Population pharmacokinetic modelling of decitabine in patients with myelodysplastic syndromes (MDS) and acute myeloid leukemia (AML).

Abstract

6551 Background: Decitabine is a deoxycytidine nucleoside derivative inhibitor of DNA-methyltransferases indicated for treatment of MDS patients and currently being investigated for the treatment of AML patients. Aim of this analysis was to build a population pharmacokinetic (PK) model that describes decitabine PK in MDS and AML patients following two different decitabine IV infusion schedules. Methods: Nonlinear mixed effect modelling was performed on plasma decitabine concentrations from 59 patients in 4 clinical trials; 1 phase 1 trial with decitabine given at 15-mg/m2 as 3-hour infusion every 8 hours on days 1-3 in a 6-week cycle; and 2 phase 2 and 1 phase 3 trials with a convenient dosing schedule of 20-mg/m2 as 1-hour infusion on days 1-5 in a 4-week cycle. Gender, weight, body surface area (BSA), body mass index, total bilirubin concentration, and calculated creatinine clearance (CrCL) were the covariates investigated in the analysis. Results: Decitabine PK was characterized by a linear 2-compartment model with rapid elimination from the central compartment and slow distribution to a peripheral compartment. For a typical patient (weight 70 kg/BSA 1.73 m2) the decitabine PK parameters following 20-mg/m2 regimen were: Vd = 69.1 L (95%CI: 49.7 - 89 L), CL = 298 L/hr (95%CI: 249 - 359 L/hr), Vp= 47.0 L (95%CI: 32.5 – 65.8 L), Q = 32.3 L/hr (95%CI: 24.4 – 43.8 L/hr). Vd was proportional to body weight while CL was proportional to BSA with inter-individual variability for Vd and CL of 70.6% and 55.7%, respectively. Intra-individual variability of decitabine concentrations was moderate (CV=31.6%). CL, Q, and Vp for the 15-mg/m2 regimen were 32.5% (95%CI: 18.4 – 43.9%) lower than for the 20-mg/m2 regimen while Vd was 74.1% (95%CI: 60.3 – 83.7%) lower. Clearance for females was 12.1% (95%CI: 0.5-25.6%) lower than for males but the effect was small relative to the inter-individual variability (55.7%) and clinically not relevant. The PK parameters were independent of time. There was no evidence of parameter dependencies on age, CrCL, total bilirubin concentration, or disease (AML or MDS). Conclusions: The PK of decitabine was well described by a linear two-compartment model.