Phase I pharmacokinetic (PK) and pharmacodynamic (PD) study of MLN8237, a novel selective aurora A kinase (AAK) inhibitor, in patients (pts) with advanced solid tumors

2565 Background: MLN8237 is a second-generation, selective AAK inhibitor, designed to avoid benzodiazepine-like effects (somnolence) observed with another agent, MLN8054. This phase I trial examines the safety, PK, and PD of MLN8237 administered orally (PO) in pts with advanced solid tumors. Methods: Pts received MLN8237 PO once daily (QD) or twice daily (BID) for 7 days (d) followed by a 14d recovery period. Tumor and skin biopsies were done serially to evaluate bioactivity measured by mitotic index and chromosome and spindle abnormalities in mitotic cells. Doses were increased in cohorts of 3–6 pts until dose-limiting toxicity (DLT) evaluated in the first cycle was seen in ≥2 pts in a cohort. Results: As of 12-Dec-08, 27 pts received MLN8237 (median 2 cycles, range 1–10+); median age was 57 yr (range 31–78). The first cohort received 5mg QD. Higher QD doses were safe in another Phase 1 study (Infante et al, EJC Suppl 2008;6[12]:90), so subsequent cohorts in this study started at 80mg QD (Table). DLTs at 60–100mg PO BID included neutropenia, pancytopenia, stomatitis, and somnolence. Somnolence (of any grade) was seen with QD dosing, but was reduced with divided doses each <100 mg. Antitumor activity was seen in a pt in cohort 4 with a large treatment-resistant pleomorphic liposarcoma. Preliminary PK analyses showed dose-dependent increases in AUC 0–24 and C max . Consistent with Aurora A inhibition, increased mitotic index in skin and tumor biopsies after MLN8237 dosing was coupled with decreased bipolar spindles and aligned chromosomes in mitotic cells from tumor biopsies. Conclusions: At the doses and schedule evaluated, MLN8237 is tolerable with BID dosing x7d and exhibits favorable PK, PD, and clinical antitumor activity. Additional planned cohorts include extended dosing for 14–21d in 28–35d cycles. The results support future phase 2 development. [Table: see text] [Table: see text]