Unraveling the Enigma of Melanoma Brain Metastasis: New Horizons in Mechanisms, Diagnosis, and Therapy

1. Introduction

Melanoma is an aggressive form of skin cancer and the 5^th leading cause of cancer-related deaths worldwide [1]. While frontline treatment strategies have greatly improved patient outcomes over the last decade, melanoma remains a deadly disease. Importantly, the majority of melanoma-related deaths result from distant metastases [1,2]. In the United States, an estimated 8,430 melanoma-related deaths occurred in 2025 [3]. Although melanoma incidence continues to increase, mortality rates began decreasing in 2013, largely due to the development of both targeted therapies and immunotherapies [4]. However, while the 5-year survival rate for patients with localized melanoma approaches 100%, survival still drops below 35% for patients with distant metastatic disease despite current treatments.

Melanoma arises from melanocytes, pigment-producing cells that synthesize melanin, which serves as a protective barrier against ultraviolet (UV) radiation. Melanocytes are derived from neural crest cells during embryonic development and reside in the basal layer of the epidermis [5]. Historically, melanoma progression has been described as a linear, stepwise process in which metastatic dissemination does not occur until late-stage disease progression, as outlined by the Clark model staging system [6]. This model outlines critical steps in melanoma progression, beginning with the formation of benign melanocytic and dysplastic nevi, followed by primary tumor growth, and ultimately culminating in metastasis, driven by the accumulation of genetic alterations [7]. In this framework, metastasis is restricted to advanced stages of disease and is attributed to a limited subset of tumor clones with metastatic potential. However, accumulating evidence indicates that melanoma progression is frequently non-linear and biologically more complex than this traditional model suggests.

Melanoma is characterized by a pronounced capacity for metastatic dissemination, driven by biological features that facilitate vascular invasion, immune evasion, and colonization of distant organs. Metastasis is initiated when malignant cells from superficial primary lesions invade the deeper dermal layers, enabling entry into the vascular and lymphatic compartments. Following intravasation, melanoma cells disseminate to regional and distant sites via lymphatic and hematogenous circulation [8]. Melanoma cells also acquire endothelial-like properties that enhance survival within the circulation and facilitate adhesion and extravasation at secondary sites (Figure 1). In parallel, the expression of pro-inflammatory and pro-angiogenic mediators supports tumor vascularization and metastatic outgrowth [9,10,11].

Metastatic progression is the primary determinant of melanoma prognosis. Notably, metastatic dissemination frequently precedes clinical detection of the primary lesion, limiting the effectiveness of surgical resection alone and necessitating aggressive multimodal treatment strategies, as removal of the primary lesion does not reliably prevent further spread [1]. Among metastatic sites, melanoma exhibits a particularly high propensity for brain colonization, which represents a major source of morbidity and mortality. An estimated 40–75% of patients with stage IV disease develop intracranial metastases, which is associated with a median survival of ~9 months from the time of brain metastasis diagnosis in the modern treatment era [12].

Although metastatic melanoma has historically been associated with poor outcomes, recent advances in immune checkpoint inhibition and targeted therapies have begun to improve survival, including in patients with brain involvement [13,14,15,16,17,18,19]. This review summarizes current insights into the mechanisms underlying melanoma metastasis, with a particular emphasis on brain colonization, and highlights emerging therapeutic strategies and future research directions.

2. Mechanisms of Melanoma Metastasis

Cutaneous melanoma is distinguished by an exceptionally high mutational burden and intratumoral heterogeneity, reflecting the involvement of numerous genetic drivers in disease initiation and progression. This extensive genetic complexity poses substantial challenges for therapeutic targeting and contributes to the rapid emergence of treatment resistance [20,21,22,23]. Notably, melanoma recurrence rarely occurs at the primary tumor site and instead frequently presents as metastatic lesions at distant sites [24], suggesting that dissemination likely occurs prior to surgical removal of the primary lesion.

The likelihood of melanoma metastasis is influenced by multiple factors, ranging from tumor microenvironment cues to intrinsic changes in cell state and differentiation [25]. Activation of diverse oncogenic cell signaling pathways is a hallmark of melanoma development, progression, and metastasis. Among these, aberrant activation of the mitogen-activated protein kinase (MAPK) pathway occurs in the majority of melanoma cases, most commonly through oncogenic mutations in BRAF, which are present in over 50% of all cutaneous melanoma cases and represent the most frequently altered oncogene in this malignancy (Figure 2). The most common oncogenic alteration, BRAF^T1799A, encodes the BRAF^V600E oncoprotein kinase which induces a strong gain-of-function phenotype and leads to constitutive BRAF signaling with sustained ERK activation.

Clinically, BRAF-mutant melanomas exhibit an increased propensity for brain metastasis [26,27], with specific mutational subtypes (e.g., V600K) associated with distinct survival outcomes [28]. Beyond proliferative signaling, BRAF mutations have been implicated in melanogenesis, phenotypic plasticity, and immune modulation within the tumor microenvironment, collectively facilitating metastatic spread [26]. Consequently, selective mutant BRAF inhibitors, in combination with MEK inhibitors, have become a cornerstone of targeted therapy for patients with BRAF-mutant melanoma [29].

NRAS, which encodes a membrane-associated GTPase in the RAS family upstream of the RAF family kinases, is mutated in approximately 20% of cutaneous melanomas and was among the earliest oncogenic drivers identified in this disease [30]. NRAS transduces receptor-mediated signals through the Ras-regulated>RAF>MEK>ERK MAPK cascade, leading to ERK-mediated phosphorylation of numerous cytoplasmic and nuclear targets, including the lineage-defining transcription factor MITF [23,31,32] (Figure 2). Although targeting RAS has historically proven challenging in numerous solid tumor types, earning its reputation as “undruggable”, recent advancements in targeted therapies have begun to show promise. This is further discussed in Section 5.

Activation of RAS isoforms (H-, N-, or K-) also triggers downstream activation of the phosphatidylinositol 3-kinase (PI3K) pathway, another critical driver of melanoma progression. PI3K activation leads to downstream phosphorylation of the serine/threonine kinase, AKT, which promotes cell survival through inhibition of pro-apoptotic mediators such as Bad. In melanoma, PI3K pathway activation is associated with enhanced tumor growth, angiogenesis, metastatic dissemination, and therapeutic resistance [33] (Figure 2).

Importantly, silencing of PTEN or activation of PI3K (H1047R) has minimal effects on normal melanocytes, indicating that these alterations likely do not drive melanoma initiation but promote its disease progression [34]. While it is well-established that melanomagenesis requires numerous cooperating events beyond activation of common drivers such as BRAF or NRAS, melanoma metastasis likewise depends on additional genetic events, the majority of which converge on activation of PI3K lipid signaling [35,36]. Specifically, loss of PTEN has been shown to cooperate with the BRAF^V600E mutation to drive metastatic melanoma [37]. In parallel, activation of AKT can independently promote melanoma metastasis to the lung and brain, even in the absence of PTEN inactivation [38,39,40,41,42]. Notably, melanomas with brain metastases frequently exhibit upregulated PI3K signaling (Figure 2), and although pharmacologic inhibition of this pathway can suppress tumor growth in preclinical and early clinical settings, therapeutic efficacy has been limited by dose-limiting toxicities and the requirement for simultaneous inhibition of multiple paralogs in the pathway [43,44,45,46].

2.1. Epithelial-to-Mesenchymal Transition [EMT] and Plasticity in Melanoma

Epithelial-to-mesenchymal transition (EMT) is a key driver of metastasis in many solid tumors. Although not epithelial in origin, melanoma cells can exhibit an EMT-like transition that shares molecular parallels with classical EMT in epithelial cancers, involving transcription factors such as ZEB1, TWIST1, and TCF4 that repress differentiation markers and facilitate metastatic propensity. This EMT-like transition is particularly relevant to central nervous system (CNS) dissemination in melanoma, where brain-metastatic variants exhibit upregulated EMT-associated genes, enabling enhanced aggressiveness and adaptation to the cerebral microenvironment during invasion and colonization. Through this EMT-like process, melanoma cells acquire mesenchymal traits that enhance motility, invasiveness, and resistance to therapy, as well as evasion of immune surveillance, thereby facilitating vascular transit and traversal of the blood-brain barrier (BBB) (21). This phenotypic plasticity is orchestrated by autocrine and paracrine signaling networks that induce extensive transcriptional reprogramming [47].

A central regulator of melanocyte biology and melanoma behavior is microphthalmia-associated transcription factor (MITF), a lineage-specific transcription factor essential for melanocyte survival [48,49,50,51]. MITF controls a broad transcriptional program encompassing melanogenesis, cell-cycle regulation, mitochondrial biogenesis, tricarboxylic acid cycle metabolism, and lipid utilization [52]. In melanoma, dysregulated MITF activity promotes multiple oncogenic phenotypes, including enhanced proliferation, invasion, immune evasion, and resistance to therapy. Importantly, MITF functions according to a “rheostat” model, in which low intracellular levels are associated with cell-cycle arrest and increased invasive capacity, whereas moderate to high expression promotes proliferation and cellular de-differentiation [53,54,55].

Upstream signaling pathways, including the MAPK pathway described above, frequently converge on MITF regulation, while additional pathways further modulate MITF regulation to drive melanoma progression and immune escape. The Wnt signaling pathway, a highly conserved regulator of development and tissue homeostasis, signals through both canonical β-catenin-dependent and noncanonical β-catenin-independent mechanisms. Specifically, MITF expression is directly regulated by β-catenin as part of the canonical Wnt signaling pathway, where it acts as a co-activator in complex with TCF/LEF family transcription factors that bind to the MITF promoter [56,57]. The MITF promoter is also activated by the lineage-specific transcription factor SOX10 [58,59,60,61], a marker strongly associated with metastatic melanoma in patient samples [62].

Although the role of β-catenin in melanoma metastasis has been debated, owing in part to reports that it can suppress migration, critical studies have demonstrated that expression of mutant β-catenin significantly increased metastasis in BRAF- or NRAS-driven melanoma mouse models [63,64]. A defining feature of EMT is the loss of epithelial adhesion mediated by E-cadherin coupled with a gain of mesenchymal markers such as N-cadherin, a switch that promotes cellular motility and survival during immune-mediated stress [65,66]. In melanoma, oncogenic NRAS and BRAF signaling drives EMT-like transcriptional programs, including a shift from ZEB2/SNAIL2 to ZEB1/TWIST1 expression, which has been linked to enhanced invasiveness and metastatic competence [10,67,68].

Additional regulators, such as autocrine secreted protein acidic and rich in cysteine (SPARC) signaling and upregulation of the homeobox transcription factor PRRX1, further reinforce EMT through repression of E-cadherin and activation of β-catenin and TGF-β pathways, respectively [22,55,69,70,71]. Collectively, these EMT-like programs promote the phenotypic flexibility required for melanoma cells to extravasate, adapt, and proliferate within the brain microenvironment.

2.2. Tumor Microenvironment Contributions to Metastasis

The tumor microenvironment (TME) surrounding melanoma brain metastasis plays a central role in immune evasion, therapeutic resistance, as well as metastatic outgrowth, and has therefore emerged as a critical focus of clinical and translational research. The melanoma brain metastatic TME is highly heterogenous, comprising immune and stromal cell populations, extracellular matrix (ECM) components, and a wide array of soluble factors including cytokines, chemokines, and growth factors. A major focus of TME investigation has centered on the immune landscape within this niche and the mechanisms of melanoma cell immune evasion.

A principal mechanism of immune escape involves activation of inhibitory immune checkpoints that suppress cytotoxic T lymphocyte (CTL) function. Melanoma cells induce upregulation of CTLA-4 and PD-1 on CTLs, leading to impaired T-cell activation and exhaustion in the context of chronic antigen exposure [72]. These pathways are therapeutically targeted by immune checkpoint inhibitors (ICIs), including monoclonal antibodies against CTLA-4 and PD-1, which aim to restore antitumor immunity [73,74,75,76,77]. Despite these advances, resistance to immune checkpoint blockade remains a major therapeutic obstacle in this disease [73,75,78,79].

Beyond T cells, innate and adaptive immune populations also contribute to immune evasion in melanoma brain metastasis. Neutrophils have emerged as key mediators of therapeutic resistance, with non-responding intracranial tumors exhibiting neutrophil-rich microenvironments despite comparable levels of T-cell infiltration [80]. In particular, IL-8–expressing neutrophil subpopulations, along with those producing interferons and calprotectin, promote angiogenesis, aggressive tumor growth, and EMT, suggesting that neutrophil-driven inflammation contributes to immune escape and metastatic persistence within the central nervous system [81].

B-cells also represent a prominent immunomodulatory population in melanoma brain metastases, occurring at higher frequencies than in brain metastases derived from breast or lung cancers [82]. Regulatory B-cells secrete anti-inflammatory cytokines such as IL-10, which suppress CTL activity and impair antigen presentation, thereby attenuating antitumor immune responses [83].

Stromal cells further shape the immunosuppressive and pro-metastatic niche of melanoma brain metastases. Although normally essential for maintaining BBB integrity and immune homeostasis, stromal cells can undergo tumor-induced reprogramming that supports metastatic growth, inflammation, and immune evasion [84]. Activated stromal signaling pathways, including YAP1 downstream of β-catenin, promote EMT-like transition through induction of a ZEB2-to-ZEB1 transcriptional switch in adjacent melanoma cells and may also drive tumor vasculogenesis [85,86,87]. Age-associated fibroblast changes further exacerbate metastatic progression, including secretion of the Wnt antagonist sFRP2, which suppresses β-catenin, MITF, and APE1, thereby promoting angiogenesis, metastasis, and resistance to BRAF-targeted therapies [19].

Remodeling of the ECM is another defining feature of melanoma brain metastasis. Loss of basement membrane integrity, particularly through collagen IV degradation, facilitates tumor invasion, while increased matrix stiffness promotes hypoxia, focal adhesion signaling, drug resistance, and metastatic fitness [88,89]. Aged fibroblasts contribute to ECM remodeling by increasing endothelial permeability, impairing lymphatic function, and promoting visceral metastasis. These fibroblasts also secrete ceramide lipids that are taken up by melanoma cells via fatty acid transport protein 2 (FATP2), encoded by the SLC27A2 gene, supporting resistance to oxidative stress and targeted therapies such as BRAF/MEK inhibitors [90,91].

Finally, melanoma cells actively remodel their metastatic niche through intercellular communication. The secretion of nerve growth factor receptor (NGFR)-containing extracellular vesicles enhances lymphangiogenesis and tumor-endothelial adhesion, thereby facilitating metastatic dissemination and colonization of the brain [92].

3. Melanoma Brain Metastasis: Unique Challenges and Mechanisms

Brain metastases occur far more frequently than primary brain tumors, and melanoma represents the third most common primary malignancy to metastasize to the brain, following lung and breast cancers [11,93,94,95]. Although lung cancer accounts for the greatest absolute number of brain metastases due to its higher overall incidence, melanoma exhibits the highest relative propensity for CNS dissemination among all solid tumors [11]. Epidemiological analyses have also identified sex-based disparities in melanoma brain metastasis. Recent healthcare database studies demonstrate that male patients have a 22% higher risk of developing brain metastases, and among affected individuals, experience poorer clinical outcomes compared with women [96].

The BBB serves as a formidable obstacle to metastatic dissemination, requiring tumor cells to actively traverse endothelial interfaces to colonize the CNS [97,98]. Although the precise mechanisms remain incompletely defined, emerging evidence suggests that direct interactions between melanoma cells and cerebral endothelial cells facilitate extravasation into the brain parenchyma [99,100]. Advances in in vivo modeling and live-cell imaging have substantially improved our understanding of the melanoma brain tropism, highlighting critical points in the metastatic cascade.

Dissemination of melanoma cells to the brain is a complex, multistep process that begins with intravasation into lymphatic or blood vessels. Circulating tumor cells must survive the hostile conditions of the circulation, express adhesion molecules that mediate interactions with vascular endothelium, and ultimately extravasate across the BBB [101,102,103]. This process is particularly challenging due to the presence of tight-junction adhesions and low permeability within the brain endothelium [104].

Within the CNS microenvironment, astrocytes and microglia represent key stromal populations that critically influence metastatic progression. Astrocytes contribute to BBB integrity through their perivascular end-feet, whereas microglia function as the resident immune cells of the CNS [105,106]. Although these cells were historically viewed as protective against metastatic invasion, accumulating evidence indicates that melanoma cells can co-opt astrocytes and microglia, inducing aberrant activation states that promote tumor survival, invasion, and therapeutic resistance [107,108]. Activated microglia, in particular, have been shown to suppress apoptosis in metastatic melanoma cells, further facilitating intracranial tumor expansion [109]. Collectively, these findings suggest that CNS stromal cells, while initially tumor-suppressive, can be rapidly subverted to support metastatic progression.

In addition to microenvironmental influences, melanoma brain metastases exhibit distinct genomic and transcriptomic profiles compared with primary tumors or extracranial metastases. Specifically, next-generation sequencing has identified increased frequencies of mutations in BRAF, SETD2, PBRM1, and DICER1, along with increased PD-L1 expression, underscoring distinct genetic and epigenetic features that may contribute to CNS tropism and therapeutic resistance [110]. Supporting this, Biermann et al. performed multi-modal single-cell and spatial genomics of treatment-naive melanoma brain metastases compared with extracranial metastases and discovered that cancer cells from melanoma brain metastases adopt a ‘neural-like’ cell state, expressing genes with essential roles in synapse formation and interactions among brain niche-cells [111]. This phenotype, beyond baseline neural crest lineage, may enhance brain-metastatic organotropism, invasiveness, and drug resistance, warranting further mechanistic and therapeutic exploration.

4. Diagnostic Advances in Melanoma Metastasis

Contrast-enhanced magnetic resonance imaging (MRI) with gadolinium is widely regarded as the most sensitive imaging modality for the detection of brain metastases [112]. T1-weighted sequences acquired after contrast administration are particularly sensitive, as metastatic lesions typically appear hyperintense relative to surrounding brain tissue. T2-weighted sequences, while less specific, provide complementary information and are useful for evaluating associated hemorrhage, edema, and mass effect [113]. Accordingly, MRI plays a central role across all stages of melanoma brain metastasis management, including initial detection, lesion characterization, precise anatomic localization, and longitudinal assessment of therapeutic response [112,114,115].

Despite its status as the imaging gold standard, MRI has important limitations. Gadolinium-based contrast agents are contraindicated in patients with impaired renal function, and MRI is not feasible for individuals with incompatible implanted devices such as pacemakers or metallic hardware. In such cases, alternative imaging modalities, including positron emission tomography (PET), are employed [112]. PET imaging most commonly utilizes the glucose analogue 18F-fluorodeoxyglucose (18F-FDG), which is preferentially taken up by metabolically active tumor cells [116]. FDG-PET is therefore useful for systemic staging and early detection of metastatic disease. However, its application in the brain is limited by high baseline glucose uptake in normal neural tissue, which reduces lesion contrast and spatial resolution. PET also provides limited anatomic detail than MRI, complicating accurate delineation of tumor margins, and is susceptible to false-positive findings in regions of inflammation, infection, or other metabolically active non-neoplastic processes. Nevertheless, FDG-PET remains a valuable tool for diagnosis, staging, treatment planning, and disease monitoring, particularly when MRI is contraindicated [112,116].

Radiomics has emerged as a promising computational imaging approach that converts conventional radiographic and histopathologic data into high-dimensional, quantitative features using advanced computational algorithms [117]. By extracting subtle patterns related to tumor shape, texture, and signal intensity that are not readily discernible by visual inspection alone, radiomics has the potential to improve tumor characterization and prognostication [117,118]. This approach may be particularly valuable in melanoma, which demonstrates marked heterogeneity in lesion size, morphology, and growth behavior, and where early detection could limit metastatic progression [21]. Despite its promise, clinical implementation of radiomics is currently limited by challenges related to reproducibility, lack of standardization across imaging platforms, and variability in analytic pipelines, underscoring the need for further validation before widespread clinical implementation [117,118].

Liquid biopsy represents a minimally invasive alternative to conventional tissue biopsy and has gained increasing attention in the evaluation of metastatic melanoma. This approach involves the analysis of bodily fluids, such as blood, cerebrospinal fluid, or saliva, for tumor-derived biomarkers such as circulating tumor cells (CTCs), circulating tumor DNA (ctDNA), and extracellular vesicles. Liquid biopsies enable real-time monitoring of disease burden, capture temporal genomic evolution, and facilitate assessment of therapeutic response in real time [119]. However, limitations include incomplete detection of tumor heterogeneity and the potential for false-positive or false-negative results [120].

CTCs, also known as circulating melanoma cells (CMCs), are shed from primary or metastatic lesions into the circulation and are of particular interest given melanoma’s high metastatic potential. Elevated CTC levels correlate with poorer prognosis and reduced overall survival. To persist in circulation, these cells undergo phenotypic changes, including epithelial-mesenchymal transition [120,121]. Similarly, ctDNA released through tumor cell apoptosis or active secretion provides molecular insight into actionable mutations, such as BRAF alterations, enabling treatment selection and response monitoring [122].

Exosome-based analyses represent an additional and promising diagnostic modality. Exosomes are small, stable extracellular vesicles released by both normal and malignant cells and are readily detectable in bodily fluids including blood and urine. They contain tumor-specific cargo, such as proteins, lipids, DNA, and RNA, that reflects the molecular characteristics of the originating tumor [120,123,124,125]. Selective packaging of exosomal contents, together with surface markers indicative of tissue origin, enables enrichment and characterization of tumor-derived vesicles. Increasing evidence suggests that combining multiple liquid biopsy approaches, such as ctDNA and exosome analysis, enhances sensitivity, reduces false-negative rates, and provides complementary molecular information [120]. Importantly, exosomes have been implicated in therapeutic resistance mechanisms in melanoma [126,127]. Specifically, exosome-mediated transfer of platelet-derived growth factor (PDGFR) to melanoma cells can activate the PI3K signaling pathway allowing bypass of BRAF and MAPK pathway inhibition [128].

Recent advances in liquid biopsy research have also focused on extracellular RNA, driven in part by large-scale initiatives such as the NIH-funded Extracellular RNA Communication Consortium [129]. Extracellular RNA molecules are protected from degradation through association with ribonucleoprotein complexes (e.g., Argonaute-2), binding to high- or low-density lipoproteins, or encapsulation within extracellular vesicles [130,131]. Integrating multiple extracellular RNA detection strategies has been shown to further enhance the sensitivity and clinical utility of liquid biopsy approaches [120].

Micrometastases represent an early and clinically occult stage of brain dissemination in which tumor cells colonize the CNS without forming detectable macroscopic lesions or producing neurological symptoms. Early identification of micrometastatic disease could enable timely therapeutic intervention and improve patient outcomes; however, detection remains challenging due to minimal tumor burden and weak biological signal [132]. Critically, reports of melanoma transmission from organ donors to recipients years after apparent disease resolution provide compelling evidence for long-term micrometastatic persistence [133].

Conventional imaging modalities often lack the resolution and specificity to distinguish micrometastases from treatment-related changes such as pseudoprogression or radiation necrosis, as well as from primary brain tumors or metastases of other origins [134,135]. Similarly, existing metabolic assays are insufficiently sensitive or specific to reliably detect micrometastatic disease. Ongoing research is therefore focused on identifying molecular and imaging biomarkers specific to melanoma brain metastases, including cerebrospinal fluid markers of tumor presence and neural tissue injury [136]. Advanced imaging techniques, such as amino acid–based PET tracers, are also under investigation. Notably, O-(2-¹⁸F-Fluoroethyl)-l-tyrosine (FET) PET has demonstrated a sensitivity of approximately 90%, highlighting its potential utility in detecting early intracranial melanoma involvement [137,138].

5. Therapeutic Strategies for Melanoma Brain Metastasis

The current standard of care for melanoma brain metastases encompasses three principal therapeutic modalities: systemic therapy, neurosurgical resection, and radiation therapy, delivered either as stereotactic radiosurgery (SRS) or whole-brain radiotherapy (WBRT) [139]. Treatment selection is guided by multiple factors, including symptom burden, number and size of lesions, molecular tumor profile, and patient performance status [16,140]. Historically, melanoma brain metastases were associated with a dismal prognosis, with median survival estimates of approximately 4–6 months. However, the advent of immunotherapies and advances in radiation delivery have led to meaningful survival improvements in survival for patients [12,141].

Neurosurgical resection is recommended for selected patients with large (e.g., >2.5 cm), symptomatic brain metastases, particularly those associated with hemorrhage or significant mass effect. Surgical intervention provides rapid decompression, symptomatic relief, and durable local control, and may improve overall survival in appropriately selected cases. However, surgery may be contraindicated for lesions located within critical brain regions or in patients with elevated perioperative complication risks [142,143].

Radiation therapy remains a cornerstone of intracranial disease management. SRS delivers high-dose radiation with submillimeter precision to discrete lesions while minimizing exposure to surrounding brain tissue. In contrast, WBRT exposes the entire brain to radiation and is generally reserved for patients with extensive intracranial disease or those who are not candidates for surgery or SRS. Although WBRT may provide symptomatic benefit and modest survival benefit, its use is limited by cumulative neurotoxicity and the risk of long-term cognitive decline [144,145]. Comparative studies have demonstrated no significant difference in recurrence rates between SRS and WBRT, with a trend toward improved survival in patients treated with SRS [146]. Consequently, current treatment paradigms favor systemic immunotherapy for patients with asymptomatic or small-volume brain metastases [147]. For patients with asymptomatic disease who progress on immunotherapy and harbor one to three lesions, SRS is preferred over WBRT [15,147]. The most favorable outcomes are observed in patients undergoing surgical resection followed by adjuvant SRS [15].

Systemic therapy for melanoma brain metastases consists primarily of immunotherapy and molecularly targeted agents. Systemic therapy alone has historically demonstrated limited efficacy, likely due to restricted drug penetration across the BBB, while radiotherapy alone offers only marginal benefit over supportive care and is frequently associated with neurotoxicity and subsequent cognitive decline [148,149]. Conventional chemotherapeutic agents, including dacarbazine and temozolomide, which can cross the BBB, have shown limited clinical benefit in this setting [150,151]. More favorable outcomes have been observed in patients undergoing surgical resection followed by adjuvant immunotherapy, although immune-related adverse events (IRAEs) frequently limit treatment durability[152].

Recent trials have highlighted the potential benefits of neoadjuvant immunotherapy. Studies comparing neoadjuvant vs. adjuvant pembrolizumab in resectable late-stage cutaneous melanoma demonstrated improved outcomes with neoadjuvant treatment; however, these trials excluded patients with brain metastases, underscoring the continued therapeutic gap for unresectable intracranial disease [153]. In 2022, joint American Society of Clinical Oncology (ASCO), Society for Neuro-Oncology (SNO), and American Society for Radiation Oncology (ASTRO) guidelines recommended combination immunotherapy with ipilimumab plus nivolumab for patients with asymptomatic melanoma brain metastases, irrespective of BRAF mutation status. For patients with BRAF^V600E/K-mutant disease, combined dabrafenib and trametinib therapy was also recommended as an alternative first-line option [154].

Multiple studies have evaluated mutation-specific targeted therapies in melanoma brain metastases. In the Phase II BREAK-MB trial, treatment with dabrafenib achieved an overall intracranial response rate of 35% in patients with BRAF^V600E- or BRAF^V600K alterations [155]. New evidence also supports the efficacy of newer BRAF and MEK inhibitors, including recently approved encorafenib and binimetinib for patients with brain metastatic melanoma [156,157]. Other targeted approaches have yielded mixed results. For example, the PI3K inhibitor buparlisib was well tolerated but produced no intracranial responses [14], while a Phase II study of the CDK4/6 inhibitor abemaciclib failed to meet its primary endpoint despite evidence of strong BBB penetration and intracranial clinical benefit in a subset of patients [158].

Despite these advances, effective pharmacologic treatment of melanoma brain metastases remains limited by the BBB, which is composed of tightly connected endothelial cells, pericytes, and astrocytic end-feet that collectively restrict CNS drug delivery [159]. While radiation and surgery may effectively control macroscopic disease, micrometastatic lesions often remain undetected and untreated [132]. Moreover, although immunotherapies demonstrate robust extracranial activity, their efficacy within the CNS remains variable, and even newer agents with partial BBB penetration have not consistently translated into improved overall survival [160,161].

Therapeutic resistance remains a major clinical challenge in advanced metastastic melanoma. A key mechanism of therapeutic resistance arises from tumor–stromal interactions, particularly with astrocytes [13]. Melanoma cells form gap junctions with astrocytes via connexin-43, which facilitates the settlement of metastatic melanoma cells in the brain [162].

Emerging therapeutic strategies aim to overcome resistance and exploit CNS-specific vulnerabilities. Chimeric antigen receptor (CAR) T cells have demonstrated efficacy in numerous hematologic malignancies and show promise in melanoma [163,164,165,166,167]. Oncolytic viruses represent another promising approach, leveraging tumor-selective viral replication to induce direct oncolysis and stimulate antitumor immunity [168]. A Phase III clinical trial led to FDA-approval of Talimogene Laherparepvec (TVEC), a modified herpes simplex virus type 1 (HSV-1) encoding granulocyte-macrophage colony-stimulating factor (GM-CSF) for advanced melanoma [169]. However, BBB penetration remains an obstacle to oncolytic virus efficacy, and further studies are needed to assess CNS delivery of candidate viruses such as parvovirus H-1PV [170].

Additional innovative approaches include repurposing BBB-penetrant antipsychotic agents such as trifluoperazine, fluphenazine, and clozapine, which disrupt lysosomal function, induce cell-cycle arrest, and modulate the immune microenvironment in melanoma brain metastases [171,172]. While targeting RAS was long considered unfeasible, the advent of novel RAS inhibitors has recently revolutionized the treatment landscape for cancers driven by oncogenic RAS signaling and is now being clinically explored in NRAS-mutant melanoma [173,174].

Finally, recent studies have identified focal adhesion kinase (FAK) as a critical mediator of melanoma brain metastasis downstream of PI3K signaling [42,175]. In a seminal study, Almazan and colleagues demonstrated that combined inhibition of FAK and RAF/MEK signaling suppressed melanoma tumor growth, reduced brain metastasis formation, and prolonged survival, including in mice with established intracranial disease. FAK inhibition alone significantly reduced the development of brain metastases, while the combined inhibition of FAK and RAF/MEK induced robust apoptosis and tumor regression leading to prolonged survival in preclinical models [175]. Notably, this study employed the novel RAF/MEK inhibitor, avutometinib, which functions as a molecular clamp to inhibit MAPK signaling. Promising preclinical studies such as these have catalyzed ongoing clinical trials for solid tumors using these combination therapies and offer renewed hope for patients with melanoma brain metastases [176,177] (Table 1).

6. Immunological Perspectives in Melanoma Brain Metastasis

The brain microenvironment is composed of a unique cellular framework, characterized by a distinct immune landscape that differs markedly from other common sites of melanoma metastasis. While the immune landscape of the brain has been historically considered immunologically isolated due to the BBB, recent studies have challenged this paradigm, particularly in the context of brain metastases where the BBB is frequently compromised, permitting infiltration of peripheral immune cells [178]. Thus, in addition to resident microglia, macrophages, and dendritic cells, peripheral immune cells such as neutrophils and lymphocytes are also commonly present within the metastatic brain microenvironment [178].

Tumor-associated macrophages (TAMs) are highly expressed in the brain microenvironment and play a critical role in the pathogenesis of brain metastasis. Two principal TAM subpopulations reside in the CNS: microglia and border-associated macrophages (BAMs). Microglia are phenotypically plastic cells and represent the predominant immune cell population in the healthy brain [179]. Notably, the composition of TAMs is distinctly dynamic and evolves over the course of metastatic progression. In early-stage disease, characterized by smaller metastatic brain lesions, microglia comprise the majority of TAMs. However, as disease progresses and peripheral monocytes infiltrate the brain, the TAM population shifts toward monocyte-derived macrophage cells [180,181,182]. Microglia, in particular, directly interact with melanoma cells and facilitate their establishment and survival in the brain microenvironment [183,184].

The lymphocytic compartment of the brain microenvironment is primarily composed of T-cells, including both CD4⁺ and CD8⁺ populations, which represent the most phenotypically diverse immune cells in this niche [82,185,186]. Interactions between microglia cells and T-cells have been shown to exert context-dependent effects, contributing to both immunosuppression and immune activation. Several reports demonstrate that microglia-T cell interactions suppress T cell activity and promote metastatic growth [187]. Conversely, brain resident T-cells have also been shown to exhibit antigen-presentation capacity, enhancing anti-tumor immune response within the brain metastatic compartment [188,189].

Despite the challenges targeting brain metastases, growing insight into the immune microenvironment in the brain, along with recent advances in immunotherapeutic strategies capable of crossing the BBB, offers renewed promise [190]. One such strategy involves targeting TAMs, given their central role in promoting brain metastatic outgrowth. Approaches to modulate TAMs include macrophage depletion through inhibition of the colony-stimulating factor 1 receptor (CSF1R), which regulates macrophage survival and differentiation. However, while depletion strategies may counteract the pro-tumorigenic role of TAMs, they could be limited by compensatory recruitment of other myeloid cells, such as myeloid-derived suppressor cells. Another approach is reprogramming macrophages toward anti-tumor phenotypes. Tumoral interferon beta (IFNβ), often induced by radiotherapy, has been shown to repolarize TAMs in murine melanoma brain metastasis models [191]. This re-education shifts TAMs toward an immune-stimulatory state, and has the potential to improve clinical outcomes in patients, including better responses to combined radioimmunotherapy. IFNβ could act as a predictive biomarker in this context, providing mechanistic insights into synergistic therapies with immunotherapies. While preclinical studies have shown promising effects in vivo, further investigation is needed before these modalities can be translated into the clinic [184,192,193,194].

Immunotherapy has revolutionized the treatment landscape for patients with melanoma. However, baseline immune status plays a critical role in determining therapeutic response. Tumors are often described along an immunologic spectrum ranging from “hot” to “cold” [195]. While primary brain tumors are typically immunologically “cold,” brain metastases, especially originating from melanoma primary tumors, can exhibit “hot” immune profiles due to infiltrating lymphocytes and antitumor immune cells [178]. Because melanoma cells, including metastatic melanoma cells, frequently express high levels of PD-1 and PD-L1, clinical trials have focused on assessing the efficacy of immune checkpoint inhibitors in melanoma brain metastases [82].

A recent Phase II clinical trial evaluating the PD-1 antibody pembrolizumab in patients (both treatment naive and pretreated) from multiple primary tumor types, including melanoma, demonstrated an intracranial response of ~40%. However, more than half of patients experienced treatment-related adverse events, highlighting a major therapeutic obstacle. Similarly, trials evaluating the combination therapy with the CTLA-4 inhibitor ipilimumab and the PD-1 inhibitor nivolumab in melanoma brain metastases reported intracranial responses of 46-57%, yet again with adverse events in over half of all treated patients [17,19].

Recent efforts have focused on enhancing immune activation within the brain microenvironment while overcoming immunosuppressive barriers [196,197]. Encouragingly, studies in other metastatic tumor types, as well as novel immunotherapeutic combination strategies, have demonstrated substantial improvements in intracranial progression-free survival and prolonged time to brain metastasis [81,198,199]. Nonetheless, significant work is needed to improve immunotherapeutic efficacy and tolerability for patients with melanoma brain metastases who have exhausted currently available treatment options.

7. Translational and Preclinical Models for Studying Brain Metastasis

The development and application of diverse experimental models has been essential for advancing our understanding of melanoma biology, including tumor behavior, metastatic progression, therapeutic resistance, immune interactions, and early oncogenic events [200]. Current approaches encompass a broad spectrum of in vivo and in vitro systems including zebrafish models, genetically-engineered mouse models (GEMMs), tumor-derived cell lines, patient-derived xenografts (PDX), and a growing array of three-dimensional (3D) culture platforms, such as organoid models [200].

Mouse models have long formed the foundation of preclinical melanoma research and have played a central role in the development of effective therapeutic strategies. However, interspecies genetic and physiological differences can limit translational relevance, particularly in PDX models [200,201]. Notably, only ~30% of findings from animal studies advance to human clinical trials, and fewer than 8% ultimately progress beyond Phase I testing, underscoring the need for more predictive experimental systems [202,203].

Tumor-derived melanoma cell lines remain widely used due to their reproducibility, suitability for high-throughput experimentation, and ease of genetic manipulation. While these systems enable mechanistic interrogation and drug discovery efforts, they are inherently limited by selection bias toward culture-adapted clones, absence of a physiologically relevant microenvironment, and lack of critical stromal, vascular, and immune components that influence tumor behavior in vivo [204,205].

PDX models partially address these shortcomings by engrafting human tumor tissue into immunocompromised mice, thereby preserving intratumoral heterogeneity and enabling assessment of therapeutic responses across architectural, histological, phenotypic, and mutational dimensions. However, dependence on murine host stroma and lack of a fully functioning immune system limits the capability of PDX models to faithfully recapitulate human tumor-microenvironment interactions [200,206].

Advances in 3D melanoma modeling have yielded platforms that more closely mimic native tissue architecture and tumor microenvironments. Spheroid cultures, composed of aggregated melanoma cells grown in suspension, develop oxygen and nutrient gradients that generate intratumoral heterogeneity and enhance physiological relevance relative to two-dimensional cultures [207,208]. When co-cultured with additional cell types, spheroids can provide insight into early steps of the metastatic cascade; nevertheless, their capacity to model complex host–tumor interactions remain limited [209,210].

Organoid models represent a further evolution of 3D systems, incorporating layered skin cell populations within supportive scaffolds. Recent bioengineering innovations have enabled the inclusion of appendages such as hair follicles, sebaceous and sweat glands, and vascular-like structures, reducing necrosis and enhancing tissue fidelity [211]. Importantly, patient-derived melanoma organoids generated from metastatic lesions have recently been shown to be feasible, enabling personalized modeling of disease progression and therapeutic response [212,213,214].

Skin reconstruct models employ primary human keratinocytes, melanocytes, and fibroblasts embedded within extracellular matrix components to generate stratified epidermal structures that closely resemble native human skin. These platforms enable detailed investigation of cell–cell interactions, stromal influences, and melanoma initiation and progression within a customizable microenvironment [215,216].

Finally, integration of multi-omics approaches, spanning genomics, transcriptomics, proteomics, metabolomics, and epigenomics, has substantially advanced understanding of the molecular determinants that enable melanoma cells to thrive and spread in the brain. Recent studies reveal that melanoma brain metastases harbor distinct molecular programs that drive CNS tropism, immune escape, metabolic reprogramming, and therapeutic resistance [217]. Single-cell multi-omic analyses have identified a pre-metastatic melanoma subpopulation with upregulated neural adhesion genes (e.g. NRG3, NCAM1) that predisposes tumors to brain colonization [111,217,218]. Complementary genomic and transcriptomic profiling reveals that melanoma brain metastases frequently harbor mutations such as loss of function PTEN and display brain-like gene expression signatures not observed in extracranial lesions, underscoring the importance of context-specific modeling platforms [111,217,218].

8. Clinical Management and Multidisciplinary Approaches

Management of melanoma brain metastases requires a comprehensive, multidisciplinary approach involving neuro-oncologists, medical oncologists, radiation oncologists, neurosurgeons, rehabilitation specialists, and palliative care teams, each of whom plays a critical role in optimizing patient outcomes. Given the complexity and heterogeneity of intracranial disease, coordinated care across specialties is essential for effective symptom control, therapeutic decision-making, and longitudinal management.

Neuro-oncologists are central to the care of patients with melanoma brain metastases, as these individuals frequently experience a broad spectrum of neurological complications arising from both tumor burden and treatment-related toxicities [112,219]. Clinical manifestations vary according to lesion location and extent and may include seizures, neurocognitive decline, focal neurological deficits, and altered mental status. Accordingly, neuro-oncologic management emphasizes precise pharmacologic and supportive interventions tailored to individual neurological symptoms, with the overarching goal of preserving function and quality of life [112,219].

Medical oncologists oversee systemic disease management, including tissue biopsy, molecular and genomic profiling, and administration of systemic therapies such as immune checkpoint inhibitors and targeted agents. Although they do not directly perform surgical or radiation-based interventions, medical oncologists play a key role in therapeutic strategy selection, coordination of multidisciplinary care, management of treatment-related adverse effects, and provision of patient education and longitudinal support [220].

Diagnostic radiologists contribute to accurate detection and characterization of brain metastases through advanced imaging techniques, while radiation oncologists determine the appropriateness, modality, and dosing of radiotherapy and manage treatment-related toxicities in close collaboration with the broader care team [221,222].

Despite the substantial symptom burden associated with intracranial disease, palliative care and quality-of-life considerations remain underemphasized in patients with melanoma brain metastases. Patients commonly experience debilitating symptoms, including headaches, seizures, fatigue, and progressive neurological impairment, which significantly disrupt daily functioning. Early integration of palliative care has been shown to improve quality of life and reduce symptom severity, particularly when sustained for longer than three months [223,224]. In addition to addressing physical and psychological distress, palliative care facilitates informed decision-making and aligns treatment strategies with patient goals. Given the rapid and often unpredictable progression of brain metastases, early multidisciplinary involvement, including neurology, psychiatry, oncology, and supportive care services, is essential for optimizing patient-centered outcomes [220,224].

As therapeutic options continue to evolve, there is a growing need for expanded research and education focused on the genetic, molecular, and biological drivers of melanoma metastasis to enable more personalized treatment strategies. Advances in genomic profiling, including liquid biopsy approaches such as CTC analysis and ctDNA detection, now permit minimally invasive acquisition of molecular data that can inform therapeutic selection, monitor disease progression, and detect emerging resistance mechanisms [112].

In parallel, predictive modeling approaches that integrate clinical variables, imaging data, germline genetic markers, and computational analyses are increasingly being used to assess metastatic risk and guide individualized management [21,119,120,135]. These models have demonstrated improved accuracy in early detection and prognostication, supporting more tailored treatment paradigms. Although limitations remain, particularly in model generalizability and data standardization, emerging tools such as liquid biopsy platforms and artificial intelligence-driven radiomics hold significant promise for enhancing metastatic risk prediction and further refining precision care in melanoma [112,225,226].

9. Future Directions and Research Gaps

Future research on melanoma brain metastases converges on a few major challenges, including, overcoming therapeutic barriers to drug delivery across the BBB, advancing biomarker identification to detect early metastatic spread, and further elucidating mechanisms of how melanoma spreads to the brain. A major priority of metastatic melanoma research is to identify therapeutics that can reach the brain microenvironment, as the BBB prevents adequate drug penetration and can lead to uneven drug distribution [97]. This, along with unique immune profiles and an immunosuppressive microenvironment in the brain, makes current systemic treatments for melanoma brain metastases less effective [227]. Concurrent efforts in biomarker discovery aim to identify molecular signatures to help predict patients with a higher risk of brain metastasis in order to guide therapeutic options. Multi-omics platforms (such as genomic, transcriptomic, proteomic, and spatial profiling approaches) and liquid biopsies are leveraged to uncover new predictive biomarkers and therapeutic targets [228].

Equally important is expanding our mechanistic understanding of how melanoma cells invade and adapt in the brain microenvironment. From reaching the BBB to co-opting astrocytes and microglia to evade the immune system and ultimately propagate in the brain. Emerging approaches such as multi-omic profiling and AI-driven modeling are promising to accelerate progress. For example, a recent multi-omic analysis revealed distinct immunosuppressive and metabolic tumor microenvironments in melanoma brain metastases [218]. Moreover, machine learning models can integrate radiomic, genomic, and clinical data in order to improve prognostic predictions and aid in personalizing treatment [229]. By addressing these gaps to improve drug delivery across the BBB, establish robust biomarkers for early metastatic spread, and further identifying brain colonization, the field will continue to make strides to deliver more effective and personalized interventions for patients with melanoma brain metastases.

10. Conclusions

Melanoma brain metastasis is unique in its aggressive tumor-intrinsic programs and targeting the CNS-microenvironment is uniquely challenging, thus, exemplifying one of the most clinically challenging manifestations in cancer [1,11]. Advances in molecular biology, multi-omic profiling, and preclinical modeling have clarified mechanisms of CNS tropism, including lineage-specific transcriptional programs, metabolic rewiring, and stromal co-option by astrocytes and microglia [35,69,107,183,193,230]. However, these insights are only the beginning in translating effective intracranial therapies. Clinically, progress in immune checkpoint blockade and targeted MAPK pathway inhibitors have improved outcomes for some patients [17,43,75,76,152,153,157,175,231,232,233], but the BBB, intratumoral heterogeneity, and adaptive resistance pathways continue to limit durable therapeutic benefit for these patients [218]. Moving forward, it will be critical to further investigate and develop therapeutic strategies to deliver effective and safe CNS penetrating treatments, develop liquid biopsy approaches to detect micrometastases to guide treatment strategies, and further drive integrated multi-omics and AI analytics to deconvolute brain-specific metastatic biology and resistance mechanisms. With translational pipelines and clinical trials designed specifically for intracranial disease, there is potential to translate mechanistic discoveries into therapies to extend survival and preserve neurological function for patients with this devastating disease.