This article provides a review of the current medical management of patients with high-risk and metastatic cutaneous melanoma, including a review of the use of adjuvant interferon therapy and a discussion of adjuvant treatments under evaluation. The use of standard chemotherapeutic agents for metastatic disease is discussed, with an emphasis on developmental therapeutics using targeted agents. This discussion includes a review of the immune therapy for metastatic melanoma, including newer immunomodulatory agents and cellular therapeutics that are expected to significantly impact the care of these patients.
Cutaneous melanoma is a malignancy of the skin with a propensity for early invasion and widespread metastasis. More than 90% of melanoma is cutaneous, and the principles of treatment of cutaneous melanoma are sometimes applied to other melanomas, (ocular melanoma and so forth). However, the biologic differences between them can significantly alter their outcome and response to treatment. Melanoma is one of the few common cancers with rising incidence in the United States: a steady increase of approximately 3% per year. In 2009, an estimated 68,720 people will be diagnosed with cutaneous melanoma, with 8650 dying from the disease. In the United States, cutaneous melanoma currently ranks sixth in incidence, and 20% of patients will present with regional involvement or distant spread at the time of diagnosis. Melanoma also seems to occur in a younger age group, with most patients presenting between 25 and 65 years old.
Despite the overall advances in the fields of tumor biology and oncology therapeutics, the treatment of metastatic melanoma has remained a challenge, and the prognosis for patients with widely metastatic disease remains poor. The task of treating this malignancy has inspired studies that have dramatically expanded the knowledge of cancer biology. The early suggestion of an immune response, as demonstrated by the regression of lesions injected with Bacillus Calmette Guérin (BCG), has led to numerous studies that have advanced tumor immunology. The study of the immune response to melanoma has led to an understanding of the recognition by the immune system of tumors, and the mechanisms by which the tumor and host interplay. Melanoma has been a model for the study of tumor immunology, and a basis for studies on the recognition of tumor antigens and their presentation by the immune system. Because of the resistance of melanoma to standard treatments, it has been a platform for the study of new agents and has led to the use of cytokine therapies, cellular therapeutics, and gene therapies. In this article, the current systemic therapy for high-risk melanoma and metastatic, cutaneous melanoma is discussed.
Adjuvant therapy for melanoma
The majority of patients who present with melanoma have early stage disease, with 62.6% presenting with stage 0 or I disease, and an additional 23.1% with stage II disease. The standard therapy for localized melanoma is surgical resection. However, the relatively high rate of recurrence in subgroups of patients with melanoma suggests the need for adjuvant approaches for this disease. Decisions regarding adjuvant therapy for neoplastic diseases represent a balancing of risk of disease recurrence against the relative value of the agent used including its activity, toxicity, and cost. There are several risk factors that have been associated with melanoma, including primarily the involvement of lymph nodes (LN), the presence of ulceration of the lesion, and the level of invasion, which is best defined by the Breslow depth of invasion but also by the extent of invasion as defined by the Clark level. The staging system for melanoma was revised in 2003 to take into account the important risk factors for melanoma, with the goal of increased consistency and identification of various risk groups in patients with resected disease. The upstaging of ulcerated lesions, and the more routine determination of pathologic lymphatic involvement through the use of sentinel node evaluation, has led to better identification of risk in patients. Patients can be stratified into various risk groups: those with stage IA disease who have a very low risk of recurrence, those with stage Ib and IIa disease, who carry an intermediate risk, patients with IIb, IIc, and IIIa disease who have a higher risk, and those with stage IIIb and IIIc disease, who carry a very high risk of recurrence. These groups have a 5-year survival of approximately 95%, 65% to 90%, 45% to 70%, and 25% to 50%, respectively. The poor prognosis associated with the recurrent melanoma has led to numerous clinical trials attempting to decrease this risk. These studies have included the use of chemotherapy, immunotherapy including cytokines, vaccines, and cells, and biochemotherapy.
Adjuvant Chemotherapy
Oncologists traditionally have developed adjuvant treatments of malignant disease based on their activity in the metastatic disease. There are several cytotoxic agents that have demonstrated activity in the treatment of melanoma. Despite this, there are limited numbers of phase 3 studies evaluating the effectiveness of these agents in the treatment of high-risk melanoma. Dacarbazine (DTIC) is one of the only Food and Drug Administration (FDA)-approved cytotoxic agents for the treatment of metastatic melanoma, but it has limited benefit in the adjuvant treatment of melanoma. One of a few studies with DTIC given after surgical resection alone showed improvement in relapse-free survival (42% vs 30%), but this was not significant. A small randomized study evaluating the combination of carmustine, actinomycin-D, and vincristine showed an improvement in relapse-free survival at 5 years (29% vs 9%) but not in overall survival, and this regimen has not been pursued. Because of the promising results initially observed with nonspecific immunotherapy using BCG, a subsequent study (E1673) evaluated the activity of this agent alone or in combination with DTIC. This study, which was one of the largest trials (734 patients) of adjuvant therapy to date, included patients with resected stage I to III disease treated with BCG and DTIC, or BCG alone. There was no benefit seen with adjuvant BCG compared with observation in disease-free survival ( P = .84) or 5-year overall survival (67% vs 62%, P = .40). Nor was there any benefit with DTIC plus BCG over BCG alone.
A study from the German Dermatologic Cooperative Group treated a group of 444 patients with stage III disease with of low-dose subcutaneous interferon-α2B (IFN), the combination of IFN and DTIC 850 mg/m 2 , or observation alone. The study found a significant improvement in overall survival in patients treated with IFN alone compared with observation (59% vs 42%, P = .0045), yet no benefit was seen in the group of patients treated with the combination of IFN and DTIC (overall survival 45%), suggesting the DTIC actually interfered with the effect of IFN. This study is one of a few studies to indicate a significant benefit of low-dose IFN in the adjuvant setting, and this was seen in overall survival and not just recurrence-free survival. Data from this study and the others discussed here suggest that there is no benefit of chemotherapy in the adjuvant treatment of melanoma.
Adjuvant Immune-based Therapy
Cytokines
The suggestion that there was an immunologic response by the host to tumor led to interest in ways to enhance that response. The recognition of cytokines as mediators of immune signaling led to therapeutic trials in patients with metastatic melanoma. Intriguing cytokine responses observed in the metastatic setting triggered studies to determine their role in adjuvant treatment.
Interferon
Interferons are a complex family of proteins with immunomodulatory and antiangiogenic properties that are produced in response to viral infection or after T-cell activation. Their effects include upregulation of the histocompatibility antigen expression on tumor cells, and activation of various effector cells including natural killer (NK) cells, T cells, monocytes, and dendritic cells. IFN is the most extensively studied agent for the adjuvant treatment of melanoma, with several large trials suggesting a benefit for its use. Although there are numerous studies of IFN in the adjuvant setting, it was the pioneering work by Kirkwood using high doses that dramatically advanced the use of this agent.
Three large United States cooperative group trials have shown improvement in disease-free survival, and 2 of these trials showed overall survival benefit with the use of IFN. The cooperative group trial E1684 showed a 5-year relapse-free survival improvement (37% vs 26%, P = .002) and suggested an overall survival benefit (40% vs 37%, P = .024). This result corresponded to an increase in disease-free survival from 1 to 1.7 years and an overall survival from 2.8 to3.8 years. The treatment had the greatest benefit in patients in the patients with node-positive disease. This regimen, which used 20 MIU/m 2 , 5 days per week for 4 weeks, followed by 10 MIU/m 2 /d 3 days per week for 11 months, is associated with significant toxicities, including flulike symptoms, chronic fatigue, nausea, weight loss, myelosuppression, hepatotoxicity, and depression, requiring dose modification in most patients. Despite this, a quality of life analysis suggested that this toxicity was compensated for by the prevention of disease relapse. The data from E1684 led to FDA approval of IFN for the treatment of patients with resected stage IIB and stage III melanoma.
The intriguing results of the E1684 study led to 2 additional studies. The E1690 trial compared high-dose IFN to low-dose IFN and a control. This study also showed an improvement in 5-year relapse-free survival benefit (44% vs 35%, P = .05); however, it failed to show an overall survival benefit. The trial again suggested the importance of dose of the agent, as the low-dose IFN arm had an intermediate relapse-free survival rate, which was not statistically significant. A follow-up analysis of pooled data from the E1684 and E1690 trials showed that patients treated with IFN had continued improvement in relapse-free survival, with a reduction of risk of relapse of 23% ( P = .006), but no improvement in overall survival. Qtwist Analysis has shown an improvement in the quality of life associated survival. Finally, the E1694 trial compared high-dose IFN to a ganglioside vaccine, GM2-KLH/QS21, and again showed an improvement in 5-year relapse-free survival (62% vs 49%, P = .0007) and also showed an overall survival benefit (78% vs 73%, P = .015). This result again supported the benefit of IFN, although critics have pointed out that this may have been a negative effect of the vaccine rather than a positive effect of the IFN. However, a small study, E2696, showed that relapse-free survival was better with the combination of IFN plus GM2-KLH vaccine compared with vaccine alone, lending support to the value of IFN. Questions surrounding the use of IFN include the importance of dosing and schedule of administration. The European Organization for Research and Treatment of Cancer ( EORTC) 18,952 study evaluated higher dose IFN, using 10 MIU subcutaneous daily 5 days per week for 4 weeks followed by 11 months at 10 MIU 3 days per week for 11 months. Relapse-free survival was improved, but not overall survival. Another trial (E1697) is an ongoing randomized trial that uses only the first 4 weeks of treatment for patients with intermediate and higher-risk disease.
Another question surrounding the use of IFN is the level of benefit related to nodal involvement. In the Sunbelt melanoma trial, 3619 patients underwent sentinel LN staging based on histology or molecular analysis using reverse transcriptase-polymerase chain reaction (RT-PCR). Patients with single LN involvement were randomized to high-dose IFN or observation, and patients with more than one LN or extranodal extension were treated. There was no statistical benefit for patients with a single positive LN who were treated with IFN versus observation in 5-year disease-free survival (70.2% vs 73.2%) or overall survival (75.4% vs 72.9%). As expected, the patients with single LN involvement had significantly better survival than patients with more advanced nodal involvement (44.5% disease-free survival, 52.9% overall survival). No benefit was observed with IFN for patients with LN involvement as detected by RT-PCR.
Potential ways to improve on the results IFN-based therapy include adjustment of the schedule, to improve efficacy as well as to decrease toxicity, and its combination with other agents including other biologic agents, chemotherapy, and vaccines. One of the observations of the data from the E1684 was the early separation of the curves between the IFN and observation arms, suggesting the importance of the priming phase of IFN. The E1697 study is a randomized study that compares high-dose IFN for 4 weeks versus observation, that continues to accrue patients with high-risk stage II or early stage III disease. A study from Greece of 364 evaluable patients showed no difference in the disease-free or overall survival in patients treated with induction therapy alone or induction plus 1 year of IFN treatment. The E1697 study randomized patients to 4 weeks of IFN to observation and continues to accrue patients. The Italian melanoma trial is evaluating sequential induction phase versus the standard E1684 regimen, and to date has demonstrated a higher dose achieved, but evaluation is ongoing.
The pegylation of IFN alters its pharmacokinetics, allowing for decreased frequency of administration with the potential for improved toxicity-benefit profile. The EORTC conducted a trial of 1256 patients with resected stage III disease in which patients were randomly assigned to observation or pegylated interferon-α2b (PEG-IFN) 6 μg/kg per week for 8 weeks induction then 3 μg/kg per week. Despite the intended duration of 5 years of treatment, the median length of treatment was only 12 months, with discontinuation in 31% of patients related to toxicity. There was an improvement in 4-year recurrence-free survival in the IFN-treated patients compared with the observation group (45.6% vs 38.9%, hazard ratio 0.82, P = .01). The activity of IFN as a monotherapy led to its evaluation in combination therapy. The addition of interleukin-2 to IFN has been studied but did not show a benefit. Although the E2696 trial was not designed to compare the combination of vaccine with IFN to IFN alone, the data were similar to those seen with IFN alone in the previous E1684 and E1690 trials. The combination of IFN with the allogeneic melanoma cell lysate vaccine, Melacine, also failed to show improvement over IFN alone.
Granulocyte-macrophage colony-stimulating factor
Another agent with potential adjuvant activity is granulocyte-macrophage colony-stimulating factor (GM-CSF). In vitro, GM-CSF can activate macrophages to become cytotoxic for melanoma cells and may mediate proliferation, maturation, and migration of dendritic cells (DC). In an early single-arm study, 48 patients with stage III disease received adjuvant GM-CSF, 125 μg subcutaneously for 14 days every 28 days. The data showed an improved survival compared with matched historical controls (37.5 vs 12.2 months, P <.001). Overall the drug was well tolerated, with some myalgias, weakness, and fatigue. In a follow-up study presented in abstract form, of 98 patients treated with GM-CSF for 3 years, a melanoma-specific survival rate of 60% was observed, and this was significantly prolonged compared with a group of 142 patients treated for 1 year with GM-CSF (hazard ratio 0.61, P = .047). It is noteworthy that 2 patients treated for 3 years with GM-CSF developed acute myeloid leukemia. In a small study of 42 patients with high-risk (stage III B/C, IV), resected melanoma, GM-CSF (125 μg) was administered subcutaneously for 14 days every 28 days. Median overall survival was 65 months. GM-CSF treatment resulted in an increase in mature DC, and was associated with remission or delayed recurrence. Despite the encouraging results in this very high-risk population, randomized phase 3 data are still required to clearly demonstrate the activity of this agent. Data from the E4697 study that has completed accrual should provide a clearer answer to the value of GM-CSF as a single agent in the adjuvant setting for high-risk melanoma. Studies are already looking at the combination of this agent to try to improve its efficacy.
Adjuvant Vaccines
Although vaccines have been used for the treatment of patients with metastatic disease, they lend themselves to adjuvant therapy in which there is minimal residual disease, there is less tumor heterogeneity, there is more time for the therapy to have its effect, and effects of the tumor on host immune function are minimized. The historic understanding of nonspecific immunotherapies and the viral vaccine production, as well as the brilliant work done to identify the specific antigen recognized by the immune system, has led to an array of vaccine strategies. Vaccines tested in the adjuvant setting include tumor cells, gangliosides, and tumor-associated antigen peptides.
Tumor cell vaccines
Probably the best known and widely studied of the melanoma vaccines is the tumor cell lysate vaccine called Melacine, which is an allogeneic melanoma cell lysate combined with a “detoxified” Freund adjuvant. In the original study by Mitchell and colleagues in stage IV patients, signs of clinical regression were observed in 5 of the 17 patients with measurable disease. Patients with resected stage II melanoma treated with Melacine showed a prolongation of relapse-free survival and overall survival. However, this result occurred only in patients with certain human leukocyte antigen (HLA) types. A phase 3 study compared Melacine to the “Dartmouth” chemotherapy regimen consisting of DTIC, cisplatin, lomustine, and tamoxifen, and showed no significant difference in response rate (7.1% vs 10%). The SWOG-9035 trial was a phase 3 study of patients with T3N0 intermediate risk melanoma. The 600 eligible patients were randomized to receive Melacine or observation. There was a similar relapse-free survival between patients receiving vaccine (65%) and those on the observation arm (63%). On further evaluation, the patients who had matching HLA type to the allogeneic lines were more likely to derive clinical benefit. In a study comparing Melacine and low-dose versus high-dose IFN, there were no differences in clinical outcome, but toxicity was lower in the low-dose IFN arm.
In a recent study testing a modified tumor cell vaccine strategy, patients with early stage melanoma were treated with a mixture of apoptotic/necrotic allogeneic melanoma cell lines loaded onto DC. One patient with stage IIC disease and 7 of 8 patients with stage III disease had disease-free survival at 49.5 months Very few other vaccine strategies using the adjuvant activity of DC have been tested in high-risk patients.
Tumor-antigen peptide vaccines
Several small studies have explored the efficacy of tumor-associated antigenic peptides (gp100 only or gp100 plus tyrosinase) with or without interleukin (IL)-2, GM-CSF, Montanide, DC, and incomplete Freund adjuvant. Peptide-specific T cells were induced in most patients. However, the lack of clinical efficacy in these studies prompted a recent vaccine study that incorporates a larger number of antigenic epitopes to broaden the antitumor immune response. This prospective, randomized clinical trial (3318) tested the ability of a vaccine with 12 defined, shared melanoma peptides from melanocytic differentiation proteins and cancer testis antigens to induce antitumor immune responses. This vaccine was compared with a 4-peptide vaccine with only melanocytic differentiation peptides. The analysis indicates that patients who developed a T-cell response have a higher probability of longer disease-free survival ( P = .041), and that the clinical outcome correlates significantly with the T-cell response to certain peptides. Cumulative responses to the 12-peptide vaccine were greater compared with the 4-peptide vaccine ( P = .12). Preliminary data suggest that median survival for the entire patient population on this trial approaches 3 years, which is a favorable finding compared with published data for this patient population.
Ganglioside vaccines
Gangliosides are immunogenic glycolipids, and the GM2 ganglioside is overexpressed on the surface of the most melanomas. Vaccines were developed in the 1980s, and one of the early studies compared vaccination with a GM2/BCG vaccine compared with BCG alone. The study showed that GM2 antibody was associated with a prolonged disease-free interval and survival. Although the trial failed to show a statistically significant improvement in disease-free or overall survival, it was underpowered, with only 58 patients enrolled to the GM2/BCG arm. More recent studies have used a vaccine with GM2 conjugated to the keyhole limpet hemocyanin protein and QS21 as an adjuvant. The EORTC18961 was the largest adjuvant trial to date, with 1314 patients with stage II disease. Patients were randomized to treatment with GM2-KLH21 (n = 657) or observation (n = 657), and had no difference in relapse-free survival (75.1% vs 77.8%) or overall survival (89.2% v 92.4%), with a hazard ratio of 1.00. In an Eastern Cooperative Oncology Group (ECOG) study of 880 melanoma patients, there was no difference in outcome between patients treated with GM2-KLH plus BCG versus IFN alone.
Adoptive Immunotherapy
Adoptive immunotherapy for melanoma is based on the transfer of effector cells such as tumor infiltrating lymphocytes (TIL) to mediate cancer regression. Most of these studies have evaluated their efficacy in patients with large tumor burdens. A recent study evaluated infusions of TIL in combination with IL-2 versus IL-2 alone in patients with regional LN recurrence but no visceral metastasis. Eighty-eight patients were randomized to receive 2 doses of autologous TIL plus IL-2 (n = 44) or IL-2 alone (n = 44). There was no difference in relapse-free survival for the patients who received TIL/IL-2 versus IL-2 alone. However, in patients with 1 tumor-invaded LN (n = 34), the relapse rate was lower in patients who received TIL/IL-2 (30%) versus IL-2 alone (68%) ( P = .0219). There was no difference in relapse rate in patients with more than 1 tumor-invaded LN. Overall survival was higher in patients who received TIL/IL-2 (73%) versus IL-2 alone (32%) with 1 tumor-invaded LN. Survival was not different in patients with more than one tumor-invaded LN. A current phase 3 study is evaluating the effects of adjuvant TIL in patients with 1 tumor-invaded LN.
Adjuvant Biochemotherapy
Although a benefit for the combination of chemotherapy and nonspecific immunotherapy had not been observed in previous trials, the high response rates associated with treatment with biochemotherapy (DTIC, cisplatin, velban, IFN, and IL-2) in the metastatic setting led to a comparison of its benefit compared with high-dose IFN in the adjuvant setting. The study enrolled 138 of the intended 200 patients randomized to receive standard high-dose IFN as described in the E1684 study, IFN 10 million units/m 2 subcutaneously for 52 weeks, or biochemotherapy. This study was ended early due to slow accrual. At the time of closure, a futility analysis was performed that showed no difference in 2-year (66% vs 68%) or 5-year (58% vs 59%) disease-free survival or in overall survival. Another study, Intergroup S0008, is similarly studying biochemotherapy versus standard IFN in a randomized phase 3 study.
Summary of adjuvant therapy
There remain many unanswered questions regarding the optimal treatment of patients with resected melanoma and the optimal use of these adjuvant therapies available. However, the data with IFN have led to FDA approval of the agent for the adjuvant therapy for patients with intermediate- or high-risk melanoma, and it has now gained acceptance as a standard agent for patients with melanoma larger than 4 mm or with LN involvement. The 2009 National Comprehensive Cancer Network (NCCN) guidelines suggest IFN as a category 2B option for patients with stage IIb or greater disease. Despite this, the recommendation is not universal, and other investigators have concluded that there is currently no standard adjuvant therapy following resection of melanoma. Innovation of newer modalities and patient participation in clinical studies will ultimately lead to improvements in the care of patients with increased risk of recurrent disease.
Summary of adjuvant therapy
There remain many unanswered questions regarding the optimal treatment of patients with resected melanoma and the optimal use of these adjuvant therapies available. However, the data with IFN have led to FDA approval of the agent for the adjuvant therapy for patients with intermediate- or high-risk melanoma, and it has now gained acceptance as a standard agent for patients with melanoma larger than 4 mm or with LN involvement. The 2009 National Comprehensive Cancer Network (NCCN) guidelines suggest IFN as a category 2B option for patients with stage IIb or greater disease. Despite this, the recommendation is not universal, and other investigators have concluded that there is currently no standard adjuvant therapy following resection of melanoma. Innovation of newer modalities and patient participation in clinical studies will ultimately lead to improvements in the care of patients with increased risk of recurrent disease.
Therapy for metastatic melanoma
The treatment of metastatic melanoma remains a challenge for the clinician. Although the goal of early systemic adjuvant therapy for melanoma has been a central focus in the systemic management of the disease, there remains a high rate of recurrence. The poor outcomes associated with standard cytotoxic chemotherapeutic agents have led to development of numerous alternative therapies for melanoma treatment, particularly the use of immunotherapeutic treatments. An understanding of tumor biology piloted the development of agents with a more specific target of action on the tumor. The tremendous success of targeted agents in other diseases, such as imatinib (Gleevac) in chronic myelogenous leukemia, and the relative success of other targeted agents in renal cell carcinoma (sorafanib, sutinib, temsirolimus) has prompted trials of such targeted agents in melanoma as well.
In discussing the treatment of metastatic melanoma, it is important to consider outcome measurement. Treatment benefit is usually measured in terms of response rate, and disease-free or overall survival rates. Response rate may overestimate the benefit because these responses may be of only short duration. Overall and disease-free survival may serve to underrepresent the benefit to the individual patient. Measurements of disease stabilization are often used to try to reflect the benefit of some of the newer agents that have antiangiogenic or antiproliferative effects, but do not necessarily induce dramatic tumor shrinkage. In addition, the response rates with immunotherapeutic regimens are often low but may result in prolonged disease-free survival, and thus may actually be very important to the outcome for the individual patient.
Recurrent melanoma is associated with dismal clinical outcomes. Much effort has been extended to improve response rates by evaluating the efficacy of chemotherapy, combination chemotherapy, apoptosis-inducing agents, targeted agents, immune-based therapy, and biochemotherapy.
Chemotherapy
Melanoma is considered a chemotherapy-resistant disease, and systemic chemotherapy has failed to significantly improve the survival of patients with nonresectable metastatic melanoma. The disease frequently becomes refractory to the agents even after initial responses are observed. Despite the lack of curative effect for the patient with advanced metastatic disease, chemotherapy continues to play a role in palliation of the disease. Although many agents have been tested and used for melanoma treatment, single-agent chemotherapy has generally been considered ineffective. One of the challenges in decision-making for the medical oncologist has been the often encouraging results of single-institution trials that are not confirmed by subsequent phase 2 trials, or the conflicting data between studies and their interpretation. Despite the poor overall outcome with these agents, they are still in common use in the clinic.
Alkylating agents
Dacarbazine (DTIC) is the most active single agent, and is the only cytotoxic agent that has been approved for the treatment of advanced melanoma. DTIC is therefore the standard against which other treatments are tested. DTIC is a prodrug of the alkylating agent 5-(3-methyltriazen-1-yl)imidazole-4-carboximide (MTIC). The drug is generally well tolerated, with nausea as its major side effect, which can be controlled with current antiemetic therapy. The response rate for single-agent DTIC in melanoma ranges from 15% to 25%, but these are generally of short duration (3–6 months), and the complete response rate is only 5%. Because of its activity, DTIC has been tested alone or in combination therapy.
Temozolomide (TMZ) is a relatively new alkylating agent. TMZ is similar to DTIC in that it is a prodrug that is converted to MTIC. In contrast to DTIC, TMZ spontaneously converts to MTIC under physiologic conditions. TMZ has the advantage of being 100% bioavailable, and can therefore be administered orally, which significantly improves the ease of use of the agent. In addition, TMZ has extensive tissue penetration including penetration of the blood-brain barrier and the cerebral spinal fluid. TMZ was shown to have an objective response rate of 21% (12 of 56 patients) in a phase 2 study, with a median survival time of 5.5 months. In a randomized phase 3 study, TMZ had a slightly higher response rate (13.5% vs 12.1%) and median survival (7.7 vs 6.4 months) than DTIC, neither of which was statistically significant.
Attempts have been made to improve the results with TMZ, primarily through changes in schedule or by the addition of additional agents. A phase 2 study was done to further assess the impact of the extended schedule used in the 2 previous studies (75 mg/m 2 /d for 6 weeks every 8 weeks). TMZ cytotoxicity is mediated through methylation of DNA, and a prolonged schedule has the potential to deplete methyl guanine methyltransferase (MGMT) levels, an enzyme that is involved in DNA repair, with improved cell killing. However, the response rate was 12.5% with no complete responses and an estimated 18-month survival of 27%, which was similar to trials with the standard 5-day regimen, and there was no correlation between MGMT and response. A trial with the MGMT inhibitor lomeguatrib similarly did not suggest an improved therapeutic ratio that would warrant its use in melanoma. A study combining TMZ with thalidomide demonstrated a 32% response rate in 38 patients treated. However, a follow up study could not demonstrate an improvement with this regimen. IFN has similarly been administered in combination with TMZ, and a phase 2 study that used a combination of PEG-IFN with TMZ resulted in a 31% response rate in 35 patients with metastatic melanoma, including 3 complete responses. A study by the Dermatologic Oncology Group using TMZ and PEG-IFN showed an 18% response rate and a survival of 9.4 months. The large randomized study by the Dermatologic Cooperative Oncology Group comparing TMZ using the 5-day regimen with or without standard IFN given 3 days per week again showed an improvement in response rate (24% vs 13%) but failed to show a statistical improvement in survival (9.7 vs 8.4 months). Although TMZ has not been approved by the FDA for the treatment of metastatic melanoma, it has become a standard agent for the treatment of patients who are not candidates for cytokine or experimental studies.
The nitrosoureas are a group of alkylating agents that act by cross-linking DNA. Carmustine (BCNU) and lomustine have similar response rates to that of DTIC (10%–20%) but are generally associated with more toxicity, including myelosuppression and alopecia. In clinical practice, these agents have a limited role as single agents, but have been used in combination chemotherapy. Fotemustine is a chloroethyl nitrosourea that more rapidly crosses the blood-brain barrier, which has been more extensively studied in Europe. In a phase 3 trial, fotemustine showed improved response rates over DTIC (15% vs 7%), but this did not translate into a survival advantage (7.3 vs 5.6 months) and although the drug is currently available as first-line therapy in some European countries, it is currently not available in the United States.
The platinum compounds, cisplatin and carboplatin, have a modest activity in patients with melanoma, with a response rate of approximately 15% to 20%. Attempts made to improve the results with cisplatin have resulted in increased toxicity. Cisplatin and carboplatin have found more use in combination with other agents.
Microtubule inhibitors
Microtubular toxins and microtubular disassembly inhibitors have both been used in patients with metastatic melanoma, and have some activity. The vinca alkaloid vinblastine, based on its modest activity and limited toxicity, has primarily been used in combination therapy. In a small study, vinorelbine (Navelbine) was used in combination with tamoxifen, with a 20% response rate, 3 of the 30 patients having a response of longer than 12 months. Other studies with vinorelbine have not demonstrated activity either as first-line or second-line therapy.
Paclitaxel and docetaxel are microtubule disassembly inhibitors with antitumor activity in a variety of neoplastic diseases. Paclitaxel has been evaluated in several phase 1 and 2 studies, and has demonstrated an approximately 12% to 16% response rate in previously untreated patients. Paclitaxel is commonly used in combination with carboplatin in other malignancies, and was similarly tested in melanoma. Rao and colleagues reported a 26% response rate in second-line therapy for patients with melanoma, and this regimen is now being used as a “backbone” for the addition of other agents, despite the limited activity of paclitaxel alone or in combination with carboplatin observed in other studies. Docetaxel is another microtubule disassembly inhibitor. A phase 2 docetaxel study showed a 12.5% response rate in melanoma with one of the patients having a durable complete response, and it still being actively studied in other combinations, with some benefit.
The NCCN guidelines for treatment of melanoma in 2009 list several therapies for the treatment of metastatic disease. The recommended agents include the alkylating agents DTIC and TMZ, either as single agents or as combinations, the microtubule inhibitor paclitaxel alone or with a platinum agent, and IL-2. However, all of these recommendations are level 2B, indicating the lack of a regularly effective agent for this disease.
Combination Chemotherapy
A common method to try to improve responses is to combine agents with additive or synergistic activity with nonoverlapping toxicity. There are a numerous combinations of chemotherapy for melanoma that have been and are being developed and studied. These regimens have generally employed DTIC or, more recently, TMZ. The most notable of these regimens is the CVD regimen, which combines cisplatin, velban, and DTIC, and the Dartmouth regimen, which combines cisplatin, carmustine, DTIC, and tamoxifen. In a large phase 3 study comparing the CVD regimen to DTIC alone, there was a trend toward improved response and survival. The Dartmouth regimen originally resulted in a 55% response rate in the initial series of 20 patients with metastatic melanoma. Despite initial promising results, subsequent phase 3 studies have not confirmed an advantage for combination chemotherapy. Although the response rate with combination was improved with the Dartmouth regimen over DTIC alone (18.5% vs 10.2%), this did not translate into an overall improvement in median survival (7.7 vs 6.3 months). Despite these data, the use of combination chemotherapy is still commonly used in the care of patients.
Apoptosis-inducing Agents
Oblimersen is a cytotoxic agent shown to downregulate Bcl-2 and increase apoptosis in human cancer xenografts. In a large clinical study of 771 patients, the combination of oblimersen sodium plus DTIC was shown to increase progression-free survival over DTIC alone (2.6 vs 1.6 months, P <.001). Although an improvement in overall survival could not be clearly shown (9.0 vs 7.8 months, P = .077), oblimersen did increase survival compared with DTIC alone in patients who had a normal lactate dehydrogenase (LDH) at baseline (11.4 vs 9.7 months, P = .02). Oblimersen was well tolerated, with some increase in myelosuppression. In addition to showing a clinical improvement, the study further indicates the heterogeneity of the patients with melanoma and the need to clearly identify patient populations when studying treatments for this disease. The FDA has not approved oblimersen for the treatment of melanoma. One aspect that led to the lack of FDA approval was the delay in follow-up scans in the oblimersen group compared with the group treated with DTIC, which created a time-lag bias, indicating the importance of care in study design and conduction. An additional phase 3 trial is ongoing to further evaluate this agent.
Elesclomol (STA-4783, Synta) is an inducer of reactive oxygen species (ROS), or heat shock proteins, which leads to apoptosis in melanoma. In a double-blind randomized multicenter study, paclitaxel 80 mg/m 2 weekly was given alone or in combination with elesclomol (213 mg/m 2 ). The combination showed an improvement in objective response rate (15% vs 3.6%), with one complete response in the 53 patients in the combination treatment. This result was also associated with an improvement in progression-free survival (3.7 vs 1.8 months, hazard ratio 0.53, P = .035) with a 1-year overall survival rate of 49%. This agent is now undergoing phase 3 testing in the SYMMETRY trial.
Targeted Agents in Melanoma
New cellular and molecular techniques are changing the practice of oncology. These molecular techniques are providing a means for sorting out the heterogeneity inherent in such a complex process as cancer and its interactions with the host. These techniques have provided prognostic information in breast cancer patients using gene array analysis of tumors, as is currently done with the OncoTypeDX analysis ; they have also started to allow predictions of tumor response to chemotherapeutic agents. The recent finding that colonic tumors that have a kras mutation are resistant to the EGF-receptor antagonist allows sparing of the patient from the expense and toxicity of the agents. These molecular techniques have allowed for a much more detailed understanding of the cell signaling pathways present in melanoma, which can then be used as targets for therapy. These targeted agents are designed to block cell signaling, to disrupt the pathways involved in angiogenesis, growth, and proliferation, and to enhance tumor apoptosis. These techniques are likely to lead to the individualization of treatment based on assessments of the genetic and epigenetic aspects of the tumor as well as the host.
Tamoxifen
The identification of estrogen receptors in melanoma led to initial trials of hormonal therapy for the disease, and tamoxifen, an estrogen receptor antagonist, might be considered one of the first targeted agents used for the therapy for melanoma. Tamoxifen was initially used as a single agent and then in combination with various chemotherapeutic regimens. Although initial studies suggested a benefit for tamoxifen as a single agent in the treatment of metastatic melanoma, subsequent studies showed a response rate of only 5%. In addition, evaluation of melanoma samples using immunostaining failed to demonstrate estrogen receptors. Tamoxifen could have several other effects including effects on angiogenesis, synergic effects with chemotherapy, and reversal of multidrug resistance. A phase 3 study demonstrated an improvement in response (28% vs 12%, P = .03) and survival (48 weeks vs 29 weeks, P = .02) after DTIC with tamoxifen compared with DTIC alone. In a phase 2 trial of the Dartmouth regimen, the omission of tamoxifen similarly was shown to reduce response rates. A phase 3 study of dacarbazine and carboplatin with or without tamoxifen showed no difference in response or survival. A multi-arm study showed no benefit from the addition of tamoxifen to DTIC or DTIC and IFN. A subsequent study showed no loss of activity to the Dartmouth regimen when tamoxifen was removed. A meta-analysis of the studies with tamoxifen further indicated that, when added to chemotherapy, it did not improve the response rate or survival. Mori and colleagues demonstrated that patients with estrogen receptor α (ER-α) methylation correlated with survival in patients treated with a biochemotherapy regimen containing tamoxifen, suggesting that the variability in the trials may be related to the ER-α methylation status of the patients in the study.
Vascular inhibitors
The production of vascular endothelial growth factor (VEGF) has been implicated in tumor-induced angiogenesis, and inhibition of VEGF-induced angiogenesis can suppress the growth of tumors in murine models. Bevacizumab (Avastin) is a humanized murine monoclonal antibody that binds to VEGF with high affinity, and has been shown in clinical studies to have activity in colon, lung, and breast cancer. In studies in patients with melanoma, some disease stabilization has been observed. In a study by Varker, 25% of treated patients had disease stabilization of 24 to 146 weeks with bevacizumab alone or bevacizumab plus IFN. Bevacizumab has been used in combination with several agents. A study using the combination of paclitaxel and bevacizumab showed an overall survival at 12 months of 43.3%, suggesting some benefit to the regimen.
In a phase 2 study, 53 patients were treated with bevacizumab (given as 10 mg/kg every other week) combined with carboplatin, and weekly paclitaxel (Taxol). This regimen was associated with a 17% response rate (partial remission), and another 30 (57%) achieved stable disease for at least 8 weeks. Median progression-free survival and median overall survival were 6 months and 12 months, respectively. Although a promising agent for control of disease, further randomized studies will be needed to further assess the impact of bevacizumab on the treatment of melanoma.
Axitinib is another antiangiogenic agent that acts by selective inhibitor of the VEGF receptor (VEGFR), which inhibits VEGR-1, -2, and -3. In a single-agent study in 32 patients, Axitinib had an overall response rate of 19% with 1 complete response. An additional 9 patients had stable disease of at least 16 weeks.
Sorafanib
The relative resistance of melanoma to chemotherapy is partly explained by its constitutive activation of cell survival pathways such as the mitogen-activated protein kinase pathway and phosphoinositol-3 kinase, which interfere with the apoptotic pathway. Sorafanib (BAY43-9006, Nexevar) is a tyrosine kinase inhibitor with activity against the RAF serine/threonine kinases, although it is also a multikinase inhibitor with activity against VEGFR-2 and -3. Sorafanib is FDA-approved for the treatment of renal cell carcinoma and hepatocellular carcinoma. In a phase 2 study of 37 patients with metastatic melanoma, no tumor responses were seen although 19% experienced stable disease. Sorafenib was evaluated in combination with paclitaxel and carboplatin in a phase 1 study of 39 patients with advanced cancer (24 with melanoma). Sorafenib-related adverse events were observed in 69% of patients. In patients with melanoma, there was 1 complete response and 9 partial responses. The E2603 study is a phase 3 study randomizing 800 patients to carboplatin and paclitaxel plus sorafanib or a placebo. The results of this study are pending.
Immune-based Therapy
Several observations suggested that the host generates an immunologic response to melanoma, which has inspired the search for an understanding of the biology of the process as well as research aimed at harnessing the response to treatment. This search was originally suggested by the spontaneous regression of melanoma. Nonspecific immunotherapy studies using intratumoral injections of BCG generated further interest in the concept of immunotherapy for this tumor. The identification of TIL and the characterization of these cells has documented that this immune response can be very specific, and has led to studies of the specific antigens recognized by T lymphocytes in the context of the major histocompatibility complex (MHC). Research has also focused on the escape mechanisms, including ongoing studies of the tolerance to tumor that takes place in the host and methods to overcome this resistance.
Cytokines
The most important therapeutic agent for melanoma treatment has been the T-cell growth factor IL-2. IL-2 has an overall response rate of 16%, with a 6% complete response rate. More importantly, these responses have been durable, with prolonged disease response. Patients with a partial response have a median response duration of 5.9 months, and the median duration of response for patients with a complete response has not been reached. Observations that suggested a role of host immunologic responses has led to the development of cytokine therapy for melanoma, with a focus on IL-2 and IFN.
Interleukin-2
In contrast to IFN, which has both immunologic and antiproliferative properties, IL-2 has no direct antitumor effects, and exerts its function purely through its ability to stimulate an immunologic response by lymphocytes. In early studies with the agent, Grimm and colleagues described the lysis of tumor cells by lymphocytes cultured in T-cell growth factor (IL-2). In addition to its ability to enhance the cytotoxic activity of HLA-specific T cells, IL-2 also induced HLA-unrestricted killing of tumor cells by lymphocytes exposed to the agent in vitro. This lymphokine-activated killer (LAK) activity could be induced in both NK and CD3/CD8+ T cells. In mouse models, IL-2 was able to reduce or eliminate pulmonary metastases in an experimental tumor model using methylcholanthrene-induced sarcomas, and this effect was dose dependent.
The preclinical studies led to the National Cancer Institute Surgery Branch clinical treatment regimen using high-dose IL-2. Patients with metastatic melanoma or renal cell carcinoma were treated with 600,000 to 720,000 units/kg of IL-2 every 8 hours on days 1 to 5 and 15 to 19, and the course was repeated depending on the response. Some patients received concurrent administration of LAK cells. This treatment resulted in a 15% to 20% response rate. Further studies showed similar responses, and a follow-up report of the 270 patients treated on 8 trials with high-dose IL-2 showed a 16% objective response rate with 10% partial and 6% complete responses, which were durable. Based on these data, IL-2 has been approved by the FDA as a treatment for patients with metastatic melanoma. The median duration of response for complete responders exceeded 59 months, and patients with responses greater than 30 months remained progression-free. A study of 374 patients given high-dose intravenous bolus IL-2 (720,000 IU/kg) from July 1988 to December 1999 in the Surgery Branch of the National Cancer Institute (NCI) reported an overall objective response rate of 15.5%. Some patients received IL-2 with a vaccine. Clinical responses were higher (54%) in patients with only subcutaneous or cutaneous metastases compared with patients with disease at other sites (12.4%) ( P = .000001).
The toxicities associated with high-dose IL-2 have been the main obstacle to its widespread use in the treatment of patients. Toxicities generally include a capillary leak syndrome manifested by fluid retention, oliguria, and hypoxemia. This syndrome is often associated with cardiac side effects, including arrhythmias. Other common toxicities include gastrointestinal symptoms including diarrhea and nausea, skin irritation and desquamation, confusion, and hematologic toxicities with thrombocytopenia. The availability of guidelines for the use and administration of high dose IL-2 has made consistent delivery of the drug much easier.
Interferon
IFN was initially studied in the 1980s for the treatment of metastatic melanoma. Initial phase 1 and 2 studies were associated with an overall response rate of approximately 16%. Up to a third of the responses observed with IFN were complete, with some durable responses. Responses have generally been in patients with small-volume cutaneous or soft tissue disease. IFN has primarily been used in combination with other agents, and a meta-analysis of multiple trials reported a higher overall response in the IFN-containing regimens (24% vs 17%). The availability PEG-IFN has allowed better convenience of dosing. Although IFN has a place in the treatment of metastatic disease, it has found much more use as an adjuvant treatment.
Immunomodulatory agents
Studies have slowly elucidated the complex process of immune regulation and have allowed for these pathways to alter the response of the host to tumors. T-cell activation requires the interaction of the costimulatory molecules B7.1 (CD80) and B7.2 (CD86) with the CD28 antigen on the T cells. The interaction of these same costimulatory molecules with cytotoxic T-lymphocyte antigen-4 (CTLA-4) causes inhibition of the lymphocytes, leading to impaired cellular immune functions. These T-cell activation and inhibitory signaling molecules represent potential targets to manipulate the natural immune response to tumors. Antibodies to CTLA-4 may act to enhance the immune response to weak antigens, as seems to be the case with immunogenic tumors. In a murine model, the use of these antibodies in conjunction with a GM-CSF vaccine was able to induce complete regression of B16 tumors in mice. Human anti-CTLA4 antibodies that have been in human clinical testing include ticilimumab (tremelimumab, CP-675,206, Pfizer) and ipilimumab (MDX-020, Bristol-Myers Squibb, Medarex Inc). In the initial phase 1 study of tremelimumab in patients with melanoma, 2 of 29 patients achieved a complete response and 2 experienced partial responses. In addition, 5 others had stabilization of their disease. Patients experienced dose-limiting toxicities, and autoimmune phenomena included diarrhea, dermatitis, vitiligo, panhypopituitarism related to hypophysis, and hyperthyroidism. A subsequent phase 1/2 study showed responses in 8 of 84 patients. The A3671009 phase 3 trial evaluated 630 patients with advanced melanoma, and randomized patients to standard chemotherapy, consisting of DTIC or TMZ. However, the trial was discontinued when interim data found that tremelimumab did not demonstrate superiority over standard chemotherapy. Pfizer has already stated its intention to determine whether tremelimumab showed benefit in any of the advanced melanoma patients who took part in the trial. Analysis of the current phase 3 trial may help identify certain patients who show a better response to tremelimumab, and guide future development. There was no difference in response rate for patients treated with tremelimumab compared with chemotherapy (9.1% vs 10.1%) or progression-free survival (18.6% vs 14.1%). Further study outcomes are expected.
In an early study of the other CTLA4 antibody, ipilimumab was given to 14 patients in conjunction with gp100 210M melanoma peptide vaccine. Responses were seen in 3 patients (2 complete and 1 partial). Treatment was associated with significant immune-related adverse events. In a follow-up study, 139 patients were treated with ipilimumab either alone (85 patients) or in conjunction with a peptide vaccine. A response was observed in 23 patients (17%), with 3 complete responses that were durable (29+, 52+, and 53+ months). In this study and another, 20% to 48% of patients given ipilimumab developed autoimmune toxicity. A meta-analysis of 42 phase 2 studies including 2100 patients reported a 1-year survival rate of approximately 25% for patients with stage III/IV melanoma.
Phase 2 and 3 clinical trials with anti-CTLA-4 antibodies have resulted in 7% to 15% objective response rates in patients with metastatic melanoma. The activity of anti-CDLA-4 antibodies has been explored as monotherapy, and in combination with vaccines, other immunotherapies such as IL-2 and chemotherapy such as DTIC. The response rate in patients who experience grade 3 or 4 autoimmune toxicities is higher (36%) compared with the response rate in patients who do not experience autoimmune toxicity (5%–11%). Current studies with ipilimumab to provide data on efficacy and survival with an active control group include a phase 3 randomized, double-blind study to assess ipilimumab in combination with DTIC versus DTIC alone in patients with stage III/IV melanoma, and a phase 3 study of ipilimumab given as adjuvant therapy in patients with high-risk stage III melanoma. These larger studies will help to demonstrate the utility of anti-CTLA-4 therapy in prolonging clinical responses in patients with melanoma.
New, innovative immune-based treatment strategies for melanoma include those that antagonize receptors that suppress the immune response (CTLA-4, PD-1) and those that activate receptors that amplify the immune response (CD40 on antigen-presenting cells, 4-1BB [CD137] and OX40 on T cells). Development of these new strategies may prove beneficial, especially in increasing the efficacy of standard treatments for melanoma.
Vaccines
Vaccines for melanoma are designed to boost immune reactions against a malignancy that is already established. The review by Rosenberg and colleagues of vaccine therapy in 323 patients with melanoma at the NCI stated an overall objective response rate of 2.6%. When the review was extended beyond the NCI vaccine studies in patients with melanoma to 35 reports of 765 patients with common cancers, the objective response rate was 3.3%. The vaccine with the highest response rate included DC (7.1% objective response rate). In NCI studies combining melanoma vaccines, the overall objective response rate was 13% for patients receiving IL-2 alone (n = 379) and 16% for patients receiving IL-2 and vaccine (n = 305). Higher response rates occurred in vaccinated patients with subcutaneous or cutaneous disease only versus those with visceral disease. Despite low response rates, most vaccines were successful in generating easily detectable, vaccine-specific immune responses. Many groups currently seek to discover the factors that turn a detectable antitumor immune response into an effective antitumor immune response.
DC have been used in vaccination strategies to take advantage of their ability to stimulate T-cell responses and then regulate the response. Generation of DC is technically feasible for most laboratories. However, the most effective method to display the tumor antigen on the DC surface in the appropriate context to induce effective antitumor response remains under investigation. A recent review of 38 DC vaccination studies including 626 patients with metastatic melanoma reported a clinical response rate of 30%, with 3% complete and 6% partial responses, and 21% stable disease. Clinical response correlated with the use of peptide antigens ( P = .03), the use of any helper antigen/adjuvant ( P = .002), and induction of antigen-specific T cells ( P = .0004). Alternative approaches to peptide loading of DC in recent studies include adenoviral transduction, RNA transfection, tumor lysate loading, and tumor cell:DC fusions. Immunizing tumor antigens include melanocyte lineage antigens (Melan-A/MART-1, gp100, tyrosinase), cancer testis antigen (MAGE-A3, NY-ESO-1) and antigens that are overexpressed (p53, survivin). Butterfield and collegues transduced DC with an adenoviral vector encoding full-length MART-1. Results from this study and 2 others support a correlation between clinical benefit and determinant spreading (immune responses to nonvaccine antigens). Two recent studies tested the ability of autologous or allogeneic tumor lysate-loaded DC to induce clinical responses in patients with metastatic melanoma. Responses to the immunizing antigens were observed, but no patients achieved an objective clinical response even with the addition of low-dose or high-dose IL-2.
A potential obstacle to induce effective antitumor immunity in vivo arises from the host immune environment. Most studies immunize with DC in an immunocompetent host, where host immune tolerance can impede antitumor reactivity and negate clinical benefit in patients with advanced disease. Current studies take advantage of homeostatic mechanisms to boost tumor reactivity by lymphodepletion before DC immunization. In a pilot study, 16 patients were treated with a recombinant IL-2/diphtheria toxin conjugate (ONTAK, Denileukin Diftitox). Transient depletion of T cells with subsequent repopulation coincided with de novo appearance of melanoma-specific CD8+ T cells. Objective responses were observed in 4 of 16 patients. A current clinical trial is testing the efficacy of a combination of ONTAK before DC immunization (NCT00056134). Other studies have used lymphodepletion followed by infusion of tumor-reactive T cells with vaccine administration or after a vaccine-priming regimen, with limited clinical benefit.
Adoptive T-cell therapy
The identification in animal tumor models of immunologic control of tumors and. more significantly. the development of tumor immunity led to the search for immune effector cells that are active during cytokine therapy, and those that result in long-term immunity. One of the first cellular therapies was based on finding that lymphocytes develop the capacity to kill tumors in an HLA-independent fashion after exposure to IL-2. These LAK cells were found to kill NK-resistant cell lines, and represented both stimulated NK cells and CD3/CD8 cells. Despite strong preclinical data and preliminary results reported by Rosenberg and colleagues, a prospective randomized trial comparing high-dose IL-2 given with and without LAK showed no significant difference in response or survival.
The observation that immune cells that infiltrate a tumor have reactivity against the tumor and can be expanded has led to a series of clinical trials using TIL for the treatment of melanoma. Clinical responses were observed in these studies, but responses were transient and difficult to contribute to TIL because of concurrent IL-2 administration. The most effective cellular therapy for patients with melanoma has been infusion of TIL after a lymphodepleting chemotherapy preparative regimen. In 3 clinical trials with increasing intensity of lymphodepletion, objective clinical responses were observed in 52 of 93 patients (56%). The ability of the TIL to respond to HLA-matched or autologous melanoma cells has been prerequisite for their clinical use. The long culture time to generate and expand TIL, and to develop antitumor reactivity in TIL is not optimal, because the life expectancy of a patient with advanced melanoma is short. Murine studies have shown that a long culture time and multiple restimulations are inversely correlated with therapeutic efficacy. Cells that are cultured longer (older cells) have longer telomeres, and are associated with reduced clinical responses and shorter persistence in vivo. A new study by Rosenberg’s group demonstrated a similar frequency of tumor-reactive cells in TIL generated via a shorter culture (41%) compared with standard TIL (38%). The “young” TIL contained a higher frequency of CD4+ cells than standard TIL. Other phenotypic markers expressed on the “young” TIL (CD27, CD28) and longer length of their telomeres were linked in previous studies with T-cell persistence and survival. Rosenberg and colleagues have initiated a clinical trial to test the efficacy of “young” TIL and high-dose IL-2 after nonmyeloablative lymphodepletion in patients with metastatic melanoma. The authors’ group and a site in Israel have a similar clinical trial in progress.
It is not always possible to obtain tumor from which to generate TIL for a patient with melanoma, and less likely that melanoma-reactive TIL will be generated. Several techniques have evolved to provide alternatives to the requirement to establish TIL cultures. The discovery and cloning of melanoma-associated antigens has provided a tool with which to generate melanoma-specific T cells from peripheral blood. Dudley and colleagues and Yee and colleagues infused cloned T cells targeting gp100 or MART-1 with and without IL-2 into patients with advanced melanoma. These studies and others involving the transfer of T-cell clones and CD8+ T-cell lines demonstrated the safety and feasibility of this approach but also a lack of clinical efficacy. In 2008, Hunder and colleagues reported a complete response in a patient with metastatic melanoma infused with cloned CD4+ NY-ESO-1-reactive T cells. The cells persisted for at least 3 months and the response was durable after 22 months. Dudley and colleagues lymphodepleted the patients before T-cell infusion to prolong persistence of the T cells by eliminating the competition for homeostatic cytokines. Patients were treated with an escalating dose of lymphodepleting chemotherapy before adoptive transfer of CD8+ melanoma-reactive T-cell clones and IL-2. No objective responses or persistence of cloned T cells were observed.
The T-cell receptors (TCRs) of highly reactive cells have been cloned and transferred into peripheral blood lymphocytes (PBL), creating a new population of polyclonal T cells with antitumor reactivity. In the first adoptive cell transfer study using genetically engineered autologous T cells, T cells expressing a TCR reactive with MART-1 were shown to persist and express the transgene long-term in vivo. Two of 16 patients underwent regression of metastatic disease. In a follow-up study, 2 of 14 patients underwent regression of disease after being treated with MART-1 TCR-transduced PBL following lymphodepleting chemotherapy. Newer studies will test the efficacy of higher affinity TCRs, and TCRs against cancer antigens present on common epithelial cancers.
Biochemotherapy
The lack of improved survival with combination chemotherapy despite an improved response rate indicates the need to improve the durability of the responses seen. Biochemotherapy regimens combine the chemotherapy agents with immune-based therapy in an effort to improve the responses and durability of the remissions. The initial biochemotherapy studies combined chemotherapy with IFN. A study by Falkson and colleagues reported an improvement in response and response duration when IFN was added to DTIC. However, a follow up study comparing DTIC, DTIC with tamoxifen, DTIC with IFN, or the combination of all 3 drugs showed no significant difference in response or survival.
Because of the durable responses seen with IL-2, biochemotherapy regimens were designed to combine chemotherapy, IL-2, and IFN. In the initial studies a sequential approach was used, and initial studies by Legha and colleagues used an alternating and a sequential approach. The sequential regimen produced an overall response rate of 64% with 21% complete responses. More significant was the duration of response of more than 3 years in those patients who achieved a complete response. Multiple phase 2 studies also observed a high response rate, and suggested survival improvement over CVD alone.
Other studies have not been able to show the benefits of biochemotherapy. In a study of cisplatin, DTIC, and tamoxifen alone or in combination with IL-2 and IFN, there was a similar response with the combination (27% vs 44% P = .071), but no improvement in survival (15.8 vs 10.7 months, P = .052).
A large phase 3 study, E3695, evaluated 395 patients randomized to CVD or biochemotherapy (BCT) with CVD, IL-2, and IFN. This study showed a modest increase in response rate for BCT over CVD (19.5% vs 13.8%, P = .140) and an improved progression-free survival (4.8 vs 2.9 months; P = .015), but this did not translate into an overall survival advantage (9.0 vs 8.7 months). A meta-analysis of 18 trials similarly showed that biochemotherapy was associated with an improvement in response rate, but not an improvement in survival. A large review of the M.D. Anderson data, however, comparing patients treated with biochemotherapy versus chemotherapy with or without IFN, showed that biochemotherapy was associated with an improved overall response rate 52% versus 35% as well as improved 5-year survival (17% vs 7%, P = .0004) and 10-year survival (15% vs 5%, P = .0001), suggesting a benefit for biochemotherapy over chemotherapy. Although the phase 3 data would suggest against an improved survival with biochemotherapy, these regimens are still being developed and studied, and the data suggest some patients will benefit from these aggressive regimens.
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
