Variability in response to clopidogrel: how important are pharmacogenetics and drug interactions?

Abstract

Clopidogrel is a pro-drug which is converted to an active metabolite that selectively blocks ADP-dependent platelet activation and aggregation. The main enzyme responsible for activating clopidogrel is the cytochrome P450 (CYP) isoenzyme CYP2C19, which is polymorphic. There is a growing body of literature showing that carriers of variant CYP2C19 alleles have impaired ability to metabolize clopidogrel (i.e. poor metabolizers), which is associated with decreased inhibition of platelet aggregation and increased cardiovascular risk. Some proton pump inhibitors are also metabolized by the CYP2C19 enzyme and often given together with clopidogrel to reduce gastrointestinal side effects. In particular, omeprazole has been shown to inhibit the CYP-mediated metabolism of clopidogrel, and some studies have shown that the combination was associated with a higher incidence of cardiovascular adverse reactions than clopidogrel given alone. However, a recent randomized controlled trial demonstrated no significant difference in adverse cardiovascular events for patients on the combination of clopidogrel and omeprazole compared with clopidogrel alone. This current review aims to summarize the role of pharmacogenetics and drug interactions in determining variability in response to clopidogrel.

Introduction

Clopidogrel is an oral thienopyridine derivative which inhibits platelet activation and aggregation by irreversibly blocking the platelet adenosine diphosphate (ADP) P₂Y₁₂ receptor [1]. In patients with atherosclerotic vascular disease, clopidogrel was superior to aspirin in terms of preventing ischaemic stroke, myocardial infarction (MI) or vascular death, with a relative risk reduction of 8.7% (P = 0.043) [2]. Addition of clopidogrel to aspirin (dual antiplatelet therapy) also significantly improved cardiovascular outcomes in patients with acute coronary syndrome (ACS) [3–5] and those undergoing percutaneous coronary intervention (PCI) [6–8]. Despite the well documented benefits of clopidogrel, a substantial number of patients still experience recurrent major adverse cardiovascular events.

Significant inter-individual variability in antiplatelet response to clopidogrel has been observed [9]. Various potential mechanisms have been proposed and it is likely to be multi-factorial [10]. Patients demonstrating poor antiplatelet response have been labelled as ‘clopidogrel resistant’ or ‘clopidogrel non-responders’. Other nomenclatures like ‘low-responders’, ‘suboptimal-responders’, ‘hypo-responders’ and ‘semi-responders’ have also been used in the literature. ‘Clopidogrel resistance’ is still a poorly defined and somewhat confusing entity due to the lack of universally accepted definition, lack of standardized laboratory method for platelet function testing and the use of different assays and agonists in various studies. As a result, the reported incidence of clopidogrel resistance varied widely from 4% to 30% [11].

One significant source of the variability in clopidogrel responsiveness comes from the pharmacokinetics of clopidogrel. Clopidogrel is a pro-drug, which requires metabolism by the hepatic cytochrome P450 (CYP) enzymes to form the active thiol metabolite. The main enzyme responsible for activating clopidogrel is CYP2C19. Both genetic (e.g. polymorphism) and acquired (e.g. drug–drug interactions) alteration of CYP2C19 may affect the efficacy of clopidogrel metabolism [12]. In particular, carriers of loss-of-function CYP2C19 alleles have impaired metabolism of clopidogrel, which results in decreased inhibition of platelet aggregation and increased cardiovascular risk. In March 2010, the US Food and Drug Administration (FDA) added a ‘box warning’ that some patients may be poor metabolizers of clopidogrel due to low CYP2C19 activity [13]. Tests are available to identify patients with genetic polymorphisms, and use of other antiplatelet agents (e.g. prasugrel and ticaglelor) or alternative dosing strategies should be considered for poor metabolizers.

Another area of intense debate is the possible interaction between clopidogrel and proton pump inhibitors (PPI), which are often prescribed together with clopidogrel to reduce the risk of gastrointestinal bleeding. Omeprazole, in particular has been shown to inhibit the CYP-mediated metabolism of clopidogrel, and some studies have shown that combination of clopidogrel and PPI was associated with a higher incidence of adverse cardiovascular events. However, this was largely based on observational studies and hence subjected to various confounding factors. Besides, patients receiving PPI were generally older with more comorbid illnesses and hence intrinsically more susceptible to recurrent cardiovascular events. The current review will summarize the impact of pharmacogenetics and drug interactions on variability in response to clopidogrel.

Measurement of the antiplatelet effect of clopidogrel

Platelet function testing plays a critical role in determining clopidogrel responsiveness. There are two categories of tests to monitor effects of clopidogrel. The vasodilator-stimulated phosphoprotein (VASP) phosphorylation assay is specific to the platelet inhibitory effects of clopidogrel via signalling through P₂Y₁₂. An alternative approach is to use ADP as the stimulus and observe one of a number of end points, including turbidimetric platelet aggregometry, impedance platelet aggregometry, the VerifyNow P₂Y₁₂ assay (Accumetrics, San Diego, CA), Plateletworks (Helena Laboratories, Beaumont, TX), platelet surface-activated GP IIb/IIIa, platelet surface P-selectin, leucocyte-platelet aggregates, the TEG Platelet Mapping system (Haemoscope, Niles, IL) and the Impact cone and plate(let) analyzer (Siemens Healthcare Diagnostics, Inc., Deerfield, IL). However, correlations between assays are modest, and concordance in defining suboptimal response is poor [14].

Pharmacogenetics and variability of clopidogrel response

CYP2C19 loss-of-function polymorphism is associated with clopidogrel resistance

CYP2C19 is the main liver enzyme responsible for formation of the active thiol metabolite of clopidogrel. CYP3A4, CYP3A5, CYP2C9 and CYP2B6 are also involved. Genetic polymorphism of CYP2C19 can lead to significant phenotypic variation in its activity and hence affect the metabolism of clopidogrel. Hulot et al. first demonstrated that young healthy subjects carrying the *2 allelic variant (i.e. CYP2C19 *1*2 heterozygotes) had impaired responsiveness to clopidogrel compared with the CYP2C19 wild-type homozygotes (*1*1) [15]. Similar studies involving healthy individuals from other ethnic groups have shown consistent findings [16–18]. The CYP2C19*2 allele was also associated with higher platelet aggregability and residual platelet reactivity (RPA) in patients with ACS [19, 20], as well as those undergoing PCI [21]. Current evidence suggests there are three major variants of CYP2C19. The CYP2C19*1 allele has fully functional metabolism of clopidogrel whereas the CYP2C19*2 and CYP2C19*3 alleles have no functional metabolism. These two alleles account for most of the loss-of-function alleles in Caucasian (85%) and Asian (99%) patients classified as poor metabolizers [22, 23]. The CYP2C19 *4, *5, *6, *7, *8 and other alleles are less common, but may be associated with absent or reduced metabolism of clopidogrel.

CYP2C19 loss-of-function polymorphism is associated with worse clinical outcome

There is ample evidence that CYP2C19 polymorphism is associated with reduced antiplatelet response to clopidogrel, and high RPA has been shown to be associated with worse clinical outcomes in acute MI [24] and post-PCI patients [25–27]. Studies on the direct effect of possession of loss-of-function CYP2C19 alleles on clinical outcome are summarized in the following section.

EXCELSIOR (Impact of Extent of Clopidogrel-Induced Platelet Inhibition during Elective Stent Implantation on Clinical Event Rate)

In this prospective, single centre study from Germany, 797 patients undergoing elective PCI and treated with aspirin plus clopidogrel were followed up for 1 year [21]. Residual platelet activity was measured before discharge from hospital. High pre-discharge RPA was associated with a 3-fold increased risk of death and MI at 1 year, as well as a nearly 8-fold increased risk of death and MI in patients who received at least one drug-eluting stent (DES). Although the presence of the CYP2C19*2 allele was associated with significantly higher RPA and high RPA was associated with worse clinical outcome, the investigators were unable to demonstrate a direct impact of CYP2C19*2 on clinical outcome. Five patients (2.0%) reached the endpoint among carriers of the CYP2C19*2 allele, compared with 19 patients (3.4%) among wild-type homozygotes (P = 0.371). Of note, only 280 out of the 797 patients received DES. In this small subgroup, the event rate was 3.3% among carriers of the *2 allele and 2.1% among wild-type homozygotes (P = 0.684). The authors attributed this to the fact that the current study was not powered to address this issue with the given sample size.

AFIJI (Appraisal of risk Factors in young ischaemic patients Justifying aggressive Intervention)

In this on-going prospective, multicentre registry from France, 378 young patients (aged 18 to 45 years) with acute MI were followed up 6 monthly since 1996 [28]. Among these patients, 259 received clopidogrel treatment for at least 1 month (median 1.07 years) and were tested for the CYP2C19*2 allele. In this study, 72% of patients were wild-type homozygotes (*1*1), 25% were heterozygotes (*1*2) and 3% were homozygotes (*2*2). Overall, CYP2C19*2 carriers had a 3.7-fold increased risk of cardiovascular death, non-fatal MI, or urgent revascularization, mainly due to a significant increase in the rate of non-fatal MI. The risk of stent thrombosis was also significantly increased by 6-fold in carriers.

TRITON-TIMI 38 (Trial to Assess Improvement in Therapeutic Outcomes by Optimizing Platelet Inhibition with Prasugrel-Thrombolysis in Myocardial Infarction)

Mega et al. first performed a pharmacokinetic study and demonstrated that in healthy subjects treated with clopidogrel, carriers of at least one CYP2C19 reduced-function allele had 32.4% relative reduction in plasma exposure to the active metabolite of clopidogrel as well as 9% absolute reduction in maximal platelet aggregation, as compared with non-carriers [29]. The investigators then analyzed the data from the TRITON-TIMI 38 study [30], which compared prasugrel with clopidogrel in patients with ACS and scheduled to have PCI. Among clopidogrel-treated subjects in TRITON-TIMI 38, carriers had a 53% higher risk of death from cardiovascular causes, MI or stroke, as compared with non-carriers (hazard ratio, HR 1.53, P = 0.01) and more than 3-fold increase in the risk of stent thrombosis (HR 3.09, P = 0.02).

RECLOSE (Low Responsiveness to Clopidogrel and Sirolimus- or Paclitaxel-Eluting Stent Thrombosis)

In this study from Italy, Giusti et al. evaluated the effect of the CYP2C19 polymorphism on the occurrence of stent thrombosis and the composite endpoint of cardiac mortality and stent thrombosis in 772 patients undergoing PCI at 6 months [31]. CYP2C19*2 polymorphism was shown to be an independent risk factor for stent thrombosis (HR 3.43, P = 0.047) or the composite of stent thrombosis and cardiac mortality (HR 2.70, P = 0.049). The RECLOSE study also showed that the CYP2C19*2 polymorphism was an independent risk factor for stent thrombosis or the composite end point of cardiac mortality and stent thrombosis when the effect of the polymorphism was evaluated in a model adjusted for high on-clopidogrel platelet reactivity evaluated by ADP induced platelet aggregation (ADP-RPR), traditional cardiovascular risk factors, clinical and procedural risk factors for stent thrombosis (such as chronic total occlusion, multivessel disease, bifurcation lesion, myocardial infarction, total stent length, left ventricular ejection fraction). These data suggested that the mining of the CYP2C19*2 polymorphism as a risk factor for adverse cardiovascular events is only partially linked to its association with the ADP-RPR phenotype observed in the acute periprocedural phase, and that other genetic and acquired determinants of ADP-RPR besides the CYP2C19*2 polymorphism might have a role in determining the clinical outcome in these high risk vascular patients.

FAST-TIMI (French Registry of Acute ST-Elevation and Non–ST-Elevation Myocardial Infarction)

Simon et al. assessed the relationship of various allelic variants of genes to the risk of death from any cause, non-fatal stroke or MI during 1 year of follow-up in 2208 patients presenting with acute MI in a nationwide French registry [32]. Patients carrying any two CYP2C19 loss-of-function alleles (*2, *3, *4, or *5) had a significantly higher event rate than patients with none (HR 1.98). Among the 1535 patients who underwent PCI during hospitalization, the rate of cardiovascular events among patients with two CYP2C19 loss-of-function alleles was 3.58 times higher than those with none.

Other studies

Sibbing et al. studied the impact of the CYP2C19 loss-of-function polymorphism and the occurrence of stent thrombosis in another large study in France involving 2485 patients undergoing PCI [33]. The incidence of definite stent thrombosis within 30 days following PCI was significantly higher in CYP2C19*2 allele carriers than CYP2C19 wild-type homozygotes (HR 3.81, P = 0.007). The risk of stent thrombosis was highest in CYP2C19 *2*2 homozygotes. Shuldiner et al. also found that patients with the CYP2C19*2 variant were more likely to have a cardiovascular ischaemic event or death during 1 year of follow-up after PCI (HR 2.42, P = 0.02) [34].

From these studies, the magnitude of increased cardiovascular risk ranged from 53% in the TRITON-TIMI 38 study to nearly 6-fold in the AFIJI study. However, it should be noted that none of the studies was randomized and thus the possibility of bias and confounding variables cannot be excluded. In a recent meta-analysis involving 9685 patients (91.3% underwent PCI and 54.5% had ACS), carriers of reduced-function CYP2C19 alleles had a significantly increased risk of cardiovascular death, MI or stroke, as compared with non-carriers [35]. Overall, carriers of one reduced-function allele had a 1.55-fold and carriers of two reduced-function alleles had a 1.76-fold increased risk. The risk of stent thrombosis was also significantly higher in both carriers of one (HR 2.67) and two (HR 3.97) reduced-function alleles. In another meta-analysis involving 8043 patients with coronary artery disease treated with clopidogrel, Sofi et al. showed that CYP2C19*2 polymorphism was associated with a 1.96-fold increased risk of major adverse cardiovascular events (P = 0.02) [36].

Other potential mechanisms of genetic variability of clopidogrel

The effect of polymorphisms including CYP3A4, CYP3A5 and CYP2B6 are less well established. Polymorphisms in the genes encoding P-glycoprotein (an efflux transporter) and purinergic receptor P₂Y₁₂ (the active site for clopidogrel) have been studied. Current evidence suggests the adenosine 5′-triphosphate-binding cassette gene, ABCB1, may partly contribute to variability in clopidogrel responsiveness by affecting the oral bioavailability of clopidogrel [37]. Recently, another enzyme called paraoxonase-1 (PON1) was shown be another crucial enzyme for activation of clopidogrel, with the rate of activation governed by the Q192R polymorphism [38]. In patients with coronary artery disease who underwent stent implantation and received clopidogrel therapy, PON1 QQ192 homozygous individuals had significantly lower PON1 plasma activity, lower plasma concentrations of clopidogrel active metabolite, lower platelet inhibition and higher risk of stent thrombosis than RR192 homozygous individuals. The potential therapeutic implications of PON1 warrents further investigation.

Role of pharmacogenetic testing

The aim of pharmacogenetic testing is to identify patients at risk of poor response to standard dosage of clopidogrel such that alternative treatment strategies or dosage escalation can be initiated earlier. This will hopefully minimize the risk of adverse cardiovascular events including cardiovascular mortality, recurrent ischaemic events and stent thrombosis. However, the latest FDA recommendation neither mandated nor explicitly recommended CYP2C19 genetic testing in patients prescribed clopidogrel and only suggested that health professionals should be aware that tests are available to determine patients' CYP2C19 status.

Numerous commercial pharmacogenetic tests for genotype variants in CYP2C19 are available (Table 1). These tests differ in their genotyping methodology, specimen required (e.g. whole blood, saliva, buccal swab) and availability. Most of these tests cover the most common alleles (*1, *2 and *3). Patients can then be categorized into extensive metabolizers (e.g. *1*1), intermediate metabolizers (e.g. *1*2) and poor metabolizers (e.g. *2*2 or *2*3). A new allele, CYP2C19*17, has been described which is associated with increased enzymatic activity and patients who are *1*17 may be regrded as ultra-rapid metabolizers.

Table 1.

Examples of pharmacogenetic tests for clopidogrel response

Test	Company	CYP2C19 variants included	Sample type	Methodology
AccuType™ CP	Quest Diagnostics	*1, *2, *3, *4, *5	Whole blood or saliva	Polymerase chain reaction Single nucleotide primer extension reaction
Plavitest	Genelex Corporation	*1, *2, *3, *4, *5, *6, *7, *8, *17	Whole blood or sterile buccal swabs	Extended CYP2C19 DNA mutation panel
Medications panel	Navigenics, Inc.	Unspecified	Saliva	TaqMan® SNP Genotyping assays
Clopidogrel efficacy	23andMe	*1, *2, *3, *4, *8, and *17	Saliva	Polymerase chain reaction Single nucleotide primer extension reaction
Drug response (Medication)	Pathway Genomics	Unspecified	Saliva	DNA testing
Clopidogrel activation	Matrix Genomics	*1, *2, *3, *4, *5, *6, *7, *8, *9, *10, *17	Buccal swab	DNA testing
Cytochrome P450 2C19 (CYP2C19) 10 mutations	ARUP Laboratories	*2, *3, *4, *5, *6, *7, *8, *9, *10, *17	Whole blood	Polymerase chain reaction and detection primer extension
CYP450 2C19 (Plavix™) Gene test/Genetic physician consult	MyMedLab, Inc.	Unspecified	Buccal swab	DNA testing
Clopidogrel genetic test	TheranostiCs Lab	Unspecified	Buccal swab	DNA testing
Verigene® CYP2C19 XP nucleic acid test	Nanosphere	*1A, *2, *3, *4, *5A, *6, *7, *8, *9, *10 and *17	Unspecified	DNA testing
Verigene CYP2C19 nucleic acid test	Nanosphere	*1A, *2A, *3	Unspecified	DNA testing

ACCF/AHA recommendations

Recently, the American College of Cardiology Foundation (ACCF) and the American Heart Association (AHA) published a Clinical Alert in response to the FDA's new box warning [39]. It suggests that information regarding the predictive value of pharmacogenetic testing is still very limited and the current evidence base is insufficient to recommend either routine genetic or platelet function testing at the present time. This is because data concerning cross validation of the techniques, the corresponding reliability, specificity and reproducibility are still limited. Neither is there any convincing evidence to suggest routine pharmacogenetic testing improves clinical outcomes. The additional cost involved is another concern. Therefore, clinical judgment is required to assess clinical risk and variability in patients considered to be at increased risk. Genetic testing to determine if a patient is a poor metabolizer may be considered before starting clopidogrel therapy in patients believed to be at moderate or high risk for poor outcomes, like patients undergoing elective high-risk PCI procedures (e.g. treatment of extensive and/or very complex disease).

Platelet function testing is another possible option to select patients with suboptimal antiplatelet response. However, evidence regarding this aspect is still very limited. The latest AHA guideline suggests that only in patients in whom subacute thrombosis may be catastrophic or lethal (unprotected left main, bifurcating left main, or last patent coronary vessel), platelet aggregation studies may be considered and the dose of clopidogrel increased to 150 mg day^–1 if less than 50% inhibition of platelet aggregation is demonstrated (Class IIb recommendation, level of evidence: C) [40].

Currently, there is still a lack of prospective studies which evaluate the effect of different antiplatelet treatments tailored to individual characteristics of patients, such as genetic profile, residual platelet reactivity, drug–drug interactions, and traditional and procedural risk factors on clinical outcomes. Such kinds of studies are urgently needed. Several studies, such as GRAVITAS (NCT00645918), DANTE (NCT00774475), TRIGGER-PCI (NCT00910299) and ARCTIC (NCT00827411) are being conducted to evaluate the merit of changing therapy on the basis of platelet reactivity and their results are eagerly awaited.

Drug interactions and clopidogrel resistance

Another potential mechanism of clopidogrel resistance relates to the competitive inhibition of CYP isoenzymes required for activation of clopidogrel to its thiol metabolite. Various drugs have been implicated, including statins, calcium channel blockers and PPIs. Current evidence suggests that statins can be safely used together with clopidogrel and evidence regarding negative interactions with calcium channel blockers is limited [13]. However, the potential interaction between PPIs and clopidogrel has been an area of intense debate.

Bleeding complications are the major adverse effect associated with clopidogrel. While intracanial bleeding carries the highest mortality, gastrointestinal haemorrhage is the most frequent site of bleeding. Bleeding risk of dual antiplatelet therapy is substantially higher than monotherapy with aspirin or clopidogrel and the AHA and American College of Gastroenterology guidelines recommend empirical use of PPIs in patients receiving dual antiplatelet therapy [41]. Besides, PPIs are often used for other indications such as peptic ulcer disease and gastro-oesophageal reflux disease. The majority of PPIs are metabolized by the CYP2C19 isoenzyme. Omeprazole is predominantly metabolized by this pathway followed by esomeprazole, pantoprazole and lansoprazole. Rabeprazole is the PPI least metabolized via CYP2C19 and the majority of its metabolism is via a non-enzymatic pathway [42]. As a result, PPIs may in theory dampen the antiplatelet effect of clopidogrel by competitive antagonism and lead to increased cardiovascular risk.

Studies showing negative interactions between PPIs and clopidogrel

Gilard et al. first reported in an in vivo study that in patients receiving aspirin and clopidogrel for coronary angioplasty, platelet reactivity as determined by the VASP phosphorylation test was significantly higher in patients on concurrent omeprazole therapy [43]. Results of this observational study served as a background for the subsequent OCLA (Omeprazole CLopidogrel Aspirin) study [44]. In this study, 124 patients on dual antiplatelet therapy after PCI were randomized to omeprazole 20 mg daily or placebo for 1 week. The clopidogrel effect, expressed as platelet reactivity index (PRI) was tested on days 1 and 7 in both groups by the VASP phosphorylation test. Omeprazole significantly decreased the antiplatelet effect of clopidogrel. This was the first randomized study showing that omeprazole had a detrimental effect on the antiplatelet effect of clopidogrel. However, it should be noted that this study employed platelet reactivity index as a surrogate marker for clopidogrel efficacy. Also, the study was conducted for only 7 days. In clinical practice, patients are usually given at least 1 month of dual antiplatelet therapy for bare metal stents and 1 year for drug eluting stents. Nevertheless, this study did provide important information that the use of omeprazole during the early stages of PCI may lower the efficacy of clopidogrel, which theoretically may increase the risk of stent thrombosis and adverse cardiovascular outcome. Compared with omeprazole, pantoprazole and esomeprazole seemed not to attenuate the effect of clopidogrel in post-PCI patients in early observational studies [45, 46]. However, these studies provided very limited insight into whether such a reduced antiplatelet response could translate into a worse clinical outcome.

There were at least five large retrospective studies showing that concomitant PPI use with clopidogrel was associated with adverse cardiovascular outcomes [47–51]. A few of these studies deserve further discussion. The Clopidogrel Medco Outcomes Study was a large retrospective cohort study which followed more than 16 000 post-PCI patients on clopidogrel therapy for 1 year [48]. Patients were at least 80% compliant with clopidogrel and on PPIs for an average of 9 months. It compared clinical outcomes of patients taking pantoprazole, esomeprazole, omeprazole and lansoprazole. Early results were published in abstract form and showed that in patients without a preceding cardiovascular event, the use of PPIs with clopidogrel was associated with a 1.79-fold (adjusted) increased risk of hospitalization for stroke, MI, angina or coronary artery bypass grafting (CABG). A more pronounced effect was seen among patients with a preceding cardiovascular event (adjusted HR 1.86). The final results of this study are not yet published, but will certainly have important implications.

In a multi-centre, retrospective cohort involving 8205 ACS patients on clopidogrel, Ho et al. found that clopidogrel plus a PPI was associated with significantly higher risk of hospitalization for recurrent ACS and revascularization procedures, but not for all-cause mortality [50]. The adjusted odds ratio was 1.25 for death or rehospitalization, 1.86 for hospitalization because of recurrent ACS and 1.49 for any revascularization procedures. It should be noted that in Ho et al.'s study, the PPI group included a statistically significantly higher rate of chronic obstructive pulmonary disease (COPD), diabetes, previous MI, congestive heart failure, liver and renal disease. After adjustment of these factors, the hazard ratio was only modestly increased (1.25). Other potentially important but unknown confounders may not be adjusted and jeopardize the interpretation. In another retrospective cohort, Juurlink et al. found current use of PPIs (defined as use of PPI within 30 days) was associated with an increased risk of re-infarction among 734 patients re-admitted with recurrent MI within 90 days after index acute MI (adjusted HR 1.27) [51]. No association was found with more distant exposure to PPIs. Pantoprazole, which does not inhibit CYP2C19, was not associated with re-admission for myocardial infarction. It should be noted in this study, however, that there was a significant difference in baseline use of cardio-protective agents such as angiotensin converting enzyme inhibitors, β-adrenoceptor blockers and statins. Important cardiovascular risk factors like smoking status, blood pressure control as well as lipid profile were not known. Ethnicity was not matched either, which has important bearing given the variability in genetic expressions of CYP2C19.

Studies showing no interactions between PPIs and clopidogrel

These results have certainly spawned concerns regarding the negative interaction between PPI and clopidogrel, and also led to an earlier FDA warning that specifically discouraged concurrent use of clopidogrel and omeprazole. However, some other studies failed to demonstrate such a relationship. For instance, both the FAST-MI [32] and AFIJI [28] study found no association between omeprazole use (the predominant PPI studied) and adverse cardiovascular outcomes. A few important issues need to be addressed before attributing any observed adverse clinical outcome to the negative interaction between PPIs and clopidogrel. In clinical practice, patients receiving PPIs as ulcer prophylaxis tend to have a higher risk of bleeding. One explanation is that these patients tend to be older and have more medical comobidities, rendering them more susceptible to bleeding events. The adverse events observed in retrospective studies may just reflect an intrinsically higher risk of adverse cardiovascular events in this particular group of patients. Another possibility is that PPIs may have a direct negative impact on the cardiovascular outcome. Dunn et al. showed from the data of the CREDO (Clopidogrel for the Reduction of Events During Observation) study that PPI use was independently associated with the 28 day incidence of death or MI or urgent revascularization (HR 1.6, P = 0.022) and 1 year incidence of death or MI or stroke (HR 1.5, P = 0.012) in the overall trial population [52]. Recently, registry data from Denmark also showed that use of a PPI in patients who did not receive clopidogrel was associated with a 1.29-fold higher risk of rehospitalization for MI or stroke or cardiovascular death in 1 year [53]. These results have further complicated the matter. Last but not least, most PPIs have short plasma half-lives of elimination at approximately 1 h. Being weak bases, they rapidly accumulate within the parietal cell secretory canaliculi and are unlikely to accumulate elsewhere in the body. Since clopidogrel and PPIs are both very rapidly metabolized by the CYP450 system, the chance of interaction would appear to be minimal from the pharmacological point of view [54].

O'Donoghue et al. [55] analyzed the populations from two randomized trials, the PRINCIPLE-TIMI 44 trial [56] and the TRITON -TIMI 38 trial [30]. The PRINCIPLE-TIMI 44 trial was a double-blind, two-phase crossover study that randomized 201 patients undergoing cardiac catheterization with planned PCI to either prasugrel (60 mg loading dose, 10 mg daily maintenance dose) or to high dose clopidogrel (600 mg loading dose, 150 mg daily maintenance dose). In this study, among patients treated with clopidogrel, the mean inhibition of platelet aggregation was significantly lower in those treated with a PPI at 2 h, 6 h and 18–24 h than in those not on a PPI. After 15 days of maintenance therapy with clopidogrel 150 mg daily, patients on a PPI still tended to have a lower mean inhibition of platelet aggregation (P = 0.06). In the TRITON-TIMI 38 trial, in patients randomly assigned to clopidogrel, no significant association was found between use of a PPI and risk of the primary endpoint after adjustment for potential confounders and the propensity to treat with a PPI (HR 0.94, 95% CI 0.80, 1.11, P = 0.46). Use of a PPI was not associated with increased risk of MI, stent thrombosis, or a decreased risk of bleeding. The Clopidogrel and the Optimization of Gastrointestinal Events (COGENT) trial was the first randomized clinical study of the effect of the interaction between clopidogrel and omeprazole on cardiovascular and gastrointestinal outcomes [57]. In a double-blind fashion, patients with ACS or undergoing PCI were randomized to receive a fixed-dose combination pill containing either clopidogrel and delayed-release omeprazole or clopidogrel alone. All patients received aspirin. Unfortunately, the trial was stopped early because the sponsor declared bankruptcy. However, preliminary results revealed no significant difference in cardiovascular outcomes for patients on clopidogrel and omeprazole compared with clopidogrel alone (4.9% in the omeprazole group vs. 5.7% in the placebo group, P = 0.96). Furthermore, adverse gastrointestinal events were significantly fewer in patients on clopidogrel and omeprazole (1.1% with omeprazole vs. 2.9% with placebo, P < 0.001). Thus, omeprazole appears to be safe and may offer gastrointestinal protection to patients on dual antiplatelet therapy.

In a recent meta-analysis on the use of PPIs and major adverse cardiovascular events (MACE) in patients treated with clopidogrel, which involved 13 studies and 48 674 patients, Hulot et al. showed that the impact of PPI use was significant only in high-risk patients [58]. In patients with an annual rate of MACE less than 10%, PPI use had no significant influence (HR 1.01, P = 0.84). In contrast, in patients with an annual rate of MACE more than 10%, the effect of PPI use was highly significant (HR 1.49, P < 0.001). In this meta-analysis, PPI users also had an increased risk for major adverse cardiovascular events as compared with nonusers (HR 1.41, P < 0.001). Whether the detrimental effect of PPI use in high-risk patients is related to CYP2C19 inhibition or other mechanisms are involved warrants further evaluation and a causative link between PPI use and clopidogrel response cannot be inferred from this meta-analysis.

ACCF/ACG/AHA recommendations

Based on the current evidence, patients on clopidogrel can continue PPI use if an indication exists. Patients at high risk of gastrointestinal bleeding should be given PPIs as prophylaxis. Although some authors suggest omeprazole can be safely used, others suggest following the FDA recommendation that omeprazole should be avoided until there is clear-cut evidence from larger randomized controlled trials. Up until now, it is still not known if the interaction with clopidogrel is a class effect or specific to certain drugs of this class. It is important, however, that all physicians should balance an individual's gastrointestinal bleeding risks and cardiovascular risks, assess the need for a PPI for each patient, as these drugs are easily overused. The latest ACCF/ACG/AHA 2010 Expert Consensus Document [59] suggests that PPIs are recommended to reduce GI bleeding among patients with a history of upper GI bleeding and are appropriate in patients with multiple risk factors for gastrointestinal bleeding (e.g. advanced age, concomitant use of warfarin, steroids, NSAIDs or H. pylori infection). However, routine use of PPIs is not recommended for patients at lower risk of gastrointestinal bleeding, who have much less potential to benefit from prophylactic therapy.

Conclusion

Clopidogrel resistance is multifactorial and still not entirely understood. There is accumulating evidence that pharmacogenetic factors play an important role. However, routine genetic testing for loss-of-function polymorphism of CYP450 enzymes before clopidogrel therapy is not yet evidence based. In selected cases (patients undergoing elective high-risk PCI procedures), genetic testing to determine if a patient is a poor metabolizer may be considered before starting clopidogrel. Platelet function tests should be restricted to selected cases and currently is mainly reserved for research purposes. Based on the current evidence, PPIs can still be safely used with clopidogrel if an indication exists. We suggest adhering to the latest published clinical guidelines (Table 2).

Table 2.

Current recommendations on pharmacogenetic testing [39] and concomitant use of clopidogrel and PPI [59]

•Information regarding the predictive value of pharmacogenomics testing is very limited at this time and the current evidence base is insufficient to recommend either routine genetic or platelet function testing at the present time.

•Genetic testing to determine if a patient is predisposed to poor clopidogrel metabolism (‘poor metabolizers’) may be considered before starting clopidogrel therapy in patients believed to be at moderate or high risk for poor outcomes. This might include, among others, patients undergoing elective high risk PCI procedures (e.g. treatment of extensive and/or very complex disease).

•PPIs are recommended to reduce gastrointestinal (GI) bleeding among patients with a history of upper GI bleeding and those with multiple risk factors for GI bleeding. However, routine use of PPI is not recommended for patients at lower risk of upper GI bleeding, who have much less potential to benefit from prophylactic therapy.

•Clinical decisions regarding concomitant use of PPIs and thienopyridines must balance overall risks and benefits, considering both cardiovascular and bleeding complications.

•Observational studies and a single randomized clinical trial have shown inconsistent effects on cardiovascular outcomes of concomitant use of thienopyridines and PPI. A clinically important interaction cannot be excluded, particularly in certain subgroups, such as poor metabolizers of clopidogrel.

Competing Interests

There are no competing interests to declare.