A Phase I Study of the Kinesin Spindle Protein Inhibitor Ispinesib (SB-715992) in Relapsed/Refractory Acute Leukemia

Background. Ispinesib is a novel polycylic, nitrogen-containing heterocycle small molecule that acts by inhibition of the mitotic kinesin spindle protein (KSP) HsEg5. Pre-clinical studies demonstrate activity against both murine and human solid tumor and leukemia cell lines by inhibition of adenosine diphosphate release from KSP, causing monopolar spindle formation and programmed cell death.

Methods. This was a single agent, phase I dose-escalation trial in patients with relapsed/refractory acute myeloid (AML) or lymphoid leukemia (ALL) or advanced myelodysplastic syndrome (MDS). Intravenous ispinesib was given for 3 consecutive days every 21 days. Doses were escalated from 1.5 mg/m²/day to 13.3 mg/m²/day in cohorts of 3 patients. Clinical and peripheral blood responses were monitored on at least a weekly basis and bone marrow evaluations were done every three weeks. Grade 3 or greater non-hematologic toxicities were considered dose limiting toxicities (DLT) with the exception of infectious complications, electrolyte abnormalities or self-limited gastrointestinal toxicities.

Results. Thirty-six adults (14 women, 22 men) with a median age of 66.5 years (range 28–82 years) were enrolled. The median number of previous induction treatments was 3, with a range from 0 to 8. Two patients had MDS, 5 had ALL and 29 had AML. Patients received a median number of 1.5 cycles of ispinesib, with a range from 1 to 5. The maximum tolerated dose (MTD) was determined to be 10 mg/m²/day given on 3 consecutive days. DLTs related to the study drug were grade 3 transaminitis and hyperbilirubinemia, and these developed in a total of 6 patients; one at dose level 3 mg/m²/day, 2 at dose level 7.5 mg/m²/day and 3 at dose level 13.3 mg/m²/day. Five patients developed self-limited grade 3 mucositis at dose level 13.3 mg/m²/day. Three patients died on study. One died due to disease progression, and two died both due to infection and hepatic dysfunction. None of the patients had normalization of their blood counts. Three patients with very high peripheral blast counts had rapid clearing of the blasts during treatment. Two patients had late bone marrow responses with less than 5% blasts at dose level 10 mg/m²/day after cycle 2 and cycle 3, respectively; both subsequently died of infection without count recovery. An additional three patients had a decrease in bone marrow cellularity with persistence of leukemic blasts.

Conclusions. The MTD of ispinesib in this patient population and on this schedule was higher than that observed in prior trials of non-hematologic malignancies (30 mg/m²versus 18 mg/m²). The toxicity profile was different with DLT of hepatic toxicity in this trial and DLT of neutropenia in the solid tumor trials. Ispinesib appeared to provide myelosuppression in some patients at higher dose levels; however, there appeared to be minimal effect on the leukemic marrow component. Even in the majority of patients with myelosuppression, the leukemic clone persisted and proliferated following treatment. Ispinesib was effective at clearing the peripheral blood of blasts in a minority of patients. Ispinesib, given as a single agent in this trial on this schedule, is not an effective anti-leukemic therapy.