Prognostic value of the number of biopsied sentinel lymph nodes for Chinese patients with melanoma: A single‐center retrospective study

Abstract Background Sentinel lymph node biopsy (SLNB) helps to determine accurate pathological stages and facilitates strategies for regional disease control in melanoma. However, whether the number of biopsied sentinel lymph nodes (SLNs) influences the patients' survival is rarely investigated. Methods Acral or cutaneous melanoma patients with no history of nodal disease who received SLNB in Fudan University Shanghai Cancer Center (FUSCC) from January 1, 2017, to December 31, 2021 were retrospectively enrolled. Clinicopathological variables including Breslow index, ulceration, number of positive SLNs, SLN/non‐SLN status were analyzed. The pathologic nodal (pN) stage and pathological stage were defined. Results A total of 381 eligible patients were enrolled in this study, of whom 132 (34.7%) patients were diagnosed with SLN‐positive. The median number of biopsied SLNs was 2 (range: 1 to 20). Different numbers of biopsied SLNs did not influence the release‐free survival (RFS) of the general patients. However, patients with >2 SLNs had a longer RFS than those with 1–2 SLNs in T4, N1a group and those who rejected complete lymph node dissection (CLND). Conclusions In patients with T4 melanomas, N1a melanomas and those that did not undergo a CLND, the prognosis of those with three or more SLNs retrieved seemed to be improved.

is an independent prognostic factor of melanoma, as well as primary tumor thickness. 2,3SLNB is suggested for primary melanoma patients with Breslow thickness more than 0.8 mm, or less than 0.8 mm with risk factors, for instance, regression, ulceration, neurovascular invasion, and young age. 36][7][8][9] Macro-metastasis of SLN invasion was associated with higher risk of non-SLN involvement and always indicated a poor prognosis with respect to micrometastases. 10wever, few studies have concentrated on the prognostic role of other indicators of SLNB.One major concern is the accuracy of SLNB, which depends on the lymphatic mapping and adequacy of nodal biopsy.The number of biopsied SLNs was rarely discussed in previous studies, because SLNB is generally considered as a minimally invasive procedure to prevent limb swelling and numbness.On the other hand, more than one node is detected in clinical practice during SLNB.The biopsy of all identified SLNs might lead to more accurate pathological stagging.According to the stagging criteria, more regional nodes involved with metastatic melanoma is always classified as higher stage and is related to bad prognosis.However, Jakub et al. found if the positive lymph nodes is confined to the SLNs without positive non-SLNs, the number of regional lymph nodes involved with metastatic disease does not affect DFS and OS. 11In addition, inadequate number of SLNs was found to increase the positivity rate in non-SLNs. 12Thus, except for a minimal requirement on the number of lymph nodes evaluated after CLND, 13 it is still essential to indicate minimal SLNs that should be biopsied during SLNB.Therefore, the present single-center retrospective study aimed to further investigate the prognostic value of the number of biopsied SLNs for Chinese patients with melanoma.

| PATIENTS AND METHODS
Primary cutaneous or acral melanoma patients with no clinical nodal disease, who underwent SLNB at Fudan University Shanghai Cancer Center (FUSCC; Shanghai, China) from January 1, 2017, to December 31, 2021 were retrospectively enrolled in the study.Patients with incomplete medical data or with a follow-up of less than 6 months were excluded from the study.Clinicopathological variables, including gender, age, Clark level, Breslow thickness, ulceration, number of positive SLNs, SLN/non-SLN status, and adjuvant therapy, were recorded.
The pathologic nodal (pN) stage and pathological stage of patients were defined based on the 8th edition of the American Joint Committee on Cancer (AJCC) staging manual. 14

| SLNB
SLNB was routinely performed under lymphatic mapping through lymphoscintigraphy by technetium-99 sulfur colloid, methylene blue dye, or both.The pathological examination of resected SLN was described in our previously published study. 15Briefly, each of SLNs was dissected at 3 mm thickness or along the longest axis on the largest surface, fixed by 3.7% neutral formaldehyde, then dehydrated, conventionally embedded into paraffin and finally stained with hematoxylin and eosin (HE).The metastatic status of SLNs was also determined by immunohistochemical staining of HMB45, S-100, SOX10 and Melan A. The antibodies against S-100 and Melan A were purchased from Dako Co., Ltd.(Copenhagen, Denmark).The antibody against HMB45 was purchased from MaiXin Biotechnologies Co., Ltd.(Fuzhou, China).The antibody against SOX10 was purchased from Gene Tech Co., Ltd.(Shanghai, China).Each section was observed under a light microscope by two pathologists.The majority of patients with metastatic SLNs received CLND within 1 month.

| Follow-up
Patients were assessed through clinical examinations, such as routine physical checkup, ultrasound, computed tomography and/or magnetic resonance imaging every 3 months for the first 2 years, every 6 months for 3-5 years, and then annually.Patients were followed up through reexamination or telephone follow-up until death or December 31, 2021.The survival of patients was censored at the date of the last follow-up (December 31, 2021).Relapse-free survival (RFS) was defined as the interval between radical surgery and the first recurrence or distant metastasis.Recurrence or metastasis was confirmed by pathology or imaging follow-up.

| Statistical analyses
Categorical variables were described by the frequency and component ratio.The Pearson's chi-squared test or the Fisher's exact test was used for univariable analysis between patients with 1 to 2 biopsied SLNs and those with more than 2 biopsied SLNs.The Kaplan-Meier method and the logrank test were adopted to describe and compare the impacts of some prognostic factors (SLN-positivity, acral subtype, N1a stage, T4 stage and SLN-positivity without CLND) on RFS between these two groups.p-Values less than .05were considered statistically significant.The statistical analysis was performed using SPSS 22.0 software (IBM, Armonk, NY, USA).This study was approved by the Ethics Committee of FUSCC, and every participant signed an informed consent form before enrollment.

| Baseline characteristics
A total of 381 eligible patients were enrolled in this study, and their median age was 58 years, ranging from 19 to 89 years.

| Number of biopsied SLNs and positivity rates
As the median number of biopsied SLNs was equal to 2 (range: 1 to 20), patients were assigned into two groups: 216 (56.7%) patients with 1 to 2 biopsied SLNs in group one, and 165 (43.3%) patients with more than 2 biopsied SLNs in group two.Upon Pearson's chi-squared test and the Fisher's exact test, no significant difference in the clinicopathological variables, such as gender, age, Breslow index, ulceration rate, N stage, TNM stage, gender, the Clark level, gene mutation, and pattern of relapses between the two groups.Of all these patients, 243 (63.8%) patients received adjuvant therapy and no significant difference was found in adjuvant therapies between these two groups (Table 1).As melanoma of acral type was dominant in our country and in which lower extremity was the most common primary tumor location, groin became the most common SLNB basin of these two groups, accounting for 85.2% and 58.8% respectively.The rate of SLN-positivity was 34.7% in group one and 34.5% in group two, while the rate of non-SLN-positivity in the two groups was 8.3% and 3.0% (p = .161),respectively.It was revealed that in patients with Breslow >4 mm and N1a status, the rate of non-SLN-positive was 40.7% (11/27) in group one, which was higher than that in group two (17.6%,3/17), but not statistically significant (p = .109).Furthermore, there were more cases with acral melanoma (n = 164, 75.9%, p = .003),more patients with melanoma at the lower extremities (n = 176, 81.5%, p < .001),and a higher rate of SLNB in groin (n = 184, 85.2%, p < .001) in group one (Table 1).Besides, 17 (7.9%)patients in group one and 14 (8.5%) patients in group two declined CLND, and no significant difference was detected.

| DISCUSSION
In this retrospective study involving Chinese melanoma patients with no nodal disease who received SLNB, identification and retrieval of 3 and more SLNs did not correspond in a different RFS compared to patients with only 1 or 2 SLNs.In patients with 3 or more SLN, there was a trend towards a longer RFS for patients with severe Breslow thickness (>4 mm), N1a stage, and those with SLN-positive without plan of CLND.This result implied the underling importance of the number of SLNs in melanoma, which was rarely mentioned before.
In melanoma, SLN is mainly identified by using lymphoscintigraphy after injecting radioactive technetium adjacent into the primary lesion, or injecting methylene blue dye preoperatively into four quadrants at derma layer around the primary site.Detection of the SLNs was then finished by using a gamma probe and visual observation of methylene blue dye lymphatic drained into the regional lymph node field.7][18] Indocyanine green is a new method mapping the lymph node without using radioactive tracer and without the risk of tattooing that is associated with the use of blue dye. 19However, success rates for SLN identification are insufficient, because they merely indicate the proportion of patients in which at least one SLN was identified and removed, while they do not define the proportion of patients in which all SLNs were identified by accurate preoperative lymphoscintigraphy. 20cording to the final reports of German Dermatologic Cooperative Oncology Group trial (DeCOG-SLT) and Multi-Center Selective Lymphadenectomy Trial II (MSLT-II), immediate CLND did not bring survival benefits to SLN-positive patients. 21,224][25] Thus, whether CLND is still unbeneficial for OS in Asian patients requires validation in a larger sample-sized study, although it is the question if such a trial will ever be conducted.As Chinese patients always have more tumor burden, higher rates of non-SLN metastasis and adjuvant systemic therapy in acral melanoma might be less effective than in other cutaneous melanomas, CLND for patients with SLN-positive is still strongly recommended in daily practice, followed by adjuvant systemic therapy.Patients who rejected CLND were featured by a slight tumor burden in lymph nodes, primary site Similarly, the median number of SLNs was also 2 in the present study, which was in accordance to the results of previous studies. 22,26The present study showed that different numbers of biopsied SLNs had no influence on the RFS in the general population, while no significant difference was found in the exact number of positive lymph nodes or the rate of SLNB-positive between the two groups.
However, patients with >2 SLNs had a longer RFS than those with 1-2 SLNs in T4, N1a group and those who rejected CLND.The underlying reasons might be summarized as follows: (i) identification of more SLNs might be associated with a higher accuracy of estimating N stage, and a lower false-negative rate; (ii) the median number of SLNs in acral subtype was less than that of the cutaneous subtype, indicating that for acral melanoma, especially in patients with lesions at lower extremities and biopsied in groin, there might be a higher risk of incomplete nodal detection; (iii) for primary lesion site with Breslow thickness >4 mm and a higher rate of SLN-positive, evaluation of adequate SLNs might be more critical; and (iv) a higher number of biopsied SLNs-positive reduced the probability of residual tumor, and lowered the relapse risk in patients who rejected CLND.In addition, the rate of non-SLN-positive was 40.7% in cohort with 1-2 SLNs in patients with Breslow >4 mm and N1a stage, which was higher than that of cohort with >2 SLNs (17.6%).This result was in accordance with findings from Rossi et al. that fewer nodes excised was a risk factor for non-SLNs involvement of metastasis. 12This might further explain why SLN-positive (>2 SLNs) patients without plan of CLND had a significantly longer RFS.As higher SLNs were extracted from the basin, the residual tumor burden in lymph nodes decreased correspondingly.Thus, 3 and more SLNs seem essential for patients with Breslow thickness >4 mm and those with no plan of CLND.
Although the present study provided new insights into the effects of the number of SLNs on the RFS of melanoma patients, its limitations are also noteworthy.Due to the retrospective design of the study, the risk of selection bias is inevitable, indicating the necessity of further prospective study.Furthermore, the effects of different adjuvant therapies on the disease relapse and OS were not assessed.
It is noteworthy that targeted therapy might be more efficacious than other adjuvant therapies, such as immunotherapy, during the early period of disease.In addition, data were collected from one singlecenter, demonstrating that further multicenter studies are required to indicate the potential influence of the number of SLNs on the prognosis of melanoma patients and to explore its intrinsic mechanism.

| CONCLUSIONS
In conclusion, we found that the number of identified and retrieved SLNs did not significant impact on RFS.In patients with T4 melanomas, N1a melanomas and those that did not undergo a CLND, the prognosis of those with three or more SLNs retrieved seemed to be improved.Three and more SLNs are essential for patients with Breslow thickness >4 mm and those with no plan of CLND.

F I G U R E 1
Relapse-free survival of patients with different numbers of biopsied sentinel lymph nodes (SLNs).(A) All patients; (B) Patients with SLN-positive; (C) Patients with acral subtype.F I G U R E 2 Relapse-free survival of patients with different numbers of biopsied SLNs in specific subtypes.(A) Patients with N1a stage; (B) Patients with T4 stage; (C) Patients with SLN-positive but without completion lymph node dissection (CLND).
Comparison of clinicopathological variables between groups with different numbers of sentinel lymph nodes.shallow layer of skin, and older age.In this study, the rate of CLND in patients with SLN-positive was 73.8%.