Purpose Bortezomib-melphalan-prednisone (VMP) has improved overall survival in multiple myeloma. This randomized trial compared VMP plus thalidomide (VMPT) induction followed by bortezomib-thalidomide maintenance (VMPT-VT) with VMP in patients with newly diagnosed multiple myeloma. Patients and Methods We randomly assigned 511 patients who were not eligible for transplantation to receive VMPT-VT (nine 5-week cycles of VMPT followed by 2 years of VT maintenance) or VMP (nine 5-week cycles without maintenance). Results In the initial analysis with a median follow-up of 23 months, VMPT-VT improved complete response rate from 24% to 38% and 3-year progression-free-survival (PFS) from 41% to 56% compared with VMP. In this analysis, median follow-up was 54 months. The median PFS was significantly longer with VMPT-VT (35.3 months) than with VMP (24.8 months; hazard ratio [HR], 0.58; P < .001). The time to next therapy was 46.6 months in the VMPT-VT group and 27.8 months in the VMP group (HR, 0.52; P < .001). The 5-year overall survival (OS) was greater with VMPT-VT (61%) than with VMP (51%; HR, 0.70; P = .01). Survival from relapse was identical in both groups (HR, 0.92; P = .63). In the VMPT-VT group, the most frequent grade 3 to 4 adverse events included neutropenia (38%), thrombocytopenia (22%), peripheral neuropathy (11%), and cardiologic events (11%). All of these, except for thrombocytopenia, were significantly more frequent in the VMPT-VT patients. Conclusion Bortezomib and thalidomide significantly improved OS in multiple myeloma patients not eligible for transplantation.