The molecular biology of melanoma

INTRODUCTION — Melanoma is among the most common cancers in the United States, causing approximately 98,000 new cases and 8000 deaths annually [1]. Worldwide, melanoma accounts for approximately 300,000 new cases and 60,000 deaths annually [2]. (See "Melanoma: Epidemiology and risk factors", section on 'Epidemiology'.)

A majority of patients with localized disease can be treated successfully with surgical resection. Individuals who present with or subsequently develop distant metastases are typically treated with systemic therapy, such as checkpoint inhibitor immunotherapy or molecularly targeted therapy. (See "Overview of the management of advanced cutaneous melanoma".)

An increasing understanding of melanocyte biology and melanoma pathogenesis has led to the development of targeted therapies and has the potential for continued major improvements in the care of patients with advanced melanoma. The most important breakthroughs have been the discovery of the mitogen-activated protein kinase (MAPK) pathway as the key signaling pathway, and the critical role of microphthalmia-associated transcription factor (MITF).

This topic provides an overview of melanocyte biology and the important molecular alterations of genes and signaling pathways that are critical to the development and pathogenesis of melanoma. The clinical results of targeted agents that are developed based upon this information are discussed separately. (See "Systemic treatment of metastatic melanoma with BRAF and other molecular alterations".)

MELANOCYTE BIOLOGY — Melanocytes are the pigment-containing cells of the skin that produce melanin in response to stimuli such as ultraviolet (UV) radiation. Melanocytes also make the pigment that determines skin and hair color.

Melanocytes are derived from pluripotent neural crest stem cells. Melanocyte development is modulated by KIT and microphthalmia-associated transcription factor (MITF), factors that are mutated or amplified oncogenes in a fraction of melanomas [3].

Alpha-melanocyte stimulating hormone (MSH) is a propigmentation peptide that stimulates the melanocortin 1 receptor (MC1R) on the surface of melanocytes [4]. The MC1R gene exists in numerous variant (polymorphic) forms; nonsignaling variants produce the red hair/fair skin phenotype, which is associated with an increased risk of melanoma and nonmelanoma skin cancers [5].

Intracellular cAMP induction downstream of MC1R activates the expression of MITF, which stimulates transcription of numerous genes that participate in the conversion of tyrosine into melanin pigments. Stimulation of cutaneous pigmentation following UV exposure (tanning) is thought to occur via keratinocyte DNA damage, p53-mediated induction of proopiomelanocortin (POMC)/MSH expression, followed by secretion of MSH and stimulation of MC1R in epidermal melanocytes [6].

Cutaneous UV exposure may induce behavioral effects, perhaps related to endorphin components of the POMC/MSH complex, which may produce an addiction to sun bathing and, therefore, contribute to the increase in melanoma incidence observed over recent decades [7]. "Rescue" of this cAMP-MITF-pigment cascade has been achieved via topical drug delivery in mouse models and confers protection against UV-induced carcinogenesis [8,9].

MAPK PATHWAY — The mitogen-activated protein kinase (MAPK) pathway is activated in almost all melanomas (figure 1) [10]. In nonmalignant cells, the interaction between a growth factor receptor and its ligand is required to activate this pathway. This leads to a series of events that promote cellular growth and survival. The RAS family members are G-proteins, which serve as critical mediators in the transduction of such signals.

Results of The Cancer Genome Atlas (TCGA) and other large scale genomic analysis efforts in melanoma have identified hot spot mutations in NRAS in 25 to 30 percent of cutaneous melanomas, which are thought to be important drivers of oncogenesis [11,12]. A somatic mutation in the NRAS gene can cause constitutive activity of the NRAS protein, which is thus incapable of being "turned off." This leads to the serial activation of serine/threonine kinases, which promotes cell cycle progression, cellular transformation, and increased cell survival. In addition, this cascade of events may be mediated through the overexpression and/or hyperactivation of various growth factor receptors, such as c-Met, epidermal growth factor receptor (EGFR), and KIT, as well as through the loss of function of the NF1 tumor suppressor gene that acts to suppress NRAS signaling [13-16]. In fact, the NF1 mutation is considered a third major genomic subset of cutaneous melanomas, behind BRAF and NRAS mutations, and is present in 13 to 14 percent of cases [11,12]. These subsets are largely but not entirely mutually exclusive, and cutaneous melanomas may harbor activating alterations in more than one of these genes (eg, coalterations in NF1 and NRAS) [11].

The most important downstream mediators of activated RAS are the serine/threonine kinases BRAF and CRAF, which are activated following RAS binding (figure 1) [17,18]. In contrast to BRAF, activation of the pathway by CRAF requires additional steps, whereas activation of BRAF alone is sufficient. This may explain why an activating mutation in the BRAF kinase domain is present in 40 to 50 percent of patients with melanoma, but there are no descriptions of activating mutations of CRAF [19-21].

Following activation, RAF, typically in the form of either a homo- or heterodimer, interacts with the MAPK/extracellular signal-regulated kinase (ERK) kinase MEK, thereby initiating MEK phosphorylation, which in turn leads to an activating phosphorylation of ERK, its only known substrate [17,22,23]. The activation of ERK leads to a pro-growth and transforming signal, through its interaction with a number of molecules, which appears to be critical to the pathogenesis of many malignancies.

Molecular characterization of melanocytic lesions — BRAF mutations appear to be an acquired event that occurs in early invasive melanoma and leads to clonal expansion and tumor progression. The BRAF mutation, thus, is not a founder event but rather facilitates malignant transformation with the acquisition of subsequent oncogenic stimuli. Evidence supporting this interpretation includes the following:

●BRAF mutations are common in melanocytic nevi, vertical growth phase melanomas, and metastatic melanoma (70 to 80, 40 to 50, and 40 to 50 percent, respectively). However, BRAF mutations are rarely detected in radial growth phase melanomas or in situ melanoma (10 and 6 percent, respectively), which are thought to be the initial malignant lesions prior to development of frankly invasive lesions [21,24-26].

●Polyclonality (ie, BRAF mutated cells mixed with BRAF wild-type cells) has been observed in both atypical nevi and primary melanomas, though such polyclonality has not been seen in individual metastatic tumors nor in tumors from distinct sites in an individual patient [10,27,28].

These findings can be viewed as contradicting the popularly held belief that BRAF mutation precedes all other oncogenic events in BRAF mutant melanoma, based primarily on the fact that BRAF mutations are present in 70 to 80 percent of dysplastic nevi [24]. However, the precise relationship between nevi and melanomagenesis remains incompletely understood. Further elucidation of the mechanisms of transformation in BRAF mutant melanoma is needed. Irrespective of the mechanism, however, when these BRAF mutations occur in invasive melanoma, the resultant constitutive activation of MEK (and subsequently ERK) leads to oncogenesis through the promotion of cellular growth and opposition of apoptosis, as well as to the strict reliance of these cells upon the signaling cascade [29].

Genetic abnormalities in melanoma — Genomic mutations and/or aberrations are present in the majority of melanomas [10-12,30]. These can lead to activation of the MAPK pathway and create an oncogenic addiction to the mutated or hyperactivated protein. Within this pathway, the patterns of mutation frequencies for each subtype differ significantly [30-38].

Cutaneous melanoma and precursor lesions — In cutaneous melanoma, BRAF and NRAS mutations are the most frequently observed (40 to 50 and 15 to 20 percent of cases, respectively) [21,39-42]. BRAF and NRAS mutations are more common in cutaneous melanomas without chronic sun damage (CSD) compared with those associated with CSD (60 versus 6 to 22 percent and 20 to 22 versus 0 to 15 percent, respectively), whereas mutations in c-kit are more common in CSD skin than in non-CSD skin (15 to 30 versus <1 percent in White individuals and up to 11 percent in Asian populations) [30,43,44]. Except in rare cases and in the setting of BRAF inhibitor therapy resistance, activating BRAF, NRAS, and c-kit mutations are mutually exclusive [45]. Additionally, BRAF V600E mutations are more common in females, younger patients, and in non-CSD skin than the second most common BRAF mutation BRAF V600K [46,47].

Analysis of adjacent precursor lesions (nevi) together with invasive melanomas helped to assign information regarding the order of acquisition of melanoma mutations [48]. Unequivocally, benign nevi were seen to contain BRAF V600E in all cases studied. Additionally, 77 percent of intermediately aggressive precursors or frankly invasive lesions contained telomerase reverse transcriptase (TERT) promoter mutations. Loss of both cyclin-dependent kinase inhibitor 2A (CDKN2A) alleles and occurrence of genomic copy number alterations were observed only in invasive melanomas.

Classification of BRAF mutations — Our understanding of BRAF signaling has improved, leading to a more accurate but complex categorization of MAPK driver alterations in cutaneous melanomas beyond BRAF, NRAS, and NF1 [49]. Physiologic BRAF signaling in melanocytes occurs through ERK via BRAF-BRAF homodimers and BRAF-CRAF heterodimers. In turn, ERK serves as an important feedback regulator to BRAF via upstream inhibition of RTKs and RAS-mediated BRAF/CRAF heterodimer signaling to maintain homeostasis.

Studies of BRAF signaling have identified three main classes of oncogenic BRAF alterations that alter this relationship and achieve RAS-independent growth signaling in distinct, therapeutically relevant ways [50,51]. Among cutaneous melanomas with BRAF mutations, a majority are class 1 alterations (90 to 95 percent), although additional activating BRAF (class 2 and 3) alterations have also been described (5 to 10 percent).

●Class 1 alterations – Class 1 alterations are BRAF V600 alterations to -E, -K, or less often -R and represent the most common single MAPK driver alteration in cutaneous melanomas. Other rare BRAF V600 amino acid mutations (eg, M-, D-, and G-) are also considered to be class 1 alterations. (See "Systemic treatment of metastatic melanoma with BRAF and other molecular alterations", section on 'BRAF V600 mutation variants'.)

This mutation results in BRAF monomers (rather than physiologic dimers) that signal downstream through MEK and ERK and are insensitive to ERK-mediated feedback inhibition of RAS-mediated BRAF homo- and heterodimerization. Inhibitors of both BRAF monomers (vemurafenib, dabrafenib) and MEK (trametinib) have been shown to prolong overall survival in patients with cutaneous, derived, advanced melanoma containing the mutations in BRAF V600. Furthermore, combination therapy with BRAF and MEK inhibitors (dabrafenib plus trametinib, vemurafenib plus cobimetinib, encorafenib plus binimetinib) is more effective than single-agent BRAF inhibitor therapy, without a substantial increase in toxicity. (See "Systemic treatment of metastatic melanoma with BRAF and other molecular alterations", section on 'Choice of BRAF plus MEK inhibitor therapy'.)

●Class 2 alterations – Class 2 alterations, which include missense mutations such as K601E and D597N as well as oncogenic BRAF fusions, alter the RAS-binding domain such that they are insensitive to physiologic inhibition by ERK similar to class 1 alterations; however, they signal efficiently as mutant homodimers [50]. As a result, BRAF monomer inhibitors like vemurafenib would paradoxically activate class 2 mutant homodimer signaling. Clinical trials are ongoing in the use of inhibitors of BRAF dimer signaling in advanced melanomas.

●Class 3 alterations – Class 3 alterations include BRAF D594N, which abrogates the kinase activity of BRAF signaling and leads to low ERK signaling. Its oncogenic potential lies in the loss of feedback inhibition of upstream RTKs rather than high BRAF kinase activity seen in class 1 and 2 lesions [51]. As a result, BRAF class 3 mutant melanomas are frequently coaltered with upstream RTK or RAS drivers of the MAPK pathway such as NRAS, KIT, or NF1. Successful targeted inhibition of these tumors may lie in targeting the upstream "amplified" driver activity.

Uveal melanomas — The molecular pathogenesis of uveal melanoma is distinct from that of cutaneous melanoma and other melanoma subtypes, including conjunctival melanoma. Advances in understanding the molecular pathogenesis of uveal melanoma may eventually provide important opportunities for targeted therapy in patients with metastatic disease. (See "Metastatic uveal melanoma".)

Overview — The genomic landscape of uveal melanoma is characterized by a small number of highly recurrent mutations arranged in two clusters [52].

●Mutations without prognostic relevance – An early cluster of mutually exclusive initiating mutations in GNAQ, GNA11, PLCB4, and CYSLTR2 are present in >95 percent of uveal melanomas and are not prognostically relevant. (See 'Mutations without prognostic relevance' below.)

●Mutations with prognostic relevance – By contrast, a second cluster of near mutually exclusive "progression mutations" in BAP1, SF3B1, and EIF1AX are present in patients with primary uveal melanoma, appear to have prognostic significance, and are used to determine post-treatment surveillance. (See 'Mutations with prognostic relevance' below and "Initial management of uveal and conjunctival melanomas", section on 'Posttreatment systemic surveillance'.)

•BAP1 – BAP1 mutations occur in almost 50 percent of patients (with 2 to 3 percent of patients harboring germline BAP1 mutations) and are associated with the highest risk of metastases [31,32,53].

•SF3B1 – SF3B1 mutations are present in approximately 19 to 30 percent of patients and are associated with an intermediate risk of metastases [52,54,55].

•EIF1AX – EIF1AX mutations occur in approximately 15 to 24 percent of patients and are associated with the lowest risk of metastases [33,52,56-58].

Importantly, uveal melanomas rarely if ever harbor mutations in BRAF, NRAS, KIT, or other recurrent alterations seen in cutaneous melanoma [31,32]. (See "Systemic treatment of metastatic melanoma with BRAF and other molecular alterations".)

While most uveal melanomas have a low mutational burden [52,58], a small subset with germline mutations in the DNA base excision repair factor MBD4 have a high mutation burden and may be more likely to respond to checkpoint inhibitor immunotherapy [59]. (See "Tissue-agnostic cancer therapy: DNA mismatch repair deficiency, tumor mutational burden, and response to immune checkpoint blockade in solid tumors", section on 'Tumors with high mutational burden'.)

Mutations without prognostic relevance — Mutually exclusive mutations in GNAQ, GNA11, PLCB4, and CYSLTR2 are frequently present in uveal melanoma, but lack prognostic significance.

●GNAQ and GNA11 – In patients with uveal melanoma, the MAPK pathway is upregulated by activating point mutations in either of the homologous G-proteins GNAQ or GNA11, which together are observed in over 90 percent of uveal melanomas [31,32,53-55]. These mutations are either at the Q209 position of exon 5 or, less frequently, at the R183 position in exon 4. These mutations, while common in uveal melanoma, are not prognostically relevant.

The oncogenic potential of these mutations and their ability to activate the MAPK pathway through the upregulation of protein kinase C (PKC) has been confirmed in murine models. Additionally, the potential importance of this pathway in patients with uveal melanoma was illustrated by a series of 187 patients with uveal melanocytic tumors [32]. Somatic mutations were identified in approximately 80 percent of cases. In 139 blue nevi, a cutaneous lesion associated with uveal melanoma, mutations were present in over 60 percent of cases. Of potential importance, GNAQ mutations were proportionally more common in blue nevi and primary uveal melanomas, while GNA11 mutations predominated in metastatic uveal melanoma deposits.

By contrast, these somatic mutations in GNAQ and GNA11 were absent in all nine cases of conjunctival melanoma, and a somatic mutation in these genes was identified in only 1 of 273 cases of extraocular melanoma (0.4 percent).

Additional studies on agents targeting the MAPK pathway in uveal melanoma are discussed separately. (See "Metastatic uveal melanoma", section on 'Molecularly targeted agents'.)

●PLCB4 and CYSLTR2 – Most of the remaining tumors without mutations in GNAQ and GNA11 harbor mutations in other members of Gaq signaling pathway, including the downstream effector of Gaq signaling PLCB4 [60], and the G protein coupled receptor CYSLTR2 [61]. Mutations in PLCB4 and CYSLTR2 are also mutually exclusive of those in GNAQ and GNA11 [62]. These mutations are also not prognostically relevant.

A recurrent hotspot mutation of CYSLTR2 at position 129, a substitution of leucine for glutamine, is associated with gain of function of the G-protein coupled receptor [61]. Mutations in CYSLTR2 are present in 3 to 4 percent of uveal melanomas [61,62] and are mutually exclusive of GNAQ and GNA11 mutations.

Together, mutations in either a G-protein coupled receptor or the alpha-subunits of a G-protein coupled receptor occur in over 96 percent of patients [62].

Mutations with prognostic relevance — Mutually exclusive "progression mutations" in BAP1, SF3B1, and EIF1AX appear to have prognostic significance. These mutations are used to determine the post-treatment surveillance of patients with primary uveal melanoma. (See "Initial management of uveal and conjunctival melanomas", section on 'Posttreatment systemic surveillance'.)

●BAP1 – The BAP1 gene is a nuclear deubiquitinase located on chromosome 3p21.1 that functions as a tumor suppressor through multiple incompletely understood mechanisms [63]. Mutations in the BAP1 gene are identified in almost half of primary uveal melanomas and are associated with the highest risk of metastatic disease.

Inactivating mutations in BAP1, associated with loss of the other allele through whole chromosome 3 loss, were initially identified in up to 50 percent of primary uveal melanomas and in the vast majority of metastasizing class 2 uveal melanomas (including one patient with a germline BAP1 mutation) [32]. Subsequent population-based studies have indicated that about 2 to 3 percent of all patients with uveal melanoma harbor germline BAP1 mutations [31,53].

Monosomy of chromosome 3 has been associated with the development of metastatic disease in patients with uveal melanoma but is rare in patients with cutaneous melanoma [64-69]. As an example, genetic analysis of fine needle aspirates found that partial or complete monosomy of chromosome 3 was present in 27 and 25 percent of cases in a series of 500 uveal melanomas [69]. The presence of complete monosomy was associated with significantly poorer survival at three years.

Somatic mutations in the gene encoding BAP1 on chromosome 3 have been identified in 26 of 31 uveal melanomas (84 percent), with a gene expression profile consistent with a high metastatic risk (greater than 80 percent at five years) [33]. By contrast, mutations in BAP1 were present in only 1 of 26 patients (4 percent) classified as being at low risk for metastases. Loss of function of this tumor suppressor gene is thought to result from the combination of a somatic mutation in one allele and the loss of one chromosome 3 allele.

●SF3B1 and EIF1AX – Mutations in SF3B1 and in EIF1AX are associated with a more favorable prognosis and late onset of metastasis.

•SF3B1 – SF3B1 encodes the splicing factor 3B subunit 1, which is involved in pre-messenger-RNA splicing. Change-of-function point mutations occurring mostly at codon 625 are present in approximately 19 to 30 percent of primary uveal melanomas [52,54-58] and were associated with splicing defects that contribute to tumor progression by poorly understood mechanisms [60,61]. SF3B1 mutations are mostly mutually exclusive with BAP1 mutations and are associated with a lower rate and later median onset of metastatic disease [33,70].

•EIF1AX – EIF1AX encodes the X-linked eukaryotic translation initiation factor 1A, which stimulates transfer of methionine transfer RNA to the small ribosomal subunit. Somatic mutations in EIF1AX occur in approximately 15 to 24 percent of primary uveal melanomas, and mostly mutually exclusive with BAP1 and SF3B1 mutations [33,52,58]. Uveal melanomas containing mutations in EIF1AX, but not in BAP1 or SF3B1, are associated with the best prognosis among uveal melanoma prognostic subgroups [33]. Most mutations in EIF1AX involve small, in-frame alterations affecting the N terminus of the protein that function in tumorigenesis by poorly understood mechanisms [55].

SF3B1, EIF1AX, and BAP1 mutations are almost mutually exclusive, indicating alternative molecular pathways leading to distinct clinical outcomes. For example, in one study, when compared with wild type, SF3B1 mutation was associated with a lower rate of progression to metastases (progression-free survival of approximately 80 percent at five years) [58]. It is not known how much of this improvement is due to the presence of these mutations versus the absence of the BAP1 mutation. Further studies are required to validate the positive prognosis associated with SF3B1 and EIF1AX mutations.

Conjunctival melanoma — Unlike uveal melanoma, conjunctival melanoma is associated with activation of the MAPK pathway, with mutations present in either BRAF or NRAS in approximately one-half of cases [71]. The ultraviolet (UV) signature displayed by many conjunctival melanomas suggests they may have more in common with intermittently sun-exposed cutaneous melanomas than with uveal or mucosal melanomas.

Acral and mucosal melanoma — In mucosal and acral melanomas, KIT is the most frequently mutated gene (15 to 40 percent of cases). BRAF and NRAS mutations are much less common, occurring in less than 10 percent of cases overall. Imatinib, an inhibitor of kit, a cell surface receptor that can activate RAS, has demonstrated activity in this population. (See "Treatment of metastatic mucosal melanoma", section on 'KIT mutation'.)

Mutation status and prognosis — The prognosis of patients with advanced melanoma is influenced by the specific mutations present in a specific tumor.

●Melanomas with somatic mutations of either NRAS or BRAF are associated with a poorer prognosis. However, patients whose tumors contain BRAF mutations have improved response rates and survival when treated with BRAF plus MEK inhibitors. (See "Tumor, node, metastasis (TNM) staging system and other prognostic factors in cutaneous melanoma", section on 'Mutation status' and "Systemic treatment of metastatic melanoma with BRAF and other molecular alterations", section on 'Dabrafenib plus trametinib'.)

●Patients with acral or mucosal melanoma that contain KIT mutations have a poorer prognosis compared with similar patients whose tumors do not contain identifiable KIT mutations, although they have improved response rates and survival when treated with imatinib or other KIT inhibitors. (See "Locoregional mucosal melanoma: Epidemiology, clinical diagnosis, and treatment".)

●In patients with ocular melanoma, somatic mutations in either GNAQ or GNA11 do not appear to be associated with poor prognosis (at least relative to the small group of patients with tumors lacking either mutation). Mutation in the BAP1 gene, which is thought to regulate cellular growth control via as yet incompletely understood mechanisms, does appear to be associated with increased risk of metastasis and worsened prognosis. (See 'Uveal melanomas' above and "Initial management of uveal and conjunctival melanomas", section on 'Posttreatment systemic surveillance'.)

ROLE OF MITF — Microphthalmia-associated transcription factor (MITF) is crucial to melanin production. (See 'Melanocyte biology' above.)

MITF also has important roles in cell cycling during melanocyte differentiation, melanocyte invasion during physiologic migration, and the promotion of melanocyte survival [72]. While these functions are critical to normal melanocyte biology, MITF dysregulation can contribute to the pathogenesis of melanoma [73].

MITF amplification occurs in approximately 20 percent of melanomas and appears to be associated with poorer survival [73,74]. Additionally, a specific germline mutation in MITF has been associated with increased nevus count, non-blue eye color, and elevated melanoma risk [75,76].

Mechanistic data have revealed an intriguing relationship between MITF and the mitogen-activated protein kinase (MAPK) pathway, which is relevant to therapies targeting BRAF. MITF serves as a direct substrate for phosphorylation at serine 73 by MAPK [77]. One of the critical sequelae of this phosphorylation is the targeting of MITF for ubiquitin modification, resulting in rapid degradation of MITF protein [78]. A key consequence of this homeostatic circuitry is that BRAF antagonists block MITF phosphorylation, ubiquitination, and degradation, resulting in its stabilization and profound upregulation of certain (if not all) MITF transcriptional target genes. Some of these target genes are known antigens in immunologic recognition of melanomas, such as Mart1 or gp100. Thus, BRAF antagonism may enhance the antigenicity of melanomas via enhanced MITF stability and subsequent upregulation of melanocytic antigen expression [79]. Additionally, MITF was shown to directly control expression of PGC1a, which induces mitochondrial respiration and an attendant metabolic shift from glycolysis to oxidative phosphorylation [80].

CLINICAL IMPLICATIONS — Small molecule inhibitors have been developed that effectively target BRAF monomers, MEK, and KIT mutations in patients with melanoma. Inhibitors of BRAF homo- and heterodimers as well as mitogen-activated protein kinase (MAPK)/extracellular signal-regulated kinase (ERK) are under development

BRAF inhibitors — Potent selective inhibitors of BRAF monomers, such as vemurafenib, encorafenib, and dabrafenib, have induced tumor regression and prolonged overall survival in patients with metastatic melanoma that contains the mutated form of BRAF V600E. In such patients, these agents are administered in combination with MEK inhibitors. (See "Systemic treatment of metastatic melanoma with BRAF and other molecular alterations", section on 'BRAF V600 mutant disease'.)

●Mechanisms of action – Treatment with BRAF inhibitors leads to a reduction in levels of phosphoextracellular responsive kinase (pERK) in tumors containing the BRAF V600E mutation, which is associated with clinical response [81,82]. However, selective BRAF inhibitors may paradoxically hyperstimulate RAF kinases within cells in which the RAS pathway is activated upstream of a wild-type BRAF kinase [83-85]. The addition of an MEK inhibitor to a BRAF inhibitor blocks this paradoxical activation of the MAPK pathway.

●Mechanisms of resistance – Virtually every patient treated with an inhibitor of BRAF as monotherapy eventually has disease progression [86]. No consistent mechanism of resistance has been identified.

Resistance has not been associated with the development of a second mutation that impairs the binding of the treatment drug to BRAF, a resistance mechanism observed in targeted therapy in other malignancies. Several resistance mechanisms have been described that typically involve tumor cell reactivation via the MAPK pathway through a variety of alternative means.

Studies using BRAF V600E mutated cells that were generated to exhibit acquired resistance to BRAF inhibitors provide insights into how BRAF mutated cells survive BRAF inhibition. Several mechanisms of resistance have been identified, each of which has been investigated in at least a small number of tumor cells:

•Bypass mechanisms within the MAPK pathway can restore ERK activation, regardless of ongoing BRAF inhibition. Reestablishment of signaling can be achieved through upregulation of receptor tyrosine kinases (ie, platelet-derived growth factor receptor beta [PDGFRB], human epidermal growth factor 2 [HER2]) [87,88], activation of NRAS via mutation [88-90], upregulation of CRAF [87,91], activating mutations of MEK [90,92,93], activation of the serine/threonine MAPK kinases (COT) [87], overexpression of mutant BRAF [94], and loss of NF1 [95].

•Shortened forms of the BRAF protein may be produced due to altered RNA processing [96]. These modified forms of the BRAF protein can activate the MAPK pathway even in the presence of a BRAF inhibitor. This abnormality was identified in 6 of 19 patients with acquired resistance to vemurafenib.

•Signaling through the parallel growth and survival PI3K pathway, which can be initiated by insulin growth factor receptor 1 (IGF-1R) expression or AKT1 mutation, is an alternative mechanism of acquired resistance that has been described and fits the definition of an outside-of-pathway bypass mechanism [97,98].

While complete primary resistance to BRAF inhibition (ie, active progression during initiation of therapy) is seen in less than 10 percent of patients with BRAF mutant melanoma treated with vemurafenib [99], only a small fraction of patients experience complete remissions, suggesting the existence of mechanism(s) that acutely limit the tumor lethality of BRAF inhibition in most patients.

Preclinical studies suggest that elevated pretreatment levels of CRAF as well as baseline CCND1 amplification in tumors, leading to downstream overexpression of cyclin D1 and enhanced CDK4 expression, are pretreatment biomarkers worth further investigation [91,100]. In evaluation of baseline samples obtained from patients with BRAF mutant melanoma prior to treatment with dabrafenib, loss of the cyclin-dependent kinase inhibitor CDKN2A (p16^INK4a) and amplification of CCND1 were associated with poorer outcome to therapy [101]. Additionally, high pretreatment expression of the anti-apoptotic BCL-2 family member BCL2A1, as well as elevated stromal and plasma hepatocyte growth factor, and loss of phosphatase and tensin homolog (PTEN) expression are associated with poorer outcomes to BRAF inhibitor therapy [101-104]. Finally, contemporary studies have identified a metabolic pathway through which BRAF inhibition upregulates mitochondrial respiration via MITF and PGC1a in a response that limits the ability of BRAF inhibition to kill melanoma cells [80].

MEK inhibitors — MEK is a downstream mediator of RAS and RAF activation in the MAPK pathway. Although the MAPK pathway is activated in most melanomas, the presence of BRAF V600E mutation correlates strongly with response to MEK inhibition in murine melanoma xenograft models [105].

Various MEK inhibitors (eg, trametinib, cobimetinib, binimetinib, and selumetinib) have clinical evidence of activity, particularly in patients with characteristic BRAF mutations. MEK inhibitors are administered in combination with BRAF inhibitors in patients with metastatic melanoma and a targetable BRAF mutation. (See 'BRAF plus MEK inhibitors' below.)

●Trametinib – Trametinib (either as a single agent or in combination with the BRAF inhibitor dabrafenib) prolongs survival in patients with tumors harboring BRAF V600 mutations. In addition, trametinib is possibly useful in patients with NRAS and GNAQ/GNA11 mutations. (See "Systemic treatment of metastatic melanoma with BRAF and other molecular alterations", section on 'Dabrafenib plus trametinib'.)

●Cobimetinib – The activity of cobimetinib (in combination with the BRAF inhibitor vemurafenib) in patients with metastatic melanoma is discussed separately. (See "Systemic treatment of metastatic melanoma with BRAF and other molecular alterations", section on 'Vemurafenib plus cobimetinib'.)

●Binimetinib – Binimetinib has clinical efficacy in combination with encorafenib in patients with BRAF V600E mutations and as a single agent in patients with NRAS mutations. These data are discussed separately. (See "Systemic treatment of metastatic melanoma with BRAF and other molecular alterations", section on 'Encorafenib plus binimetinib' and "Systemic treatment of metastatic melanoma with BRAF and other molecular alterations", section on 'NRAS mutation'.)

●Selumetinib – Selumetinib in combination with chemotherapy (temozolomide or dacarbazine) has limited efficacy in patients with uveal melanoma. These data are discussed separately. (See "Metastatic uveal melanoma", section on 'Molecularly targeted agents'.)

BRAF plus MEK inhibitors — The combination of BRAF plus MEK inhibitors (eg, dabrafenib plus trametinib, vemurafenib plus cobimetinib, encorafenib plus binimetinib) is safe and effective in patients with BRAF V600 mutant melanoma. The clinical trial data supporting these combinations are discussed separately. (See "Systemic treatment of metastatic melanoma with BRAF and other molecular alterations", section on 'Choice of BRAF plus MEK inhibitor therapy'.)

KIT inhibitors — Activating mutations of KIT have been demonstrated in only a small percentage of all melanoma cases. KIT inhibitors have demonstrated encouraging activity in highly selected populations of patients with melanoma. These data are discussed separately. (See "Systemic treatment of metastatic melanoma with BRAF and other molecular alterations", section on 'KIT mutations (acral and mucosal melanoma)'.)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Basics topics (see "Patient education: Melanoma skin cancer (The Basics)")

●Beyond the Basics topics (see "Patient education: Melanoma treatment; localized melanoma (Beyond the Basics)" and "Patient education: Melanoma treatment; advanced or metastatic melanoma (Beyond the Basics)")

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Melanoma screening, prevention, diagnosis, and management".)

SUMMARY

●Pathways associated with melanoma – Research into the molecular pathogenesis of melanoma has identified the mitogen-activated protein kinase (MAPK) pathway (figure 1) and microphthalmia-associated transcription factor (MITF) as key factors in the development of melanoma. (See 'MAPK pathway' above and 'Role of MITF' above.)

●Common mutations – Although the MAPK pathway is activated in almost all melanomas, different abnormalities in the pathway are associated with specific subtypes of melanoma. (See 'Molecular characterization of melanocytic lesions' above.)

The most common alterations include:

•BRAF and NRAS in cutaneous and conjunctival melanomas. (See 'Cutaneous melanoma and precursor lesions' above and 'Classification of BRAF mutations' above and 'Conjunctival melanoma' above.)

•KIT in acral and mucosal melanomas. (See 'Acral and mucosal melanoma' above.)

•GNAQ and GNA11 (among others) in uveal melanomas. Other mutations such as BAP1, SF3B1, and EIF1AX have prognostic significance and are used to determine posttreatment surveillance. (See 'Uveal melanomas' above.)

●Clinical significance – Understanding of the importance of the MAPK pathway and BRAF mutations has led to the development of combination BRAF/MEK inhibitors, which significantly prolong survival in patients with metastatic melanoma containing a BRAF V600 mutation. (See 'Clinical implications' above and "Systemic treatment of metastatic melanoma with BRAF and other molecular alterations".)

ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges David E Fisher, MD, PhD, who contributed to earlier versions of this topic review.