SEARCH

SEARCH BY CITATION

Keywords:

  • clinicopathological study;
  • Japanese patients;
  • malignant melanoma;
  • metastasis;
  • sentinel lymph node

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Case Report
  5. Discussion
  6. Acknowledgment
  7. References

Sentinel node biopsy (SNB) for malignant melanoma began to be performed in our department in 1997. A total of 121 patients underwent SNB. Sentinel node (SN) were identified in 39 (83.0%) of the 47 cases in which the blue dye method alone was used, and in 71 (95.9%) of the 74 cases by combination of dye, radioisotope, and γ probe methods. The excised SN was cut through its longest meridian, and the cut surfaces were stained with hematoxylin and eosin (HE) and examined for the presence (SN+) or absence (SN) of melanoma cells. If no melanoma cells were detected, serial sections were prepared and examined by sequential staining with HE, for S-100 protein, with HE, for HMB-45, with HE, and for Melan-A. In this study minute metastasis (SNm+) was defined as tumor cells newly identified in the immunostained section. The thickness of tumors ranged 0–38.0 mm, and their mean thickness was 4.5 mm. There were 39 SN+ cases (39/110 [39 + 71]; 35.5%) and seven SNm+ cases (7/110 [39 + 71]; 6.4%). The 5-year survival rate was significantly higher in the SN group than in the SN+ group and SNm+ group combined (P = 0.0002). The survival rate of the SNm+ group was not significantly different from that of the SN+ group (P = 0.3848). The 5-year survival rate of the SN+ group and SNm+ group combined was significantly higher than that of the group with clinically unequivocal metastatic nodes (P = 0.0001). Accurate SNB results will provide important prognostic information for Japanese patients with melanoma.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Case Report
  5. Discussion
  6. Acknowledgment
  7. References

Although cutaneous malignant melanoma is relatively rare in Japan, Japanese patients often have thicker primary lesions than white patients, and the relative frequency of each subtype differs considerably between Japanese and white subjects, probably reflecting anatomical differences. Superficial spreading melanoma is the most common type of melanoma in whites, whereas acral lentiginous melanoma is the most common type in the Japanese population.

There have been many reports on the clinical usefulness of sentinel node biopsy (SNB) in regard to the prognosis of malignant melanoma since it was first proposed in 1992.1 Sentinel nodes (SN) are the lymph nodes that directly receive lymph flow from the primary lesion, and the concept that the initial lymph node metastasis occurs in the SN is called the “SN concept”. The SN concept has been established as valid in regard to malignant melanoma, and decisions regarding the extent of surgical resection have been based on whether an SN metastasis is present. When the SN metastases are negative, resection of the primary lesion is deemed sufficient, and preventive lymph node dissection may be unnecessary.1 Because regional lymph node dissection can be omitted when SN are negative for lymph node metastasis, SNB has been widely performed in malignant melanoma to reduce the extent of resection and consequent surgical stress whenever possible. Effective utilization of SNB requires accurate identification and histopathological diagnosis of the SN, and dye and radioisotope methods of SN mapping can be used. The dye method alone has limited capacity to accurately identify the SN, and a combination of the dye and radioisotope methods is recommended.2 In this study, we used a combination of the dye and radioisotope methods since 2000. The SN was cut into two equal halves through the longest meridian, and serial sections were removed from one cut face of the node, according to the method described by Cochran et al.,3,4 and they were stained in sequence with hematoxylin–eosin (HE) for S-100 protein, with HE for HMB-45, and with HE and for Melan-A. The clinicopathological significance of SNB was also assessed. We have already reported finding that this staining method was useful for diagnosis of lymph node metastasis in our small group of melanoma patients.4

Case Report

  1. Top of page
  2. Abstract
  3. Introduction
  4. Case Report
  5. Discussion
  6. Acknowledgment
  7. References

A total of 121 malignant melanoma patients underwent SNB. SN were not identified in 11 of 121 cases. SN (202) were excised from 110 malignant melanoma patients (56 men [50.9%] and 54 women [49.1%], aged 8–84 years [mean 57.4 years]) who underwent SNB at the National Cancer Center Hospital (Tokyo, Japan) between 1997 and 2005. None of the patients were found to have clinical or radiographic evidence of lymph node metastasis during the preoperative examinations. The clinical characteristics and characteristics of the primary melanomas of 110 patients are shown in Table 1. The numbers of patients according to tumor thickness were: less than 1 mm, 24 (21.8%); 1–2 mm, 21 (19.1%); 2–4 mm, 23 (20.9%); more than 4 mm, 39 (35.5%); and unknown, three (2.7%). The thickness of tumors ranged 0–38.0 mm, and their mean thickness was 4.49 mm. Ulceration was present in 27 patients and absent in 83 patients.

Table 1. Patients’ (n = 110) characteristics and distribution of clinicopathological features (1997–2005)
 No. of patients (%)
  1. SLN, sentinel lymph node.

Sex
 Female54 (49.1)
 Male56 (50.9)
Age (year)
 Mean57.4
 Minimum–maximum8–84
Pathological type
 Nodular melanoma8 (7.3)
 Lenitigo maligna melanoma2 (1.8)
 Superficial spreading melanoma41 (37.3)
 Acral lentiginous melanoma59 (53.6)
Primary site
 Head and neck10 (9.1)
 Lower extremity20 (18.2)
 Sole32 (29.1)
 Toenail14 (12.7)
 Trunk10 (9.1)
 Upper extremity24 (21.8)
Ulceration
 Absent83 (75.5)
 Present27 (24.5)
Primary melanoma Breslow thickness (mm)
 Range (mm)0–38.0
 Mean (mm)4.49
 <1 mm24 (21.8)
 1–2 mm21 (19.1)
 2–4 mm23 (20.9)
 >4 mm39 (35.5)
 Unknown3 (2.7)
Site of SLN (n)
 Head and neck10 (9.1)
 Axilla29 (26.4)
 Groin70 (63.6)
 Axilla + groin1 (0.9)
 No. of SLN per case (mean)1.8

SNB

Sentinel node biopsy was performed after identification by the dye method alone in 47 (42.7%) of the 121 patients and by a combination of the dye and radioisotope methods in the other 74 patients (67.3%).

Dye method

Sentinel node biopsy was performed according to the method described by Morton et al.1 Patent blue V (1%) was injected into the skin at approximately 10 sites around the primary malignant melanoma lesion or the surgical scar after resection of the primary lesion, and 15 min later the skin was incised, and the lymph vessels that had stained blue in the regional lymph node area were traced. Lymph nodes that had stained blue were identified as SN and biopsied.

Dye and radioisotope method

In accordance with the method described by Mcmasters,5 approximately 24 h before the biopsy 74 MBeq of 99mTc-tin colloid was injected into the skin at four sites around the primary lesion or the surgical scar that remained after resection of the primary lesion. Lymphoscintigraphy was performed immediately after the injections and again 3 h after the injections, and the position of the SN was identified by using a γ-probe (Navigator GPS [RMD Instruments, MA, USA] or Neoprobe 2000 [Neoprobe, Dublin, OH, USA]). The dye was injected at 10 sites around the primary lesion or the surgical scar 15 min before the biopsy. The skin was incised over the site where the SN had been identified by lymphoscintigraphy by using the γ-probe as a guide, and the lymph nodes that had stained blue were excised. The excised lymph nodes were irradiated with the γ-probe again, and the scintillation count was measured. When more than one node was stained, the nodes accounting for 10% of the maximum count measured with the γ-probe were considered to be the SN.

Histopathological examination

We bisected the node though its longest meridian and evaluated one cut surface for tumor cells after routine HE staining. 10 serial 4-μm sections were also prepared from one cut face of the node by the method described by Cochran et al.,3 and stained in sequence with HE for S-100 protein, with HE for HMB-45, and with HE for Melan-A6–13 (Table 2). Immunostaining for S-100 protein, HMB-45 and Melan-A was performed by the avidin-biotin-peroxidase complex (ABC) method. After deparaffinization in xylene and rehydration in ethanol, the sections were heated in citrate buffer (10 nmol/L, pH 6.0) at 120°C for 10 min for antigen retrieval. Endogenous peroxidase was blocked with 1% hydrogen peroxidase in methanol for 20 min. An SN whose maximal cut surfaces were stained with HE and was positive for metastasis was recorded as SN+. In the cases in which no melanoma cells were detected, if tumor cells were identified in all of the immunostained sections after serial sections were prepared and HE staining and immunostaining were performed, the node was recorded as minute metastasis (SNm+). An SN in which no metastasis was detected even after HE staining and immunostaining was recorded as SN.

Table 2. Antibodies for immunohistochemical staining
MarkerCloneSourceDilutionStaining pattern
S-100 proteinRabbit policlonalDako1:2000Nuclear + cytoplasmic staining
Melanoma antihumanHMB-45Dako1:10Cytoplasmic staining
Melan-AA103Dako1:80Cytoplasmic staining

Statistical analysis

Overall survival (OS) time was defined as the period between the day of surgery and death, and melanoma-specific survival (MSS) time was defined as the period between the day of surgery and death from malignant melanoma. The Kaplan–Meier method was used to compare the distribution of OS time and MSS time, and the log–rank test was used to compare the curves.

Five variables, namely, thickness of the primary lesion, presence or absence of ulceration, SN metastasis, age and sex, were subjected to multivariate analysis using a Cox proportional hazards model.

Identification of SN

An SN was identified in 39 (83.0%) of 47 cases in which the dye method alone was used to identify SN, and no SN were identified in the other eight cases (17.0%). The regional lymph node areas in these eight cases were in the neck (one case), the axillary region (three cases) and the inguinal region (four cases).

Sentinel nodes were identified in 71 of the 74 cases (95.9%) in which both the dye method and the radioisotope method were performed together. The overall SN identification rate in both groups was 90.9% (110/121). One SN was found in 55 cases (50.0%), two in 32 cases (29.1%), three in 13 cases (11.8%), four in eight cases (7.3%), five in one case (0.9%) and six in one case (0.9%). The mean number of SN per patient was 1.8.

Histopathological examination

Of the 110 cases in which SN were identified, HE staining revealed that 39 were SN+ (35.5%) and seven were SNm+ (6.4%) (Table 3). All cases were S-100 protein, HMB-45 and Melan-A positive. In seven of the SNm+ cases, melanoma cells were observed on re-cut HE-stained sections (Fig. 1).

Table 3. Seven minute metastasis cases
No.AgeSexBreslow (mm)UlcerPrimary sitePathological typeLN lesionSurvival (months)RecurrenceDead or alive
  1. ALM, acral lentiginous melanoma; NM, nodular melanoma; SSM, superficial spreading melanoma.

159F9.5SoleALMInguinal49+Dead
284M9Lower extremityNMInguinal25+Dead
367M1.4SoleALMInguinal91Alive
465M23TrunkNMAxillary83+Alive
544F5.5Lower extremitySSMInguinal23+Dead
649F1.1Upper extremitySSMAxillary65Alive
776M2.8+Upper extremityALMAxillary16+Dead
image

Figure 1.  Minute metastasis in a sentinel node. The node was bisected through its longest meridian and one cut surface evaluated for tumor cells after routine hematoxylin–eosin (HE) staining. Ten serial 4-μm sections were also prepared from one cut face of the node by the method described by Cochran et al. and stained in sequence with HE for S-100 protein, with HE for HMB-45, and with HE and for Melan-A. Minute metastasis was present in all immunostaining sections.

Download figure to PowerPoint

Survival

In the 46 SN+ and SNm+ cases, the 5-year OS rate was 59.9%, and the median survival time (MST) was 53 months, and in the SN group the 5-year OS rate was 92.1% and the MST was 71 months (Table 4). The survival rate was significantly higher in the 64 SN cases than in the 46 SN+ and SNm+ cases (log–rank test, = 0.0002) (Fig. 2).

Table 4. OS and MSS of the sentinel nodes negative for metastasis (SN group), and sentinel nodes positive for metastasis and positive for minute metastasis (SN+ and SNm+ groups) combined
 CasesOS (%)MST (months)P (log-rank test)MSS (%)P (log-rank test)
  1. OS, overall survival; MST, median survival time; MSS, melanoma-specific survival.

SN6492.1710.000296.2<0.0001
SN+ + SNm+4659.95362.7
image

Figure 2.  Overall survival of the sentinel nodes negative for metastasis (SN group), and sentinel nodes positive for metastasis and positive for minute metastasis (SN+ and SNm+ groups) combined.

Download figure to PowerPoint

Figure 3 shows that the survival rate was significantly higher in the 64 SN cases than the 39 SN+ cases (log–rank test, = 0.0007) and significantly higher in the 64 SN cases than in the seven SNm+ cases (log–rank test, P = 0.0002). The 5-year OS rate in the 39 SN+ cases was 63.1%, and the MST was 58 months. The 5-year OS rate in the SNm+ group was 42.9%, and the MST was 49 months (Table 5). There were no significant differences in OS rates between the SN+ group and the SNm+ group (Fig. 3).

image

Figure 3.  Overall survival of the sentinel nodes negative (SN group), and sentinel nodes positive for metastasis (SN+ group), and sentinel nodes positive for minute metastasis (SNm+ group).

Download figure to PowerPoint

Table 5. OS and MSS of the sentinel nodes positive for metastasis (SN+ group) and sentinel nodes positive for minute metastasis (SNm+ group)
 CasesOS (%)MST (months)P (log–rank test)MSS (%)P (log–rank test)
  1. OS, overall survival; MST, median survival time; MSS, melanoma-specific survival.

SN+3963.1580.384866.60.2602
SNm+742.94942.9

The patients in the therapeutic lymph node dissection (TLND) group had unequivocally enlarged lymph nodes clinically, and therefore underwent radical lymph node dissection without SNB. The background of the TLND group between 1997 and 2005 is shown in Table 6. TLND was performed in 33 malignant melanoma patients (18 men [54.5%] and 15 women [45.5%]). Mean age was 52.4 years (range 21–77 years). The numbers of patients according to tumor thickness were: less than 1 mm, one (3.0%); 1–2 mm, three (9.1%); 2–4 mm, seven (21.2%); more than 4 mm, 14 (42.4%); and unknown, eight (24.3%). The thickness of tumors ranged 0.65–25.0 mm, and their mean thickness was 4.3 mm.

Table 6. Patients’ characteristics and distribution of clinicopathological features in the TLND group (1997–2005)
 No. of patients (%)
  1. SLN, sentinel lymph nodes.

Sex
 Female15 (45.5)
 Male18 (54.5)
Age (year)
 Mean52.4
 Minimum–maximum21–77
Pathological type
 Nodular melanoma6 (18.2)
 Lenitigo maligna melanoma1 (3.0)
 Superficial spreading melanoma13 (39.4)
 Acral lentiginous melanoma4 (12.1)
 Unknown9 (27.3)
Primary site
 Head and neck6 (18.2)
 Lower extremity5 (15.2)
 Sole2 (6.1)
 Toenail2 (6.1)
 Trunk13 (39.4)
 Upper extremity0 (0)
 Unknown5 (15.2)
Primary melanoma Breslow thickness (mm)
 Range (mm)0.65–25
 Median (mm)4.3
 0 ≤ 1 mm1 (3.0)
 1–2 mm3 (9.1)
 2–4 mm7 (21.2)
 >4 mm14 (42.4)
 Unknown8 (24.3)
Site of SLN (n)
 Head and neck7 (21.2)
 Axilla9 (27.3)
 Groin17 (51.5)
 Popliteal fossa0 (0)

Patients in the SN+ group and SNm+ group combined underwent radical lymph node dissection immediately after confirming that an SN was metastasis-positive. OS and MST were compared between the SN+ and SNm+ groups combined and the TLND group. The 5-year OS rate and MST were 59.9% and 53 months, respectively, in the SN+ and SNm+ groups combined (n = 46), and 20.2% and 17 months, respectively, in the TLND group (n = 33) indicating that the prognosis for the TLND group was significantly poorer (log–rank test, = 0.0001) (Fig. 4) (Table 7).

image

Figure 4.  Comparison between overall survival of the sentinel nodes positive for metastasis (SN+ group) and sentinel nodes positive for minute metastasis (SNm+ group) combined and the therapeutic lymph node dissection (TLND) group.

Download figure to PowerPoint

Table 7. OS and MSS of the sentinel nodes positive for metastasis (SN+ group) and sentinel nodes positive for minute metastasis (SNm+ group) combined and the TLND group
 CasesOS (%)MST (months)P (log–rank test)MSS (%)P (log–rank test)
  1. OS, overall survival; MST, median survival time; MSS, melanoma-specific survival; TLND, therapeutic lymph node dissection.

SN+ and SNm+4659.9530.00162.70.001
TLND3320.21727.3

Cox multivariate analysis

The outcome of the malignant melanoma patients was not associated with sex or age (≤59 vs ≥60 years). The hazard ratio was 2.014 (95% confidence internal [CI] = 0.850–4.773, = 0.112) for tumor thickness (≤4.0 vs >4.0 mm), 3.431 (95% CI = 1.512–7.783, = 0.003) for the presence of ulceration, and 3.211 (95% CI = 1.306–7.900, = 0.011) for SN metastasis, indicating that ulceration and SN metastasis are risk factors for a poor outcome (Table 8).

Table 8. Cox multivariate analysis of the prognostic factors for patients who underwent sentinel node biopsy
  Hazard ratio95% CIP
  1. CI, confidence interval.

Tumor thickness (mm)≤4.0 vs 4.0<2.0140.850–4.7730.112
UlcerationNo vs Yes3.4311.512–7.7820.003
Sentinel lymph node status (positive vs negative)No vs Yes3.2111.306–7.9000.011
Age (years)≤59 vs 60≤1.6890.767–3.7220.193
SexFemale vs male0.8350.372–1.8770.662

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Case Report
  5. Discussion
  6. Acknowledgment
  7. References

Of the 110 cases identified in which SN were excised, seven (6.4%) were judged to be SNm+. In all seven cases, tumor cells were detected in deeper serial sections that had been immunochemically stained. Blaire et al.14 reported detecting microscopic metastases by preparing serial sections and performing immunohistochemical staining for HMB-45 in four of 66 patients (6%) with node-negative melanoma. Yu et al.15 reported detecting microscopic metastases in 11 (11.7%) of 94 cases by examining deeper serial sections and performing immunohistochemical staining for S-100 protein, HMB-45, NK1C3 and MART-1. Cochran showed that histopathological examination of the cut surfaces of SN exposed by cutting them through their longest meridian alone is insufficient, and that careful examination of immunostained serial sections is necessary to avoid overlooking very small metastases.16 Accurate diagnosis of SN metastasis leads to accurate staging of malignant melanomas. Since the 2002 revision of the UICC-TNM classification17 in which a tumor is classified as stage III when lymph node metastasis is present, detection of minute metastasis by immunostaining results in an increase in stage from stage II to stage III. When SN metastasis was not detected on HE staining, the prognosis was favorable when immunostaining was also negative, but significantly worse when immunostaining was positive, even if it was only slightly stained. In our study, recurrence and metastasis occurred in 13 of the 39 cases (33.3%) diagnosed as SN+, in five of the seven cases (71.4%) diagnosed as SNm+ and in six of the 64 cases (9.4%) diagnosed as SN. Ross et al.18 reported that immunohistochemistry upstaged the disease prognosis in 8% of patients (6/76). In previous reports, the rate of recurrence/metastasis in SN patients was 0–14% (mean 7.5%).5,19–23 In our study, recurrence/metastasis occurred in 71.4% of the SNm+ patients, suggesting that SNm+ cases are a high risk group. Ling-xi et al.24 reported that in five of their seven cases deposits of melanoma cells were detected only in re-cut sections, and they recommended routine examination of both HE-stained sections and sections immunostained for S-100 protein and HMB-45. The results of the Kaplan–Meier statistical analysis in our study indicated that both the OS time and the MST were significantly shorter in SN+ patients, including SNm+ patients, than in SN patients. Radical lymph node dissection should be performed in both SN+ patients and SNm+ patients the same as in lymph node-positive stage III patients, even though extensive lymph node metastases have not been demonstrated.

Furthermore, the OS rate of 20.2% in the 33 TLND cases was significantly lower than the 59.9% in the 46 SN+ and SNm+ cases, suggesting that early diagnosis of lymph node metastasis by SNB and radical lymph node dissection even though no lymph nodes are palpable, is a reliable surgical strategy. It is especially important to make an accurate diagnosis in the SNm+ cases and treat them as stage III cases. Based on the above findings in our retrospective study of Japanese patients, SN metastasis is a more reliable prognostic factor in malignant melanoma than tumor thickness.

The above findings indicate that accurate assessment of melanoma patients for SN metastasis and minute metastasis not only reduces surgical stress and avoids unnecessary preventive lymph node dissection, which are the original aims of SNB, but also leads to the performance of surgery at the most appropriate time in cases requiring radical lymph node dissection.

Acknowledgment

  1. Top of page
  2. Abstract
  3. Introduction
  4. Case Report
  5. Discussion
  6. Acknowledgment
  7. References

This work was supported in part by a Grant-in-Aid for Cancer Research (21-7-6) from the Japanese Ministry of Health, Labor and Welfare.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Case Report
  5. Discussion
  6. Acknowledgment
  7. References
  • 1
    Morton DL, Wen DR, Wong JH et al. Technical details of intraoperative lymphatic mapping for early stage melanoma. Arch Surg 1992; 127: 392399.
  • 2
    Krag DN, Weaver DL, Alex JC, Fairbank JT. Surgical resection and radiolocalization of sentinel lymph node in breast cancer using a gamma probe. Surg Oncol 1993; 2: 335339.
  • 3
    Cochran AJ, Roberts AA, Saida T. The place of lymphatic mapping and sentinel node biopsy in oncology. Int J Clin Oncol 2003; 8: 139150.
  • 4
    Noro S, Yamamoto A, Yamazaki N et al. Sentinel lymph node biopsy and pathological examination for malignant melanoma in the dermatology division of National Caner Center Hospital. Jpn J Dermatol 2004; 114: 1524 (In Japanese).
  • 5
    McMasters KM, Reintgen DS, Ross MI et al. Sentinel Lymph node biopsy for melanoma: how many radioactive nodes should be removed? Ann Surg Oncol 2001; 8: 192197.
  • 6
    Dorvault CC, Weilbaecher KN, Yee H et al. Microphthalmia transcription factor. Cancer 2001; 93: 337343.
  • 7
    King R, Weilbaecher KN, McGill G, Cooley E, Mihm M, Fisher DE. Microphthalmia transcription factor. A sensitive and specific melanocyte marker for melanoma diagnosis. Am J Pathol 1999; 155: 731738.
  • 8
    King R, Googe PB, Weilbaecher KN, Mihm MC Jr, Fisher DE. Microphthalmia transcription factor expression in cutaneous benign, malignant melanocytic,and nonmelanocytic tumors. Am J Surg Pathol 2001; 25: 5157.
  • 9
    Granter SR, Weilbaecher KN, Quigley C, Fletcher CD, Fisher DE. Microphthalmia transcription factor: not a sensitive or specific marker for the diagnosis of desmoplastic melanoma and spindle cell (non-desmoplastic) melanoma. Am J Dermatopathol 2001; 23: 185189.
  • 10
    Chang KL, Folpe AL. Diagnostic utility of microphthalmia transcription factor in malignant melanoma and other tumors. Adv Anat Pathol 2001; 8: 273275.
  • 11
    Miettinen M, Fernandez M, Franssila K, Gatalica Z, Lasota Z, Sarlomo-Rikala M. Microphthalmia transcription factor in the immunohistochemical diagnosis of metastatic melanoma: comparison with four other melanoma markers. Am J Surg Pathol 2001; 25: 205211.
  • 12
    Busam KJ, Jungbluth AA. Melan-A, a new melanocytic differentiation marker. Adv Anat Pathol 1999; 6: 1218.
  • 13
    Busam KJ, Chen YT, Old LJ et al. Expression of Melan-A (MART1) in benign melanocytic nevi and primary cutaneous malignant melanoma. Am J Surg Pathol 1998; 22: 976982.
  • 14
    Baisden BL, Askin FB, Lange JR, Westra WH. HMB-45 immunohistochemical staining of sentinel lymph nodes: a specific method for enhancing detection of micrometastases in patients with melanoma. Am J Surg Pathol 2000; 24(8): 11401146.
  • 15
    Yu LL, Flotte TJ, Tanabe KK et al. Detection of microscopic melanoma metastases in sentinel lymph nodes. Cancer 1999; 86: 617627.
  • 16
    Cochran AJ, Balda BR, Starz H et al. The Augsburg Consensus. Techniques of lymphatic mapping, sentinel lymphadenectomy and completion lymphadenectomy in cutaneous malignancies. Cancer 2000; 89: 236241.
  • 17
    Sobin LH, Wittekind CH. Skin tumors, malignant melanoma of skin. In: International Union Against Cancer (UICC): TNM Classification of Malignant Tumors, 6th edn. New York: Wiley-Liss, 2002; 126130.
  • 18
    Ross GL, Shoaib T, Scott J et al. The impact of immunohistochemistry on sentinel node biopsy for primary cutaneous malignant melanoma. Br J Plast Surg 2003; 56: 153155.
  • 19
    Lingam MK, Mackie RM, McKay AJ. Intraoperative identification of sentinel lymph node in patients with malignant melanoma. Br J Cancer 1997; 75: 15051508.
  • 20
    Gogel BM, Kuhn JA, Ferry KM et al. Sentinel lymph node biopsy for melanoma. Am J Surg 1998; 176: 544547.
  • 21
    Statius Muller MG, Borgstein PJ, Pijpers R et al. Reliability of the sentinel node procedure in melanoma patients: analysis of failures after long-term follow-up. Ann Surg Oncol 2000; 7: 461468.
  • 22
    Clary BM, Brady MS, Lewis JJ, Coit DG. 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 2001; 233: 250258.
  • 23
    Harlow SP, Krag DN, Ashikaga T et al. Gamma probe guided biopsy of the sentinel node in malignant melanoma: a multicentre study. Melanoma Res 2001; 11: 4555.
  • 24
    Li LX, Scolyer RA, Ka VS et al. Pathologic review of negative sentinel lymph nodes in melanoma patients with regional recurrence: a clinicopathologic study of 1152 patients undergoing sentinel lymph node biopsy. Am J Surg Pathol 2003; 27: 11921202.