Analysis of antiplatelet effect of ticlopidine in humans: modeling based on irreversible inhibition of platelet precursors in bone marrow.

Abstract

The relationship between plasma concentration of ticlopidine and its inhibitory effect on platelet aggregation in human was analyzed using a pharmacokinetic/pharmacodynamic (PK/PD) model. The data of plasma concentration and inhibitory effect on platelet aggregation were taken from the literature. A two-compartment open model was fitted to plasma ticlopidine concentrations. Assuming that ticlopidine acts on platelet precursors in the bone marrow, the apparent reaction rate constant of ticlopidine and platelet precursors (K), apparent transformation rate constant of platelet precursors (kr) and apparent elimination rate constant of platelets (ke) were estimated. The estimated values +/- S.D. were 1.01 +/- 1.08 ml micrograms-1 hr-1 for K, 0.265 +/- 0.259 hr-1 for kr and 0.0747 +/- 0.0112 hr-1 for ke. The antiaggregation effects of ticlopidine on platelets after administration of 100, 200, and 300 mg (bid for 8 days) were simulated using the PD parameters of K, kr, and ke. While the antiaggregation effect reached steady state within 3-4 days without dose dependency of the interval, the maximum effect increased with dose. Furthermore, changing the elimination rate constant of ticlopidine from the central compartment in the model significantly changed the duration of inhibitory effect of ticlopidine on platelet aggregation. Therefore, the reported long duration of antiplatelet effect after discontinuation of ticlopidine, which is believed to be irreversible binding to the platelet, might have been partially caused by the delayed plasma elimination after a long therapy of ticlopidine. On the other hand, the mean life-span of platelets in the blood estimated by 1/ke after administration of ticlopidine was 14 hr, far below the life-span of platelets in the blood. For a more detailed analysis of the antiplatelet effect of ticlopidine, the possible contribution of reversible binding of the drug to glycoprotein IIb/IIIa should be considered in future PK/PD models.