Melanoma, a highly heterogeneous tumor, is comprised of a functionally diverse spectrum of cell phenotypes and subpopulations, including stromal cells in the tumor microenvironment (TME). Melanoma has been shown to dynamically shift between different transcriptional states or phenotypes. This is referred to as phenotype switching in melanoma, and it involves switching between quiescent and proliferative cell cycle states, and dramatic shifts in invasiveness, as well as changes in signaling pathways in the melanoma cells, and immune cell composition in the TME.
1. Introduction
Malignant melanoma is one of the most aggressive, heterogeneous and treatment-resistant cancers
[1], which exhibits among the highest mutation rates when compared to other cancers
[2]. Malignant melanoma is highly heterogeneous and is associated with multiple transcriptionally distinct cell phenotypes. These distinct phenotypes are plastic in nature and can switch from one phenotypic state to another to escape both targeted therapy and immunotherapy
[3].
Approximately 75% of deaths from skin cancer are due to malignant melanoma. While the 5-year survival rate is over 90% for localized melanoma, this drops to 16% for distant disease, indicating that metastasis is the main reason for poor outcomes
[4]. Recent advances in genomic technologies, especially single cell-based sequencing, and the availability of online gene expression datasets, have allowed the characterization of the different subsets of melanoma based on gene signatures and enabled categorization of them as distinct phenotypes.
Melanoma originates from the malignant proliferation of the pigment-producing melanocyte cells, and occurs predominantly in fair-skinned populations, although it can affect darker-skinned populations as well. Cutaneous melanoma predominantly occurs due to excessive ultra-violet irradiation exposure and mutations occurring in the melanocyte cells located in the basal layer of the epidermis in the skin. Melanoma occurs as well on the acral skin on the palms of the hand and soles of the feet to which darker-skinned populations are also susceptible. Additional forms of melanoma occur in the mucosal surfaces, the uveal tract, and leptomeninges
[5][6]. The Clark model for melanoma progression depicts a series of histopathological transformations of melanocytes in progressing to malignant melanoma and the subsequent development of invasion and metastasis
[7], involving the tightly regulated switching of cellular phenotypes. Melanoma exhibits characteristics in many measurable traits from early to advanced stages, including proliferation, metastatic potential, and therapeutic resistance. The progression from non-invasive to invasive melanoma resembles epithelial–mesenchymal transition (EMT), a well-characterized process of phenotypic change that is associated with metastatic progression in epithelial cancers
[4]. Both genetic and epigenetic changes contribute to the transformation of melanocytes into metastatic malignant melanoma cells
[5][6]. Much, however, remains unknown about melanoma tumor heterogeneity, and its role in disease progression and treatment response.
Targeted therapy and immune checkpoint inhibitors (ICI) are the major treatment options for metastatic melanoma
[8][9]. Once melanoma has metastasized, treating metastatic melanoma is very challenging. Irreversible genetic events, such as mutations, are believed to play a key role in melanoma initiation, potentially leading to targeted therapy resistance. However, oncogenic mutations are not responsible for the entirety of dynamic changes observed at the transcriptional and phenotypic level during melanoma progression. A multitude of distinct stable phenotypic cell states can arise within the same melanoma during metastatic dissemination
[10], which are unique in terms of molecular and functional profiles, and these characteristics vary between each cell type. Moreover, this phenotypic diversity can facilitate interconversion between phenotypic states in response to drug challenges, and melanoma can evolve to form new drug-tolerant phenotypes, making it difficult to treat
[11].
Melanoma is a highly aggressive type of skin cancer
[12], exhibiting the ability to switch between different transcriptional states, due to its neuroectodermal origin, with high phenotypic plasticity
[13]. Indeed, melanoma’s aggressiveness is at least in part due to its phenotypic plasticity, whereupon early in the tumor evolution, melanoma metastasizes to lymph nodes, distant tissues, and organs, involving the acquisition of functions such as migration and invasion, intravasation, survival in the circulation, extravasation, and colonization at secondary tumor sites
[14]. Malignant melanoma cells acquire the capacity to undergo transient and reversible morphological and functional changes by hijacking the embryonic neural crest invasion program, such that metastatic melanoma cells exploit neural crest-related receptor tyrosine kinases to increase plasticity and facilitate invasion, while primary melanocytes may actively suppress these responses under the same microenvironmental conditions
[15]. Among the most well-characterized molecular changes that signify a shift in melanoma cell behavior, linked to phenotypic plasticity, and acquisition of migration and invasion, is alterations in the expression level of
microphthalmia-associated transcription factor (MITF). Very low or absent expression of
MITF is characteristic of invasive melanoma cells, while high expression of
MITF characterizes non-invasive melanoma cells
[14] (
Figure 1).
Figure 1. Depiction of phenotype switching during melanoma progression and the development of immunotherapy and drug resistance, based on the expression of MITF and MITF-related genes in the same tumor bed. Melanoma cells expressing low levels of MITF correspond to a slow-cycling and pro-invasive state (similar to “mesenchymal-like”), whereas higher levels of expression of MITF correlate with a proliferative and melanocytic state. Undifferentiated melanomas/neural-crest like (on the right side) melanomas lack activated immune cells, while melanocytic melanomas (left side) are composed of immunologically relatively more active immune cells/hot tumor microenvironment, which is also associated with melanocytic melanomas being relatively more responsive to immunotherapy or targeted drugs.
Thus, melanoma cells can exhibit different phenotypic states based on the heterogeneous expression of
MITF in the same tumor bed, regardless of whether the cell lines have a mainly invasive (
MITF negative) or proliferative (
MITF positive) phenotype in vitro
[14]. For instance, Konieczkowski et al.
[13] revealed that sensitive melanomas display a high expression of
MITF and downstream MITF-mediated up-regulation of differentiation markers such as
TYRP1,
MLANA, and
PMEL. The same study also showed that resistant melanomas possess low
MITF expression, but high levels of inflammatory nuclear factor ‘kappa-light-chain-enhancer’ of activated B-cells (NF-ĸB) signaling and the receptor tyrosine kinase (RTK)
AXL [13]. That study supported the “phenotype switch” model, in which adaptive switching between different phenotypes in response to the tumor microenvironment (TME) is crucial for melanoma progression. However, phenotype switching is not obligatory for the invasion and metastatic spread of melanoma cells
[16].
Although phenotype switching resembles epithelial–mesenchymal transition (EMT), EMT is not an appropriate term to describe this process in melanoma, as melanocytes are not epithelial cells. Yet, EMT-like processes play a key role during the formation and migration of neural crest cells. Neural crest cells are a multipotent, migratory, transient cell population that migrate through the vertebrate embryo to infiltrate different organs and differentiate in various cell lineages including melanocytes. Melanocytes can be regarded as a product of an embryonic neural crest EMT-like process
[16].
Melanoma heterogeneity results from diverse cancer cell evolution during the disease course. As progressively more subpopulations of tumor and stromal cells develop in association with melanoma, the more challenging successful treatment becomes, such that in advanced melanoma, treatment frequently follows a pattern involving tumor growth, followed by metastasis, reduced efficacy of therapeutic treatment and, finally, resistance to treatment. Treatment strategies designed for the effective targeting of tumor heterogeneity in melanoma could lead to better therapeutic options and better outcomes for patients.