Sentinel lymph node biopsy in cutaneous melanoma: outcome after 5-years follow-up

ARTICLE

Auteur(s) : Elsa Cuchet¹, Nicole Pinel², Denis Corcella³, Jean-Philippe Vuillez⁴, Jacques Lebeau³, François Moutet³, Marc Colonna⁵, Marie-Thérèse Leccia¹

¹Dermatological Clinic, Department of Medecine, CHU Albert Michallon, 38043 Grenoble cedex 09
²Department of histopathology, CHU Albert Michallon, 38043 Grenoble cedex 09
³Department of plastic and reconstructive Surgery, CHU Albert Michallon, 38043 Grenoble cedex 09
⁴Department of Nuclear Medecine, CHU Albert Michallon, 38043 Grenoble cedex 09
⁵Radiotherapy, Department of oncology, CHU Albert Michallon, 38043 Grenoble cedex 09

accepté le 27 Avril 2007

Materials and methods

The different stages of the technique were carried out according to the recommendations of the French Cutaneous Cancerology Group.

The Breslow thickness was superior or equal to 1.5 mm with a decrease of this thickness to 0.75 mm when other bad histo-prognostic factors were also present: ulceration, regression, vascular or lymphatic embols. Clinical examination, chest x-ray and abdominal/pelvic echography were performed. Exclusion criteria included mucosal involvement, presence of a palpable lymph node, large initial excision of the melanoma with reconstruction flap, Karnofsky performance status lower than 80, patient aged under 18 years and pregnancy. The clinical characteristics of the patients and anatomopathological features of the tumours are shown in tables 1 and 2.

Detection of the sentinel lymph node was performed after the initial excision of the melanoma prior to more extensive surgery. The technique was performed within the two months following the initial biopsy-excision and in most cases under general anaesthesia.

Preoperative cutaneous lymphoscintigraphy was performed with the radiocolloid available in our country: technitium (^99mTc) Rhenium sulfur, Nanocis^®. After tracer injection, a dynamic scan of the lymphatic outflow was obtained, followed by early and delayed static images of the lymphatic drainage basin acquired by a gamma-camera. The skin overlying the highest activity emission point was marked.

A peroperative intradermal injection of blue dye was sometimes administered, at the beginning of the intervention, at the site of the primary melanoma and was performed by one of the surgeons in order to facilitate detection.

Excision of the sentinel lymph node was performed and ex vivo radioactivity of the node was evaluated as well as in vivo residual radioactivity (counts per second) to allow the removal of residual marked lymph nodes.

The sentinel lymph node was then immediately transported to the pathology laboratory for histopathological examination. Fixation was done using Formalin Fixed Zinc. Two to 3 mm macroscopic sections were performed and included in paraffin. A few sections were carefully examined by HES (hematoxylin eosin saffron). Immunohistochemistry was then carried out on deparaffinated sections, at three different levels with a distance of 150 to 200 microns between each section extremity. Three slides at each section level were stained, giving a total of 9 slides (4 μm per section). The antibodies used were anti-PS100, anti-HMB45 and anti-Melan A antibodies.

Patients with a positive sentinel lymph node underwent elective complete lymph node dissection.

Follow-up consisted of a clinical evaluation every 3 months and imagery (chest X ray abdominal/pelvic echography) every six months.
Table 1 Clinical characteristics of the patients

Statistical analysis

The Khi2 independence test was used for qualitative variables.

Survival rates were evaluated using the non-parametric Kaplan-Meier method.

Comparison of the survival between the different groups was performed using the Log-rank test.

Results

During surgery the colorimetric method was associated in 22.5% of cases, with the use of a peroperative probe. In most cases (85.8%) the samples were taken under general anaesthesia. In 8 of the 62 patients, the number of sentinel lymph nodes removed through surgery was smaller than the number of sentinel nodes detected through lymphoscintigraphy, mostly due to technical reasons and constraints due to surgical access (paravertebral, deep sternal/clavicular and thoracic localizations).

The rate of complications associated with the excision of the sentinel lymph node was very low with early complications in only 8% of cases (including seroma in 4.8%, haematoma in 1.6% and for one patient the development of a blue colouring of the entire cutaneous surface in the absence of associated methaemoglobinaemia). Only one late complication involved an algoneurodystrophy of the upper limb.

Eleven patients (17.7% of patients) had one or several positive sentinel lymph nodes. The percentage of positive sentinel lymph nodes was respectively 9%, 27.3%, 45.5% and 18.8% for Breslow thickness ≤ 1 mm, between 1.01 to 2 mm, > 2.01 to 4 mm and > 4 mm (table 3). The diagnosis was established through immunohistochemistry in 54.7% of cases. The positivity of the sentinel lymph node was closely correlated with the Breslow thickness except for melanoma > 4 mm on account of the small number of patients.

Median follow-up was of 44 months and ranged from 11 to 67 months. Patients were examined every three months. Melanoma recurrence was observed in 17 patients, 6 of whom had a positive sentinel lymph node. Therefore, 54.5% (6/11) of patients with a positive sentinel lymph node relapsed and only 21.5% (11/51) of patients with a negative sentinel lymph node. Thirteen of the 62 patients presented with a local regional recurrence (isolated in-transit metastases, lymph node metastases alone, or the association of both). Among these 13 patients, 5 had a positive sentinel lymph node representing 45.4% of cases (5/11) and 8 had a negative sentinel lymph node representing 15.6% of cases (8/51).

There was a significant difference in terms of local regional recurrences between patients with a positive and patients with a negative sentinel lymph node (p = 0.017).

Eleven patients out of 62 (17.7%) presented with distant metastases.

Among these eleven patients, 5 of the 11 patients (45.4%) had a positive sentinel lymph node and 6 of the 51 patients (11.7%) had a negative sentinel lymph node.

Statistical analysis revealed a significant difference in terms of metastatic evolution between positive and negative sentinel lymph node patients (p = 0.054).

In the overall study population the recurrence-free survival rate at 36 and 60 months was of 77% and 70% respectively.

In patients with a negative sentinel lymph node the recurrence-free survival rate at 36 and 60 months was of 85% and 78% respectively. In patients with a positive sentinel lymph node this rate was of 45% and 36% respectively. A statistically significant difference therefore exists in terms of recurrence-free survival between the two populations depending on the presence or absence of a positive sentinel lymph node (p = 0.0046) (figure 1).

The rate of overall survival in the study population as a whole at 36 and 60 months was of 91% and 77% respectively.

In patients with a negative sentinel lymph node, the overall survival rate at 36 and 60 months was of 94% and 85% respectively. In patients with a positive sentinel lymph node this rate at 36 and 60 months was of 82% and 47%. A statistically significant difference in terms of overall survival therefore exists between these two populations depending on the presence or absence of a positive sentinel lymph node (p = 0.0019) (figure 2).
Table 3 Correlation between positive sentinel lymph node and Breslow thickness

Discussion

Our results on the preoperative detection of the sentinel lymph node are excellent. Indeed, at least one sentinel lymph node was detected in 100% of patients by lymphoscintigraphy with an average of 1.7 nodes per patients as opposed to an average of 2 to 2.5 nodes per patient in the literature. Lymphoscintigraphy has allowed the detection of atypical drainage in 24% of cases (10 to 28% in different studies) and has allowed the detection of 3.2% intermediate lymph nodes. Surgical removal was successful in 98% of cases (failures involved lymph nodes that were too deep). In our series, the total number of lymph nodes removed for each patient reduced with time as the surgeon gained in experience. The problems associated with the specificity of surgical excision are real, due to the lack of homogeneity in the criteria concerning the definition of a sentinel or non-sentinel lymph node, on account of the radioactivity emitted. This question has led to many discussions and McMasters’ team proposed removing all lymph nodes that emit 10% or more of the ex-vivo radioactivity detected in the most radioactive node [4]. In our study, we observed very few early (8%) or late (1.6%) complications associated with the removal of the sentinel lymph node, which is in agreement with the data of the literature [5-7].

The results in our series in terms of anatomopathological detection of metastases are identical to those observed in various other studies: 17.7% of sentinel lymph nodes contained metastases compared to an average of 20% in the literature using the same methods: standard histological techniques and immunohistochemistry using anti-PS 100, anti-HMB45 and anti-Melan-A antibodies [8-10]. The anatomopathological technique and the detection of micrometastases required training to differentiate between capsular naevi and micrometastases. It is important to emphasize the lack of homogeneity in the terms used in the literature to define the type of metastasis present in sentinel lymph nodes (metastases, micrometastases, isolated tumor cells) making the interpretation by the anatomic pathologist very difficult.

In our study we confirm that the use of immunohistochemistry techniques is essential for the detection of micrometastases, since 54.7% of the latter had not been diagnosed after HES staining alone. According to various studies, standard histological techniques underestimate the presence of positive sentinel lymph nodes in 10 to 36% of cases [11-13]. The search for metastases by tyrosinase RT-PCR is not carried out in our center as its use is currently controversial, as it has not been standardized and can lead to false positives [14-16]. Concerning other methods to detect occult lymph node metastasis, Alonso et al. compared ^99mcTc-MIBI scintigraphy to sentinel lymph node biopsy and found that this technique had insufficient sensitivity, with a false negative rate of 17% which is more than with the surgical procedure [17].

The role of the lymph node tumor burden in melanoma patients staged by sentinel lymph node biopsy is also under evaluation in order to optimise the prognostic role of the anatomopathological diagnosis. The study of Ranieri et al. showed that the tumor burden, evaluated from the largest tumor deposit, was a significant prognostic factor associated with recurrence-free and overall survival [18]. Carlson et al. also demonstrated that patients with a metastatic deposit > 2 mm in diameter in the sentinel lymph node had a significantly decreased survival rate [19].

It is currently admitted that a positive sentinel lymph node is correlated with the Breslow thickness. In our study, even if the differences observed were not statistically significant (most certainly due to the small sample size), the percentage of positive sentinel lymph nodes increased with the Breslow thickness (table 3).

The prognostic value of the detection of the sentinel lymph node has also been confirmed in our series. Indeed, the rates of local regional and distant recurrences were statistically more important in the group of patients with micrometastases: 54.5% of recurrences in patients with a positive sentinel lymph node as opposed to 21.5% for patients with a negative sentinel lymph node, even though elective lymph node dissection had been performed. Our results are concordant with the studies of Jansen et al., Statius Muller et al., Liszkay et al. [6, 20, 21]. In addition, recurrence-free and overall survival rates at 60 months were higher in patients with a negative sentinel lymph node. Indeed, recurrence-free survival at 36 and 60 months was 85% and 78% respectively for patients with negative sentinel lymph nodes, whereas the rates were 45% and 36% respectively for patients with positive sentinel lymph nodes (p = 0.0046). The overall survival rate was 94% at 36 months and 85% at 60 months for patients with negative sentinel lymph nodes, as opposed to 82% at 36 months and 47% at 60 months for patients with positive sentinel lymph nodes (p = 0.0019).

So, our experience confirms that the sentinel lymph node procedure in cutaneous melanoma is reliable and reproducible. It does, however, require a close collaboration between dermatologists, nuclear medicine physicians, plastic surgeons and anatomic pathologists. Surgical and anatomopathological techniques require training in order to obtain optimal results. Standardization is essential to maintain quality control. In our experience, postoperative complications were rare and minor. Since 2002, we have extended the application of this technique to melanomas with a Breslow thickness that was superior or equal to 1 mm. Our results concerning recurrence-free and overall survival also confirm the prognostic value of the sentinel lymph node status, which determines with precision the stage of the disease at diagnosis and can be used to set up homogeneous groups of patients for adjuvant medical treatment protocols.

But the enthusiasm generated by this technique must not make us forget the ultimate goal of this procedure, which is to improve patient survival. The therapeutic value of this method has still not been really proven. Morton et al. in their Multicenter Selective Lymphadenectomy Trial (MSLT) evaluated the clinical effect of immediate complete lymphadenectomy in patients with tumor-positive sentinel nodes compared to delayed lymphadenectomy carried out at the time of clinical recurrences. The analysis of their results did not reveal a significant difference in melanoma-specific survival between the two study groups and, particularly, no benefit for overall survival. They only showed a minimal disease free survival advantage for immediate complete lymphadenectomy for tumor-positive sentinel nodes [22]. Our results, despite the limited number of patients, also clearly showed that immediate complete lymphadenectomy for positive sentinel nodes did not really improve patient survival.

Moreover, the significance of isolated tumor cells in sentinel lymph nodes is not known at this time and could play a positive role in immunosurveillance [23]. So it is essential to define more precise histological criteria (such as tumor burden) and/or immune markers to determinate the good indications for elective lymph node dissection [24]. In conclusion, sentinel node status is certainly a pronostic factor but in the absence of a real benefit on overall survival, the present question is to decide if it is reasonable to carry out complete lymphadenectomy for positive sentinel nodes (isolated tumor cells and/or micrometastases).

Acknowledgements

Conflict of interest: none.

References

2 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 2001; 19(16): 3622-34.

3 Trost O, Danino AM, Dalac S, Herve C, Moutel G, Malka G. [Is sentinel l ymph node procedure in early-staged cutaneous melanoma really safe?]. Ann Chir Plast Esthet 2005; 50(2): 113-7.

4 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(3): 192-7.

5 Blumenthal R, Banic A, Brand CU, Ris HB, Lardinois D. Morbidity and outcome after sentinel lymph node dissection in patients with early-stage malignant cutaneous melanoma. Swiss Surg 2002; 8(5): 209-14.

6 Jansen L, Nieweg OE, Peterse JL, Hoefnagel CA, Olmos RA, Kroon BB. Reliability of sentinel lymph node biopsy for staging melanoma. Br J Surg 2000; 87(4): 484-9.

7 Morton DL, Cochran AJ, Thompson JF, et al. Sentinel node biopsy for early- stage melanoma: accuracy and morbidity in MSLT-I, an international multicenter trial. Ann Surg 2005; 242(3): 302-11; (discussion 311-3).

8 Jacobs IA, Chevinsky AH, Swayne LC, Magidson JG, Britto EJ, Smith TJ. Gamma probe-directed lymphatic mapping and sentinel lymphadenectomy in primary melanoma: Reliability of the procedure and analysis of failures after long-term follow- up. J Surg Oncol 2001; 77(3): 157-64.

9 Doting MH, Hoekstra HJ, Plukker JT, et al. Is sentinel node biopsy beneficial in melanoma patients? A report on 200 patients with cutaneous melanoma. Eur J Surg Oncol 2002; 28(6): 673-8.

10 Berk DR, Johnson DL, Uzieblo A, Kiernan M, Swetter SM. Sentinel lymph node biopsy for cutaneous melanoma: the Stanford experience, 1997-2004. Arch Dermatol 2005; 141(8): 1016-22.

11 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. Multicenter Selective Lymphadenectomy Trial Group. Ann Surg 1999; 230(4): 453-63; (discussion 463-5).

12 Gershenwald JE, Colome MI, Lee JE, et al. Patterns of recurrence following a negative sentinel lymph node biopsy in 243 patients with stage I or II melanoma. J Clin Oncol 1998; 16(6): 2253-60.

13 Messina JL, Glass LF, Cruse CW, Berman C, Ku NK, Reintgen DS. Pathologic examination of the sentinel lymph node in malignant melanoma. Am J Surg Pathol 1999; 23(6): 686-90.

14 Shivers SC, Wang X, Li W, Joseph E, et al. Molecular staging of malignant melanoma: correlation with clinical outcome. JAMA 1998; 280(16): 1410-5.

15 Blaheta HJ, Ellwanger U, Schittek B, et al. Examination of regional lymph nodes by sentinel node biopsy and molecular analysis provides new staging facilities in primary cutaneous melanoma. J Invest Dermatol 2000; 114(4): 637-42.

16 Rimoldi D, Lemoine R, Kurt AM, et al. Detection of micrometastases in sentinel lymph nodes from melanoma patients: direct comparison of multimarker molecular and immunopathological methods. Melanoma Res 2003; 13(5): 511-20.

17 Alonso O, Martinez M, Delgado L, et al. Comparison of ^99mTc-MIBI scintigraphy and sentinel node biopsy in the detection of occult lymph node metastases from cutaneous melanoma. Eur J Dermatol 2003; 13(5): 449-54.

18 Ranieri JM, Wagner JD, Azuaje R, et al. Prognostic importance of lymph node tumor burden in melanoma patients staged by sentinel node biopsy. Ann Surg Oncol 2002; 9(10): 975-81.

19 Carlson GW, Murray DR, Lyles RH, Staley CA, Hestley A, Cohen C. The amount of metastatic melanoma in a sentinel lymph node: does it have prognostic significance? Ann Surg Oncol 2003; 10(5): 575-81.

20 Statius Muller MG, van Leeuwen PA, de Lange-De Klerk ES, et al. The sentinel lymph node status is an important factor for predicting clinical outcome in patients with Stage I or II cutaneous melanoma. Cancer 2001; 91(12): 2401-8.

21 Liszkay G, Peley G, Sinkovics I, et al. Clinical significance of sentinel lymph node involvement in malignant melanoma. Pathol Oncol Res 2003; 9(3): 184-7.

22 Morton D, Thompson J, Cochran A, et al. Sentinel-Node Biopsy or Nodal Observation in Melanoma. N Engl J Med 2006; 335(13): 1307-17.

23 Woods GM, Malley RC, Muller HK. The skin immune system and the challenge of tumour immunosurveillance. Eur J Dermatol 2005; 15(2): 63-9.

24 Vuylsteke RJ, Borgstein PJ, van Leeuwen PA, et al. Sentinel lymph node tumor load: an independent predictor of additional lymph node involvement and survival in melanoma. Surg Oncol 2005; 12(6): 1-9.