Phase III Trial Comparing Concurrent Biochemotherapy With Cisplatin, Vinblastine, Dacarbazine, Interleukin-2, and Interferon Alfa-2b With Cisplatin...

Phase III Trial Comparing Concurrent Biochemotherapy With Cisplatin, Vinblastine, Dacarbazine, Interleukin-2, and Interferon Alfa-2b With Cisplatin, Vinblastine, and Dacarbazine Alone in Patients With Metastatic Malignant Melanoma (E3695): A Trial Coordinated by the Eastern Cooperative Oncology Group

J Clin Oncol. 2008 Dec 10;26(35):5748-5754, MB Atkins, J Hsu, S Lee, GI Cohen, LE Flaherty, JA Sosman, VK Sondak, JM Kirkwood

The addition of immunotherapy to cisplatin and dacarbazine chemotherapy has been shown to increase response rates vs immunotherapy or chemotherapy alone in patients with metastatic melanoma. The addition of interferon alfa-2b (IFN-ą) to cisplatin, vinblastine, and dacarbazine (CVD) plus interleukin-2 (IL-2) resulted in further improvement in response rates and a marginally significant, 3-month improvement in survival time in 1 study. This combination regimen showed no survival benefit in 2 other studies, however. Promising results of studies in which immunotherapy and chemotherapy were administered concurrently, rather than sequentially, prompted the Eastern Cooperative Oncology Group (ECOG) to conduct a randomized phase III trial to evaluate a biochemotherapy (BCT) regimen of IL-2 and IFN-ą immunotherapy concurrent with CVD in 395 patients with progressive metastatic melanoma. In this ECOG trial (E3695), BCT did not improve overall survival (OS) over CVD alone and resulted in greater toxicity.

Patients randomized to the CVD arm received cisplatin, vinblastine, and dacarbazine, with antiemetics and dexamethasone. Patients randomized to the BCT arm received CVD, plus IL-2, IFN-ą, cephalexin, and granulocyte colony-stimulating factor, as well as hydration and antiemetic therapy, prophylactic acetaminophen and ranitidine, and antipruritics, antidiarrheal agents, and anxiolytics as needed. In both treatment arms, cycles were repeated at 3-week intervals for a maximum of 4 cycles. Dosage was modified in response to grade 3 or higher toxicity.

Of the 416 patients were enrolled, 395 had analyzable data, including 195 in the CVD arm and 200 in the BCT arm. There were 3 patients in the CVD arm and 5 patients in the BCT arm who did not receive their assigned therapy. Significantly fewer patients in the BCT arm than in the CVD arm received full doses of their therapies (P < .01). A total of 10 patients in the CVD arm and none in the BCT arm underwent >4 cycles of therapy.

Nearly three-quarters (73%) of patients in the CVD arm and 95% of patients in the BCT arm experienced grade 3 or higher toxicity (P = .001). The most common toxic effects included leukopenia, granulocytopenia, thrombocytopenia, anemia, infection, nausea, vomiting, hepatic and metabolic abnormalities, hypotension, and fatigue. With the exception of granulocytopenia and infection, all toxicities were significantly more common in patients who received BCT than in those who received CVD. Deaths due to treatment-related toxicity occurred in 3 patients in the CVD arm (myocardial infarction, hypotension, and infection) and 2 patients in the BCT arm (infection and renal failure).

Overall, 94% of patients died. Median OS was similar in the CVD and BCT arms (8.7 vs 9.0 months; hazard ratio [HR] = 0.95; 95% confidence interval [CI], 0.8-1.17; P = .639). At 1 year, 36.9% of patients in the CVD arm and 41% in the BCT arm were alive. Median progression-free survival was significantly shorter in the CVD recipients than in the BCT recipients (2.9 vs 4.8 months; HR = .0.77; 95% CI, 0.63-0.94; P = .015). At 6 months, 25.0% and 38.9% of patients were progression free.

Response rates were similar in the CVD and BCT groups (13.8% vs 19.5%; P = .140). Complete responses occurred in 4.6% and 2.5% of patients in the CVD and BCT groups, respectively, and these responses were durable (>lasting 2 years) in 6 CVD-treated patients but only 2 BCT-treated patients. Among patients with complete or partial responses, median response duration was similar in the CVD and BCT arms (9.4 vs 6.1 months; HR = 1.47; 95% CI, 0.83-2.60; P = .181).
The E695 trial represents the largest and most definitive phase III trial of BCT ever conducted. Given that BCT improved progression-free survival but not overall survival in this trial and also was more toxic than CVD, the ECOG investigators conclude that BCT should not be considered the standard of care for patients with advanced melanoma