Article Text
PDF
Abstract
Objective To investigate the influence of CYP2C19 genetic polymorphisms on clinical outcomes of intracranial aneurysms treated with stent-assisted coiling.
Methods Between September 2014 and October 2015, we prospectively recruited 215 patients with intracranial aneurysms who were treated with stent-assisted coiling. CYP2C19 genotypes were determined and clopidogrel response was tested. The primary endpoints included symptomatic or silent ischemic events, and bleeding events. The secondary endpoint was clinical outcome at 3 months.
Results Of the 215 patients, 108 (50.2%) were classified as intermediate metabolizers (IMs, CYP2C19*1/*2, *1/*3), 76 (35.3%) as extensive metabolizers (EMs, CYP2C19*1/*1) and 31 (14.4%) as poor metabolizers (PMs, CYP2C19*2/*2, *2/*3, *3/*3). Carriers of CYP2C19 loss-of-function (LOF) alleles (*2 or *3, p=0.001), especially PMs (p=0.004), had an increased risk for clopidogrel resistance. After the procedures, cerebral ischemic events occurred in 69 patients (32.1%) and bleeding was seen in 20 patients (9.3%). In comparison with IMs and PMs, EMs had a lower risk for ischemic events (21.1% vs 37.0% and 41.9%, p=0.02 and 0.027, respectively) and a relatively higher risk for bleeding events (18.4% vs 5.6% and 0%, p=0.006 and 0.01, respectively). Based on multivariate analysis, the carriage of CYP2C19 LOF alleles (p=0.032) and clopidogrel resistance (p=0.047) were considered as predictors of cerebral ischemic events, and EMs were significantly associated with bleeding (p=0.002). Posterior circulation aneurysms (p=0.038), hemorrhagic history (p=0.001) and poor metabolic genotypes (p=0.001) could result in poor clinical outcomes (modified Rankin Scale >2).
Conclusions CYP2C19 genetic polymorphisms had significant influence on the antiplatelet effect of clopidogrel, and could be considered as risk factors of ischemic or bleeding events and even clinical outcomes of patients with intracranial aneurysms treated with stent-assisted coiling.
- Aneurysm
- Complication
- Platelets
- Stent
Statistics from Altmetric.com
Introduction
Stent-assisted coiling has been used for wide-necked, complex-shaped and dissecting intracranial aneurysms. Neurovascular stents protect coils from protruding into the parent artery, and also create a mesh for endothelialization.1 However, the use of stents enhances the thrombotic activity and platelet aggregation, which may result in ischemic complications.2 For intracranial aneurysms treated with stent-assisted coiling, oral antiplatelet therapy (mainly clopidogrel and aspirin) has been recommended as an important protocol to decrease the incidence of procedure-related thrombosis.3 ,4
Clopidogrel, which plays a central role in inhibiting platelet activation or aggregation, is a prodrug requiring biotransformation by cytochrome P450 (CYP) 2C19 enzymes to form active metabolites.5 The expression and function of enzymes that depend on the CYP2C19 genotypes mainly affect the transforming metabolites, and then influence the response of clopidogrel.6 Recently, numerous reports have focused on the role of CYP2C19 genetic polymorphisms (such as CYP2C19*1, CYP2C19*2, CYP2C19*3, CYP2C19*17) in treating acute coronary syndrome (ACS) or ischemic stroke.7–12 In comparison with studies that explored the pharmacogenomics of clopidogrel in patients with cardiovascular disease or cerebral ischemic stroke, there were few reports referring to intracranial aneurysms. Additionally, the practice of individualized antiplatelet therapy and standard predictors for clopidogrel response have not been well investigated.13
This prospective observational cohort study identified the association among CYP2C19 genetic polymorphisms, clopidogrel response and the clinical outcomes of intracranial aneurysms treated with stent-assisted coiling.
Materials and methods
Study population
Between September 2014 and October 2015, 215 patients were prospectively recruited in our institution. Each patient gave written informed consent to participate in the study and their privacy was strictly protected. The protocol of this study was approved by our ethics committee. Inclusion criteria were the following: (1) 18–80 years of age; (2) wide-necked intracranial aneurysms; (3) treated with stent-assisted coiling. Exclusion criteria included: (1) treatment with anticoagulants, thrombolytic agents and other antiplatelet drugs within 2 weeks; (2) acute ruptured aneurysms; (3) severe cardiovascular diseases or cerebral ischemic stroke; (4) significant coagulopathy; (5) severe hepatic or renal dysfunction, malignant diseases, chronic inflammatory diseases or infectious conditions at study entry; (6) missing the clinical follow-up. Baseline demographic data, including age, sex, clinical presentation, cerebrovascular risk factors (smoking and drinking history, diabetes, hypertension, hypercholesterolemia) and aneurysm characteristics (location, size, occlusion grade), were recorded.
CYP2C19 genotyping
The CYP2C19 genetic polymorphisms were extracted from blood samples and detected using the PCR-restriction fragment length polymorphisms technique. Polymorphisms were analyzed for wild-type alleles (*1) and loss-of-function (LOF) alleles (*2 and *3). The other alleles (CYP2C19*4, *5, *6, *7, *8 and *17) were much less common and were not tested in this study. Individuals who were carriers of *1/*1 (the normal copy that possesses full enzymatic activity) were classified as ‘extensive metabolizers (EMs)’. Those with one *2 or *3 allele (*1/*2 or *1/*3) were classified as ‘intermediate metabolizers (IMs)’, and those with two *2 or *3 alleles (*2/*2, *2/*3 or *3/*3) as ‘poor metabolizers (PMs)’.
Platelet function testing
Platelet function testing was available in each case. The baseline blood sample was drawn before the administration of standard dual antiplatelet therapy (aspirin and clopidogrel). In order to assess preoperative platelet inhibition, a secondary blood sample was collected immediately before the embolization procedure. A thromboelastography (TEG) hemostasis system (Haemoscope Corporation, Niles, Illinois, USA) was used to test the platelet inhibition. The agonists of adenosine diphosphate (ADP) were added to measure the platelet inhibition of P2Y12 receptors. The maximum amplitude (MA) represented the MA with thrombin-stimulated platelets and fibrin meshwork (the maximum clot strength). The ADP percentage inhibition and MA (MA-ADP, ADP-induced clot strength) were used to measure the response of clopidogrel. The cut-off values of the inhibition rate of ADP <30% and MA-ADP >50 were defined as the clopidogrel resistance, respectively.4 ,14
Endovascular treatments
The endovascular procedure was performed under general anesthesia. During the procedure, a bolus of heparin was administered using 3000 IU, and then 1000 IU per hour. Each recruited patient received dual antiplatelet therapy (clopidogrel, 75 mg/day; aspirin, 100 mg/day) for at least 3 days before the procedure. After the procedure, patients were given clopidogrel (75 mg/day) for 3 months and aspirin (100 mg/day) for at least 6 months. Three types of stents (LVIS (MicroVention), Enterprise (Codman) and Solitaire AB (eV3)) were used. The occlusion rate was assessed for the treated aneurysm by a three-point Roy–Raymond Scale (complete occlusion, neck remnant and residual sac).
Clinical endpoints
The primary endpoints included cerebral ischemic events and bleeding during the 3 months’ follow-up. Ischemic events were evaluated by clinical examination or MRI with diffusion-weighted imaging (DWI) and were divided into two categories:(1) symptomatic ischemic complications, such as a newly developed transient ischemic attack or permanent ischemic infarctions within the period of clopidogrel therapy; and (2) clinically silent acute thromboembolic phenomena, which were located in the territory of the stented vessel, as represented by DWI on postoperative day 1.15 MRI data were independently reviewed by two neuroradiologists. After the procedures, all patients (100%) received clinical neurological examination, and 193 patients (89.8%) underwent MRI-DWI. The bleeding events were defined as intracranial hemorrhages identified by CT or any hemorrhages outside the cranium which appeared once patients received clopidogrel therapy and which remitted when treatment was stopped. The secondary endpoint was clinical outcome, which was assessed by modified Rankin Scale (mRS) at 3 months. In this study, clinical follow-up was performed by telephone or outpatient interview. An mRS score ≤2 was defined as a good clinical outcome, while mRS >2 was considered to be a poor clinical outcome.
Statistical analysis
Continuous variables were presented as mean±SD or expressed as frequencies and percentages. The selected single nucleotide polymorphisms were evaluated by Pearson's χ2 test for the Hardy–Weinberg equilibrium. Analyses of an independent sample t-test and χ2 test were used to assess the differences among EMs, IMs and PMs for continuous and categorical variables, respectively. Univariate and multivariate logistic regression analysis was used to identify independent contributions of CYP2C19 genotypes and other variables to endpoints and clinical outcomes. A two-sided p value <0.05 was considered to be significant. The statistical analysis was performed using SPSS V.17.0 software (SPSS Inc, Chicago, Illinois, USA).
Results
Two hundred and fifteen participants were recruited to the study: 129 (60%) women and 86 (40%) men. The age of patients ranged from 18 to 79 years (mean±SD, 52.3±10.2 years). The allele frequencies of CYP2C19*1, *2 and *3 were 60.5%, 32.8% and 6.7%, respectively. The proportion of genotypes for each allele in the population were consistent with those predicted by the Hardy–Weinberg equilibrium for polymorphisms (p>0.05). The distribution of CYP2C19 polymorphisms is shown in table 1. One hundred and thirty-nine (64.7%) patients were carriers of at least one CYP2C19 LOF allele (IM: 108 (50.2%), PM: 31 (14.4%)) and 76 (35.3%) patients were non-carriers (EM). Table 2 shows the baseline characteristics and subgroup analysis of the EMs, IMs and PMs. There were no statistically significant differences in the distribution of age, sex, other cerebrovascular risk factors and aneurysm characteristics in the three subgroups (all p>0.05).
Genotype frequency of CYP2C19
Characteristics of the study population
CYP2C19 genotypes and platelet reactivity
Table 3 shows the variation of ADP inhibition rate and MA-ADP in accordance with the CYP2C19 genotypes (EMs, IMs and PMs). Sixty-eight cases (31.6%) were shown to have clopidogrel resistance. The EMs had a relatively higher ADP inhibition rate and a lower MA-ADP than the IMs and PMs (all p<0.001) (table 3). Thus, a significantly smaller proportion of EM cases experienced clopidogrel resistance (table 3). Univariate regression analysis demonstrated that the carriage of CYP2C19 LOF alleles (p=0.001, OR=0.307, 95% CI 0.154 to 0.609), especially PMs (p=0.004, OR=0.316, 95% CI 0.145 to 0.687), was significantly associated with clopidogrel resistance.
Comparison analysis based on CYP2C19 genotypes
CYP2C19 genotypes and endpoints
Cerebral ischemic events occurred in 69 patients (32.1%) after procedures, including 21 (9.8%) with transient ischemic attack, 14 (6.5%) with ischemic stroke and 34 (15.8%) with silent acute thromboembolic phenomena, which were detected by DWI on postoperative day 1. For the symptomatic ischemic complications, 26 cases were confirmed by clinical examination within 24 hours and 9 cases were delayed ischemia (3 days to 3 months). It was notable that among the ischemic cases, 16 were CYP2C19 LOF allele non-carriers (*1/*1, EM) and 53 were LOF allele carriers (40 as IMs, 13 as PMs). In comparison with IMs (p=0.02) and PMs (p=0.027), the risk of the cerebral ischemic events in EMs was significant lower (table 3). Based on multivariate analysis, the CYP2C19 LOF alleles and clopidogrel resistance were associated with cerebral ischemic events for the intracranial aneurysms treated with stent-assisted coiling (table 4, p=0.032 and 0.047). Additionally, bleeding occurred in 20 cases, of which 3 were postoperative intracranial hemorrhages and 17 were hemorrhages outside the cranium. Of these cases, few (35%, 7/20) were detected within 3 days after procedures. Table 3 shows 14 (18.4%, 14/76) cases in EMs and 6 (5.6%, 6/108) cases in IMs. The EMs had a higher proportion of bleeding events than IMs (p=0.006) and PMs (p=0.01) (table 3), and could be considered as an independent risk factor of bleeding events (p=0.002, table 5).
Predictors of cerebral ischemic events
Predictors of bleeding events
According to the final follow-up results, 12 (5.6%, 12/215) patients had poor clinical outcomes (mRS>2). Of these patients, eight had cerebral ischemic infarction and continuing neurological deficits. On multivariate regression analysis, the location of the posterior circulation, intracranial hemorrhagic history and PMs (*2/*2, *2/*3 or *3/*3) were significantly associated with poor clinical outcomes (table 6).
Predictors of poor clinical outcomes according to univariate and multivariate logical regression analysis
Discussion
The response to clopidogrel varies widely among individuals. Different responses to clopidogrel may be related to CYP2C19 genetic polymorphisms.13 Several studies have shown that carriers with CYP2C19 alleles *2 or *3 had significantly lower levels of the active metabolite of clopidogrel, and could be considered as risk predictors for clinical outcomes in ACS or patients with cerebrovascular ischemic stroke. McDonough et al16 found that CYP2C19 IMs or PMs had higher residual platelet reactivity than EMs or UMs (ultrarapid metabolizers) during the follow-up period. Wallentin et al17 also reported that carriers of the reduced function CYP2C19 allele variants showed a higher rate of clopidogrel resistance, and the occurrence of cardiovascular death, myocardial infarction or stroke, mainly within the first 30 days after the treatment for ACS. Qiu et al5 indicated that carriers of CYP2C19 LOF alleles had 3.01-fold increased risk of poor prognosis during 6 months’ follow-up compared with non-carriers in patients with ischemic stroke treated with clopidogrel. Another study showed the similar clinical results after the clopidogrel treatment.18
Ischemic disease studies have focused mainly on clopidogrel hyporesponsiveness and identified ischemia as phenotypic outcomes because hemorrhagic complications were relative rare.19 Owing to the different physiopathologic mechanism and the potential risk of rupture, inadequate antiplatelet therapy for patients with intracranial aneurysm might cause fatal complications. A loading dose of 300 mg clopidogrel or its long-term use, which was the standard dosage for a cardiovascular stent-placement procedure, might increase the risk of bleeding or rebleeding in intracranial aneurysms.13 ,20 On the other hand, insufficient platelet inhibition has a significant association with the incidence of thromboembolic complications.15 Thus, experience from coronary artery diseases or cerebral ischemic stroke could not readily be extrapolated to intracranial aneurysms, and the residual platelet reactivity or clopidogrel response should be paid more attention.
In this study, we found that the carriers of CYP2C19*2 or *3 alleles had a relatively lower ADP inhibition rate and higher MA-ADP than CYP2C19 LOF non-carriers (*1/*1). That meant the carriage of at least one CYP2C19 LOF allele, especially the PMs (*2/*2, *2/*3, *3/*3), was associated with a high risk of clopidogrel resistance. This result is in line with existing data for ischemic diseases.21 ,22 It was notable that the weak metabolic gene of CYP2C19 decreased the activity of cytochrome P450 enzymes and influenced the process of transforming metabolites, thus reducing the effective use of clopidogrel and causing ‘clopidogrel resistance’. After stent-assisted coiling procedures, patients with intracranial aneurysm and clopidogrel resistance were prone to experience acute or delayed in-stent thrombosis, and even cerebral ischemic infarctions. Thus, the carriage of CYP2C19 LOF alleles and clopidogrel resistance could be considered as an independent risk predictor of postprocedural cerebral ischemic events in intracranial aneurysms. Our study also showed a statistically significant association between carriers of two *2 or *3 alleles (PM, *2/*2, *2/*3 or *3/*3) and poor clinical outcomes. Consequently, these results provided evidence supporting CYP2C19 genetic testing for Chinese populations, in whom the prevalence of the CYP2C19 LOF allele is high, and might allow clinicians to personalize antiplatelet therapy.
The CYP2C19*17 allele enhanced platelet inhibition and clopidogrel response, thereby possibly increasing the risk of bleeding complications.23–25 Among the 215 eligible blood samples, there were no homozygotes of the CYP2C19*17 allelic variant; this result was similar to that obtained by Ham et al.26 Additionally, we found an unexpected result—namely, that carriers of *1/*1 (EM) were significantly associated with intracranial hemorrhages or any hemorrhages outside the cranium. It was not clear that the influence of this polymorphism on the activity of clopidogrel was similar to the ultrarapid metabolizer phenotype (*17/*17), and hence bleeding events. Further investigation is needed.
Recently, TEG has been widely used for the assessment of platelet function. Unlike other specific platelet function tests, TEG focuses on platelet contribution to clot strength, rather than platelet activation and aggregation; and could be used to monitor coagulation disorder.27 A study demonstrated that the TEG MA parameters might be better than light transmittance aggregometry, which ignores the important contribution of platelet–fibrin interactions to both thrombosis and hemorrhage.28 Wang et al29 found that the ADP inhibition was lower and MA-ADP was higher in patients with ischemic events, and demonstrated that MA-ADP was a predictor for ischemic events after stent treatment. In our study, we also confirmed that clopidogrel resistance could be reflected by ADP inhibition and MA-ADP. These results were similar to those of Tang et al.30
Limitations
Our study had some limitations. First, the patients presented with acute ischemic lesions for several reasons, such as the dislodgement of microthrombi which arose during the stent procedures, plaque cracking and migrating to distal vessels, and the stent itself causing embolism. Second, we used only TEG to measure the resistance to antiplatelet agents because no other method of platelet function testing was available in our center. Third, the sample size was small, so a larger series might be needed to observe hemorrhagic and thromboembolic complications. Fourth, there were no homozygotes of the CYP2C19*17 allele (ultrarapid metabolizer phenotype) that might have increased the incidence of bleeding. More data deserve further investigation. Fifth, this was an observational cohort study designed to investigate the influence of CYP2C19 genetic polymorphisms on clinical outcomes in intracranial aneurysms, thus we did not follow a preventative antiplatelet regimen based on the information of TEG and CYP2C19 genetic polymorphisms. In a further study, we will adjust the antiplatelet regimen according to the previous data.
Conclusion
Carriage of the CYP2C19 LOF allele had significant influence on the antiplatelet effect of clopidogrel and could result in cerebral ischemic events. Patients who were EMs might have an increased risk for bleeding. The importance of testing CYP2C19 genetic polymorphisms and platelet function should be considered to predict clinical outcomes in patients with intracranial aneurysms treated with stents and clopidogrel. Further research is needed to focus on individualized and optimized therapeutic regimens of antiplatelet drugs.
Acknowledgments
We are grateful to Dr Xinjian Yang, Dr Chuhan Jiang, Dr Aihua Liu, Dr Peng Jiang, Dr Shiqing Mu and Dr Ming Lv for the endovascular treatment and data.
References
Footnotes
Contributors Conception and design: HG, YJ and YL. Acquisition of data: HG, HJ, PL and HH. Data analysis and interpretation: HG, XL, YL and HR. Drafting the article: HG. Final approval of the version: all authors.
Funding This work was supported by the National Natural Science Foundation of China, grant number (81471166 and 81171078).
Competing interests None declared.
Patient consent Obtained.
Ethics approval Institutional review board approval was obtained from Beijing Tiantan Hospital's ethics committee for this study.
Provenance and peer review Not commissioned; externally peer reviewed.