Dongwoo Kang, SaeHeum Song, Ophelia Yin, Dolly Parasrampuria, Raymond Miller
Daiichi Sankyo Pharma Development
Objectives: To describe the relationship between the exposure of Edoxaban (a direct Factor Xa inhibitor) and the activity of intrinsic Factor Xa (iFXa) and bleeding events.
Methods: A population PK model of edoxaban was built using 13 Phase I and 5 Phase II studies including 11,444 PK samples from 1,624 subjects. Individual post hoc concentration values were used to model the iFXa data from a 3-month Phase II study[1] evaluating the safety of the four doses of edoxaban (30mg QD, 60mg QD, 30mg BID, and 60mg BID) in patients with NVAF compared to warfarin. The iFXa was determined with a two-stage chromogenic method (Biogenic, Tokyo, Japan). A dynamic binding model was fit to the iFXa data from 585 patients who received edoxaban. For both the PK and PD models of edoxaban, the first order conditional estimation with interaction (FOCEI) estimation as implemented in NONMEM V.7.2 was used to obtain estimates of the model parameters and the variance-covariance matrix. Logistic regression analysis was then employed to find the best predictor for the bleeding event.
Results: A two-compartment PK model adequately described the plasma edoxaban concentration data from the 18 studies. The dynamic binding model fit the iFXa data better than other PD models (eg, equilibrium model) due to the mechanism of edoxaban binding to the intrinsic FXa. Logistic regression analysis showed that the time duration of iFXa suppressed at a certain threshold value or lower best correlated with the bleeding events observed from the Phase II study, and that the observed bleeding order matched the order of the time duration for the threshold of 15% activity: 30 mg QD (5.5 % / 9.2 hr), 60 mg QD (7.3 % / 13.7 hr), 30 mg BID (12.7 % / 18.8 hr), 60 mg BID (18.3 % / 21.6 hr). Therefore, 60 mg QD could be administered with less risk of bleeding than 30 mg BID at the same daily dose of 60 mg.
Conclusions: The established PK and PD models of edoxaban have good predictability with respect to the observed data. The prolonged period of iFXa suppression, despite the same total daily dose, may provide a biological explanation to the greater bleeding risk observed from 30mg BID compared to 60mg QD. The 60mg QD and 30mg QD regimens showed similar and less bleeding risk to the warfarin control treatment (8.0%), respectively, and were chosen for the phase 3 trials [2,3].
References:
[1] Weitz JI, Connolly SJ, Patel I, Salazar D, Rohatagi S, Mendell J, Kastrissios H, Jin J, and Kunitada S. Randomised, parallel-group, multicentre, multinational phase 2 study comparing edoxaban, an oral factor Xa inhibitor, with warfarin for stroke prevention in patients with atrial fibrillation. Thrombosis and Haemostasis 104(3):633-641, 2010.
[2] Giugliano RP, et al. Edoxaban versus warfarin in patients with atrial fibrillation. The New England Journal of Medicine 369(22):2093-2104, 2013. [3] The Hokusai-VTE Investigators. Edoxaban versus warfarin for the treatment of symptomatic venous thromboembolism. The New England Journal of Medicine 369(15):1406-1415, 2013.
Reference: PAGE 23 () Abstr 3258 [www.page-meeting.org/?abstract=3258]
Poster: Drug/Disease modeling - Other topics