Isabelle L Ray-Coquard, Julien Domont, Emmanuelle Tresch-Bruneel, Emmanuelle Bompas, Philippe A Cassier, Olivier Mir, Sophie Piperno-Neumann, Antoine Italiano, Christine Chevreau, Didier Cupissol, François Bertucci, Jacques-Olivier Bay, Olivier Collard, Esma Saada-Bouzid, Nicolas Isambert, Corinne Delcambre, Stéphanie Clisant, Axel Le Cesne, Jean-Yves Blay, Nicolas Penel

The aim of this randomized, phase II trial was to explore the activity and safety of adding bevacizumab to paclitaxel once per week in treatment of angiosarcomas (AS). Patients were treated with paclitaxel alone (90 mg/m(2) per week for six cycles of 28 days each; arm A) or with paclitaxel combined with bevacizumab (10 mg/kg once every 2 weeks; arm B). In the combination treatment arm, bevacizumab was administered after the six cycles of chemotherapy as maintenance therapy (15 mg/kg once every 3 weeks) until intolerance or progression occurred. Stratification factors were superficial versus visceral AS and de novo versus radiation-induced AS. The primary end point was the 6-month progression-free survival (PFS) rate, which was based on RECIST, version 1.1. Statistical assumptions were P0 = 20%, P1 = 40%, a = 10%, and b = 20%. P0 was the PFS rate at 6 months defining inactive drug, and P1 was the PFS rate at 6 months defining promising drug. A total of 52 patients were enrolled, and 50 were randomly assigned in 14 centers. The most common primary sites were the breast (49%) and skin (12%). There were 17 (34%) visceral and 24 (49%) radiation-induced AS. The performance status was 0 in 24 patients (49%) and 1 in the remaining 25 patients (51%). The median follow-up time was 14.5 months. Both treatment regimens were considered active, with 6-month PFS rates of 54% (14 of 26) in arm A and 57% (14 of 24) in arm B. The median overall survival rates were 19.5 months in arm A and 15.9 months in arm B. Toxicity was higher with the combination arm and included one fatal drug-related toxicity (intestinal occlusion). The primary objective was met in both treatment arms. However, the present data do not support additional clinical investigation of combined paclitaxel/bevacizumab for the treatment of advanced AS.