The randomized, open-label, phase III trial HOVON-65/GMMG-HD4 was designed to evaluate the efficacy of bortezomib prior to HDM for response and progression-free survival (PFS) in patients with newly diagnosed MM. The trial was performed in 75 referral centers in the Netherlands and Belgium (HOVON group) and Germany (GMMG group). Patients with Salmon & Durie (SD) stage II or III, age 18-65 years inclusive, were randomly assigned to 3 cycles of VAD (vincristine 0.4 mg, adriamycine 9 mg/m2 days 1-4, dexamethasone 40 mg days 1-4, 9-12, and 17-20) or PAD (bortezomib 1.3 mg/m2 days 1,4,8,11, adriamycine 9 mg/m2 days 1-4, dexamethasone 40 mg days 1-4, 9-12, and 17-20). No thrombosis prophylaxis was given. Stem cells were mobilized using the CAD regimen, including cyclophosphamide 1000 mg/m2 iv day 1, and G-CSF. After induction therapy, all patients were to receive 1 or 2 cycles of high-dose melphalan (HDM) 200 mg/m2 with autologous stem cell rescue followed by maintenance with thalidomide 50 mg daily (VAD arm) or bortezomib, 1.3 mg/m2 once every 2 weeks (PAD arm) for 2 years. Between May 4, 2005 and May 16, 2008, 833 patients were randomized. After the trial was closed, we here report the planned interim analysis data on response after induction and HDM-1 of the initial 300 (150 per arm) randomized patients. The 2 randomization arms were equal for SD stage of disease, ISS stage, and distribution of chromosomal abnormalities. 137 patients (91%) completed PAD or 136 (91%) completed VAD and 132 patients (88%) in each arm completed HDM-1. Full dose bortezomib could be administered in 95% (PAD1), 89% (PAD2) and 85% (PAD3) of patients. Successful stem cell apheresis was achieved in all 137 PAD treated patients who received CAD mobilization . Peripheral polyneuropathy CTC grade 3-4 during PAD vs VAD was 16% vs 6%, while DVT/pulmonary embolism was diagnosed in 3% during VAD and 4% during PAD. Responses were assessed according to EBMT criteria including VGPR and nCR after PAD/VAD, after HDM-1 and best response on protocol treatment. Complete Response (CR/nCR), Very Good Partial Response (VGPR) and Partial Response (PR) in both arms were compared by logistic regression (table 1). Deletion of chromosome 13q or presence of t(4;14) did not have a significant impact on VGPR or (n)CR. We conclude that PAD induces significantly more PR+VGPR+(n)CR as compared with VAD, and that this effect is sustained after HDM-1.