The history of systemic treatments for advanced cutaneous melanoma, which is reviewed thoroughly in the paper by Anderson et al, can best be characterized as a series of unrealized hopes.
Dacarbazine has been used as a single agent to treat melanoma for approximately 20 years. It has a response rate of only 20%, and responses to this agent in patients with visceral metastases have been reported to be less frequent. Unfortunately, since dacarbazine was approved by the FDA, no other drug has proved to be more efficacious.
Combination chemotherapy has shown initial promise in several small phase II trials but has not yet shown more efficacy than dacarbazine alone in randomized phase III trials. For example, a regimen consisting of cisplatin, vinblastine, and dacarbazine (CVD) had a response rate of approximately 45% in a phase II trial; however, when this combination was compared directly against single-agent dacarbazine, it was not found to be superior.
The Dartmouth regimen, which consists of dacarbazine, cisplatin (Platinol), carmustine (BiCNU), and tamoxifen, was reported to have a 55% response rate in a trial of 20 patients in 1984 , and several other phase II trials of this regimen subsequently reported similar results. However, all of these were small studies with wide confidence intervals, the lower limits of which overlapped the range of responses for dacarbazine alone. The Dartmouth regimen is more expensive and is expected to be significantly more toxic than single-agent dacarbazine. Currently, we are conducting a multi-institutional, cooperative group phase III trial designed to assess definitively whether the response rate of the Dartmouth regimen is superior to that of dacarbazine alone.
Similarly, the results of hormonal therapy have been disappointing. Although tamoxifen lacks significant activity against melanoma as a single agent, the initial trials had suggested that tamoxifen was a crucial component of the Dartmouth regimen; when tamoxifen was omitted from the regimen, the response rate dropped from 52% to only 10%. Recently, a series of randomized trials has sought to determine whether the addition of tamoxifen to chemotherapy regimens improves response rate and survival. The largest of these trials, conducted by the National Cancer Institute of Canada, showed no difference in response rate or survival between the Dartmouth regimen with or without tamoxifen .
Initial Promise of Biologic Agents Still Unfulfilled
Biologic agents, most notably interferons and interleukin-2 (IL-2), have been studied extensively for the treatment of advanced melanoma. Interferon-alfa has been shown to yield response rates ranging up to 22% and is possibly comparable to but no better than dacarbazine. At high doses, interferon is probably more toxic than dacarbazine. Interestingly, trials utilizing "uninterrupted" (daily or three-times-weekly) schedules of administration of interferon-alfa yielded higher response rates than those using "intermittent" dosing. In addition, clinical response was inversely related to tumor bulk, and the time to response was lengthy, with many responses occurring only after 6 months of therapy. At present, the main potential of inteferon-alfa as a single agent seems to be in the adjuvant setting in patients with stage III or IV melanoma who have been rendered free of disease by surgery but who are at high risk for recurrence.
High-dose IL-2 has yielded response rates ranging from 10% to 26%. However, administration of high-dose IL-2 by bolus is extremely toxic and requires hospitalization for several days. The combination of IL-2 with lymphokine-activated killer (LAK) cells was not found to be more efficacious than IL-2 alone. A cost-benefit analysis comparing IL-2 to standard chemotherapy for the treatment of advanced melanoma might be very illuminating.
Combination biologic therapy, namely IL-2 plus interferon, has also been evaluated in patients with advanced melanoma. Again, while the initial phase II trials appeared to show promise, a subsequent randomized trial failed to show a significant difference in response rate and survival when the combination was compared to IL-2 alone.
At present, no single chemotherapeutic or biologic agent or combination of drugs has clearly demonstrated increased survival compared to dacarbazine alone in patients with advanced melanoma. Therefore, dacarbazine should remain the standard first-line systemic therapy, albeit a woefully inadequate one, for the treatment of these patients. The major adverse reaction to dacarbazine is severe nausea and vomiting, which was reported to affect more than 90% of patients receiving this drug. Thus, it would not be overly cynical to state that the development of effective antiemetic agents, such as the 5-hydroxytryptamine3 (HT3) antagonists, which have made dacarbazine a reasonably well-tolerated agent, represents the most significant advance in the treatment of advanced melanoma since the advent of dacarbazine.
New Treatment Strategies May Represent Progress
Nevertheless, new treatment strategies that may represent significant progress in the treatment of advanced melanoma are being developed and evaluated. One such strategy is biochemotherapy, or chemoimmunotherapy, ie, the combination of chemotherapy with biologic agents. The potential of biochemotherapy has been suggested by the results of a couple of studies. Falkson et al  reported that dacarbazine plus interferon-alfa-2b (Intron A) was superior to dacarbazine alone with respect to response rate, duration of response, and survival.
Also, BOLD-IFN (bleomycin, Oncovin, lomustine, and dacarbazine plus human leukocyte interferon) has been reported to have a response rate of 62% with a 13% complete response rate, whereas BOLD alone had yielded a response rate of approximately 20%. Among patients with liver metastases, a subgroup in whom objective responses have been difficult to achieve, 7 of 10 patients experienced a response to BOLD-IFN. Interestingly, in contrast to the observation concerning the effect of the dosing schedule on efficacy when interferon was used as a single agent, intermittent dosing of interferon appeared to yield better results than uninterrupted dosing in BOLD-IFN .
The optimism generated by these studies should be tempered. The two arms in the Falksons' study were not balanced with respect to gender or age, two factors that have been reported to be predictive of survival in patients with advanced melanoma. In addition, three other trials failed to demonstrate that the combination of dacarbazine and interferon-alfa was superior to dacarbazine alone; however, it should be noted that these trials used lower doses of interferon-alfa than did the Falksons' trial [5-7].
The Eastern Cooperative Oncology Group (ECOG) is conducting a randomized trial comparing dacarbazine alone with dacarbazine plus interferon-alfa as administered by the Falksons. The Falksons' regimen, if it proves to be superior, might be enhanced in future trials by the addition of other chemotherapeutic and biologic agents, but the dose of interferon and its associated toxicity would probably preclude addition of IL-2.
The combination of chemotherapy with both IL-2 and interferon-alfa has been evaluated in several phase II trials. For the combination of the Dartmouth regimen and IL-2 and interferon-alfa, Anderson et al note that the complete response rate was 15% and that the median survival of the complete responders was more than 15 months; this is an erroneous and misleading way of analyzing survival. The median survival for all patients entered into the trial was 11.5 months .
The authors have conducted a series of phase II trials of CVD in combination with interferon-alfa and IL-2 in patients with advanced melanoma . They report that a "sequential" schedule, denoted "BIO\CVD," produced a higher response rate, prolonged progression-free survival (8 vs 4 months), and prolonged overall survival (12 vs 9 months) than historical controls treated with CVD alone. However, BIO\CVD is quite toxic and requires lengthy hospitalization with its attendant expense. The prolongation of survival reported for this regimen is quite short, and if quality of life were considered in the analysis, it is possible that this regimen would not represent an improvement over chemotherapy alone. In addition, it is questionable whether this regimen will be amenable to implementation by oncologists practicing in community hospitals. Furthermore, the patients treated with this protocol were significantly younger than the typical patient with stage IV melanoma, and it might be difficult to use this regimen to treat older patients without unacceptable toxicity.
The authors also report that a trial of "concurrent" biochemotherapy, designed to decrease toxicity and cost, suggested that this strategy is superior to chemotherapy alone in terms of the response rate. However, this superiority did not translate into prolonged progression-free or overall survival.
Will Biochemotherapy Prove to Be a Major Advance?
Anderson et al are among the foremost proponents of biochemotherapy. However, I am skeptical that this strategy as it is currently being implemented will prove to be a major advance in the treatment of melanoma. As detailed above, combination chemotherapy has not yet been shown to be superior to single-agent chemotherapy, nor has combination immunotherapy been shown to be superior to single-agent immunotherapy. Furthermore, most of the phase II trials that have evaluated either interferon-alfa or IL-2 in combination with various single chemotherapeutic agents or standard combination chemotherapy regimens have failed to suggest that biochemotherapy is superior to chemotherapy alone.
The proponents of biochemotherapy emphasize the high response rates of these regimens. The more important question is whether they result in improved survival. Studies at our center , as well as other institutions, indicate that approximately 10% of patients with stage IV melanoma survive for at least 3 years, and no individual treatment strategy has been shown to significantly improve this "long-term survival" rate. I believe that this is a standard against which biochemotherapy and other new treatment strategies must be measured.
Although the results of phase II trials of biochemotherapy may appear to be encouraging, the superiority of these regimens over standard treatment for advanced melanoma can be proven only by randomized trials. As these trials are designed, it must be remembered that survival and quality of life are the key end points. Unless these new treatment strategies improve long-term survival without excessive toxicity and cost, they will inevitably be deemed to be disappointments. Furthermore, survival among patients with advanced melanoma varies significantly. Therefore, as models predicting survival of patients with stage IV melanoma are developed and validated, it will be necessary that they be considered when designing trials in which survival is an end point in order to prevent bias and to facilitate comparisons among trials.
Better Understanding of Tumor Biology and Immunology
Certainly, it is an interesting biologic question as to whether immunotherapy could have additive or even synergistic effects when it is administered in combination with cytotoxic chemotherapy. However, most of the studies examining biochemotherapy are doing so in a strictly empiric fashion without a strong biologic foundation. It is likely that significant progress in the treatment of advanced melanoma will result not from an empiric combining of extant agents, but rather from strategies based on a more detailed understanding of tumor biology and immunology.
Given the increasing incidence of melanoma, the heightening of public and media awareness, and the inadequacy of the standard therapeutic regimens, it is understandable that new, promising strategies for the treatment of advanced melanoma engender unbridled optimism. However, over the past 20 years, we were often excited by phase II trials that appeared to promise progress, only to be disappointed when these results failed to hold up in phase III trials. Biochemotherapy represents a much-needed hope for those of us who care for melanoma patients, but only rigorous evaluation will determine whether this hope will be realized.
1. Del Prete SA, Maurer LH, O'Donnell J, et al: Combination chemotherapy with cisplatin, carmustine, dacarbazine, and tamoxifen in metastatic melanoma. Cancer Treat Rep 68:1403-1404, 1984.
2. Rusthoven J, Quirt I, Iscoe N, et al: A randomized, placebo-controlled trial comparing BCNU (B), dacarbazine (D), and cisplatin (P) versus BDP and high-dose tamoxifen in the treatment of metastatic melanoma. Proc Am Soc Clin Oncol 14:413, 1995.
3. Falkson CI, Falkson G, Falkson HC: Improved results with the addition of interferon-a2b to dacarbazine in the treatment of patients with metastatic malignant melanoma. J Clin Oncol 9:1403-1408, 1991.
4. Pyrhonen S, Hahka-Kemppinen, Muhonen T: A promising interferon plus four-drug chemotherapy regimen for metastatic melanoma. J Clin Oncol 10:1919-1926, 1992.
5. Kirkwood J, Ernstoff M, Giuliano A, et al: Interferon-a2a and dacarbazine in melanoma. J Natl Cancer Inst 82:1062-1063, 1990.
6. Thomson DB, Adena K, McLeod, et al: Interferon-a2a does not improve response or survival when combined with dacarbazine in metastatic malignant melanoma: Results of a multi-institutional Australian randomized trial. Melanoma Res 3:133-138, 1993.
7. Bajetta E, Di Leo A, Zampino MG, et al: Multicenter randomized trial of dacarbazine alone or in combination with two different doses and schedules of interferon-a2a in the treatment of advanced melanoma. J Clin Oncol 12:806-811, 1994.
8. Richards J, Mehta N, Ramming K, et al: Sequential chemoimmunotherapy in the treatment of metastatic melanoma. J Clin Oncol 10:1338-1343.
9. Legha S, Buzaid A, Ring S, et al: Improved results of treatment of metastatic melanoma with combined use of biotherapy and chemotherapy. Proc Am Soc Clin Oncol 13:394, 1994.
10. Frank SJ, Reilmann R, Applewhite J, et al: Predictors of long-term survival in stage IV melanoma. Proc Am Soc Clin Oncol 14:411, 1995.