Pharmacometric characterization of dabigatran hemodialysis

Abstract

Background: Hemodialysis has been shown to be a useful method of decreasing dabigatran plasma levels in situations that require rapid elimination of this thrombin inhibitor. However, there is currently no clinical recommendation for the accelerated/optimized elimination of dabigatran via hemodialysis (e.g., flow rates, filter type, duration of dialysis).

Objectives: The primary objective of the present work was to characterize, via pharmacometric methods, the effects of different blood flow rates in hemodialysis on the pharmacokinetics of dabigatran, using data from a dedicated phase I dialysis study of end-stage renal disease (ESRD) patients. In addition, the effects of various clinically relevant hemodialysis settings were evaluated by simulation to assess their potential use in non-ESRD situations.

Methods: Seven patients with ESRD were investigated in an open-label, fixed-sequence, two-period comparison trial. A population pharmacokinetic model was developed to fit the data and then used for various simulations. Data analyses were performed using NONMEM(®), Berkeley Madonna, or SAS.

Results: The pharmacokinetics of dabigatran were best described by a two-compartment model with first-order absorption and a lag time. In addition to total body clearance in ESRD subjects, a first-order dialysis clearance was implemented which was greater than zero during hemodialysis and zero during the interdialytic periods. The relationship between the dialysis clearance and the blood flow rate was best described by the Michaels function. Simulations showed that varying clinically relevant dialysis settings such as filter properties or flow rates had only minor effects. Dialysis duration had the strongest impact on dabigatran plasma concentration. The observed geometric mean redistribution effect after hemodialysis was low (<16 %). The final model was successfully evaluated through the prediction of plasma concentrations from a case report undergoing dialysis.

Conclusions: This analysis allowed the influences of various hemodialysis parameters on the dabigatran plasma concentration to be predicted in detail for the first time. Dialysis duration was identified as having the strongest impact on the reduction in dabigatran plasma concentration. The model developed here can potentially serve as a tool to provide guidance when considering the use of hemodialysis in patients who have received dabigatran.

Fig. 1

Schematic representation of the study design for a single period. PK pharmacokinetic

Fig. 2

a Schematic illustration of the final pharmacokinetic model. b Relationship between blood flow rate and apparent dialysis clearance of dabigatran for a dialysate flow rate of 700 mL/min and an estimated mass transfer-area coefficient of 313 mL/min

Fig. 3

Goodness-of-fit plots of the final population pharmacokinetic model. Observed plasma concentrations vs. population predictions (a) and individual predictions (b). Solid lines indicate lines of identity. Conditional weighted residuals vs. population predictions (c) and time after first dose (d)

Fig. 4

Simulated percentage changes in dialysis clearance and the resulting changes in reduction in plasma concentration for various dialysis settings compared to the reference. Reference settings were: blood flow rate (BFR) = 200 mL/min, dialysate flow rate (DFR) = 700 mL/min, filter = large surface, high flux (Gambro PF-210H). For details, see “Methods”

Fig. 5

Simulated time profile of plasma concentration for the maximum redistribution effect observed after a 10-h dialysis in a patient with a creatinine clearance of 30 mL/min (solid line) and without dialysis (dotted line)

Fig. 6

a Predicted reduction in plasma concentration (%) vs. duration of dialysis displayed for patients with different creatinine clearance values. b Predicted plasma concentrations vs. duration of dialysis for three different initial concentrations representing the observed median and the 90th and 99.6th percentiles of the trough plasma concentrations of the 150 mg dose group in the RE-LY trial. CL _CR creatinine clearance

Fig. 7

External prediction of the time profile of the plasma concentrations measured in a patient undergoing hemodialysis. Measured values are shown as open circles. Predicted median and 5th and 95th percentiles are represented as lines