Sentinel Lymphadenectomy Does Not Increase the Incidence of In-Transit Metastases in Primary Melanoma
http://www.100md.com
《临床肿瘤学》
the Roy E. Coats Research Laboratories of the John Wayne Cancer Institute at Saint John’s Health Center, Santa Monica, CA
ABSTRACT
PURPOSE: Recent reports by European investigators suggest that sentinel lymphadenectomy (SLND), a mainstay of melanoma diagnosis and treatment planning, increases the risk of in-transit metastasis (ITM) and should be abandoned. This study compared the incidence of ITM after wide local excision (WLE), WLE plus SLND (SLND), or WLE plus elective lymphadenectomy (ELND) for primary melanoma.
PATIENTS AND METHODS: Review of our prospective database identified 4,412 patients who underwent WLE (n = 2,771), SLND (n = 1,016), or ELND (n = 625) for stage I/II melanoma (1971 through 2002). The incidence of ITM overall and as a first recurrence was examined before and after computerized prognostic matching of treatment groups. Intergroup statistical comparisons used 2 analysis and log-rank test.
RESULTS: The incidence of ITM increased with Breslow depth, Clark level, and T stage. Although overall incidence of ITM was significantly higher (P = .0008) after ELND (6.56%) versus WLE (3.36%) or SLND (3.64%), the ELND group had higher risk primaries. Treatment groups matched by T stage (1,875 patients; 625 per group) or by age, sex, Breslow depth, and primary location (1,680 patients; 560 per group), showed no significant differences in ITM overall or as a first recurrence.
CONCLUSION: There is no relationship between SLND and ITM. Recent reports to the contrary reflect analysis of significantly smaller cohorts not matched for confounding variables such as T stage. The phase III Multicenter Selective Lymphadenectomy Trial will definitively settle the issue; until then, use of SLND, the most accurate staging procedure for early-stage melanoma, should continue.
INTRODUCTION
Since its introduction at the 1990 Annual Meeting of the Society of Surgical Oncology,1 and its first published description in 1992,2 intraoperative lymphatic mapping with sentinel lymphadenectomy (SLND) and selective complete lymph node dissection has become a mainstay for the diagnosis and treatment planning of early-stage melanoma.3 SLND was introduced as an alternative to elective lymphadenectomy (ELND) in the management of the regional lymph nodes in early-stage melanoma. ELND has remained controversial since 1892, when Snow4 first recommended routine complete lymphadenectomy in all primary melanoma patients without clinical evidence of regional metastasis. The concept initially was that if metastatic melanoma progresses sequentially from the primary site to the regional lymph nodes and then to more distant sites, early removal of these nodes would interrupt the metastatic cascade.
Results of multiple retrospective single-center studies support Snow’s approach. These studies report a survival advantage when patients without clinical evidence of nodal metastasis undergo routine ELND versus wide local excision (WLE) alone. However, four randomized prospective studies have failed to show any overall survival advantage for ELND versus observation after WLE.5-8 These studies raised the possibility that regional metastasis may be a marker of systemic disease instead of the primary metastatic pathway. Distant site metastasis can occur even without involvement of the regional nodes. To further confuse the issue, the Intergroup Melanoma Trial demonstrated a survival advantage with ELND for certain subsets of patients with intermediate-thickness primary tumors.7
The value of ELND is also controversial because of the cost and potential complications of the procedure. Because only approximately 20% of patients with intermediate-thickness melanoma will have metastasis to regional nodes,9 80% of ELND patients will not benefit from the procedure and are at risk for acute and chronic problems from ELND.10 Obviously, the costs of operative and postoperative care are much greater when WLE is followed by ELND.11
As described by Morton et al,1-3 resection and histopathologic examination of the sentinel lymph node preserves the diagnostic advantage of ELND without its potential morbidity. Intraoperative lymphatic mapping is based on the hypothesis that melanoma metastasizes to one or more lymph nodes representative of the regional basin (sentinel lymph nodes). Patients whose sentinel nodes (SNs) are negative for tumor do not have nodal metastasis and need not undergo complete nodal dissection. Refinements in technique (preoperative lymphoscintigraphy, gamma counter use) and increased ability to identify SN micrometastasis (multiple sectioning, immunohistochemistry, carbon dye) have increased the accuracy of SLND for staging nodal basins at risk for metastasis from primary cutaneous melanoma. Multiple centers have demonstrated that SN-negative patients have a better prognosis than SN-positive patients. The accuracy and success of SLND worldwide led the WHO to declare intraoperative lymphatic mapping and SLND the standard of care for melanoma.12 SLND certainly has become standard practice in the United States. The results of the phase III Multicenter Selective Lymphadenectomy Trial (MSLT)13 should provide additional validation of SLND as standard of care.
Despite such widespread acceptance and the almost universally held belief that regional nodal status is the most important predictor of survival in metastatic melanoma, several publications recently have suggested that SLND should be abandoned as a staging procedure because it might increase the risk of in-transit metastasis (ITM).14-16 The first of these reports by Estourgie et al14 cited a 23% incidence of ITM in patients with a positive SN. These studies report a higher incidence of ITM during long-term follow-up of patients who underwent WLE plus SLND versus WLE alone. They state that SLND adversely affects the pattern of spread and recurrence of melanoma. In a review of the SLND literature, Thomas and Clark15 calculated a 9% overall incidence of local metastasis or ITM; the incidence decreased to 5.7% in patients who had tumor-negative SNs and did not undergo completion lymphadenectomy, and increased to 20.9% in those who underwent completion lymphadenectomy because of tumor-positive SNs. They concluded that SLND increases the risk of ITM by two- to four-fold and should not be widely used until validated by randomized controlled trials.
To examine the validity of these reports, we reviewed the ITM recurrence rate in patients who underwent SLND at our melanoma referral center, which has a uniquely extensive clinical database. We compared the incidence of ITM after WLE alone, WLE plus SLND, and WLE plus ELND for primary cutaneous melanoma. Because we could conceive of no physiologic reason for SLND to increase ITM, we hypothesized that there would be no significant difference in the rate of ITM among the three treatment groups.
PATIENTS AND METHODS
Patients
Patients who had undergone WLE alone or with SLND or ELND for primary cutaneous melanoma were identified by review of the John Wayne Cancer Institute (JWCI) database. This computerized database contains the records of all 11,000 patients seen by our staff since April 1971. Records are complete to within 30 months of last follow-up or death for 94% of patients in the database. This database represents a 30-year prospective audit of the results of therapy by JWCI staff. All patients who underwent treatment by our group at the University of California, Los Angeles Division of Surgical Oncology (1971 to 1991) or at JWCI (1991 to 2002), or were referred to us within 4 months of primary diagnosis for postoperative treatment/follow-up during the same period, were eligible for this study.
Outcome Measures and Statistical Methods
The rate of ITM as a first recurrence and the overall rate of ITM were determined for each treatment group. Overall ITM was defined as the incidence of ITM recurrence at any time after diagnosis and treatment of primary cutaneous melanoma. Estimated survival curves for overall ITM and ITM as a first recurrence were constructed by the nonparametric Kaplan-Meier method. Differences in survival distributions were tested by the log-rank method according to different levels of risk factors. Differences in intergroup frequency distributions and proportions were analyzed by 2 analysis.
To reduce intergroup variability, outcomes were reassessed after matching patients by various prognostic factors. Matched pairs were identified by a computer program that matched at a 1:1 ratio based on covariates selected for matching. For matched-pair analysis, survival times were censored for patients who were alive at the last follow-up or who died without evidence of melanoma.
RESULTS
Complete data were available for 4,412 patients with American Joint Committee on Cancer stage I or II melanoma. Of these, 2,771 underwent WLE alone, 1,016 underwent WLE plus SLND, and 625 underwent WLE plus ELND. Most primary tumors were on the trunk or extremities (83%) and most (86.4%) had a Clark level of II, III, or IV. Approximately half (53.9%) of the patients were male, and each treatment group had more males than females. Most patients (72.2%) were no older than 50 years; the SLND group contained the highest percentage of patients older than 50 years (51.1%).
Complete data on Breslow thickness were available for 3,832 of the 4,412 patients. Almost 83% of the WLE group had primary tumors 2 mm or less. Melanomas thicker than 2 mm represented 27.8% of the SLND group, 36.4% of the ELND group, and 17.2% of the WLE group. Clark level analysis yielded more striking intergroup differences (Table 1): almost half of the WLE group (46.9%) had Clark level I or II primary tumors, compared with 10.5% and 3.6% of SLND and ELND groups, respectively.
Primary lesions were ulcerated (9.3%), nonulcerated (45%), or had an unknown ulceration status (45.9%). Among patients whose ulceration status was known, the rate of ulceration was 13.2%, 14.2%, and 6.7% for SLND, ELND, and WLE groups, respectively. All three treatment groups had an approximately equal distribution of melanomas in extremity, head and neck, and trunk sites.
For all patients, overall ITM occurred in 93 WLE patients (3.36%), 37 SLND patients (3.64%), and 41 ELND patients (6.56%); the overall incidence of ITM was significantly higher for the ELND group than the two other treatment groups (P = .0008). The median time to onset of overall ITM was 24.5, 26.9, and 17.8 months in WLE, SLND, and ELND groups, respectively. ITM as a first recurrence occurred in 44 WLE patients (1.59%), 17 SLND patients (1.67%), and 16 ELND patients (2.56%); there was no significant intergroup difference (P = .2405). Median time to onset for ITM as a first recurrence was 28.3, 28.3, and 17.9 months for WLE, SLND, and ELND groups, respectively. The earlier onset of overall and first-recurrence ITM in the ELND group was not significant (P = .9240 and 0.4476, respectively).
As shown in Table 2, the overall incidence of ITM increased with the depth of the primary lesion. Because this direct relationship could bias statistical comparisons of ITM rates, patients from each group were matched by T stage. The computer program identified 1,875 patients matched by T stage (625 in each treatment group). As listed in Table 3, there was no significant difference (P = .3362 and P = .8465, respectively) in rates of overall ITM (5.8%, 4.6%, or 6.6%) or ITM as a first recurrence (2.2%, 2.1%, 2.6%) in matched patients from WLE, SLND, and ELND groups. The time to ITM overall or as a first recurrence also was not significantly different among sets of patients matched by T stage (P = .9240 and P = .4476, respectively).
Because intergroup differences in other characteristics of the primary melanoma could bias statistical analysis of ITM rates, patients from each treatment group were matched by age, sex, Breslow depth, and primary site. Ulceration was not used for matching because data were not available for approximately half of the study group. The computer program identified 1,680 patients matched for age, sex, Breslow depth, and primary site (560 in each treatment group). Overall ITM recurrence rates were 6.3% for WLE, 3.9% for SLND, and 6.1% for ELND; corresponding rates of ITM as a first recurrence were 1.8%, 1.4%, and 2.3%, respectively. These differences were not significant (Table 3).
Before patients were matched, Kaplan-Meier survival curves for the entire cohort of patients showed no significant treatment-related differences in the rate of ITM as a first recurrence (P = .3048; Fig 1), but WLE and SLND groups fared significantly better than the ELND group with respect to overall ITM (P = .0029; Fig 2). After patients in the three treatment groups were matched by T stage, the Kaplan-Meier survival curves for overall ITM showed no differences (P = .4476; Fig 3). Similarly, survival curves for patients matched by age, sex, primary site, and Breslow thickness showed no significant treatment-related differences in overall ITM (P = .1843; Fig 4).
The status of the regional lymph nodes, which is the most significant prognostic factor in patients with early-stage melanoma, was available only for patients in SLND and ELND groups. The overall incidence of ITM was 8.2% for node-positive SLND patients, 3.1% for node-negative SLND patients, 10.4% for node-positive ELND patients, and 5.7% for node-negative ELND patients (Table 4). The higher incidence of ITM in SLND and ELND patients with tumor-positive nodes reflects the higher rate of thicker, more ulcerated lesions in these subgroups. Primary lesions thicker than 2 mm were found in 24% of node-negative SLND patients, more than 55% of node-positive SLND patients, 30% of node-negative ELND patients, and 66% of node-positive ELND patients. In the subgroup of SLND and ELND patients whose records included data on ulceration, the rate of ulceration was approximately 25% in node-positive patients and 12% in node-negative patients.
DISCUSSION
Although SLND is widely accepted as a minimally invasive method of staging the regional lymph nodes for melanoma, its use is still being challenged. McMasters et al17 eloquently stated four major reasons why SLND should be performed. First, SLND improves accuracy of staging and provides valuable prognostic information to guide subsequent follow-up and treatment. Second, SLND facilitates early therapeutic lymph node dissection for patients with nodal metastases. Third, SLND identifies patients who are candidates for adjuvanttherapy with interferon alfa-2b, which at present is the standard of adjuvant treatment. Finally, SLND identifies homogeneous patient populations for entry into clinical trials of novel adjuvant therapy.
However, does SLND increase the risk of ITM, and thereby decrease a patient’s chance of long-term disease-free survival According to Thomas et al,15 the reported incidence of ITM is 2.5% to 6.3% after WLE versus 9.0% after SLND. The incidence jumps to 20.9% when the SN is positive for tumor. However, these figures are based on a literature review of 1,406 patients, only 206 of whom had positive sentinel nodes. Moreover, the population is pooled from small studies that did not control for confounding variables such as thickness of the primary melanoma. Our large review of patients matched by important prognostic factors quite clearly shows that SLND does not increase the incidence of ITM. In fact, the incidence of ITM, overall or as a first recurrence, may be lower after SLND than WLE alone or WLE plus ELND. Similar data was also reported by van Poll et al18 from the Sydney Melanoma Unit and by investigators from M.D. Anderson (T.M. Pawlik and J.E. Gershenwald, unpublished data, January 2004) in reviews of their experience with large numbers of patients.
Almost all reports agree that SN status is the most important predictor of survival in melanoma. Gershenwald et al19 reported a 6.53 hazard ratio for survival in SN-positive patients—higher than any other prognostic factor. Proponents of SLND state that SN staging can distinguish patients with a relatively favorable prognosis from those who require additional surgery, adjuvant therapy, and more rigorous follow-up. Opponents of SLND state that the survival advantage of completion lymphadenectomy plus postoperative adjuvant therapy and careful monitoring for SN-positive patients does not outweigh any possible survival risks from SLND. Certainly ITM is a survival risk in patients with melanoma: the 10-year survival rate for patients with local or ITM recurrence and lymph node involvement is only 28% to 35%.15 However, our data failed to identify any increased incidence of ITM after SLND versus WLE alone or WLE with ELND. Moreover, the decreased survival of patients with ITM and tumor-positive regional nodes may reflect nodal metastasis; patients with American Joint Committee on Cancer stage III disease have a 10-year survival rate of approximately 35% to 40%, depending on the extent of nodal involvement and the ulceration status of the primary lesion.9
Critics of SLND have also argued that as yet there is no agreement on adjuvant therapy for node-positive patients. Studies of interferon from the Eastern Cooperative Oncology Group confirmed a disease-free survival benefit,20,21 but it is not clear whether interferon alfa can improve overall survival. Although the question regarding interferon has yet to be answered, there are other clinical trials of adjuvant therapies available. Several randomized clinical trials are currently examining vaccines, biochemotherapy, and other adjuvant treatments that could prove to be more effective than interferon alfa.
Opponents of SLND state that no randomized controlled studies have yet shown a survival advantage for SLND in SN-positive patients. This issue awaits final results of the MSLT, the largest multicenter, randomized, prospective study of SLND.13 However, even if SLND does not improve survival, the procedure has inherent value because of its high predictive accuracy for metastasis in the regional lymph basin. Unless more accurate and less invasive staging tests are developed, the use of SLND should be continued.17 Its minimal invasiveness, convenience, simplicity of concept, and cost effectiveness justify current use of SLND in patients with clinically localized cutaneous melanoma.
In conclusion, review of more than 4,000 patients treated at John Wayne Cancer Institute during a 30-year period has not shown an increased incidence of ITM after SLND versus WLE or ELND. When patients with similar primary risk factors were compared, there was no association between SLND and ITM. There has also been no survival disadvantage, either disease free or overall, in those SLND patients who subsequently developed ITM. These observations have been made in other institutions that treat a large volume of melanoma patients.18 Any suggestion that SLND be abandoned is not justified. This conclusion will be further validated once the results of the MSLT are published. Simply put, SLND is a minimally invasive, low-morbidity, cost-effective, and accurate method to assess the status of the regional node basin in primary melanoma. The status of the regional basin as determined by SLND or complete lymph node dissection is the most important predictor of survival in melanoma. Although additional studies must be performed to determine how to best apply this information in the adjuvant treatment and follow-up of high-risk melanoma patients, we strongly believe that SLND should continue as the standard of care in patients with primary cutaneous melanoma.
Authors’ Disclosures of Potential Conflicts of Interest
The authors indicated no potential conflicts of interest.
NOTES
Supported by grant CA29605 from the National Cancer Institute and by funding from the Amyx Foundation Inc (Boise, ID); Alice Johnson McKinney, the Harold J. McAlister Charitable Foundation (Los Angeles, CA); Nancy and Carroll O’Connor (Los Angeles, CA); the George Hoag Family Foundation (Los Angeles, CA [R. Essner]); and the Saban Family Foundation (Los Angeles, CA [R. Essner]).
Presented at the Annual Meeting of the Society of Surgical Oncology, Atlanta, GA, March 3-6, 2005.
Authors’ disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Morton D, Cagle L, Wong J, et al: Intraoperative lymphatic mapping and selective lymphadenectomy: Technical details of a new procedure for clinical stage I melanoma. Presented at Annual Meeting of the Society of Surgical Oncology, Washington, DC, May 22, 1990
Morton DL, Wen DR, Wong JH, et al: Technical details of intraoperative lymphatic mapping for early stage melanoma. Arch Surg 127:392-399, 1992
Morton DL, Hoon DSB, Cochran AJ, et al: Lymphatic mapping and sentinel lymphadenectomy for early-stage melanoma: Therapeutic utility and implications of nodal microanatomy and molecular staging for improving the accuracy of detection of nodal micrometastases. Ann Surg 238:538-549, 2003
Snow H: Melanotic cancerous disease. Lancet 2:872, 1892
Sim FH, Taylor WF, Ivins JC, et al: A prospective randomized study of the efficacy of routine elective lymphadenectomy in malignant melanoma: Preliminary results. Cancer 41:948-956, 1978
Veronesi U, Adamus J, Bandiera DC, et al: Delayed regional lymph node dissection in stage I melanoma of the skin of the lower extremities. Cancer 49:2420-2430, 1982
Balch CM, Soong SJ, Bartolucci AA, et al: Efficacy of an elective regional lymph node dissection of 1 to 4 mm thick melanomas for patients 60 years of age and younger. Ann Surg 224:255-263, 1996
Cascinelli N, Morabito A, Santinami M, et al: Immediate or delayed dissection of regional nodes in patients with melanoma of the trunk: A randomized trial—WHO Melanoma Programme. Lancet 351:793-796, 1998
Morton DL, Wanek L, Nizze JA, et al: Improved long-term survival after lymphadenectomy of melanoma metastatic to regional nodes: Analysis of prognostic factors in 1134 patients from the John Wayne Cancer Clinic. Ann Surg 214:491-499, 1991
Urist NM, Maddox WA, Kennedy JE, et al: Patient risk factors and surgical morbidity after regional lymphadenectomy in 204 melanoma patients. Cancer 51:2152-2156, 1983
Essner R, Conforti A, Kelley MC, et al: Cost-conscious management of the inguinal nodes in early-stage melanoma. Melanoma Res 7:S29, 1997
WHO: WHO declares lymphatic mapping to be standard of care for melanoma. Oncology 13:288, 1999
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. Ann Surg 230:453-463, 1999
Estourgie SH, Nieweg OE, Valdes Olmos RA, et al: Review and evaluation of sentinel node procedures in 250 melanoma patients with median follow up of 6 years. Ann Surg Oncol 10:681-688, 2003
Thomas JM, Clark MA: Selective lymphadenectomy in sentinel node-positive patients may increase the risk of local/in-transit recurrence in malignant melanoma. Eur J Surg Oncol 30:686-691, 2004
Estourgie SH, Nieweg OE, Kroon BBR: High incidence of in-transit metastases after sentinel node biopsy in patients with melanoma. Br J Surg 91:1370-1371, 2004
McMasters KM, Reintgen DS, Ross MI, et al: Sentinel lymph node biopsy for melanoma: Controversy despite widespread agreement. J Clin Oncol 19:2851-2855, 2001
van Poll D, Thompson JF, McKinnon JG, et al: A sentinel node biopsy procedure does not increase the incidence of in-transit recurrence in patients with primary melanoma. Ann Surg Oncol (in press)
Gershenwald JE, Thompson W, Mansfield PF, et al: Multi-institutional melanoma lymphatic mapping experience: The prognostic value of sentinel lymph node status in 612 stage I or II melanoma patients. J Clin Oncol 17:976-983, 1999
Kirkwood JM, Strawderman MH, Ernstoff MS, et al: Interferon alfa-2b adjuvant therapy of high-risk resected cutaneous melanoma: The Eastern Cooperative Oncology Group Trial EST 1684. J Clin Oncol 14:7-17, 1996
Kirkwood JM, Ibrahim J, Sondak V, et al: High- and low-dose interferon alfa-2b in high risk melanoma: First analysis of intergroup trial E1690/S9111/C9190. J Clin Oncol 18:2444-2458, 2000(John C. Kang, Leslie A. W)
ABSTRACT
PURPOSE: Recent reports by European investigators suggest that sentinel lymphadenectomy (SLND), a mainstay of melanoma diagnosis and treatment planning, increases the risk of in-transit metastasis (ITM) and should be abandoned. This study compared the incidence of ITM after wide local excision (WLE), WLE plus SLND (SLND), or WLE plus elective lymphadenectomy (ELND) for primary melanoma.
PATIENTS AND METHODS: Review of our prospective database identified 4,412 patients who underwent WLE (n = 2,771), SLND (n = 1,016), or ELND (n = 625) for stage I/II melanoma (1971 through 2002). The incidence of ITM overall and as a first recurrence was examined before and after computerized prognostic matching of treatment groups. Intergroup statistical comparisons used 2 analysis and log-rank test.
RESULTS: The incidence of ITM increased with Breslow depth, Clark level, and T stage. Although overall incidence of ITM was significantly higher (P = .0008) after ELND (6.56%) versus WLE (3.36%) or SLND (3.64%), the ELND group had higher risk primaries. Treatment groups matched by T stage (1,875 patients; 625 per group) or by age, sex, Breslow depth, and primary location (1,680 patients; 560 per group), showed no significant differences in ITM overall or as a first recurrence.
CONCLUSION: There is no relationship between SLND and ITM. Recent reports to the contrary reflect analysis of significantly smaller cohorts not matched for confounding variables such as T stage. The phase III Multicenter Selective Lymphadenectomy Trial will definitively settle the issue; until then, use of SLND, the most accurate staging procedure for early-stage melanoma, should continue.
INTRODUCTION
Since its introduction at the 1990 Annual Meeting of the Society of Surgical Oncology,1 and its first published description in 1992,2 intraoperative lymphatic mapping with sentinel lymphadenectomy (SLND) and selective complete lymph node dissection has become a mainstay for the diagnosis and treatment planning of early-stage melanoma.3 SLND was introduced as an alternative to elective lymphadenectomy (ELND) in the management of the regional lymph nodes in early-stage melanoma. ELND has remained controversial since 1892, when Snow4 first recommended routine complete lymphadenectomy in all primary melanoma patients without clinical evidence of regional metastasis. The concept initially was that if metastatic melanoma progresses sequentially from the primary site to the regional lymph nodes and then to more distant sites, early removal of these nodes would interrupt the metastatic cascade.
Results of multiple retrospective single-center studies support Snow’s approach. These studies report a survival advantage when patients without clinical evidence of nodal metastasis undergo routine ELND versus wide local excision (WLE) alone. However, four randomized prospective studies have failed to show any overall survival advantage for ELND versus observation after WLE.5-8 These studies raised the possibility that regional metastasis may be a marker of systemic disease instead of the primary metastatic pathway. Distant site metastasis can occur even without involvement of the regional nodes. To further confuse the issue, the Intergroup Melanoma Trial demonstrated a survival advantage with ELND for certain subsets of patients with intermediate-thickness primary tumors.7
The value of ELND is also controversial because of the cost and potential complications of the procedure. Because only approximately 20% of patients with intermediate-thickness melanoma will have metastasis to regional nodes,9 80% of ELND patients will not benefit from the procedure and are at risk for acute and chronic problems from ELND.10 Obviously, the costs of operative and postoperative care are much greater when WLE is followed by ELND.11
As described by Morton et al,1-3 resection and histopathologic examination of the sentinel lymph node preserves the diagnostic advantage of ELND without its potential morbidity. Intraoperative lymphatic mapping is based on the hypothesis that melanoma metastasizes to one or more lymph nodes representative of the regional basin (sentinel lymph nodes). Patients whose sentinel nodes (SNs) are negative for tumor do not have nodal metastasis and need not undergo complete nodal dissection. Refinements in technique (preoperative lymphoscintigraphy, gamma counter use) and increased ability to identify SN micrometastasis (multiple sectioning, immunohistochemistry, carbon dye) have increased the accuracy of SLND for staging nodal basins at risk for metastasis from primary cutaneous melanoma. Multiple centers have demonstrated that SN-negative patients have a better prognosis than SN-positive patients. The accuracy and success of SLND worldwide led the WHO to declare intraoperative lymphatic mapping and SLND the standard of care for melanoma.12 SLND certainly has become standard practice in the United States. The results of the phase III Multicenter Selective Lymphadenectomy Trial (MSLT)13 should provide additional validation of SLND as standard of care.
Despite such widespread acceptance and the almost universally held belief that regional nodal status is the most important predictor of survival in metastatic melanoma, several publications recently have suggested that SLND should be abandoned as a staging procedure because it might increase the risk of in-transit metastasis (ITM).14-16 The first of these reports by Estourgie et al14 cited a 23% incidence of ITM in patients with a positive SN. These studies report a higher incidence of ITM during long-term follow-up of patients who underwent WLE plus SLND versus WLE alone. They state that SLND adversely affects the pattern of spread and recurrence of melanoma. In a review of the SLND literature, Thomas and Clark15 calculated a 9% overall incidence of local metastasis or ITM; the incidence decreased to 5.7% in patients who had tumor-negative SNs and did not undergo completion lymphadenectomy, and increased to 20.9% in those who underwent completion lymphadenectomy because of tumor-positive SNs. They concluded that SLND increases the risk of ITM by two- to four-fold and should not be widely used until validated by randomized controlled trials.
To examine the validity of these reports, we reviewed the ITM recurrence rate in patients who underwent SLND at our melanoma referral center, which has a uniquely extensive clinical database. We compared the incidence of ITM after WLE alone, WLE plus SLND, and WLE plus ELND for primary cutaneous melanoma. Because we could conceive of no physiologic reason for SLND to increase ITM, we hypothesized that there would be no significant difference in the rate of ITM among the three treatment groups.
PATIENTS AND METHODS
Patients
Patients who had undergone WLE alone or with SLND or ELND for primary cutaneous melanoma were identified by review of the John Wayne Cancer Institute (JWCI) database. This computerized database contains the records of all 11,000 patients seen by our staff since April 1971. Records are complete to within 30 months of last follow-up or death for 94% of patients in the database. This database represents a 30-year prospective audit of the results of therapy by JWCI staff. All patients who underwent treatment by our group at the University of California, Los Angeles Division of Surgical Oncology (1971 to 1991) or at JWCI (1991 to 2002), or were referred to us within 4 months of primary diagnosis for postoperative treatment/follow-up during the same period, were eligible for this study.
Outcome Measures and Statistical Methods
The rate of ITM as a first recurrence and the overall rate of ITM were determined for each treatment group. Overall ITM was defined as the incidence of ITM recurrence at any time after diagnosis and treatment of primary cutaneous melanoma. Estimated survival curves for overall ITM and ITM as a first recurrence were constructed by the nonparametric Kaplan-Meier method. Differences in survival distributions were tested by the log-rank method according to different levels of risk factors. Differences in intergroup frequency distributions and proportions were analyzed by 2 analysis.
To reduce intergroup variability, outcomes were reassessed after matching patients by various prognostic factors. Matched pairs were identified by a computer program that matched at a 1:1 ratio based on covariates selected for matching. For matched-pair analysis, survival times were censored for patients who were alive at the last follow-up or who died without evidence of melanoma.
RESULTS
Complete data were available for 4,412 patients with American Joint Committee on Cancer stage I or II melanoma. Of these, 2,771 underwent WLE alone, 1,016 underwent WLE plus SLND, and 625 underwent WLE plus ELND. Most primary tumors were on the trunk or extremities (83%) and most (86.4%) had a Clark level of II, III, or IV. Approximately half (53.9%) of the patients were male, and each treatment group had more males than females. Most patients (72.2%) were no older than 50 years; the SLND group contained the highest percentage of patients older than 50 years (51.1%).
Complete data on Breslow thickness were available for 3,832 of the 4,412 patients. Almost 83% of the WLE group had primary tumors 2 mm or less. Melanomas thicker than 2 mm represented 27.8% of the SLND group, 36.4% of the ELND group, and 17.2% of the WLE group. Clark level analysis yielded more striking intergroup differences (Table 1): almost half of the WLE group (46.9%) had Clark level I or II primary tumors, compared with 10.5% and 3.6% of SLND and ELND groups, respectively.
Primary lesions were ulcerated (9.3%), nonulcerated (45%), or had an unknown ulceration status (45.9%). Among patients whose ulceration status was known, the rate of ulceration was 13.2%, 14.2%, and 6.7% for SLND, ELND, and WLE groups, respectively. All three treatment groups had an approximately equal distribution of melanomas in extremity, head and neck, and trunk sites.
For all patients, overall ITM occurred in 93 WLE patients (3.36%), 37 SLND patients (3.64%), and 41 ELND patients (6.56%); the overall incidence of ITM was significantly higher for the ELND group than the two other treatment groups (P = .0008). The median time to onset of overall ITM was 24.5, 26.9, and 17.8 months in WLE, SLND, and ELND groups, respectively. ITM as a first recurrence occurred in 44 WLE patients (1.59%), 17 SLND patients (1.67%), and 16 ELND patients (2.56%); there was no significant intergroup difference (P = .2405). Median time to onset for ITM as a first recurrence was 28.3, 28.3, and 17.9 months for WLE, SLND, and ELND groups, respectively. The earlier onset of overall and first-recurrence ITM in the ELND group was not significant (P = .9240 and 0.4476, respectively).
As shown in Table 2, the overall incidence of ITM increased with the depth of the primary lesion. Because this direct relationship could bias statistical comparisons of ITM rates, patients from each group were matched by T stage. The computer program identified 1,875 patients matched by T stage (625 in each treatment group). As listed in Table 3, there was no significant difference (P = .3362 and P = .8465, respectively) in rates of overall ITM (5.8%, 4.6%, or 6.6%) or ITM as a first recurrence (2.2%, 2.1%, 2.6%) in matched patients from WLE, SLND, and ELND groups. The time to ITM overall or as a first recurrence also was not significantly different among sets of patients matched by T stage (P = .9240 and P = .4476, respectively).
Because intergroup differences in other characteristics of the primary melanoma could bias statistical analysis of ITM rates, patients from each treatment group were matched by age, sex, Breslow depth, and primary site. Ulceration was not used for matching because data were not available for approximately half of the study group. The computer program identified 1,680 patients matched for age, sex, Breslow depth, and primary site (560 in each treatment group). Overall ITM recurrence rates were 6.3% for WLE, 3.9% for SLND, and 6.1% for ELND; corresponding rates of ITM as a first recurrence were 1.8%, 1.4%, and 2.3%, respectively. These differences were not significant (Table 3).
Before patients were matched, Kaplan-Meier survival curves for the entire cohort of patients showed no significant treatment-related differences in the rate of ITM as a first recurrence (P = .3048; Fig 1), but WLE and SLND groups fared significantly better than the ELND group with respect to overall ITM (P = .0029; Fig 2). After patients in the three treatment groups were matched by T stage, the Kaplan-Meier survival curves for overall ITM showed no differences (P = .4476; Fig 3). Similarly, survival curves for patients matched by age, sex, primary site, and Breslow thickness showed no significant treatment-related differences in overall ITM (P = .1843; Fig 4).
The status of the regional lymph nodes, which is the most significant prognostic factor in patients with early-stage melanoma, was available only for patients in SLND and ELND groups. The overall incidence of ITM was 8.2% for node-positive SLND patients, 3.1% for node-negative SLND patients, 10.4% for node-positive ELND patients, and 5.7% for node-negative ELND patients (Table 4). The higher incidence of ITM in SLND and ELND patients with tumor-positive nodes reflects the higher rate of thicker, more ulcerated lesions in these subgroups. Primary lesions thicker than 2 mm were found in 24% of node-negative SLND patients, more than 55% of node-positive SLND patients, 30% of node-negative ELND patients, and 66% of node-positive ELND patients. In the subgroup of SLND and ELND patients whose records included data on ulceration, the rate of ulceration was approximately 25% in node-positive patients and 12% in node-negative patients.
DISCUSSION
Although SLND is widely accepted as a minimally invasive method of staging the regional lymph nodes for melanoma, its use is still being challenged. McMasters et al17 eloquently stated four major reasons why SLND should be performed. First, SLND improves accuracy of staging and provides valuable prognostic information to guide subsequent follow-up and treatment. Second, SLND facilitates early therapeutic lymph node dissection for patients with nodal metastases. Third, SLND identifies patients who are candidates for adjuvanttherapy with interferon alfa-2b, which at present is the standard of adjuvant treatment. Finally, SLND identifies homogeneous patient populations for entry into clinical trials of novel adjuvant therapy.
However, does SLND increase the risk of ITM, and thereby decrease a patient’s chance of long-term disease-free survival According to Thomas et al,15 the reported incidence of ITM is 2.5% to 6.3% after WLE versus 9.0% after SLND. The incidence jumps to 20.9% when the SN is positive for tumor. However, these figures are based on a literature review of 1,406 patients, only 206 of whom had positive sentinel nodes. Moreover, the population is pooled from small studies that did not control for confounding variables such as thickness of the primary melanoma. Our large review of patients matched by important prognostic factors quite clearly shows that SLND does not increase the incidence of ITM. In fact, the incidence of ITM, overall or as a first recurrence, may be lower after SLND than WLE alone or WLE plus ELND. Similar data was also reported by van Poll et al18 from the Sydney Melanoma Unit and by investigators from M.D. Anderson (T.M. Pawlik and J.E. Gershenwald, unpublished data, January 2004) in reviews of their experience with large numbers of patients.
Almost all reports agree that SN status is the most important predictor of survival in melanoma. Gershenwald et al19 reported a 6.53 hazard ratio for survival in SN-positive patients—higher than any other prognostic factor. Proponents of SLND state that SN staging can distinguish patients with a relatively favorable prognosis from those who require additional surgery, adjuvant therapy, and more rigorous follow-up. Opponents of SLND state that the survival advantage of completion lymphadenectomy plus postoperative adjuvant therapy and careful monitoring for SN-positive patients does not outweigh any possible survival risks from SLND. Certainly ITM is a survival risk in patients with melanoma: the 10-year survival rate for patients with local or ITM recurrence and lymph node involvement is only 28% to 35%.15 However, our data failed to identify any increased incidence of ITM after SLND versus WLE alone or WLE with ELND. Moreover, the decreased survival of patients with ITM and tumor-positive regional nodes may reflect nodal metastasis; patients with American Joint Committee on Cancer stage III disease have a 10-year survival rate of approximately 35% to 40%, depending on the extent of nodal involvement and the ulceration status of the primary lesion.9
Critics of SLND have also argued that as yet there is no agreement on adjuvant therapy for node-positive patients. Studies of interferon from the Eastern Cooperative Oncology Group confirmed a disease-free survival benefit,20,21 but it is not clear whether interferon alfa can improve overall survival. Although the question regarding interferon has yet to be answered, there are other clinical trials of adjuvant therapies available. Several randomized clinical trials are currently examining vaccines, biochemotherapy, and other adjuvant treatments that could prove to be more effective than interferon alfa.
Opponents of SLND state that no randomized controlled studies have yet shown a survival advantage for SLND in SN-positive patients. This issue awaits final results of the MSLT, the largest multicenter, randomized, prospective study of SLND.13 However, even if SLND does not improve survival, the procedure has inherent value because of its high predictive accuracy for metastasis in the regional lymph basin. Unless more accurate and less invasive staging tests are developed, the use of SLND should be continued.17 Its minimal invasiveness, convenience, simplicity of concept, and cost effectiveness justify current use of SLND in patients with clinically localized cutaneous melanoma.
In conclusion, review of more than 4,000 patients treated at John Wayne Cancer Institute during a 30-year period has not shown an increased incidence of ITM after SLND versus WLE or ELND. When patients with similar primary risk factors were compared, there was no association between SLND and ITM. There has also been no survival disadvantage, either disease free or overall, in those SLND patients who subsequently developed ITM. These observations have been made in other institutions that treat a large volume of melanoma patients.18 Any suggestion that SLND be abandoned is not justified. This conclusion will be further validated once the results of the MSLT are published. Simply put, SLND is a minimally invasive, low-morbidity, cost-effective, and accurate method to assess the status of the regional node basin in primary melanoma. The status of the regional basin as determined by SLND or complete lymph node dissection is the most important predictor of survival in melanoma. Although additional studies must be performed to determine how to best apply this information in the adjuvant treatment and follow-up of high-risk melanoma patients, we strongly believe that SLND should continue as the standard of care in patients with primary cutaneous melanoma.
Authors’ Disclosures of Potential Conflicts of Interest
The authors indicated no potential conflicts of interest.
NOTES
Supported by grant CA29605 from the National Cancer Institute and by funding from the Amyx Foundation Inc (Boise, ID); Alice Johnson McKinney, the Harold J. McAlister Charitable Foundation (Los Angeles, CA); Nancy and Carroll O’Connor (Los Angeles, CA); the George Hoag Family Foundation (Los Angeles, CA [R. Essner]); and the Saban Family Foundation (Los Angeles, CA [R. Essner]).
Presented at the Annual Meeting of the Society of Surgical Oncology, Atlanta, GA, March 3-6, 2005.
Authors’ disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
Morton D, Cagle L, Wong J, et al: Intraoperative lymphatic mapping and selective lymphadenectomy: Technical details of a new procedure for clinical stage I melanoma. Presented at Annual Meeting of the Society of Surgical Oncology, Washington, DC, May 22, 1990
Morton DL, Wen DR, Wong JH, et al: Technical details of intraoperative lymphatic mapping for early stage melanoma. Arch Surg 127:392-399, 1992
Morton DL, Hoon DSB, Cochran AJ, et al: Lymphatic mapping and sentinel lymphadenectomy for early-stage melanoma: Therapeutic utility and implications of nodal microanatomy and molecular staging for improving the accuracy of detection of nodal micrometastases. Ann Surg 238:538-549, 2003
Snow H: Melanotic cancerous disease. Lancet 2:872, 1892
Sim FH, Taylor WF, Ivins JC, et al: A prospective randomized study of the efficacy of routine elective lymphadenectomy in malignant melanoma: Preliminary results. Cancer 41:948-956, 1978
Veronesi U, Adamus J, Bandiera DC, et al: Delayed regional lymph node dissection in stage I melanoma of the skin of the lower extremities. Cancer 49:2420-2430, 1982
Balch CM, Soong SJ, Bartolucci AA, et al: Efficacy of an elective regional lymph node dissection of 1 to 4 mm thick melanomas for patients 60 years of age and younger. Ann Surg 224:255-263, 1996
Cascinelli N, Morabito A, Santinami M, et al: Immediate or delayed dissection of regional nodes in patients with melanoma of the trunk: A randomized trial—WHO Melanoma Programme. Lancet 351:793-796, 1998
Morton DL, Wanek L, Nizze JA, et al: Improved long-term survival after lymphadenectomy of melanoma metastatic to regional nodes: Analysis of prognostic factors in 1134 patients from the John Wayne Cancer Clinic. Ann Surg 214:491-499, 1991
Urist NM, Maddox WA, Kennedy JE, et al: Patient risk factors and surgical morbidity after regional lymphadenectomy in 204 melanoma patients. Cancer 51:2152-2156, 1983
Essner R, Conforti A, Kelley MC, et al: Cost-conscious management of the inguinal nodes in early-stage melanoma. Melanoma Res 7:S29, 1997
WHO: WHO declares lymphatic mapping to be standard of care for melanoma. Oncology 13:288, 1999
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. Ann Surg 230:453-463, 1999
Estourgie SH, Nieweg OE, Valdes Olmos RA, et al: Review and evaluation of sentinel node procedures in 250 melanoma patients with median follow up of 6 years. Ann Surg Oncol 10:681-688, 2003
Thomas JM, Clark MA: Selective lymphadenectomy in sentinel node-positive patients may increase the risk of local/in-transit recurrence in malignant melanoma. Eur J Surg Oncol 30:686-691, 2004
Estourgie SH, Nieweg OE, Kroon BBR: High incidence of in-transit metastases after sentinel node biopsy in patients with melanoma. Br J Surg 91:1370-1371, 2004
McMasters KM, Reintgen DS, Ross MI, et al: Sentinel lymph node biopsy for melanoma: Controversy despite widespread agreement. J Clin Oncol 19:2851-2855, 2001
van Poll D, Thompson JF, McKinnon JG, et al: A sentinel node biopsy procedure does not increase the incidence of in-transit recurrence in patients with primary melanoma. Ann Surg Oncol (in press)
Gershenwald JE, Thompson W, Mansfield PF, et al: Multi-institutional melanoma lymphatic mapping experience: The prognostic value of sentinel lymph node status in 612 stage I or II melanoma patients. J Clin Oncol 17:976-983, 1999
Kirkwood JM, Strawderman MH, Ernstoff MS, et al: Interferon alfa-2b adjuvant therapy of high-risk resected cutaneous melanoma: The Eastern Cooperative Oncology Group Trial EST 1684. J Clin Oncol 14:7-17, 1996
Kirkwood JM, Ibrahim J, Sondak V, et al: High- and low-dose interferon alfa-2b in high risk melanoma: First analysis of intergroup trial E1690/S9111/C9190. J Clin Oncol 18:2444-2458, 2000(John C. Kang, Leslie A. W)