Unraveling the pharmacokinetic interaction of ticagrelor and MEDI2452 (Ticagrelor antidote) by mathematical modeling

Abstract

The investigational ticagrelor‐neutralizing antibody fragment, MEDI2452, is developed to rapidly and specifically reverse the antiplatelet effects of ticagrelor. However, the dynamic interaction of ticagrelor, the ticagrelor active metabolite (TAM), and MEDI2452, makes pharmacokinetic (PK) analysis nontrivial and mathematical modeling becomes essential to unravel the complex behavior of this system. We propose a mechanistic PK model, including a special observation model for post‐sampling equilibration, which is validated and refined using mouse in vivo data from four studies of combined ticagrelor‐MEDI2452 treatment. Model predictions of free ticagrelor and TAM plasma concentrations are subsequently used to drive a pharmacodynamic (PD) model that successfully describes platelet aggregation data. Furthermore, the model indicates that MEDI2452‐bound ticagrelor is primarily eliminated together with MEDI2452 in the kidneys, and not recycled to the plasma, thereby providing a possible scenario for the extrapolation to humans. We anticipate the modeling work to improve PK and PD understanding, experimental design, and translational confidence.

Hemostatic effects of the ticagrelor antidote MEDI2452 in pigs treated with ticagrelor on a background of aspirin

“…There was a slight delay between MEDI2452 plasma exposure and the pharmacodynamic effect, as maximal MEDI2452 exposure occurred at 5 min and ADP-induced aggregation was close to normal at 60 min and all time-points after 60 min, apart from 90 min, when it was not significantly different from the ASA group. This is in line with previously reported mouse data [19,20]. The relative inhibition induced by ASA alone could not be calculated, as no sample was collected before the ASA dosing.…”

“…3). The published ticagrelor-MEDI2452 mathematical interaction model developed in the mouse [20], and here translated to the pig, predicted the MEDI2452 pharmacokinetics reasonably well, including the time courses of total and free plasma concentrations of ticagrelor and AR-C124910XX (Data S1).…”

“…To induce significant ticagrelor-dependent bleeding in the model, the ticagrelor dose was selected to yield a supra-therapeutic plasma concentration, and to achieve high inhibition of adenosine diphosphate (ADP)-induced platelet aggregation throughout the duration of the experiment. The MEDI2452 dose, 600 mg kg À1 , was selected as a high, investigational dose where the exposure would exceed the exposure of ticagrelor and induce rapid elimination of the free fraction of ticagrelor and AR-C124910XX as predicted by a pharmacokinetic interaction model translated from the published mouse ticagrelor-MEDI2452 model [20] (Data S1).…”

“…2). The reason for this increase in total concentrations is that MEDI2452 acts like a ticagrelor buffer in the plasma compartment and redistributes ticagrelor and AR-C124910XX from non-plasma (tissue) compartments to the plasma compartment [20]. In the absence of MEDI2452, the concentrations of total ticagrelor and AR-C124910XX in the urine were just above LLOQ (0.010 lM), whereas the addition of MEDI2452 resulted in a marked increase in mean concentrations of total ticagrelor and AR-C124910XX in the urine (Fig.…”

“…Ticagrelor reduced survival (4/9 vs. 8/8, P < 0.05), reduced survival time (169 vs. 240 [240-240] min, P < 0.05) and increased bodyweight-adjusted total blood loss (37 vs. 16 [13][14][15][16][17][18][19][20][21][22][23][24][25] mL kg À1 , P < 0.05) compared with ASA alone (Fig. 5).…”