Basic & Clinical Pharmacology & Toxicology, 2014, 115, 352–359

Doi: 10.1111/bcpt.12222

Semi-Mechanistic Modelling and Simulation of Inhibition of Platelet Aggregation by Antiplatelet Agents Hwi-yeol Yun1,†, Wonku Kang2,3,†, Byung-yo Lee1, Sunkyung Park2, Young-Ran Yoon4, Jin Yeul Ma5 and Kwang-il Kwon1 1 College of Pharmacy, Chungnam National University, Daejeon, Korea, 2College of Pharmacy, Yeungnam University, Gyeongsan, Korea, 3College of Pharmacy, Chung-Ang University, Seoul, Korea, 4Department of Biomedical Science and Clinical Trial Center, Kyungpook National University, Daegu, Korea and 5Korean Medicine (KM)-Based Herbal Drug Development Group, Korea Institute of Oriental Medicine, Daejeon, Korea

(Received 4 December 2013; Accepted 10 February 2014) Abstract: Antiplatelet agents are a class of pharmaceuticals that decrease platelet aggregation and thus inhibit thrombus formation. We examined the relationships between plasma concentrations of antiplatelet agents (triflusal, clopidogrel and cilostazol) and the platelet aggregation inhibitory effect after dosing. We used triflusal, cilostazol and clopidogrel for the development of a semimechanistic PK/PD model. The drugs chosen are used widely and reflect various mechanisms of antiplatelet agents. Time courses of plasma concentrations of the antiplatelet agents and their platelet aggregation effects were analysed using ADPAT V. Pharmacokinetic profiles were fitted to an extended parent–metabolite pharmacokinetic model, based on a two-compartment model, and the pharmacodynamic effects of the agents were fitted to a platelet aggregation effect model that consisted of the following parameters: Ks, the active-form platelet synthesis rate constant; K, the apparent reaction rate constant of the agent and active-form platelets; Kel-PRP, the apparent rate constant of platelets; and e, an intrinsic activity parameter. This semi-mechanistic PK/PD model described well the relationship between plasma concentrations of antiplatelet agents and platelet aggregation effects. In addition, the estimated parameters were suitable for the explanation of the agents and also have a good correlation with platelet characteristics, such as platelet half-life and platelet aggregation baseline effects. Specifically, we discovered the strong correlations between estimated K parameter and in vitro drug activity. We conclude that this semi-mechanistic PK/PD model explained drug PK/PD characteristics well and will be useful for accurate predictions of antiplatelet effect in the clinical situations.

Platelets, 2–3 lm irregularly shaped cell fragments, function primarily in the formation of blood clots. The average platelet lifespan is generally 5–10 days, and the normal platelet count in a healthy individual is between 150,000 and 450,000/lL blood [1]. Platelets have important roles in thrombosis; thus, antiplatelet therapies are widely used in the primary and secondary prevention of and treatment for thrombotic cerebrovascular or cardiovascular disease. Antiplatelet drugs have various mechanisms of action that involve inhibition of platelet adhesion, aggregation, release and activation (Fig. 1). Although various drugs are used in this area, the most commonly used agents include triflusal, clopidogrel, cilostazol and aspirin. Cilostazol is a selective inhibitor of phosphodiesterase type 3 (PDE3), with a therapeutic effect of increasing cyclic adenosine monophosphate (cAMP). An increase in cAMP results in an increase in the active form of protein kinase A (PKA), which is directly related to the inhibition of platelet aggregation. PKA also prevents the activation of myosin lightchain kinase, an enzyme that is important in the contraction of smooth muscle cells, thereby exerting a vasodilatory effect [2]. Author for correspondence: Kwang-il Kwon; College of Pharmacy, Chungnam National University, Daejeon 305-764, Korea (fax +82-42823-6781; e-mail [email protected]). Jin Yeul Ma; Korean Medicine (KM)-Based Herbal Drug Development Group, Korea Institute of Oriental Medicine, Daejeon 305-811, Korea (fax +82-42-868-9573; e-mail [email protected]). † These two authors contributed equally to this work.

Triflusal and its main metabolite (2-hydroxy-4-trifluoromethyl benzoic acid) are selective platelet antiaggregants, blocking the cyclooxygenase (COX) enzymes, inhibiting thromboxane A2 (TXA2) and preventing aggregation. Triflusal has the same mechanism of antiplatelet aggregation as aspirin and is commonly indicated for the prevention of cardiovascular events, such as stroke, the acute treatment for cerebral infarction and myocardial infarction and thromboprophylaxis in atrial fibrillation [3]. Clopidogrel is a pro-drug, the action of which may be related to an ADP receptor on platelet membranes. The drug specifically and irreversibly inhibits the P2Y12 subtype of the ADP receptor, which is important in the activation of platelets and eventual cross-linking by the protein fibrin. Platelet inhibition can be demonstrated 2 hr after a single oral dose of clopidogrel, but the onset of action is slow, so that a loading dose of 300 mg is usually administered [4]. In general, the relation between plasma concentration and the effect of platelet aggregation has been reported by empirical research method. Pharmacokinetic/pharmacodynamic study of antiplatelet agents via modelling approach has not been reported. PK/PD modelling and simulation is used to support model-based drug development and useful for real applications such as learning, decision-making, study optimization and therapeutic candidate drug selection of antiplatelet agents (Deredot et al., 1999; Lee et al., 2011; Hochholzer et al., 2005; Woo et al., 2002). The objective of this study was to examine the relationship between plasma concentrations of antiplatelet agents (triflusal,

© 2014 Nordic Association for the Publication of BCPT (former Nordic Pharmacological Society)

MODELLING AND SIMULATION OF ANTIPLATELET AGENTS

353

The study was conducted in accordance with good clinical practice and was approved by the local ethics committee at each centre. All individuals provided written informed consent.

Fig. 1. Mechanisms of action of antiplatelet drugs.

clopidogrel and cilostazol) and platelet aggregation inhibitory effects after dosing to assess the usefulness of semi-mechanistic pharmacokinetic (PK)–pharmacodynamic (PD) modelling in describing this relationship. Accurate modelling should enable the prediction of the PD profiles of other drugs with differing activity. Methods Data collection. We collected data retrospectively from three studies and, in total, 82 healthy Koreans participated. All individuals were selected from among those passing an initial clinical screening process, including a physical examination and laboratory tests. Demographic data for the individual are shown in Table 1.

Study design and PK sample analysis. The study designs varied and are summarized in Table 2. In study 1 (n = 20), the individuals were given a single oral dose of 100 mg of cilostazol. Blood samples were collected in heparin-treated tubes before and at pre-dose, 0.5, 1, 2, 3, 4, 6, 8, 10, 24, 32 and 48 hr after the drug administration to determine the plasma concentrations of cilostazol. To determine ADPinduced platelet aggregation, blood samples were collected at predose, 1, 2, 3, 4, 6, 8, 10, 12 and 24 hr after administration of cilostazol. In study 2 (n = 7), the individuals received a single oral dose of 900 mg of triflusal. Blood samples were collected in heparin-treated tubes before and at pre-dose, 0.25, 0.5, 0.75, 1, 1.5, 2, 3, 4, 5, 6, 7, 8, 10, 12, 15, 24, 32 and 48 hr after the drug administration to determine the plasma concentrations of triflusal. To determine arachidonic acidinduced platelet aggregation, blood samples were collected at pre-dose, 1, 2, 3, 4, 6, 8, 10, 24 and 48 hr after administration of the drug. In study 3 (n = 35), individuals received a single 900 mg triflusal oral loading dose on day 1, followed by 600 mg/day maintenance doses (two 300 mg capsules once daily) on days 2–9. To determine the plasma concentrations of HTB (active metabolite), serial blood samples were collected at pre-dose, 24, 48, 96, 144, 168, 192, 192.5, 193, 194, 196, 199, 202 and 216 hr (from day 1 to day 10) after administration of the loading dose. To determine arachidonic acidinduced platelet aggregation, blood samples were collected at 0 (predose), 24, 48, 96, 144, 168, 192, 196, 202 and 216 hr after administration of the loading dose. In study 4 (n = 20), individuals received a single 300 mg clopidogrel oral loading dose on day 1, followed by 75 mg/day maintenance doses on days 2–7. To determine the plasma concentrations of clopidogrel and its metabolites (clopidogrel carboxyl metabolite and

Table 1. Demographic characteristics of individuals.

Study Study Study Study

1 2 3 4

(Cilostazol) (Triflusal single dose) (Triflusal multiple dose) (Clopidogrel)

No. of patients

Age (year)

20 7 35 20

22.5 23.2 24.1 24.3

   

BWT (kg)

2.6 1.5 1.7 2.7

63.2 68.4 70.7 70

   

HT (cm)

13.1 10.3 9.0 8.2

167.2 172.5 176.3 174

   

9.5 4.3 5.0 6.3

Table 2. Summary of the studies included in this analysis. Study 1

Study 2

Study centre No. of individuals Drug Dose

A 20 Cilostazol (Pletal) 100 mg single dose

A 7 Triflusal (Disgren) 900 mg single dose

PK sampling times

Pre-dose, 0.5, 1, 2, 3, 4, 6, 8, 10, 24, 32 and 48

PD sampling times

Pre-dose, 1, 2, 3, 4, 6, 8, 10, 12 and 24

Pre-dose, 0.25, 0.5, 0.75, 1, 1.5, 2, 3, 4, 5, 6, 7, 8, 10, 12, 15, 24 and 48 Pre-dose, 1, 2, 3, 4, 6, 8, 10, 24, 48

Analytical method for PK Analytical method for PD

Validated LC/UV method

Validated LC/UV method

Study 3

Study 4

B 35 Triflusal (Disgren) Multiple dose: 900 mg 9 1 day, 600 mg 9 8 days Pre-dose, 24, 48, 96, 144, 168, 192, 192.5, 193, 194, 196, 199, 202 and 216 Pre-dose, 24, 48, 96, 144, 168, 192, 196, 199, 202 and 216 Validated LC/MS/MS method

B 20 Clopidogrel (Plavix) Multiple dose: 300 mg 9 1 day, 75 mg 9 6 days Pre-dose, 0.5, 1, 1.5, 2, 4, 6, 8, 12 and 24 Pre-dose, 2, 4, 8, 12, 24, 48, 72, 96, 120, 144, 146, 148, 152, 156, 216 and 288 Validated LC/MS/MS method

Platelet aggregation turbidity test

A: Chungnam National University, B: Kyungpook National University.

© 2014 Nordic Association for the Publication of BCPT (former Nordic Pharmacological Society)

HWI-YEOL YUN ET AL.

354

clopidogrel thiol metabolite), serial blood samples were collected at pre-dose, 0.5, 1, 1.5, 2, 4, 6, 8, 12 and 24 hr after administration of the loading dose. To determine ADP-induced platelet aggregation, blood samples were collected at 0 (pre-dose), 2, 4, 8, 12, 24, 48, 72, 96, 120, 144, 146, 148, 152, 156, 216 and 288 hr after administration of the loading dose. For the PK analysis, 7 mL of blood was collected into tubes containing sodium heparin (Vacutainer; BD BioSciences, Franklin Lakes, NJ, USA), and for the PD analysis, 3 mL of blood was collected into tubes containing 0.109 mM sodium citrate (3.8%; Vacutainer; BD, Belliver Industrial Estate, Plymouth, UK). The tubes with blood samples for PK analysis were subjected to centrifugation (1208 9 g, 10 min.) and stored at 80°C. The plasma concentrations of cilostazol [5] and triflusal + HTB (the main metabolite of triflusal) [6] after a single oral dose were analysed using validated LC/UV methods and of clopidogrel and the carboxylic acid and thiol metabolites of clopidogrel (the two major metabolites of clopidogrel) [7], triflusal + HTB [8] after multiple oral doses were analysed simultaneously using a validated LC/MS/MS method. Platelet aggregation test. To obtained platelet-rich plasma (PRP), a citrated tube of blood was mixed gently by inverting three to five times and then centrifuged (10 min., 160 9 g). PRP was removed and the remaining specimen was centrifuged again (10 min., 2000 9 g) to obtain platelet-poor plasma (PPP). In total, 500 lL of PRP was collected into microtubes, which were sealed with caps to prevent pH changes caused by direct exposure to air and then put on ice. Platelet aggregation was assessed using a photometric method in Chrono-log aggregometer cuvettes containing 250 lL of PRP or PPP maintained at 37°C for 5 min. After the addition of 1 lM arachidonic acid or 10 lM ADP to stimulate platelet aggregation in PRP, aggregation was monitored for 5 min. Platelet aggregation results are expressed as the maximum percentage change in optical density, with PPP used as a reference. Analyses of blank plasma samples from three individuals were performed. Intraday precision was assessed by replicate analysis (n = 2 for four wells) on three different days. Precision was calculated as (S.D./ mean) 9 100 (%). The intra- and interday precision ranged from 2.1 to 6.8% and from 2.1 to 7.6%, respectively. The inhibition of platelet aggregation (IPA,%) was calculated using the following formula:

IPAt ð%inhibitionÞ ¼

MPA0  MPAt  100 MPA0

ð1Þ

where MPA is the maximum platelet aggregation at each scheduled time-point, MPA0 is the MPA at baseline (predose), MPAt is the MPA at time t, and IPAt is the IPA at time t (8). Pharmacokinetic model. Pharmacokinetic analyses were performed using non-compartmental and compartmental methods. The area under the plasma concentration-versus-time curve (AUC) was calculated using the trapezoidal rule and extrapolated to infinity. The time course of the plasma concentrations was used to determine the maximum plasma concentration (Cmax) and the time (Tmax) to reach Cmax. The elimination rate constant (kel) was obtained by linear regression of the terminal phase, and the calculated elimination half-life (t½) was 0.693/kel. We also used a modified parent–metabolite compartmental model based on a two–compartment model. Models were constructed as a series of differential equations that were solved numerically and were fitted to the data using ADAPT V (Biomedical Simulation Resource, Los Angeles, CA, USA). Fitting was performed using maximum likelihood estimation under the assumption that the standard deviation of the measurement error was a linear function of the quantity measured. The following information (obtained with ADAPT V) was used to

evaluate the goodness of fit and the quality of the parameter estimates: coefficients of variation in parameter estimates, parameter correlation matrix, sums of squares of residuals, visual examination of the distribution of residuals and the Akaike information criterion. We used a coefficient of variation of

Semi-mechanistic modelling and simulation of inhibition of platelet aggregation by antiplatelet agents.

Antiplatelet agents are a class of pharmaceuticals that decrease platelet aggregation and thus inhibit thrombus formation. We examined the relationshi...
369KB Sizes 5 Downloads 3 Views