Association of CYP2C19 Polymorphisms With Clopidogrel Reactivity and Clinical Outcomes in Chronic Ischemic Stroke

Abstract

Background: CYP2C19 variants are associated with the antiplatelet effects of clopidogrel against recurrent cardiovascular events. However, it remains unknown whether the elapsed time from stroke onset affects the relationship between the genetic variants and such events. To address this, we conducted a prospective cohort study to determine the effect of CYP2C19 variants on clinical outcomes in the chronic phase.

Methods and Results: In total, 518 Japanese non-acute stroke patients treated with clopidogrel were registered at 14 institutions. Patients were classified into 3 clopidogrel-metabolizing groups according to CYP2C19 genotype: extensive metabolizer (EM: *1/*1), intermediate metabolizer (IM: *1/*2 or *1/*3), and poor metabolizer (PM: *2/*2, *2/*3, or *3/*3). Antiplatelet effects of clopidogrel were assessed by adenosine diphosphate (ADP)-induced platelet aggregation and vasodilator-stimulated phosphoprotein (VASP) phosphorylation. The endpoint was composite cerebrocardiovascular events (CVEs). In 501 successfully followed-up patients, the median time from index stroke to enrollment was 181 days. There were 28 cardiovascular and 2 major bleeding events. There were no significant differences in the rates of cardiovascular events among the groups.

Conclusions: Despite associations between CYP2C19 variants and on-clopidogrel platelet reactivity, there was no significant difference in rates of CVEs in the chronic stroke phase among the 3 clopidogrel-metabolizing groups of CYP2C19 variants.

Clopidogrel, a platelet P2Y₁₂ receptor blocker, has been demonstrated to reduce the risk of stroke and cardiovascular events in patients with prior ischemic events and is widely used for secondary prevention in patients with stroke or cardiovascular disease.^1,2 Clopidogrel is also an inactive prodrug that must be converted to a biologically active metabolite by hepatic cytochrome P450 (CYP) enzymes.³ In the 2-step oxidative process of biotransformation, CYP2C19 is the major enzyme involved in the generation of the active metabolite of clopidogrel, and its genetic variants can affect the drug’s antiplatelet function. Among the several genetic variants of CYP2C19, the 2 most frequent variants,⁴ CYP2C19*2 and CYP2C19*3, are known as loss-of-function (LOF) variants, whereas the gain-of function variant CYP2C19*17 is associated with increased enzyme function of CYP2C19.⁵ Within CYP2C19 LOF variants, the prevalence of intermediate metabolizers (IMs) or poor metabolizers (PMs) is much higher in Asian populations than in Caucasian populations.⁶ Thus, clopidogrel should be used with caution in Asian patients.

In stroke patients, a previous meta-analysis identified a significant association between CYP2C19 variants and increased composite vascular events.⁷ To our knowledge, all related studies have consistently showed a relationship between CYP2C19 variants and platelet reactivity.^8–12 However, several studies have failed to show a connection between the genetic variants and the rate of cardiovascular events.^13–15 To address these inconsistent results, we hypothesized that differences in the elapsed time from stroke onset might affect the relationship between the genetic variants and cardiovascular events because most studies enrolled stroke patients in the acute phase (within 1 week of the index of stroke).^{8–12,15–20} We therefore prospectively investigated whether CYP2C19 variants affected cardiovascular events during 2 years of follow-up in the chronic phase of stroke.

Methods

This study was a prospective, multicenter cohort study consisting of 14 stroke institutions in Japan (ClinicalTrials.gov Identifier: NCT02711410). It assessed platelet function and CYP2C19 variants in patients treated with clopidogrel for the secondary prevention of stroke. The study protocol was approved by the ethical review committee of each institution. All methods were performed in accordance with relevant guidelines and regulations and written informed consent was given by all the participants.

Study Population

Patients were recruited between October 2009 and March 2012 at each institution. We enrolled males and females aged 20 years or older who (1) had experienced an ischemic stroke or transient ischemic attack (TIA) (excluding cardiogenic embolism) in the 3 years prior but not in the past month; (2) had received long-term clopidogrel therapy (75 mg once a day) for secondary prevention of stroke (for at least 1 month); and (3) were willing and able to provide written informed consent. The exclusion criteria were malignancies, congenital bleeding tendency, atrial fibrillation, concomitant use of an oral anticoagulant agent, platelet count of <100×10⁹/L or >450×10⁹/L within 3 months of enrollment, or modified Rankin Score >4.

Genotyping of CYP2C19 Variants

Genomic DNA was extracted from peripheral blood leukocytes. The genotype of the CYP2C19*2 (c.681G>A, rs4244285), *3 (c.636G>A, rs4986893), and *17 (c.-806C>T, rs12248560) variants was determined by the TaqMan genotype discrimination method (Applied Biosystems, Foster City, CA, USA) using commercially available primers and probes purchased from the Assay-on-Demand system.

Patients were classified into three CYP2C19 metabolizer groups according to genotype as follows: (1) extensive metabolizer (EM) not bearing the genetic variants CYP2C19*2 or CYP2C19*3 (*1/*1); (2) IM carrying one LOF allele (*1/*2, *1/*3); and (3) PM carrying 2 LOF alleles (*2/*2, *2/*3, *3/*3). There has been controversy concerning regarding the phenotype associated with the CYP2C19*2/*17 and *3/*17 genotypes, which are rarely found in Asian populations.^21,22 Therefore, in the present study, patients with the CYP2C19*17 genotype (n=10) were excluded from analysis.

Platelet Aggregation Assay

Residual platelet reactivity was measured approximately 1 month after enrollment in the study. Fasting blood samples were collected from the antecubital vein. Citrate-anticoagulated whole blood was centrifuged at 190 g for 10 min at room temperature within 1 h of collection to obtain platelet-rich plasma. ADP-induced platelet aggregation was determined by a change in light transmission (LT) induced by platelet aggregation in platelet-rich plasma after the addition of ADP (final concentration, 5 or 10 μmol/L) by LT aggregometry (PA-200 or PA-20, Kowa, Japan; HEMA TRACER 313M, model PAT-2A; or PATI analyzer PAM-12C, MC Medical, Tokyo).²³ The percentage of maximal LT was calculated on the basis of a 100% reference value that was the mean of the maximal LT determined using the same machine but with healthy volunteers.

Vasodilator-Stimulated Phosphoprotein (VASP) Index Analysis

Fasting blood samples for the VASP index analysis were collected, as described for the platelet aggregation assay. The VASP index analysis was performed at 2 institutions: the National Cerebral and Cardiovascular Center and Kobe City Medical Center General Hospital. The blood samples collected at other institutions, excluding those obtained from Kobe City Medical Center General Hospital, were transported to the National Cerebral and Cardiovascular Center at room temperature and analyzed within 48 h of collection. VASP index phosphorylation analysis was performed using a PLT VASP/P2Y₁₂ assay kit (Biocytex, Marseille, France). In brief, a citrated blood sample was incubated with prostaglandin E1 (PGE1) or with both PGE1 and ADP 10 mol/L for 10 min and fixed with paraformaldehyde. Following this, the platelets were permeabilized and labeled with a CD61 phycoerythrin-labeled platelet-specific antibody and a FITC-labeled phosphorylated VASP-specific mouse monoclonal antibody or a negative isotype control antibody. The samples were then analyzed on a FACSCalibur^TM flow cytometer (BD Biosciences, Tokyo, Japan). The extent of VASP phosphorylation was expressed as the VASP index, which was calculated from the mean fluorescence intensity (MFI) of samples incubated with PGE1 or PGE1 and ADP according to the following formula: VASP index=[(MFI(PGE1)–MFI(PGE1+ADP))/MFI(PGE1)×100. The ratio of phosphorylated to dephosphorylated VASP protein is directly dependent on the level of activation of the P2Y₁₂ receptor, thereby permitting specific analyses of the biological efficacy of P2Y₁₂ receptor blockers.

Clinical Outcomes

Patients were examined at 1 month and every 6 months for 2 years after enrollment. The primary outcome of the study was recurrent cerebrocardiovascular events (CVEs), including ischemic stroke, TIA, acute myocardial infarction (AMI), urgent revascularization, other thromboembolic disease, and cardiovascular death. Secondary outcomes included major bleeding, including intracranial hemorrhage, documented retroperitoneal bleed, and bleeding episodes that caused ≥2 g/dL decline in the hemoglobin level or required at least 1 unit of red cell transfusion. Ischemic stroke was defined as a new neurological deficit of vascular origin lasting at least 24 h, based on radiological evidence of stroke. TIA was defined as a transient episode of neurological dysfunction caused by focal brain ischemia that improved within 24 h. Urgent revascularization was defined as being an emergency and requiring unexpected hospitalization. AMI was defined as cell necrosis of the myocardium in the setting of ischemic symptoms and ECG changes (ST-segment elevation or depression, development of Q waves) and/or rising of specific cardiac biomarkers. Cardiovascular death was defined as death from stroke, MI or documented sudden cardiac death. All outcomes were adjudicated by the independent Clinical Events Committee, which comprised 2 medical doctors who were unaware of the patients’ genetic information and laboratory assay results.

Statistical Analysis

The sample size of the present study was determined on the basis of previous reports as follows: the composite event rate was estimated to be 10% over 2 years in previous studies.¹ We assumed a hazard ratio (HR) of 2.0 for the recurrence of CVEs in the IM and PM groups, compared with the EM group.^24,25 In addition, we assumed a prevalence of 70% among patients in the IM and PM groups.²⁶ Using the precision method, a sample size of 498 patients was determined to produce an approximated two-sided 95% confidence interval (CI) for the HR with a width equal to ±0.126 on a log-transformed scale (the 95% CI should be 1.76–2.27). Continuous variables are presented using the interquartile range (IQR), whereas categorical variables are expressed as frequencies and percentages. In addition, categorical variables were compared using the χ² test or Fisher’s exact test. Continuous data were compared by Kruskal-Wallis test. Platelet reactivity data and the VASP index were compared among groups by one-way analysis of variance (ANOVA) analysis. The cumulative event rates of the outcomes were estimated by the Kaplan-Meier method and compared among the 3 metabolizer groups (EM, IM, and PM) using the log-rank test. To confirm the robustness of the results, sensitivity analysis (case-deleted study) was performed.

The HR and 95% CI were calculated using a Cox proportional hazard regression model. A multivariable Cox proportional hazard regression analysis was also performed on the primary outcomes using 6 variables achieving P<0.2 in the univariate analysis and sex to avoid potential confounding variables in this cohort study. Wilcoxon’s rank sum test was used to evaluate differences in the residual platelet reactivity between the presence/absence of CVEs.

For all statistical analyses, P-values were two-sided. P<0.05 indicated a significant difference. All statistical analyses were performed using R for windows version 3.5.0 (www.r-project.org).

Results

Patients’ Characteristics

In total, 518 patients were registered, but 7 were excluded because of lack of CYP2C19 genotype data, and the 10 patients with the CYP2C19*17 allele because of the limited effects of that allele on platelet reactivity. Consequently, 501 Japanese patients were eligible for analysis. The allelic frequencies of CYP2C19 *1, *2, and *3 in the 501 patients were 60.6%, 29.6%, and 9.8%, respectively. The distribution of the clopidogrel-metabolizing groups according to CYP2C19 genotypes was as follows: 36.3%, 48.5%, and 15.2% for the EM (n=182; *1/*1), IM (n=243; *1/*2, *1/*3), and PM (n=76; *2/*2, *2/*3, *3/*3) groups, respectively (Table 1). The baseline demographics and characteristics of the patients are shown in Table 2. At baseline, the median age of the sample was 68 years and the proportion of males was 72.7%. The leading stroke subtype was atherothrombotic (43.9%). With regard to major comorbidities, hypertension and diabetes mellitus were present in 78.8% and 32.9% of patients, respectively. Incidence of concomitant drugs, statins, proton-pump inhibitors, and other antiplatelet drugs was not different among groups.

Table 1. Genotyping Results in Patients Treated With Clopidogrel

	CYP2C19*2
CYP2C19*3
Genotype	G/G (*1*1)	G/A (*1*2)	A/A (*2*2)
G/G (*1*1)	182 (36.3)	188 (37.5)	40 (8.0)
G/A (*1*3)	55 (11.0)	29 (5.8)	0
A/A (*3*3)	7 (1.4)	0	0

Data reflect the number of patients (percentage). The participants included 182 extensive metabolizers (EMs) without polymorphisms for CYP2C19*2 or CYP2C19*3 (36.3%), 243 intermediate metabolizers (IMs) carrying 1 loss-of-function allele (*1/*2 or *1/*3) (48.5%), and 76 poor metabolizers (PMs) carrying 2 loss-of-function alleles (*2/*2, *2/*3, *3/*3) (15.2%).

Table 2. Clinical Characteristics of Extensive, Intermediate, and Poor Metabolizers According to CYP2C19 Genotype

	EM (n=182)	IM (n=243)	PM (n=76)	Total (n=501)	P value
Male, n (%)	139 (76.4)	172 (70.8)	53 (69.7)	364 (72.7)	0.36
Age, years	68 (61–75)	68 (62–74)	63 (58–70)	68 (61–74)	0.003
Previous stroke subtype or TIA, n (%)
Atherothrombotic	79 (43.4)	113 (46.5)	28 (36.8)	220 (43.9)	0.33
Lacunar	62 (34.1)	82 (33.7)	24 (31.6)	168 (33.5)	0.92
Other or undetermined	17 (9.3)	26 (10.7)	9 (11.8)	52 (10.4)	0.81
TIA	24 (13.2)	22 (9.1)	15 (19.7)	61 (12.2)	0.04
Time from stroke onset to study entry, days	173 (82–397)	182 (63–518)	178 (90–497)	181 (73–463)	0.68
Carotid artery stenosis, n (%)	100 (54.9)	124 (51.0)	35 (46.1)	259 (51.7)	0.41
Hypertension, n (%)	151 (83.0)	189 (77.8)	55 (72.4)	395 (78.8)	0.14
Diabetes mellitus, n (%)	63 (34.6)	81 (33.3)	21 (27.6)	165 (32.9)	0.54
Current smoker, n (%)	26 (14.3)	35 (14.4)	10 (13.2)	71 (14.2)	0.96
Chronic renal failure, n (%)*	8 (4.4)	21 (8.6)	1 (1.3)	30 (6.0)	0.03
Previous coronary artery disease, n (%)*	12 (6.6)	13 (5.3)	3 (3.9)	28 (5.6)	0.72
BMI, kg/m2	24.1 (22.1–25.8)	23.4 (21.1–25.4)	23.7 (22.2–25.8)	23.7 (21.6–25.6)	0.04
BP on recruitment, mmHg
Systolic	134 (124–148)	136 (123–148)	139 (126–148)	136 (124–148)	0.45
Diastolic	78 (70–87)	79 (71–86)	80 (74–90)	79 (71–86)	0.15
Concomitant drugs, n (%)
Aspirin	15 (8.2)	25 (10.3)	8 (10.5)	48 (9.6)	0.74
Cilostazol*	10 (5.5)	10 (4.1)	3 (3.9)	23 (4.6)	0.81
Proton-pump inhibitor	41 (22.5)	44 (18.1)	14 (18.4)	99 (19.8)	0.50
Statins	107 (58.8)	163 (67.1)	48 (63.2)	318 (63.5)	0.21
NSAIDs*	10 (5.5)	10 (4.1)	1 (1.3)	21 (4.2)	0.37

Categorical variables are presented as number of patients (percentage). All continuous variables are presented as median (interquartile range). Kruskal-Wallis test was performed on continuous variables. Chi-squared test was performed for most categorical variables. *Fisher’s exact test was performed. BMI, body mass index; BP, blood pressure; NSAIDs, nonsteroidal anti-inflammatory drugs; TIA, transient ischemic attack. Other abbreviations as in Table 1.

Effects of CYP2C19 Genotype on the Antiplatelet Effects of Clopidogrel

Of the 501 patients, 1 patient did not complete ADP-induced platelet aggregation analysis, and 52 did not complete the VASP index, so they were excluded from the study. The lowest residual ADP-induced platelet aggregation values in patients treated with clopidogrel were observed in the EM group. Residual platelet reactivity was significantly higher in the IM and PM groups than in the EM group, according to 5 μmol/L (EM: 51.5±17.5%, IM: 59.7±16.2%, PM: 69.4±17.1%, P<0.0001) and 10 μmol/L ADP-induced platelet aggregation (EM: 58.6±16.6%, IM: 65.8±15.0%, PM: 74.6±15.0%, P<0.0001) analyses. The VASP index was also significantly higher in the IM and PM groups than in the EM group (EM: 43.3±14.9%, IM: 52.6±15.9%, PM: 66.6±14.3%, P<0.0001). Thus, CYP2C19 variants were significantly associated with residual platelet reactivity and the VASP index.

CYP2C19 Genotype and Clinical Outcomes

The median time from the index stroke to enrollment was 181 days. The 2-year follow-up (median: 728 days; IQR: 700-746) was completed in 89.6% (449/501) of the enrolled patients.

In total, 28 (5.59%) CVEs occurred in the followed-up patients, including 17 cases of ischemic stroke, 7 cases of TIA, 2 cases of AMI, and 2 cases of urgent revascularization. No cardiovascular deaths were recorded. Four major bleeding events occurred. Kaplan-Meier curves and the number of patients at risk are shown in Figure 1. The cumulative CVE rate in the EM, IM, and PM groups was 5.5%, 5.8%, and 5.3%, respectively (not significantly different among groups; log-rank P=0.98). As shown in Figure 2A,B, the single case-deleted analysis also showed consistent results. Cox proportional hazard regression analysis showed no significant difference in the primary endpoints among the 3 clopidogrel-metabolizing groups [EM vs. PM: HR=0.98 (95% CI=0.31–3.11), P=0.98; EM vs. IM: HR=1.08 (95% CI=0.48–2.42), P=0.86] (Table 3). In the multivariable Cox proportional hazard regression analysis to check for potential confounding variables, there are no other significant variables related to clinical outcomes (Supplementary Table).

Figure 1.

Kaplan-Meier analysis of cumulative cerebrocardiovascular events rate based on CYP2C19 genotype shows no significant difference among extensive metabolizers (EMs), intermediate metabolizers (IMs), and poor metabolizers (PMs) (log-rank P=0.98).

Figure 2.

Case-deleted analysis for CYP2C19 genotypes. (A) EM vs. IM group; (B) EM vs. PM group. Sensitivity analysis was used to estimate the effect of the hazard ratio (HR) by deletion of individual cases. Both case-deleted analyses showed no significant change in HR and confirmed the robustness of the results. EM, extensive metabolizer; IM, intermediate metabolizer; PM, poor metabolizer.

Table 3. Event Rates and HRs for CVEs and Major Bleeding According to CYP2C19 Genotype

Event	EM (n=182)	IM (n=243)	PM (n=76)	Total (n=501)
CVE, n (%)
Ischemic stroke	5 (2.7)	10 (4.1)	2 (2.6)	17 (3.4)
TIA	2 (1.1)	3 (1.2)	2 (2.6)	7 (1.4)
AMI	1 (0.5)	1 (0.4)	0 (0)	2 (0.4)
Urgent revascularization	2 (1.1)	0 (0)	0 (0)	2 (0.4)
Cardiac death	0 (0)	0 (0)	0 (0)	0 (0)
Total	10 (5.5)	14 (5.8)	4 (5.3)	28 (5.6)
HR (95% CI)	Ref.	1.08 (0.48–2.42) P=0.86	0.98 (0.31–3.11) P=0.97
Major bleeding, n (%)	1 (0.5)	3 (1.2)	0 (0)	4 (0.8)
HR (95% CI)	Ref.	2.27 (0.24–21.77) P=0.48	–‡	–‡

Data are presented as the number of patients or hazard ratio (percentage or 95% CI). AMI, acute myocardial infarction; CI, confidence interval; CVE, cerebrocardiovascular event; HR, hazard ratio. Other abbreviations as in Tables 1,2. ^‡No events were recorded in the PM group, and thus the HR and 95% CI could not be estimated.

Antiplatelet Effects of Clopidogrel and Clinical Outcomes

Figure 3 shows the effect of ADP-induced platelet aggregation (5 or 10 μmol/L ADP) and the VASP index on clinical outcomes. There was no significant difference in residual platelet reactivity or clinical outcomes among the 3 groups as assessed by 5 μmol/L ADP-induced platelet aggregation (no CVEs: 58.3±17.8%, CVEs: 56.1±18.8%, P=0.75) (Figure 3A), 10 μmol/L ADP-induced platelet aggregation (no CVEs: 64.5±16.3%, CVEs: 63.9±20.1%, P=0.94) (Figure 3B), and the VASP index (no CVEs: 49.9±18.3%, CVEs: 47.3±22.6%, P=0.73) (Figure 3C).

Figure 3.

Effects of ADP-induced platelet aggregation and the VASP index on clinical outcomes. The horizontal line within each box represents the median, and the lower and upper borders of each box represent the 25th and 75th percentiles, respectively. I bars represent the maximum and minimum values of the observation. Platelet reactivity was compared across cerebrocardiovascular events (CVEs) using a Wilcoxon test model. There was no significant difference in residual platelet reactivity and clinical outcomes among the 3 groups assessed by (A) 5 μmol/L ADP-induced platelet aggregation (no CVEs: 58.3±17.8%, CVEs: 56.1±18.8%, P=0.75), (B) 10 μmol/L ADP-induced platelet aggregation (no CVEs: 64.5±16.3%, CVEs: 63.9±20.1%, P=0.99), and (C) VASP index (no CVEs: 51.5±17.0%, no CVEs: 52.0±19.6%, P=0.77). ADP, adenosine diphosphate; VASP, vasodilator-stimulated phosphoprotein.

Discussion

To clarify the relationship between CYP2C19 variants and clinical outcomes in patients treated with clopidogrel in the chronic stroke phase, we registered 518 Japanese non-acute stroke patients at 14 institutions across the country using a prospective study design. In total, 501 patients were eligible and successfully followed up for at least 2 years, during which time 28 CVEs and 4bleeding events occurred.

The main result of our study was that CYP2C19 variants did not influence the long-term clinical outcomes in the chronic stroke phase despite a clear relationship between genotype and platelet reactivity. Even taking into consideration potential confounding factors such as sex, age, stroke type and other stroke factors, the results maintained consistency. To our knowledge, this is the first prospective long-term (2 years) follow-up study limited to chronic stroke patients (median time from index stroke=6 months) (Table 4).

Table 4. Previous Studies of the Relationship Between CYP2C19 Variants and Clinical Outcomes in Stroke Patients Excluding Surgical Management

	Study	Year	Study type	Country	Center	No. of subjects	Enrollment timing*	Follow-up period	Outcomes reported	Event rate during follow-up period	Significant relationship between CYP2C19 variants and outcomes	Significant relationship between CYP2C19 variants and platelet reactivity
1	Jia et al8	2013	Cohort	China	Single	259	Stroke onset	6 months	NIHSS, mRS	NA	Significant	Significant (ADP)
2	Spokoyny et al27	2014	Retrospective cohort	USA	Single	43	NA	NA	Ischemic stroke, TIA	20.90%	Significant	NA
3	Zhang et al9	2014	Prospective cohort	China	Single	95	Stroke onset	6 months	Stroke, other ischemic vascular events	18.95%	Significant	Significant (VASP)
4	Han et al13	2015	Prospective cohort	China	Single	247	Within 14 days after onset	12 months	MACE, stroke, TIA, vascular death	23.30%	Not significant	Significant (VerifyNow†)
5	Jeong et al16	2015	Prospective cohort	Korea	Single	76	5 days after onset	5 days	Early recurrence on DWI	47.40%	Significant	NA
6	McDonough et al14	2015	Post hoc analysis	Worldwide	Multi	522	Within the last 6 months	3.4 years (mean)	Stroke, major bleeding	Recurrent stroke 5.3%, Major bleeding 4.7%	Not significant (except for white subcortical stroke patients)	NA
7	Qiu et al10	2015	Prospective cohort	China	Single	211	Within 7 days after onset	6 months	Stroke, MI, cardiovascular death	15.20%	Significant	Significant (ADP, PAC-1#, CD62P#, CD63#)
8	Sun et al17	2015	Prospective cohort	China	Multi	625	Within 7 days after onset	12.7 months (mean)	Vascular death, ischemic stroke, MI, bleeding	13.60%	Significant	NA
9	Hoh et al15	2016	Prospective cohort	USA	Multi	188	Within 7 days after onset	12 months	TIA, stroke, MI, death	14.90%	Not significant	NA
10	Wang et al19	2016	Post hoc analysis	China	Multi	1,463	Within 24 h after onset	3 months	Stroke, MI, vascular death	8.30%	Significant	NA
11	Wang et al18	2016	Prospective cohort	China	Multi	321	Within 7 days after onset	12 months	Ischemic stroke, MI, vascular death	20.20%	Significant	NA
12	Yi et al12	2016	Prospective cohort	China	Multi	363	Within 3 days after onset	6 months	Ischemic stroke, TIA, death	9.87%	Significant	Significant (ADP)
13	Tomek et al11	2018	Retrospective cohort	Czech Republic	Single	130	Within 7 days after onset	14.9 months (mean)	Ischemic stroke, TIA, MI, vascular death	14.60%	Significant	Significant (LTA)
14	Yi et al20	2018	Prospective cohort	China	Multi	570	Within 2 days after onset	5 years	Ischemic stroke, TIA, MI, death	18.40%	Not significant	NA
15	Our study	2018	Prospective cohort	Japan	Multi	501	Within 3 years after onset (actual mean span: 181 days after onset)	2 years	Ischemic stroke, TIA, MI, urgent revascularization, other thromboembolic disease, cardiovascular death	5.60%	Not significant	Significant (ADP, VASP)

*Time elapsed from stroke onset to the enrollment in study. ^†P2Y₁₂ test kit to detect the platelet reactivity (Accumetrics, San Diego, CA, USA). ^#Platelet activation markers to identify platelet function. ADP, adenosine diphosphate; DWI, diffusion weighted image; LTA, light transmittance aggregometry; MACE, major adverse cardiac events; mRS, modified Rankin Scale; NA, not applicable; NIHSS, National Institutes of Health Stroke Scale; TIA, transient ischemic attack; VASP, vasodilator-stimulated phosphoprotein.

A meta-analysis of stroke prevention studies revealed that carriers of LOF alleles of CYP2C19 *2, *3, and *8 were at increased risk of stroke compared with noncarriers,⁷ but other studies showed negative results (Table 4).^13–15 This discrepancy may originate from differences among studies in the time elapsed from stroke onset. In Table 4, except for 1 study not showing enrollment timing and follow-up period,²⁷ 8 of 11 (72.7%) studies that observed CVEs within 1 year after stroke onset demonstrated a significant association with CYP2C19 variant.^{8–12,16–20} However, 2 of 4 (50.0%) longer-term studies (>1 year) of stroke patients have shown a significant association of CYP2C19 variants and risk of stroke.^11,20 Moreover, 10 of 11 (90.9%) studies that recruited subjects within 1 week after the index stroke demonstrated significant associations with CYP2C19 variants.^{8–12,16–20} In contrast, none of 3 studies that recruited subjects within >2 weeks of the index stroke showed significant relationships. In a cardiological study, Arima et al revealed that CYP2C19 LOF alleles do not significantly affect clinical outcomes in patients with stable angina.²⁶ Another study in patients with coronary artery disease reported a significant association of CYP2C19 variants with worse outcomes within the first year although the adverse effects of the variants were significantly reduced thereafter.²⁸ We also reported that neither high on-treatment platelet reactivity nor high residual cyclooxygenase-1 activity was significantly associated with stable cardiovascular patients treated with aspirin.²⁹ These findings are consistent with the notion that despite the continuing influence of CYP2C19 variants on platelet aggregation, the recurrence of CVEs seems to be affected by the time from stroke onset. In the Clopidogrel in High-risk patients with Acute Non-disabling Cerebrovascular Events (CHANCE) trial, Wang et al reported that CYP2C19 LOF variants were associated with poor prognosis at 3 months after ischemic stroke; however, an add-on of clopidogrel to aspirin in this trial did not improve the prognosis between 3 and 12 months after the onset of stroke.³⁰ One plausible explanation is that although residual platelet reactivity is a surrogate marker, the effects of unknown risk factors may have a significant influence on long-term outcomes after stroke.

Study Limitations

Several need to be considered in the current study. (1) The low rate of recurrent CVEs (5.6% over 2 years) could explain the lack of difference in CVE rates among the genetic groups, as a result of weak statistical power. However, to evaluate the robustness (internal validity) of this negative result, we performed case-deletion influence diagnostics as a sensitivity analysis, allowing identification of influential observations that may affect parameter estimates and model fitting conclusions.³¹ Even if removed on a case-by-case basis, there was no significant change in the HR (Figure 2), thus confirming the statistical robustness of our results. (2) We did not recruit stroke patients in the acute phase, preventing comparison with chronic phase patients. (3) We assessed platelet reactivity at a single point approximately 1 month after enrollment only, ruling out determining whether residual platelet reactivity was high at the time of recurrent CVEs. (4) We cannot exclude the meaningful effects of other genetic variants, which may have affected responsiveness to clopidogrel.

Conclusions

Despite associations between CYP2C19 variants and on-clopidogrel platelet reactivity, there was no significant difference in the outcome of CVEs in the chronic stroke phase. To identify the important period after stroke in relation to CYP2C19 variants, further studies are needed to confirm these results.

Acknowledgments

We thank the members of the Clinical Event Committee: Dr. Misa Nakano at Toyonaka Municipal Hospital and Dr. Takuma Maeda at the National Cerebral and Cardiovascular Center. We thank all patients who participated in this study and Kazuyoshi Nakai for the platelet reactivity assays.

Disclosures

Dr. Nakagawara, Uchiyama, Yasaka, Kitagawa, and Nagatsuka have received speaker’s honoraria from Sanofi-Aventis. The other authors report no conflicts.

Grants

This study was supported in part by grants-in-aid from the Ministry of Health, Labour and Welfare of Japan (H20-Junkanki-Ippan-016) and the Intramural Research Fund of the National Cerebral and Cardiovascular Center of Japan.

Supplementary Files

Please find supplementary file(s);

http://dx.doi.org/10.1253/circj.CJ-18-1386

References

• 1.   CAPRIE Steering Committee. A randomised, blinded, trial of clopidogrel versus aspirin in patients at risk of ischaemic events (CAPRIE). Lancet 1996; 348: 1329–1339.
• 2.   Yusuf S, Zhao F, Mehta SR, Chrolavicius S, Tognoni G, Fox KK, et al. Effects of clopidogrel in addition to aspirin in patients with acute coronary syndromes without ST-segment elevation. N Engl J Med 2001; 345: 494–502.
• 3.   Savi P, Herbert JM, Pflieger AM, Dol F, Delebassee D, Combalbert J, et al. Importance of hepatic metabolism in the antiaggregating activity of the thienopyridine clopidogrel. Biochem Pharmacol 1992; 44: 527–532.
• 4.   Hulot JS, Bura A, Villard E, Azizi M, Remones V, Goyenvalle C, et al. Cytochrome P450 2C19 loss-of-function polymorphism is a major determinant of clopidogrel responsiveness in healthy subjects. Blood 2006; 108: 2244–2247.
• 5.   Sim SC, Risinger C, Dahl ML, Aklillu E, Christensen M, Bertilsson L, et al. A common novel CYP2C19 gene variant causes ultrarapid drug metabolism relevant for the drug response to proton pump inhibitors and antidepressants. Clin Pharmacol Ther 2006; 79: 103–113.
• 6.   Bauer T, Bouman HJ, van Werkum JW, Ford NF, ten Berg JM, Taubert D. Impact of CYP2C19 variant genotypes on clinical efficacy of antiplatelet treatment with clopidogrel: Systematic review and meta-analysis. BMJ 2011; 343: d4588.
• 7.   Pan Y, Chen W, Xu Y, Yi X, Han Y, Yang Q, et al. Genetic polymorphisms and clopidogrel efficacy for acute ischemic stroke or transient ischemic attack: A systematic review and meta-analysis. Circulation 2017; 135: 21–33.
• 8.   Jia DM, Chen ZB, Zhang MJ, Yang WJ, Jin JL, Xia YQ, et al. CYP2C19 polymorphisms and antiplatelet effects of clopidogrel in acute ischemic stroke in China. Stroke 2013; 44: 1717–1719.
• 9.   Zhang S, Lai X, Li W, Xiong Z, Xu A, Xu A, et al. VASP phosphorylation and genetic polymorphism for clopidogrel resistance in Chinese patients with non-cardioembolic ischemic stroke. Thromb Res 2014; 134: 1272–1277.
• 10.   Qiu LN, Sun Y, Wang L, Han RF, Xia XS, Liu J, et al. Influence of CYP2C19 polymorphisms on platelet reactivity and clinical outcomes in ischemic stroke patients treated with clopidogrel. Eur J Pharmacol 2015; 747: 29–35.
• 11.   Tomek A, Mat’oska V, Frydmanova A, Magerova H, Sramek M, Paulasova-Schwabova J, et al. Impact of CYP2C19 polymorphisms on clinical outcomes and antiplatelet potency of clopidogrel in caucasian poststroke survivors. Am J Ther 2018; 25: e202–e212.
• 12.   Yi XY, Lin J, Wang YF, Zhou Q, Wang C, Cheng W, et al. Association of cytochrome P450 genetic variants with clopidogrel resistance and outcomes in acute ischemic stroke. J Atheroscler Thromb 2016; 23: 1188–1200.
• 13.   Han Y, Lv HH, Liu X, Dong Q, Yang XL, Li SX, et al. Influence of genetic polymorphisms on clopidogrel response and clinical outcomes in patients with acute ischemic stroke CYP2C19 genotype on clopidogrel response. CNS Neurosci Ther 2015; 21: 692–697.
• 14.   McDonough CW, McClure LA, Mitchell BD, Gong Y, Horenstein RB, Lewis JP, et al. CYP2C19 metabolizer status and clopidogrel efficacy in the secondary prevention of small subcortical strokes (SPS3) study. J Am Heart Assoc 2015; 4: e001652.
• 15.   Hoh BL, Gong Y, McDonough CW, Waters MF, Royster AJ, Sheehan TO, et al. CYP2C19 and CES1 polymorphisms and efficacy of clopidogrel and aspirin dual antiplatelet therapy in patients with symptomatic intracranial atherosclerotic disease. J Neurosurg 2016; 124: 1746–1751.
• 16.   Jeong TD, Kim SM, Kim HJ, Lee W, Kwon SU, Min WK, et al. CYP2C19 genotype and early ischemic lesion recurrence in stroke patients treated with clopidogrel. J Stroke Cerebrovasc Dis 2015; 24: 440–446.
• 17.   Sun W, Li Y, Li J, Zhang Z, Zhu W, Liu W, et al. Variant recurrent risk among stroke patients with different CYP2C19 phenotypes and treated with clopidogrel. Platelets 2015; 26: 558–562.
• 18.   Wang Y, Cai H, Zhou G, Zhang Z, Liu X. Effect of CYP2C19*2 and *3 on clinical outcome in ischemic stroke patients treated with clopidogrel. J Neurol Sci 2016; 369: 216–219.
• 19.   Wang Y, Zhao X, Lin J, Li H, Johnston SC, Lin Y, et al. Association between CYP2C19 loss-of-function allele status and efficacy of clopidogrel for risk reduction among patients with minor stroke or transient ischemic attack. JAMA 2016; 316: 70–78.
• 20.   Yi X, Lin J, Zhou J, Wang Y, Huang R, Wang C. The secondary prevention of stroke according to cytochrome P450 2C19 genotype in patients with acute large-artery atherosclerosis stroke. Oncotarget 2018; 9: 17725–17734.
• 21.   Scott SA, Sangkuhl K, Stein CM, Hulot JS, Mega JL, Roden DM, et al. Clinical pharmacogenetics implementation consortium guidelines for CYP2C19 genotype and clopidogrel therapy: 2013 update. Clin Pharmacol Ther 2013; 94: 317–323.
• 22.   Fricke-Galindo I, Cespedes-Garro C, Rodrigues-Soares F, Naranjo ME, Delgado A, de Andres F, et al. Interethnic variation of CYP2C19 alleles, ‘predicted’ phenotypes and ‘measured’ metabolic phenotypes across world populations. Pharmacogenomics J 2016; 16: 113–123.
• 23.   Kameda S, Sakata T, Kokubo Y, Mitsuguro M, Okamoto A, Sano M, et al. Association of platelet aggregation with lipid levels in the Japanese population: The Suita study. J Atheroscler Thromb 2011; 18: 560–567.
• 24.   Mega JL, Close SL, Wiviott SD, Shen L, Hockett RD, Brandt JT, et al. Cytochrome p-450 polymorphisms and response to clopidogrel. N Engl J Med 2009; 360: 354–362.
• 25.   Collet JP, Hulot JS, Pena A, Villard E, Esteve JB, Silvain J, et al. Cytochrome P450 2C19 polymorphism in young patients treated with clopidogrel after myocardial infarction: A cohort study. Lancet 2009; 373: 309–317.
• 26.   Arima Y, Hokimoto S, Akasaka T, Mizobe K, Kaikita K, Oniki K, et al. Comparison of the effect of CYP2C19 polymorphism on clinical outcome between acute coronary syndrome and stable angina. J Cardiol 2015; 65: 494–500.
• 27.   Spokoyny I, Barazangi N, Jaramillo V, Rose J, Chen C, Wong C, et al. Reduced clopidogrel metabolism in a multiethnic population: Prevalence and rates of recurrent cerebrovascular events. J Stroke Cerebrovasc Dis 2014; 23: 694–698.
• 28.   Peng Y, Chen M, Liu XJ, Liu W, Li Q, Chai H, et al. The CYP2C19 genotype does not impact the long-term prognosis of patients with coronary artery disease. Atherosclerosis 2013; 227: 106–111.
• 29.   Nagatsuka K, Miyata S, Kada A, Kawamura A, Nakagawara J, Furui E, et al. Cardiovascular events occur independently of high on-aspirin platelet reactivity and residual COX-1 activity in stable cardiovascular patients. Thromb Haemost 2016; 116: 356–368.
• 30.   Wang Y, Pan Y, Zhao X, Li H, Wang D, Johnston SC, et al. Clopidogrel with aspirin in acute minor stroke or transient ischemic attack (CHANCE) trial: One-year outcomes. Circulation 2015; 132: 40–46.
• 31.  Cook RD, Weisberg S. Residuals and influence in regression. New York: Chapman and Hall. 1982. [Retrieved from the University of Minnesota Digital Conservancy, http://hdl.handle.net/11299/37076]