Combined Cardioprotectant and Antithrombotic Actions of Platelet P2Y12 Receptor Antagonists in Acute Coronary Syndrome

Abstract

Since the P2Y₁₂ receptor antagonists were first introduced, they have been extensively tested in patients with acute coronary syndrome and are now standard of care. These antiplatelet drugs are very effective in reducing subsequent cardiovascular events, stent thromboses, and mortality in patients with acute myocardial infarction undergoing reperfusion therapy. Although the prevailing view is that their benefit derives from their antithrombotic properties, other unrelated pleiotropic effects appear to be equally beneficial. Accumulating clinical and animal evidence indicates that, if present at the time of reperfusion, these drugs have a direct anti-infarct effect similar to that of ischemic postconditioning. Four oral antagonists have been developed in rapid succession: ticlopidine, clopidogrel, prasugrel, and ticagrelor. Each agent had a more consistent and rapid onset of action than the previous one, and this has correlated with improved clinical outcomes when given early in treatment. Unfortunately, gut absorption causes an appreciable delay in the onset of effect, especially when morphine is used, and the constant push to minimize the door-to-balloon time has made it difficult to achieve adequate platelet inhibition at the time of percutaneous coronary intervention with an oral agent. An intravenous P2Y₁₂ antagonist such as cangrelor may optimize treatment because it produces nearly maximal inhibition of platelet aggregation within minutes. If antiplatelet agents do protect through postconditioning’s mechanism, then they would render any other intervention that protects through that mechanism redundant. Indeed, animals treated with cangrelor cannot be further protected by pre- or postconditioning. However, interventions that use a different mechanism such as mild hypothermia or cariporide, a Na⁺-H⁺ exchange blocker, do add to cangrelor’s protection. Future research should be directed toward identifying interventions that can augment the protection from antiplatelet therapy and finding a way to optimize P2Y₁₂ inhibition at reperfusion in all patients.

Platelet Inhibitors and Infarct Size Reduction

Outcomes for patients with ST-segment elevation myocardial infarction (STEMI) have improved remarkably over the past decade. The 3-month mortality was 9.4% in the COMPlement inhibition in myocardial infarction treated with thromboLYtics (COMPLY) trial (2000-2002), ¹ and it was 3.2% in the δ-PROTEin kinase C for the reducTION of infarct size-acute myocardial infarction (PROTECTION-AMI) trial (2008-2010). ² Mochly-Rosen and Grimes ² compiled data from 10 trials over that time period, and the trend toward steady improvement can clearly be seen. Although a number of factors may have contributed, the increased use of antiplatelet drugs during percutaneous coronary intervention (PCI) in patients with STEMI is notable. Figure 1 shows the influence of clopidogrel, a platelet P2Y₁₂ adenosine diphosphate (ADP)-receptor blocker, aspirin, and the glycoprotein (GP) IIb/IIIa inhibitor abciximab. In 7193 patients, loading with clopidogrel or abciximab prior to stenting reduced 1-year mortality by one-third from that seen with aspirin. ³ Combining clopidogrel and abciximab was no better than either alone. When Heestermans et al ⁴ examined the Zwolle acute myocardial infarction registry, they noticed that infarct size and 30-day mortality were reduced in patients who received an antiplatelet agent prior to rather than after balloon dilatation of the coronary artery, indicating that the drugs seemed to protect against an injury that occurred in the first minutes of reperfusion.

Figure 1.

Kaplan-Meier curves for 4 treatments of patients undergoing primary percutaneous coronary intervention followed for 1 year. GP indicates glycoprotein. Adapted with permission from Witkowski et al. ³

A major factor in outcomes after a myocardial infarction is infarct size. We will review many studies which show that platelet inhibitors limit infarct size in both animals and humans. We will explore possible mechanisms of this salvage. Some investigators propose that these antiplatelet drugs prevent microthrombi that impair reperfusion of the ischemic tissue, while others suggest that they prevent platelets from activating neutrophils which attack otherwise viable tissue. We have proposed that at least the P2Y₁₂ inhibitors condition the heart and activate protective signaling as seen in pre- and postconditioning independent of any effect on platelet aggregation.

Thrombosis and Myocardial Infarction

The pathophysiology of acute myocardial infarction is clear. Rupture of a coronary atherosclerotic plaque exposes thrombogenic elements in the plaque to platelets and fibrinogen and thrombosis ensues. ⁵ Following PCI, endothelium is disrupted exposing thrombogenic elements in the vessel wall to flowing blood, and deployment of intravascular stents introduces thrombogenic elements into the coronary artery. Thus, thrombosis is a pathologic process that should be prophylactically inhibited to prevent complications of PCI. In the 1980s and 1990s, 2 types of antithrombotic agents could be used chronically. Coumarin derivatives such as warfarin interfered with hepatic production of clotting factors, and, therefore, impeded the coagulation cascade, while aspirin blocked the enzyme cyclooxygenase and production of thromboxane A₂, a potent ligand released by activated platelets which causes platelet aggregation. Platelet activation and aggregation are early steps in thrombus formation. Patients with STEMI have increased platelet reactivity and higher levels of platelet–monocyte and platelet–neutrophil aggregates. ⁶ The extent of platelet aggregation on admission correlates with the area of microvascular obstruction assessed by contrast-enhanced magnetic resonance imaging (no-reflow phenomenon), and infarct size correlates with the extent of no-reflow. ⁶ Therefore, aspirin and anticoagulants such as warfarin interfered with different phases of the thrombotic process, and each was used in patients with myocardial infarction undergoing coronary revascularization procedures (see subsequently).

Infusion of platelets into animal hearts subjected to ischemia/reperfusion increases myocardial necrosis but only if they had first been activated by pharmacologic agents or ischemia. ⁷ Nonactivated platelets infused into isolated hearts subjected to ischemia/reperfusion increase the extent of ischemia ⁸ and adversely affect postreperfusion hemodynamics. ⁹ Platelet aggregation is minimized in mice with deficiency of Gq, ¹⁰ GPVI (platelet collagen receptor), ^11,12 or P-selectin (protein responsible for platelet-endothelium interaction), ¹³ and infarct size is smaller than that in wild-type animals following ischemia/reperfusion. Hence, platelet inhibition should be beneficial to the ischemic heart.

Clinical Trials With Aspirin, Ticlopidine, and Coumarin Derivatives (Warfarin-Like Agents)

The Antiplatelet Trialists Collaboration analyzed 145 randomized trials of prolonged antiplatelet therapy, mainly aspirin. ¹⁴ Vascular events were significantly less common in patients with vascular disease which included acute myocardial infarction. Because of aspirin’s success in minimizing vascular events in individuals with coronary and peripheral arterial disease, efforts were expended to identify other antiplatelet agents. Thrombosis is initiated by tethering of platelets to endothelial cells via P-selectin followed by their binding to exposed collagen through GPVI receptors. These events trigger a signaling cascade leading to cross-linking of platelets and aggregation. This process is amplified by activation of platelet surface receptors for ADP and thromboxane A₂, both released by the activated platelet.

Ticlopidine is a thienopyridine which irreversibly blocks the platelet’s P2Y₁₂ ADP receptors. It was developed for aspirin-intolerant patients or for use with aspirin when dual antiplatelet treatment was needed. The effect of combination of ticlopidine and aspirin on incidence of death, myocardial infarction, coronary artery bypass grafting, and repeat angioplasty at 30 days after coronary artery stenting was compared to that of combination of a coumarin derivative and aspirin in patients with either acute coronary syndrome (ACS) or stable angina pectoris. ¹⁵ There was a 75% decrease in cardiac events in the ticlopidine group (P < .01). All events in the antiplatelet group occurred in the first week, while events in the anticoagulant group occurred up to day 26. Occlusion of the stented vessel was observed in 0.8% of the patients treated with ticlopidine and aspirin and in 5.4% of patients treated with warfarin and aspirin (P < .004). Thus, dual anti-platelet therapy was much more effective than conventional anticoagulant treatment in prevention of vascular complications following stenting, mainly because of fewer in-stent thrombi. These results were confirmed in the STent Anticoagulation Restenosis Study (STARS). ¹⁶ Although the 2 regimens of ticlopidine and aspirin and aspirin and an oral anticoagulant in the Full ANticoagulaTion versus ASpirin and TIClopidine (FANTASTIC) trial produced comparable clinical events, ticlopidine treatment resulted in fewer subacute stent occlusions (P = .01). ¹⁷

Early Clopidogrel Trials

Despite ticlopidine’s clinical efficacy, it increased the risk of neutropenia and thrombotic thrombocytopenic purpura and caused gastrointestinal symptoms often requiring drug discontinuation. It was supplanted by clopidogrel, a second irreversible P2Y₁₂ receptor antagonist with a much lower hematologic risk. The CLopidogrel ASpirin Stent International Cooperative Study (CLASSICS) ¹⁸ examined 3 drug regimens initiated in the first 6 hours following stenting and continued for 4 weeks: daily maintenance dose (75 mg) of clopidogrel and aspirin; loading dose of clopidogrel (300 mg) followed by daily maintenance dose and aspirin; and ticlopidine and aspirin. There was no difference in incidence of vascular events in ticlopidine and combined clopidogrel groups, but patients treated with ticlopidine had significantly more neutropenia, thrombocytopenia, and rashes. A comparison of ticlopidine and aspirin versus clopidogrel (loading dose of 600 mg and daily maintenance dose of 75 mg) and aspirin initiated immediately before stenting failed to show any difference in vascular events after 1 month or scintigraphic infarct size. ¹⁹ Thus, clopidogrel was established as an effective drug in patients receiving stents and it had fewer disqualifying side effects.

In the Clopidogrel in Unstable angina to prevent Recurrent Events (CURE) trial, major adverse vascular events were decreased by clopidogrel in patients with non-STEMI (NSTEMI). Most were treated medically with either clopidogrel (loading dose of 300 mg and maintenance dose of 75 mg) and aspirin or placebo and aspirin for 3 to 12 months, and major adverse vascular events were decreased by clopidogrel by nearly 20% (P < .001). ²⁰ For patients with stenting, PCI was performed a median of 6 days after initiation of treatment. ²¹ There was a 42% decrease in major vascular events in clopidogrel-treated patients. But this benefit of clopidogrel was not shared by all but rather was dependent on the time the treatment was started. The decrease in vascular events was exclusively observed in those individuals in whom stenting was performed less than 48 hours after initiation of clopidogrel treatment (P = .038). ²²

In the CLopidogrel as Adjunctive ReperfusIon Therapy (CLARITY) studies, ^23,24 aspirin and clopidogrel (300 mg loading dose and 75 mg maintenance dose) in patients with STEMI treated with fibrinolysis yielded 20% to 46% risk reduction over placebo at 30 days. Clopidogrel improved patency of the infarct-related artery by up to 36% (P < .001). Coronary angiography was performed a median of 3 days following initiation of drug. Thus, clopidogrel diminished the chance of recurrent thrombosis following fibrinolysis.

Clopidogrel’s Shortcomings

Clopidogrel is an oral prodrug. It first must be absorbed from the gut and then metabolized by cytochrome P450-dependent pathways in the liver to produce active metabolites that can interact with P2Y₁₂ receptors. In patients with ACS, a loading dose of 300 mg resulted in peak inhibition of only 20% to 30% of platelet aggregation in approximately 6 hours. ²⁵ Inhibition was only slightly less at 24 hours. Clinicians were concerned about clopidogrel’s modest effect on platelet aggregation and fretted that its slow onset of effect might result in inadequate inhibition of platelet function in the critical hours immediately after PCI. In the Clopidogrel for the Reduction of Events During Observation (CREDO) trial, 300 mg of clopidogrel or placebo was administered between a minimum of 3 hours and maximum of 24 hours before planned PCI. ²⁶ There was no difference in vascular events at 28 days among clopidogrel and placebo groups until pretreatment occurred more than 15 hours before PCI when clopidogrel reduced relative risk by 66.6% (P = .011). A meta-analysis of trials of clopidogrel pretreatment versus administration of a loading dose just before or immediately after PCI confirmed that adequate pretreatment significantly decreased vascular events in the follow-up period of 7 to 365 days. ²⁷

Loading With Clopidogrel

There was a clear benefit of effective platelet inhibition at the time of stenting. But the drive to shorten the interval to less than 90 minutes between hospital arrival of the patient with STEMI and commencement of revascularization made platelet inhibition at start of stenting difficult to achieve. Accordingly, the efficacy of larger loading doses was examined. Loading with 600 mg in patients with ACS produced peak platelet inhibition within 2 hours and loading with 300 mg showed peak platelet inhibition only after 6 hours. ²⁵ However, it still took 5 to 6 hours for the peak effect, 35% to 45% inhibition of platelet aggregation, to be reached. The higher loading dose consistently decreased subsequent cardiovascular events, ^{28 –31} stent thrombosis, ^30,31 infarct size, ³² and microvascular obstruction, ³² and a meta-analysis of 10 studies showed that loading doses of clopidogrel greater than 300 mg resulted in fewer cardiac deaths or myocardial infarctions than loading doses of 300 mg. ³³ In 1 study, there was no difference in periprocedural ischemic events or cardiovascular complications at 30 days in patients with stable angina pectoris treated with either 300 mg or 600 mg of clopidogrel immediately before PCI. ³⁴ But platelets would not have been inhibited until several hours after the procedure with either dose, possibly accounting for lack of clinical effect. However, in neither Intracoronary Stenting and Antithrombotic Regimen: Rapid Early Action for Coronary Treatment-4 (ISAR-REACT) ³⁵ nor PRAGUE-8 ³⁶ trials did early or late administration of 600 mg clopidogrel before elective PCI have an effect on cardiovascular adverse events at 7 to 30 days. The benefit of clopidogrel loading seems to be more pronounced in patients with ACS in whom infarction is often present suggesting an anti-infarct effect with early administration.

Clopidogrel does not protect all patients. In fact, 24% to 46% treated with 300 mg clopidogrel ^25,29,37 and 15% to 21% treated with 600 mg clopidogrel ^25,29 had little or no inhibition of platelet aggregation 6 to 19 hours following treatment. Gurbel et al ³⁸ measured platelet reactivity serially in patients who received a loading dose of 300 mg clopidogrel before elective coronary artery stenting and then a daily maintenance dose of 75 mg. After 2 hours of the loading dose, 63% still had less than 10% inhibition of platelet aggregation. Thirty-one percent of the patients continued to be nonresponders at 24 hours and 5 days and 15% at 30 days. Loading with 600 mg clopidogrel in patients with NSTEMI ³⁹ and STEMI ⁴⁰ left 36% and 55%, respectively, with high platelet reactivity at the time of PCI. Nonresponsiveness (definition is dependent on the particular study, eg, less than 10% change in platelet aggregation from baseline to post-treatment blood sample in response to 5 μmol/L ADP ³⁸ ) had adverse clinical implications. Nonresponders had more frequent elevations of creatine phosphokinase-MB following PCI, ³⁷ a significantly higher likelihood of subacute stent thrombosis, ⁴¹ and possibly a higher likelihood of adverse cardiovascular events. ³⁹

Likely causes of clopidogrel nonresponsiveness could be poor intestinal absorption or inadequate metabolism to the active metabolite. Biotransformation of the prodrug to the active metabolite by cytochrome P450 enzyme system is affected by the CYP2C19 gene. ⁴² The CYP2C19*2 variant allele leads to poor metabolism. In the year following deployment of drug-eluting stents, myocardial infarction occurred in 8.4%, 10.9%, and 44.4% of the patients with 0, 1, and 2 variant alleles, respectively (P = .016). There are still other factors that affect clopidogrel’s pharmacokinetics that have not yet been identified. ⁴³

Improving the Pharmacology of Oral Antiplatelet Agents

A third-generation thienopyridine was introduced because of the variable response to clopidogrel. Prasugrel, like clopidogrel, is a prodrug requiring hepatic metabolism to become an active metabolite that irreversibly binds to platelet P2Y₁₂ receptors and blocks platelet aggregation. But it is metabolized by different enzymes and has a more consistent effect in patients.

Inhibition of aggregation of platelets from rats fed prasugrel was 10-fold greater than that seen with clopidogrel. ⁴⁴ Similar observations were made in patients with stable coronary artery disease treated with either prasugrel or clopidogrel. ⁴⁵ In patients undergoing elective PCI, there was 74.8% inhibition of platelet activity 6 hours after 60 mg prasugrel compared to 31.8% in patients treated with 600 mg clopidogrel (P < .0001). ⁴⁶ In addition to prasugrel’s greater peak platelet antiaggregatory effect, the onset of effect is more rapid. ⁴⁷ The median time to reach ≥20% inhibition of platelet aggregation was 30 minutes for prasugrel and 1.5 hours for clopidogrel (P < .001). Although 24% of the patients treated with clopidogrel had insignificant inhibition of platelet reactivity, all patients responded to prasugrel. ⁴⁷

In the TRIal to assess Improvement in Therapeutic Out-comes by optimizing platelet inhibitioN with Prasugrel-Thrombolysis in Myocardial Infarction (TRITON-TIMI) ⁴⁸ 38 patients with ACS scheduled for PCI were treated with either prasugrel (loading dose 60 mg and daily maintenance dose 10 mg) or clopidogrel (loading dose 300 mg and daily maintenance dose 75 mg) for 6 to 15 months. Approximately 75% of the patients in each group received the study drug either during the PCI procedure or immediately after its completion, while 25% of the patients were treated prior to PCI. Prasugrel resulted in a 19% relative risk reduction in cardiovascular adverse events (P < .001) ⁴⁸ and an overall lower risk of myocardial infarction (P < .0001). ⁴⁹ Similar advantages of prasugrel over clopidogrel were observed in patients with STEMI, including a decrease in stent thrombosis (2.8% for clopidogrel and 1.6% for prasugrel, P = .023). ⁵⁰ However, in patients with unstable angina pectoris or NSTEMI not undergoing revascularization but treated with maintenance doses of prasugrel or clopidogrel, there was no difference in cardiovascular events after a median follow-up of 17 months. ⁵¹ So the important clinical difference between the 2 drugs was seen in those being revascularized.

Ticagrelor Is a Direct P2Y₁₂ Inhibitor

Although prasugrel had a favorable pharmacokinetic and genetic profile, treatment with it as established in the TRITON-TIMI 38 protocol ⁴⁸ could be initiated only after diagnostic angiography had revealed the patient to be a candidate for PCI. That provides little time for effective platelet inhibition before PCI. ⁴⁸ Montalescot and colleagues ⁵² treated patients with NSTEMI with either a 60 mg loading dose of prasugrel just before PCI or 30 mg a median of 4.3 hours before PCI and another dose of 30 mg when it was known PCI was needed. Although only the latter group would have had any platelet inhibition at reperfusion, there was no difference in subsequent cardiovascular complications in the 2 groups. Perhaps there was still no sufficient platelet inhibition at the time of the procedure in patients pretreated with one-half of the standard loading dose, or the ischemic period was too long to allow salvage of the myocardium.

The oral drug ticagrelor is a triazolopyrimidine. It is not a prodrug and is a reversible, noncompetitive inhibitor of the platelet P2Y₁₂ receptor. ⁵³ In healthy volunteers ^53,54 and patients with stable coronary artery disease, ^54,55 loading doses of ticagrelor produced faster and more consistent platelet inhibition than either 300 mg⁵³ or 600⁵⁵ mg of clopidogrel. In patients with stable coronary artery disease, 180 mg ticagrelor led to 41% inhibition of platelet aggregation after 0.5 hours, 88% inhibition at 2 hours, and a maximal inhibition of 93% shortly thereafter. ⁵⁵ Corresponding inhibition by 600 mg clopidogrel was 8%, 38%, and 58%. After 5 days of maintenance treatment with ticagrelor, platelet inhibition ranged from 61% to 100% compared to 11% to 100% after 14 days of maintenance clopidogrel treatment. ⁵³ The rate-controlling step is still the time needed for intestinal absorption. In healthy volunteers, the median t _max for absorption was 1.5 to 3 hours. ⁵³

The PLATelet inhibition and patient Outcomes (PLATO) study in patients with ACS established ticagrelor’s superiority to clopidogrel. ⁵⁶ Ticagrelor decreased occurrence of combined vascular side effects, myocardial infarction, death from vascular causes, and death from any cause (P < .001). The advantage of ticagrelor over clopidogrel was seen regardless of whether the latter’s loading dose was 300 or 600 mg, and regardless of whether an invasive or noninvasive strategy was planned. Ticagrelor resulted in lower event rates throughout the follow-up period of 6 to 12 months. ⁵⁷

In spite of the encouraging clinical data for both prasugrel and ticagrelor, these are still oral preparations. Therefore, there is an obligatory delay between drug administration and biologic effect, probably the result of slow intestinal absorption. Although in healthy volunteers the effects of prasugrel and ticagrelor can be realized 1 to 2 hours after dosing, ⁵⁵ that was not the case in patients with ACS. Parodi et al ⁵⁸ measured platelet reactivity in patients with STEMI at baseline and 2, 4, 8, and 12 hours after loading doses of 60 mg prasugrel or 180 mg ticagrelor. High residual platelet reactivity was found in 44% and 60% of patients loaded with prasugrel and ticagrelor, respectively, at 2 hours. The mean time to achieve clinically significant platelet inhibition was 3 ± 2 hours and 5 ± 4 hours with prasugrel and ticagrelor, respectively. The independent predictor of this delay in biologic effect of the antagonists was morphine use. Possibly morphine slowed intestinal motility.

Thus, in high-risk patients, the antiplatelet effect of the most potent oral agents is still delayed, and PCI is potentially performed without adequate platelet inhibition. This problem is magnified by the desire to shorten door-to-balloon times as much as possible, thus decreasing any pretreatment interval.

Cangrelor Is an Intravenous Antiplatelet Agent

Cangrelor has a chemical structure similar to that of ticagrelor and is an intravenous, reversible blocker of the platelet P2Y₁₂ receptor. It has a half-life of <5 minutes and, like ticagrelor, is a consistent, potent, and rapid inhibitor of platelet aggregation. ⁵⁹ Hence, it could be administered to a patient minutes before PCI with the expectation that platelet inhibition could be satisfactorily effected by the time of stent deployment. Two large clinical trials yielded mixed results. In the Cangrelor versus standard tHerapy to Achieve optimal Management of Platelet InhibitiON (CHAMPION)-PLATFORM trial, the effect of a cangrelor bolus (30 μg/kg) within 30 minutes prior to PCI followed by a 2 to 4-hour infusion (4 μg/kg/min) in patients with unstable angina or NSTEMI was compared to that of intravenous placebo plus 600 mg clopidogrel administered at the end of the PCI. ⁶⁰ Cangrelor-treated patients also received 600 mg clopidogrel at the end of the infusion. At 48 hours, there was no difference between the 2 groups in the composite of death, myocardial infarction, or ischemia-driven revascularization. However, stent thrombosis (P = .02) and death from any cause (P = .02) were decreased by cangrelor. In CHAMPION-PCI, the protocol was modified to compare cangrelor as administered previously to clopidogrel 600 mg given simultaneously with the start of the placebo infusion 30 minutes before PCI in patients with ACS. ⁶¹ Again, there were no differences in the 2 groups at 48 hours. Both trials were designed to detect periprocedural complications of PCI.

Whether either CHAMPION trial revealed any benefit from cangrelor’s rapid onset of effect over that from oral drugs was subsequently shown to depend on the working definition of myocardial infarction. The primary end point in the CHAMPION trials was driven by occurrence of periprocedural myocardial infarction. Because of the difficulty in diagnosing periprocedural myocardial infarction when cardiac biomarkers are already elevated, an expert consensus panel redefined how myocardial infarction should be diagnosed. ⁶² Using this new universal definition, early ischemic events, mainly periprocedural infarcts, in the combined CHAMPION studies were reduced (P = .037), and the incidence of stent thrombosis was halved (P = .018) by cangrelor. ⁶³ The recently completed double-blind, placebo-controlled CHAMPION-PHOENIX trial generally followed the CHAMPION PCI protocol. ⁶⁴ Patients were to undergo either urgent or elective PCI, and the universal definition of myocardial infarction was used. The odds of an early ischemic event were 22% lower with cangrelor than with clopidogrel (P = .005), and a reduction in the rate of periprocedural myocardial infarction accounted for most of the benefit. The rate of stent thrombosis at day 2 was decreased by 43% (P = .01). Thus, cangrelor appears to be an attractive option for treatment of patients undergoing PCI. The CHAMPION trials concentrated on unstable angina and patients with NSTEMI and was not designed to detect a direct anti-infarct effect but rather focused on stent thrombosis and periprocedural reinfarction. At the time of this writing, cangrelor has not received Food and Drug Administration approval.

There are other platelet inhibitors such as the GPIIb/IIIa inhibitors which act at sites other than the platelet’s P2Y₁₂ receptor. They are also quite effective and seem to have a direct anti-infarct effect as well. ⁶⁵

Direct Cardioprotective Effects of Antiplatelet Agents

Aspirin and thienopyridine and triazolopyrimidine P2Y₁₂ antagonists inhibit platelet function and elicit important clinical effects. Because the unifying activity of these agents is platelet inhibition, it has been assumed that these drugs work by blocking thrombosis. Clearly, the effects on stent thrombosis are the result of an antithrombotic action. However, other clinical effects and especially the ability to reduce myocardial infarction and its sequelae may have other explanations. This search for other possibilities has been fueled by a curious anomaly identified in clinical studies of ischemic postconditioning in which the ischemic myocardium is reperfused in a staccato fashion by introducing several brief reperfusion/reocclusion cycles before unimpeded reflow. ^66,67 It is believed that these brief reocclusion periods maintain sufficient tissue acidosis to produce temporary closure of the injurious mitochondrial permeability transition pore (MPTP), ⁶⁸ while the brief reperfusions allow reintroduction of sufficient oxygen to trigger a complex signal transduction pathway ⁶⁷ that leads to long-lasting closure of MPTP and prevention of necrosis.

In a small proof-of-concept study (33 patients), Staat et al ⁶⁹ noted that ischemic postconditioning significantly reduced infarct size compared to that in patients without postconditioning. However, patients were recruited prior to 2005 and approximately half did not receive any antiplatelet agent. And most of those treated with clopidogrel received a suboptimal loading dose of 300 mg. Because of evidence that P2Y₁₂ inhibitors might limit infarct size in animals (see subsequently), these investigators performed a retrospective analysis of their clinical database. ⁷⁰ Although treatment with either clopidogrel or postconditioning alone showed a trend toward smaller infarcts, neither result was significantly different from that seen in the untreated patients. Only the combined clopidogrel pretreatment and postconditioning group had significantly smaller infarcts than the untreated group. Clopidogrel loading seemed as potent as postconditioning in protecting the heart against infarction; when the 2 were combined, there was an unambiguous reduction in infarct size. Did clopidogrel and postconditioning act synergistically through differing mechanisms to provide additive effects or did they protect by the same mechanism but were given in suboptimal doses?

Other clinical studies of postconditioning have yielded conflicting conclusions. Freixa et al ⁷¹ and Tarantini and colleagues ⁷² failed to observe any protective effect of postconditioning. Since all patients in these studies were pretreated with 300 or 600 mg clopidogrel, one could conclude that postconditioning added no additional benefit to clopidogrel loading alone. A third study also showed no effect of ischemic postconditioning in the entire study population, although smaller infarcts were seen in patients with large areas at risk. ⁷³ One can speculate that clopidogrel itself had a maximal protective effect and that any postconditioning protection was pre-empted. Two other studies support the infarct-sparing properties of postconditioning. ^74,75 Clopidogrel was used in the latter studies but its timing of administration was not specified. If platelet inhibition is required at the moment of reperfusion to protect, then clopidogrel’s inhibition of platelets may not have been fully established at the time of PCI leaving the patient receptive to the infarct-sparing effect of postconditioning.

Platelet Inhibitors in Animals

The intriguing possibility of a direct protective effect of antiplatelet agents and even a conditioning action can best be addressed in experimental animals. Unlike the case in most preclinical studies, the platelet antiaggregatory agents described previously have already been demonstrated to have beneficial effects in man undergoing revascularization for acute myocardial infarction. Therefore, the scientific question is not whether they are beneficial but rather what is their mechanism. Animal models of temporary coronary occlusion and subsequent reperfusion closely mimic the clinical PCI scenario and are, therefore, quite useful in this quest.

Barrabés and colleagues ⁷⁶ noted that activated platelets from patients with acute myocardial infarction infused into isolated rat hearts before onset of ischemia/reperfusion increased infarct size, but platelets from healthy volunteers had no effect. Platelets trigger thrombosis following plaque rupture to cause coronary occlusion and myocardial infarction, as well as microembolization of platelet-rich aggregates, platelet-mediated vasoconstriction, and platelet-mediated inflammatory reaction in the ischemic myocardium facilitated by recruitment of neutrophils. ^9,77 P-selectin-deficient hearts have less leukocyte retention following ischemia/reperfusion ^13,77,78 and smaller infarcts. ^13,78 As noted previously, perfusion of previously ischemic hearts with platelets activated in a second animal following ischemia/reperfusion results in deterioration of left ventricular function and larger myocardial infarcts. ^7,13 And mice with deficiencies of either platelet receptor GPVI ^11,12 or signaling protein G_q ¹⁰ have decreased platelet aggregation and smaller infarcts following ischemia/reperfusion. It was hypothesized that blockade of platelet activation leads to smaller infarcts and better function of reperfused myocardium.

Investigations of blockade of platelet aggregation in ischemia/reperfusion in animals have mostly evaluated GPIIb/IIIa antagonists. Kingma et al ⁷⁹ occluded a canine coronary artery for 60 minutes and reperfused it for 3 hours. An experimental GPIIb/IIIa inhibitor which completely blocked in vitro platelet aggregation significantly decreased infarct size when administered at reperfusion. However, because there was no effect on myocardial flow during reperfusion, Kingma et al postulated that the mechanism of protection was not related to blood flow but perhaps a direct protective effect on heart muscle by an unknown action. This was the first suggestion that the salutary effect of a platelet inhibitor might not be related to inhibition of platelet aggregation but might be the result of an as yet unidentified direct cardioprotective effect.

Kunichika and colleagues ⁸⁰ made similar observations in dogs with tirofiban, a GPIIb/IIIa antagonist, which decreased infarct size and increased myocardial blood flow within the risk area. The agent’s utility was attributed to improvement in microvascular flow. Tirofiban also diminished the area of no-reflow in pigs during reperfusion and produced a small decrease in infarct size. ⁸¹ In a canine model of coronary thrombosis followed by angioplasty to effect reperfusion tirofiban administered a few minutes before recanalization improved myocardial blood flow following reperfusion, decreased no-reflow zone, and diminished myocardial infarct size. ⁸² It was concluded that inhibition of platelet aggregation prevented microthromboembolism.

Wang et al studied ticagrelor ⁸³ and cangrelor ⁸⁴ in dogs with coronary thrombosis from combined mechanical and electrolytic injury. Reperfusion was induced by a fibrinolytic agent. Both drugs decreased infarct size and improved tissue blood flow. Cangrelor also lessened the incidence of coronary artery reocclusion, cyclic flow reduction in the injured coronary artery segment, and platelet accumulation in the distal microcirculation. These studies are, however, plagued by 2 shortcomings. Because platelet inhibition would have hastened thrombolysis and thus shortened the ischemic period, it would have obscured any direct cardioprotective effect of the drug on infarction. Also collateral flow was not measured further complicating infarct size analysis. ^85,86

Although these studies support correlation between inhibition of platelet aggregation, improvement in myocardial perfusion, and decreased infarct size, there have been several negative investigations. The deleterious effect of addition of activated pig platelets to perfused, isolated rat hearts undergoing ischemia/reperfusion was not blocked by tirofiban, ⁷ and a second GPIIb/IIIa inhibitor had no effect on infarct size in pigs experiencing ischemia/reperfusion despite its prevention of cyclic flow reduction in the reperfused coronary artery. ⁸⁷ In both of these studies, the pharmacologic agent produced profound inhibition of platelet aggregation. Therefore, the effect on platelets seemed to be divorced from the effect on infarct size. As already noted, when this same group of investigators added activated platelets from patients with acute myocardial infarction to isolated rat hearts undergoing ischemia/reperfusion, myocardial injury was increased. ⁷⁶ However, the injury appeared to be significantly attenuated if the collected activated platelets were first mixed with either abciximab or clopidogrel. But clopidogrel is a prodrug and could not have been metabolized in vitro to its active metabolite. Therefore, any observed improvement must have been a nonspecific effect related to the preparation rather than a result of the drug’s antiplatelet activity. In 2 additional studies in which tirofiban ⁸¹ and ticagrelor ⁸³ demonstrated cardioprotective effects, clopidogrel was ineffective.

Looking for a Direct Cardioprotective Effect of P2Y₁₂ Blockers

These preclinical studies demonstrate a profound anti-infarct effect of several classes of platelet inhibitors. But there is poor consensus on the mechanism of action of these agents. The prevailing hypothesis that these antiplatelet agents are having beneficial effects because of their antithrombotic action is unproved. Clinical studies cited previously demonstrating decreased stent thrombosis in patients treated with antiplatelet agents implicate an antithrombotic effect. But is there something else? Is there a direct protective pleiotropic effect as postulated by Kingma et al? ⁷⁹ Is there a dissociation between an antiplatelet effect and cardioprotection?

We evaluated a variety of platelet inhibitors in rabbits subjected to 30-minute coronary occlusion/3-hour reperfusion. ⁸⁸ An intravenous bolus of 60 μg/kg cangrelor administered several minutes before reperfusion followed by a constant infusion of 6 μg/kg/min decreased platelet aggregation by more than 94%. Although 38% of the ischemic zone infarcted in control rabbits, cangrelor resulted in significant salvage of reperfused myocardium and only 19% of the risk zone infarcted, similar to protection of ischemic postconditioning. Cardioprotection following ischemic postconditioning and that initiated by cangrelor were similar. Protection was abolished if either the first reocclusion of postconditioning ⁶⁷ or administration of the cangrelor bolus was delayed until 10 minutes after release of the 30-minute index coronary occlusion. This observation would explain why a beneficial effect in the clinical trials seemed to correlate with the degree of inhibition of platelet aggregation at the time of stent deployment.

Postconditioning’s protection is a product of signal transduction, and a number of key signaling components have been identified in its pathway. ⁸⁹ We tested 7 inhibitors known to block signaling required for postconditioning’s protective effect: wortmanin and LY294002 (phosphatidylinositol 3 kinase/Akt antagonists), PD98059 (antagonist of mitogen-activated protein ERK kinase [MEK] 1/2, and, therefore, extracellular signal-regulated protein kinases 1 and 2), 5-hydroxydecanoic acid (putative blocker of mitochondrial K_ATP channels), 8-(p-sulfophenyl)theophylline (non-selective antagonist of adenosine receptors), MRS1754 (selective antagonist of adenosine A_2B receptors), and 2-mercaptopropionylglycine (scavenger of reactive oxygen species and blocker of redox signaling). All abolished cangrelor’s protection (Figure 2). It is important to note that none restored platelet reactivity. ⁸⁸

Figure 2.

Infarct size in open chest rabbits expressed as a percentage of the risk zone after 30-minute regional ischemia and 180-minute reperfusion. Cangrelor infusion reduced infarct size but its protection was abolished by Wort, LY, PD, SPT, MPG, or 5HD. All of these block protection from pre- and postconditioning but none affects infarct size in untreated hearts. *P < .001. Reprinted with permission from Rösen et al ⁸⁸ . Wort indicates wortmannin; LY, LY294002; PD, PD98059; SPT, 8-(p-sulfophenyl)theophylline; MPG, 2-mercaptopropionylglycine; 5HD, 5-hydroxydecanoic acid.

The observation that postconditioning and cangrelor-induced platelet blockade have identical kinase and receptor “fingerprints” strongly supports the hypothesis that the signaling and mechanism of protection of the 2 interventions are the same. This assumption is further supported by the lack of additive protection when cangrelor infusion and ischemic postconditioning are applied to the same heart. ⁸⁸ Hence, we propose that cangrelor is a bona fide-conditioning agent with the direct myocardial cardioprotective action hypothesized by Kingma et al ⁷⁹ in 2000. Since all of the antagonists and receptor blockers noted previously which aborted cangrelor’s cardioprotective effect had no influence on the drug’s platelet antiaggregatory effect, signal transduction rather than any effect on aggregation per se causes the protection. Yet we found that the magnitude of protection was correlated with the degree of suppression of aggregation indicating that P2Y₁₂ blockade was common to both processes.

We also examined other antiplatelet and anticoagulant agents for possible cardioprotective activity. Clopidogrel, 15 mg/kg, had to be fed to rabbits for 2 days before significant platelet blockade was achieved. It also limited infarct size, and protection was blocked by either wortmanin or MRS1754 administered just before reperfusion indicating signal transduction was involved. Neither inhibitor interfered with inhibition of aggregation further suggesting that prevention of coagulation was not the cause of protection. ⁸⁸ Aspirin blocked platelet aggregation from low-dose arachidonic acid but not from ADP. It did not significantly affect infarction. Heparin actually tended to increase infarct size. ⁸⁸

We additionally studied cangrelor in primate ⁹⁰ and rodent ⁹¹ hearts. In both species, cangrelor was very protective when administered just before reperfusion. In macaque hearts, cangrelor’s protective effect was equivalent to that of ischemic postconditioning, ⁹⁰ whereas in rats cangrelor and ischemic preconditioning, an intervention in which the brief cycles of ischemia/reperfusion precede the prolonged coronary occlusion, were equipotent. ⁹¹ In monkeys, an antibody to platelet GPVI also decreased infarct size, ⁹⁰ indicating an additional intervention which both diminished platelet aggregation and was cardioprotective.

We examined effects of ticagrelor, a third P2Y₁₂ receptor antagonist, in rats treated with 20 mg/kg by gavage 2 hours before coronary occlusion. ⁹¹ Ticagrelor blunted peak platelet aggregation by 84%. It also decreased infarction from 45% of the risk zone in control animals to 26%, and this cardioprotective effect was abolished by wortmanin and chelerythrine, a protein kinase C antagonist, both of which block conditioning’s protection. ⁸⁹ Neither interfered with ticagrelor’s inhibitory effect on platelet reactivity.

Thus, several pharmacologic and biologic interventions that blocked platelet aggregation were also infarct sparing. Curiously, however, although the antiplatelet effect linked all of the agents, it was not the determining factor for the cardioprotective effect. Cangrelor in isolated rabbit hearts perfused with platelet-free Krebs buffer was not protective, suggesting that some blood element, presumably platelets, is somehow involved. ⁸⁸ It is difficult to understand how receptor inhibition could trigger an intracellular signaling cascade that eventually leads to cardioprotection. Further investigations are ongoing.

Combining Antiplatelet Drugs With Other Cardioprotective Interventions

Ischemic preconditioning ⁹¹ or postconditioning ⁸⁸ offered no additive protection against infarction when combined with cangrelor. We assume that animals (or patients) properly treated with a P2Y₁₂ inhibitor are already in a conditioned state and attempts to further condition them will be ineffective. However, we have found other known protective interventions that can be combined with cangrelor to produce even more potent protection. ⁹¹ Both mild whole-body hypothermia (32°C-35°C) ^{92 –94} and cariporide, ⁹⁵ a Na⁺-H⁺ exchange blocker, reduce infarct size. But unlike conditioning and cangrelor, mild hypothermia and cariporide are most protective when applied during ischemia rather than reperfusion. Accordingly, protection of ischemic preconditioning could be added to that of cooling, ⁹² and protective effect of pharmacological postconditioning with AMP579 could be added to that of cariporide. ⁹⁵ In rats, peritoneal lavage with ice-cold saline 10 minutes before coronary occlusion lowered blood temperature to 32°C to 33°C and decreased infarction to 25% of risk zone, equivalent to that seen with cangrelor treatment just before reperfusion. ⁹¹ Combination of the 2 halved infarction to 14% of myocardium at risk. Cariporide’s protective effect (27% infarction) is also comparable to cangrelor’s. Combination of cangrelor and cariporide nearly halved infarction to 16% of the risk zone. Combination of all 3 interventions again halved infarction to only 6% of the risk zone.

Conclusions

These preclinical investigations raise several issues which have potentially important clinical implications. First, we believe that the antiplatelet agents that we have evaluated (clopidogrel, ticagrelor, and cangrelor) are truly conditioning agents. And, if true, this may help to explain the variable results of recent clinical trials of ischemic postconditioning and postconditioning-mimetic agents. The evidence indicates that if patients are treated with an optimal dose of an antiplatelet agent, their myocardium will already be protected. So addition of other conditioning interventions such as ischemic postconditioning or cyclosporine ⁹⁶ should have little additional effect. Future studies should concentrate on optimizing the dose and timing of existing antiplatelet agents to be sure platelet aggregation is inhibited at the time of stent deployment. The slow onset of action with oral drugs could be alleviated by use of an intravenous preparation like cangrelor which produces a maximal antiplatelet effect quickly. Additionally, if a patient pretreated with cangrelor must be sent for urgent surgical revascularization, cessation of the infusion restores normal platelet reactivity in minutes. Of course, after treatment with cangrelor during the revascularization procedure, long-term treatment would then be maintained with an oral agent.

Second, today’s patients with ACS are already enjoying the benefit of the platelet inhibitors but myocardial infarction has still not been abolished. So what is next? Unfortunately, further shortening of door-to-balloon times may not produce additional benefits. ⁹⁷ Instead we propose that interventions that can yield additional protection on top of that produced by a properly administered P2Y₁₂ blocker be sought. The 2 interventions discussed previously are currently available. Ultra-rapid cooling can be instituted in the catheterization suite (Velocity trial, Clinicaltrials.gov, NCT01655433). Although hypothermia obviously can’t be initiated before the onset of ischemia, application during early ischemia can still be beneficial. ⁹² Also since hypothermia protects through signaling, a pharmacologic agent may be found to replace the mechanical cooling process. Although cariporide increased the incidence of stroke when administered to surgical patients over a prolonged period, ⁹⁸ other Na⁺-H⁺ exchange blockers have no reported safety issues and could possibly even be administered by emergency medical personnel early in the ischemic period. Cardioprotection in the future may be initiated with a combination of protective interventions, all with additive effects.