Abstract
From January 2003 to March 2010, a prospective study was undertaken at the National Cancer Research Institute of Genoa in 15 patients with melanoma who had local recurrence (LR) or a few (≤3) in-transit metastases and clinically-negative regional lymph nodes with the aim of defining: i) the feasibility of sentinel node re-staging (r-sN) of the regional nodal basin; ii) the prognostic value of sentinel node status, and iii) the potential benefit in terms of disease-free survival and overall survival in patients with an histologically-positive sentinel node undergoing therapeutic regional lymph node dissection. Preoperative lymphoscintigraphy was performed to identify the r-sN: the radiotracer was intra-dermally injected around the LR or in-transit metastasis. Moreover, 10 min prior to the operative procedure, 0.5 ml intradermal injection of Patent-Blue-V dye was given around each LR or in-transit metastasis site, so that r-sN identification was achieved by both visualization of the nodal blue dye staining and the information supplied by gamma-detection probe. At least one sentinel node was intra-operatively identified in each patient, and a tumor-positive r-sN was required in four out of fifteen patients. The interval between the diagnosis of primary melanoma and the onset of recurrence was longer, although not significantly, in patients with tumor-negative r-sN, a compared to tumor-positive r-sN (49±47 months vs. 25±19 months, p=0.342). There was a trend toward an improved 1-, 3-, and 5-year disease-free survival and overall survival in patients with tumor-negative r-sN a compared to tumor-positive r-sN. Hence, the r-sN proved to be a feasible and accurate staging procedure even in patients with a few localizations of LR or in-transit metastases (≤3). r-sN identified those with a more favorable prognosis, supporting an aggressive therapeutic approach in the natural history of their disease; moreover, an unnecessary regional lymph node dissection was safely avoided in 11 out of 15 73.3% patients because they had a tumor-negative r-sN.
Recurrent melanoma is a typical manifestation of metastatic disease with a dismal prognosis; the estimated incidence is between 3% to 7%, and it occurs more frequently in patients with extremity location of the primary melanoma, higher Breslow thickness, and previous histologically-positive sentinel lymph node (sN) (1, 2). Locoregional recurrences of melanoma may present as: local recurrence (LR) at the primary excision scar; satellite lesions within 2 cm from the primary site; true in-transit metastasis located between 2 cm from the primary site and the first station regional lymph nodes; and nodal basin recurrence, that is recurrent disease within the scar of a complete regional lymphadenectomy (nodal and soft tissue) which are not classified as in-transit metastasis (1).
The treatment of clinically-negative regional lymph nodes in patients with LR and in-transit metastasis, without a previous regional lymphadenectomy, is somewhat controversial. As a matter of fact, while in the 1990s there was a general consensus for adding prophylactic lymph node dissection to the surgical treatment of recurrence (excision or isolated limb perfusion), this therapeutic recommendation has not been included in more recent guidelines (3-6). On these grounds sN biopsy, a well-known staging procedure in patients with primary melanoma, might theoretically also be helpful in recurrent disease (LR/in-transit metastasis) both as a prognostic marker and for a selective approach to regional lymph node dissection (7-8).
However, due to the paucity of eligible patients, data regarding sN biopsy in this particular subset of patients are rather scant mostly as regards the number of in-transit metastases amenable to sN biopsy, the selection of the most suitable site of injection of the radiotracer for lymphatic mapping, the accuracy of this staging procedure, and the potential survival advantage related to a selective regional lymph node dissection.
For these reasons, a prospective observational study was undertaken in patients with LR and/or with a few (≤3) in-transit metastasis and clinically-negative regional lymph nodes in order to define the feasibility of sN re-staging (r-sN) of the regional nodal basin, the prognostic value of sN status, as well as the potential benefit in terms of disease-free survival (DFS) and overall survival (OS) in patients with an histologically positive sN undergoing therapeutic regional lymph node dissection.
Patients and Methods
From January 2003 to March 2010, 15 patients with LR and/or in-transit metastasis, who had undergone previous sN biopsy without regional lymphadenectomy because of a tumor-negative sN, were selectively recruited to undergo r-sN of the regional nodal basin. The study protocol was approved by the Ethics Committee of the National Cancer Research Institute of Genoa; all patients were informed about the study procedures and gave their written consent.
The r-sN procedure was performed only in patients with a few localizations of LR and/or in-transit metastases (≤3) and without clinical evidence of regional adenopathy or distant metastases. The confirmation of the absence of systemic disease was accomplished by means of a complete staging work-up including magnetic resonance imaging (MRI) of the brain and whole-body positron-emission tomography (PET)-computerized tomography (CT).
The following data were prospectively collected: patient age and gender; characteristics of the primary melanoma (Breslow thickness, Clark level of invasion, growth pattern, anatomic site, and presence of ulceration); the interval between treatment of the primary melanoma and diagnosis of LR and/or in-transit metastasis; the number of LR and/or in-transit metastasis and type of surgery on recurrent disease. Data regarding the r-sN procedure included: the number of lymphatic drainage basins; the number of retrieved sNs; the ratio of histologically-positive sNs to the total number of sNs; the number of positive lymph nodes in patients undergoing regional lymphadenectomy. Disease staging was defined based on the categories of the TNM Staging System, especially distinguishing patients with LR and/or satellitosis or in-transit metastasis without metastatic regional nodes who were classified as having N2c disease from those with associated metastatic regional nodes who were classified with N3 disease (9).
r-sN procedure. Preoperative lymphoscintigraphy was performed in order to identify the node fields receiving direct lymphatic drainage. This process involved intradermal injections of 30 to 40 MBq of 99mTc-albumin nanocolloid (Nanocoll GE Healthcare, Milan, Italy) around the LR or in-transit metastasis, followed by early (5 min) and delayed (20 min) imaging with a gamma-scintillation camera. In the case of multiple recurrences (≤3), the total dose of radiotracer was equally distributed into each lesion. The nuclear medicine physician marked the ‘hot-spot’ of the sN on the overlying skin of the regional basin. The r-sN procedure was performed within 24 h of the radiocolloid injection so that residual radioactivity in the lymph nodes could be measured intraoperatively with a hand-held gamma-detecting probe. Ten minutes prior to the operative procedure, 0.5 ml intradermal injection of Patent-Blue-V dye was given around each LR or in-transit metastasis site, so that sN identification was achieved by both visualization of the nodal blue dye staining and the information supplied by gamma-detection probe (combined technique). Two experienced nuclear medicine physicians independently performed visual interpretation of the scans. The sN was submitted for permanent histological examination: it was initially bisected through its longest median in the hilar plane and, if larger than 9 mm, additional parallel 2-mm slices were cut. Each slice was formalin-fixed and paraffin-embedded. From each tissue block, six 50-micron and six 150-micron sections were cut; sections V, VI, IX and X were immunostained with antibody to MART1 and S-100 proteins, whereas the others were stained with H&E (10). Whenever the r-sN was found to be histologically positive, patients underwent a complete lymph node dissection.
Statistical analysis. DFS was defined as the interval between the r-sN procedure and diagnosis of subsequent locoregional or distant metastasis; overall survival was defined as the interval between the primary treatment of melanoma and the most recent follow-up assessment or death. Survival analysis was accomplished by means of Kaplan-Mayer method (Kaplan-Meir product limit estimator). Survival plots obtained without stratification were supplied with corresponding confidence intervals.
Results
The characteristics of patients and primary melanomas are reported in Table I. In five patients, the recurrence of melanoma was diagnosed within 24 months from diagnosis of primary melanoma and in 10 patients more than 24 months after the primary treatment, with a median interval of 35 months (range=3-179 months). In six patients, there was an isolated LR (n=2) and in-transit metastasis (n=4), while in nine patients there were two (n=5) or three in-transit metastases (n=4). Complete surgical excision with histologically-negative margins of the LR and/or in transit metastasis was obtained in each patient.
In all 15 patients, the preoperative lymphoscintigraphy identified only one drainage basin, and at least one sN was always identified intraoperatively; in nine out of fifteen patients, more than one sN was retrieved from the operative specimen (Table II). Two cases of rather unusual sN nodal basin were observed: popliteal nodes and epitrochlear nodes. At least one histologically-positive sN was identified in four out of fifteen patients, the tumor deposits always being detected by H&E staining. Tumor-positive sNs were detected in 3 out of 12 patients with recurrent melanoma of the extremities, and in 1 out of 3 patients with recurrence of truncal melanoma. All patients with tumor-positive r-sN underwent complete regional lymphadenectomy and two patients had additional tumor-positive lymph nodes in the surgical specimen. No patient with tumor-negative r-sN subsequently developed a nodal recurrence in the same regional basin. No postoperative complication was reported following both r-sN biopsy and complete regional lymph node dissection.
The median interval between surgical treatment of the primary melanoma and diagnosis of LR and/or in-transit metastasis was 25±19 months (SD) in patients with positive r-sN and 49±47 months (SD) in those with negative r-sN (p=0.342) (Figure 1).
Median follow-up after r-sN was 36 months (range=5-105 months). With regard to the type of relapse after the treatment of recurrence, among 11 patients with tumor-negative r-sN, six patients died due to systemic disease, three are still alive with stable disease (distant metastases), and two have remained disease-free. Two out of four patients with positive r-sN died in the following three and 22 months with regional and distant recurrence, while the remaining two patients are still alive with stable disease (distant metastasis) after systemic therapy, 69 and 105 months after the r-sN procedure.
There was a trend towards an improved 1-, 3-, and 5-year DFS and survival in patients with tumor-negative r-sN, compared to tumor-positive r-sN (Figures 2 and 3).
Discussion
Cutaneous and subcutaneous locoregional recurrences of melanoma may present as LR, satellitosis, or in-transit metastasis; they are likely to arise from tumor cells entrapped into lymphatic vessels located between the primary site and the regional nodal basin (11). In-transit metastases are frequently associated with a poor prognosis due to the co-existence of sub-clinical systemic disease, with a 5-year DFS of 36%, and a 10-year survival ranging from 20% to 37% (12-14). Patients with limited disease, that is one or a few in-transit metastases (<5), usually undergo complete surgical excision with histologically-negative margins, with 5-year survival rates of 30% to 40% (15). Conversely, hyperthermic isolated limb perfusion or infusion is the treatment of choice in patients with unresectable in-transit metastases of the extremities (5).
The synchronous management of clinically-negative regional lymph nodes in patients without a previous regional lymphadenectomy is somehow controversial. While in the 1990s there was a general consensus for adding prophylactic lymph node dissection to the surgical treatment of the recurrence (excision or isolated limb perfusion), this therapeutic recommendation has not been included in more recent guidelines (3-6). However, the risk of occult nodal involvement is rather high in this subset of patients, and the sN procedure might be theoretically helpful for the pathological staging of a clinically-negative regional nodal basin with the aim of performing a more selective approach to complete lymph node dissection (7, 8).
However, literature data have been collected only in a few patients with questions regarding the selection of eligible patients, the technical details and the accuracy of the procedure, and the potential survival advantage related to this selective regional lymph node dissection.
In our experience, patients with one or a few (≤3) LR/in-transit metastases were mapped by means of injection of the radiotracer into each site of recurrence, based on the concept that finite regions of the skin drain into specific lymph nodes within the regional nodal basin, as originally suggested by Morton et al. (16). On the one hand, the feasibility of the procedure was confirmed by the finding of at least one r-sN in all patients, notwithstanding the previous sN biopsy of the primary melanoma. On the other hand, no false-negative result occurred as patients never developed regional node recurrence after a tumor negative r-sN, thus indicating the high accuracy of the lymphatic mapping, even in the case of up to three in-transit metastases.
Based on these findings and the hypothesis that in-transit metastasis should arise from tumor cells entrapped within the lymphatic ducts between the primary site and the regional lymphatic basin, the radio-tracer injection should preferably be performed into each site of recurrence instead of the scar of primary melanoma, as suggested by Dewar et al. (17). In fact, their failure to detect the correct sN in one patient after injection of the radiotracer into the in-transit deposit might be related to neoplastic embolization of regional lymphatic vessels, as suggested by the extensive involvement of the regional node basin (11 tumor-positive lymph nodes out of 23) in that particular patient, thus determining a diversion of the lymphatic mapping toward a false-negative lymph node.
As regards to the prognostic value of sN re-staging in this clinical subset, allows definition of patients with LR and/or in-transit metastasis and associated regional node metastasis (N3, stage IIIC) with a worse prognosis (5-year survival rate of 26%), compared to patients without regional lymph node metastasis (N2c, stage IIIB) and an estimated 5-year survival rate of 53% (9, 18). This fact, coupled with the observation that patients with a tumor-negative r-sN are later more likely to develop locoregional recurrence, might suggest a more favorable natural history of their disease, supporting a more aggressive therapeutic approach.
Regarding the therapeutic value of r-sN staging, the rather low number of patients recruited in this clinical series does not allow for any firm conclusion to be drawn as to its impact on survival. However, it is worth noting that in 11 out of 15 patients, the tumor-negative status of the r-sN safely allowed avoidance of an unnecessary regional lymph node dissection because none of these patients developed a regional recurrence. On these grounds, r-sN staging may certainly represent a valuable instrument for selecting patients with LR and/or in-transit metastasis and clinically-negative regional nodes who should not undergo complete regional lymph node dissection. Whether r-sN could identify patients with a low nodal tumor burden who might benefit from an early lymph node dissection with a positive impact on survival needs to be determined on a wider sample of patients.
Conclusion
The r-sN proved to be feasible and accurate staging procedure, even in patients with a few localizations of LR and/or in-transit metastasis (≤3) without evidence of regional nodal disease, in order to identify patients with a more favorable prognosis, supporting an aggressive therapeutic approach in the natural history of their disease; moreover, it was possible to avoid an unnecessary regional lymph node dissection safely whenever a tumor-negative r-sN was found.
Footnotes
-
Financial Disclosure
The Authors Marco Gipponi, Nicola Solari, Davide Giovinazzo, Paola Queirolo, Sergio Bertoglio, Giuseppe Villa, Marina Gualco, Dario Bleidl, and Ferdinando Cafiero have no financial interest in any of the products, devices, or drugs mentioned in this article.
- Received February 5, 2014.
- Revision received April 11, 2014.
- Accepted April 16, 2014.
- Copyright© 2014 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved