Abstract CT019: Population pharmacokinetics of an aurora kinase A inhibitor, LY3295668 erbumine (AK-01), in patients with locally advanced or metastatic solid tumors

Background: The aurora kinase family plays a vital regulatory role in mitotic and meiotic events, with aurora kinase A (AurA) critical to centrosome maturation, mitotic spindle formation and checkpoint activation. LY3295668 erbumine (AK-01) is an inhibitor of AurA with 1000-fold selectivity for inhibition of AurA over aurora kinase B (AurB). Pharmacokinetic-pharmacodynamic (PK-PD) nonclinical models have shown that 90% inhibition of AurA phosphorylation (pAurA) over an extended period (≥ 16 hours per day) was required to yield efficacy in H446 (small cell lung cancer) xenograft models. Nonclinical plasma LY3295668 concentrations associated with 90% pAurA inhibition (IC90) were estimated previously using a direct sigmoidal relationship between PK and pAurA. In this first-in-man study, PK of LY3295668 in patients with advanced/metastatic cancer were evaluated. Human PK profiles were simulated subsequently and compared to pAurA IC90 to estimate a human efficacious dose range. Methods: LY3295668 erbumine was administered to 13 patients with locally advanced or metastatic solid tumors over a dose range of 25 to 75 mg given twice daily (BID). In the first cycle, serial plasma PK samples were collected up to 8 hours on Day 1 and Day 15. Additional trough samples were obtained prior to dosing on Day 2, on Day 8 and at discontinuation if applicable. PK parameter estimations were performed using non-linear mixed effects models in NONMEM v7.3. Results: The population PK model was developed using 146 plasma concentrations collected from 13 patients. A 2-compartment PK model with first order absorption best described the disposition of LY3295668. Following oral administration, LY3295668 was rapidly absorbed with peak concentrations occurring within 1 to 2 hours. The estimated elimination half-life was approximately 21 hours; therefore, steady state was attained within 4 to 5 days of dosing. Simulations of plasma concentration-time profiles showed that at the maximum tolerated dose (MTD) of 25 mg BID, 90% of patients are expected to achieve steady-state plasma concentrations greater than the pAurA IC90 for the entire day. Patients with dose limiting toxicities (DLTs) had the highest model-predicted exposures amongst the 13 patients. Conclusion: At the MTD of 25 mg BID, steady state LY3295668 plasma concentrations are maintained above the pAurA IC90 for the entire dosing interval, exceeding the minimum requirements for efficacy associated with non-clinical xenograft mode