Malignant melanoma in the 21st century: the emerging molecular landscape

Abstract

Malignant melanoma presents a substantial clinical challenge. Current diagnostic methods are limited in their ability to diagnose early disease and accurately predict individual risk of disease progression and outcome. The lack of adequate approaches to properly define disease subgroups precludes rational treatment design and selection. Better tools are urgently needed to provide more accurate and personalized melanoma patient management. Recent progress in the understanding of the molecular aberrations that underlie melanoma oncogenesis will likely advance the diagnosis, prognosis, and treatment of melanoma. The emerging pattern of molecular complexity in melanoma tumors mirrors the clinical diversity of the disease and highlights the notion that melanoma, like other cancers, is not a single disease but a heterogeneous group of disorders that arise from complex molecular changes. Understanding of molecular aberrations involving important cellular processes, such as cellular signaling networks, cell cycle regulation, and cell death, will be essential for better diagnosis, accurate assessment of prognosis, and rational design of effective therapeutics. Defining an individual patient's unique tumor characteristics may lead to personalized prediction of outcomes and selection of therapy. We review the emerging molecular landscape of melanoma and its implications for better management of patients with melanoma.

FIGURE 1. The CDKN2A locus and cell cycle control

The CDKN2A locus on chromosome 9p21 has an unusual structure because it encodes for 2 overlapping but very distinct proteins: p16^INK4A and p14^ARF. This is accomplished through selective use of an alternative first exon (exon E1a in p16^INK4A and exon E1b in p14^ARF). Although structurally very different, both protein products act as negative regulators of cell cycle progression. The p16^INK4A protein inhibits the activation of CDK4 and CDK6 by cyclin D1 (CCND1), thereby preventing the subsequent phosphorylation of RB1. Underphosphorylated RB1 sequesters the transcription factor E2F and prevents it from inducing the progression from G1 to S phase of the cell cycle. The absence of functional p16^INK4A, therefore, leads to hyperphos-phorylation of RB1 with resulting release of E2F and uninhibited cell cycle progression. In contrast, p14^ARF regulates tumor protein 53 (p53) activity by inhibiting MDM2, a ubiquitin ligase that otherwise targets p53 for degradation by proteasome. High levels of p14^ARF stabilize p53, permitting it to induce p21^WAF1/CIP1, a cell cycle inhibitor that blocks CDK2/cyclin E (CCNE1)–mediated phosphorylation of RB1. In the absence of functional p14^ARF, uncontrolled ubiquitination and degradation of p53 removes this important cell cycle brake, leading ultimately to hyperphosphorylation of RB1 and cell cycle progression. P = phosphate. For expansion of other abbreviations, see Glossary on page 846.

FIGURE 2. The MAPK and PI3K/AKT pathways

Activation of various receptor tyrosine kinases (VEGFR, C-KIT, C-MET, FGFR, IGF1R, and others) by their cognate ligands (VEGF, SCF, HGF, FGF, and IGF1/2, respectively) leads to proliferation and survival of melanoma cells through two interacting signal transduction pathways. Ligand availability through autocrine loops (eg, VEGF), paracrine secretion (eg, IGF1), as well as IGFBPs can modulate activation of these pathways. Within the MAPK pathway, activated receptors lead to SHC-mediated activation of RAS and propagation of signaling through RAF, MEK (also known as MAP2K), and MAPK (also known as ERK). Activated MAPK transduces signals that regulate multiple cell processes including proliferation, differentiation, angiogenesis, and survival. Isolated activation of the MAPK pathway by oncogenic stimuli, such as mutated BRAF, results in expression and secretion of IGFBP7, which, in turn, suppresses MAPK activity and induces senescence, as seen in benign nevi. Melanomas seem to circumvent this negative feedback loop by blocking the expression of IGFBP7 through yet unidentified mechanisms. The second essential signaling pathway stimulated by receptor tyrosine kinases is the PI3K/AKT pathway. PI3K, a lipid kinase, catalyzes the phosphorylation of phosphatidylinositol (PI). Triple phosphorylation of PI is necessary for the association of AKT with the cell membrane and its subsequent activation by PDK1. The key regulatory component is PTEN, a phosphatase that regulates PI3K and SHC phosphorylation. RAS may also regulate PI3K activity. Through activated AKT, a number of key mitogenic processes are activated including inhibition of apoptosis, survival gene transcription, cell cycle progression, protein translation, and cell growth and proliferation. The balance between AKT activation and PTEN regulatory activity is a key determinant in cell cycle progression. FGF = fibroblast growth factor; HGF = hepatocyte growth factor; IGF = insulinlike growth factor; P = phosphate; VEGF = vascular endothelial growth factor; ? = mechanism unknown. For expansion of other abbreviations, see Glossary on page 846.

FIGURE 3. Apoptosis pathway

Apoptotic cell death can be initiated through two main pathways: the death receptor (extrinsic) pathway and the mitochondrial (intrinsic) pathway. The death receptor pathway is activated by the binding or FAS or TRAIL ligands to their receptors (DR4 or DR5), stimulating receptor aggregation. This aggregation stimulates recruitment of FADD and caspase (casp) 8 activation, which is regulated by cFLIP (also known as CFLAR) and growth factor signaling through the MAPK pathway. Casp 8 activation leads to casp 3 cleavage, which initiates multiple proapoptotic processes, including CAD stimulation of DNA cleavage. Through the mitochondrial pathway, prosurvival signaling through AKT activation stimulates phosphorylation of BAD, which allows BCL2 to exert its antiapoptotic effects by blocking proapoptotic proteins NOXA and BAX. However, dephosphorylated BAD blocks BCL2 by heterodimerization (BAD/BCL2), which allows proapoptotic proteins to form pores in the mitochondria. This process then releases apoptogenic factors from the mitochondrial intermembrane space, including cytochrome C (Cyt C), APAF1, and casp 9. These factors form the so-called apoptosome, which stimulates apoptosis through casp 3 cleavage. Recent data have suggested that the proteasome regulates the activity of the proapoptotic protein, NOXA, as proteasomal inhibition with bortezomib induces NOXA up-regulation and apoptosis in melanoma cell lines. P = phosphate. For expansion of other abbreviations, see Glossary on page 846.

FIGURE 4. MITF signaling

MITF plays the important role of a master regulator of melanocyte-specific transcription. Several pathways affect the level of MITF expression by acting on the MITF promoter. Endothelin signaling is currently the least well characterized. Better understood are the WNT and α-MSH pathways, which independently are capable of increasing MITF transcriptional activity. Specifically, α-MSH binds to the G protein–coupled MC1R, which acts through adenylate cyclase to stimulate binding of CREB1 to the MITF promoter. The activity of MITF protein is affected by several growth factors including C-KIT (also known as KIT) and hepatocyte growth factor (HGF), which stimulate MAPK-mediated MITF phosphorylation. In the phosphorylated state, MITF has higher transcriptional activity. Both increased expression of MITF and its activation by phosphorylation stimulate the transcription of MITF target genes, including those involved in melanin production (TYR, TRP1, and DCT), genes important for survival (BCL2), and numerous markers important in melanoma (MLANA, TRPM1, and SILV). E cadherin and β-catenin signaling. β-Catenin functions at the cell membrane forming a link between E cadherin and the actin cytoskeleton as well as in the cytoplasm where it participates in WNT signaling. The WNT acts through the G protein–coupled receptor Frizzled to suppress GSK3B, leading to decreased β-catenin phosphorylation, which ordinarily leads to β-catenin degradation. Such WNT signaling thus increases β-catenin levels, which causes β-catenin translocation to the nucleus, where it interacts with members of the TCF/LEF family and activates transcription of target genes, including MITF. P = phosphate; ? = mechanism unknown. For expansion of other abbreviations, see Glossary on page 846.

FIGURE 5. Stem cells in melanoma

Melanoma stem cells, also known as tumor-initiating cells, are thought be a subpopulation of tumor cells that display many attributes characteristic of normal tissue stem cells. This cellular population grows slowly, is capable of self-renewal, and is relatively resistant to drugs and toxic treatments. These attributes have potentially critical implications for cancer therapy. Current therapies, which aim to interfere with rapid cell proliferation, would likely be ineffective at destroying the melanoma stem cell compartment. As illustrated in the Figure, classic chemotherapy treatment, when successful, would primarily target the bulk of the tumor composed of rapidly proliferating daughter cells but would not eradicate slowly cycling, drug-resistant melanoma stem cells. Subsequent repopulation of the rapidly proliferating daughter cells thus leads to renewed tumor growth with clinical melanoma relapse.