Apixaban in End-stage Kidney Disease : A Pharmacokinetics Study

RATIONALE

Chronic Kidney Disease (CKD) is one of the epidemics of modern times. In the recent National Health and Nutrition Examination Survey (NHANES), the overall prevalence of CKD (stages 1 to 4) increased from 10.0% in 1988-1994 to 13.1% in 1999-2004 (1). Mounting data point to the lethal synergy between chronic kidney disease (CKD) and cardiovascular disease (CVD) (2,3,4) Atrial fibrillation (AF) is an important co-morbid condition in patients with CKD. While prevalence differs from region to region, it is estimated that between 5 and 10% of dialysis patients have AF (5). Analyses of the US renal data system (USRDS) demonstrate that the prevalence of AF in patients with hemodialysis continues to rise (6). Prevalent AF is positively associated with all-cause mortality and stroke, similar to that of the general population (5).

The choice of the optimal anticoagulation regimen for hemodialysis patients with AF is hampered by lack of randomized controlled trials. Observational data support the use of the CHADS2-score for risk stratification of AF in HD (5). Cost-utility analyses suggest that warfarin appears to be the optimal therapy to prevent thromboembolic stroke in hemodialysis patients with AF (7). These analyses however extrapolate the decrease in risk of stroke observed in randomized trials of oral vitamin K antagonist therapy in the general population (7). Both retrospective and prospective observational data however observe an increased risk for stroke in HD patients receiving oral vitamin K antagonist therapy for AF (8,5).

Recent findings shed a worrisome light on the pleiotropic effects of oral vitamin K antagonist therapy in patients with advanced chronic kidney disease. Besides promoting undercarboxylation of the pro- and anticoagulant factors (II, VII, IX, X, protein S, protein C) several other glutamine-domain containing proteins remain undercarboxylated. Matrix-Gla protein is a powerful calcification inhibitor. Nonphosphorylated matrix-Gla protein has been associated with overall mortality and cardiovascular mortality in HD patients (9). Observational data suggest that patients treated with oral vitamin K antagonists demonstrate higher valvular and coronary calcium (10). Prospective trials currently investigate the effects of vitamin K supplementation on the evolution of vascular calcification in patients with end-stage kidney disease and on aortic valve calcifications.

Apixaban is a novel oral direct factor Xa inhibitor; In patients with atrial fibrillation, apixaban was superior to warfarin in preventing stroke or systemic embolism, caused less bleeding, and resulted in lower mortality (the ARISTOTLE trial) (11). An important benefit of apixaban as compared to other recent NOACs, e.g. dabigatran and rivaroxaban, is that the elimination is less dependent on kidney function as only 25% is excreted renally and 75% is excreted through the hepatobiliary system, with a mean elimination half-life of 8-15 hours in healthy adults.

Given its favorable outcome profile compared to oral vitamin K antagonists in patients with normal kidney function and in patients with mild to moderate kidney disease and given the potential serious side-effects of oral vitamin K antagonists in end-stage kidney disease, apixaban may be an attractive alternative for systemic anticoagulation in dialysis patients. The pharmacokinetics of apixaban in end-stage renal disease is not well characterized.

HYPOTHESIS The pharmacokinetics profile of apixaban allows safe use in patients with end-stage renal disease

STUDY DESIGN Open label non-randomized phase II pharmacokinetics study.

STUDY AIMS

Primary:

To determine inter-dialytic pharmacokinetics of Apixaban

Secondary:

• To determine intra-dialytic pharmacokinetics of Apixaban
• Dose finding of apixaban in patients treated with hemodialysis

DOSING AND TIMING Part A - interdialytic kinetics Dose titration

1. single dose 2.5 mg post-dialysis
2. single dose 5 mg post-dialysis As study drug is administered immediately following dialysis, part A will provide inter-dialytic PK data (primary aim)

Part B - intra- plus interdialytic pharmacokinetics Dose titration

1. single dose 2.5 mg 30 minutes pre-dialysis
2. single dose 5 mg 30 minutes pre-dialysis As study drug is administered 30 minutes pre-dialysis, part B will provide intra-dialytic kinetic PK data (secondary aim). As we expect that study drug effect is not completely lost at the end of the dialysis session, we will continue sampling blood at fixed time-points after end of dialysis. These data may contribute to interdialytic PK (primary study aim).

NUMBER OF STUDY PARTICIPANTS

We aim to include 6 patients for each part and dosage:

Part A - Interdialytic kinetics 2.5 mg N = 6 5 mg N = 6

Part B - Intra- plus interdialytic kinetics 2.5 mg N = 6 5 mg N = 6

Based on previous publications on single dose pharmacokinetics of apixaban in healthy volunteers (12), we expect that 6 patients per part will provide sufficient data to generate PK profiles.

SCHEDULED BLOOD SAMPLING

Part A - Interdialytic kinetics (post-dialysis dosing)

• Sampling at start of first dialysis (pre)
• Sampling at end of dialysis (post, t = 0)
• Sampling for pharmacokinetics (t =0.5, 1, 2, 4, 8, 24 hours post dialysis)
• Sampling at start of second dialysis (t = 44 hours)
• Sampling at end of second dialysis (t = 48 hours)

Part B - Intradialytic kinetics (pre-dialysis dosing)

• Sampling at start of first dialysis (pre)
• Sampling for pharmacokinetics ( t=0.25, 0.5, 1, 1.5,2, 3,4 hours after dialysis start)
• Sampling at end of dialysis (post, t = 0)
• Sampling for pharmacokinetics (t =0.5, 1, 2, 4, 24 hours post dialysis)
• Sampling at start of second dialysis (t = 44 hours)
• Sampling at end of second dialysis (t = 48 hours)

ANALYTICAL TECHNIQUES

Apixaban concentrations (ng/mL) will be determined based on a calibration curve with Apixaban (Apixaban Calibrator and controls, Hyphen).

Conventional anticoagulation used for hemodialysis is by heparin or low molecular weight heparins (LMWH). This will lead to significant interference with the proposed chromogenic anti-Xa activity assays. To exclude interference of anticoagulation for hemodialysis during the study period, the routine anticoagulation will be switched to regional citrate anticoagulation according to local protocol during the study period. The following will be applied.

• One week wash-out period of (low molecular weight) heparin, during which hemodialysis anticoagulation is according to the local regional citrate anticoagulation protocol (no use of (low molecular weight) heparins).
• During the study period hemodialysis anticoagulation is according to the local regional citrate anticoagulation protocol (no use of (low molecular weight) heparins).

PHARMACOKINETICS ANALYSES

Anti-Xa activity values (IIU/mL) will be converted to apixaban concentration data (ng/mL) based on the previously demonstrated linear relationship (e.g. Frost et al., 2014 - PMID: 24697979). Apixaban concentration-time profiles will be generated and observed values for the descriptive PK parameters Cmax (peak plasma concentration) and time to Cmax (Tmax) will be determined directly from these profiles. PK profiles will be further analyzed with non-compartmental analysis (NCA). Briefly, the slope (λ) of the terminal phase of the concentration-time profile will be determined by log-linear regression on the appropriate number (typically at least 3) of data points. The terminal (elimination) half-life (t1/2, λ) will be calculated from Ln(2)/λ. The area under the curve (AUC) between administration (time 0) and the last measurable data point (AUC0-T) will be calculated with the 'Lin up/Log down' trapezoidal method. The AUCT-∞ will be obtained from the last measureable concentration divided by λ, and will be summed with AUC0-T to obtain AUC0-∞ (total exposure). Similarly, after plotting plasma concentration-time products as a function of time, Area under the first Moment Curve (AUMC) values will be calculated. The mean residence time (MRT, h) will be obtained from AUMC/AUC. Mean half-life (t1/2) will be calculated as Ln(2)×MRT. Oral ('apparent') clearance (Cl/F) values will be obtained by dividing dose by AUC. The volume of distribution at steady (Vdss/F) will be obtained as MRT x Cl/F. As alternative to NCA, the feasibility of analyzing the data with a 1-compartmental model with first order absorption will be explored as well. Summary statistics will be tabulated for all relevant PK parameters: (i) the descriptive parameters Cmax, Tmax and AUC0-∞ (total exposure), (ii) MRT and mean t1/2, (iii) the primary (fundamental) PK parameters Cl/F and Vd/F. Reported parameters will enable appropriate comparison with previously published data on apixaban PK in various populations. For the purpose of dose finding, AUC0-∞ (total exposure) represents the key parameter.