Rivaroxaban population pharmacokinetic and pharmacodynamic modeling in Iranian patients

What is known and objective: Although predictable pharmacokinetic and pharmacodynamic of rivaroxaban allow fixed dosing regimens without routine coagulation monitoring, there is still the necessity to monitor and predict the effects of rivaroxaban in specific conditions and different populations. The current study was designed and conducted to analyze the rivaroxaban population pharmacokinetics in Iranian patients and establish a pharmacokinetic/pharmacodynamic model to predict the relationship between rivaroxaban concentration and its anticoagulant activity.

Methods: A sequential nonlinear mixed effect pharmacokinetic/pharmacodynamic modeling method was used to establish the relation between rivaroxaban concentration and anti-factor Xa activity, prothrombin time, and activated partial thromboplastin time (aPTT) as pharmacodynamic biomarkers in a population of sixty-nine Iranian patients under treatment with oral rivaroxaban. Rivaroxaban plasma concentration was quantified by a validated high-performance liquid chromatography-tandem mass spectrometry.

Results and discussion: The typical population values (inter-individual variability%) of the oral volume of distribution and clearance for a one-compartment model were 61.2 L (21%) and 3.68 L·h^-1 (61%), respectively. Creatinine clearance and Child-Turcotte-Pugh score were found to affect the clearance. A direct link linear structural model best fitted the data for both prothrombin time and aPTT. The baseline estimates of aPTT and prothrombin time in the population were 35.0 (15%) and 12.6 (2%) seconds, respectively. The slope of the relationship between apTT, prothrombin time, and rivaroxaban concentration was 0.033 (28%) and 0.018 (54%) s·ml·ng^-1 , respectively. The selected model for anti-factor Xa activity consisted of a direct link inhibitory E_max model with Hill coefficient. The maximum level of inhibition (E_max ) was 4 IU·ml^-1 . The concentration of rivaroxaban producing 50% of the maximum inhibitory effect (EC₅₀ ) was 180 (24%) ng·ml^-1 , and Hill coefficient (γ) was 1.44 (108%). No covariates showed a statistically significant effect on PT and activated partial thromboplastin time prolonging properties and anti-factor Xa activity.

What is new and conclusion: Our results confirmed that pharmacokinetic/pharmacodynamic models similar to those of the other studies describe the relationship between the rivaroxaban concentration and its anticoagulant effect in Iranian patients. However, considerable differences were observed in the parameters of the pharmacodynamics-pharmacokinetic models with the results of other reports that can explain the unpredictable effects of rivaroxaban in some patients.