The treatment of advanced melanoma remains a great challenge, with patient survival dictated primarily by the site and pace of the disease.[1] Patients with melanoma and distant metastatic disease have a 5-year survival rate of less than 5%. Median survival for these patients ranges from 6 to 9 months, but it can vary depending on several prognostic parameters. For instance, patients with metastases to the skin, subcutaneous tissue, distant lymph nodes, or lungs have median survivals ranging from 12 to 15 months, whereas patients with metastases in the liver, brain, or bones have median survivals of only 3 to 4 months.[1] Treatment options include chemotherapy, immunotherapy, and the combination of chemotherapy and interleukin (IL)-2 (Proleukin) and interferon (IFN)-alfa, referred to as biochemotherapy. This article will provide a brief overview of the major treatment options, with a special focus on the results of cisplatin(Drug information on cisplatin)-based biochemotherapy regimens in patients with advanced melanoma. Chemotherapy Alone The most active chemotherapeutic drugs in the treatment of advanced melanoma include dacarbazine(Drug information on dacarbazine) (DTIC-Dome), temozolomide(Drug information on temozolomide) (Temodar), cisplatin, the 2-chloroethylnitrosoureas (carmustine [BiCNU], lomustine [CeeNU], and fotemustine(Drug information on fotemustine)), the vinca alkaloids (vincristine and vinblastine(Drug information on vinblastine)), and the taxanes (paclitaxel and docetaxel(Drug information on docetaxel) [Taxotere]).[2] Observed overall response rates to these single agents have usually ranged from 7% to 20%, with complete remissions observed in fewer than 5% of patients.[2] Furthermore, responses are usually not durable. Although no single agent is clearly more active than dacarbazine alone, a recent randomized phase III study comparing dacarbazine (n = 117) with fotemustine (n = 112) showed a trend for a higher response rate (15.5% vs 7.2%, P = .053) and overall survival (7.4 vs 5.8 months, P = .073) favoring the fotemustine arm.[3] Temozolomide
Temozolomide is an imidotetrazine derivative that, at physiologic pH, spontaneously converts to MTIC, the active metabolite of dacarbazine.[2] It has the advantages of being absorbed orally and possessing better central nervous system penetration. In a phase II study in patients with metastatic melanoma, temozolomide was well tolerated and produced objective responses in 21% of patients, including complete responses in 5%.[4] Of interest, one of four patients with cerebral metastases exhibited a partial response. A subsequent phase III trial compared temozolomide (200 mg/m²/d orally for 5 days every 4 weeks) to dacarbazine (250 mg/m²/d intravenously for 5 days every 3 weeks) in 305 patients without brain metastases.[ 5] Temozolomide produced apparent improvement in median progression-free survival (1.9 vs 1.5 months) and health-related quality of life relative to dacarbazine, but had no significant impact on response rate (13.5% vs 12.1%) or overall survival (7.7 vs 6.4 months).[5] Of note, in a subset of patients who had regularly scheduled computed tomography scans of the head, fewer central nervous system relapses were observed in the patients receiving temozolomide. Using novel dosing schedules and combinations, investigations of temozolomide are continuing. Regimens involving multiple doses per day or prolonged daily exposure have been studied in an effort to circumvent DNA repair processes and more optimally coordinate with radiation therapy schedules. A temozolomide dose of 75 mg/m² daily for 6 weeks has been determined to be the maximum tolerated dose for the prolonged daily schedule. The pilot experience of Hwu et al suggested a synergistic interaction between temozolomide and thalidomide(Drug information on thalidomide) (Thalomid). The authors subsequently conducted a phase II study of temozolomide, 75 mg/m²/d for 6 weeks, with thalidomide, 200 mg at bedtime (escalating to 400 mg if well tolerated), in 38 patients without brain metastases. They observed 1 complete and 11 partial responses, for an overall response rate of 32%.[6] The results of a randomized phase II study with a total of 180 patients assigned to temozolomide alone, temozolomide plus IFN-alfa, and temozolomide plus thalidomide showed overall response rates of 9%, 18%, and 15%, respectively, providing further support for a synergistic interaction between temozolomide and thalidomide.[7] Furthermore, temozolomide and thalidomide were better tolerated than temozolomide and IFNalfa. Hwu et al[8] more recently reported preliminary data in patients with central nervous system metastases and identified two complete responses and one partial response among 11 treated patients. These data are clearly encouraging, and this combination is undergoing further evaluation by other investigators. Dacarbazine/Oblimersen
Another interesting and promising strategy is the combination of dacarbazine with bcl-2 antisense (oblimersen sodium, Genasense).[9] Preliminary results of a recently completed phase III randomized trial comparing dacarbazine alone vs dacarbazine plus oblimersen were recently released.[10] A total of 771 patients were included in this large study, which showed an increased response rate (11.7% vs 6.8%, P = .019) and time to progression (74 vs 49 days, P = .0003) for the combination arm but only a trend toward improved overall survival (9.1 vs 7.9 months, P = .18). Other Combinations
The next step will be to incorporate oblimersen with other agents, particularly in biochemotherapy strategies. Most chemotherapeutic agents are inactive in previously treated patients. Paclitaxel(Drug information on paclitaxel), however, has shown activity even in patients who failed prior chemotherapy. For instance, Zonder et al[11] reported a 13% response rate with weekly paclitaxel (150 mg/m² over 1 hour for 6 weeks followed by 2 weeks' rest) among 15 treated patients and Bedikian et al[12] reported a response rate of 15.6% in 32 assessable patients treated with a regimen of paclitaxel at 90 mg/m² IV over 80 minutes every 4 days for three cycles, repeated every 3 weeks. Combination chemotherapy, including well-known regimens such as CVD (cisplatin/vinblastine/dacarbazine) and the Dartmouth regimen (cisplatin/carmustine/ dacarbazine/tamoxifen), has produced responses in about 10% to 30% of patients (in phase III studies), but durable complete responses have been rare.[2] In three phase III studies comparing combination chemotherapy with single-agent dacarbazine, results showed only a slight increase in response (not statistically significant) favoring the chemotherapy combination but no improvement in overall survival.[ 2] In this author's opinion, combination chemotherapy should be considered when the objective response is important, as for instance, in patients with metastases in the head and neck region or brachial plexus involvement. Immunotherapy Alone The disappointing results observed with chemotherapy alone shifted the attention of many investigators to the biologic agents, of which IFN-alfa and IL-2 have been the most extensively studied. High-dose IL-2 results in overall responses in approximately 15% of patients, with durable remissions in about 5%.[13,14] Despite the low response rate, high-dose IL-2 has curative potential. Because the response rate to highdose IL-2 is low and toxicity high, great effort has been made to identify predictors of response. The National Cancer Institute (NCI) experience sug- gests that patients with only subcutaneous and/or skin metastases have a significantly higher chance of responding (overall response: 53%) compared with patients with visceral metastases (overall response: 10% to 15%).[15] Furthermore, a recent study from the NCI suggested that pretreatment molecular profiling may allow us to better separate responders from nonresponders.[16] Lower-dose IL-2 regimens, IFN-alfa, or other immunotherapy approaches have generally produced lower response rates and few durable remissions.[17] Based on preclinical data that suggested a synergistic interaction between IFN-alfa and IL-2, these biologics have been combined in various clinical studies of patients with advanced melanoma. Overall response rates have ranged from 10% to 41%, with the average being 20%.[2] In a small randomized trial, however, the combination of IL-2 plus IFN-alfa failed to produce a statistically significant higher response rate than IL-2 alone.[18] Novel Strategies
A promising and novel strategy was recently reported by Rosenberg's group and consisted of adoptive transfer of highly selected tumor-reactive T cells directed against overexpressed self-derived differentiation antigens after a nonmyeloablative regimen with cyclophosphamide(Drug information on cyclophosphamide) (Cytoxan, Neosar) and fludarabine (Fludara).[19] Of 13 patients who previously failed standard therapy, including high-dose IL-2, 6 achieved a partial response. Five of the responders demonstrated signs of autoimmune reaction such as vitiligo and anterior uveitis. This approach, although complex and expensive, is very promising and will likely be evaluated in other cancers. High-dose IL-2 (NCI regimen) was also evaluated in patients who failed prior biochemotherapy. In the University of Pittsburgh experience, of 18 patients treated, 3 achieved a complete remission.[20] In our experience at the Hospital Sirio-Libanes, of 10 patients treated, 1 patient achieved a complete remission of 6-month duration, and 3 achieved a partial response. All responses were of rapid onset. A representative patient, who did not respond to concurrent biochemotherapy and achieved a solid partial response to high-dose IL-2 is shown in Figure 1. Another strategy still under evaluation is the combination of subcutaneous IL-2 with histamine dihydrochloride. A subgroup analysis of a randomized phase III trial comparing subcutaneous IL-2 therapy with and without histamine suggested an improvement in overall survival in patients with liver metastases.[21] This finding, however, remains to be confirmed by another randomized trial specifically addressing this patient cohort. Biochemotherapy The limited results observed with chemotherapy and immunotherapy alone have prompted many investigators to empirically combine chemotherapy drugs with IL-2 and IFN-alfa, referred to as "biochemotherapy" or "chemoimmunotherapy."[1] Of all the currently employed treatment modalities for metastatic melanoma, cisplatin- based regimens combined with biologic agents such as IFN-alfa and IL-2 appear to have attained the highest response rates.[1] The combination of IL-2 or IFN-alfa with non-cisplatinbased regimens was disappointing and rapidly abandoned. Phase II Studies of Cisplatin- Based Biochemotherapy
Phase II studies of cisplatin-based biochemotherapy regimens have shown overall response rates ranging from 40% to 60%, with complete remission rates on the order of 10% to 20%.[22-34] Durable remissions exceeding 5 years were seen in approximately 5% to 10% of patients. Of interest, relapses occurring beyond 2 years were distinctly uncommon, suggesting that these patients exhibiting durable responses may be "cured."[22,24-26] Furthermore, re sponse to biochemotherapy is usually rapid and observed after the first cycle. A representative case is shown in Figure 2. The results of select phase II studies with IL-2 administered intravenously are shown in Table 1.[22-28,32-34] One of the most popular regimens developed at M. D. Anderson Cancer Center is the CVD regimen administered concurrently with IL-2 and IFNalfa for a maximum of six cycles. Tumor responses were observed in 34 of 53 patients (64%), with 20% complete responses and 9% durable complete responses.[25] In another study, this regimen was modified in an effort to reduce toxicity. Modifications included antibiotic and granulocyte colony-stimulating factor (G-CSF [Neupogen]) prophylaxis, prohibition of long-term central venous access, and restriction to a maximum of four cycles of therapy. Tumor responses were seen in 19 of 40 evaluable patients (response rate: 48%) including 8 complete responses.[28] Biochemotherapy was also evaluated as secondline therapy in melanoma patients and found to have negligible activity in this patient population, producing an overall response rate of only 6% compared with 37% in previously untreated patients.[28] O'Day et al recently reported the use of IL-2 and granulocyte macrophage colony-stimulating factor (GMCSF [Leukine]) as maintenance therapy in patients who achieved a partial response or stable disease on biochemotherapy. Preliminary results of this pilot experience were encouraging in terms of progression-free and overall survival, compared with historical controls.[35] A larger phase II multicenter trial has been initiated in an effort to confirm these results.

• Efforts to Reduce Toxicity-Many of these biochemotherapy regimens require inpatient treatment because they have substantial toxicity. A number of investigators have endeavored to develop regimens with less toxicity that could be administered in the outpatient setting (Table 2).[29-32,36] They simplistically replaced IL-2 administered intravenously with the subcutaneous route and frequently reduced the IL-2 dose to improve tolerance. Unfortunately, biochemotherapy regimens involving subcutaneous administration and lower doses of IL-2 appear to produce lower response rates than were generally observed with regimens involving intravenous IL-2. For example, Flaherty et al[32] reported the results of a randomized phase II trial of two outpatient biochemotherapy regimens involving dacarbazine, cisplatin, and IFN-alfa with IL-2 administered either intravenously or subcutaneously. There were 16 responses including 5 complete responses in the 44 patients who received the IV IL-2 regimen (response rate: 36%), compared with only 6 responses (including 1 complete) in the 36 patients assigned to receive the subcutaneous IL-2 regimen (response rate: 17%).[32] In addition, a study in which patients were randomly assigned to receive either the Dartmouth regimen or the Dartmouth regimen preceded by IL-2 administered subcutaneously (day -2 to 0) and followed by IFNalfa (days 1 to 3) showed no difference in response rate (22% for biochemotherapy vs 27% for the Dartmouth regimen alone), median duration of response (2.8 vs 2.5 months), or survival (5 vs 5.5 months).[36] The lower overall response rates for the biochemotherapy regimens involving subcutaneous IL-2 in these two studies relative to the intravenous IL-2 biochemotherapy regimens mentioned previously suggest a potential schedule, dose, and route of administration effect for IL-2 in biochemotherapy combinations.

Results of Meta-analysis
Two recent meta-analyses suggested improved response rates and possibly improved survival for combinations involving cisplatin, IL-2, and IFN-alfa, compared with either chemotherapy or immunotherapy alone. In one analysis of 631 patients, biochemotherapy regimens produced a response rate of 45% compared to 21% and 15% with IL-2 and IFN or IL-2 alone, respectively. Survival, however, was not significantly different between groups (10.5 months), with 20% and 10% survival rates at 2 and 5 years.[37] Another meta-analysis that analyzed 154 studies involving over 7,000 patients revealed the highest response rate of 47% for patients who received cisplatin, dacarbazine, IL-2, and IFNalfa.[ 38] The results of these two metaanalyses clearly support the phase II data that show a higher response rate for biochemotherapy compared to chemotherapy or immunotherapy alone.