesions may constitute ‘‘significant’’-risk lesions. Corsetti et al.7 and Lowe et al.14 recommend SLNB in all thin melanomas with Clark level III and IV. Bedrosian et al.9 suggested that patients with thin melanoma lesions with vertical growth phase should be evaluated for SLNB. There are few data on the utility of SLNB in this patient population, secondary to the small number of patients who have undergone SLNB for lesions 1 mm.
In our series of 184 patients with melanoma lesions 1 mm, we found 12 (6.5%) with metastatic disease within the SLN. Prior studies have ranged from 0% to 7.8% but have differed in their definition of ‘‘thin’’ melanomas and inclusion criteria (Table 5). These figures may overestimate the true node-positivity rates in all patients with thin melanomas. It is difficult to determine the true rate of SLNB positivity in thin melanomas because of the variable selection bias for performance of the procedure inherent in all retrospective studies.
|
1 mm. In the smaller disease subset, .75 to 1.0 mm, only mitotic index was significant. A tumor with a Breslow tumor thickness of .6 mm was the thinnest lesion found to have metastatic disease within the SLN in this series. All metastatic disease within the SLN involved lesions with a Clark III level of invasion or greater. This observation suggests that the dense dermal lymphovascular plexus at the papillary and reticular dermis skin junction (Clark level III) may be a biologically important boundary before lymphangioinvasion occurs.7 An increased rate of recurrence in melanoma lesions with high mitotic activity has been previously reported.4,11,19,20 Others have suggested that mitotic index may be a useful criterion for selecting thin-melanoma patients for SLNB staging.11,20,21
Disease-free and overall survival were significantly associated with SLNB results in patients with melanomas 1 mm. This must be interpreted carefully in light of our small sample size. However, this observation is consistent with prior reports of melanoma patients with SLNB-staged T2 to T4 lesions2 and in larger series.
Our findings point to a role for SLN staging in selected patients with thin melanoma—particularly patients with a Breslow thickness of .75 to 1.0 mm, especially if other risk factors are present, such as an increased mitotic index (more than two mitoses per square millimeter) and Clark level IV or V. Although the data do not show ulceration as a predictor of the SLNB result, we would also recommend SLNB for these patients on the basis of our prior work and that of others.22,23
Others have questioned the application15,17 and cost-effectiveness24 of SLNB in patients with thin melanomas. Muller et al.15 reported no positive SLNs in 75 patients with melanomas thinner than .90 mm and suggested that SLNB in patients with a Breslow thickness <.90 mm was ‘‘superfluous’’ unless certain other risk factors were present. Jacobs et al.17 found that only 2 (3%) of 65 patients with thin melanomas had positive SLNB results and that no patient with a lesion <.75 mm had metastatic disease within the SLN. This suggests that only a small percentage of patients may benefit from tumor upstaging by SLNB. Small sample sizes and differences in referral patterns and selection criteria for performing SLNB may account for these findings.
SLNB status is a highly significant predictor of outcome in thin-melanoma patients. However, SLNB should not be used in all thin melanomas. We found no positive SLNB result in melanoma<.6 mm deep. In such lesions, the false-negative rate of the SLNB procedure may exceed the likelihood of identifying nodal metastases. However, these data support SLNB in selected thin-melanoma patients with a Breslow thickness 
.75 mm—particularly those with Clark level of invasion of IV or higher or an intermediate or higher mitotic index. Further investigation is warranted with larger series and longer follow-up to formulate specific recommendations for the use of SLNB in patients with thin melanomas.
Received for publication April 14, 2004. Accepted for publication December 19, 2005.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1 mm level III and IV melanomas are higher risk lesions for regional failure and warrant sentinel lymph node biopsy. Ann Surg Oncol 2000; 7:456–60.[Abstract]1 mm) with vertical growth phase. Ann Surg Oncol 2000; 7:262–7.[Abstract]This article has been cited by other articles:
 |
|
 |
|
  B. E. Wright, R. P. Scheri, X. Ye, M. B. Faries, R. R. Turner, R. Essner, and D. L. Morton Importance of Sentinel Lymph Node Biopsy in Patients With Thin Melanoma Arch Surg, September 1, 2008; 143(9): 892 - 900. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. J. Lim, B. W. Feig, H. Wang, K. K. Hunt, M. A. Rodriguez-Bigas, J. M. Skibber, V. Ellis, K. Cleary, and G. J. Chang Sentinel Lymph Node Evaluation Does Not Improve Staging Accuracy in Colon Cancer Ann. Surg. Oncol., January 1, 2008; 15(1): 46 - 51. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. L. Pennie, S. L. Soon, J. B. Risser, E. Veledar, S. D. Culler, and S. C. Chen Melanoma Outcomes for Medicare Patients: Association of Stage and Survival With Detection by a Dermatologist vs a Nondermatologist Arch Dermatol, April 1, 2007; 143(4): 488 - 494. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
