The frequency and clinicopathological significance of NRAS mutations in primary cutaneous nodular melanoma in Indonesia

Abstract Background Melanoma is a lethal skin malignancy with a high risk of metastasis, which prompts a need for research on treatment targets and prognostic factors. Recent studies show that the presence of neuroblastoma RAS viral oncogene homolog (NRAS) mutation can influence cell growth in melanomas. The NRAS mutation, which stimulates the mitogen‐activated protein kinase (MAPK) signaling pathway, is associated with a lower survival rate. However, evidence from Indonesia population is still very rare. Further understanding of the role of NRAS mutations in Indonesian melanoma cases will be crucial in developing new management strategies for melanoma patients with NRAS mutations. Aims To explore the frequency of NRAS mutations and their clinicopathological associations in patients with primary nodular cutaneous melanoma in Central Java and Yogyakarta, Indonesia. Methods and results Fifty‐one paraffin‐embedded tissue samples were collected from primary nodular skin melanoma cases between 2011 and 2019 from the two largest referral hospitals in Yogyakarta and Central Java, Indonesia. The NRAS mutation status was evaluated using qualitative real‐time polymerase chain reaction (qRT‐PCR). The association of NRAS mutation was analyzed with the following: age, gender, location, lymph node metastasis, ulceration, mitotic index, tumor‐infiltrating lymphocytes (TILs), necrosis, tumor thickness, lymphovascular invasion (LVI), and tumor size. NRAS mutations were detected in 10 (19.6%) samples and predominantly observed (60%) in exon 2 (G12). These mutations were significantly correlated with lymph node metastases (p = .000); however, they were not associated with other variables analyzed in this study. Conclusions The prevalence of NRAS mutations in primary nodular cutaneous melanoma cases from Indonesia is consistent with previous studies and is significantly associated with increased lymph node metastases. However, the predominant mutation detected in exon 2 (G12) is different from previous studies conducted in other countries. This suggests that melanoma cases in Javanese people have different characteristics from other ethnicities.


| INTRODUCTION
Although considered rare compared with basal and squamous cell carcinomas, melanoma is a lethal skin malignancy that has a high risk of metastasis. The incidence rate of skin melanoma varies throughout the world. In western countries, melanoma is relatively frequent, especially in light-skinned populations. Recent research shows that the highest incidences are in Queensland, Australia, and Auckland, New Zealand. 1 The incidence of melanoma is low in Asian populations, which is approximately 0.25 per 100 000 population in 2018, according to the Global Cancer Observatory. 2 The prognosis of cutaneous melanoma is influenced by several clinical and histopathological factors. Clinical factors include age, sex, and anatomic location. Histopathological factors consist of Breslow tumor thickness, ulceration, Clark's anatomic level, tumor volume, growth patterns, mitosis number, radial and vertical growth phase, regression, tumor vascularity, lymphovascular invasion (LVI), angiotropism, histological type, and tumor-infiltrating lymphocytes (TILs). 3 Recent research shows that cell growth in melanomas is also influenced by the NRAS gene. The presence of NRAS mutations stimulates the RAS-RAF-extracellular signal-regulated kinase/mitogenactivated protein kinase (MAPK) signaling pathways that can interfere with cell cycle regulation and prosurvival pathways and increase cell proliferation. 4 NRAS mutations occur in 15%-25% of melanoma cases, most often in exons 2 (codon 12) and 3 (codon 61). The presence of NRAS mutations was associated with a lower survival rate. 5 Until now, no prognostic marker has been validated for melanoma. Further understanding of melanoma with NRAS mutations will be crucial in developing new management strategies for melanoma patients with NRAS mutations. 6 The study of cutaneous nodular melanoma in Asia has been scarce because of its uncommon nature. Although the frequency is low, nodular melanoma is a major contributor to skin cancer-related mortalities. One study conducted in Australia found nodular melanomas represented 14% of the invasive melanomas and shockingly caused 43% of the deaths. 7 This poor outcome implores further elucidation on this particular type of melanoma.
Further research concerning NRAS mutations and its associations is needed because of the lack of evidence in Indonesia. This study aimed to examine the frequency of NRAS mutations and their associations with clinicopathological properties in patients with primary nodular cutaneous melanoma in Central Java and Yogyakarta, Indonesia.

| METHODS
A retrospective cross-sectional study was conducted in the Department of Anatomical Pathology Dr. Sardjito Hospital, Sleman, Yogyakarta and Dr. Soeradji Tirtonegoro Hospital, Klaten, Central Java. Both hospitals are the major referral hospitals in Yogyakarta and Central Java Province, Indonesia. We collected and analyzed 51 paraffin-embedded tissue samples from primary nodular cutaneous melanoma cases from 2011 to 2019. The ethnicity of the patients was Javanese, which is one of the major ethnic groups in Indonesia.
The existence of NRAS mutation was examined using qualitative real-time polymerase chain reaction (qRT-PCR). The DNA source was The clinicopathological data, including age, sex, anatomic location, lymph node metastasis, ulceration, mitosis number, TILs, necrosis, tumor thickness, LVI, and tumor size, were collected. Hematoxylineosin stained slides were examined under a microscope for ulceration, tumor thickness, lymph node metastasis, LVI, necrosis, and TILs. The existence or lack of lymph node metastasis was evaluated by examining the lymph node biopsy. Tumor thickness was estimated from the granular layer to the deepest invasion of the tumor and subsequently categorized as ≤4 or >4 mm. Ulceration was defined as the combination of a full-thickness epidermal defect (including the absence of stratum corneum and basement membrane), evidence of host response (i.e., fibrin deposition and neutrophils), and thinning, effacement, or reactive hyperplasia of the surrounding epidermis. 8 LVI was defined as the presence of tumor cells within the blood vessel and/or lymph vessel, encasing the tumor identical to the primary cutaneous melanoma cells. Necrosis was defined as the existence of an area of necrotic cells occupying at least one-fourth high-power field (0.07 mm 2 ). TILs were defined as migrating lymphocytes from the blood vessels to the surrounding tumor stroma and categorized into present or absent. Non-brisk and brisk TILs are classified into present category. Tumor size was measured from the length of the larger axis in millimeters after formalin tissue fixation. 9,10 Immunohistochemistry was performed using a manual method from ScyTek Laboratories (ScyTek Laboratories Inc., Utah, USA). In this study, we used the CellMarque Ki-67 (SP6) monoclonal antibody (Sigma-Aldrich Co., Oakville, Canada). Tonsil tissue slides were used as a positive control. Paraffin blocks were sliced as thick as 3 μm, incubated, deparaffinized, and rehydrated. Antigen retrieval was conducted using Tris EDTA solution (Vivantis Inc., California, USA) at pH 9 for 20 min at 95 C. Slides were submerged for 15-20 min in 3% hydrogen peroxide, soaked in primary antibody for 60 min, labeled with horseradish peroxidase polymer for 10-20 min and diaminobenzidine for 3 min. Finally, slides were counterstained using hematoxylin for 0.5-1 min and then covered by the coverslips. The mitotic index was counted based on the percentage of positively stained nuclei per 1000 tumor cells and then categorized as <20% and ≥20%. The associations between NRAS mutation status and clinicopathological parameters were analyzed using the chi-squared or Fisher's exact tests.
The association between NRAS mutation and the clinicopathological parameters is shown in Table 1. Representative results of Ki-67 proliferative index measurement are shown in Figure 1. NRAS mutation was highly correlated with lymph node metastasis (p = .000). No significant association was found between NRAS mutation and other variables analyzed in this study.
The association between NRAS mutation and clinicopathological parameters 6 (12) 29 Previous studies demonstrated that NRAS mutations occur more commonly in nodular subtype melanomas than in other subtypes and often develop in the skin with continuous ultraviolet exposure. 20 In this study, all tumor samples were from Asian races, specifically Javanese ethnicity, which is the most common ethnic group in Indonesia. Since Indonesia is located on the equator, high-intensity sunlight is continuously present throughout the year, which may cause skin damage because of the high cumulative sun damage (CSD), particularly in older people. Melanomas that arise due to sun exposure are generally dominated by NF1 and NRAS mutations. 30,31 The sun exposure is also related to the acral lentiginous type of melanoma, which generally has a thin tumor thickness. 32 This study also showed that NRAS mutation was predominantly present in exon 2 (G12). This finding contradicts some previous studies that demonstrated that the predominant site of NRAS mutation is in exon 3 (Q61). 5 more common in mucosal melanomas and NRAS exon 3 (Q61) mutations in nodular and superficial spreading melanomas. 33 The Bucheit study sample numbered 136, but only 58 (42%) samples were a combination of nodular type melanoma and superficial spreading, and 50 samples (86%) were with mutations in exon 3 (Q61). 33 In this study, we used samples from nodular type melanoma only.
In this study, we found a significant association between lymph node metastasis and NRAS mutation. This finding is in concordance with the results of the study by Sheen et al that revealed that melanoma with NRAS mutations has the nature and pattern of aggressive growth. This group of melanoma tends to develop lymph node and even distant metastases. 13 Sentinel lymph node (SLN) biopsy is a mandatory procedure to achieve sufficient data about lymph node metastasis even with a small number of samples. SLN positivity itself has been accepted as a prognostic factor for tumor recurrence and poor prognosis. 34 NRAS-mutant tumors are likely to act aggressively, especially in the early stages in the high-risk melanoma population. In our study, 45 (88%) patients had lymph node metastasis. Accordingly, most of the patients had a higher risk of recurrence and poor prognosis. All patients with NRAS mutation had lymph node metastasis, proving that this mutation has an important role in more aggressive tumor behavior.
We found a predominance of female patients in our study. Some studies show that women who suffer from melanoma have a good prognosis, even patients with lymph node metastasis. The gender influence in the prognosis of the disease is related to differences in thickness, ulceration, and anatomic location of the melanoma between women and men. The presentation of melanoma is thinner, and ulceration is less common in women than in men. 35 Thicker tumors tend to have higher mitotic rates. 36 This study demonstrated no statistically significant association between NRAS mutations and anatomic locations. In a previous study by Sheen et al in Taiwan with 119 samples, the extremities were the most common location for melanomas. 13 Lee et al also found the same finding in the United Kingdom with 1972 patients observed. 37 NRAS mutation was significantly associated with tumor location, especially the extremities. 35 In this study, we found no significant between the existence or lack of NRAS mutation and age, gender, and anatomical site. This result differs from the previous study by Sheen et

CONFLICT OF INTEREST
The authors declare that they have no competing interests.

DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available on request from the corresponding author upon reasonable request.