|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Journal of Clinical Oncology, Vol 24, No 27 (September 20), 2006: pp. 4464-4471 © 2006 American Society of Clinical Oncology. DOI: 10.1200/JCO.2006.06.3198 Sentinel and Nonsentinel Node Status in Stage IB and II Melanoma Patients: Two-Step Prognostic Indicators of Survival
From the Nuclear Medicine Unit, Scientific Directorate, Pathology Unit, Istituto Nazionale Tumori, Milan; Pathology Unit, Unit of General Surgery, San Pio X Hospital; and the Melanoma and Sarcoma Unit, Istituto Nazionale Tumori, Milan, Italy Address reprint requests to Natale Cascinelli, MD, Istituto Nazionale Tumori, via Venezian 1, 20133 Milan, Italy; e-mail: direzionescientifica{at}istitutotumori.mi.it
PURPOSE: To evaluate the prognostic significance of sentinel node biopsy in the management of stage IB and II melanoma patients, and to evaluate the status of nonsentinel nodes as a "second step key factor" to assess the prognosis of these patients. PATIENTS AND METHODS: We conducted an analysis of data collected in a prospective database. RESULTS: From February 1994 to June 2005, 1,108 consecutive patients with stage IB and II melanoma were submitted to sentinel node biopsy; 176 patients (15.9%) had occult node metastases. The frequency of positive nodes increased with increasing Breslow's thickness. The largest diameter of metastatic foci and their localization within the lymph node were associated with the risk of nonsentinel node metastases only. The 5-year survival of patients with positive sentinel nodes was 81.4% in patients with one positive node and 39.6% in patients with two positive nodes (P = .056). Multivariate analysis indicated that status of sentinel nodes is a key factor and that sex and Breslow's thickness maintain statistically significant relevance. Ulceration, which was associated with survival when considered as single factor (P < .001) had no impact on survival in the multivariate analysis (P = .10). To evaluate the relevance of metastases to nonsentinel nodes, we identified four groups of patients. CONCLUSION: Evaluation of the sentinel node is a useful procedure to identify patients to be submitted for complete lymph node dissection. The procedure makes it possible to assess the best prognosis of patients.
Despite the fact that sentinel node biopsy is widely used to determine the pathologic status of regional lymph nodes in stage IB and II melanoma patients, some controversies still exist. The strongest position against the procedure has recently been taken by Medalie and Ackerman,1 who stated that sentinel node biopsy should be abandoned. To contribute to the discussion of this very important issue, to verify whether or not a wait-and-see policy is acceptable in positive sentinel node patients outside ongoing clinical trials, and to identify homogeneous groups of patients to be considered for studies on adjuvant treatments, we reviewed the data of the 1,108 patients submitted to lymphatic mapping and sentinel node dissection at the National Cancer Institute in Milan, Italy, and San Pio X Hospital in Milan, Italy.
From February 1994 to June 2005, 1,108 consecutive patients with a primary cutaneous melanoma thicker than 1 mm or with Clark’s levels IV and V were submitted to lymphatic mapping and sentinel node lymphadenectomy, irrespective of Breslow’s thickness and without any clinical evidence of metastatic disease. Of these patients, 852 were treated at the National Cancer Institute and 256 at the San Pio X Hospital. All data related to our case series were recorded in a prospective database.
Lymphoscintigraphy and Intraoperative Sentinel Node Detection The patients received four intradermal injections of 99mTc-colloid particles, each one around the surgical scar. The activity for each injection was 10 to 15 MBq in a volume of nearly 0.1 mL (total, 0.4 to 0.5 mL). After radioactive tracer administration, 15 minutes dynamic planar acquisition was carried out with a gamma camera (Prism 1000 XP by Picker), followed by sequential static images every 5 minutes for up to 1 hour or until the identification of the sentinel node. To obtain a better anatomic image, a transmission image with a 57Coflood was used. Once the pattern of lymphatic distribution was assessed and the regional basin of lymphatic drainage identified, the skin corresponding to scintigraphic hot spots was marked. In the operating room the day after the lymphoscintigraphy, 0.5 to 1 mL of Patent Blue-V (Herst Eller Guebert, Brussels, Belgium) or methylene blue dye was injected immediately before surgery through a 25-gauge needle inserted intradermally on either side of the excisional scar in order to detect the sentinel lymph node with greater accuracy. The dye was pushed into the lymphatics by gentle massage. A sodium iodide crystal scintillator gamma probe (C-Trak Automatic System, Care Wise Medical Products, Morgan Hill, CA) or a cesium iodide crystal scintillator gamma probe (Navigator, Gamma Guidance System, Norwalk, CA) was used to detect sentinel nodes, using the scintigraphic images and skin marks as a guide. The surgeon was helped by the blue dye to identify any sentinel nodes and the node with the highest count rate was considered for resection. Also considered were the other lymph nodes with count rates ranging from 100% to 20% of the hottest node; all these were considered sentinel nodes. The surgical identification of sentinel nodes started within 20 minutes of the blue dye injection. In 27 patients, more than one sentinel node basin was present. The key feature of any sentinel node is that it receives direct lymphatic drainage from the primary tumor site. The preoperative lymphoscintigrams revealed that the cutaneous lymphatic drainage pathways were variable, and several pathways that drained to sentinel nodes in unexpected sites were observed, as already described.4,5 There were a few patients in whom it was not possible to identify the sentinel node(s) with certainty; these patients received a random biopsy of a node in the same basin and were excluded from the analysis. All patients with positive sentinel nodes were submitted to complete lymph node dissection of the involved lymph node basin.
Pathologic Examination Nonsentinel node samples. If the sentinel node contained metastatic melanoma based on H&E or immunohistochemistry results, the patient subsequently underwent a standard complete lymph node dissection of the involved nodal basin. The nonsentinel nodes were isolated from fresh node-containing fat. The number of nodes, size of the largest node, and gross appearance was recorded. All nodes were routinely fixed and paraffin embedded. Sections were stained with H&E and, in cases of doubt, immunohistochemistry was also performed. Positive sentinel node histopathologic analysis. For the purposes of this study, all sentinel node specimens involved by melanoma were re-evaluated by a panel of pathologists from the National Cancer Institute and San Pio X Hospital. Metastatic deposit location and morphology. The metastatic deposit within each sentinel node was classified as subcapsular, combined subcapsular and parenchymal, parenchymal, multifocal, or extensive, as previously described by Dewar et al.8 The number of deposits, their sizes, and the presence of extracapsular invasion were also recorded. When more than one sentinel node was found, these evaluations were done in the node colored by blue dye and in the node with the highest count.
Statistical Analysis
Table 1 shows patient characteristics; the sentinel node positivity rate was 15.9% (176 of 1,108 patients). As listed in Table 2, the risk of occult regional node metastases increased with Breslow’s thickness, Clark’s levels, and ulceration. The last two factors lost any statistically significant value once adjusted by maximum tumor thickness (Table 3).
Table 4 lists the histologic characteristics of the metastatic deposits. In most patients (84.7%), only one node was the site of metastases. In addition, none of the characteristics of the metastatic deposits evaluated in the node as previously described was significantly correlated with the risk of metastasis to more than one sentinel node.
Thirty-three (18.7%) of the 176 patients with a positive sentinel node presented additional secondary deposits in nonsentinel nodes at complete lymph node dissection. Table 5 lists that the largest diameter of metastatic foci and the localization of the metastatic foci within the lymph node were associated with the risk of a positive nonsentinel node at a statistically significant level (P = .022 and P = .009, respectively). Seven of the 33 nonsentinel positive nodes were associated with metastatic foci less than 1 mm in maximum diameter and two of 33 were associated with subcapsular localization.
Since the median exposure time was 61.1 months, we evaluated survival. Table 6 lists that sex, age, site of drainage, Clark’s levels, Breslow’s thickness, ulceration, and the status of sentinel nodes had a statistically significant impact on survival.
Table 7 lists the projected 5-year survival rates of patients with positive sentinel nodes; it is important to note that the only factor related with survival was the number of positive sentinel nodes involved (P = .056). The overall survival rate at 5 years was 81.4% for patients with one positive node, which dropped to 39.6% for patients with two positive nodes. The major difference and the borderline value of P is most likely due to the small number of patients with two positive sentinel nodes (n=24), since the time of follow-up was not different between the two groups.
Multivariate analysis (Table 8) was performed on the 643 patients for whom ulceration was defined and for whom the status of sentinel nodes as a key factor was indicated in the assessment of the prognosis of patients with primary melanoma, even if Breslow’s thickness and age maintained a statistically significant relevance. Since ulceration was not related to survival (P = .10), the same analysis was carried out on the total series (Table 9); in this larger group, sex became significantly related with survival.
To evaluate the relevance of the combined information on the status of sentinel and nonsentinel nodes, we identified and studied four groups of patients: (A) patients with negative sentinel nodes who never developed regional node metastases during follow-up; (B) patients with positive sentinel nodes and negative nonsentinel nodes at complete lymph node dissection; (C) patients with positive sentinel nodes and secondary deposits in nonsentinel nodes at complete lymph node dissection; (D) patients with negative sentinel nodes who developed clinically detectable node metastases at the same regional basin during follow-up. The last group included 47 patients (5.0% of 932 sentinel node-negative patients) who developed clinically detectable node metastases in the basin where the sentinel node was found to be negative with a mean delay of 19.5 months (range, 2.5 to 46.1 months). The 5-year overall survival was 92.6% for group A and 82.0% for group B (P = .004), and 47.1% for the goup C compared with 60.1% of group D (P = .134). Survival rates of the four groups is shown in Fig 1.
Table 10 shows the outcome of multivariate analysis when these four groups of patients, instead of the status of nodes, were considered. Patients who never developed regional node recurrence after a negative sentinel node biopsy and patients with positive sentinel nodes and all negative nonsentinel nodes at the time of complete node dissection had a highly significant better rate of survival than patients with positive sentinel and nonsentinel nodes. The observed differences in survival rates between patients who developed regional node metastases following a negative sentinel node and patients who were found to have both positive sentinel and positive nonsentinel nodes at complete lymph node dissection did not reach statistical significance. Table 10 also lists that Breslow’s thickness greater than 2 mm loses the significant impact on survival.
Data of this analysis confirm that the procedure of lymphatic mapping and sentinel lymphadenectomy is the most reliable method of nodal staging available to identify stage IB and II melanoma patients without occult metastatic deposits in the regional nodes who do not need any additional surgical treatment. In our series, 84.9% of 1,108 patients were found to have a negative sentinel node and did not receive additional surgery. The risk that these patients could develop a palpable node was low and comparable with that described in the literature.12,13 Two observations may support the hypothesis that recurrence in the same basin is due to the loss of single cell metastases. First, the frequency of positive nonsentinel nodes at completion of lymph node dissection in patients with a previous positive sentinel node with single cells (one of 29; Table 5); second, survival rates of patients who developed a recurrence (Fig 1). The only factor indicating the risk of occult nodal metastases was Breslow’s thickness (Table 3), as reported in previous studies,14-17 thus confirming the accuracy and predictability of tumor thickness as the most dominant prognostic factor.18-21 After the article by Balch et al22 was published, which indicated the relevance of Clark’s levels IV and V to assess the prognosis of melanoma patients with a primary tumor thickness of less than 1 mm, we decided to make this group of patients candidates for sentinel node biopsy also . All patients with a positive sentinel node underwent a complete lymph node dissection of the regional basin. The main question open at present is: is there a real reason to indicate a complete node dissection in patients with stage IB and II melanoma and positive sentinel nodes? The reason for asking this question is that all published studies, including the object of this article,18-21,23-26 showed that only approximately 20% of patients with positive nodes also had metastases in nonsentinel nodes. It is said that by continuing to perform complete lymph node dissections, we perform surgical procedures that do not give benefit to patients in 80% of cases. In our opinion, this is not really true because by using the sentinel node dissection procedure to avoid node dissection in sentinel node–negative patients, we avoided approximately 80% of the useless procedures (84.5% in this series). The risk of observing a recurrence in the basin where the sentinel node was found to be negative was 5%, which is acceptable for all of us, because a surgical procedure can be avoided in 932 patients who have a 95% chance of not benefiting from it. Because 18.7% of the patients with positive sentinel nodes in our series had nonsentinel node metastases, by avoiding complete lymph node dissection in patients with a positive sentinel node we increased the risk of recurrence from 5% to 23.7%, because occult metastases to nonsentinel nodes will most likely become clinically evident also. The second question is: do we have a chance to evaluate a higher risk of having occult metastases in nonsentinel nodes? To answer this question, some authors8,18-21,23-26 have classified the microanatomic location and morphologic characteristics of secondary deposits and have correlated these factors to the presence of additional metastases in nonsentinel nodes. In our series, we evaluated several characteristics of microdeposits in sentinel nodes that would potentially be able to assist us in evaluating the risk of metastases in nonsentinel nodes; the largest diameter of microscopic deposits and the localization of metastatic deposits were associated with the risk of metastases to nonsentinel nodes at a statistically significant level (P = .02 and P = .009, respectively). However, contrary to the results obtained by Dewar et al,8 in our series we could not relate the subcapsular location with the absence of metastases in nonsentinel nodes. Of the 33 patients with nonsentinel node metastases, 26 patients (78.8%) were associated with sentinel node metastases with a diameter of 1 mm or greater and 20 patients (76.9%) were associated with a localization of sentinel node metastases at an intraparenchymal site. As may be observed from Table 5, if we take these data into consideration, we can reduce complete node dissection in approximately 50% of sentinel node–positive patients. In terms of numbers, it means approximately 90 fewer surgical procedures. Is it reasonable that by reducing a relatively small number of surgical procedures we increase the risk of recurrences at the site of sentinel node biopsy from 5% to 8.9% in these 90 patients in whom we were unable to assess the status of nonsentinel nodes? In addition, we have to take into consideration the real meaning of sentinel node dissection. Waiting for the final results of the MSLT 1 study that might indicate a benefit in terms of survival, sentinel node dissection allows only the best assessment of prognosis in stage IB and II melanoma patients. Results of this study clearly indicate that the prognosis of these patients may be defined by a two-step procedure. The first step is the histologic examination of sentinel nodes, which gives the first differentiation between a group of patients with good prognosis constituted by sentinel node–negative patients and a group of patients with an intermediate prognosis constituted by patients with positive nodes. A second level of knowledge is mandatory in sentinel node–positive patients: the status of nonsentinel nodes. This second step makes it possible to differentiate a group of patients with a good prognosis constituted by 143 individuals with positive sentinel nodes and negative nonsentinel nodes (group B) from a relatively small group of patients with positive sentinel and nonsentinel nodes with a poor prognosis (group C). This second step allows us to identify patients at different risk levels of death to be considered for trials aimed at evaluating the efficacy of adjuvant treatments. In conclusion, it seems that at present, outside of a clinical trial that may show equal results comparing the wait-and-see policy with immediate complete dissection of regional nodes, a complete node dissection is necessary in sentinel node–positive patients to achieve the best assessment of prognosis of stage IB and II melanoma and to identify those patients who, having only positive sentinel nodes and negative nonsentinel nodes, have a good prognosis. While waiting for the results of the ongoing trial (MSLT 2), more studies are necessary for a better evaluation of the microstaging of secondary deposits in sentinel nodes, aimed at reducing the risk of underestimating occult secondary deposits in nonsentinel nodes.
The authors indicated no potential conflicts of interest.
Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.
1. Medalie N, Ackerman AB: Sentinel node biopsy has no benefit for patients whose primary cutaneous melanoma has metastasized to a lymph node and therefore should be abandoned now. Br J Dermatol 151:298-307, 2004[CrossRef][Medline] 2. Morton DL, Wen DR, Wong JH, et al: Technical details of intraoperative lymphatic mapping for early stage melanoma. Arch Surg 127:392-399, 1992 3. 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 230:453-465, 1999[CrossRef][Medline] 4. Thompson JF, Uren RF, Shaw HM, et al: The location of the sentinel lymph node in cutaneous melanoma: New insights into lymphatic anatomy. J Am Coll Surg 189:195-206, 1999[CrossRef][Medline] 5. Cochran AJ, Baildar BR, Starz H, et al: The Augsburg Consensus: Techniques of lymphatic mapping, sentinel lymphadenectomy, and completion lymphadenectomy in cutaneous malignancies. Cancer 89:236-241, 2000[CrossRef][Medline] 6. Karimipour DJ, Lowe L, Su L, et al: Standard immunostains for melanoma in sentinel lymph node specimens: Which ones are most useful? J Am Acad Dermatol 50:759-764, 2004[CrossRef][Medline] 7. Starz H: Pathology of the sentinel lymph node in melanoma. Semin Oncol 31:357-362, 2004[CrossRef][Medline] 8. Dewar DJ, Newell B, Green MA, et al: The microanatomic location of metastatic melanoma in sentinel lymph nodes predicts nonsentinel lymph node involvement. J Clin Oncol 22:3345-3349, 2004 9. Kaplan EL, Meier P: Nonparametric estimation from incomplete observations. J Am Stat Assoc 53:457-481, 1958[CrossRef] 10. Mantel H, Haenszel W: Statistical aspects of the analysis of data from retrospective studies of disease. J Natl Cancer Inst 22:719-748, 1959[Medline] 11. Cox DR: Regression models and life tables (with discussion). J Royal Stat Soc 34:187-220, 1972 (Series B) 12. McCarthy WH, Shaw HM, Thompson JF, et al: Time and frequency of recurrence of cutaneous stage I malignant melanoma: Guidelines for follow-up. Surg Gynecol Obstet 166:497-502, 1988[Medline] 13. Clary BM, Brady MS, Lewis JL, 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] 14. Soong SJ, Shaw HM, Balch CM, et al: Predicting survival and recurrence in localized melanoma: A multivariate approach. World J Surg 16:191-195, 1992[CrossRef][Medline] 15. Cascinelli N, Marubini E, Morabito A, et al: Prognostic factors for stage I melanoma of the skin: A review. Stat Med 4:265-278, 1985[Medline] 16. Averbook BJ, Russo LJ, Mansour EG: A long-term analysis of 620 patients with malignant melanoma at a major referral center. Surgery 124:746-756, 1998[CrossRef][Medline] 17. Buzaid AC, Ross MI, Balch CM, et al: Critical analysis of the current American Joint Committee on Cancer Staging System for cutaneous melanoma and proposal of a new staging system. J Clin Oncol 15:1039-1051, 1997 18. Reeves ME, Delgado R, Busam KJ, et al: Prediction of nonsentinel lymph node status in melanoma. Ann Surg Oncol 10:27-31, 2003[CrossRef][Medline] 19. Cochran AJ, Wen DR, Huang RR, et al: Prediction of metastatic melanoma in nonsentinel nodes and clinical outcomes based on the primary melanoma and the sentinel node. Mod Pathol 17:747-755, 2004[CrossRef][Medline] 20. Starz H, Siedlecki K, Balda BR: Sentinel lymphonodenectomy and S-classification: A successful strategy for better prediction and improvement of outcome in melanoma. Ann Surg Oncol 11:162S-168S, 2004 (suppl)[CrossRef][Medline] 21. Scolyer RA, Li LL, McCarthy SW, et al: Micromorphometric features of positive sentinel lymph nodes predict involvement of nonsentinel nodes in patients with melanoma. Am J Clin Pathol 122:532-539, 2004[CrossRef][Medline] 22. Balch CM, Soong SJ, 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 23. Starz H, Balda BR, Kramer KU, et al: A micromorphometry based concept for routine classification of sentinel lymph node metastases and its clinical relevance for patients with melanoma. Cancer 91:2110-2121, 2001[CrossRef][Medline] 24. Wagner JD, Gordon MS, Chuang TY, et al: Predicting sentinel and residual lymph node basin disease after sentinel lymph node biopsy for melanoma. Cancer 89:453-462, 2000[CrossRef][Medline] 25. McMaster KM, Wong SL, Edwards MJ, et al: Frequency of nonsentinel lymph node metastases in melanoma. Ann Surg Oncol 9:137-141, 2002[CrossRef][Medline] 26. Fink AM, Weihsengruber F, Spangl B, et al: S-classification of sentinel lymph node biopsy predicts the results of complete regional lymph node dissection. Melanoma Res 15:267-271, 2005[CrossRef][Medline] Submitted March 7, 2006; accepted July 24, 2006.
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2006 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
2 test and by the Fisher’s exact test when the sample size was small. The multiple logistic regression model was used to estimate the importance of the major prognostic factors such as Clark’s level, Breslow’s thickness, and ulceration for the presence of metastatic deposit in the sentinel node. Patient deaths regardless of cause (melanoma related and others) were considered as events in overall survival. The date of sentinel node biopsy was time zero for survival analysis. The cumulative overall survival rates were calculated by the Kaplan-Meier method.
