The Safety, Pharmacokinetics and Pharmacodynamics of KW-2478, a Novel Hsp90 Antagonist, in Patients with B-Cell Malignancies: A First-in-Man, Phase I, Multicentre, Open-Label, Dose Escalation Study

Treatment of cancer cells with Hsp90 inhibitors/antagonists results in cell cycle arrest, destabilisation and apoptosis. This ability of Hsp90 antagonists to modulate tumour microenvironments and initiate the selective degradation of various factors needed for cell proliferation and survival make them potential candidates for the treatment of B-cell malignancies such as multiple myeloma (MM), chronic lymphocytic leukemia (CLL) and B-cell non-Hodgkin’s lymphoma (NHL). KW-2478 is a novel non-ansamycin, non-purine analogue antagonist for Hsp90. This study investigated KW-2478 in patients (aged ≥18 y) with relapsed/refractory MM, CLL or NHL. Fifteen patients (5 cohorts of 3) received escalating doses of KW-2478 at 14, 28, 47, 71, 99 mg/m², IV over 60 min, once daily on Days 1 to 5 of a 14-day cycle. Patients could receive further cycles for up to one year, with a dose escalation option. Study objectives were to determine safety, tolerability, pharmacokinetic and pharmacodynamic profiles of KW-2478. Safety was assessed throughout the study (AEs, dose-limiting toxicities [DLTs], vital signs, ECG, physical examination, safety laboratory tests and visual ophthalmological examination with electroretinogram [ERG]). DLT was defined as an event during the first cycle which was considered related to KW-2478 and either led to treatment delay, persisted beyond Day 14, was a clinically significant ERG change, was a grade 4 or clinically significant grade 3 non-hematological event, or was a grade 4 hematological toxicity not related to primary disease that had not recovered to ≤ grade 3 by Day 14. On Days 1 and 5 (first cycle), blood samples were collected at pre-dose, 50 min post start of infusion and 10 and 30 min, and 1, 2, 4 and 8 h post infusion for KW-2478 pharmacokinetic analysis. Blood was sampled for pharmacodynamic analysis at pre-dose and 8 h on Days 1 and 5. Hsp70 expression in PBMCs was analyzed by Western blot. Results showed good tolerability with no DLTs between KW-2478 doses of 14 and 99 mg/m². Patients received a median of three treatment cycles though one MM patient underwent 19 treatment cycles and had two dose escalations with no toxicity. Overall, six patients (40%) had KW-2478 related toxicities (seven grade 1, seven grade 2 and two grade 3) and there were three grade 4 toxicities, not related to KW-2478. There were no trends noted for any specific toxicities associated with KW-2478, though one grade 1 and one grade 2 KW-2478 related episodes of hypertension were observed in a single patient, leading to hospitalization overnight. Two grade 3 related episodes of QTc interval prolongation were reported in one patient. Overall, KW-2478 related toxicities led to dose interruption (6%) or withdrawal from the study (13%). Two patients died from disease progression and no patients died from treatment related causes. Peak plasma concentrations were observed at the end of the infusion, after which KW-2478 plasma concentrations decayed in a biphasic manner with a dose-independent half-life of approximately 6 h. For Day 1, maximum plasma concentration (C_max) and area under the KW-2478 concentration-time curve from time zero to infinity (AUC_0–∞) values increased in a dose-dependent manner at all doses. Repeated 5-day administration had no effect on plasma concentration at all doses. Mean [range] Day 5 KW-2478 C_max varied from 656 [482–1050] ng/mL at the 14 mg/m² dose to 2977 [2440–3640] ng/mL at 99 mg/m². Mean [range] Day 5 KW-2478 AUC to last measurable concentration (AUC_0-t) varied from 836 [532–1336] ng/mL at the 14 mg/m² dose to 3465 [2997–4318] ng/mL at 99 mg/m². Pharmacodynamic data were variable but there was a trend toward increased Hsp70 expression with increasing doses of KW-2478. This trend was most evident at 71 and 99 mg/m², at 8 h post dose on Day 5. Once-daily infusions at doses of up to 99 mg/m² achieved exposure levels similar to those that showed anti-tumor activity in pre-clinical models including models of human MM. Hsp70 induction data suggested that these exposure levels may be sufficient to affect function of Hsp90 in B-cell malignancies. Overall, KW-2478 was well tolerated with no DLTs at doses up to 99 mg/m² and demonstrated a predictable pharmacokinetic profile in its target population. The study is proceeding with a KW-2478 dose of 132 mg/m²; further dose escalation is planned.