ALBERT

Description: Background: Heat shock protein 90 supports the function of client proteins in multiple myeloma (MM) including IL-6 and IGF-R1. Inhibition of HSP90 leads to protein degradation and apoptosis of MM cells, including patient-derived cells resistant to bortezomib or immunomodulators. KOS-953, a new formulation of 17-AAG, is a potent HSP90 inhibitor in phase 1 and 2 trials that suppresses cell surface expression of IL-6, IGF-R1 and their downstream signaling molecules, including the PI-3K/Akt/IKK-a and Ras/Raf/MAPK pathways. In vivo, KOS-953 extends the survival of mice after MM cells are injected and home to the marrow (〉250 days in KOS-953-treated mice vs 29 days in the control group; p=0.0001). Aims: To define the MTD, toxicity, PK and pharmacodynamics of KOS-953 in patients (pts) with relapsed and relapsed, refractory MM. Methods: Pts receive an IV infusion of KOS-953 twice weekly for 2 out of 3 weeks. PK and PD are performed following the 1st and 4th infusion. Parent drug and its active metabolite are quantified in plasma. PD is assessed in bone marrow aspirates (BM) with MM cells examined for apoptosis, proliferation, pAKT, and change in HSP, IL-6 and IGF-R1 expression. Peripheral blood lymphocytes (PBLs) are assayed for change in HSP70/90 levels. Response is assessed by EBMT criteria. Results: Twenty -two pts (6 F/16 M; median age 64 yrs; median number of prior regimens 4, range 2–19; 41% prior SCT, 100% prior thalidomide, 82% prior bortezomib) were enrolled in 4 dose levels (150, 220, 275 and 340 mg/m2). Two episodes of DLT have been observed (both G3–4 elevations in liver transaminases): one pt at 220 mg/m2 with hepatic infiltration with MM (who continued with G1-2 elevations following dose reduction) and 1/6 pts at 340 mg/m2. Assessment of 420 mg/m2 is planned. Overall, G ≥ 3 drug-related toxicity was rare, consisting of AST/ALT (n=3), anemia, vomiting, memory impairment and rash (n=1 in each case). Common G1-2 drug related toxicity included diarrhea (n=9), fatigue (n=6), elevated transaminases (n=5), myalgias (n=5), infusional reactions (n=5) and rash (n=4). Elimination half-life (parent) equaled 2.8±1.0 hrs (5.1+/−1.0 hrs for the metabolite). Clearance equaled 45.0±3.0 L/hr with Vss 137±43 L. No change in PK was observed comparing the 1st and 4th infusion. At 340 mg/m2: AUCtot and Cmax were 18997 ng/mL*hr and 9355 ng/mL (parent); 13121 ng/mL*hr and 1644 ng/mL (metabolite). For the parent, AUC and Cmax increased linearly with dose. Comparing BMs (screening and 4 hr following 1st dose), CD138+ cells showed decreased pAKT (p=0.04) and increased apoptosis (p=0.06). PBLs showed reactive induction of HSP70 following KOS-953. Responses include 1 MR (150 mg/m2); 1 PR (220 mg/m2); 1 PR (340 mg/m2 with confirmation pending). An additional 6 pts have stable disease (≥ 3 cycles). Conclusions: KOS-953 has been successfully dose escalated from 150 - 340 mg/m2 with manageable toxicity and without treatment-emergent neuropathy. PK studies show linear kinetics, supporting further dose escalation to define MTD. PD studies suggest activity in heavily pre-treated MM patients, with clinical benefit (PR/MR or SD) observed in 9/22 pts (41%).