This phase III, randomized, double-blind, placebo-controlled study was conducted to evaluate the efficacy and safety of sorafenib with carboplatin and paclitaxel (CP) in patients with advanced melanoma who had progressed on a dacarbazine- or temozolomide-containing regimen. A total of 270 patients were randomly assigned to receive intravenous paclitaxel 225 mg/m2 plus intravenous carboplatin at area under curve 6 (AUC 6) on day 1 of a 21-day cycle followed by either placebo (n = 135) or oral sorafenib 400 mg (n = 135) twice daily on days 2 to 19. The primary efficacy end point was progression-free survival (PFS); secondary and tertiary end points included overall survival and incidence of best response, respectively. The median PFS was 17.9 weeks for the placebo plus CP arm and 17.4 weeks for the sorafenib plus CP arm (hazard ratio, 0.91; 99% CI, 0.63 to 1.31; two-sided log-rank test P = .49). Response rate was 11% with placebo versus 12% with sorafenib. Dermatologic events, grade 3 thrombocytopenia, diarrhea, and fatigue were more common in patients treated with sorafenib plus CP versus placebo plus CP. In this study, the addition of sorafenib to CP did not improve any of the end points over placebo plus CP and cannot be recommended in the second-line setting for patients with advanced melanoma. Both regimens had clinically acceptable toxicity profiles with no unexpected adverse events. A trial of similar design for the first-line treatment of patients with advanced melanoma (intergroup trial E2603) is currently ongoing.