The treatment of patients with melanoma and positive sentinel lymph node biopsy (SLNB) has drastically changed after the publication of two clinical trials that advised against performing complete lymph node dissection (CLND) due to the lack of benefit in melanoma-specific survival (MSS).1,2 Clinical practice guidelines on the management of melanoma have incorporated this change in their recommendations, though not uniformly.3–6 Thus, while the performance of CLND in these patients has significantly decreased in clinical practice,7–12 it is still recommended and performed in some clinical contexts considered high risk, such as the presence of lymphovascular invasion or immunosuppression. CLND has also been maintained in cases with clinical characteristics for which there is less evidence, as these were underrepresented in the above-mentioned clinical trials, such as extracapsular extension of lymph node metastasis, involvement of three or more nodes, metastasis >1mm in size, melanoma located in the head and neck, or involvement of two or more lymphatic regions.13–15

Change in the surgical treatment of patients with regional metastasis has coincided with an increase in survival, both relapse-free and melanoma-specific, achieved in the last decade thanks to new adjuvant therapies.16–19 Among the recommended options in the guidelines for stage III patients are dabrafenib plus trametinib for patients with BRAF (serine/threonine protein kinase B-raf) V600 mutation (not funded in Spain) and nivolumab or pembrolizumab regardless of BRAF status (only funded in Spain for stages IIIC and IIID during the study period).6,20

In our region, there are no studies analyzing the surgical treatment of patients with melanoma with a positive SLNB in recent years. Understanding clinical practice could contribute to better optimization of health care resources. The primary endpoint of this study was to evaluate the progression and current status of CLND indication in patients with melanoma and positive SLNB. Secondary endpoints included analyzing clinical and pathological characteristics associated with CLND, and determining whether CLND impacts survival in these patients.

A total of 449 patients who underwent a positive SLNB were included, 19 of whom were excluded due to incomplete data regarding dates, inclusion errors, lack of follow-up, or localization in mucosa (Fig. 1). A total of 430 patients were analyzed, and their descriptive data are shown in Table 1. Regarding gender, cases were more common in men (n=233 [54.2%]) vs women. The most common age group was between 60 and 80 years (n=169 [39.3%]), followed by the 45–60 years group (n=154 [35.8%]). The least common group was elderly patients >80 years (n=35 [8.1%]), and this age group also had the lowest rate of CLNDs (only 4 [11.4%]; p=0.02). The trunk was the most common tumor location (n=179 [41.7%]), followed by the lower limbs (n=81 [18.9%]), and the head and neck (n=75 [17.5%]). Regarding tumor thickness, 37.3% (n=157) had a Breslow thickness of 2.1mm up to 4mm, while thin tumors ≤1mm accounted for 6.89% of the sample (n=29). There was a trend towards increased CLND proportional to tumor thickness and pathological T stage. The most common histological subtype was SSM (n=214 [53.6%]), followed by MN (n=115 [28.8%]). For SLNB metastases, most cases had a single metastatic lymph node (n=305 [70.9%]), and metastasis size was >1mm in 55.8% (n=213). CLNDs increased both in cases with ≥2 positive nodes and those with metastasis sizes ≥0.1mm.

Figure 1.

Flow diagram of patients with cutaneous melanoma and a positive sentinel lymph node biopsy from 2017 through 2022.

(0.11MB).

Over the study period, the largest group consisted of 144 patients (33.7%) who received only adjuvant therapy. CLND was performed alone in 49 patients (11.4%), and the combination of CLND plus adjuvant therapy occurred in 97 cases (22.7%). Observation was chosen for 138 patients (32.2%) (Table 1). Stage distribution is shown in Fig. 1 (supplementary data). Importantly, these proportions changed over the years of the study. In fact, CLND in these patients with positive SLNs was predominantly performed early in the study period (2017), becoming marginal in the most recent year, 2022 (Fig. 2). Similarly, treatment strategies shifted significantly, with CLND decreasing and adjuvant therapy becoming more prominent during the study period (Fig. 3).

Figure 2.

Distribution of patients with and without complete lymph node dissection from 2017 through 2022.

(0.11MB).

Figure 3.

Distribution of patients based on the management received after positive results in the selective sentinel lymph node biopsy from 2017 through 2022.

(0.15MB).

In univariate analysis, variables associated with CLND were tumor thickness >4mm (OR, 2.7 [95% CI, 1.1–7.1]; p=0.04), nodular melanoma histological subtype (OR, 1.7 [95% CI, 1.1–2.8]; p=0.03), ≥2 positive nodes (OR, 1.7 [95% CI, 1.1–2.6]; p=0.013), lymph node metastasis size >1mm (OR, 11.4 [95% CI, 2.6–49]; p=0.001), Turin hospital (OR, 4 [95% CI, 2.2–7.2]; p=0.001), the combined group (OR, 3.1 [95% CI, 1.7–5.7]; p=0.001), and extracapsular extension (OR, 5.2 [95% CI, 2.1–13.1]; p<0.001). Age >80 years was inversely associated with CLND (OR, 0.3 [95% CI, 0.1–0.9]; p=0.02). Multivariate logistic regression analysis found independent variables associated with CLND were age >80 years (OR, 0.3 [95% CI, 0.1–1]; p=0.05), MLA histological subtype (OR, 2.94 [95% CI, 1.95–6.92]; p=0.01), metastasis size >1mm (OR, 7.91 [95% CI, 1.71–36.47]; p=0.008), extracapsular extension of metastasis in SLNs (OR, 4.81 [95% CI, 1.51–15.28]; p=0.008), Turin hospital (OR, 4.4 [95% CI, 2.11–9.32]; p<0.001), and the combined group (OR, 3.34 [95% CI, 1.58–7.08]; p<0.001) (Table 2).

Univariate analysis showed no differences in various survival measures (LRFS, RRFS, DSS, and OS) between patients undergoing CLND vs observation. However, when analyzed by treatment type (Fig. 4), worse survival was observed in patients undergoing CLND alone vs others (p=0.04). Survival differences by stage are shown in Fig. 2 (supplementary data).

Figure 4.

Local recurrence-free survival, regional lymphatic recurrence-free survival, disease-specific survival, and overall survival using the Kaplan–Meier method based on treatment received in patients with melanoma and positive sentinel lymph node biopsy (n=430).

(0.51MB).

This is the first study describing the treatment of patients with melanoma and positive SLNB in a selection of centers in Southern Europe (Spain and Italy). A notable decrease in the number of CLNDs performed has been observed, and conversely, there has been an increasingly frequent use of adjuvant therapy in these patients.

Based on the Multicenter Selective Lymphadenectomy Trial-II (MSLT-II) and the German Dermatologic Cooperative Oncology Group-Selective Lymphadenectomy Trial (DeCOG-SLT), there is clear evidence that performing CLND in patients with a positive SLNB does not benefit melanoma-specific survival vs routine ultrasound-based lymph node follow-up.1,2 A recent meta-analysis even found that, paradoxically, the observation group had a slight survival advantage at 3 and 5 years. These results are consistent and can be explained by the way melanoma spreads. Its hematogenous spread is independent of lymphatic dissemination and is not progressive or orderly as previously hypothesized during the 1990s.25–27

The findings of this study have also been reported in previous observational studies conducted in other countries, including a major international multicenter study7 and two large population-based studies from the U.S. National Cancer Database,8,12 which reflect a significant decrease in the number of CLNDs in recent years.28

The landmark clinical trials on the topic of discussion are no stranger to criticism, particularly since most patients included (∼70%) had a small tumor burden (<1mm), meaning that high-risk recurrence patients, such as those with a large tumor burden in the sentinel node, extracapsular extension, >3 affected nodes, >2 affected regions, or even those with melanoma located in the head and neck region, were underrepresented. For example, the head and neck region are absent in the DeCOG-SLT and underrepresented in the MSLT-II, which observed a trend toward better survival in the CLND group, but without reaching statistical significance (HR, 1.60 [0.96–2.66]; p=0.07) with 113 vs 128 patients in the dissection vs observation groups.

Since our study is based on the routine clinical practice, it more accurately reflects the characteristics most widely seen in patients with positive SLNB. For example, more than half of the patients had a tumor burden in the node >1mm. This and other high-risk factors, such as extracapsular extension, were associated with the performance of CLND, as was having a melanoma thicker than 4mm. Conversely, elderly patients (>80 years) underwent CLND less frequently. In literature, most published studies have also found an inverse relationship between CLND and age.8,10–12 Other related factors include the number of affected nodes and tumor size.7,11 Localization has been associated with a higher frequency of CLND, with melanoma in the head and neck region more likely to lead to lymphadenectomy, while localization in the lower limbs is inversely related.7,12 Decision-making regarding melanoma in the head and neck region is especially controversial, not only because of the results of the MSLT-II but also due to the contradictory findings in observational studies in this localization.29–31 Some argue in favor of CLND in this region for better regional control of the disease, as recurrence could lead to significant morbidity.32 Furthermore, CLND of cervical nodes has lower morbidity vs other regions such as axillary and inguinal, which can experience up to 50% complications, including infection, seroma, and lymphedema. However, the higher complication rates following inguinal CLND likely explain the trend of avoiding CLND in melanoma located in the lower limbs, as reported by several studies.33

No differences in survival were found between patients undergoing CLND and those in the observation group in univariate analysis. However, it would be interesting to compare survival across patients who underwent CLND vs observation, particularly for those with high-risk melanoma and positive SLNB (extracapsular extension, high tumor burden, >3 affected nodes, >2 affected lymphatic regions, and head and neck localization).13–15 Yes, the current trend assumes that the biological behavior and the rationale for observation vs CLND are independent of these characteristics.

Variability in hospital practices regarding the recommendation of CLND in these patients has been observed, reflecting differences in the temporal adaptation to the clinical practice guidelines across centers, and differences in the availability of clinical trials, appropriate radiological follow-up, or status of reference centers.12

Our study shows the growing role of adjuvant therapy in recent years, with nearly 60% of patients receiving systemic adjuvant therapy. More than half of these patients did not undergo CLND, particularly in the later years of the study. These data are consistent with treatment patterns described in the literature.7,8,11

Even before the publication of the MSLT-II and DeCOG-SLT data, studies had shown improved relapse-free survival in melanoma with positive SLNB on adjuvant immunotherapy.12,16,17 However, these studies only included patients who had undergone CLND, so the technique continued to be recommended initially. A new study on the efficacy of systemic adjuvant therapy in patients with melanoma and positive SLNB who did not undergo CLND showed a 67% improvement in relapse-free survival at 24 months in the adjuvant therapy group.19

As limitations, it should be noted that this is a retrospective study, the sample size is smaller vs other similar studies, and there are differences in the number of cases included across hospitals. However, the geographical distribution of the reference centers involved gives the study generalizability and current relevance. Furthermore, this is the only study of its kind conducted in centers in the Mediterranean area.

Based on the recent evidence included in current versions of clinical practice guidelines, there has been a significant decrease in the number of CLNDs performed in patients with melanoma and positive SLNB. This study demonstrates this same decrease in reference hospitals in our area. At the same time, there is an increased use of adjuvant therapy in this patient group, particularly since 2019. However, CLND continues to be performed in certain cases in clinical practice, as its recommendation has not been entirely eliminated from the guidelines, due to the presence of high-risk groups where the role of lymphatic dissection remains unclear. Further studies are needed to evaluate the effect of CLND in high-risk melanoma patients with positive SLNB.

This research has not received specific funding from public agencies, commercial sectors, or nonprofit organizations for the participating centers, except for research conducted by the Melanoma Working Group at Hospital Clínic in Barcelona, which is supported by the Center for Biomedical Research on Rare Diseases (CIBERER) from Instituto de Salud Carlos III (Madrid, Spain); AGAUR 2017_SGR_1134 and the CERCA Program from the Generalitat de Catalunya, Spain; a Research Grant from “Fundación Científica de la Asociación Española Contra el Cáncer” GCB15152978SOEN, Spain; the European Commission under the 6th Framework Program, Contract No. LSHC-CT-2006-018702 (GenoMEL), the 7th Framework Program, Diagnoptics; the European Commission under the HORIZON2020 Framework Program, iTobos and Qualitop; and the European Commission under the Horizon Europe Program, HORIZON-MISS-2021-CANCER-02, MELCAYA (reference 101096667). This research was partly supported by grants from the Health Research FundP.I. 18/00419 and 22/01467, Spain. Part of the work was conducted at the Esther Koplowitz Center in Barcelona, Spain.

None declared.