Sentinel Lymph Node Predictors in Melanoma of Breslow Thickness 0.8-1.0 mm

Introduction: Current melanoma staging guidelines consider all patients staged T1b to have the same metastatic risk and recommends that sentinel lymph node biopsy (SLNB) be considered in this disparate group. The goals of this study were to specifically determine predictors of sentinel lymph node positivity and those predictive for obtaining a SLNB in melanomas of Breslow thickness 0.8 mm to 1.0 mm, which has not been previously studied.


INTRODUCTION
favorable prognosis with low probability of metastatic spread. 2 However, a small percentage of these patients may go on to develop advanced disease. Based on the shared 5%-10% metastatic risk in patients with primary melanoma of Breslow thickness < 0.8 mm with ulceration and 0.8 mm -1.0 mm with or without ulceration, these two different patient cohorts are staged as T1b and the National Comprehensive Cancer Network (NCCN) melanoma guidelines recommend considering SLNB in these two populations. 3 The potential metastatic risk in patients with thin melanomas has led to an extensive attempt to determine predictors of sentinel lymph node (SLN) positivity in thin melanomas, of which primary tumor thickness and ulceration have been shown to be the most important factors associated with a positive sentinel lymph node. 4 Additional clinicopathologic factors, for which a prognostic value is less clear due to variability among studies, include Clark level, mitotic rate, lymphovascular invasion, anatomic site, tumor infiltrating lymphocytes, regression, and patient characteristics such as gender and age. [5][6][7][8][9][10][11][12] Interestingly, melanomas of Breslow thickness between 0.8 mm to 1.0 mm are considered to have the same risk for positive SLN regardless of any other associated risk factors, and the NCCN recommends considering SLNB in this population, along with thinner melanomas (< 0.8 mm) with ulceration 3 . However, if ulceration and other patient and tumor characteristics are independent predictive factors of progression and survival, we would expect that patients whose melanomas are associated with other risk factors would have a higher likelihood for regional metastasis. To our knowledge, no studies have been conducted regarding SLN predictors in this specific population. Therefore, we sought to determine predictive factors for performing SLNB and those predictive of positive SLN in this particular cohort.
After approval by the University Hospitals Cleveland Medical Center Institutional Review Board (IRB). Electronic records of patients with biopsy-proven melanoma were reviewed at our institution from January 1, 1997 through July 31, 2019. Patients who met criteria for Breslow thickness between 0.8 mm and 1.0 mm were initially identified through our electronic pathology record system, Copath. Medical records were then reviewed to identify patients who did and did not have a SLNB at the time of primary tumor excision. Patients were excluded if they had a previous history of other thicker melanomas or if no data on SLN outcomes were available. Patient demographics, primary tumor characteristics, and nodal status data were collected.

Statistical methods
Study covariates included in the analyses were age at diagnosis, gender, anatomic site, Breslow thickness, histologic type, dermal mitotic rate (MR), ulceration, regression, Clark level, lymphovascular invasion (LVI) and tumor infiltrating lymphocytes (TIL). Age at diagnosis was categorized into groups (< 60 years, ≥ 60 years). MR was grouped as < 1 or ≥ 1. Tumor site was designated as head and neck, trunk, upper extremities, lower extremities, or genitalia. Ulceration, regression, and LVI, and TIL were classified as being present or absent.
The primary aim of this study was to identify predictors for performing a SLNB. Multiple logistic regression was used to investigate associations between study covariates and METHODS whether a SLNB was performed. Estimates of odds ratios (OR) and 95% confidence intervals (CI) were provided to evaluate the effects of these associations. Using this analytical framework, we further investigated predictors for SLNB positivity (within SLNBperformed patient group). A p-value ≤ 0.05 was considered statistically significant. All statistical analyses were performed by using R version 3.6.2 (2019).
A total of 10,206 patients diagnosed with primary cutaneous melanoma were identified, of which 458 met criteria for Breslow thickness between 0.8 mm and 1.0 mm. A total of 87 patients were excluded, leaving a cohort of 361 patients available for analysis (Figure 1). Of these, 223 (62.8%) underwent SLNB and 138 (51.3%) did not ( Within the subset of patients that underwent SLNB, 20 (8.9%) were found to be positive and 203 (91.1%) negative ( Table 3). The majority of patients were male (53.8%), however the positive SLNB group was predominately composed of female patients (70.0%).
Univariate analyses found significant differences in patient gender, with females having an increased risk for positive SLNB (OR=3.66, 95% CI=1.34-11.67, p= 0.016) as compared to males. No other variables were significant. To our knowledge, this is the first study reporting predictors for performing SLNB in patients with thin melanomas, specifically those measuring 0.8-1.0 mm. Our results showed that younger age (<60 years), Breslow thickness (≥0.9 mm), mitoses (>1), and presence of TIL were all predictive factors for performing SLNB at our institution. The strong evidence in support of Breslow tumor thickness and mitoses as factors independently predictive of melanoma outcomes led to their incorporation into AJCC staging. 1 At our institution, we follow AJCC/NCCN guidelines to stage melanomas and provide recommendations. Although mitoses are no longer used in staging of thin melanomas, this was a risk factor used in previous AJCC staging editions and explains  why mitotic rate may have been considered when recommending SLNB in this specific group.

DISCUSSION
When management recommendations are not clearly stated in NCCN guidelines, we follow an evidence-based approach to provide appropriate recommendations. Although younger age (<60 years) is not used as part of melanoma staging per AJCC, this patient-specific factor has been shown to be an important independent predictor of melanoma outcomes 13   been demonstrated in several studies. [9][10][11][12] Our results show that younger age (<60 years) is a predictive factor for performing SLNB and thus aligns with currently published data demonstrating an increased risk for positive SLNB in this population.
Why the presence of TIL is a predictive factor for performing SLNB is less clear. There is conflicting evidence regarding the association between TIL and regional nodal metastasis in melanoma. Although some studies have reported a lower risk for positive SLNB in melanomas with TIL, 14-18 others have not shown such a relationship. [8][9][19][20][21] Absence of TIL has been demonstrated to be an independent predictor of SLN positivity. 14,22 It is unlikely that patients in our cohort underwent SLNB based on TIL alone given the available evidence. The results are possibly due to the relatively small number of cases. Further studies are needed to specifically determine the significance of TIL in thin melanomas, if any.
Although several studies have attempted to evaluate clinicopathologic predictors of SLN positivity in thin melanomas, no study has specifically looked at melanomas measuring 0.8 to 1.0 mm in thickness. Thus, we reviewed our cohort to determine which characteristics were predictive of SLN positivity in this specific population. Our results revealed that female gender significantly increases the odds of having a positive SLN. This correlates with the findings from Wright et al. 10 but contrasts with other studies in which male gender was found to increase regional metastatic risk. 9,23 Given the inconsistency among reports, no consensus currently exists regarding the association between gender and risk for metastasis in thin melanomas.
Ulceration has been reported to be an important risk factor in patients with thin melanomas. 7,9,[24][25][26][27][28][29][30][31] As a result, AJCC designates a more advanced stage for melanomas with Breslow thickness <0.8 mm with ulceration as compared to melanomas <0.8 mm without ulceration. Interestingly, melanomas of Breslow thickness 0.8-1 mm are given the same stage regardless of ulceration status. This grouped staging is likely the result of variability in data and lack of a strong correlation between ulceration and metastatic risk as thin melanomas become thicker. Our study, which focuses on this very specific group, revealed zero patients with ulcerated melanomas in the positive SLNB group. All patients with ulcerated melanomas (n=11) had a negative SLNB. Thus, no increased risk for SLN metastasis was found based on ulceration alone.
Other patient and tumor characteristics variably reported in the literature to increase the risk for positive SLN, including age, Breslow thickness, mitotic rate, Clark level, TIL, and regression, were not found to significantly increase the risk for positive SLNB in this group.
Limitations to our study include its retrospective and single institution nature, likely introducing selection bias, and a limited number of cases available for review, making it difficult to statistically assess for significant differences. Several other studies have encountered a similar limitation given that certain tumor traits, such as ulceration and SLN metastasis, in thin melanomas are uncommon. Moreover, we attempted to study survival outcomes in this group, but there were not enough cases with a positive SLN or survival data available for a significant statistical analysis.
In summary, our study represents the first analysis of predictors of performing SLNB and of increased risk for positive SLN in patients with melanomas of Breslow thickness between 0.8 and 1.0 mm. Our results revealed that younger age (<60 years), increased Breslow thickness (≥0.9 mm), mitotic rate >1, and presence of TIL were predictive factors of performing SLNB. Female gender was found to be the only statistically significant factor to increase the risk for positive SLN. We believe these findings can potentially improve management recommendations and enhance patient care.