We have previously shown that inhibitors of the hsp90 molecular chaperone (including geldanamycin, 17-allylamino-17-demethoxy-geldanamycin (17-AAG) and other members of the ansamycin family) have in vitro and in vivo activity against MM cells (including drug-sensitive and -resistant cells) and confer sensitization to several other anti-tumor agents. An extensive program of phase I clinical trials in solid tumors has documented biological activity of 17-AAG with manageable hsp90 inhibitor-related toxicities and has identified maximum tolerated doses for further clinical development. These first trials involved an intravenous formulation of 17-AAG in DMSO and egg-phospholipids. As such, it is believed that the formulation may have contributed to some of the observed side effects of 17-AAG administration; and conversely, that alternative formulations of 17-AAG in clinically inert excipients might potentially allow for administration of higher doses of 17-AAG and more favorable clinical outcomes. In this study, we describe the in vitro and in vivo clinical profile of KOS-953, a cremophor-based 17-AAG formulation. In our in vitro studies, KOS-953 exhibits comparable anti-MM activity to DMSO-based 17-AAG, including activity against a broad spectrum of MM cell lines and primary cells which are both sensitive and resistant to cytotoxic chemotherapeutics, proteasome inhibitor bortezomib, thalidomide or its derivatives. Consistent with prior experience on DMSO-based formulations of 17-AAG, KOS-953 triggered intracellular depletion of several key kinases, including Akt, Raf, IKK-a; decreased the constitutive and cytokine-induced activity of NF-κB; and suppressed expression of diverse intracellular anti-apoptotic proteins (e.g. FLIP, XIAP, survivin); leading to tumor cell sensitization to other pro-apoptotic agents (e.g. cytotoxic chemotherapy, PS-341). Importantly, we evaluated the in vivo anti-tumor activity of KOS-953 in our SCID/NOD mouse model of diffuse MM bone lesions. In this model, mice with bioluminescence imaging-confirmed diffuse skeletal lesions were randomly assigned to receive either KOS-953 (50 mg/kg i.p., twice weekly) or equal volume of vehicle. Mice in both cohorts were followed-up serially by whole-body real-time fluorescence imaging and whole-body bioluminescence imaging, models previously validated in separate studies of our group. The primary endpoint of the study was the overall survival of mice, defined as time between injection of tumor cells and sacrifice for hind limb paralysis, moribund state, or death. KOS-953 treatment was associated with prolongation of overall survival of mice (p<0.05, log-rank test), and was well tolerated, without vital organ tissue damage in histopathologic analyses. These findings indicate that alternative 17-AAG formulations, such as cremophor-based ones, can deliver biologically-active doses of 17-AAG and achieve anti-tumor responses. Coupled with our accumulating experience on the role of hsp90 for MM cell proliferation, survival and drug-resistance, the current study therefore provides the preclinical framework for further clinical evaluation of hsp90 inhibitors, such as KOS-953, to improve treatment outcome in MM.