Pharmacokinetics and Dialytic Clearance of Apixaban During In Vitro Continuous Renal Replacement Therapy

Abstract

Objective: To evaluate the transmembrane clearance (CLTM) of apixaban during modeled in vitro continuous renal replacement therapy (CRRT), assess protein binding and circuit adsorption, and provide initial dosing recommendations. Design: In vitro pharmacokinetic (PK) study.Setting: University research laboratory.Subjects: Not applicable. Interventions: Apixaban was added to the CRRT circuit and serial, undiluted pre-filter bovine blood samples were collected along with analogous post-filter blood and effluent samples. All experiments were performed in duplicate using continuous veno-venous hemofiltration (CVVH) and hemodialysis (CVVHD) modes, with varying filter types (M150 and HF1400), flow rates (2 and 4 L/h), and point of CVVH replacement fluid dilution (pre, post, and pre/post filter). Concentrations of apixaban and urea were quantified via liquid chromatography-tandem mass spectrometry. Plasma PK parameters for apixaban were estimated via noncompartmental analysis in WinNonlin. CLTM was calculated via the estimated area under the curve (AUC) and by the product of the sieving/saturation coefficient (SC/SA) and flow rate. Two and three-way ANOVA models were built to assess the effects of mode, filter type, flow rate, and point of dilution on CLTM by each method. Optimal doses were suggested by matching the AUC observed in vitro to the systemic exposure demonstrated in Phase 2/3 studies of apixaban. Linear regression was then utilized to provide dosing estimations for flow rates from 0.5-5 L/h. Measurements and Main Results: Mean adsorption to the HF1400 and M150 filters differed significantly at 38% and 13%, respectively, while mean (±SD) percent protein binding was 70.81±0.01%. Effect of CVVH point of replacement fluid dilution did not differ across filter types, although CLTM was consistently significantly higher during CRRT with the HF1400 filter compared to the M150. The three-way ANOVA demonstrated improved fit when CLTM values calculated by AUC were used (adjusted R2 0.87 vs. 0.52), and therefore, these values were used to generate optimal dosing recommendations. Linear regression revealed significant effects of filter type and flow rate on CLTM by AUC, suggesting doses of 5-10 mg BID may be needed for flow rates ranging from 0.5-5 L/h respectively. Conclusion: CLTM of apixaban during CRRT resulted in estimated dosing recommendations ranging from 5 mg BID for flow rates ≤3 L/h up to 7.5-10 mg BID for rates >3 L/h, depending on filter type, in order to match target systemic exposure thresholds. The safety and efficacy of these proposed dosing regimens warrants further investigation in clinical studies.