CHICAGO The first two randomized trials to assess the addition of sorafenib to chemotherapy for advanced melanoma exhibited mixed results, according to presentations at the annual meeting of the American Society of Clinical Oncology.
A randomized, 17-center, phase II study of 101 chemotherapy-naive patients showed a 50% improvement in progression-free survival and a 62% improvement in time to progression when sorafenib (Nexavar) was added to dacarbazine (DTIC-Dome) compared with dacarbazine plus placebo.
Improved progression-free survival did not translate into a survival benefit, however. "At our last analysis, 65 of 101 patients had died, and there was no difference in median survival between the two study arms," said Dr. David F. McDermott, clinical director of the biologic therapy program at Beth Israel Deaconess Medical Center in Boston.
The second study, the 270-patient, phase III Platelet Receptor Inhibition in Ischemic Syndrome Management (PRISM) trial, tested paclitaxel plus carboplatin with or without sorafenib as second-line treatment. The trial produced negative results. Dr. Sanjiv S. Agarwala, chief of medical oncology at St. Luke's Cancer Center in Bethlehem, Pa., reported that sorafenib failed to improve progression-free survival, tumor response rates, or time-to-disease progression in metastatic melanoma patients, whose disease had progressed on a chemotherapy regimen containing dacarbazine or temozolomide (Temodal).
In his discussion of the two trials, Dr. Keith Flaherty said that although the trials had mixed results, the 6-month progression-free survival rate of 41% in the study by Dr. McDermott and colleagues "is truly the high water mark of what we've achieved to date … at least when focusing on this end point." These gains were achieved at a toxicity cost deemed "not unacceptable" by Dr. Flaherty of the division of hematology-oncology at the University of Pennsylvania Health System in Philadelphia.
The multicenter trial by Dr. Agarwala and colleagues did manage to produce data showing that the carboplatin-paclitaxel combination is "relatively active" in patients who have failed front-line chemotherapy containing dacarbazine or temozolomide, according to Dr. Flaherty. "The roughly 30% progression-free survival rate at 6 months is a number that many of us in the field believe is a sign of activity," he said.
"The front-line randomized phase II trial certainly suggests that sorafenib may be active in this setting, and I think the phase III study gives us enough evidence to say that carboplatin-paclitaxel control arm therapy is a perfectly reasonable therapy to offer patients," Dr. Flaherty concluded.
In the dacarbazine with or without sorafenib study, Dr. McDermott and his associates randomized 101 good performance status patients to receive either dacarbazine at 1,000 mg/m2 on day 1 in combination with oral sorafenib 400 mg twice daily, or dacarbazine at 1,000 mg/m2 on day 1 and two placebo tablets twice daily. Tumors were assessed at baseline and every 6 weeks, and treatment was continued until progression or intolerable toxicity.
Dose reductions due to adverse events (including grades 3 and 4 thrombocytopenia, neutropenia, nausea, and CNS hemorrhage) were more common in the sorafenib arm.
"All these toxicities were reversible, and there were no treatment-related deaths. Sorafenib-associated hand-foot syndrome, rash, hypertension, and elevated lipase [were] not greater than [have] been reported in earlier sorafenib trials," Dr. McDermott said.
The 270 chemotherapy-refractory patients in the PRISM trial had stage IV or unresectable stage III melanoma. Half were randomized to receive paclitaxel 225 mg/m2 and carboplatin AUC = 6 on day 1 every 3 weeks plus oral sorafenib 400 mg twice daily on days 2 to 19 every 3 weeks. The other half received the paclitaxel-carboplatin regimen plus an oral placebo. Both groups continued treatment until disease progression or intolerable toxicity.
The difference in progression-free survival between the sorafenib plus chemotherapy and sorafenib plus placebo arms was insignificant at 17.4 weeks and 17.9 weeks, respectively, and there were no tumor responses in either arm, according to Dr. Agarwala.
Neutropenia affected nearly half of patients similarly in both arms, while thrombocytopenia, diarrhea, hand-foot reactions, and rash were higher with sorafenib.
Both trials were sponsored by Bayer, which markets sorafenib. The ongoing Eastern Oncology Cooperative Group trial E2603 is evaluating the same regimen studied by Dr. Agarwala and colleagues in a larger patient population with unresectable locally advanced or stage IV melanoma.