Advanced melanoma has the highest per-death loss of years of potential life expectancy second only to adult leukemia. The incidence of melanoma continues to rise worldwide at approximately 3% per year, and in 2009 there were an estimated 68,720 new cases in the United States and 8650 deaths (Balch et al. 2009). According to Surveillance Epidemiology and End Results (SEER) data, roughly 4% of melanomas are metastatic at time of diagnosis. Locally advanced or recurrent unresectable, and disseminated melanoma is notoriously unresponsive to standard treatment, and is associated with a dismal prognosis. Patients with metastatic melanoma have a median survival of 8 months and 2-year survival rates of 10% to 15%. (Tsao et al. 2004). Until March of 2011, only 2 drugs were approved by the Food and Drug Administration (FDA) in the United States (US) for metastatic melanoma: interleukin 2 (IL-2) and dacarbazine (ipilimumab was approved in March of 2011 and will be discussed in this chapter). Dacarbazine is an alkalyting agent that in 1975 became the first chemotherapeutic drug approved by the US FDA for metastatic melanoma. Response rates for this regimen were 20% in a 30-year overview (Serrone et al. 2000), whereas more recent studies showed response rates in the 10-14% range (Bedikian et al. 2006; Chapman et al. 1999; Middleton et al. 2000). Temozolomide, an orally available congener of dacarbazine, was shown to be non-inferior in efficacy in comparison to dacarbazine in a randomized phase III trial (Middleton et al. 2000). Temozolomide had a similar objective response rate and overall survival compared to dacarbazine, and a significantly longer disease free survival. IL-2 was approved by the US FDA in 1998 for the treatment of metastatic melanoma. Single agent IL-2, when given in high dose intravenous boluses, leads to objective responses in 16% of patients, with complete responses in 6% and partial responses in 10%. Many patients maintained a complete response even after 7 years follow-up (Atkins et al. 2000). The toxicities of high dose IL-2 are substantial, including constitutional symptoms (fevers, chills, fatigue), hypotension, renal insufficiency, emesis, diarrhea and thrombocytopenia, and up to 2% of mortality is directly related to the treatment (Atkins et al. 2000). A systemic review of 41 randomized clinical trials including patients receiving various treatment schedules including biochemotherapy did not show improved progression-free survival (PFS) or overall survival (OS) (Eigentler et al. 2003). More recently, another meta-analysis of 18 trials involving nearly 2,500 patients with metastatic melanoma suggested an increase in response rate to biochemotherapy, but overall survival was not improved (Ives et al. 2007). Until today, biochemotherapy and combination chemotherapy built upon IL-2 and dacarbazine or temozolomide has not shown statistically significant improvement in overall survival and cannot be regarded as standard therapeutic options for metastatic melanoma. Thus, more effective treatment modalities are urgently needed.
2. Agents targeting the immune system
Increasing knowledge of T-cell regulation has uncovered potential immunologic targets for the treatment of melanoma. The interaction between antigen-presenting cells (APC) and T lymphocytes is crucial for inducing melanoma-specific T-cell responses. In addition to the antigen specific interaction between the HLA peptide complex on the APC and the T-cell receptor (TCR), several different co-stimulatory and co-inhibitory molecules modulate the T cell response. For instance, the T-cell surface molecule CD28 interacts with the B7 receptor on the APC to mediate a co-stimulatory signal (which is necessary in addition to the HLA-peptide-TCR interaction for efficient priming of the T cell), whereas the T-cell cytotoxic T-lymphocyte antigen (CTLA)-4 interacts with B7 to downregulate T-cell activation, acting as a natural “checkpoint” on the T-cell mediated immunologic response. Blocking interaction between CTLA-4 and B7 can overcome this checkpoint and enhance T-cell-mediated antitumor activity. This can be achieved by an anti-CTLA-4 monoclonal antibody (mAb). To date, two fully human anti-CTLA-4 mAb’s have been developed for metastatic melanoma: tremelimumab (CP-675,206; Pfizer Inc., New York) and ipilimumab (MDX-010; Bristol-Myers, New Jersey).
In the initial phase I study, tremelimumab was relatively well tolerated and demonstrated encouraging clinical activity including several complete responses. However, in a randomized phase III study comparing single-agent tremelimumab with either dacarbazine or temozolomide in patients with advanced melanoma, tremelimumab failed to demonstrate a significant improvement in OS in comparison to standard chemotherapy (Ribas 2008). On the other hand, ipilimumab, in a randomized phase III clinical trial comparing ipilimumab alone versus gp100 vaccine alone to the combination of ipilimumab and gp100 vaccine, resulted in improved OS of nearly four months (median survival duration of 10.1 and 10.0 months in the ipilimumab arm and the combined arm, respectively, in comparison to 6.4 months in the vaccination alone (HR 0.66; 95% CI = 0.51-0.87; p =.033 and HR 0.68; 95% CI = 0.55-0.85; p <.001, respectively). A total of 676 HLA-A*0201–positive patients with unresectable stage III or IV melanoma, whose disease had progressed after one or more of the following therapeutic regimens: dacarbazine, temozolomide, fotemustine, carboplatin, or interleukin-2, were included in this study. This was the first randomized clinical trial that showed a statistically significant improvement in OS for metastatic melanoma (Hodi et al. 2010). This agent has now received FDA approval for use in metastatic melanoma.
The activity and side-effect profile of anti-CTLA-4 antibodies have several characteristics that reflect their immune-mediated mechanism of action. Objective responses observed in patients with metastatic melanoma with either tremelimumab or ipilimumab were seen in approximately 7-10% of patients. Remarkably, as much as 70% of responses were durable (Hodi et al. 2010; Ribas 2008). The unique pattern of response to CTLA-4 mAbs such as initial apparent progression of disease, even with emergence of new lesions, followed by regression and responses over the course of several months to years, has been seen with these agents (Hamid 2007). The recognition of this unusual kinetics of response has led to a
proposal of immune-related response criteria (irRC) when evaluating treatment responses to anti-CTLA-4 mAb’s (Wolchok JD 2009) (table 1). Immune related response criteria should replace currently accepted Response Evaluation Criteria in Solid Tumors (RECIST) for the evaluation of melanoma patients treated with immunotherapy. Indeed, it is possible that clinical efficacy of tremelimumab may have been missed in the phase III clinical trial owing to the failure to recognize that immune response criteria are needed for adequate response assessment.
The growing clinical experience with anti–CTLA-4 mAbs also identifies a constellation of autoimmune side effects, which have been designated “immune-related adverse events” or irAEs (Di Giacomo et al. 2010). These irAEs, including rash, autoimmune colitis, hepatitis,
|New, measurable lesions|
(i.e., ≥5 × 5 mm)
New, nonmeasurable lesions (i.e., <5 × 5 mm)
|Always represent PD|
Always represent PD
Changes contribute to defining
BOR of CR, PR, SD, and PD
Disappearance of all lesions in two consecutive observations not less than 4 wk apart
≥50% decrease in SPD of all index lesions compared with baseline in two observations at least 4 wk apart, in absence of new lesions or unequivocal progression of non-index lesions
50% decrease in SPD compared with baseline cannot be established nor 25% increase compared with nadir, in absence of new lesions or unequivocal progression of non-index lesions
At least 25% increase in SPD compared with nadir and/or unequivocal progression of non-index lesions and/or appearance of new lesions (at any single time point)
|Incorporated into tumor burden|
Do not define progression
(but preclude irCR)
Contribute to defining irCR
(complete disappearance required)
Disappearance of all lesions in two consecutive observations not less than 4 wk apart
≥50% decrease in tumor burden compared with baseline in two observations at least 4 wk apart
50% decrease in tumor burden
compared with baseline cannot be established nor 25% increase compared with nadir
At least 25% increase in tumor burden compared with nadir (at any single time point) in two consecutive observations at least 4 wk apart
and less frequently hypophysitis and uveitis tend to be not harmful provided the clinician is aware of these unusual side effects, monitors the symptoms adequately, and knows when to intervene. IrAEs are thought to be a result of non-specific or cross-reactive tissue damage caused by activated T-cells. The most commonly reported rash is a macular and papular exanthema affecting trunk and extremities, which has been reported in 47–68 % of patients (O'Day et al. 2010; Weber et al. 2009; Wolchok et al. 2010). These dermatologic irAEs are generally mild and symptomatic relief is provided by antihistamines, while grade 3/4 autoimmune dermatitis is successfully treated with slowly tapered corticosteroids. At our institution, we reported the unique occurrence of hair depigmentation following therapy with CTLA-4 monoclonal antibody that correlated with durable clinical responses (figure 2). Six of 43 patients developed sudden hair depigmentation starting with the eyebrows, and continued to have either complete depigmentation of all body hair or the development of diffuse vitiligo. The median time to depigmentation was 10 months. All of these patients achieved partial response or complete response that sustained during the follow-up period of 24-36 months (Pavlick 2010).
Diarrhea due to immune-mediated colitis is the most frequent gastrointestinal irAE; if untreated, it may lead to serious complications such as intestinal perforation (<1%) (Beck et al. 2006). Although grade 3–4 diarrhea is generally reversible with standard anti-inflammatory therapies, experience suggests that patients with colitis treated with high-dose corticosteroids combined with mesalamine, and with seeming resolution of the adverse effects, can have an early recrudescence of the symptoms requiring either prolonged corticosteroid administration, tumor necrosis factor (TNF) blockade with infliximab, or prolonged bowel rest with total parenteral hyperalimentation (Yang et al. 2007). Other irAEs such as hepatitis, iridocyclitis and endocrinopathies including hypophyisitis, hypothyroidism and adrenal insufficiency are uncommon but worth mentioning. Hepatitis generally presents as asymptomatic rise in the levels of alanine aminotransferase (ALT) and aspartate aminotransferase (AST). In most cases, a delayed schedule of anti-CTLA 4 mAb leads to normalization of these levels. However, unlike other irAEs, hypophysitis, which occurs in 1.5% of patients treated with ipilimumab (Hodi et al. 2010) often does not improve with high-dose corticosteroids, and permanent hormone replacement therapy is required. Similarly the frequency of hypothyroidism and adrenal insufficiency is 1.5% each (Hodi et al. 2010). Interestingly, development and increasing severity of irAEs generally associate with tumor regression and with a prolonged time to relapse; nevertheless, this association is not absolute and patients without evidence of irAEs have also experienced significant clinical responses (Downey et al. 2007; Kahler and Hauschild 2010).
Melanoma is notorious for its propensity to metastasize to the brain. Up to 75% of patients develop this complication in autopsy series (de la Monte et al. 1983; Patel et al. 1978), and at least one third to one half of patients in clinical practice are found to have one or more brain metastases during the clinical course of their metastatic disease (Amer et al. 1978; Sampson et al. 1998). Brain metastasis portends a grave prognosis. The currently available treatment options for brain metastasis from melanoma include surgical and radiation therapies which may improve the outcome of selected patients with brain metastasis. The most favorable outcomes have been reported in patients who are candidates for stereotactic radiosurgery (Fife et al. 2004). Even though systemic treatment using cytotoxic agents, such as temozolomide in the US, often is employed in the setting of brain metastasis by virtue of the ability of temozolomide to cross the blood-brain barrier, its impact on overall survival remains unproven (Agarwala et al. 2004; Antonadou et al. 2002; Hwu et al. 2005). Up to date, no single treatment modality has been systematically tested in a randomized clinical trial to demonstrate an impact on disease-specific survival. Encouragingly, treatment with CTLA-4 mAb has resulted in durable control of brain metastases (Hodi et al. 2008b; Margolin et al. 2010; Schartz et al. 2010). While the phase III trials with ipilimumab excluded patients with prior or current brain metastasis, a parallel expanded access trial of ipilimumab enrolled patients who were later found to have small brain metastases that had not been treated with radiation. Among those patients, several were noted to have tumor regression in the brain. This unexpected observation led to a clinical trial initiated by the Cytokine Working Group specifically designed for melanoma patients with one or more brain metastases that did not require immediate surgery or radiotherapy. In this trial, patients were candidates for ipilimumab therapy if they had no systemic disease, untreated disease, or were no longer responding to a prior regimen for systemic disease. Brain metastases and extracranial disease were assessed separately, using standard criteria for tumor assessment based on the World Health Organization system and a composite global response status. The results of this trial, reported at the 2010 ASCO meeting, were encouraging. At week 12, for brain alone, the disease control rate was 21.5% (5 of 51 patients achieved partial response, and 6 of 51 patients achieved stable disease); the duration of brain disease control ranged from 3 to 15 months (Lawrence 2010).
The programmed death 1 (PD-1) receptor is a negative regulator of antigen-activated T cells (Fourcade et al. 2009). It bears homology to CTLA-4 but provides distinct co-inhibitory signals. The cytoplasmic domain of PD-1 contains 2 tyrosine signaling motifs that can attenuate the TCR/CD28 signal (Parry et al. 2005). There are two known ligands for PD1: B7-H1/PD-L1 (hereafter PD-L1), the predominant mediator of PD-1–dependent immunosuppression, and B7-DC/PD-L2. PD-L1 is expressed by many tumors including melanoma, and its interaction with PD-1 resulted in tumor escape in experimental models (Iwai et al. 2002). Blockade of the PD-1: PD-L1 pathway in combination with prolonged antigen stimulation with melanoma cells augments the number of cytokine-producing, proliferating, and total NY-ESO-1-specific CD8 +T cells. In murine tumor models, the inhibition of PD-1: PD-L1 interactions was shown to reverse the generation of functional anergy and exhaustion in CD 8+ T cells (Chikuma et al. 2009), and to restore a considerable proportion of the effector activity of CD8+ T cells (Blank et al. 2004), which can reinstate their antitumor effects (Blank et al. 2004; Iwai et al. 2002; Strome et al. 2003). Current data indicate that there is a correlation between the degree of PD-L1 expression and the vertical growth of primary tumors in melanoma (Hino et al. 2010). Multivariate analysis demonstrated that PD-L1 expression is an independent prognostic factor for melanoma (Hino et al. 2010). Therefore, blockade of PD-1: PD-L1 interaction represents a rational strategy for cancer immunotherapy.
MDX-1106 (BMS-936558/ONO-4538) is a fully human immunoglobulin G4 (IgG4) mAb specific for PD-1. The drug binds PD-1 with high affinity and blocks its interaction with both PD-L1 and PD-L2. A phase I study of single agent MDX-1106 in refractory solid tumors was conducted in 39 patients with advanced metastatic non-small cell lung carcinoma (NSCLC), melanoma, castrate-resistant prostate cancer, renal cell carcinoma (RCC), or colorectal carcinoma (CRC). Although efficacy was not the primary endpoint of this phase I study, of the 39 treated patients, 1 durable complete response (CRC) and 2 partial responses (melanoma, RCC) were seen. Two additional patients (melanoma, NSCLC) had significant lesional tumor regressions, which did not meet criteria for PR. This study suggested a more benign immune-related toxicity profile for anti-PD-1 mAb than one seen associated with anti-CTLA-4 mAb. Only 1 patient with metastatic ocular melanoma developed grade 3 inflammatory colitis following five doses (1 mg/kg) administered over 8 months, which responded to steroids and infliximab, while grade 2 immune related adverse events occurred in 3 patients presenting with polyarticular arthropathies requiring oral steroids and hypothyroidism requiring hormone replacement. (Brahmer et al. 2010). Another phase I trial was conducted on 16 patients with metastatic disease including melanoma. Objective responses were documented in 37.5% of patients lasting 3-13+ months; half of the patients had melanoma and there were few irAEs (Sznol 2010).
2.3. Adoptive cell therapy
The transfer of tumor specific T cells has emerged as a promising therapeutic strategy for melanoma. Adoptive cell therapy (ACT) was pioneered at the Surgery Branch of the National Cancer Institute (NCI; Bethesda, MD), and to date, has shown some of the most impressive response rates (up to 72%) in patients with metastatic melanoma. The most developed approach is based on the ex vivo selection of highly reactive, tumor-infiltrating lymphocytes (TIL), their activation and numerical expansion before reinfusion to the autologous tumor-bearing host (Dudley et al. 2005). Adoptive transfer after a nonmyeloablative conditioning regimen was shown to result in the persistent clonal repopulation of T-cells in patients with metastatic melanoma, with the transferred cells proliferating in vivo, displaying functional activity, and trafficking to tumor sites (Dudley et al. 2002).
By eliminating competition for endogenous serum cytokines, lymphodepletion may affect survival and proliferation of the adoptively transferred TIL. Data from animal models suggested that increased levels of lymphodepletion could improve ACT efficacy. In murine models, lymphodepletion seemed to enhance the antitumor effects of transferred T cells in vivo by several mechanisms including the elimination of suppressive CD4+, CD25+ T-regulatory lymphocytes, the elimination of cellular “sinks” for homeostatic cytokines such as IL-7 and IL-15, and the engagement of toll-like receptors on antigen-presenting cells after damage to the integrity of the gut epithelial lining (Antony et al. 2005; Gattinoni et al. 2005; Paulos et al. 2007). Cyclophosphamide and fludarabine have been employed for nonmyeloablative lymphodepletion. In a series of 35 patients, Dudley et al. showed that adoptive transfer of autologous TIL after nonmyeloablative but lymphodepleting chemotherapy with cyclophosphamide and fludarabine followed by high dose IL-2 resulted in objective responses in 51% of heavily pretreated patients with metastatic melanoma (Dudley et al. 2005). Further increasing intensity of host preparative lymphodepletion by adding total-body irradiation (TBI) of either 2 or 12 Gy prior to cell transfer in cyclophosphamide and fludarabine lymphodeleted patients was shown to generate response rates up to 52% and 72%, respectively. Of the 25 patients studied in the 12 Gy TBI arm, there were 4 complete responders and at least 2 substantial partial responders (Dudley et al. 2008).
The caveat of adoptive cell therapy is its complexity of several critical preparative steps that are both labor intensive and costly which may be the major obstacle to the widespread application of this approach. Up to 4 to 6 weeks of cell culture are required to obtain adequate numbers of reactive lymphocytes for adoptive transfer. Additionally, much akin to the limitation of high dose IL-2, only patients with an excellent functional status are be able to withstand the intensity of adoptive cell therapy which may also involve TBI in addition to IL-2. On the other hand, as the data from the NCI suggest, such potent therapy in highly selected patients does have the potential to induce durable responses and a potential cure. To date, however, only small phase II clinical trials have been completed and no phase III study has been initiated. Meanwhile, further innovations beyond current TIL technology are underway. Current strategies seek to improve the yield of TIL cultures, to use vaccines to stimulate the transferred T cells in vivo, and to employ gene therapy using genetically engineered lymphocytes that express highly reactive T-cell receptors (TCRs) with specific anti-melanoma differentiation antigen activity (gp-100 or MART-1), produce IL-2 or other molecules promoting an effective immune response. The gene therapy approach is of particular interest because the anti-tumor effector T cells could be rapidly expanded for adoptive transfer. Furthermore, it could be of tremendous benefit to patients who do not have autologous reactive TIL available and would otherwise be ineligible for adoptive immunotherapy (Johnson et al. 2009; Yang et al. 2010). The objective response rates in trials employing gene therapy have been reported to be 20% to 45%, and durable tumor regression at all disease sites including the brain has also been observed (Johnson et al. 2009; Morgan et al. 2006; Robbins et al. 2011).
3. Agents targeting oncogenic signaling pathways
The MAPK pathway (figure 3) has been of keen interest in melanoma in particular since Davies et al. first reported that 66% of melanomas harbor activating somatic missense mutations in the BRAF gene (V600E), leading to constitutive activation of this pathway (Davies et al. 2002; Meier et al. 2005). The RAF/MEK/ERK pathway is an important regulator of growth, survival and migration and it is constitutively activated in many human cancers ( Fecher et al. 2008 ). Physiologically, the RAF/MEK/ERK signaling pathway is activated upon binding of extracellular ligands to growth factor receptors with intrinsic tyrosine kinase activity such as epidermal growth factor receptor (EGFR), c-Kit, platelet-derived growth factor receptor (PDGFR), vascular endothelial growth factor receptor
(VEGFR), and fibroblast growth factor receptor (FGFR). Signaling through the RAF/MEK/ERK pathway eventually leads to the transcription of hundreds if not thousands of genes related to cellular proliferation, survival, apoptosis, tumor invasion, and motility ( Fecher et al. 2008 ; Shields et al. 2007). Another kinase signaling pathway involves PI3K/AKT activating mTOR which promotes signaling to the nucleus and activates genes involved in cell growth and proliferation (Stahl et al. 2004). Studies have shown that a high proportion of melanomas carry alterations in the PI3K/AKT pathway (PTEN deletion or AKT amplification) independent of the BRAFV600E mutation (Haluska et al. 2007; Vivanco and Sawyers 2002). These findings reveal a complex, interlinked network of cellular signaling pathways that contain several potentially synergistic therapeutic targets for metastatic melanoma.
Sorafenib (Bay 43-9006) was the first RAF inhibitor to enter early clinical trials. It is a small-molecule, multi-targeted tyrosine kinase inhibitor that blocks EGFR, c-Kit, Flt-3, PDGF and VEGF in addition to BRAF (Adnane et al. 2006; Wilhelm et al. 2004). Although Sorafenib does not have meaningful activity in melanoma as a single agent (Eisen et al. 2006; Ott et al. 2010), its use in combination with dacarbazine or temozolomide led to superior objective responses and progression –free survival compared to historical response rates to these agents (Amaravadi et al. 2009). However, in a phase III randomized trial (E2603) comparing carboplatin, paclitaxel and sorafenib versus carboplatin, paclitaxel and placebo in chemotherapy-naïve patients, the futility analysis demonstrated no benefit of the three drug combination compared to the two drug chemotherapy combination. In the PRISM study, the addition of Sorafenib to the combination of carboplatin and paclitaxel as second-line treatment after chemotherapy with dacarbazine or temozolomide failed to improve progression-free survival, response rate and time-to-disease progression in metastatic melanoma patients (Hauschild et al. 2009).
Since then, second generation selective RAF inhibitors have been developed and are actively tested in various clinical trials. RG7204 (Roche Pharmaceuticals, formerly PLX4032, Plexxikon) is an oral, highly selective inhibitor of the oncogenic V600E mutant BRAF kinase, which showed promising results in early clinical studies. In a dose-finding phase I trial, 11/16 (68%) of patients with mutant BRAF metastatic melanoma achieved PR and four patients had minor responses leading to a PFS of 8-9 months (Flaherty 2009). A dose extension phase I trial with 32 patients demonstrated an objective response rate of 81% (2 CRs, 24 PRs) (Flaherty et al. 2010). The median PFS among these patients was more than 7 months. RG7204 is generally well tolerated with rash, photosensitivity, arthralgia and nausea. Of note, 31% of patients developed grade 3 squamous cell carcinoma (SCC), keratoacanthoma (KA) type. The median time to the appearance of a cutaneous squamous cell carcinoma was 8 weeks with no reported involvement of other organs. Treatment with RG7204 was not interrupted by the appearance of these skin lesions and the majority of them were resected (Flaherty et al. 2010). On the basis of these promising results, a phase II trial (BRIM 2) is now fully accrued and a phase III trial (BRIM 3) met its primary end points with RG7204 improving both PFS and OS (Chapman, PB, N Engl J Med. 2011 Jun 30;364(26):2507-16. Epub 2011 Jun 5) comparing RG7204 to dacarbazine in untreated patients with BRAF V600E mutant metastatic melanoma.
GSK 2118436 is another oral, highly potent and selective inhibitor of the V600E/K/D mutant BRAF. In a phase I/II study, treatment with GSK 2118436 led to a decrease in FDG-PET metabolic uptake with 11/14 (79%) of melanoma patients showing a decrease from baseline (range -5 to -100%), and 18/ 30 (60%) patients demonstrated a > 20% tumor decrease by RECIST at first restaging (8-9 wks). GSK 2118436 showed good tolerability and low grade nausea, vomiting, fatigue, headaches and skin changes (including low grade SCC) were the main adverse effects (Kefford 2010). A phase II study of GSK2118436 as salvage therapy (NCT01153763), and a phase III study of GSK2118436 versus dacarbazine (NCT01227889) as front line therapy for mutant BRAF metastatic melanoma patients are underway.
Judging from the data available to date and the experience in our center, selective RAF inhibitors are relatively well tolerated, high grade adverse events are uncommon. Unique side effects associated with these novel drugs are the emergence of KA and SCC. These lesions can appear as early as 2 weeks after the initiation of therapy and are reported in 15–30% of the patients (Arnault et al. 2009; Flaherty et al. 2010). Histologically, KA is almost indistinguishable from a well differentiated SCC. In contrast to KA, SCC is a malignant lesion that does not regress spontaneously and has the potential for metastasis. The biologic mechanism and natural history of these lesions in the context of RAF inhibitors, in comparison to their spontaneous counterparts, are currently unknown (Arnault et al. 2009; Robert et al. 2010).
The serine/threonine tyrosine kinase MEK acts downstream from RAF in the RAF/MEK/ERK pathway; its inhibition is an attractive anticancer strategy as it has the potential to block upregulated signal transduction through this pathway, regardless of the upstream position of the oncogenic aberration. AZD6244 (ARRY-142886, Array BioPharma, AstraZeneca) is a selective MEK1/2 inhibitor that has shown preclinical activity (Yeh et al. 2007) and demonstrated clinical activity in a phase I study with a relatively benign toxicity profile (Adjei et al. 2008). These results led to a randomized phase II trial comparing AZD6244 versus temozolomide in chemotherapy naïve advanced melanoma patients (Dummer 2008). A total of 200 patients were enrolled with 104 and 96 patients randomized to AZD6244 and temozolomide, respectively; those who progressed on temozolomide could crossover to AZD6244. Results showed a trend toward OS in patients with BRAF mutations in the AZD6244 arm (Dummer 2008). At the 2010 ASCO annual meeting, the early results of a phase I study with AZD6244 in combination with docetaxel, dacarbazine or temsirolimus suggests the presence of BRAF mutation was significantly associated with clinical responses and increased time to progression. (Patel 2010). Several phase II studies of AZD6244 are underway for advanced melanoma patients with BRAF mutations as front line therapy: for treatment-naïve patients versus temozolomide (NCT00338130), in combination with dacarbazine versus dacarbazine alone (NCT00936221), as well as salvage therapy (NCT00866177). GSK1120212 is another potent and selective inhibitor of the MEK1/2 enzymes with promising anti-tumor activity in a phase I clinical trial, resulting in response rate > 70% in advanced melanoma patients with known BRAF mutations, including one patient who was previously treated with PLX4032 (Infante 2010). This drug is currently investigated in phase II and III clinical trials for advanced BRAF mutant melanoma patients who were either previously treated with a BRAF inhibitor or not. (NCT01245062, NCT01037127).
Although targeting the MAPK pathway is a promising new therapeutic approach for the treatment of melanoma, and treatment with selective BRAF and MEK inhibitors can induce high response rates, the limited duration of these responses in most patients, most likely because of emerging resistance to these inhibitors represents a significant clinical challenge. Molecular redundancy, in part due to the existence of RAF isoforms and signaling through alternative oncogenic pathways, such as PI3K/AKT/mTOR pathway (Jiang et al. 2011; Paraiso et al. 2010), receptor tyrosine kinase (PDGFRβ)-dependent pathway (Nazarian et al. 2010) and COT (MAP3K8) (Johannessen et al. 2010), may provide the melanoma cells’ escape mechanisms to specific pathway inhibitors and underscore their ability to adapt to pharmacological challenges (Jiang et al. 2011; Paraiso et al. 2010). In preclinical models, it has been reported that acquired resistance of melanoma cells to the BRAF inhibitors was associated with rebound activation of the RAF/MEK/ERK pathway (Paraiso et al. 2010). In line with this finding, activating signals to downstream MEK/ERK has been shown to switch to ARAF (Villanueva et al. 2010) or CRAF (Montagut et al. 2008; Villanueva et al. 2010) via N-RAS upregulation (Nazarian et al. 2010) to overcome the effect of BRAF inhibition. Moreover, the majority of melanoma cells harboring the BRAF V600E mutation retained the wild-type BRAF allele which could be rescued from the effects of BRAF knock-down by extracellular growth factors such as basic fibroblast growth factor, hepatocyte growth factor or endothelin-1 (Christensen and Guldberg 2005).
Imatinib is a selective tyrosine kinase inhibitor with multiple targets, including c-Kit and PDGR receptors, and has shown to be highly efficacious in chronic myelogenous leukemia and GIST tumors (Heinrich et al. 2000). Initial phase II trials with this agent in melanoma were disappointing with no objective responses (Ugurel et al. 2005; Wyman et al. 2006). However, gain of function mutations, gene amplifications and over-expression of c-kit were subsequently reported in 30-40% of mucosal, acral and cutaneous melanomas with chronic sun damage (Curtin et al. 2006). Impressive tumor regression was documented in a patient with mucosal melanoma who carried a mutation in the juxtamembranous domain of c-Kit (exon 11) in response to single agent imatinib (Hodi et al. 2008a). Moreover, preclinical studies showed sensitivity of c-Kit mutant mucosal melanoma, providing a rationale for the use of imatinib in this melanoma type. Preliminary results of a phase II trial evaluating the effect of imatinib in patients with metastatic melanoma with c-kit aberrations were presented at ASCO 2009. Over 30% of patients achieved a response (complete and partial response), whereas 50% had disease stability (Carvajal 2009). Another phase I/II study to define safety and efficacy of imatinib in combination with temozolomide in patients with unresectable, stage III/IV melanoma is currently underway. After the data on c-kit alterations became available when the trial was already in progress, patients with mucosal, acral and chronic sun damage melanomas were preferentially enrolled in the phase II part of the study. Early results of the trial were presented at ASCO 2008 ( Fecher 2008 ). Of the 23 patients treated, 16 had been enrolled in phase I and 7 in phase II. The combination was well tolerated and demonstrated anti-tumor activity in melanoma. Of the 7 patients treated in the phase II trial, 1 patient had a CR and 6 had PRs ( Fecher 2008 ). Phase II studies with a second generation c-kit inhibitor (nilotinib) in first or second line therapies for advanced melanoma with c-Kit mutation or amplification are ongoing (NCT01168050, NCT01099514). In addition, the multi-targeted receptor tyrosine kinase inhibitor sunitinib has shown potential efficacy in patients with c-kit mutated melanoma in an early phase clinical trial (Minor 2010).
Resistance of mutant BRAF melanoma cells to RAF/MEK inhibition may also be due to activation of other survival signaling pathways such as the PI3K/AKT/mTOR pathway (figure 3) resulting in melanoma development and progression (Shao and Aplin 2010; Stahl et al. 2004; Villanueva et al. 2010). Recent reports suggested a significant correlation of increased PI3K-AKT activity with resistance to RAF/MEK inhibitors in melanoma (Gopal et al. 2010; Villanueva et al. 2010) and that inhibition of the PI3K/AKT/mTOR pathway could suppress MEK inhibitor-induced activation of AKT and resulted in synergistic cell killing with a MEK inhibitor (AZD6244) (Gopal et al. 2010). This provides a rationale for combinatorial therapy leading to dual inhibition of both RAF/MEK/ERK and PI3K/AKT/mTOR pathways. A phase II trial of the mTOR inhibitor temsirolimus (CCI-779) combined with the MEK inhibitor AZD6244 is currently recruiting treatment-naïve patients with BRAF mutant advanced melanoma (NCT01166126). In addition, inhibition of the PI3K/AKT/mTOR signaling pathway was found to sensitize melanoma cells to chemotherapeutic agents such as cisplatin and temozolomide in vitro as evidenced by enhanced apoptosis and suppressed invasive tumor growth (Sinnberg et al. 2009). A Phase II study of the mTOR inhibitor everolimus in combination with paclitaxel and carboplatin in patients with metastatic melanoma is in progress (NCT01014351).
4. Other targeted drugs currently in clinical development
4.1. PARP Inhibitors
Poly-ADP ribose polymerase (PARP) is a key enzyme in DNA repair. ADP- ribosylation is involved in DNA excision repair and inhibition of the enzyme enhances the cytotoxicity of DNA damaging agents (Durkacz et al. 1980). PARP inhibition in BRCA-deficient breast cancers has shown a favorable therapeutic index. In a phase II trial, olaparib, a novel poly-ADP ribose polymerase (PARP) inhibitor, achieved an objective response rate of 41% in women with BRCA1 or BRCA2 mutations and advanced breast cancer (Tutt et al. 2010). In melanoma, PARP-1 expression has been shown to correlate with tumor thickness (Staibano et al. 2005), a poor prognostic factor for melanoma, and over-expression of PARP-1 is correlated with recurrence and/or progression of the disease (Csete et al. 2009), suggesting that it potentially is as a promising new therapeutic target. AG014699 is the first PARP inhibitor to enter a clinical trial and was studied in combination with temozolomide in a phase II study with 40 chemotherapy-naïve metastatic melanoma patients. Of the 20 evaluable patients, there were 4 confirmed PRs and 4 SDs (Plummer 2006).
4.2 Antibody-drug conjugate
Glycoprotein NMB (GPNMB) is over-expressed in a variety of cancers including melanoma. CR011-vcMMAE is an antibody-drug conjugate comprised of a fully-human monoclonal antibody directed at the extracellular domain of GPNMB linked to a potent tubulin destabilizing agent, monomethyl auristatin E (MMAE). The enzyme-sensitive linker is designed to be stable in the bloodstream and to release MMAE inside tumor cells, resulting in cancer cell death. A phase I/II study of CR011-vcMMAE in patients with advanced, pretreated melanoma has completed accrual (NCT00412828). At the ASCO annual meeting in 2010, CR-vcMMAE at 1.88mg/kg, when given every three weeks, was reported to produce an objective response rate of 15% (Hamid 2010).
4.3. Antibodies against integrins
Integrins of the αv family, such as αvβ3 and αvβ5, are implicated in tumor-induced angiogenesis and are thought to play a role in tumor growth. CNTO 95 is a fully human monoclonal antibody directed against αv Integrins, which has demonstrated anti-tumor and anti-angiogenic activities in animal models (Trikha et al. 2004). Results of a phase I study to assess safety and pharmacokinetics of CNTO 95 in patients with advanced refractory solid tumors demonstrated good tolerability (Jayson 2004). A phase I/II multicenter, randomized and double-blind study to assess the safety and efficacy of CNTO 95, alone and in combination with dacarbazine, in stage IV melanoma patients has been completed and result pending publication (NCT00246012). Another phase I trial of CNTO 95 in patients with metastatic melanoma and angiosarcoma recently demonstrated a manageable toxicity profile, and encouragingly, 2 of 18 patients with melanoma, who had failed several therapeutic regimens, sustained stable disease 5 and 7 months, respectively (O'Day et al. 2011).
Understanding the molecular biology of melanoma and unraveling of several signaling pathways over the last few years, as well as advances in immunology have led to the development of a number of promising novel therapeutic treatment options for metastatic melanoma. Immune-checkpoint blockade with ipilimumab has demonstrated increased overall survival in a randomized phase III clinical study (Hodi et al. 2010). Other immune-related approaches, such as anti-PD1/PD-L1 and adoptive cell therapy, are on the horizon. The identification of the BRAF V600E mutation led to the development of new agents that specifically block the MAPK pathway. Improved PFS and OS as a result of treatment with the V600E mutant BRAF kinase inhibitor RG7204 in BRIM 3 have been reported. Encouraging response rates with imatinib therapy in patients with mucosal, acral, and sun-damaged skin melanomas that harbor mutations or amplification in the c-kit gene have also been documented. These results herald the era of individualized therapies based on specific tumor genotypes, of which scientifically thoughtful combinations will usher in a new standard of care for metastatic melanoma in the near future.