We performed an observational retrospective study including all patients with cutaneous melanoma of the trunk with negative SLNB, diagnosed between October 2004 and September 2017 at Coimbra University Hospital Centre, Coimbra, Portugal. Medical records were analyzed from the first dermatologic consultation until the last evaluation or death. Patients with at least 12 months of follow-up, or less, in case of earlier progression to nodal or metastatic disease were included.

Two groups were considered for comparison: patients with multiple-LBD and single-LBD. Patients were considered to have multiple-LBD when SLNB was performed in more than one regional lymph node basin, regardless of the total number of nodes excised.

The endpoints included the first nodal or metastatic disease after negative SLNB in a patient not submitted to any other medical or surgical treatment so far.

Patients’ clinical features (age, gender and tumor localization on the trunk), melanoma histological features (Breslow depth, ulceration, and histological subtype), the pattern of lymphatic drainage (number and localization of node basins explored and sentinel lymph nodes excised), nodal recurrence-free survival (nodal-RFS), metastatic recurrence-free survival (metastatic-RFS) and overall recurrence-free survival (overall-RFS) were reviewed.

Since the study was conducted within a 14-year time period, differences in the approach to the regional and distant metastatic disease (namely different indications for complete lymph node dissection, or metastatic disease surgery; indications for systemic therapy of metastatic disease with chemotherapy, targeted therapy or immune checkpoint inhibitor therapy; and more recently adjuvant therapy) were registered.

However, for the purpose of this study, these changes in the guidelines could not impact the present study’s results since they did not influence the drainage pattern and only recurrence-free survivals (i.e., before the effect of any possible complete lymph node dissection, or systemic therapy either for metastatic disease or in the adjuvant setting) were considered.

The diagnosis of melanoma was based upon excisional biopsy or incisional biopsy followed by radical excision and histopathology analysis. An excisional biopsy was performed with narrow margins. Wide local excision with a margin of 1 or 2 cm depending on Breslow thickness.

Patients were selected for SLNB according to the recommended guidelines at that time.

A 2-day protocol using [99mTc] Tc-human serum albumin nanocolloid was used to perform a SLN scintigraphy: On day-1, the injections were made at the primary melanoma site. Immediately after the injection a dynamic imaging study with the field of view centered on the lesion was started. Static images 20-minutes and 2-hours after injection were acquired. Two and half hours after tracer injection, a SPECT with low-dose CT scan without contrast was made. On day 2 SLN biopsy was attempted on all nodal basins that had been identified by preoperative lymphoscintigraphy. For this, a portable gamma-probe was used intraoperatively, to detect the emission of radiation from SLN, thus allowing its identification and excision.

Nodal basins included right and left axilla, neck, groin, scapular, supraclavicular, intercostal, or submandibular. In-transit or satellite nodes were not considered as a separate basin.

During the study period, the lymphoscintigraphy protocol remain unchanged and was performed by the same nuclear medicine team. The SLNB procedure was also accomplished by the same dermatologic surgical team of the hospital.

IBM SPSS® software, version 25.0 (Armonk, NY: IBM Corp.) was used. Only p-values of non-parametric tests were reported since most continuous variables had a non-normal distribution. The Chi-Square test was used to compare categorical variables between multiple and single-LBD groups. Kaplan-Meier method and log-rank test were used to compare survival estimates between groups. Logistic regression was used to test for the significance of the predictor variables. Statistical significance was set at p-value ≤0.05.

During the study period, 206 patients had a diagnosis of melanoma of the trunk and were selected to perform SLNB in the present study’s institution. Of these, 147 had negative SLNB and were eligible for the present study. The patients who were lacking clinical or pathological information or did not fulfill the required follow-up time were excluded and a total of 135 patients were included.

The clinical characteristics of the patients and pathological features of primary melanoma, according to the nodal drainage pattern, are summarized in Table 1.

Seventy-eight patients (57.8%) were males, with a median age of 69 years (33–95). Melanoma was located in the upper trunk in 101 patients (74.8%), in the inferior trunk in 34 (25.2%), in the front in 32 (23.7%), and in the back in 103 (76.3%).

Histologic subtypes of melanoma identified were superficial spreading (60), nodular (39), nevoid (1), spitzoid (1), desmoplastic (1), or other (7). In 26 cases the histologic subtype classification was missing, despite having information about all other studied variables.

Median Breslow depth was 1.7 mm and ulceration was present in 47 (34.8%) patients.

During the follow-up period, 3 patients (2%) had an only nodal recurrence, 24 (17.8%) developed exclusively metastatic recurrence and 9 (6.7%) had both nodal and metastatic recurrence.

Multiple-LBD was identified in 61 (45.2%) patients, and in most of them (88.5%) only 2 basins were explored. In 35 (57.3%) patients with multiple-LBD, there was bilateral symmetric identification of SLN, being both axilla in 30 (46.6%) patients and both groins in 5 (8.2%). In 6 patients, 3 basins were explored, and in a single patient, drainage occurred to 4 basins.

Of the 74 patients with single-LBD, the SLN was located mostly in the axilla (63), and less often on the groin (7), cervical area (1), or other uncommon basins (3) (scapular and supraclavicular) (Table 2).

Posterior trunk melanomas were more often associated with multiple-LBD (51.5%), while in anterior trunk melanomas single-LBD was more common (75%), being these differences statistically significant.

Any other clinical or histological feature significantly associated with the drainage pattern (multiple-LBD and single-LBD).

The median follow-up period was 5.6 years (range 1–15 years).

Ten of the 74 (13.5%) patients with single-LBD developed nodal recurrence compared with 2 of the 61 (3.6%) patients with multiple-LBD (p = 0.04) (Fig. 1 and Table 1).

Figure 1.

Lymph node basin drainage of patients with nodal recurrence. Single-LBD, Single Lymph node Basin Drainage; Multiple-LBD, Multiple Lymph node Basin Drainage.

(0.18MB).

In the single-LBD group, 5 patients had a nodal recurrence in the same basin identified in SLNB, meaning a false negative rate of SLNB of 6.76% (5/74). In another 5 patients with single-LBD, nodal recurrence developed in a regional basin not identified in SLNB.

Among the 2 patients with multiple-LBD, only 1 developed nodal recurrence in a territory not previously studied by SLNB, meaning a false negative rate of 1.63% (1/61). The overall false-negative rate for SLNB was 4.4% (6/135).

In survival analysis, nodal-RFS was significantly longer in the group with multiple-LBD, than in the group with single-LBD (175.6 vs. 138.7 months; p = 0.04) (Fig. 2).

Figure 2.

(A) Kaplan-Meier: Nodal Recurrence-Free Survival (NRFS) was significantly more favorable in patients with multiple-LBD. (B and C) No significant differences in Distant Recurrence-Free Survival (DRFS) and Overall Recurrence-Free Survival (ORFS) were found, according to drainage pattern. Single-LBD, Single Lymphatic Basin Drainage; Multiple-LBD, Multiple Lymphatic Basin Drainage. ** Statistically significant difference in survival curves.

(0.26MB).

On univariate analysis, single-LBD (HR = 4.34; p = 0.05) and Breslow depth (HR = 1.36; p = 0.05) were the only factors significantly associated with a higher risk of nodal recurrence, in negative-SLNB patients.

After adjustment for Breslow depth, the single-LBD remained associated with a higher risk of nodal recurrence (HR = 4.54; p = 0.05) (Table 3).

Although not statically significant, there was a tendency for a lower risk of nodal disease when the number of basins increased (HR = 0.24; p = 0.06).

The number of lymph nodes excised did not correlate with nodal disease risk (HR = 0.75; p = 0.32).

No significant differences in metastatic-RFS and overall-RFS were found between single or multiple-LDB patients (Fig. 2). On logistic regression, the patter of lymphatic drainage did not affect the risk of metastatic (distant) or overall (nodal and metastatic) disease recurrency (p = 0.27 and p = 0.12, respectively).