Phase 1, Pharmacokinetic (PK) and Pharmacodynamic (PD) Study of Intravenous Alvespimycin (KOS-1022) in Patients with Refractory Hematological Malignancies.

Background: Alvespimycin (KOS-1022), an Hsp90 inhibitor and a derivative of geldanamycin, is in phase 1 trials investigating a variety of intravenous and oral schedules. Compared to 17-AAG (the first Hsp90 inhibitor to enter clinical testing), KOS-1022 is ~3–5 fold more potent (comparing in vitro cytotoxicity or the MTD in toxicology studies); its formulation is water-soluble. The objectives of this trial were to establish the MTD and recommended phase 2 dose in patients with refractory hematologic malignancies; to assess the safety, to document responses and to characterize the PK and PD of KOS-1022.

Methods: Escalating doses of KOS-1022 were given IV over 1 hour twice weekly for 2 out of 3 weeks; premedications were not given. Plasma KOS-1022 concentrations (1^st and 4^th infusion in Cycle 1) were quantitated by LC/MS/MS. Pre and on-study CD34+ bone marrow and peripheral blasts were analyzed by flow cytometry to quantify HSP70/90, pAKT/total AKT and markers of apoptosis and proliferation. Toxicity was assessed by CTCAE v 3.0; responses in AML pts were assessed using Cheson JCO 2003 criteria.

Results: 24 pts were enrolled at doses of 8 (n=4), 16 (n=7), 24 (n=11) and 32mg/m2 (n=2). All were AML, except 1 CML pt; 3 pts had FLT3 mutations. Median age 72; median ECOG PS 1; most (n=21) patients had had 2 or 3 prior induction regimens. Cardiac DLT was noted in 2 patients treated at 32 mg/m2 (acute myocardial infarction and elevation of troponin; significant co-morbidity existed for both pts: prior myocardial infarction on post-mortem/CHF/ventricular hypertrophy and rapidly progressive AML with troponin elevation during induction chemotherapy 6 weeks prior to study). Cardiotoxicity was not notable at the three lower doses. Common drug-related toxicities (all Grade 1–2): fatigue 29%, nausea 19%, diarrhea 14%, arthralgias and dizziness 14%. Grade 3–4 non-hematological toxicity (except for the DLT noted above) was not observed. KOS-1022 PK: Cmax was roughly dose-proportional and clearance (CL) showed no clear dose- or time-dependency. The median Cmax, CL, and elimination half-life for the 24 mg/m2 cohort (n=8) were 291 ng/mL, 16.0 L/hr/m2, and 18.1 h, respectively, after dosing on Day 1. Comparing exposure on Day 1 and Day 11 at 24 mg/m2 (n=6), AUC values were 1951 and 2168 ng*hr/mL, respectively; pre-infusion drug levels at this dose on Day 11 were 8 ng/mL. In the six pts with matched pairs of bone marrow aspirates, increased apoptosis by mitochondrial potential was observed on Day 15, after four infusions; (p=0.027). Activity in AML: 3 pts with CRi; one patient remained stable on study x 9 cycles (three of these pts had wild-type FLT3 and one had unknown FLT3 status).

Conclusions: Encouraging signs of antileukemia activity were observed in a refractory population. Using a twice weekly schedule, 24 mg/m2 was well tolerated in this population. KOS-1022 plasma PK are generally linear over this dose range. Apoptosis of bone marrow cells by flow cytometry was observed after 4 infusions.