Introduction
The incidence of malignant melanoma has increased worldwide over recent years and it is currently a significant public health problem (1,2). Ultraviolet radiation, which directly damages DNA, is the significant risk factor for the pathogenesis of melanoma (1,3). The early detection of melanoma and evaluation of melanoma tissue biomarkers are important for patient risk stratification, personalized diagnostics and treatment (4,5).
The current World Health Organization (WHO) classification of skin tumors subdivides melanoma on the basis of solar elastosis assessed by dermal elastic fibers, and measures cumulative sun damage (CSD) (3). According to WHO classification, there are currently 3 classes of melanomas: Those associated with high CSD, those associated with low CSD and nodular melanomas (3,6). Solar elastosis is usually apparent in superficially spreading and lentigo malignant melanoma, the so-called high CSD melanoma. Desmoplastic melanoma is associated with increased solar elastosis. The most common subtype of high CSD melanoma is superficially spreading melanoma, which usually begins with early radial growth followed by vertical growth and invasion of the dermis (3). Acral, mucosal, uveal and spitzoid melanomas are not associated with CSD, or are characterized by low CSD. Nodular melanomas are usually characterized as a low CSD type with early progression to vertical growth (3).
While the advent of novel personalized treatments of melanoma based on BRAF inhibitors and immunotherapies have reduced mortality rate over the last decade, advanced and metastatic melanomas still remain difficult to treat (7–10). Therefore, early diagnostic and risk stratification for the progression of melanoma is of particular importance. However, rare melanoma histopathological subtypes can make diagnosis challenging (3).
Therefore, the biomarkers of early-stage melanoma for the prediction of melanoma clinical behavior is of particular importance. It has been shown that such clinicopathological characteristics such tumor size, tumor type, tumor invasiveness (Breslow thickness, Clark level, lymphovascular invasion, neurotropisms), ulceration and tumor mitosis activity are significant prognostic factors for the development and progression of melanoma (3,6,11). In addition, it has been demonstrated that tumor-infiltrating lymphocytes could stratify melanoma with low and high risk progression (12–14).
The development of melanoma is closely related to somatic and epigenetic changes. Different mutations have been implicated in its pathogenesis and evolution. Recent genomic classification subdivides melanoma into 4 major subtypes based on the pattern of the most prevalent significantly mutated genes: mutant BRAF, RAS, NF1 and triple-WT (wild type) (15). Advances in molecular pathology and assessment of genetic biomarkers are increasingly used in clinical practice for diagnosis, personalized treatment and the prognosis of melanoma. Modern treatment guidelines are focused on the assessment of genetic biomarkers of melanoma (1,16,17).
The assessment of the BRAF gene mutation is of particular importance (3). BRAF mutations are observed in 40–60% of all primary malignant melanoma cases (16–20). The BRAF mutation is usually observed in younger patients, in non CSD skin and in superficial spreading melanoma, whereas NRAS mutational melanoma were characterized for nodular subtype and CSD skin (1,16–20). The BRAF gene is located on chromosome 7 and encodes a cytoplasmic serine-threonine kinase. BRAF plays a role in MAP-kinase (MAPK) pathway activation, contributing to cellular growth, differentiation, survival and proliferation (21). The mutations of the BRAF gene are generally located in codon 600 of the BRAF gene. The most common mutation observed in up to 90% of cases, resulted of transversion of T to A at nucleotide 1,799 position (T1799A). Less common mutations included substitutions of V for lysine (V600K), arginine (V600R) and leucine, V600M (22). Previous studies showed that the BRAF V600E mutation is usually observed in younger patients and at the body extremities, whereas V600K mutations are associated with older patients and usually found at the head and neck (14–19,23).
The RAS gene family includes genes that encode the G proteins which are responsible for: cell growth, proliferation and differentiation. RAS gene family consists of 3 main genes-NRAS, KRAS, and HRAS (15,19). The NRAS gene is most frequently mutated at hotspots in exon 2 (codons 12 and 13) and exon 3 (codon 61) (15,19). Recent evidence showed that in up to 20–30% of cases, NRAS mutations co-existed with BRAF mutations. Patients with both BRAF and NRAS mutations had poorer prognoses than those with the BRAF mutant melanoma alone (24–26). Generally, NRAS mutations are independent of BRAF mutations, but dual expression has been reported (25). The association of NRAS mutations with the degree of solar elastosis suggests that NRAS is closely related to the mutations induced by UV irradiation. Previous studies showed that the NRAS mutation is also associated with decreased immune responses in peritumoral melanoma tissue and a more advanced tumor stage (26). However, the association of NRAS mutational status with histopathological characteristics in early-stage melanoma is still poorly understood.
The current study's objective was to compare the NRAS and BRAF mutation status with the clinicopathological characteristics of patients with Stage IA-IIC melanoma.
Discussion
In the current study, genotype-phenotype associations were assessed in 118 patients with Stage IA-IIC malignant invasive melanoma, according to the AJCC classification. Histopathological examination of melanoma is currently a gold standard for the diagnosis of melanoma. In addition, such histopathological characteristics of invasive cutaneous melanoma as tumor size and type, lymphovascular invasion, ulceration, Breslow thickness, Clark invasion level, mitotic rate and disease are well established powerful prognostic and predictive factors for melanoma (3,6). The development of melanoma is closely related to somatic and epigenetic changes. Activated mutations of the oncogenes BRAF and NRAS are of particular importance in melanoma progression (15–20). BRAF personalized treatment of melanoma significantly improved patients' prognosis (18–20), however, the prognostic value of BRAF and NRAS mutation and its association with clinical and histopathological characteristics is still controversial, especially in early-stage melanoma.
Our study showed that BRAF mutations and NRAS mutations were identified in 56.8 and 29.6% of cases respectively. Furthermore, while BRAF mutational status was not associated with PFS, the NRAS mutational status did significantly correlate with PFS. In patients with BRAS and NRAS co-mutant melanoma, the PFS was significantly poorer compared to the BRAF mutant melanoma.
BRAF mutations in primary melanomas have been observed at a rate of 22–72% (14–20). In our study, the frequency of BRAF mutation falls within this range. Over 90% of the mutations in BRAF result in substitution of the valine at position 600, resulting in activation of the downstream effectors of the RAS-RAF-MEK-MAPK pathway (1). The associations of BRAF V600 mutational status and Breslow thickness, patient gender, Breslow thickness and peritumoral lymphocytes infiltration was revealed, supporting our previous evidence (14). Previous studies showed the different value of BRAF mutational status in association with clinicopathological characteristics and PFS of melanoma (27–33). Our study did not find any association between the BRAF mutational status and PFS of melanoma.
It has been demonstrated that BRAF mutational status correlates to younger age groups and females (34,35). However, some studies also demonstrated associations with the male gender (29). In our study BRAF mutation correlated to female gender and older age. This observation could be explained by the fact that only early-stage melanoma patients were enrolled in our study. Some studies showed the importance of immunological tolerance mechanisms in the development of BRAF mutant melanoma (14,36). This study confirmed our previous results which showed that TIL infiltration is associated with BRAF mutational status (14,37). It seems that the assessment of TIL is beneficial for the risk stratification of melanoma and therefore it should be included in the routine histopathological assessment of melanoma.
Previous studies of BRAF and NRAS com-mutant melanoma have been discordant. NRAS gene mutation was found in 15–25% of melanoma cases (27,38). In our study NRAS mutation was observed in up to 30% of melanoma cases.
We assume therefore that the high prevalence of NRAS mutations could be explained by the older median age of the enrolled patients in our study, e.g., 67 years. It has been demonstrated that patients with NRAS mutant melanoma compared with BRAF mutant melanoma were usually older (>55 years) with a previous history of UV exposure. NRAS mutant melanoma is commonly found in upper extremities and characterized by increased Breslow tumor thicknesses (1,27,38).
These results showed that NRAS mutations are associated with Breslow thickness, nodular melanoma tumor type, mitotic activity and lymphovascular invasion. It was assumed that NRAS mutations in primary stage IA-IIC melanoma could have potentially relevant predictive value. The NRAS gene is most frequently mutated at hotspots in exon 2 (codons 12 and 13) and exon 3 (codon 61) (38). The NRAS mutation characteristic for nodular melanoma is localized in sun-damaged skin (39).
Nevertheless, the value of NRAS mutation on disease progression and prognosis is still controversial. Some studies showed that the NRAS mutation was associated with a favorable prognosis (40). In contrast, other studies demonstrated that NRAS gene mutation was associated with a poorer prognosis (38,41,42). Other studies did not find any significant association between the NRAS mutation and the prognosis of melanoma (37,43,44).
Similarly in stage IV melanoma, the data for the NRAS mutation is also controversial. While one study suggested that the NRAS-mutated tumor genotype in metastatic Stage IV melanoma was associated with increased overall survival compared to the BRAF-mutated and WT tumor genotypes (40). Other studies had the opposite results and did not support this evidence (43,44).
It has been demonstrated that NRAS mutation status was an independent predictor of shorter survival after a diagnosis of stage IV melanoma (45). It could be suggested that molecular mechanisms involving NRAS genetic pathway could be different between metastatic Stage IV and early-Stage IA-IIC melanoma. These results, in line with previous studies, demonstrated that NRAS mutations are associated with higher Breslow's thickness and poor disease prognosis (38,41,42).
In addition, our study showed that NRAS mutations are associated with increased mitotic activity of the tumor and lymphovascular invasion, which could be one of potential explanations of the aggressive behavior of those tumors which carried a NRAS mutation. Furthermore, it was demonstrated that NRAS mutational status in primary Stage IA-IIC melanoma is a powerful predictive factor, significantly associated with progression free survival.
NRAS personalized treatment of melanoma is challenging. Lonafarnib and tipifarnib have been studied for NRAS mutant melanomas (1). In addition, selective MEK inhibitors could have potential benefit in the treatment of NRAS mutant melanoma (45). However, the potential significant value of our study is in the finding of a significant predictive value of NRAS mutations for Stage IA-IIC melanoma. Therefore, routine assessment of NRAS mutations in Stage IA-IIC melanoma could be potentially beneficial for the prediction of disease progression. It should be stressed that our study subjects included only those with local disease, e.g., at the time of diagnosis patients did not have local recurrence, regional lymph node metastasis or distant metastasis.
The value of the current study is in the demonstration that Stage I and II NRAS mutant melanoma is characterized by poorer progression free survival and is associated with histopathological characteristics responsible for tumor growth such as Breslow thickness, mitotic activity and lymphovascular invasion. Further research for patients with advanced and metastatic melanoma should evaluate the role of BRAF and NRAS mutational status in disease progression.
Several limitations of our study should be mentioned. A significantly higher number of case-cohort with equal gender distribution would be beneficial. At the same time, the strength of the present study was the demonstration of significant role of NRAS mutational status in patients with early-stage IA-IIC non-metastatic melanoma. All patients were enrolled from the single oncology hospital, which treats up to 85% of all melanoma cases in Latvia.
In our study BRAF and NRAS mutation status was assessed by digital droplet PCR (ddPCR) using BRAFV600, NRAS Q61 and NRAS G12/G13 Screening Assay. The PCR testing is the gold standard for BRAF and NRAS mutation testing according to ASCO and CAP protocols. The immunohistochemistry is cost-effective method compared to PCR and the value, specificity, and sensitivity of BRAF and NRAS immunohistochemistry should be addressed in future studies for general melanoma testing.
In conclusion, the patients with NRAS and NRAS/BRAF co-mutant Stage IA-IIC melanoma had poorer progression free survival when compared to the NRAS wild and BRAF mutant melanomas. The NRAS assessment in melanoma in routine clinical practice is beneficial for the risk stratification of disease progression. Our results highlighted the value of NRAS personalized treatment in patients with invasive melanoma.