|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Journal of Clinical Oncology, Vol 23, No 3 (January 20), 2005: pp. 518-524 © 2005 American Society of Clinical Oncology. DOI: 10.1200/JCO.2005.00.075 Prognostic Implications of Multiple Lymphatic Basin Drainage in Patients With Truncal MelanomaFrom the Departments of Surgery, Epidemiology and Biostatistics, and Pathology, Memorial Sloan-Kettering Cancer Center, New York, NY Address reprint requests to Mary Sue Brady, MD, Gastric and Mixed Tumor Service, Memorial Sloan-Kettering Cancer Center, 1275 York Ave, New York, NY 10021; e-mail: bradym{at}mskcc.org
PURPOSE: Lymphatic drainage to multiple basins (MLBD) is frequently observed in patients with truncal melanoma undergoing sentinel lymph node (SLN) mapping. Recently published data suggest that patients with MLBD are at increased risk of nodal metastases compared with those with single lymphatic basin drainage (SLBD). We studied the impact of MLBD on SLN positivity and survival. PATIENTS AND METHODS: We identified 266 patients with truncal melanoma undergoing SLN mapping and biopsy from 1995 to 2001. MLBD was defined as lymphoscintigraphic and intraoperative identification of an SLN in more than one nodal basin. Clinical and pathologic variables were recorded and analyzed for their impact on survival. RESULTS: MLBD occurred in 76 patients (29%), and SLBD occurred in 190 patients (71%). Clinical and pathologic variables were similar between the two groups, although there were more males in the MLBD group (78% v 64%; P = .034). Patients with MLBD did not have higher risk for positive SLNs compared with those with SLBD (22% v 21%, respectively). Five-year survival for patients with MLBD was less favorable than that of patients with SLBD (68% v 78%, respectively; log-rank P = .04). Multivariate analysis revealed that increasing Breslow thickness (P < .001), SLN metastasis (P < .001), and MLBD (P = .04) were independent predictors of poor overall and relapse-free survival. The negative prognostic implication of MLBD remained significant when only patients with negative SLNs were analyzed (relative risk, 2.7; P = .03). CONCLUSION: MLBD in patients with truncal melanoma undergoing SLN mapping is associated with a less favorable survival compared with patients with SLBD, independent of SLN status.
The strongest prognostic indicators for melanoma include sentinel lymph node (SLN) status, Breslow thickness, and tumor ulceration.1 To a lesser degree, truncal as opposed to extremity location is also known to be an independent predictor of adverse outcome.2 The reason for this is unknown, but may be due to the propensity for truncal melanomas to drain to more than one lymphatic basin.3 In fact, patients with truncal melanoma presenting with clinically evident lymph node metastases in more than one lymphatic bed are known to have a particularly poor prognosis when compared with patients with palpable nodal metastasis in a single basin.4,5 The introduction and widespread use of lymphatic mapping and sentinel lymph node biopsy over the last decade has made lymphoscintigraphy a standard approach in the evaluation of patients with cutaneous melanoma at risk for lymph node metastasis.6 Indeed, the pathologic status of the SLN is the most powerful predictor of outcome in patients with clinically negative regional lymph nodes.7 Drainage to more than one regional nodal basin is observed in 5% to 49% of patients with truncal melanoma undergoing SLN mapping.5,8-11 No one has previously studied the prognostic implications of multiple lymphatic basin drainage (MLBD) as a clinical observation distinct from the pathologic status of the regional nodes. Two recent studies report contradictory findings. A study from University of Texas M.D. Anderson Cancer Center (Houston, TX), involving 281 patients, reported MLBD in 31% of patients.12 They reported that patients with MLBD were twice as likely to have a positive SLN compared with those with single lymphatic basin drainage (SLBD; 30% v 16%; P = .01). A second study of 702 patients from the Sydney Melanoma Unit (Sydney, Australia) reported a 46% incidence of MLBD in patients with truncal melanoma.13 These investigators found SLN metastases in 16% of patients, and MLBD was not found to increase the risk of a positive SLN. Survival was poor in 13 patients with SLN metastases in two or more basins, nine of whom died within 2 years of diagnosis. It has been the clinical impression of one of the investigators (M.S.B.) that SLN status is less predictive of outcome in patients with truncal melanoma compared with patients with extremity primary site. For this reason, and because of published data suggesting a higher risk of SLN metastasis in MLBD patients with truncal melanoma, we reviewed our experience with patients with truncal melanoma undergoing SLN biopsy at our institution from 1995 to 2001. Our aims were to determine the impact of MLBD on incidence of SLN metastasis and survival in patients with truncal melanoma undergoing SLN mapping.
Approval to conduct the study was obtained from the institutional review board of Memorial Sloan-Kettering Cancer Center (New York, NY). We then identified 285 patients with primary cutaneous melanoma of the trunk who underwent SLN mapping and biopsy at Memorial Sloan-Kettering Cancer Center from July 1995 to November 2001. All patients had histologically confirmed melanoma and clinically negative regional lymph nodes. Clinical and pathologic characteristics and follow-up information were collected retrospectively for analysis. Follow-up information was collected through August 1, 2002. Lymphoscintigraphy was performed in all patients to identify draining lymphatic basins. This was accomplished by injecting 400 µCi of technetium-99m–sulfur colloid into the dermis adjacent to the tumor or biopsy site on the day of surgery. Gamma imaging was then performed documenting the location of the SLN(s). These films were provided to the surgeon at the time of operation. Our lymphatic mapping technique combines preoperative lymphoscintigraphy with intraoperative injection of isosulfan blue dye. Before wide excision of the primary melanoma, 0.5 to 1.0 mL of isosulfan blue dye (Lymphazurin 1%, Hirsh Industries, Inc., Richmond, VA) is injected intradermally adjacent to the lesion or biopsy site. All radioactive nodes or blue-stained nodes are removed at the time of SLN biopsy. Bed counts less than 10% of the hottest SLN indicate adequate removal of all SLNs in a regional nodal basin. SLNs were examined using serial sectioning and hematoxylin and eosin staining. Beginning in late 1997, immunohistochemistry (S-100, HMB-45) was routinely applied if hematoxylin and eosin staining did not reveal evidence of metastatic disease.
All patients underwent wide excision of the primary site based on tumor thickness, with margins consistent with current standards. Patients with melanoma Eight nodal basins were defined for the purposes of this study (with right and left designated accordingly): cervical or supraclavicular, axillary, internal mammary, and inguinal. In-transit nodes were not considered a separate basin. MLBD was defined as drainage to two or more of the named basins as visualized on preoperative lymphoscintigraphy (Fig 1).
Nineteen of the 285 patients were excluded from the study. In seven patients this was due to lack of radiotracer migration to any lymph node basin on lymphoscintigraphy. Twelve other patients with MLBD by lymphoscintigraphy who were explored in only a single lymph node basin were also excluded. These patients had weak radiotracer uptake visualized in one of the mapped basins, which was not detectable using the hand-held gamma-probe in the operating room. Their exclusion is based on incomplete pathologic information from one of the scintigraphically identified nodal basins.
Statistical comparisons between groups were assessed using a t test or
The clinical characteristics of all 266 patients are summarized in Table 1. Nearly two-thirds of the patients were male, and the median age of the patient group was 56 years (range, 12 to 88 years). There was no difference in median age between the two study groups. MLBD was observed in 29% of the patients. More males, however, were present in the MLBD group (78% v 64%; P = .03).
Pathologic information is provided in Table 2. There was no significant difference between patients with MLBD versus SLBD with regard to median Breslow depth of the primary tumor (1.7 v 1.8 mm, respectively) or presence of primary tumor ulceration (29% v 30%, respectively). The average number of SLNs removed was 2.2 in the SLBD group, and 3.4 in the MLBD group. Twenty-one percent of all patients were found to harbor metastases in their SLNs. Almost identical rates of SLN metastases were observed in the SLNB (21%) and the MLBD (22%) groups. Four patients with MLBD had positive SLNs in two or more basins. Of the 56 patients with SLN metastases, 52 underwent completion lymphadenectomy. Ten percent (five of 52) of these patients were found to have additional lymph node metastases in their lymphadenectomy specimens.
Follow-up and disease recurrence data are summarized in Table 3. At a median follow-up of 27 months, 20% of patients had recurrent disease. Recurrence occurred in 17% of patients with SLBD, and 25% of those with MLBD. Most patients (86%) who developed recurrence had a systemic recurrence, with a minority of patients sustaining isolated locoregional recurrence (13%). Isolated locoregional recurrence was more common in the SLBD than in the MLBD group (18% v 5%; P = not significant).
Univariate analysis of RFS revealed four factors associated with unfavorable prognosis: SLN micrometastasis, increasing Breslow depth category, ulceration of the primary lesion, and male sex (Table 4). Univariate analysis of factors affecting OS revealed that male sex was no longer significant (relative risk, 1.7; P = .15). Ulceration of the primary tumor, increased Breslow thickness category, SLN metastasis, and MLBD were associated with a significantly increased risk of death (Table 5). Patients with MLBD had a less favorable RFS distribution, although this was of borderline significance (Fig 2). The OS survival distribution of patients with MLBD, however, was significantly less favorable compared with those with SLBD (Fig 3).
Multivariate analysis revealed that SLN micrometastasis, increasing Breslow depth category, ulceration of the primary lesion, and MLBD were independent predictors of less favorable RFS (Table 6) and OS (Table 7). Ulceration of the primary lesion was an independent predictor of RFS (P = .01) but not OS (P = .09). Male sex had no significant impact on RFS or OS in multivariate analysis (P = .11 for both models).
We then analyzed the 210 patients who had negative SLNs to determine whether MLBD remained independently predictive of outcome. Increasing Breslow depth category and ulceration remained significant for both RFS and OS (Tables 8 and 9). MLBD was of borderline statistical significance for RFS (P = .057) and was independently predictive of poor OS (P = .03). In addition, patients with truncal melanoma and negative SLNs had a less favorable Kaplan-Meier survival distribution (RFS and OS) when negative SLNs were identified in more than one regional nodal basin compared with patients with negative SLNs in a single basin (Fig. 4 and 5).
We observed MLBD in 29% of our patients with truncal melanoma, consistent with other large series, where the reported frequency of MLBD ranges from 30% to 40%.12,13 No major differences in clinical features, except for sex, were detected between the SLBD and MLBD groups. Although the MLBD group included more males than the SLBD group, both groups had a predominance of males (> 60%). There were no significant differences between the two groups with regard to the strongest prognostic factors (ie, SLN status, primary tumor thickness, or primary tumor ulceration). In our experience MLBD is not a risk factor for SLN metastases. SLN metastases occurred in 21% of patients, and the frequency of metastasis to the SLN was not different between the SLBD and MLBD groups. Accordingly, our results are consistent with the experience reported from the Sydney Melanoma Unit13 and contrary to those reported from the M.D. Anderson Cancer Center.12 The 21% incidence of SLN metastasis reported in this study for melanomas of the trunk is somewhat higher than the 17% rate of SLN metastasis observed for all melanomas treated at our institution.14 Certainly, the higher rate of SLN metastasis observed in truncal versus extremity melanomas may partly explain the negative prognostic implications associated with truncal location of the primary lesion. Truncal primary lesion retains prognostic significance even in patients with positive regional lymph nodes, however, suggesting that additional factors negatively influence clinical outcome in addition to a greater propensity for lymphatic metastasis.15 Our study is the first to identify MLBD as an independent predictor of poorer survival in patients with truncal melanoma. To ensure that this observation was accurate, we evaluated the impact of MLBD on the subgroup of patients with negative nodes. Patients with MLBD had a relative risk of death of 2.7 compared with those with SLBD, despite negative regional lymph nodes. A conclusive explanation for the negative prognostic significance of MLBD cannot be offered from the data presented in this study. Primary melanomas that exhibit MLBD may secrete vascular endothelial growth factors that facilitate the development of peritumoral lymphatics. In a study of murine melanoma, Padera et al16 demonstrated that mice bearing orthotopically implanted B16 melanoma overexpressing vascular endothelial growth factor C exhibited an increase in lymphatic metastasis when compared with control animals bearing mock transfected tumors. Interestingly, this appeared to be due to an increase in density and diameter of peritumor lymphatics by the vascular endothelial growth factor C–expressing tumors. MLBD may be a manifestation of a primary tumor with increased potential for the production of lymphangiogenic factors, with resultant enhanced metastatic potential. However, this does not explain our observation that patients with MLBD had no increase in risk of SLN metastasis. It may be that MLBD occurs in tumors with a greater propensity for hematogenous metastasis, or that factors secreted by the primary tumor allow regional lymph node metastasis to transgress the lymph node and thereby evade pathologic detection. Our survival data confirm the strong prognostic significance of tumor thickness, ulceration, and SLN status in patients with cutaneous melanoma in addition to MLBD. Univariate and multivariate analysis revealed that each of these factors was a highly statistically significant and independent predictor of survival. In summary, approximately 30% of patients with primary cutaneous melanoma of the trunk will demonstrate MLBD. In our experience, MLBD is not a risk factor for SLN metastases, but it is an independent predictor of adverse OS and RFS. This association may explain the negative prognostic significance of truncal primary site in patients with cutaneous melanoma.
The authors indicated no potential conflicts of interest.
We thank Annie DiMario for editorial assistance.
Presented at the 39th Annual Meeting of the American Society of Clinical Oncology, Chicago, IL, May 31-June 3, 2003. Authors' disclosures of potential conflicts of interest are found at the end of this article.
1. Balch CM, Reintgen DS, Kirkwood JM, et al (eds). Cancer: Principles and Practice of Oncology. Philadelphia, PA, Lippincott-Raven, 1997, pp 1947-1994 2. 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). Ann Surg Oncol 7:87-97, 2000[CrossRef][Medline] 3. McMasters KM: Multiple nodal basin drainage in truncal melanomas. Ann Surg Oncol 7:249-250, 2000[CrossRef][Medline] 4. Barth RJ, Venzon DJ, Baker AR: The prognosis of melanoma patients with metastases to two or more lymph node areas. Ann Surg 214:125-130, 1991[Medline] 5. Coit DG, Rogatko A, Brennan MF: Prognostic factors in patients with melanoma metastatic to axillary or inguinal lymph nodes. Ann Surg 214:627-636, 1991[Medline] 6. Ross MI, Reintgen DS, Balch CM: Selective lymphadenectomy: Emerging role of lymphatic mapping and sentinel node biopsy in the management of early stage melanoma. Semin Surg Oncol 9:219-223, 1993[Medline] 7. Bonenkamp JJ, Daley C, Colman D, et al: Drainage to multiple lymphatic fields is not an independent prognostic factor in truncal melanoma. Melanoma Res 11:S78-S79, 2001 (suppl 1) 8. Berger DH, Feig BW, Podoloff D: Lymphoscintigraphy as a predictor of lymphatic drainage from cutaneous melanoma. Ann Surg Oncol 4:13-18, 1997[CrossRef][Medline] 9. Fisher EB, Lewis VL, Griffith BH, et al: The role of cutaneous lymphoscintigraphy in determining regional lymph node drainage of truncal melanoma. Ann Plast Surg 28:506-510, 1992[CrossRef][Medline] 10. Lock-Anderson J, Rossing N, Drzewiecki KT: Pre-operative cutaneous lymphoscintigraphy in malignant melanoma. Cancer 63:77-82, 1989[CrossRef][Medline] 11. Norman J, Wells K, Kearney R, et al: Identification of lymphatic drainage basins in patients with cutaneous melanoma. Semin Surg Oncol 9:224-227, 1993[Medline] 12. Porter GA, Ross MI, Berman RS, et al: Significance of multiple nodal basin drainage in truncal melanoma patients undergoing sentinel lymph node biopsy. Ann Surg Oncol 7:256-261, 2000[CrossRef][Medline] 13. Bonenkamp JJ, Daley C, Colman M, et al: Drainage to multiple lymphatic fields is not an independent prognostic factor in truncal melanoma. Melanoma Res 11:S78-S79, 2001 14. Clary BM, Brady MS, Lewis JJ, et al: Sentinel lymph node biopsy in the management of patients with primary cutaneous melanoma: Review of a large single-institutional experience with an emphasis on recurrence. Ann Surg 233:250-258, 2001[CrossRef][Medline]
15. Balch CM, Soong S, 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 19:3622-3634, 2001
16. Padera TP, Kadambi A, Di Tomaso E: Lymphatic metastasis in the absence of functional intratumoral lymphatics. Science 296:1883-1886, 2002 Submitted July 16, 2004; accepted October 13, 2004.
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2005 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
1 mm in Breslow depth underwent excision with 1-cm margins. Those with melanoma more than 1 and 
2 cm were obtained for patients with melanoma > 2 mm in Breslow depth. Patients with metastasis detected in the SLN were offered completion lymph node dissection of the affected nodal basin. 
2 test where applicable. Relapse-free survival (RFS) and overall survival (OS) distributions were estimated using the Kaplan-Meier method. Univariate comparisons of survival distributions were made using the log-rank test. Multivariate analyses were assessed via the Cox proportional hazards model. For all analyses, a P value of 
