Clopidogrel resistance and clopidogrel treatment failure

INTRODUCTION — Clopidogrel, an oral platelet P2Y₁₂ receptor blocker, is used with aspirin in patients who undergo coronary artery stenting or who have an acute coronary syndrome (ACS) to reduce the risk of subsequent cardiovascular events such as stent thrombosis or recurrent ACS. Nonetheless, adverse cardiovascular events occur despite the use of dual antiplatelet therapy and are partly attributable to the variable pharmacodynamics of aspirin and P2Y₁₂ receptor blockers.

This topic will address whether there is a clinical role for screening patients for P2Y₁₂ drug resistance and the management of patients who have clinical events despite clopidogrel therapy.

The issues concerning aspirin nonresponse and resistance are presented separately. (See "Nonresponse and resistance to aspirin".)

The indications for P2Y₁₂ inhibitor use for the treatment of coronary artery disease are discussed elsewhere. (See "Acute ST-elevation myocardial infarction: Antiplatelet therapy", section on 'Patients receiving primary PCI' and "Antithrombotic therapy for elective percutaneous coronary intervention: General use", section on 'P2Y12 receptor blockers' and "Acute non-ST-elevation acute coronary syndromes: Early antiplatelet therapy" and "Long-term antiplatelet therapy after coronary artery stenting in stable patients", section on 'Information for patients'.)

The indications for P2Y₁₂ inhibitor use in stroke are discussed separately. (See "Long-term antithrombotic therapy for the secondary prevention of ischemic stroke", section on 'Clopidogrel'.)

DEFINITIONS

"Nonresponsiveness" to an antiplatelet drug is a pharmacodynamic phenomenon where there is no clinically meaningful change in platelet function after treatment. In studies employing light transmittance aggregometry, a change in maximal aggregation ≤10 percent from baseline, using adenosine diphosphate (ADP) as the agonist, is defined as "resistance." The percent change between pretreatment and on-treatment platelet function (ie, "responsiveness") can be categorized as "nonresponsive" (≤10 percent), "hypo-responsive" (10 to 20 percent), or "responsive" (>20 percent).

Heightened platelet reactivity (HPR), also called high on-treatment platelet reactivity, to ADP during clopidogrel therapy indicates persistent suboptimal response of the P2Y₁₂ receptor (clopidogrel target). It is based on one measurement of on-treatment platelet reactivity. Usually, the cut points of HPR have been linked to clinical risk of thrombosis based on nonrandomized comparisons of stented patients. The evidence to support these cut points is described elsewhere. (See 'HPR and thrombotic events' below.)

Clopidogrel treatment failure is best defined as the occurrence of a thrombotic event/ischemic event during clopidogrel therapy in patients with HPR. Treatment failure may result from patient noncompliance and/or inadequate antiplatelet response to clopidogrel. As multiple signaling pathways mediate platelet activation and the occurrence of thrombotic events, a treatment strategy directed against a single pathway, cannot be expected to prevent the occurrence of all events.

In this context, not all ischemic events in patients taking clopidogrel may be due to lack of clopidogrel pharmacodynamic effect. There are multiple mechanisms involved in the genesis of a thrombotic event that may overcome the pharmacodynamic effects of clopidogrel. Therefore, a patient taking clopidogrel after stenting may have a new ischemic event in the presence of potent response to clopidogrel as measured by a platelet function assay.

PREVALENCE — It has been estimated that between 16 and 50 percent of patients treated with clopidogrel have high on-treatment platelet reactivity (HPR) [1]. HPR is dependent on the cut-off value used for each assay: the higher the cut-off, the lower the percent of patients with HPR. In addition, poor biologic response is determined by factors other than HPR. (See 'Potential explanations' below.)

Clopidogrel response variability was initially demonstrated by measuring adenosine diphosphate-induced (ADP) platelet aggregation, as well as p-selectin and activated GP IIb/IIIa expression, at baseline and serially for 30 days after stenting in patients treated with a 300 mg clopidogrel load followed by 75 mg daily therapy. Some patients had no demonstrable antiplatelet effect; when the absolute difference between pre- and post-treatment platelet aggregation was ≤10 percent, patients were labelled regarded as "resistant" [2]. The prevalence of resistant patients was 31 percent at day five post-stenting, and it fell to 15 percent at day 30. (figure 1).

In addition, the level of platelet reactivity early after a standard clopidogrel regimen for coronary stenting was directly related to the pre-treatment reactivity. It has been reported that patients treated with clopidogrel (or prasugrel) who were hyper-responsive to ADP before treatment were likely to be hyper-responsive to ADP during platelet P2Y₁₂ receptor blocker treatment [3,4].

A reduction in platelet reactivity over time (after initiation of therapy) has been observed in some [5-8], but not all [9], studies. For example, two studies have shown a decrease in the prevalence of high on-treatment platelet reactivity of up to 50 percent at day 30 compared to day one post-stenting [2,6,7]. The reason for this time-dependent decrease is not clear.

POTENTIAL EXPLANATIONS — Clopidogrel response variability has been attributed to variability in active metabolite generation that is caused by: variable absorption, which is influenced by an ABCB1 gene polymorphism [1,10]; functional variability in CYP isoenzyme activity, which is influenced by single nucleotide polymorphisms (SNPs); and drug-drug interactions. The latter two are discussed below.

Variation in clopidogrel metabolism — Clopidogrel is a pro-drug that requires conversion into an active metabolite for biologic activity after oral administration. Approximately 85 percent of absorbed clopidogrel is hydrolyzed by human carboxylesterase-1 into an inactive carboxylic acid metabolite, and 15 percent is metabolized into an active metabolite by hepatic cytochrome P450 (CYP). Hepatic biotransformation involves a two-step oxidative process. In the first step, the thiophene ring of clopidogrel is oxidized to form 2-oxo-clopidogrel by CYP2C19, CYP1A2, and CYP2B6. In the second step, CYP3A4, CYP2B6, CYP2C19, and CYP2C9 catalyze the formation of the active metabolite (R-130964). It has been proposed that CYP2C19 is the major enzyme involved in the generation of clopidogrel active metabolite [11].

Among several polymorphisms of CYP2C19, the two most frequent variants associated with loss-of-function (LoF SNPs) are CYP2C19*2, a G681A mutation in exon 5 resulting in an aberrant splice site leading to the production of a truncated, nonfunctioning protein and CYP2C19*3, a G636A mutation in exon 4 resulting in a premature stop codon [12-15]. The *17 variant is associated with increased gene transcription and increased enzyme function. The prevalence of LoF SNPs varies with race; the prevalence in Eastern Asian populations is up to 65 percent, whereas it is approximately 30 percent in White populations [16].

Interaction with other drugs — Clopidogrel metabolism is potentially influenced by concomitantly administered agents, such as proton pump inhibitors (PPIs), calcium channel blockers (CCBs), or warfarin, which inhibit or enhance CYP activity or compete with clopidogrel during hepatic cytochrome P450-mediated metabolism [1]. Although the influence of these interactions on clopidogrel pharmacokinetics and pharmacodynamics has been reported, no prospective study has conclusively demonstrated a clinically important effect of these drugs in patients treated with clopidogrel.

In addition, demographic variables such as age, renal failure, diabetes, and body mass index also influence the platelet response to adenosine diphosphate (ADP) by either directly affecting platelet function or by affecting clopidogrel metabolism. The final platelet reactivity phenotype and clinical outcomes of patients treated with clopidogrel is the result of all of these influences [1].

●Drugs without clinically meaningful interaction – Some drugs could cause an interaction with clopidogrel, but trials or studies do not support a clinically important interaction. These drugs include:

•Statins – Although some studies have suggested a possible association between lipid soluble statin administration and pharmacodynamic effects of clopidogrel, the clinical relevance of these pharmacodynamics findings has not been demonstrated. Although such a relationship may be explained by the fact that some statins are metabolized by some of the same pathways as clopidogrel, the totality of evidence suggests that clinicians need not preferentially prescribe any particular statin with coadministration of clopidogrel [17-19]. (See "Statins: Actions, side effects, and administration", section on 'Drug interactions'.)

•Proton pump inhibitors – In most patients, we believe that PPIs and clopidogrel can be used together. We agree with the 2009 US Food and Drug Administration notice that suggests that patients who take clopidogrel should consult with their clinician if they are taking a PPI or considering PPI use [20,21]. Despite the known pharmacodynamic interaction between PPIs and clopidogrel, which decreases the antiplatelet effect of clopidogrel, the available evidence suggests that PPIs do not decrease the clinical efficacy of clopidogrel [22-26].

Examples of studies supporting this approach include the following:

-In a systematic review of published data, simultaneous use of a P2Y₁₂ inhibitor and a PPI did not significantly increase the risk of major adverse cardiovascular events (4.5 versus 4.7 percent in the PPI group; relative risk [RR] 0.99, 95% CI 0.76-1.28) [26].

-In one of the largest trials in the meta-analysis, 3761 patients treated with clopidogrel were randomly assigned to either omeprazole or placebo [24]. There was no significant difference in cardiovascular events (4.9 versus 5.7 percent with placebo; hazard ratio [HR] 0.99, 95% CI 0.68-1.44) and a lower rate of gastrointestinal bleeding in patients assigned to omeprazole (1.1 versus 2.9 percent with placebo; HR 0.34, 95% CI 0.18-0.63).

●Calcium channel blockers – CCBs inhibit CYP3A4, and there has been a concern that they may decrease the efficacy of clopidogrel. Some [27,28], but not all [29-31], studies have suggested a possible deleterious impact of CCB on clopidogrel efficacy. Based on the current totality of evidence, we do not recommend the restriction of CCB use in patients who are prescribed clopidogrel.

●Drugs with uncertain clinical interaction – Examples of drugs that can alter the metabolism of clopidogrel but whose clinical interaction is not well-established include:

Drugs that may increase the activity of clopidogrel by inducing CYP P450 enzymes:

•Rifampin [32,33]

•St. John’s wort [34]

Drugs that may decrease the activity of clopidogrel by inhibiting CYP P450 enzymes:

•Polyunsaturated omega-3 fatty acids [35]

•Erythromycin, troleandomycin

•Ketoconazole

Diabetes mellitus — Patients with diabetes mellitus have increased platelet activation [5,36-38] and a higher percent of circulating immature platelets [39]. These characteristics could, in theory, counteract the inhibitory effect of clopidogrel (or aspirin) on platelets [37,40-43]. The totality of available evidence is incomplete and inconsistent [44-48].

It has been postulated that patients with diabetes mellitus may require increased doses of one or both of these agents in order to show an optimal therapeutic effect [49], or may be candidates for a more potent platelet inhibitor (eg, prasugrel [50,51], ticagrelor [52-55]). (See "Prevalence of and risk factors for coronary heart disease in patients with diabetes mellitus", section on 'Platelet activation' and "Platelet biology and mechanism of anti-platelet drugs" and "Platelet biology and mechanism of anti-platelet drugs", section on 'Platelet receptors and their agonists'.)

Renal insufficiency — There is no convincing evidence chronic kidney disease (CKD) significantly influences platelet response to clopidogrel [56,57].

Noncompliance — Noncompliance with clopidogrel therapy is common and offers a plausible explanation for the lack of clinical benefit [15,58,59].

Smoking — In subgroup analyses of randomized trials of clopidogrel therapy, nonsmokers taking clopidogrel had less clinical benefit than did smokers taking clopidogrel. This effect is referred to as a "smoker's paradox" [60-63]. The stimulation of CYP1A2 associated with smoking enhances platelet inhibition by clopidogrel; smoking cessation reverses this effect [28,29,42,60-62,64-67]. Although this subgroup finding may be due to chance, lower clopidogrel active metabolite exposure and pharmacodynamic effects of clopidogrel in nonsmokers relative to smokers is a plausible alternative explanation for the smokers’ paradox [68].

The magnitude of the effect of smoking on the efficacy of clopidogrel was evaluated in a 2013 systematic review, meta-analysis, and indirect comparison of randomized trials that included nearly 75,000 patients (29 percent smokers) in which clopidogrel was compared to placebo, aspirin, or another platelet P2Y₁₂ receptor blocker [63]. Among the subgroup of smokers, patients taking clopidogrel had a 25 percent reduction in the primary composite outcome of cardiovascular death, myocardial infarction (MI), and stroke compared to the control therapy (RR 0.75, 95% CI 0.67-0.83). In the subgroup of nonsmokers, patients taking clopidogrel had a smaller 8 percent reduction (0.92, 95% CI 0.87-0.98).

Finally, it has been hypothesized, that the underlying pathobiology of thrombosis may differ between smokers and nonsmokers [69].

HPR AND THROMBOTIC EVENTS — Most, but not all, descriptive and observational analytic studies contribute to the formulation of the hypothesis of an association between "resistance" to clopidogrel or high on-treatment platelet reactivity (HPR) and subsequent cardiovascular events after percutaneous coronary intervention (PCI) or peripheral artery intervention:

Platelet function and outcomes

●In a case series of aggregometry, 60 patients undergoing primary PCI for ST-elevation MI (STEMI) who were in the lowest quartile of clopidogrel responsiveness at day six (compared to baseline levels) had the highest rates of ischemic events during follow-up [70].

●In the PREPARE POST-STENTING study, a threshold of approximately 50 percent maximal periprocedural aggregation (with 20 micromol adenosine diphosphate [ADP]) was associated with six-month ischemic event occurrence [71]. In subsequent studies, approximately 40 percent aggregation (with 20 micromol ADP) was associated with the onset of risk for stent thrombosis occurrence and approximately 40 percent platelet aggregation (5 micromol ADP) among patients receiving long-term clopidogrel and aspirin therapy before stenting was associated with 12-month ischemic event occurrence [72,73].

●In a case series of 340 patients undergoing PCI, those with high posttreatment reactivity measured by VerifyNow (>235 P2Y₁₂ reaction units [PRU]) had higher rates of cardiovascular death (2.8 versus 0 percent, p = 0.04) and stent thrombosis (4.6 versus 0 percent, p = 0.004) [74].

●In a case series of 1608 patients undergoing elective PCI using the Multiplate analyzer, low responders as indicated by upper quintile platelet reactivity (approximately 482 aggregation units (AU)*min) had a significantly higher risk of definite stent thrombosis and a higher mortality rate within 30 days compared with patients less than 482 AU*min [75].

●In a case series of 802 consecutive patients undergoing elective PCI receiving a loading dose of 600 mg clopidogrel followed by 75 mg daily [76], platelet aggregation was assessed immediately before PCI by optical aggregometry following 5 micromol ADP. Multivariate analysis confirmed platelet aggregation above the median level as a significant independent predictor of the 30-day composite of death, MI, and target lesion revascularization (relative risk [RR] 6.7, 95% CI 1.5-29).

●On-treatment platelet reactivity utilizing the VerifyNow P2Y₁₂ point-of-care assay was evaluated in a study of 683 patients with acute coronary syndromes (ACS) who were treated with aspirin and clopidogrel and were scheduled to undergo PCI [77]. In a multivariate model, a residual platelet reactivity (RPR) ≥240 PRU was a significant and independent predictor of cardiovascular death and nonfatal MI at 12 months (hazard ratio [HR] 2.5, 95% CI 1.3-5.1). No significant association was found between a high RPR and target vessel revascularization. A subsequent meta-analysis, which included this study, concluded that an RPR ≥230 PRU was associated with a significantly higher rate of the composite end point of death, MI, and stent thrombosis (HR 2.10, 95% CI 1.62-2.73), as well as a significantly higher rate for each of the individual end points of death, MI, and stent thrombosis [78].

●The prospective registry, ADAPT-DES, is the largest case series of platelet function study conducted, which was in patients treated with stenting. Platelet reactivity was assessed using VerifyNow point-of-care assays in 8582 patients (52 percent ACS patients) after successful PCI. HPR (>208 PRU) was independently associated with 30-day definite/probable stent thrombosis (HR 3.0, p = 0.005), one-year definite/probable stent thrombosis (HR 2.49, p = 0.001), and MI (HR 1.42, p = 0.01); and two-year definite/probable stent thrombosis (adjusted HR 1.84, p = 0.009) and MI (HR 1.33, p = 0.01). In addition, >208 PRU was independently associated with a lower incidence of bleeding at one year (HR 0.73, p = 0.002) and at two years (HR 0.82, p = 0.02) [79,80]. A monotonic association between successively higher PRU quintiles and two-year stent thrombosis was observed, whereas the greatest risk of clinically relevant bleeding occurred in the lowest PRU quintile, with similar risks across the four higher quintiles. These relationships remained significant in fully adjusted multivariable analyses (adjusted HR for stent thrombosis in Q5 versus Q1: 2.32, 95% CI 1.17-4.59; p = 0.02; adjusted HR for clinically relevant bleeding in Q5 versus Q1: 0.61, 95% CI 0.47-0.77; p<0.001).

ADAPT-DES provides important and relevant information to the formulation of hypotheses about patients with HPR and culprit-lesion morphology, as well as adverse plaque morphology. Specifically, three-vessel coronary artery disease, incidence of fibroatheroma, elevated percentage of plaque burden, and longer culprit lesion attenuated plaque length [81].

Those prescribed a PPI had higher HPR (odds ratio 1.38, 95% CI 1.25-1.52; p = 0.0001) and also a greater rate of postdischarge adverse outcomes (HR 1.21, 95% CI 1.04-1.42; p = 0.02) [82]. After further analysis of patients who were treated with DAPT at two years in the ADAPT DES study (comprising 46 percent of patients), those with HPR prescribed clopidogrel had higher rates of definite or probable stent thrombosis (adjusted HR 2.16, p = 0.003), MI (adjusted HR 1.35, p = 0.02), freedom from clinically relevant bleeding (adjusted HR 0.74, p = 0.002), and all-cause mortality (adjusted HR 1.36, p = 0.04) [83]. These results are useful to formulate but not test hypotheses and may be due, all or in part, to confounding by indication. In addition, however, they are compatible with the "therapeutic window of P2Y₁₂ inhibition" at moderately inhibited PRU to avoid the risk of ischemic event occurrences and bleeding and contribute to the formulation of the platelet hypothesis for which personalized antiplatelet therapy is being prescribed [84].

●Systemic review and meta-analysis of individual patient data on MACE outcomes using 13 prospective observational studies with 6478 clopidogrel-treated patients revealed that the strength of the association between MACE risk and platelet reactivity as measured by ADP-induced aggregation by conventional aggregometry was increased significantly (p = 0.04) with the number of risk factors present (age >75 years, ACS at inclusion, diabetes, and hypertension). Platelet reactivity allowed the reclassification of 44 percent of the total population to a different risk level for the outcome of MACE, mostly in intermediate or high-risk patients [85]. In this regard, it should be noted that meta-analyses of observational studies will surely reduce the role of chance but also introduce bias and confounding as the individual studies are not randomized [86,87].

●In the PRECLOP's case series of 100 patients undergoing infrainguinal angioplasty or stenting and taking clopidogrel 75 mg daily were assessed with VerifyNow prior to the procedure [88]. At one year, the primary composite end point (death, major stroke, major amputation, target vessel revascularization, and bypass) increased with increasing quartile of PRU (4, 12, 52, and 84 percent, respectively; all p<0.05 except for the first versus second quartile).

Since case series are descriptive and are hypothesis formulating but not testing, some medically managed patients do not support this hypothesis:

●In the TRILOGY randomized trial, 9326 patients with medically managed unstable angina or non-ST elevation MI were randomly assigned to treatment with either clopidogrel or prasugrel and 27.5 percent of these patients were included in the platelet function substudy [89]. Platelet function was measured serially up to 30 months after randomization by the VerifyNow assay. There was only a modest, unadjusted association between on-treatment PRU values and HPR status with ischemic event occurrence that did not persist after multivariable adjustment.

●In the ADIRE case series of 771 stable cardiovascular outpatients, platelet reactivity to ADP was not associated with three-year major cardiovascular adverse events [90].

Loss of function gene carriers and outcomes — Evidence from prospective studies of P2Y₁₂ therapy suggest higher rates of cardiovascular events in patients with loss of function CYP2C19 gene variants who are taking clopidogrel [91-95]. As examples:

●In an NIH Implementing Genomics In Practice (IGNITE) study, patients with a LoF gene variant treated with clopidogrel had higher rates MACE after PCI. After 1 year of observation, CYP2C19 LOF allele carriers who received clopidogrel were more likely to experience a MACE than those allele carriers who received alternative therapy (adjusted HR 2.21, p = 0.021). In patients who received either prasugrel or ticagrelor, patients with a CYP2C19 LOF allele had a similar risk of MACE when compared to patients without a LOF allele (adjusted HR 0.81, 95% CI 0.48-1.35; p = 0.41) [91].  

●In a meta-analysis of randomized trials designed to compare the efficacy of clopidogrel to either prasugrel or ticagrelor in CYP219 LOF gene carriers, patients with a CYP219 LOF gene variant had a higher risk of cardiovascular events with clopidogrel than with other P2Y₁₂ inhibitors (RR 1.42, 95% CI 1.2-1.7) [92]. Patients without a LoF gene variant had a similar risk of cardiovascular events with clopidogrel or alternative P2Y₁₂ therapy (RR 1.0, 95% CI 0.8-1.25). Patients with PCI comprised 77 percent of the sample.  

CLINICAL MANIFESTATIONS AND DIAGNOSIS — As mentioned above, clopidogrel treatment failure is the occurrence of a thrombotic event/ischemic event during clopidogrel therapy in patients with heightened platelet reactivity (HPR). These events are usually stent thrombosis or recurrent acute coronary syndrome (ACS). The clinical presentations of these events are discussed separately. (See "Coronary artery stent thrombosis: Clinical presentation and management", section on 'Clinical presentation' and "Diagnosis of acute myocardial infarction".)

A definitive diagnosis of clopidogrel treatment failure requires the finding of HPR on laboratory testing in a patient with either stent thrombosis or recurrent ACS. In most patients, we do not believe testing for clopidogrel hyporesponsiveness is necessary. If the patient appears to have been taking clopidogrel as prescribed, switching to a more potent antiplatelet is reasonable. (See 'Management of possible clopidogrel failure' below.)

If a decision is made to use one of the following tests to assess the degree of platelet P2Y₁₂ receptor inhibition (responsiveness to antiplatelet therapy), we are most comfortable with the VerifyNow P2Y₁₂ assay, as the largest body of data correlating platelet reactivity to clinical outcomes has been obtained with it.

The following laboratory methods are used to assess the degree of platelet P2Y₁₂ receptor inhibition (responsiveness to antiplatelet therapy):

●Point-of-care and near point-of-care assays: VerifyNow P2Y₁₂ assay, Multiplate Analyzer, thrombelastography with Platelet Mapping Assay, and platelet function assay-100 with collagen-adenosine diphosphate (ADP) cartridge. VerifyNow and Multiplate assays have been widely used to demonstrate the relation of HPR to clinical outcomes in observational studies. VerifyNow has been used in the main prospective clinical trials of personalized antiplatelet therapy [96]. With the VerifyNow P2Y₁₂ assay, values >208 PRU are required for the finding of HPR. With Multiplate, values >47 AU are required for HPR.

●ADP-induced aggregation measured by light transmittance aggregometry using platelet rich plasma. Values >46 percent with 5 micromol ADP or >59 percent with 20 micromol ADP are required for the finding of HPR.

●P2Y₁₂ receptor reactivity determined by the degree of reduction in the maximal levels of vasodilator stimulated phosphoprotein (VASP) phosphorylation (P) stimulated by P_GE1 after the addition of ADP. VASP-P is measured by flow cytometry and is specific in determining P2Y₁₂ receptor activity. With VASP, values >50 percent PRI are required for the finding of HPR.

The cut-off values presented above are derived from studies that have used receiver operating characteristic (ROC) curve analysis used to define a threshold or cut-point of on-treatment platelet reactivity associated with the optimal combination of sensitivity and specificity to identify thrombotic/ischemic risk. It should be noted that such cut-points may depend on the subset of patients studied. In fact, cut-off values have been mainly investigated in patients undergoing percutaneous coronary intervention (PCI) and different targets may be obtained in other settings depending on patient management or baseline risk profile. The consistent findings across multiple investigations support the important role of HPR in the etiology of ischemic events after PCI, including stent thrombosis, and suggest the existence of a threshold level of platelet reactivity below which ischemic events may be prevented.

The observed cut-off values for platelet reactivity noted above had a very high negative predictive value for thrombotic/ischemic event occurrence. However, the positive predictive value is low for all assays. This is consistent with the fact that although a major determinant of thrombotic events, HPR is not the sole factor responsible for these events [1].

A 2010 consensus statement proposed cut-off values based on receiver operating characteristic curve analysis for different platelet function assays to be used in future studies of personalized antiplatelet therapy [1]. A meta-analysis of studies employing the VerifyNow point-of-care assay lends further support for the potential role of monitoring of P2Y₁₂ receptor inhibitor therapy as a diagnostic marker [78].

Taken together, these data are compatible with the hypothesis that adequate protection against ischemic events with clopidogrel therapy might be achieved by overall low to moderate levels of post-treatment platelet reactivity. Some studies suggested that the prognostic value of HPR can be improved by the addition of classical clinical and procedural risk factors and suggested that a comprehensive algorithm including clinical as well as laboratory findings (platelet reactivity and genotype) should be considered in future personalized antiplatelet therapy investigations to optimize outcomes [97,98].

SCREENING — The utility of routine screening for clopidogrel hyporesponsiveness or nonresponsiveness is controversial [99]. In the absence of a well-established benefit of genotype-guided therapy, we do not routinely test patients for clopidogrel resistance with genetic testing to detect loss of function gene carriers (eg, CYP2C19 allele testing) or with platelet function testing to detect P2Y₁₂ resistance [100,101].

Some experts recommend routine testing based on biological plausibility and observational studies that suggest an association between genotype or platelet function and stent thrombosis. In contrast, most randomized trials do not show a benefit of genotype-guided [102-104] or platelet reactivity-guided [105-108] antiplatelet therapy. As examples:

●The TAILOR-PCI was a large trial of gene-guided antiplatelet therapy in the drug-eluting stent era [102]. In this trial, 5302 patients undergoing percutaneous coronary intervention (PCI; 84 percent for acute coronary syndrome [ACS]) and who were to receive 12 months of DAPT were randomly assigned to clopidogrel therapy or genotype-guided therapy. In the genotype-guided arm, CYP2C19 LoF allele carriers were given ticagrelor and wild-type allele carriers received clopidogrel; all patients in the standard-therapy arm received clopidogrel. The primary analysis only included patients with LoF alleles, which included 903 patients in the genotype-guided group and 946 patients in the standard therapy group. Genotype-guided therapy was associated with a nonsignificant 34 percent reduction in the primary endpoint of cardiovascular death, myocardial infarction (MI), stroke, stent thrombosis, or severe recurrent ischemia at 12 months (4.0 versus 5.9 percent; hazard ratio [HR] 0.66, 95% CI 0.43-1.02). The reduction in the endpoint was primarily driven by a reduction in severe recurrent ischemia (ie, ECG changes leading to urgent revascularization without a diagnosis of MI, cardiac angina). The risk of major or minor bleeding was similar in both groups. In a prespecified sensitivity analysis that assessed repeated events, genotype-guided therapy was associated with a significant 40 percent reduction in the primary endpoint compared with standard therapy (HR 0.60, 95% CI 0.41-0.89).  

●The POPULAR GENETICS trial found that genotype-guided therapy in patients with STEMI was associated with similar rates of MACE [103]. In this trial, 1242 patients were randomly assigned to gene-guided therapy and 1246 were assigned to standard care. In the gene-guided group, LoF carriers received prasugrel or ticagrelor, while all patients in the standard therapy group received either prasugrel or ticagrelor. The rate of all-cause death, myocardial infarction, definite stent thrombosis, stroke, and bleeding were similar between groups (5.1 versus 5.9 percent in the standard therapy group; absolute rate difference 0.7 percent; 95% CI 2.0 to -0.7).

●In the ANTARCTIC randomized trial, 877 ACS patients aged 75 years or older who underwent coronary stenting and were treated with prasugrel 5 mg were randomly assigned to platelet monitoring or no monitoring groups [107]. There was no difference between the two strategies in the primary composite end point of cardiovascular death, MI, stroke, stent thrombosis, urgent revascularization, and bleeding complications at 12 months (28 percent in both groups) or the single end point of bleeding events. The chief strength of ANTARCTIC is the randomized design, but limitations include the fact that, with low-dose prasugrel, the need to intensify therapy for high platelet reactivity was uncommon in the monitoring group (approximately 4 percent). In addition, almost all of the P2Y₁₂ dose adjustments in ANTARCTIC were de-intensification from prasugrel to clopidogrel in an attempt to reduce bleeding.

●The ARCTIC trial and subsequent ARCTIC-Monitoring study did not demonstrate a clear benefit of platelet function-guided antiplatelet therapy prior to PCI or during one year of follow-up and noted high variability in agreement between platelet function assays [105,106].  

MANAGEMENT OF POSSIBLE CLOPIDOGREL FAILURE — Patients who have stent thrombosis or an acute coronary syndrome (ACS) while taking clopidogrel may have clopidogrel resistance or hypo-/nonresponsiveness. As no adequately powered study has clearly demonstrated that testing for clopidogrel hyporesponsiveness improves clinically important outcomes, most of our authors and reviewers do not perform these tests.

For these patients, based on studies that have shown that more potent platelet P2Y₁₂ receptor blockers (such as ticagrelor or prasugrel) lower on-treatment platelet reactivity, most of our experts switch to one of these drugs. However, two of our authors perform platelet function testing and point-of-care genetic testing to assess for clopidogrel hyporesponsiveness in these cases and manage the patient accordingly: the finding of platelet resistance leads to a recommendation for ticagrelor or prasugrel; the finding of adequate clopidogrel effect leads to continuation of clopidogrel therapy.

None of the interventions discussed below, including the use of a more potent agent, has been shown to improve long-term clinical outcomes.

Acute coronary syndrome

Adjusting the loading dose — Achieving a platelet reactivity level below high on-treatment platelet reactivity (HPR) by selectively increasing the clopidogrel loading dose based upon laboratory evidence for clopidogrel resistance has been shown to benefit short-term outcome [109-116].

The clinical benefit of increasing the clopidogrel loading dose on reducing the incidence of subsequent adverse cardiac events was shown in a prospective randomized trial of 162 patients scheduled to undergo percutaneous coronary intervention (PCI). Clopidogrel hypo-nonresponsiveness was defined as an in vitro vasodilator-stimulated phosphoprotein phosphorylation analysis (VASP index) >50 percent after a 600 mg loading dose of clopidogrel [111]. All patients with a VASP index >50 percent were randomly assigned to a control group or VASP-guided group. In the comparison group, PCI was carried out without an additional bolus of clopidogrel, whereas in the VASP-guided group, up to three additional boluses of 600 mg of clopidogrel were given in 24-hour increments. The VASP index was then assessed 12 hours after administration until an index <50 percent was achieved, following which PCI was performed. The following findings were noted:

●In the VASP-guided group, 38 patients required two doses of clopidogrel to achieve a VASP index <50 percent, 14 required three doses, and 15 required four doses.

●Despite a total clopidogrel dose of 2400 mg, 11 patients (14 percent) remained "resistant" and received PCI. There were no major adverse cardiac events in this subgroup.

●The number of serious adverse cardiac events (two cardiovascular deaths, four acute or subacute stent thromboses, two recurrent ACS) during the one-month follow-up was significantly less in the VASP-guided group (0 versus 8 events).

Adjusting the maintenance dose — The following studies in which the maintenance dose was adjusted did not show improvement in long-term clinical outcome:

●In the GRAVITAS trial, 2214 subjects with high platelet reactivity on clopidogrel (as measured by the VerifyNow P2Y₁₂ test) following PCI with a drug-eluting stent, were assigned at random to treatment with either standard dose (no loading dose, 75 mg/day) or high dose (600 mg loading dose, 150 mg/day thereafter) clopidogrel for 60 days [6]. The primary study end point (six-month incidence of death from cardiovascular causes, nonfatal myocardial infarction, or stent thrombosis) was virtually identical in those receiving either standard or high dose clopidogrel (2.3 percent in each group; hazard ratio 1.01; 95% CI 0.58-1.76). In addition to being underpowered, there are other potential explanations for the neutral results of GRAVITAS trial [117]. In a time-dependent analysis of GRAVITAS, <208 PRU was independently associated with the 60 days primary end point (hazard ratio [HR] 0.23, 95% CI 0.05-0.98) and tended to be an independent predictor at six months (HR 0.54, 95% CI 0.28-1.04) [118]. Only a minority of patients receiving high-dose clopidogrel achieved <208, indicating that high-dose clopidogrel regimen may have been sub-optimal and a more potent intervention that reduces HPR to a greater extent would have had greater potential to improve clinical outcomes given the very low event rate. In support of this hypothesis, the ELEVATE-TIMI (Escalating Clopidogrel by Involving a Genetic Strategy - Thrombolysis In Myocardial Infarction) 56 trial showed that up to 225 mg of clopidogrel might be necessary to overcome HPR in patients carrying one loss-of-function cytochrome 2C19 gene (see below). GRAVITAS enrolled a population at low absolute risk for ischemic events despite displaying HPR. The majority of patients had stable angina, were successfully treated with PCI, and periprocedural events were not included in the primary end point. The tested pharmacologic intervention was administered more than 12 hours after PCI and the associated acute vessel injury/stent deployment, which may have been too late to blunt a platelet-related incipient lesion. Finally, it is possible that a single platelet function test may not reliably reflect the effect of clopidogrel on adenosine diphosphate-induced platelet reactivity in all patients [117].

●In the CREATIVE trial, 1078 patients undergoing PCI who had low responsiveness to clopidogrel, as assessed by thromboelastography, were randomly assigned to standard antiplatelet therapy (clopidogrel 75 mg daily plus aspirin 100 mg daily), double-dose clopidogrel (clopidogrel 150 mg daily plus aspirin 100 mg daily), or adjunctive use of cilostazol (cilostazol 100 mg twice daily plus aspirin 100 mg daily plus clopidogrel 75 mg daily) [119]. The primary end point of the incidence of major adverse cardiac and cerebrovascular events at 18 months occurred in 14.4, 10.6, and 8.5 percent of the three groups, respectively (HR 0.72, 96% CI 0.47-1.09 and 0.55, 95% CI 0.35-0.87, comparing double-dose clopidogrel or adjunctive cilostazol, respectively). There were no significant differences in the rates of major bleeding.

●The ELEVATE study of PCI demonstrated that 225 mg daily clopidogrel is required to reduce platelet reactivity (<50 platelet reactivity index as measured by VASP phosphorylation assay) in CYP2C19*2 heterozygotes to levels achieved with standard clopidogrel, 75 mg, in noncarriers [120]. Clinical outcomes were not assessed.

Use of alternative antiplatelet agents — There are no randomized trials that have tested whether the substitution (for clopidogrel) of more potent platelet P2Y₁₂ receptor blockers (ticagrelor and prasugrel) in patients with HPR lowers the risk of subsequent adverse cardiovascular outcomes. Since the use of these two drugs has been shown to lower on-treatment platelet reactivity, it may be prudent to adopt this strategy [54,121,122] and we believe this is a reasonable approach. Practitioners and patients should be aware of the higher risk of bleeding with these agents.

Further support for this approach derives from data summarized below:

●In a randomized, single-blind, cross-over study, prasugrel was more efficient in reducing platelet function compared to clopidogrel as measured by VerifyNow P2Y₁₂ assay. The prevalence of HPR was significantly lower in all patients (7.5 versus 35.8 percent), in CYP2C19*2 carriers (5.3 versus 47.4 percent), and in noncarriers (8.8 versus 29.4 percent). In another randomized, crossover study, the same authors demonstrated that 10 mg prasugrel therapy was very effective in overcoming HPR rate compared to 150 mg clopidogrel therapy in CYP2C19*2 carriers (0 versus 100 percent) [123].

●In the RAPID GENE study, 200 patients undergoing PCI were randomly assigned to either rapid point-of-care genotyping for the CYP2C19*2 allele or standard treatment. Carriers were given 10 mg prasugrel daily, while noncarriers and those in the standard treatment group were given 75 mg of clopidogrel daily [122]. The primary end point was high on-treatment platelet reactivity after one week of antiplatelet therapy, which occurred in none of the 23 carriers in the genotyping group treated with prasugrel compared with 7 of the 23 carriers (30 percent) in the standard treatment group who were treated with clopidogrel.

●In a post-hoc subgroup analysis of data from the ONSET/OFFSET and RESPOND studies, the prevalence of high on-treatment platelet reactivity, using a number of different laboratory assays, was significantly lower for those treated with ticagrelor (0 to 8 percent) than for those treated with clopidogrel (21 to 81 percent) [121]. (See "Platelet biology and mechanism of anti-platelet drugs" and "Acute non-ST-elevation acute coronary syndromes: Early antiplatelet therapy".)

●In the TRIGER-PCI randomized trial of 212 patients with stable coronary artery disease and high on-treatment platelet reactivity undergoing PCI, a 10 mg daily dose of prasugrel was effective in reducing on-treatment platelet reactivity compared to 75 mg daily dose clopidogrel [108]. However, the study was terminated early for futility because of extremely low event rates.

●In a randomized trial of STEMI patients undergoing PCI, two-hour post-dose, HPR (>208 PRU) prevalence was 46.2 and 34.6 percent in patients treated with ticagrelor and prasugrel, respectively. Efficacy, defined as reducing platelet reactivity during the first 24 hours of STEMI was similar in the two groups [124].

●In two studies, aspirin and clopidogrel response was evaluated by VerifyNow assay in patients undergoing elective PCI. Poor responders to aspirin or clopidogrel were treated with a glycoprotein (GP) IIb/IIIa receptor inhibitor which decreased 30-day as well as one-year post-PCI ischemic events without increased bleeding rates [125,126].

●Previous large scale studies of personalized antiplatelet therapy that used the VerifyNow P2Y₁₂ assay mainly recruited low-risk patients undergoing PCI and mostly used the suboptimal therapy of 150 mg daily clopidogrel to overcome HPR [6,106,108]. The latter studies failed to demonstrate the utility of personalization of antiplatelet therapy. In a study of 714 patients with ACS undergoing PCI, platelet function was measured by Multiplate analyzer, and patients with HPR on standard clopidogrel therapy were treated with high-dose clopidogrel or prasugrel at the discretion of the treating physician. In this study, the risk of all-cause death, myocardial infarction, stent thrombosis, or stroke at one year was higher in the high-dose clopidogrel group compared to patients without HPR (HR 2.27, 95% CI 1.45-3.55), whereas outcomes were similar between the prasugrel treated group and patients without HPR (HR 0.90, 95% CI 0.44-1.81) [127]. The latter outcomes were associated with significantly lower platelet reactivity in the prasugrel treated group compared with the high-dose clopidogrel treated group (p<0.0001).

●The TROPICAL ACS trial, discussed elsewhere, provides some additional evidence for switching P2Y₁₂ receptor blockers in patients with an ACS. (See "Acute non-ST-elevation acute coronary syndromes: Early antiplatelet therapy", section on 'Switching from ticagrelor or prasugrel to clopidogrel'.)

●A randomized, pharmacodynamic trial of clopidogrel 300 mg daily, ticagrelor 90 mg twice daily, or ticagrelor 60 mg twice daily in 162 aspirin-treated stable patients who underwent PCI found the following [128]:

•Both maintenance doses of ticagrelor achieved more potent and consistent platelet inhibition than clopidogrel at one month.

•There was no impact of ticagrelor on in vitro adenosine uptake or adenosine plasma levels with either ticagrelor dose compared with clopidogrel.

●A genotyping substudy of TROPIC ACS revealed that CYP2C19*2 and CYP2C19*17 carrier status in rapid metabolizers correlates with platelet reactivity in ACS patients on clopidogrel. Based on these results, the authors suggested that CYP2C19 genotyping can play a valuable role for preselecting patients who will and who may not be suitable for early guided deescalation of antiplatelet treatment. They also demonstrated that in patients treated with platelet-function-guided deescalation strategy, genotyping may not provide added benefit in predicting ischemic or bleeding risk [129].

The CREATIVE trial presented above suggests that triple therapy with clopidogrel, aspirin, and cilostazol may be beneficial. (See 'Adjusting the maintenance dose' above.)

In the HOST-REDUCE-POLYTECH-ACS trial, conducted in South Korea, 2338 ACS patients undergoing PCI were randomly assigned to the de-escalation strategy of 10 mg prasugrel for one month followed by 5 mg prasugrel versus conventional strategy of 10 mg prasugrel for one year in addition to 100 mg aspirin in both groups. The primary endpoint of net adverse clinical events (all-cause death, nonfatal MI, stent thrombosis, repeat revascularization, stroke, and bleeding events of grade 2 or higher according to Bleeding Academic Research Consortium [BARC] criteria) was similar between groups (Kaplan-Meier estimate 7.2 percent in the de-escalation group and 10.1 percent in the conventional group, a statistically significant reduction [HR 0.70, 95% CI 0.52-0.92; p equivalence = 0.012]). Ischemic risk was similar between the groups (HR 0.76, 95% CI 0.40-1.45; p = 0.40) while bleeding risk was significantly decreased in the de-escalation group (HR 0.48, 95% CI 0.32-0.73; p = 0.0007).

Stent thrombosis — The approach to additional antiplatelet therapy in patients with stent thrombosis taking clopidogrel is presented separately. (See "Coronary artery stent thrombosis: Clinical presentation and management", section on 'Long-term antiplatelet therapy'.)

LOW PLATELET REACTIVITY AND BLEEDING — While dual antiplatelet therapy (DAT) lowers the risk of ischemic events in patients undergoing percutaneous coronary intervention, the risk of bleeding is increased, particularly when more potent P2Y₁₂ receptor inhibitors are used. In this setting, low platelet reactivity (LPR) is associated with an increased risk of bleeding. A consensus document highlighting the above observations with a therapeutic window concept and updated cutoffs for high on-treatment platelet reactivity and LPR for P2Y₁₂ receptor inhibitor therapy has been published [130]. However, we do not recommend screening for low platelet reactivity, as there are insufficient data to support its use.

Platelet function testing offers the possibility of discerning the optimal drug choice and dosing strategy for an individual patient. The safety and efficacy of early de-escalation of antiplatelet treatment from prasugrel to clopidogrel guided by platelet function testing was investigated in the open-label, assessor-blinded, TROPICAL-ACS trial, which is discussed elsewhere. (See "Acute ST-elevation myocardial infarction: Antiplatelet therapy", section on 'Switching from ticagrelor or prasugrel to clopidogrel'.)

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Non-ST-elevation acute coronary syndromes (non-ST-elevation myocardial infarction)" and "Society guideline links: ST-elevation myocardial infarction (STEMI)".)

SUMMARY AND RECOMMENDATIONS

●"Resistance" or "nonresponsiveness/hyporesponsiveness" to an antiplatelet drug is a pharmacodynamic phenomenon in which there is no significant reduction in platelet function after treatment as compared to the baseline. High on-treatment platelet reactivity (HPR) is a similar concept and indicates persistent response of the P2Y₁₂ receptor (clopidogrel target) based on one measurement of on-treatment platelet reactivity. (See 'Definitions' above.)

●Most, but not all, studies have found a positive association between patients with hypo-/nonresponsiveness or HPR and subsequent thrombotic/ischemic events after percutaneous coronary intervention. (See 'HPR and thrombotic events' above.)

●Clopidogrel treatment failure is the occurrence of a thrombotic/ischemic event during clopidogrel therapy in patients with heightened platelet reactivity (HPR). These events are usually stent thrombosis or recurrent acute coronary syndrome (ACS). (See 'Clinical manifestations and diagnosis' above.)

●For patients who have been started on clopidogrel, we do not routinely test patients for clopidogrel resistance with genetic testing to detect loss of function gene carriers (eg, CYP2C19 allele testing) or with platelet function testing to detect P2Y₁₂ resistance. (See 'Screening' above.)

●The optimal approach to patients taking clopidogrel for a prior ACS (without stenting) who develop a subsequent ACS is not known. For these patients, we suggest switching from clopidogrel to a more potent platelet P2Y₁₂ receptor blocker rather than any other strategy (Grade 2C). (See 'Management of possible clopidogrel failure' above.)

●Our approach to patients taking clopidogrel who develop stent thrombosis is discussed separately. (See "Coronary artery stent thrombosis: Clinical presentation and management", section on 'Long-term antiplatelet therapy'.)