Systemic Therapy for Advanced Melanoma

Clay M. Anderson

Rolando Breier

Systemic therapy for melanoma is the least effective strategy to treat this disease as of 2011. Early detection and adequate surgical treatment are improving outcomes steadily,¹ but the ability to improve the prognosis of potentially curable regional disease or improve survival in metastatic disease with systemic agents have only been marginally impacted. Hence treatment is generally undertaken with the goal of either modestly improving survival in stage III (regional or lymph node involvement) disease or temporarily controlling metastatic disease in a minority of patients. Cytotoxic chemotherapeutic drugs are still largely ineffective, whereas immunotherapeutic agents only dramatically impact a small fraction of patients. The roles of vaccine or cellbased immunotherapy, gene therapy, or targeted therapy have yet to be defined. Progress in systemic therapy interventions is urgently needed.

GOALS OF THERAPY

Since systemic treatment for melanoma is of marginal benefit, it should should only be given with a clear view of the goal of treatment in an individual patient situation. For resected stage III patients, adjuvant treatment should have clear evidence from randomized phase III trials of improvement in overall survival (OS) or cure rate. Benefits seen in treated patients compared with historical controls, even with careful case matching, should be considered hypothesis generating only and should not be the basis for individual treatment decisions, assuming that a prospective randomized trial can be done. In the metastatic setting, where cure is quite unlikely, treatment can be given with curative intent in a minority of cases and is usually administered with palliative intent where evidence of survival benefit from systemic treatment is largely absent. Measurable disease is advisable, so early identification of disease control or response versus progression can lead to alteration of the treatment plan. Expectant palliation of radiographically progressive disease is reasonable and has the advantage of selecting patients who may need and benefit from treatment, treating the patient when healthier and better able to tolerate side effects, and assuring the presence of measurable disease.

ADJUVANT THERAPY

Since the approval in 1995 of high-dose interferon alfa-2b (IFN-α 2b) by the U.S. Food and Drug Administration (FDA) for adjuvant treatment of high-risk resected melanoma, this single standard has been available and used with modest benefit and a fairly high incidence of toxicity. The drug’s true benefit has been controversial since several studies have produced conflicting results in terms of clinical benefit.²^,³^,⁴ An assumed absolute survival benefit of approximately 10%, as demonstrated by the Eastern Cooperative Oncology Group (ECOG) 1684 study,² is the rationale for this year-long regimen with significant constitutional side effects and some rarer longer-term risks. Patients treated with interferon in the 1980s and 1990s were usually high-risk T4 and node-positive patients, whereas today’s stage III patients may only have microscopic disease in one pathologically involved lymph node and may have a better prognosis. The real benefit of IFN in such patients is not precisely known. No other adjuvant treatment has been shown to be effective in terms of OS in a randomized clinical trial. This includes several vaccines, chemotherapy, biochemotherapy (BCT), and other immunotherapy. Sargramostim, or granulocyte macrophage colony-stimulating factor (GM-CSF), had some promise based on phase II data and comparison with historical controls⁵ but in a randomized clinical trial⁶ failed to improve OS.

ESTABLISHED SINGLE-AGENT CHEMOTHERAPY

Chemotherapy is largely ineffective for melanoma. This reflects primary resistance as a basic biologic property of this cancer.⁷ Chemotherapy has never proved effective in the adjuvant setting after resection of regional or distant metastases. In the metastatic setting, no survival benefit related to chemotherapy has ever been convincingly demonstrated. Response rates for single drugs have varied from 5% to 35% in phase II studies and from 2% to 15% in phase III studies.⁸

The only FDA-approved cytotoxic chemotherapeutic drug for melanoma is dacarbazine (DTIC), which is only modestly active. The response rate has varied from 5% to 30%,⁸ but in randomized studies it has never been >15%.⁸ Complete responses are rare and the duration of response is short (2 to 4 months).⁸ Other single drugs used historically include vinblastine, carmustine, and lomustine.⁸ Their clinical activity is less or equal to that of DTIC. There is no reason to prefer one chemotherapeutic drug over another except for specific patient concerns about toxicities, convenience, and cost. One exception is temozolomide, a newer drug, which may be better tolerated and may be effective as treatment or delay the emergence of CNS metastases. Temozolomide is discussed in the subsequent text.

ESTABLISHED SINGLE-AGENT IMMUNOTHERAPY

Interferon-α

IFN-α, a type A immunostimulating IFN molecule produced by recombinant DNA technology, has been the most widely used agent to treat melanoma. It has been applied since the 1980s to treat metastatic disease, but its use in the metastatic setting is mainly of historical interest. Currently, it is primarily used as part of multiagent treatment for metastatic disease such as BCT and in the adjuvant setting. The dose in metastatic disease has ranged from 3 million units (flat dose) to 20 million units per m², 3 to 7 days a week intravenously or subcutaneously.⁹^,¹⁰ The dose-response relation is unclear and modest at best. Constitutional toxicities limit dose and duration in most cases. Response rates have varied from 5% to 15%, complete response rates have been approximately 2% to 5%, and overage duration of response is 4 months, 4 months⁹^,¹⁰ with no improvement in OS in multiple meta-analysis.¹¹^,¹²

Pegylated Interferon

The European Organization for Research and Treatment of Cancer (EORTC) 18991 randomized stage III melanoma patients to observation versus 5 years of PEG interferon.¹³ The advantages of PEG interferon are fewer side effects and weekly administration. There was a 6.7% improvement in recurrencefree survival at 4 years without improvement in OS in the subset of N1 (microscopic nodal disease) patients deriving the most benefit.

Interleukin 2

Interleukin 2 (IL-2) is another genetically engineered polypeptide cytokine discovered, cloned, and produced in the late 1970s.¹⁴ It has shown occasional dramatic benefits in metastatic melanoma and renal cell carcinoma and has been approved by the FDA for treatment of those two diseases. True cures are noted in approximately 5% of metastatic melanoma patients¹⁵ with high-dose IL-2. Virtually all cures have either a complete response or a dramatic partial response followed by surgery for residual disease. Lesser responses are generally not worthwhile, and the treatment is accompanied by significant short-term organ toxicities. Only patients with a performance status of 0-1 should be treated with patient should this drug with the good of a response and complete remission. The FDA-approved dose and schedule is 600,000 to 720,000 units per kg IV over 15 minutes every 8 hours to a maximum of 14 doses per cycle, repeated after a 9-day break to complete one course, which can be repeated after an 8-week break once or twice based on response and tolerance. Lower doses by IV bolus, SQ injection, or continuous IV infusion are inferior as a single agent,¹⁶^,¹⁷^,¹⁸ but are used in BCT or combined immunotherapy strategies in order for the overall toxicity of the regimen to be tolerable. The combination with autologous lymphocytes from blood or tumor (adoptive immunotherapy) does not lead to improved survival or response duration compared with IL-2 alone.¹⁹^,²⁰^,²¹ Histamine, which seems to potentiate IL-2 effect in animal models and in vitro, offered no advantage over low-dose IL-2 alone in metastatic disease, including those patients with liver metastases.²²^,²³

Interleukin 21

rIL-21 is a novel class I cytokine that is produced by activated CD4+ T cells and NK T cells that mediated activation of CD8+ T cells and NK cells with activity against melanoma in multiple phase I and II clinical trials.²⁴^,²⁵^,²⁶ The overall response rate (ORR) to IL-21 was 24.3% (9/37 patients) with a median duration of 5.3 months for first-line metastatic melanoma patients and 16 patients had stable disease for up to 5.3 months. rIL-21 was well-tolerated with most adverse events being fatigue, diarrhea, and myalgia, grade 1 or 2. A randomized phase II clinical trial is ongoing.

ESTABLISHED COMBINATION CHEMOTHERAPY

Combination chemotherapy increases response rates and toxicities in melanoma, without a documented increase in OS.⁸ This combination therapy should be used judiciously. A clinical trial of a new combination that appears promising is appropriate. Otherwise, combination chemotherapy should only be used if a response is needed—to improve palliation or to make surgery or radiation more applicable if a response is obtained.

The most widely used combination regimens are the BOLD regimen, the Dartmouth regimen, and the cisplatin, vinblastine, and DTIC (CVD) regimen. The bleomycin, vincristine (Oncovin), lomustine, and DTIC (BOLD) regimen was first reported in 1980, by investigators at Duke University Medical Center.²⁷ Its clinical activity was impressive in this phase II trial, with a response rate of 40% and a complete response rate of 9%. The Dartmouth or cisplatin, BCNU (carmustine), DTIC, and tamoxifen (CBDT) regimen was first described in 1984 at Dartmouth University Medical Center²⁸ and in phase II trials produced clinical activity as good or better than CVD.⁸ The CVD regimen was developed at MD Anderson Cancer Center with reported response rate of 40% and 4% complete responses.²⁹

Combination chemotherapy for advanced melanoma should be viewed as a strategy that has yet to be determined in terms of improved patient outcomes. It can be used in select cases with a specific goal in mind, but more importantly should be tested further with new agents that have more potential than current choices or as a building block for combinations of cytotoxic drugs and novel immunotherapy agents or targeted therapies. An example is carboplatin and paclitaxel combined with the B-Raf inhibitor sorafenib, which is currently undergoing a phase III intergroup trial.³⁰

Role of Tamoxifen in Melanoma Chemotherapy

In addition to testing the utility of the Dartmouth regimen relative to DTIC,³¹ investigators have compared the Dartmouth regimen with and without tamoxifen³²^,³³ and other chemotherapeutic regimens with and without tamoxifen.³⁴ Tamoxifen plus chemotherapy seems to increase responses in phase II trials but has not added incremental benefit in randomized
trials and has added toxicity and risk, especially an increase in venous thromboembolic disease.³⁵ Consequently, it has fallen from favor as a building block for combination regimens for advanced melanoma.

BIOCHEMOTHERAPY REGIMENS

In the early 1990s, various investigators began combining modestly active cytotoxic drugs with immunotherapy agents that had occasionally produced durable responses.⁸ Initial and later trials of single chemotherapeutic drugs with single immunotherapy drugs failed to show any appreciable incremental benefit compared with chemotherapy alone. Eventually, cisplatin-based chemotherapy was combined with both IL-2 and IFN, either concurrently or with the immunotherapy being given immediately after the chemotherapy. This combination consistently produced dramatic responses at multiple centers, including MD Anderson, University of Chicago, Salpetriere Hospital in Paris, and John Wayne Cancer Institute, with ORRs of 60% and higher and complete response rates up to 20%, most being durable.⁸ These results were also replicated in a multicenter phase II trial,³⁶ leading to randomized studies that attempted to show acceptable toxicity and survival benefit.

Randomized Trials

MD Anderson

Eton et al.³⁷ reported the results of a randomized clinical trial comparing CVD chemotherapy to the sequential administration of CVD and IL-2/IFN in 2002. This trial showed a modest statistically and clinically improved survival in BCT-treated patients versus chemotherapy-treated patients. Median survival was longer in the BCT-treated patients (11.9 vs. 9.2 months) and response rate was higher as well (48% vs. 25%), but there were few complete responders in either group, in contrast to numerous phase II trials in experienced centers.

National Cancer Institute

Rosenberg et al.³⁸ compared cisplatin, DTIC, and tamoxifen alone to this regimen combined with IFN and high-dose IL-2. This trial was powered to detect a difference in response rate but not for survival. The results demonstrated improved response with BCT (44% vs. 27%), but, paradoxically, better survival in the CVD group. This result has not been adequately explained, but this BCT regimen is one that is not generally used and may have excess toxicity compared with other regimens.

European Organization for Research and Treatment of Cancer Studies

Cisplatin, or cisplatin with DTIC, days 1 through 3, was the chemotherapy building block in several studies and was compared with biotherapy with both IL-2 and IFN or chemotherapy and IFN with or without IL-2 for a maximum of 4 cycles.³⁹^,⁴⁰ In the first study, cisplatin increased the response rate of the biotherapy alone from 18% to 33% and delayed the time to progression from 1.8 to 3.1 months, but there was no overall increase in survival. In the second study, median survival was 9 months in both arms, whereas 2-year survival was higher in the IL-2 containing arm (17.6% vs. 12.9%), but was not statistically different.

Intergroup

Initiated by ECOG, this large study compared concurrent BCT versus CVD chemotherapy.⁴¹ The vinblastine dose was reduced for safety reasons and the dose intensity in the BCT arm was questioned, as many patients did not maintain full doses or complete 4 cycles. There was no significant difference between the arms in terms of survival, but response rate was slightly better in the BCT arm (17.1% vs. 11.4%), as was time to progression-free survival (PFS; 5.3 vs. 3.6 months).

Others

Several other studies,⁴²^,⁴³^,⁴⁴ mostly performed outside the United States, have utilized chemoimmunotherapy treatments that would not comply with the preceding definition of BCT, and in general have not produced impressive results compared with chemotherapy alone.

Retrospective Analyses

Three retrospective analyses have been done to assess whether more general and less dramatic trends are present in treatment outcome with BCT regimens compared with chemotherapy alone.¹⁶^,⁴⁵ In these studies regimens containing chemotherapy, IL-2, and IFN are associated with a better survival compared with other treatment types. These findings are not in complete agreement with randomized trials, causing controversy regarding the true benefit of BCT. Further trials with modified regimens are needed, especially addressing the problem of early and survival-limiting CNS metastases especially in those patients with dramatic responses to up from aggressive BCT treatments.

NEWER SINGLE AGENTS

Temozolomide

As a single agent, temozolomide, an oral alkylator that has the same active metabolite as DTIC but unlike DTIC penetrates the blood-brain barrier well, is as active as DTIC and is better tolerated, as noted in a phase III trial.⁴⁶ It has not received FDA approval for melanoma treatment, however. Nevertheless, it is extensively is extensively used in patients with and without brain metastases. In retrospective reviews and post hoc analyses of phase II and phase III trials, it appears that temozolomide does delay or help prevent CNS metastases in metastatic melanoma.⁴⁷^,⁴⁸

Taxanes

Both paclitaxel and docetaxel have been utilized in metastatic melanoma.⁴⁹^,⁵⁰^,⁵¹^,⁵² The clinical activity of these two agents appears equivalent but low and variable and is similar to other commonly used single cytotoxic agents. Response rates of 10% to 20% and good tolerance have made these drugs potential building blocks for combination chemotherapeutic regimens that
are currently being tested, but they are not effective enough to be recommended as single agents against metastatic disease in any setting.

The combination of paclitaxel and carboplatin⁵⁰ in second-line metastatic melanoma produced mild clinical benefit with mostly partial response and stable disease and a median OS of 7.8 months.

ABRAXANE (nab-paclitaxel ABI-007) a solvent free, albumin bound formulation of paclitaxel at 100 mg per m² and carboplatin (AUC 2) weekly for stage IV melanoma was tested in a phase II trial for chemotherapy naïve or previously treated patients (no prior taxanes).⁵³ The chemotherapy naïve patients arm had a better response, 25.6% versus 8.8%, compared with the previously treated arm but no improvement in disease-free survival (DFS) or OS. A trial combining nab paclitaxel and carboplatin and bevacizumab is currently ongoing.

NEWER COMBINATIONS

Only gold members can continue reading. Log In or Register to continue