Impact of Cilostazol Pharmacokinetics on the Development of Cardiovascular Side Effects in Patients with Cerebral Infarction

Abstract

This study investigated the impact of polymorphisms of metabolic enzymes on plasma concentrations of cilostazol and its metabolites, and the influence of the plasma concentrations and polymorphisms on the cardiovascular side effects in 30 patients with cerebral infarction. Plasma concentrations of cilostazol and its active metabolites, and CYP3A5*3 and CYP2C19*2 and *3 genotypes were determined. The median plasma concentration/dose ratio of OPC-13213, an active metabolite by CYP3A5 and CYP2C19, was slightly higher and the median plasma concentration rate of cilostazol to OPC-13015, another active metabolite by CYP3A4, was significantly lower in CYP3A5*1 carriers than in *1 non-carriers (p = 0.082 and p = 0.002, respectively). The CYP2C19 genotype did not affect the pharmacokinetics of cilostazol. A correlation was observed between changes in pulse rate from the baseline and plasma concentrations of cilostazol (R = 0.539, p = 0.002), OPC-13015 (R = 0.396, p = 0.030) and OPC-13213 (R = 0.383, p = 0.037). A multiple regression model, consisting of factors of the plasma concentration of OPC-13015, levels of blood urea nitrogen, and pulse rate at the start of the therapy explained 55.5% of the interindividual variability of the changes in pulse rate. These results suggest that plasma concentrations of cilostazol and its metabolites are affected by CYP3A5 genotypes, and plasma concentration of OPC-13015, blood urea nitrogen, and pulse rate at the start of therapy may be predictive markers of cardiovascular side effects of cilostazol in patients with cerebral infarction.

INTRODUCTION

Cilostazol is an oral phosphodiesterase (PDE) type-3 inhibitor with antiplatelet and vasodilating properties,^1,2) and is used for the treatment of cerebral infarction, transient ischemic attack, and intermittent claudication.^3,4) Recent meta-analyses demonstrated that cilostazol is effective for the secondary prevention of ischemic stroke.⁵⁾ Headache, tachycardia, and palpitation are commonly reported as side effects due to the vasodilatory and inotropic properties of cilostazol.⁶⁾ Increases in pulse rate and palpitation were frequently observed in patients receiving cilostazol for the secondary prevention of cerebral infarction, and persisted in some patients until discontinuation of the treatment.⁷⁾ Since increased pulse rate and palpitation occasionally cause angina pectoris, these symptoms are side effects warranting close attention.

The pharmacokinetics of cilostazol show large interindividual variation among patients with intermittent claudication.⁸⁾ Cilostazol is metabolized to dehydrocilostazol (OPC-13015) through a quinone hydroxylated intermediate metabolite, OPC-13326, mainly by CYP3A4, and to monohydroxycilostazol (OPC-13213) through a hexane hydroxylated intermediate metabolite, OPC-13217, by CYP2C19 and CYP3A5.^9,10) OPC-13015 and OPC-13213 have three times and one-third more potent inhibitory activity against platelet aggregation than cilostazol, respectively,¹¹⁾ whereas pharmacological potencies of OPC-13326 and OPC-13217 remain unclear. Previous studies in healthy subjects demonstrated that genetic polymorphisms in CYP2C19 and CYP3A5 significantly affect the pharmacokinetics of cilostazol and its metabolites.^12–14) There has, however, been no report on the influence of genetic polymorphisms on the blood disposition of cilostazol in patients with cerebral infarction. Cilostazol binds extensively to plasma albumin in the bloodstream.¹⁵⁾ Although a high frequency of hypoalbuminemia was reported in acute stroke patients,¹⁶⁾ no study has examined the differences in the blood cilostazol disposition between healthy subjects and patients with cerebral infarction.

The antiplatelet effect of cilostazol was reported to be affected by its plasma concentration.^17,18) In addition, recent studies demonstrated significant correlations between platelet responses to cilostazol and CYP2C19 and CYP3A5 genotypes.^19,20) In contrast, the relationships between the cardiovascular pharmacology and pharmacokinetics of cilostazol and metabolites have not been evaluated in patients with cerebral infarction.

In the present study, we investigated the influence of CYP2C19 and CYP3A5 genotypes on the plasma disposition of cilostazol, including metabolites, and the impact of the plasma disposition of cilostazol and the two metabolites on changes in pulse rate and blood pressure in patients with cerebral infarction.

MATERIALS AND METHODS

Ethics

This study was performed in accordance with the Declaration of Helsinki and its amendments, and the protocol was approved by the Ethics Committee of Shimada General Medical Center and University of Shizuoka. Patients received information about the scientific aim of the study, and provided written informed consent.

Study Design

This observational study enrolled 30 inpatients with cerebral infarction at Shimada General Medical Center. They received 50 or 100 mg oral cilostazol (Pretaal^® OD tablet, Otsuka Pharmaceutical Co., Ltd., Tokushima, Japan) twice daily for prophylaxis against recurrence of cerebral infarction. Exclusion criteria were as follows: patients who (1) were being co-treated with other PDE inhibitors; or (2) had severe renal impairment (estimated glomerular filtration rate <30 mL/min/1.73 m²) or total bilirubin concentration >2 mg/dL. Blood sampling was performed after achieving a steady state for four days or more after the start of cilostazol treatment. Blood specimens were collected in tubes containing ethylenediaminetetraacetic acid (EDTA) dipotassium salt 12 h after the evening dose. Hypoalbuminemia was defined as a level below 3.5 g/dL in serum.

Determination of Plasma Cilostazol and Its Metabolites

Cilostazol, OPC-13015, OPC-13213, and cilostazol-d11 as an internal standard were purchased from Toronto Research Chemicals (Toronto, ON, Canada). The plasma fraction was obtained by centrifugation of whole blood at 1670 × g at 4 °C for 10 min. Plasma concentrations of cilostazol and its metabolites were determined by liquid chromatography-tandem mass spectrometry (LC-MS/MS) based on the modified method of Bramer et al.²¹⁾ Plasma samples were extracted using a solid extraction device (Oasis^® HLB, 1 cc, 30 mg; Waters, Milford, MA, U.S.A.) and evaporated using a centrifugal vaporizer (CVE-200D, Eyela, Tokyo, Japan). The residue was reconstituted with 100 µL of the LC-MS/MS eluent, and a ten µL-aliquot was subjected to LC-MS/MS. Chromatography was carried out on a Mightysil RP-18 MS column (2.0 × 150 mm; Kanto Chemical Co., Inc., Tokyo, Japan) at a flow rate of 200 µL/min. Mobile phase A consisted of acetonitrile/2 mM ammonium formate-0.1% formic acid (10 : 90 (v/v)), and phase B consisted of acetonitrile/2 mM ammonium formate-0.1% formic acid (90 : 10 (v/v)). Mass spectra were obtained using an API3200 Qtrap (SCIEX, Framingham, MA, U.S.A.). The mass transition pairs (m/z) were 370.2→288.3 for cilostazol, 368.2→286.3 for OPC-13015, 386.2→288.2 for OPC-13213, and 381.2→289.3 for the internal standard. The calibration curves in human plasma were linear over concentration ranges of 5–1000 ng/mL for cilostazol (R > 0.998), OPC-13015 (R > 0.999), and OPC-13213 (R > 0.998). The intra- and inter-day mean accuracy of cilostazol, OPC-13015, and OPC-13213 were 98.4–110.4, 92.8–105.7, and 99.7–104.6%, respectively. The intra- and inter-precisions (CV %) of cilostazol, OPC-13015, and OPC-13213 were 0.3–16.1, 0.5–12.6, and 0.3–12.7%, respectively.

CYP Genotyping

Genomic DNA was extracted from the whole peripheral blood of each patient using a SMITEST EX-R&D (Medical & Biological Laboratories, Aichi, Japan). CYP2C19*2 and CYP2C19*3 were determined using PCR restriction fragment length polymorphism (RFLP) procedures. Detection of CYP2C19*2 (rs4244285) and CYP2C19*3 (rs4986893) alleles was performed as described previously, with some modifications.^22,23) CYP3A5*3 (rs776746) alleles were determined with quantitative real-time PCR using a TaqMan^® genotyping assay (Applied Biosystems, CA, U.S.A.).²⁴⁾ With respect to CYP2C19, the patients were divided into two groups based on the genetic variants (*1, *2, and *3): *1 carrier (*1/*1, *1/*2, and *1/*3) and *1 non-carrier (*2/*2, *2/*3, and *3/*3). For CYP3A5, the patients were divided into two groups based on the genetic variants (*1 and *3): *1 carrier (*1/*1 and *1/*3) and *3/*3.

Evaluation of Cardiovascular Side Effects

Blood pressure and pulse rate were measured before the start of cilostazol therapy and three hours after administration on days 4, 5, and 6. Changes in values were compared between the baseline (day 0) and the average of three days (days 4–6).

Statistical Analysis

All statistical analyses were performed using EZR (Saitama Medical Center, Jichi Medical University, Saitama, Japan), a graphical user interface for R (The R Foundation for Statistical Computing, Vienna, Austria), and IBM SPSS Statistics Ver. 23 (IBM, New York, U.S.A.). Plasma concentrations of cilostazol and its metabolites were evaluated as absolute values and the concentration/dose (C/D) ratio. Cilostazol metabolism was estimated by the predose plasma concentration rate of the metabolites to cilostazol as a metabolic rate. The total potency-adjusted pharmacologically active moieties were calculated as the sum of the plasma concentration of cilostazol: [OPC-13015 × 3] and [OPC-13213 × 1/3]. The influence of CYP2C19 and CYP3A5 genotypes on the plasma concentrations of cilostazol and its metabolites was analyzed using the Mann–Whitney U test. Changes in blood pressure and pulse rate from the baseline were compared by the Wilcoxon signed-rank test. The correlations between plasma concentrations of cilostazol, OPC13015, and OPC 13213, and between changes in blood pressure and pulse rate and the plasma concentrations of cilostazol and its metabolites were tested by Spearman's rank correlation coefficient analysis. Correlations between changes in blood pressure and pulse rate and the plasma concentrations of cilostazol and its metabolites were determined using sigmoid curve regression in addition to linear regression. Correlation coefficients of sigmoid curves were calculated after converting all the data to positive numbers. The influence of CYP2C19 and CYP3A5 genotypes on cardiovascular side effects was analyzed using the Mann–Whitney U test. Multiple regression analysis was performed using the following possible factors: age, sex, height, body weight, CYP2C19 and CYP3A5 genotypes, concomitant use of antihypertensive medication, total bilirubin, aspartate aminotransferase, alanine aminotransferase, blood urea nitrogen, serum creatinine, serum albumin, total protein, plasma concentrations of cilostazol and its metabolites, total potency-adjusted pharmacologically active moieties, and blood pressure and pulse rate before the treatment with cilostazol. To extract significant factors, a stepwise forward selection method was used with a criterion of p < 0.1 to enter and p > 0.2 to remove. Subsequently, we estimated the final model by considering the relationship and independence of the extracted factors. Values for the intercept and coefficient were determined for each factor. All values are expressed as the median and the interquartile range (IQR). p < 0.05 was considered significant.

RESULTS

Study Populations

Table 1 shows the demographic characteristics of patients in this study. The median serum albumin level was 4.1 g/dL, and one patient had hypoalbuminemia. The allele frequencies of the observed genetic variants of CYP2C19 and CYP3A5 were as follows: CYP2C19*2 (36.7%), CYP2C19*3 (18.3%), and CYP3A5*3 (75.0%) (Supplementary Table 1). The proportions of observed CYP genotype groups were as follows: 21 (70.0%) and 9 (30.0%) patients with CYP2C19*1 and *1 non-carriers, and 13 (43.3%) and 17 (56.7%) patients with CYP3A5*1 and *3/*3 carriers, respectively. All genotype frequencies were found to be consistent with the Hardy–Weinberg equilibrium.

Table 1. Demographics of Patients at the Start of the Study

Characteristics	Number or median (interquartile range or %)
Number of patients, male/female	30, 22/8
Age (years)	71 (66–80)
Body weight (kg)	60 (49–67)
BMI (kg/m2)	23.3 (20.0–24.9)
Total protein (g/dL)	6.9 (6.7–7.3)
Serum albumin (g/dL)	4.1 (3.9–4.3)
Serum creatinine (mg/dL)	0.71 (0.63–0.85)
Blood urea nitrogen (mg/dL)	14.0 (12.4–17.4)
Total bilirubin (mg/dL)	0.76 (0.52–0.96)
Aspartate aminotransferase (IU/L)	21 (19–24)
Alanine aminotransferase (IU/L)	17 (12–23)
SBP (mmHg)	144 (134–164)
DBP (mmHg)	81 (72–90)
Pulse rate (bpm)	75 (66–80)
Concomitant medication, n (%)
Aspirin	6 (20.0)
Clopidogrel	2 (6.7)
Statin	15 (50.0)
PPIs	25 (83.3)
Lansoprazole	10
Esomeprazole	11
Omeprazole	1
Vonoprazan	3

Dates are expressed as the median with interquartile range in parentheses. Abbreviations: BMI, Body mass index; SBP, Systolic blood pressure; DBP, Diastolic blood pressure; PPIs, Proton pump inhibitors.

Influence of CYP Genotypes on Cilostazol Pharmacokinetics

Figure 1 shows that the plasma concentration of cilostazol was directly and significantly correlated with those of the metabolites, OPC-13015 and OPC-13213 (p < 0.001). The median molecular concentration rates of OPC-13015 and OPC-13213 to cilostazol were 37.8 and 13.9%, respectively (Table 2).

Fig. 1. Relationships between Plasma Concentrations of Cilostazol and Dehydrocilostazol (OPC-13015) (a), or Monohydroxycilostazol (OPC-13213) (b) in Patients with Cerebral Infarction

Correlations between plasma concentrations of cilostazol, OPC13015, and OPC 13213 were tested by Spearman's rank correlation coefficient analysis.

Table 2. Influence of CYP2C19 and CYP3A5 Genotypes on Plasma Concentrations of Cilostazol and Its Metabolites in Patients with Cerebral Infarction

	All	CYP2C19 genotype			CYP3A5 genotype
		*1 carrier	*1 non-carrier	P	*1 carrier	*1 non-carrier	P
Number of patients	30	21	9		13	17
Plasma cilostazol concentration (µmol/L)	2.24 (1.49–2.82)	2.26 (1.88–3.10)	1.67 (1.19–2.59)	0.258	2.58 (2.13–3.10)	1.91 (1.43–2.59)	0.346
Plasma OPC-13015 concentration (µmol/L)	0.795 (0.615–1.06)	0.900 (0.630–1.19)	0.700 (0.571–0.920)	0.326	0.761 (0.672–0.901)	0.992 (0.603–1.24)	0.341
Plasma OPC-13213 concentration (µmol/L)	0.294 (0.221–0.377)	0.296 (0.231–0.384)	0.242 (0.213–0.345)	0.512	0.296 (0.275–0.356)	0.242 (0.213–0.384)	0.391
Cilostazol C/D ratio (µmol/L per mg/kg)	0.711 (0.497–0.984)	0.738 (0.558–0.991)	0.491 (0.418–0.736)	0.209	0.736 (0.574–1.12)	0.643 (0.450–0.801)	0.133
OPC-13015 C/D ratio (µmol/L per mg/kg)	0.270 (0.194–0.342)	0.280 (0.190–0.375)	0.266 (0.218–0.277)	0.603	0.266 (0.190–0.280)	0.283 (0.195–0.367)	0.403
OPC-13213 C/D ratio (µmol/L per mg/kg)	0.091 (0.075–0.120)	0.092 (0.076–0.118)	0.089 (0.074–0.128)	0.874	0.104 (0.087–0.128)	0.081 (0.056–0.102)	0.0824
OPC-13015/cilostazol concentration ratio	0.378 (0.317–0.511)	0.343 (0.317–0.459)	0.515 (0.372–0.612)	0.141	0.318 (0.289–0.343)	0.482 (0.383–0.541)	0.00157*
OPC-13213/cilostazol concentration ratio	0.139 (0.096–0.167)	0.138 (0.100–0.163)	0.166 (0.093–0.186)	0.469	0.157 (0.095–0.167)	0.138 (0.100–0.166)	0.770

Plasma concentrations of cilostazol and its metabolites were evaluated as the predose plasma concentration adjusted with the cilostazol dose per body weight (C/D ratio). Data are expressed as the median with interquartile range in parentheses. The influences of CYP phenotypes on plasma concentrations of cilostazol and its metabolites were tested using the Mann–Whitney U test (* p < 0.05). CYP2C19*1 carrier and non-carrier mean carriers having at least one of CYP2C19*1 allele (*1/*1, *1/*2, and *1/*3) and the others (*2/*2, *2/*3, and *3/*3), respectively. CYP3A5*1 carrier and non-carrier mean carriers having CYP3A5*1/*1 or *1/*3, and CYP3A5*3/*3, respectively. Abbreviations: OPC-13015, dehydrocilostazol; OPC-13213, monohydroxycilostazol.

The influences of CYP2C19 and CYP3A5 genotypes on plasma cilostazol and its metabolite dispositions are shown in Table 2. No significant differences were observed in the plasma cilostazol and its metabolite C/D ratio or the metabolic rate between CYP2C19*1 carriers and non-carriers. The median plasma OPC-13213 C/D ratio was slightly lower in CYP3A5*1 non-carriers than in *1 carriers (p = 0.082). The median plasma concentration rate of OPC-13015 to cilostazol was 48.2%, and was significantly higher in CYP3A5*1 non-carriers than in *1 carriers (p = 0.002). Lansoprazole, esomeprazole, omeprazole, and clopidogrel, which can inhibit CYP2C19 activity, were concomitantly administered to 23 of 30 patients. We investigated the influence of CYP2C19 and CYP3A5 genotypes on plasma concentrations of cilostazol and the metabolites in the 23 patients (Table 3). While there was no significant influence of CYP2C19 genotypes on plasma concentrations of cilostazol and the metabolites, the plasma concentration rate of OPC-13015 to cilostazol was significantly higher in *1 non-carriers of CYP3A5 than in *1 carriers (p = 0.011). Plasma cilostazol and the OPC-13213 C/D ratio decreased slightly in *1 non-carriers of CYP3A5 compared with *1 carriers (p = 0.060 and 0.060, respectively).

Table 3. Influence of CYP2C19 and CYP3A5 Genotypes on Plasma Concentrations of Cilostazol and Its Metabolites in Patients with Concomitant Use of Lansoprazole, Esomeprazole, Omeprazole, and Clopidogrel

	All	CYP2C19 genotype			CYP3A5 genotype
		*1 carrier	*1 non-carrier	P	*1 carrier	*1 non-carrier	P
Number of patients	23	15	8		12	11
Plasma cilostazol concentration (µmol/L)	2.26 (1.36–2.85)	2.26 (1.57–3.12)	2.03 (1.27–2.59)	0.583	2.58 (1.92–3.11)	1.70 (1.34–2.43)	0.268
Plasma OPC-13015 concentration (µmol/L)	0.749 (0.606–0.987)	0.774 (0.606–0.987)	0.725 (0.613–0.943)	0.925	0.762 (0.655–0.906)	0.657 (0.541–1.01)	1.00
Plasma OPC-13213 concentration (µmol/L)	0.281 (0.211–0.370)	0.296 (0.212–0.370)	0.259 (0.212–0.366)	0.897	0.312 (0.267–0.368)	0.229 (0.174–0.349)	0.230
Cilostazol C/D ratio (µmol/L per mg/kg)	0.643 (0.471–1.02)	0.643 (0.520–1.02)	0.590 (0.419–0.850)	0.548	0.849 (0.605–1.13)	0.491 (0.435–0.722)	0.0595
OPC-13015 C/D ratio (µmol/L per mg/kg)	0.256 (0.187–0.295)	0.235 (0.176–0.295)	0.270 (0.247–0.286)	0.651	0.270 (0.188–0.296)	0.256 (0.181–0.295)	0.735
OPC-13213 C/D ratio (µmol/L per mg/kg)	0.092 (0.072–0.123)	0.092 (0.072–0.111)	0.100 (0.074–0.128)	0.975	0.108 (0.089–0.128)	0.064 (0.054–0.100)	0.0595
OPC-13015/cilostazol concentration ratio	0.372 (0.296–0.459)	0.334 (0.296–0.412)	0.453 (0.351–0.619)	0.138	0.309 (0.288–0.347)	0.458 (0.378–0.566)	0.0106
OPC-13213/cilostazol concentration ratio	0.142 (0.098–0.167)	0.142 (0.102–0.164)	0.147 (0.092–0.178)	0.771	0.160 (0.095–0.169)	0.127 (0.111–0.155)	0.689

Plasma concentrations of cilostazol and its metabolites were evaluated as the predose plasma concentration adjusted with the cilostazol dose per body weight (C/D ratio). Data are expressed as the median with interquartile range in parentheses. The influences of CYP phenotypes on plasma concentrations of cilostazol and its metabolites were tested using the Mann–Whitney U test (* p < 0.05). CYP2C19*1 carrier and non-carrier mean carriers having at least one of CYP2C19*1 allele (*1/*1, *1/*2, and *1/*3) and the others (*2/*2, *2/*3, and *3/*3), respectively. CYP3A5*1 carrier and non-carrier mean carriers having CYP3A5*1/*1 or *1/*3, and CYP3A5*3/*3, respectively. Abbreviations: OPC-13015, dehydrocilostazol; OPC-13213, monohydroxycilostazol.

Relationships between Cardiovascular Side Effects and Plasma Concentrations of Cilostazol and Its Metabolites

Table 4 shows that the median values of systolic blood pressure (SBP) and diastolic blood pressure (DBP) significantly decreased after 4 to 6 d of treatment with cilostazol (−11 and −7 mmHg, respectively, p < 0.01) in the whole population. A significant increase in pulse rate was observed (median 12 bpm, p < 0.01), and no patient developed hypotension (SBP <100 mmHg, DBP <60 mmHg) in the whole population. Tachycardia (pulse rate > 100 bpm) was seen in two of 30 patients. Figure 2 shows that the changes in pulse rate from the baseline were significantly correlated with plasma concentrations of cilostazol (R = 0.539, p = 0.002), OPC-13015 (R = 0.396, p = 0.030), OPC-13213 (R = 0.383, p = 0.037), and the amount of total potency-adjusted pharmacologically active moieties (R = 0.512, p = 0.004). In contrast, there were no significant correlations between SBP or DBP and plasma concentrations of cilostazol, its metabolites, or the amount of total potency-adjusted pharmacologically active moieties (Supplementary Figs. 1, 2). Although we also examined the association between changes in blood pressure and pulse rate and the plasma concentrations of cilostazol and its metabolites using regression analysis with sigmoid curves, we found no correlations between these factors except between pulse rate and plasma OPC-13015 concentration (R = 0.737, p < 0.001; Supplementary Table 2).

Table 4. Influence of CYP2C19 and CYP3A5 Genotypes on Cardiovascular Effects in Patients with Cerebral Infarction

	All		CYP2C19 genotype			CYP3A5 genotype
		P	*1 carrier	*1 non-carrier	P	*1 carrier	*1 non-carrier	P
Number of patients	30	—	21	9		13	17
SBP at the baseline (mmHg)	144 (134–164)	—	139 (128–164)	150 (140–163)	0.287	156 (137–164)	139 (133–159)	0.429
Changes in SBP (mmHg)	−11 (−19–1)	0.006	−11 (−15–1)	−7 (−19–0)	0.803	−11 (−29–0)	−7 (−15–2)	0.414
DBP at the baseline (mmHg)	81 (72–90)	—	80 (72–87)	88 (72–94)	0.342	84 (72–94)	80 (72–87)	0.503
Changes in DBP (mmHg)	−7 (−13–3)	0.004	−7 (−13–2)	−9 (−10–4)	0.982	−10 (−13–0)	−6 (−13–4)	0.414
Pulse rate at the baseline (bpm)	75 (66–80)	—	74 (66–78)	77 (68–83)	0.330	74 (65–77)	75 (68–80)	0.285
Changes in pulse rate (bpm)	12 (7–18)	< 0.001	14 (8–18)	10 (5–15)	0.188	14 (6–17)	12 (8–18)	0.883

Data are expressed as the median with interquartile range in parentheses. Changes in SBP, DBP, and pulse rate from the baseline values were compared by the Wilcoxon signed-rank test. The influences of CYP phenotypes on the cardiovascular effect were tested using the Mann–Whitney U test. CYP2C19*1 carrier and non-carrier mean carriers having at least one of CYP2C19*1 allele (*1/*1, *1/*2, and *1/*3) and the others (*2/*2, *2/*3, and *3/*3), respectively. CYP3A5*1 carrier and non-carrier mean carriers having CYP3A5*1/*1 or *1/*3, and CYP3A5*3/*3, respectively. Abbreviations: SBP, Systolic blood pressure; DBP, Diastolic blood pressure.

Fig. 2. Relationships between Changes in Pulse Rate and Plasma Concentrations of Cilostazol (a), Dehydrocilostazol (OPC-13015) (b), Monohydroxycilostazol (OPC-13213) (c), and the Amount of Total Potency-Adjusted Pharmacologically Active Moieties (d)

* The amount of total potency-adjusted pharmacologically active moieties was calculated as the sum of the plasma molar concentration of [cilostazol + OPC-13015 ×3 + OPC-13213 ×1/3]. Correlations between changes in pulse rate and the plasma concentrations of cilostazol and its metabolites were tested by Spearman's rank correlation coefficient analysis.

The influence of CYP2C19 and CYP3A5 genotypes on the cardiovascular side effects of cilostazol is shown in Table 4. No significant differences were observed in changes of SBP, DBP, or pulse rate from the baseline between the CYP3A5 and CYP2C19 genotype groups.

Table 5 shows that a multiple regression model, consisting of factors for plasma concentration of OPC-13015 (standardized partial regression coefficient; β = 0.454, p = 0.0026), levels of blood urea nitrogen (BUN) (β = 0.415, p = 0.0056) and pulse rate at the start of the therapy (β = − 0.480, p = 0.0017), explained 55.5% of the interindividual variability of the changes in pulse rate.

Table 5. Factors Influencing the Changes in Pulse Rate in Multiple Regression Analysis

Dependent variable	Factor	Unstandardized coefficients		Standardized coefficients β	t	P	VIF
		B (95% CI)	SE
Changes in pulse rate	Intercept	21.0 (1.38–40.7)	9.57		2.20	0.0369
	Pulse on Day 0	−0.413 (−0.656– −0.171)	0.118	−0.480	−3.50	0.00168	1.02
	BUN	0.847 (0.270–1.424)	0.281	0.415	3.02	0.00562	1.03
	Plasma concentration of OPC-13015	9.81 (3.76–15.9)	2.94	0.454	3.33	0.00260	1.01

The adjusted coefficient of determination (R²) was 0.555. Changes in pulse = 21.0 − 0.413 × [Pulse on Day 0] + 0.847 × [BUN] + 9.81 × [Plasma concentration of OPC-13015]. Abbreviations: B, partial regression coefficient; β, standardized partial regression coefficient; VIF, variance inflation factor; SE, standard error; CI, confidence interval.

DISCUSSION

To the best of our knowledge, this is the first study to investigate the direct relationships between plasma concentrations of cilostazol and the metabolites OPC-13015 and OPC-13213. The correlation coefficients between plasma cilostazol, OPC-13015, and OPC-13213 concentrations were relatively high (R > 0.6) in patients with cerebral infarction. These results indicate that the influence of polymorphisms in metabolic enzymes on the pharmacokinetics of cilostazol is relatively weak in this study population.

Although the enzymatic function of CYP3A5 sometimes overlaps with CYP3A4 in regard to substrate specificity, CYP3A4 and CYP3A5 convert cilostazol into different metabolites, OPC-13326 and OPC-13217, that are then metabolized to OPC-13015 and OPC-13213, respectively (Supplementary Fig. 3). The CYP3A5*3 genotype is frequently seen in both Japanese and Caucasian populations.²⁵⁾ Homozygous carriers of CYP3A5*3 genes lack functional activity of CYP3A5. This study demonstrated that CYP3A5*3 homozygous carriers have slightly lower plasma OPC-13213 C/D ratios and significantly higher plasma concentration rates of OPC13015 to cilostazol than *1 carriers, among patients with cerebral infarction. These results indicate that CYP3A5 plays an essential role in the metabolism of cilostazol, and the reduced enzymatic activity causes the elevated alternative pathway that converts cilostazol to OPC-13015 by CYP3A4 in patients enrolled in this study. Our results are not consistent with the previous study by Lee et al.,¹⁴⁾ in which CYP2C19 but not CYP3A5 genotypes affected the pharmacokinetics of cilostazol metabolites after a single administration in healthy volunteers. We speculate that the lack of a gene effect of CYP2C19 polymorphism on cilostazol pharmacokinetics was caused by coadministration with CYP2C19 inhibitors. In this study population, seven (33.3%), seven (33.3%), and two (9.5%) out of 21 CYP2C19*1 carrier patients (one patient took two CYP2C19 inhibitors) were co-administered lansoprazole, esomeprazole, or clopidogrel, which were reported to inhibit CYP2C19-mediated metabolism,^26,27) respectively. Table 3 shows no significant influence of CYP2C19 genotypes on plasma concentrations of cilostazol and the metabolites in patients concomitantly receiving CYP2C19 inhibitors, being consistent with the results in the total patients (Table 2). On the other hand, the cilostazol C/D ratio (p = 0.060) and plasma OPC-13213 C/D ratio were slightly lower (p = 0.060), and the median plasma concentration rate of OPC-13015 to cilostazol significantly higher in the CYP3A5*1 non-carriers than those in the *1 carriers of patients receiving CYP2C19 inhibitors. Kim et al. reported that no differences were observed in the area under the concentration–time curve of cilostazol and OPC-13015 in healthy subjects after the coadministration of clopidogrel between CYP2C19 genotype groups, while CYP3A5 genotypes affected the parameters.¹⁹⁾ Under the condition of the concomitant use of CYP2C19 inhibitors, CYP3A5 may play a more critical role in the metabolism of cilostazol, leading to a more conspicuous influence of CYP3A5 polymorphisms on cilostazol pharmacokinetics. Another reason why CYP3A5 polymorphisms affected cilostazol metabolism more strongly than CYP2C19 may be the higher enzymatic affinity of CYP3A5 compared with that of CYP2C19 in the metabolism from cilostazol to OPC-13217—an intermittent metabolite before OPC-13213 based on an in vitro study.¹⁰⁾ Cilostazol extensively binds to plasma albumin in peripheral blood,¹⁵⁾ whereas plasma protein bindings of OPC-13015 and OPC-13213 are unknown. A previous study reported that about half of patients with acute ischemic stroke developed hypoalbuminemia with a serum albumin level below 3.5 g/dL.¹⁶⁾ However, all patients except one in this study had normal albumin levels. Pharmacokinetic parameters of cilostazol in the patient with hypoalbuminemia were similar to those in patients with normal albumin (data not shown). Therefore, it was unlikely that the difference in protein binding resulted in different pharmacokinetics of cilostazol between patients with cerebral infarction and healthy subjects.

Woo et al. reported a significant correlation between plasma concentrations of cilostazol and changes in the heart rate from the baseline after a single oral dose of 100 mg cilostazol in healthy subjects.¹⁷⁾ However, they did not determine the effects of cilostazol metabolites. Although platelet aggregation inhibitory effects of cilostazol metabolites have been reported,¹¹⁾ the relationships between cardiovascular side effects and blood concentrations of cilostazol metabolites have not been investigated. This is the first study to examine the direct relationship between plasma concentrations of cilostazol metabolites and the cardiovascular side effects in patients. Angina pectoris rarely develops but is one of the serious side effects of cilostazol, caused by an increased pressure rate product accompanying an increase in pulse rate due to cilostazol. We examined changes in blood pressure and pulse rate 4 to 6 d after the start of cilostazol therapy because the apparent elimination half-life of cilostazol is approximately 11 h and plasma disposition of cilostazol reaches a steady state within 4 d.⁸⁾ In this study, significant correlations between changes in pulse rate from the baseline and plasma cilostazol, OPC-13015, and OPC-13213 concentrations, and the amount of total potency-adjusted pharmacologically active moieties were observed three hours after the administration of cilostazol. We performed multiple regression analysis in order to determine the best combination of parameters for the changes in pulse rate. The analysis revealed that the plasma concentration of OPC-13015, pulse rate at the start of cilostazol, and BUN values explained 55% of the interindividual variability of the changes in pulse rate. OPC-13015 has three times more potent antiplatelet aggregation activity than cilostazol,¹¹⁾ and plasma OPC-13015 concentration reached 37.8% of cilostazol and correlated with plasma cilostazol concentration in this study. In addition, sigmoid curve regression analysis showed that changes in pulse rate well correlated with plasma OPC-13015 concentration (R = 0.737). These results indicated that OPC-13015 plays an important role in cardiovascular effects after administration of cilostazol. We speculated therefore that OPC-13015 was one of the factors that determines changes in pulse rate. The issues mentioned above indicated that we should consider plasma OPC-13015 concentrations when evaluating the cardiovascular side effects of cilostazol in patients with cerebral infarction. Changes in pulse rate were correlated with BUN and pulse rate at the start of cilostazol treatment. BUN levels are reportedly a prognostic factor in patients with heart failure.^28,29) and a predictive indicator of the response to diuretics in acute heart failure patients.³⁰⁾ Toyonaga et al. reported that cardiovascular side effects of cilostazol were affected by the baseline heart rate.³¹⁾ Therefore, we suggest that BUN and the baseline pulse rate may be predictive factors determining the cardiovascular effect of cilostazol.

No significant difference was observed in changes in pulse rate from the baseline between the CYP2C19 and CYP3A5 genotype groups. We found that CYP3A5*3/*3 carriers had a slightly decreased plasma OPC-13213 C/D ratio and a significantly increased plasma OPC-13015/cilostazol concentration rate. Because OPC-13213 shows three times less potent inhibitory activity against platelet aggregation than cilostazol, and the plasma concentrations were approximately one-tenth those of cilostazol in this study, changes in plasma OPC-13213 concentrations are unlikely to affect the cardiovascular side effects of cilostazol. Multiple regression analysis revealed that the plasma OPC-13015 concentration was a factor determining changes in pulse rate. The plasma OPC-13015 concentration did not differ between *1 carriers and *1 non-carriers of CYP2C19 and CYP3A5 genes. Therefore, in this study population, we consider that CYP2C19 and CYP3A5 polymorphisms hardly affected the cardiovascular side effects of cilostazol.

There were some limitations in the present study. In this study, 70% of patients were concomitantly administered CYP2C19 inhibitors, which may affect the metabolic pathway of cilostazol, as mentioned above. The sample was small and from a single hospital. We did not analyze other CYP3A4-related polymorphisms such as CYP3A4*18 (rs28371759, c.878T > C), which is associated with an increase CYP3A4 activity,³²⁾ even if the frequency is low (1.3%) in the Japanese population.³³⁾

In conclusion, CYP3A5 but not CYP2C19 genotypes significantly affected the pharmacokinetics of cilostazol in patients with cerebral infarction. The plasma concentrations of cilostazol, OPC-13015, and OPC-13213, and the amount of total potency-adjusted pharmacologically active moieties, were correlated with changes in pulse rate from the baseline. Changes in pulse rate from the baseline were explained in 55% by the plasma OPC-13015 concentration, BUN, and pulse rate at the start of cilostazol treatment.

Conflict of Interest

The authors declare no conflict of interest.

Supplementary Materials

The online version of this article contains supplementary materials.

REFERENCES

• 1) Kimura Y, Tani T, Kanbe T, Watanabe K. Effect of cilostazol on platelet aggregation and experimental thrombosis. Arzneimittelforschung, 35 (7A), 1144–1149 (1985).
• 2) Uchikawa T, Murakami T, Furukawa H. Effects of the anti-platelet agent cilostazol on peripheral vascular disease in patients with diabetes mellitus. Arzneimittelforschung, 42, 322–324 (1992).
• 3) Shi L, Pu J, Xu L, Malaguit J, Zhang J, Chen S. The efficacy and safety of cilostazol for the secondary prevention of ischemic stroke in acute and chronic phases in Asian population- an updated meta-analysis. BMC Neurol., 14, 251 (2014).
• 4) Dawson DL, Cutler BS, Meissner MH, Strandness DE Jr. Cilostazol has beneficial effects in treatment of intermittent claudication: results from a multicenter, randomized, prospective, double-blind trial. Circulation, 98, 678–686 (1998).
• 5) McHutchison C, Blair GW, Appleton JP, Chappell FM, Doubal F, Bath PM, Wardlaw JM. Cilostazol for secondary prevention of stroke and cognitive decline: Systematic review and meta-analysis. Stroke, 51, 2374–2385 (2020).
• 6) Pratt CM. Analysis of the cilostazol safety database. Am. J. Cardiol., 87 (12A), 28D–33D (2001).
• 7) Gotoh F, Tohgi H, Hirai S, Terashi A, Fukuuchi Y, Otomo E, Shinohara Y, Itoh E, Matsuda T, Sawada T, Yamaguchi T, Nishimaru K, Ohashi Y. Cilostazol stroke prevention study: a placebo-controlled double-blind trial for secondary prevention of cerebral infarction. J. Stroke Cerebrovasc. Dis., 9, 147–157 (2000).
• 8) Bramer SL, Forbes WP, Mallikaarjun S. Cilostazol pharmacokinetics after single and multiple oral doses in healthy males and patients with intermittent claudication resulting from peripheral arterial disease. Clin. Pharmacokinet., 37 (Suppl. 2), 1–11 (1999).
• 9) Abbas R, Chow CP, Browder NJ, Thacker D, Bramer SL, Fu CJ, Forbes W, Odomi M, Flockhart DA. In vitro metabolism and interaction of cilostazol with human hepatic cytochrome P450 isoforms. Hum. Exp. Toxicol., 19, 178–184 (2000).
• 10) Hiratsuka M, Hinai Y, Sasaki T, Konno Y, Imagawa K, Ishikawa M, Mizugaki M. Characterization of human cytochrome p450 enzymes involved in the metabolism of cilostazol. Drug Metab. Dispos., 35, 1730–1732 (2007).
• 11) Okuda Y, Kimura Y, Yamashita K. Cilostazol. Cardiovasc. Drug Rev., 11, 451–465 (1993).
• 12) Yoo HD, Park SA, Cho HY, Lee YB. Influence of CYP3A and CYP2C19 genetic polymorphisms on the pharmacokinetics of cilostazol in healthy subjects. Clin. Pharmacol. Ther., 86, 281–284 (2009).
• 13) Yoo HD, Cho HY, Lee YB. Population pharmacokinetic analysis of cilostazol in healthy subjects with genetic polymorphisms of CYP3A5, CYP2C19 and ABCB1. Br. J. Clin. Pharmacol., 69, 27–37 (2010).
• 14) Lee HI, Byeon JY, Kim YH, Lee CM, Choi CI, Jang CG, Bae JW, Lee YJ, Lee SY. Effects of CYP2C19 and CYP3A5 genetic polymorphisms on the pharmacokinetics of cilostazol and its active metabolites. Eur. J. Clin. Pharmacol., 74, 1417–1426 (2018).
• 15) Bramer SL, Forbes WP. Effect of hepatic impairment on the pharmacokinetics of a single dose of cilostazol. Clin. Pharmacokinet., 37 (Suppl. 2), 25–32 (1999).
• 16) Dziedzic T, Pera J, Slowik A, Gryz-Kurek EA, Szczudlik A. Hypoalbuminemia in acute ischemic stroke patients: frequency and correlates. Eur. J. Clin. Nutr., 61, 1318–1322 (2007).
• 17) Woo SK, Kang WK, Kwon K. Pharmacokinetic and pharmacodynamic modeling of the antiplatelet and cardiovascular effects of cilostazol in healthy humans. Clin. Pharmacol. Ther., 71, 246–252 (2002).
• 18) Iwamoto T, Kin K, Miyazaki K, Shin K, Takasaki M. Recovery of platelet function after withdrawal of cilostazol administered orally for a long period. J. Atheroscler. Thromb., 10, 348–354 (2003).
• 19) Kim HS, Lim Y, Oh M, Ghim JL, Kim EY, Kim DH, Shin JG. The pharmacokinetic and Pharmacodynamic interaction of clopidogrel and cilostazol in relation to CYP2C19 and CYP 3A5 genotypes. Br. J. Clin. Pharmacol., 81, 301–312 (2016).
• 20) Ikeda Y, Yamanouchi J, Kumon Y, Yasukawa M, Hato T. Association of platelet response to cilostazol with clinical outcome and CYP genotype in patients with cerebral infarction. Thromb. Res., 172, 14–20 (2018).
• 21) Bramer SL, Tata PN, Vengurlekar SS, Brisson JH. Method for the quantitative analysis of cilostazol and its metabolites in human plasma using LC/MS/MS. J. Pharm. Biomed. Anal., 26, 637–650 (2001).
• 22) de Morais SM, Wilkinson GR, Blaisdell J, Nakamura K, Meyer UA, Goldstein JA. The major genetic defect responsible for the polymorphism of S-mephenytoin metabolism in humans. J. Biol. Chem., 269, 15419–15422 (1994).
• 23) Kubota T, Chiba K, Ishizaki T. Genotyping of S-mepheniytoin 4′-hydroxylation in an extended Japanese population. Clin. Pharmacol. Ther., 60, 661–666 (1996).
• 24) Kagawa Y, Yamamoto Y, Ueno A, Maeda T, Obi T. Impact of CYP2D6, CYP3A5, and ABCB1 polymorphisms on plasma concentrations of donepezil and its metabolite in patients with Alzheimer’s Disease. Ther. Drug Monit., 43, 429–435 (2021). doi: 10.1097/FTD.0000000000000823.
• 25) Ishikawa M, Kawai M, Maeda T, Kagawa Y. Prediction of neutrophil reduction using plasma paclitaxel concentration after administration in patients with uterine, ovarian, or cervical cancers in an outpatient clinic. Cancer Chemother. Pharmacol., 81, 399–411 (2018).
• 26) Li XQ, Andersson TB, Ahlström M, Weidolf L. Comparison of inhibitory effects of the proton pump-inhibiting drugs omeprazole, esomeprazole, lansoprazole, pantoprazole, and rabeprazole on human cytochrome P450 activities. Drug Metab. Dispos., 32, 821–827 (2004).
• 27) Nishiya Y, Hagihara K, Kurihara A, Okudaira N, Farid NA, Okazaki O, Ikeda T. Comparison of mechanism-based inhibition of human cytochrome P450 2C19 by ticlopidine, clopidogrel, and prasugrel. Xenobiotica, 39, 836–843 (2009).
• 28) Fonarow GC, Adams KF Jr, Abraham WT, Yancy CW, Boscardin WJ. Risk stratification for in-hospital mortality in acutely decompensated heart failure: classification and regression tree analysis. JAMA, 293, 572–580 (2005).
• 29) Miura M, Sakata Y, Nochioka K, Takahashi J, Takada T, Miyata S, Hiramoto T, Inoue K, Tamaki K, Shiba N, Shimokawa H. Prognostic impact of blood urea nitrogen changes during hospitalization in patients with acute heart failure syndrome. Circ. J., 77, 1221–1228 (2013).
• 30) Valente MA, Voors AA, Damman K, Van Veldhuisen DJ, Massie BM, O’Connor CM, Metra M, Ponikowski P, Teerlink JR, Cotter G, Davison B, Cleland JG, Givertz MM, Bloomfield DM, Fiuzat M, Dittrich HC, Hillege HL. Diuretic response in acute heart failure: clinical characteristics and prognostic significance. Eur. Heart J., 35, 1284–1293 (2014).
• 31) Toyonaga S, Nakatsu T, Murakami T, Kusachi S, Mashima K, Tominaga Y, Yamane S, Uesugi T, Kanai H, Tsuji T. Effects of cilostazol on heart rate and its variation in patients with atrial fibrillation associated with bradycardia. J. Cardiovasc. Pharmacol. Ther., 5, 183–191 (2000).
• 32) Fukushima-Uesaka H, Saito Y, Watanabe H, et al. Haplotypes of CYP3A4 and their close linkage with CYP3A5 haplotypes in a Japanese population. Hum. Mutat., 23, 100 (2004).
• 33) Yamamoto T, Nagafuchi N, Ozeki T, Kubota T, Ishikawa H, Ogawa S, Yamada Y, Hirai H, Iga T. CYP3A4*18: it is not rare allele in Japanese population. Drug Metab. Pharmacokinet., 18, 267–268 (2003).