OUP user menu

Current antiplatelet options for NSTE-ACS patients

G. Cayla , J. Silvain , S.A. O’Connor , J.-P. Collet , G. Montalescot
DOI: http://dx.doi.org/10.1093/qjmed/hcs077 935-948 First published online: 28 April 2012


Non-ST elevation (NSTE) myocardial infarction and unstable angina are the most common clinical presentations of acute coronary syndrome (ACS). Platelet activation is central to the pathogenesis of NSTE-ACS and consensus guidelines that advocate early revascularization supported by intensive antiplatelet therapy. This review examines the drugs used concurrently with aspirin as dual antiplatelet therapy in the NSTE-ACS setting. Clopidogrel represented an important therapeutic advance. However, variations in platelet response and a relatively slow onset of action compromise outcomes with clopidogrel. Evidence reviewed in this article shows that in NSTE-ACS patients, ticagrelor and prasugrel are more effective than clopidogrel and are relatively well tolerated, with an acceptable and manageable bleeding risk. The literature suggests several differences between ticagrelor and prasugrel that should allow clinicians to better tailor treatment to the patient. Head-to-head comparisons are now needed to compare directly the risks and benefits of ticagrelor and prasugrel in NSTE-ACS. Further studies also need to address other outstanding issues such as the benefits and risks of prasugrel pre-treatment and to stratify efficacy and tolerability according to diabetes mellitus (DM) and other co-morbidities. In the meantime, the issues discussed in this review should enhance clinicians’ ability to optimize and individualize NSTE-ACS treatment, thereby further reducing the morbidity and mortality associated with this common cardiovascular condition.


Despite advances in prevention, diagnosis and management, driven by an unprecedented understanding of the pathogenesis underlying acute coronary syndromes (ACS), myocardial infarction (MI) remains a common cause of death, disability, poor quality of life and preventable health-care expenditure worldwide.1 (MI refers to myocardial cell death following prolonged ischemia that is manifest as biochemical markers of myocyte necrosis in the blood.2) Non-ST elevation (NSTE) MI and unstable angina are the most common clinical presentations of ACS.3

The annual incidence of NSTE-ACS seems to be higher than ST elevation (STE) ACS. The annual incidence of hospital admissions for NSTE-ACS is approximately 3 per 1000 of the European population.4 The natural history of NSTE-ACS and STEMI also differs. An analysis of the OPERA registry, which followed 1878 patients for 1 year, found that 36.7% of STEMI and 41.5% of NSTEMI patients were re-hospitalized, whereas 16% of both groups underwent revascularization procedures. In-hospital mortality was similar in the two groups (4.6 and 4.3%, respectively), whereas 1-year mortality was 9.0 and 11.6% in STEMI and NSTEMI patients, respectively (P = 0.09, log-rank).5 Other analyses suggest that in-hospital mortality is higher following STEMI (7%) than NSTE-ACS (3–5%). While mortality is similar after 6 months (12 and 13%, respectively), over the longer term death rates were higher among those with NSTE-ACS than with STE-ACS. Indeed, after 4 years, mortality in NSTE-ACS patients was double than that in STE-ACS.4

The discordant natural history between NSTE-ACS and STE-ACS may reflect differences in patient characteristics, pathophysiological variations or both. For example, NSTE-ACS patients tend to be older and have more co-morbidities, especially diabetes mellitus (DM) and renal failure, than those with STEMI.4

Against this background, revascularization of NSTE-ACS patients aims to relieve symptoms as well as improve short- and long-term prognosis. Numerous studies suggest that early invasive treatment of NSTE-ACS improves prognosis compared with a conservative strategy, reducing the risk of severe recurrent ischemia, rehospitalization and repeat revascularization. Indeed, early revascularization reduces cardiovascular death and MI during a 5-year follow up.3 As a result, guidelines from Europe4 and the USA6 advocate early revascularization supported by intensive antiplatelet therapy in patients presenting with NSTE-ACS.

Platelet activation and aggregation propagate thrombosis following acute plaque rupture. As a result, guidelines advocate starting antiplatelet therapy ‘as early as possible’ after the diagnosis of NSTE-ACS to reduce the risk of ischemic complications and recurrent atherothrombosis.4 This review examines the drugs (clopidogrel, ticagrelor and prasugrel) used concurrently with aspirin as dual antiplatelet therapy (DAPT) in the NSTE-ACS setting.


Clopidogrel, a thienopyridine P2Y12 inhibitor, represented an important advance in the acute and maintenance treatment of NSTE-ACS. Consensus guidelines endorse the use of thienopyridines in this setting.4,6 Nevertheless, clopidogrel has certain clinically significant limitations, such as larger inter-patient variability in antiplatelet response and slower onset of action when compared with the more recently introduced antiplatelet drugs discussed later in this review (Table 1).

View this table:
Table 1

Comparison of the characteristics of clopidogrel, prasugrel and ticagrelor31,36

Drug classThienopyridineThienopyridineCyclopentyltriazolopyrimidine
Oral administrationYesYesYes
Loading dose (mg)30060180
Maintenance dose (mg)751090
Frequency of administrationOnce dailyOnce dailyTwice daily
Onset of actionDelayedRapidRapid
Offset of actionDelayedDelayedRapid
Individual variabilityLargeSmallSmall
CYP-450 activationYes (twice)YesNo
Irreversible P2Y12 inhibitionYesYesNo
Relative potencyLowHighHigh
Mean platelet inhibition (%)∼50∼70∼95
Time to peak inhibition (h)∼12a22
Half life (h)Life of plateletLife of platelet7–12
Recommended time delay between last drug dose and CABG (days)>5>7>3 > 4b
  • aWith 300 mg loading dose.

  • bTicagrelor should be discontinued 7 days before elective surgery when an antiplatelet effect is not desired36.

The CURE trial

The CURE trial exemplifies clopidogrel’s efficacy in NSTE-ACS. CURE compared clopidogrel (300 mg, then 75 mg once daily) added to aspirin with aspirin monotherapy in 12 562 NSTE-ACS patients who presented within 24 h of symptom onset. Clopidogrel reduced the prevalence of the primary outcome—a composite of cardiovascular death, non-fatal MI or stroke—compared with aspirin monotherapy [9.3 and 11.4%, respectively; relative risk (RR) 0.80, 95% confidence interval (CI) 0.72–0.90; P < 0.001]. Clopidogrel also reduced the prevalence of a further composite endpoint—the first primary outcome or refractory ischemia—to a similar extent to that seen with the primary endpoint (16.5 and 18.8%, RR 0.86, 95% CI 0.79–0.94; P < 0.001).7

CURE was inadequately powered to examine outcomes in individual subgroups. Nevertheless, more than 30 subanalyses consistently showed that adding clopidogrel to aspirin and other standard therapies reduced the prevalence of the primary outcome across a range of patients including those7:

  • with (11.5 and 14.3%) and without ST-segment deviation (7.0 and 8.6%).

  • with (11.3 and 13.7%) and without associated MI (8.6 and 10.6%).

  • with (14.2 and 16.7%) and without DM (7.9 and 9.9%).

Moreover, fewer patients receiving clopidogrel developed severe ischemia (2.8 and 3.8%, respectively; RR 0.74, 95% CI 0.61–0.90; P = 0.003), recurrent angina (20.9 and 22.9%, RR 0.91, 95% CI 0.85–0.98; P = 0.01) or radiologic heart failure (3.7 and 4.4%, RR 0.82; 95% CI 0.69–0.98; P = 0.03) than with aspirin monotherapy.7

Clopidogrel reduced the risk of cardiac events within a few hours after administration and produced a durable response in the CURE trial. The rate of cardiovascular mortality, nonfatal MI, stroke or refractory or severe ischemia was lower with clopidogrel (1.4%) compared with aspirin monotherapy (2.1%) 24 h after randomization (RR 0.66, 95% CI 0.51–0.86). Clopidogrel reduced the rate of the first primary outcome both within the first 30 days of treatment (RR 0.79, 95% CI 0.67–0.92) and between 30 days and the end of the 12 month study (RR 0.82, 95 CI 0.70–0.95). While significantly more patients developed major bleeding with clopidogrel (3.7%) than with aspirin alone (2.7%, RR 1.38; P = 0.001), rates of life-threatening bleeding and hemorrhagic strokes were not higher with DAPT.7

A further analysis stratified patients based on thrombolysis in myocardial infarction (TIMI) risk score: low risk 0–2; intermediate risk 3–4 and high risk 5–7. As expected, the rate of the primary endpoint rose markedly with increasing TIMI risk from 3.5% of those with scores of 0–1 to 22.7% with scores of 6–7. Clopidogrel consistently reduced the rate of the primary outcome compared with aspirin monotherapy in low risk (4.1 and 5.7%, respectively; RR 0.71, 95% CI 0.52–0.97; P < 0.04); intermediate risk (9.8 and 11.4%, RR 0.85, 95% CI 0.74–0.98; P < 0.03) and high-risk patients (15.9 and 20.7%, RR 0.73, 95% CI 0.60–0.90; P < 0.004).8


CURE illustrates that while standard-dose clopidogrel reduces the rate of major cardiovascular events, a considerable residual risk remains. Therefore, CURRENT-OASIS 7 compared 7 days treatment with double-dose clopidogrel (600 mg on Day 1, 150 mg on Days 2–7, then 75 mg daily) and standard-dose (300 mg on Day 1, then 75 mg daily) in 25 086 ACS patients (62.8–63.4% NSTEMI or unstable angina) expected to undergo early percutaneous coronary intervention (PCI). Patients also received either high- (300–325 mg daily) or low-dose (75–100 mg daily) aspirin. The primary outcome was cardiovascular death, MI or stroke after 30 days.9

Double dose clopidogrel did not significantly reduce the rate of the primary outcome (4.2%) compared to the standard dose [4.4% hazard ratio (HR) 0.94, 95% CI 0.83–1.06; P = 0.30]. However, major bleeding was significantly more common with double-dose clopidogrel (2.5%) than among patients receiving the standard regimen (2.0%, HR 1.24, 95% CI 1.05–1.46; P = 0.01). High-dose aspirin did not significantly alter the rate of either the primary outcome (4.2 and 4.4%, respectively; HR 0.97, 95% CI 0.86–1.09; P = 0.61) or major bleeding (2.3 and 2.3%, respectively; HR 0.99, 95% CI 0.84–1.17; P = 0.90) compared with low dose.9

In a subgroup analysis of 17 263 patients who underwent PCI, double-dose clopidogrel reduced the rate of the primary outcome (3.9 and 4.5%, respectively; HR 0.86, 95% CI 0.74–0.99; P = 0.039) and the proportion of patients who developed definite stent thrombosis (0.7 and 1.3%, respectively; HR 0.54, 95% CI 0.39–0.74; P = 0.0001) compared with the standard dose. Rates of the primary outcome did not differ significantly in those taking high- or low-dose aspirin (4.1 and 4.2%, HR 0.98, 95% CI 0.84–1.13; P = 0.76). Double-dose clopidogrel also reduced rates of the primary outcome compared with the standard dose when patients were stratified according to diagnosis: unstable angina or NSTEMI (3.6 and 4.2%, respectively; HR 0.87, 95% CI 0.72–1.06; P = 0.167) and STEMI (4.2 and 5.0%, respectively; HR 0.83, 95% CI 0.66–1.05; P = 0.117), although neither difference reached statistical significance. Major bleeding was more common with double dose (1.6%) than with standard-dose clopidogrel (1.1%, HR 1.41, 95% CI 1.09–1.83; P = 0.009). However, there was no difference in bleeding rates between high- (1.5%) and low-dose aspirin (1.3%, HR 1.18, 95% CI 0.92–1.53; P = 0.20).10

In CURRENT-OASIS 7, the difference between standard- and double-dose clopidogrel reached statistical significance only in patients who underwent PCI. However, patients received double-dose clopidogrel before PCI and treatment lasted 7 days, which does not reflect current clinical practice (see below). Therefore, this study does not support waiting for coronary anatomy results before instigating clopidogrel.

Other studies confirm that clopidogrel improves prognosis among people who present with NSTE-ACS. For example, the North American CRUSADE study, which enrolled 93 045 people presenting with NSTEMI, confirmed that using clopidogrel within 24 h of hospital admission reduced adjusted in-hospital mortality compared with those who did not receive clopidogrel on admission [odds ratio (OR) 0.68, 95% CI 0.61–0.77], despite patients not undergoing PCI. The rate of major bleeding among patients not undergoing coronary artery by-pass grafting (CABG) was 9.5% in both groups.11

Impaired responses to clopidogrel

While clopidogrel markedly improved outcomes in NSTE-ACS patients, it has certain limitations. First, clopidogrel’s benefits do not emerge until several hours after administration.12 Second, clopidogrel responsiveness shows marked inter-patient variation. For example, Buonamici et al. enrolled 804 patients successfully implanted with sirolimus- or paclitaxel-eluting stents who received a loading dose of 600 mg clopidogrel. All patients also received maintenance DAPT (aspirin 325 mg and clopidogrel 75 mg daily) for 6 months, at which time 3.1% had developed definite or probable stent thrombosis. Thirteen percent were nonresponders [defined as ≥70% platelet aggregation with 10 μmol adenosine diphosphate (ADP)]. The incidence of stent thrombosis was 8.6% in nonresponders and 2.3% in responders (adjusted HR 3.08, 95% CI 1.32–7.16; P = 0.009).13

Suboptimal responses to clopidogrel probably arise from interactions between genetic polymorphisms as well as cellular and clinical factors (Figure 1).14 Studies to date suggest that polymorphisms in the gene-encoding cytochrome p450 (CYP) 2C19 are the most important genetic factor influencing an individual patient’s response to clopidogrel, reflecting the drug’s metabolic pathways. Most of a clopidogrel dose undergoes de-esterification into inactive metabolites. Approximately 15% is converted into an active metabolite by, predominately, CYP1A2, CYP3A4/5, CYP2C9 and CYP2C19.15 Therefore, genetic polymorphisms in these enzymes potentially contribute to differences in clopidogrel responsiveness.

Figure 1.

Proposed mechanisms leading to variability in individual responsiveness to clopidogrel.14

Against this background, Hulot et al. enrolled 20 healthy subjects who were wild-type homozygotes for CYP2C19 (*1/*1) and eight subjects who were heterozygous for the loss-of-function polymorphism CYP2C19*2 (*1/*2). CYP2C19 genotype did not influence baseline platelet activity. However, platelet aggregation induced by 10 μM ADP decreased gradually during 7 days treatment with clopidogrel 75 mg once daily in *1/*1 subjects, reaching 48.9% ± 14.9% (P < 0.001). In contrast, platelet aggregation did not change significantly in *1/*2 subjects (71.8% ± 14.6%; P < 0.22 vs. baseline and P < 0.003 vs. *1/*1).16

Similarly, Brandt et al. reported that the CYP2C19*2 loss of function variant significantly reduced the area under the concentration-time curve (AUC024; P = 0.004) and maximal plasma concentration (Cmax; P = 0.020) of clopidogrel’s active metabolite. Moreover, CYP2C19*2 patients showed lower inhibition of platelet aggregation at 4 h (P = 0.003) and were more likely to meet the criteria for poor responders (P = 0.030). CYP2C9 loss of function variants (*2/*2 or *3) also predicted lower AUC024 (P = 0.043) and Cmax (P = 0.006), impaired inhibition of platelet aggregation (P = 0.046) and increased numbers of poor responders (P = 0.024). Overall, 72.7% of the poor responders carried either CYP2C19*2 or CYP2C9 *2/*2 or *3. Only 27.3% of the responders showed these genotypes (P < 0.001). In contrast, loss of function CYP2C19 or CYP2C9 polymorphisms influence neither the extent of exposure to the active metabolite of prasugrel nor the pharmacodynamic response.17

As a final example underscoring the importance of CYP2C19 polymorphisms, Collet and colleagues enrolled 259 patients aged <45 years who survived their first MI and were treated with clopidogrel for at least 1 month (median 1.07 years). The primary composite endpoint (death, MI and urgent coronary revascularization during clopidogrel treatment) was more common in patients expressing CYP2C19*2 (15 events) than people without this polymorphism (11 events; HR 3.69, 95% CI 1.69–8.05; P = 0.0005). Stent thrombosis (eight and four events, respectively; HR 6.02, 95% CI 1.81–20.04; P = 0.0009) was also more common among CYP2C19*2 carriers. The detrimental effect associated with CYP2C19*2 persisted from 6 months after the start of clopidogrel treatment until the end of follow-up (HR 3.00, 95% CI 1.27–7.10; P = 0.009). On multivariate analysis, CYP2C19*2 was the only genetic variant that independently predicted cardiovascular events (HR 4.04, 95% CI 1.81–9.02; P = 0.0006).18

Nevertheless, certain other polymorphisms—such as those in ABCB1 (also known as MDR1), which encodes P-glycoprotein—also potentially influence clopidogrel responsiveness.19 Simon et al. examined the pharmacogenomic profile of 2208 acute MI patients receiving clopidogrel. None of the single-nucleotide polymorphisms evaluated in CYP3A5, P2RY12 or ITGB3, which encodes GPIIIa, were associated with adverse outcomes. However, patients expressing two variant alleles of ABCB1 (TT at nucleotide 3435) had a higher rate of cardiovascular events than those with the CC wild-type (15.5 and 10.7%, respectively; adjusted HR 1.72, 95% CI 1.20–2.47). The cardiovascular event rate was also higher (21.5%) in patients carrying any two CYP2C19 loss-of-function alleles (*2, *3, *4 or *5) than in those with none (13.3%; HR 1.98, 95% CI 1.10–3.58). The event rate among 1535 patients who underwent PCI was 3.58-fold higher (95% CI 1.71–7.51) in people with two CYP2C19 loss-of-function alleles.20

Such findings raise the prospect of individualizing clopidogrel regimens based on genotype, although further studies are needed before pharmacogenetic testing becomes routine in clinical practice. Furthermore, despite the prospect of stratifying patients based on genotype, improving antiplatelet therapy seems to be important to reduce the residual risk among NSTE-ACS patients further than that produced by clopidogrel. The newer antiplatelet agents (prasugrel and ticagrelor), discussed in the following sections, are more potent than clopidogrel, produce less variable antiplatelet response and show a faster onset of action, and appear to meet this need.

Reversal of clopidogrel therapy

Patients who are taking antiplatelets may need to stop or temporarily suspend their treatments for different reasons, including undergoing surgery (to reduce the risk of bleeding). In order for platelets to recover normal function after stopping clopidogrel, a certain amount of time needs to be allowed for inhibited platelets to be replaced with uninhibited platelets released from the bone marrow.21

A recent cohort study assessed both platelet function recovery and the reversal of antiplatelet effects with the use of donor platelets in patients who stopped treatment with either aspirin or clopidogrel (two separate cohort groups). In patients who had been taking clopidogrel, it took 10 days for platelet function to fully recover, vs. 4 days for patients who had been taking aspirin. In the cohort receiving donor platelets, light transmission aggregation (LTA) in patients given aspirin returned to normal after adding 30% donor platelets, whereas in the clopidogrel group it took the addition of 90% donor platelets for control levels to be achieved.21


Similar to clopidogrel, prasugrel is also a member of the thienopyridine class of antiplatelet drugs. Prasugrel’s active metabolite irreversibly inactivates P2Y12 receptors through the covalent linkage of a sulphydryl group. Clopidogrel and prasugrel are equipotent at inhibiting P2Y12 receptors. However, prasugrel is clinically more efficacious than clopidogrel and produces less variable antiplatelet response, with a faster onset of action.22 Prasugrel’s active metabolite attains Cmax around 30 min after ingestion, while the elimination half-life for the unbound fraction is ∼7 h (Table 1).23

Prasugrel’s advantages over clopidogrel arise from differences in the metabolic pathways that convert the parent molecule into the active metabolite. Esterases in the gut wall, liver and plasma rapidly hydrolyse prasugrel into a thiolactibe (R-95913) that undergoes oxidation by CYP isoenzymes yielding an active thiol metabolite (R-138727). Prasugrel’s metabolites undergo renal excretion.24 CYP3A4 and, to a lesser extent, CYP2B6 convert the thiolactibe into the active metabolite, with CYP2C9 and CYP2C19 making minor metabolic contributions.15 Indeed, polymorphisms in CYP2C9, CYP2C19 1717 and ABCB1 do not significantly alter prasugrel’s clinical efficacy, pharmacokinetics or pharmacodynamics.19 These metabolic differences result in levels of prasugrel’s active metabolite that are ∼10-fold higher than the equivalent concentrations following clopidogrel.22


TRITON-TIMI 38—which enrolled 10 074 subjects with unstable angina or NSTEMI and 3534 STEMI patients—is currently the largest study comparing prasugrel and clopidogrel. TRITON-TIMI 38 was designed to enrol a larger proportion of patients with unstable angina or NSTEMI than STEMI. Therefore, the analysis of the patients presenting with unstable angina or NSTEMI would largely determine whether or not differences in primary and secondary endpoints between the prasugrel and clopidogrel groups would attain statistical significance. Patients, who were thienopyridine naive, randomly received prasugrel (60 mg loading dose, 10 mg maintenance dose) or clopidogrel (300 mg loading dose, 75 mg maintenance dose) after their coronary angiogram and only if the clinician planned PCI.25 This contrasts with the PLATO study (see below), during which patients received ticagrelor before the angiogram.

Prasugrel reduced the risk of the primary efficacy endpoint (CV death, nonfatal MI or nonfatal stroke) compared with clopidogrel by 18% in patients with unstable angina or NSTEMI (9.9 and 12.1%, respectively; HR 0.82, 95% CI 0.73–0.93; P = 0.002). Prasugrel (10.0%) was also more effective than clopidogrel (12.4%) in patients presenting with STEMI (HR 0.79, 95% CI 0.65–0.97; P = 0.02). No significant interaction emerged between outcomes with the two treatments and clinical presentation (unstable angina or NSTEMI compared with STEMI). In the overall ACS group, prasugrel reduced rates of MI (9.7 and 7.4%, respectively; P < 0.001) and urgent target vessel revascularization (3.7 and 2.5%, respectively; P < 0.001) compared with clopidogrel.25

Prasugrel also decreased the likelihood of stent thrombosis (2.4 and 1.1%; P < 0.001) in patients treated with bare-metal stents only (HR 0.52, 95% CI 0.35–0.77; P < 0.001) and in those who received at least one drug-eluting stent (HR 0.43, 95% CI 0.28–0.66; P < 0.001).25 26 In a subgroup analysis, the risk of definite or probable stent thrombosis was 57% lower (1.0 and 2.2%, respectively) in people with unstable angina or NSTEMI. The benefit in patients with unstable angina or NSTEMI was even more pronounced than in the STEMI population (42% reduction).26

TRITON-TIMI 38 suggested that prasugrel is associated with a risk of bleeding that is manageable but still higher than that seen with clopidogrel. Major bleeding occurred in 2.4% and 1.8% of the prasugrel and clopidogrel groups, respectively (HR 1.32, 95% CI 1.03–1.68; P = 0.03). Life-threatening (1.4 and 0.9%, respectively; HR 1.52; P = 0.01), nonfatal (1.1 and 0.9%, respectively; HR 1.25; P = 0.23) and fatal bleeding (0.4 and 0.1%, respectively; HR 4.19; P = 0.002) were more common during treatment with prasugrel than clopidogrel. Post hoc analyses identified three patient groups at particularly high risk of bleeding: patients with a previous stroke or transient ischemic attack; patients aged ≥75 years and patients weighing <60 kg. After excluding these high-risk patients, no significant difference emerged in the rate of major bleeding between prasugrel and clopidogrel (HR 1.24, 95% CI 0.91–1.69; P = 0.17).25

As mentioned above, NSTE-ACS patients tend to have more co-morbidities, such as DM, than people presenting with STEMI.4 A subgroup analysis of TRITON-TIMI 38 reported that prasugrel significantly reduced the incidence of the primary endpoint compared with clopidogrel among nondiabetics (9.2 and 10.6%, respectively; HR 0.86; P = 0.02) and DM patients (12.2 and 17.0%, respectively; HR 0.70; P < 0.001, Pinteraction = 0.09). No significant interaction emerged between outcome and the presenting ACS: the risk reduction was 30% in those with unstable angina or NSTEMI and 29% in STEMI.27

Prasugrel also reduced the incidence of the primary endpoint compared with clopidogrel in DM subjects taking insulin (14.3 and 22.2%, respectively; HR 0.63; P = 0.009) and those not on insulin (11.5 and 15.3%, respectively; HR 0.74; P = 0.009). Nondiabetics taking prasugrel were more likely than those receiving clopidogrel to develop major hemorrhage (2.4% vs. 1.6%, HR 1.43; P = 0.02). Rates of major hemorrhage with clopidogrel and prasugrel were similar in DM patients (2.6 and 2.5%, respectively; HR 1.06, P = 0.81, Pinteraction = 0.29). Therefore, prasugrel produced a greater (14.6%) net clinical benefit (composite of all-cause mortality, nonfatal MI, nonfatal stroke or nonfatal TIMI major bleeding not related to CABG) than clopidogrel in DM patients (19.2%, HR 0.74; P = 0.001) than in those without DM (11.5 and 12.3%, respectively; HR 0.92; P = 0.16, Pinteraction = 0.05).27

As mentioned above, CYP2C9 and CYP2C19 make minor contributions to the metabolic conversion of prasugrel to the active metabolite. Consistent with this metabolic pathway, loss of function genotypes of CYP2C9 or CYP2C19 did not influence either exposure to prasugrel’s active metabolite or the pharmacodynamic profile.17 Similarly, ABCB1 did not significantly alter clinical or pharmacological outcomes in ACS patients or healthy individuals treated with prasugrel.19

Reversal of prasugrel therapy

There are no current data on the reversal of prasugrel therapy. The summary of product characteristics for prasugrel recommends platelet transfusion for patients who need the reversal of the effects of prasugrel due to excessive bleeding.28


Ticagrelor, a cyclopentyltriazolopyrimidine, is a direct acting, reversible antagonist that binds competitively with 2Me-S-ADP at a different region of the P2Y12 receptor to the thienopyridines. Ticagrelor is not a prodrug.29 Nevertheless, the onset of action of prasugrel and ticagrelor is similar. Typically, a loading dose of 60 mg prasugrel produces a maximum 60–70% inhibition of platelet function within 2–4 h. Ticagrelor (180 mg loading dose) results in an average 50–60% inhibition of maximal platelet aggregation induced by ADP after 2–4 h. Furthermore, enzymatic degradation of ticagrelor after oral administration yields at least one active metabolite, which has similar pharmacokinetics to the parent molecule. The steady-state AUC for the active metabolite is ∼35% that of the parent compound. Cmax and maximum platelet inhibition occurs between 1 and 3 h after treatment. Ticagrelor’s plasma half-life is relatively short, which mandates twice-daily dosing (Table 1).30 In contrast, clopidogrel and prasugrel are administered once daily. Moreover, while the rapid and reversible platelet inhibition produced by ticagrelor is an advantage in patients requiring CABG or other procedures, the same pharmacodynamic profile potentially puts patients, especially following the implantation of drug-eluting stents, at risk of thrombotic events if they do not adhere to treatment adequately.31


The PLATO study offers the most compelling evidence of ticagrelor’s efficacy and safety. During PLATO, 9333 patients hospitalized for ACS received ticagrelor and placebo (180 mg loading dose followed by 90 mg twice a day), whereas 9291 subjects received clopidogrel and placebo (300–600 mg loading dose followed by 75 mg daily). All patients took aspirin and treatment began within 24 h of symptom onset. Around a quarter of the patients (24.9% ticagrelor, 25.1% clopidogrel) had DM. In the ticagrelor arm, 42.9% of those enrolled presented with NSTEMI, 37.5% with STEMI and 16.6% with unstable angina. The proportions in the clopidogrel arm were 42.5, 38.0 and 16.8%, respectively.32

Two differences between the design of PLATO and TRITON-TIMI 38 are worth noting. First, the proportions of patients with NSTE-ACS (59.5 and 59.3% in the ticagrelor and clopidogrel arms, respectively) in PLATO are considerably <74% who presented with unstable angina or NSTEMI in TRITON-TIMI 38. Second, in TRITON-TIMI, 38 patients received prasugrel after the coronary angiogram and only if the clinician planned PCI. During PLATO, patients received ticagrelor before the angiogram. These differences potentially hinder direct comparison of the two studies.

After 12 months, the primary composite endpoint (death from vascular causes, MI or stroke) had occurred in 9.8 and 11.7% of patients receiving ticagrelor and clopidogrel, respectively (HR 0.84; P < 0.001). Ticagrelor-reduced rates of other composite outcomes, MI alone (5.8 and 6.9%, respectively; P = 0.005), death from vascular causes (4.0 and 5.1%, respectively; P = 0.001) and all-cause mortality (4.5 and 5.9%, respectively; P < 0.001) but not stroke alone (1.5 and 1.3%, respectively; P = 0.22).32

Ticagrelor reduced the rate of the primary composite endpoint compared with clopidogrel in patients presenting with NSTEMI (11.4 and 13.9%, respectively; HR 0.83, 95% CI 0.73–0.94) and STEMI (8.5 and 10.1%, respectively; HR 0.84, 95% CI 0.72–0.98).32 A subgroup analysis of PLATO included 7544 patients with STE or left bundle-branch block. After 12 months, ticagrelor did not significantly reduce the frequency of the primary endpoint in this subgroup compared with clopidogrel (10.8 and 9.4%, respectively; HR 0.87, 95% CI 0.75–1.01; P = 0.07).33 The reason for the discordance between the main study results and the subgroup analysis is not clear. However, no differences in antiplatelet efficacy between the two drugs emerged in patients with unstable angina (8.6 and 9.1%, respectively; HR 0.96, 95% CI 0.75–1.22; Pinteraction = 0.41).33

In patients with unstable angina or NSTEMI, ticagrelor and clopidogrel were equally effective at reducing rates of the primary composite endpoint in low (TIMI scores 0–2; 4.2 and 4.1%, respectively; HR 1.11, 95% CI 0.53–2.31) and high-risk patients (TIMI scores 5–7; 14.4 and 15.6%, respectively; HR 0.92, 95% CI 0.79–1.07). However, ticagrelor was more effective than clopidogrel in medium-risk patients (TIMI scores 3–4; 8.2 and 10.9%, respectively; HR 0.77, 95% CI 0.64–0.92), although the test for interaction was nonsignificant (Pinteraction = 0.27).32

Ticagrelor was associated with a higher rate of major bleeding unrelated to CABG than clopidogrel according to both the study criteria (4.5 and 3.8%, respectively; P = 0.03) and the TIMI criteria (2.8 and 2.2%, respectively; P = 0.03). The PLATO definition encompasses a wider range of bleeding outcomes than the TIMI major definition, which accounts for the differences between the two criteria. No significant difference in major bleeding rates emerged between ticagrelor and clopidogrel (11.6 and 11.2%, respectively; P = 0.43) according to the study’s criteria.32

In a subgroup analysis of the PLATO study, patients with pre-existing DM (n = 4662) were more likely to show ischemic and hemorrhagic endpoints than subjects without DM (n = 13 951). On multivariable analyses, DM was associated with significantly higher incidences of the primary composite outcome (HR 1.66, 95% CI 1.51–1.82; P < 0.0001), all-cause mortality (HR 1.84, 95% CI 1.61–2.10; P < 0.0001), MI (HR 1.53, 95% CI 1.35–1.73; P < 0.0001) and major bleeding (HR 1.41, 95% CI 1.28–1.55; P < 0.0001).34

In DM patients, ticagrelor-reduced rates of the primary composite endpoint (14.1%) compared with clopidogrel (16.2%, HR 0.88, 95% CI 0.76–1.03; P = 0.07), although the DM subgroup was inadequately powered to detect differences in the primary endpoint. Ticagrelor was also associated with nonstatistically significant reductions in all-cause mortality (7.0 and 8.7%, respectively; HR 0.82, 95% CI 0.66–1.01), MI (8.4 and 9.1%, respectively; HR 0.92, 95% CI 0.75–1.13) and definite stent thrombosis (1.6 and 2.4%, respectively; HR 0.65, 95% CI 0.36–1.17) compared with clopidogrel. Ticagrelor did not increase rates of major bleeding (14.1%) compared with clopidogrel (14.8%) in patients with DM (HR 0.95, 95% CI 0.81–1.12).34

As ticagrelor is not a prodrug, cytochrome polymorphisms do not influence outcomes. In PLATO, ticagrelor reduced the incidence of the primary outcome compared with clopidogrel in patients with the CYP2C19 loss-of-function allele (8.6 and 11.2%, respectively; HR 0.77, 95% CI 0.60–0.99; P = 0.0380) as well as in subjects wild-type for this polymorphism (8.8 and 10.0%, respectively; HR 0.86, 95% CI 0.74–1.01; P = 0.0608; Pinteraction = 0.46). The event rate with clopidogrel after 30 days was higher in patients with loss-of-function than wild-type CYP2C19 alleles (5.7 and 3.8%, respectively; P = 0.028). Ticagrelor also reduced the frequency of the primary outcome compared with clopidogrel irrespective of ABCB1 genotype (Pinteraction = 0.39).35

Nevertheless, ticagrelor is a substrate for, and a mild inhibitor of, CYP3A4. Therefore, the concomitant use of strong CYP3A4 inhibitors (such as ketoconazole, clarithromycin, nefazodone, ritonavir and atazanavir) with ticagrelor is contraindicated. Co-administration with potent CYP3A inducers may decrease ticagrelor’s efficacy. Ticagrelor is also a substrate for, and a weak inhibitor of, P-glycoprotein. Therefore, ticagrelor may increase the exposure of P-glycoprotein substrates, mandating appropriate monitoring with concurrent drugs with narrow therapeutic indices such as digoxin or cyclosporine. Finally, because ticagrelor can induce asymptomatic ventricular pauses and bradycardia, physicians should exercise caution when administering concomitant drugs that induce bradycardia.36

Despite the seeming lack of clinically influential polymorphisms, clopidogrel reduced the rate of the primary endpoint in the PLATO study by ≥25% ticagrelor among patients enrolled in North America (9.6 and 11.6%, respectively; HR 1.25, 95% CI 0.93–1.67), the converse of the overall result.32 However, the sponsor’s suggestion that the discordance between the results in North America and the rest of the world potentially reflected an adverse effect associated with high-dose maintenance aspirin was ‘ultimately unconvincing’.37 Further studies now need to ascertain the cause of the unexpected result in North America.

Apart from unresolved issues regarding differences in efficacy, prescribers also need to be cognisant of ticagrelor’s propensity to elevate uric acid and creatinine concentrations, increase ventricular pauses and cause dyspnoea.38 For example, in PLATO, dyspnoea was more common with ticagrelor (13.8%) than clopidogrel (7.8%).32 The ONSET/OFFSET study assessed cardiac and pulmonary function in 123 patients with stable coronary artery disease treated for 6 weeks with ticagrelor (180 mg loading dose, then 90 mg twice daily), clopidogrel (600 mg loading dose, then 75 mg daily) or placebo. Patients taking ticagrelor were more likely to report dyspnoea (38.6%) than those in the clopidogrel (9.3%) and placebo groups (8.3%, P < 0.001). Most cases of dyspnoea were mild and/or lasted less than 24 h. Three patients discontinued ticagrelor because of dyspnoea, which tended to emerge early during treatment. In those who developed dyspnoea, symptoms emerged within 24 h in 36% and within a week in 77%. No significant changes between baseline and 6 weeks were apparent in any cardiac or pulmonary function parameters in any treatment group or among the patients who developed dyspnoea during ticagrelor treatment.39 The dyspnoea associated with ticagrelor seems to arise from an adenosine-related mechanism, perhaps triggered by afferent nerve stimulation. However, the dyspnoea does not appear to produce an objective deterioration in pulmonary function. Nevertheless, clinicians should consider the risk of dyspnoea when deciding on the composition of DAPT for patients with symptomatic pulmonary disease.40

Reversal of ticagrelor therapy

As previously noted, ticagrelor is a reversibly binding P2Y12 inhibitor, whereas both clopidogrel and prasugrel have irreversible binding effects.41 Data from one study suggest that 3–5 days are needed between discontinuation of ticagrelor therapy and surgery, less time than for clopidogrel or prasugrel therapy.42

From studies to the clinic

European guidelines for NSTE-ACS published in August 2011 recommend that all patients without contraindications should receive aspirin (loading dose 150–300 mg; maintenance dose 75–100 mg daily). The guidelines advocate adding P2Y12 inhibitors to aspirin as soon as possible and maintaining treatment for 12 months, unless contraindicated (e.g. patients at excessive risk of bleeding).4

The guidelines recommend ticagrelor (180 mg loading dose, 90 mg twice daily) for all patients (including clopidogrel pre-treated) at moderate-to-high risk of ischemic events (e.g. those showing elevated troponins), regardless of the initial treatment strategy. Clinicians should discontinue clopidogrel when treatment with ticagrelor begins. The guidelines advocate prasugrel (60 mg loading dose, 10 mg daily dose) for P2Y12-inhibitor-naïve patients, especially those with DM, in whom coronary anatomy is known and who are proceeding to PCI unless the patient shows a high risk of life-threatening bleeds or other contraindications.4

The guidelines suggest clopidogrel (300 mg loading dose, 75 mg daily dose) only when ticagrelor or prasugrel are inappropriate. A 600 mg loading dose of clopidogrel (or 300 mg at PCI following a 300 mg loading dose) is recommended for patients scheduled for an invasive strategy if ticagrelor and prasugrel are unsuitable. Clinicians could consider a 150 mg maintenance dose of clopidogrel for the first 7 days in patients managed with PCI who are not at increased bleeding risk. The guidelines suggest that neither increasing the maintenance dose of clopidogrel based on platelet function nor considering genotyping and/or platelet function testing are routine but may be appropriate in selected cases.4

In patients pre-treated with P2Y12 inhibitors, CABG and other nonemergent major surgery should be postponed for at least 5 days after cessation of ticagrelor or clopidogrel, and 7 days following prasugrel if clinically feasible and unless the patient is at high risk of ischemic events. Ticagrelor or clopidogrel can be re-started after CABG as soon as considered safe.4

American guidelines also acknowledge that thienopyridines are an important component of the management of patients with NSTE-ACS.6,43 [The guidelines did not include ticagrelor, which the Food and Drug Administration (FDA) had not approved at the time the guidelines were written.] However, the guidelines note that the evidence supporting prasugrel in NSTEMI and unstable angina come solely from TRITON-TIMI 38. Therefore, the guidelines counsel that the use of prasugrel in the clinic ‘should carefully follow how it was tested in that study’—such as commencing prasugrel only after the decision to proceed to PCI. The guidelines do not recommend using prasugrel routinely before angiography but recognize that this strategy may be appropriate, in line with the license, before catheterization in selected patients for whom a decision to proceed to angiography and PCI has been made.6

The American guidelines do not explicitly endorse prasugrel over clopidogrel or vice versa. While the composite efficacy endpoint favored prasugrel, the guidelines note that the comparison is based on a single large trial. Moreover, the clopidogrel loading dose used in TRITON-TIMI 38 was lower than is currently recommended in the American guidelines. Finally, clinicians cannot prospectively identify patients likely to show an impaired response to clopidogrel. Against this background, the guidelines suggest that considering anti-thrombotic efficacy, the risk of bleeding adverse events and the clinician’s experience using a medication currently offer the best guide for decisions about the choice of thienopyridine for individual patients.6

Areas for further study

This review highlights several areas for further research. First, the lack of head-to-head trials limits comparative evaluations of ticagrelor and prasugrel in ACS generally and NSTEMI-ACS, in particular. There is a need for direct comparisons between ticagrelor and prasugrel in NSTE-ACS patients that employ a consistent protocol with respect to the relationship between the angiogram and dosing. Second, further studies need to evaluate the role of pharmacogenomic testing to identify prospectively patients likely to show a poor response to clopidogrel. The studies should also compare ticagrelor and prasugrel against clopidogrel in pharmacogenomically defined subgroups—such as stratified by the likelihood of responsiveness based on CYP2C19 loss of function polymorphisms.

Third, NSTE-ACS patients tend to be older and have more co-morbidities, especially DM and renal failure, than those with STEMI.4 DM, for example, is associated with poorer outcomes (illustrated by the PLATO subgroup analysis discussed previously), possibly reflecting more rapid progression following presentation with the first ACS event. For example, a pooled analysis of 11 TIMI trials performed between 1997 and 2006 found that 22.4% of patients presenting with unstable angina or NSTEMI had DM, compared with 15.4% among STEMI patients (P < 0.001). Moreover, mortality after 30 days among DM patients who presented with unstable angina or NSTEMI was significantly higher than among nondiabetic subjects at 30 days (2.1 and 1.1%, respectively; P < 0.001) and 1 year (7.2 and 3.1%, respectively, P < 0.001). After adjusting for multiple confounders, DM at presentation with unstable angina and NSTEMI was independently associated with higher mortality after 30 days (OR 1.78, 95% CI 1.24–2.56) and 1 year (HR 1.65, 95% CI 1.30–2.10).44

However, in CURRENT-OASIS 7, rates of the primary outcome in DM patients who underwent PCI were not significantly different with high- (4.9%) and standard-dose clopidogrel (5.6%, HR 0.89, 95% CI 0.68–1.18; P = 0.434),10 which underscores the need for new antiplatelet treatments with proven efficacy in DM patients. Therefore, further studies in NSTE-ACS patients that stratify efficacy and tolerability according to DM and other co-morbidities would help inform the choice between antiplatelet medications in the clinic in these high-risk groups.

Finally, TRILOGY-ACS (NCT00699998) is the first study to investigate prasugrel in medically managed patients (i.e. not thienopyridine-naïve) with unstable angina or NSTEMI. This on-going phase III, double-blind, double-dummy, parallel-group, randomized, controlled trial compares aspirin and prasugrel with aspirin and clopidogrel in medically managed patients enrolled within 10 days of presenting with unstable angina or NSTEMI. TRILOGY-ACS plans to enrol approximately 10 300 patients at 800 sites globally. Of these, an estimated 7800 patients will be <75 years of age. TRILOGY-ACS will also include the largest platelet function substudy to integrate genomics, biomarkers of inflammation and hemodynamic stress, systems biology and health outcomes with measures of platelet function. Initial results are expected in 2013.

The ACCOAST study

Finally, no prospective randomized trial has compared thienopyridine pre-treatment with the same thienopyridine administered at the time of PCI in a high-risk NSTEMI population. Therefore, ACCOAST—a phase 3, multicenter, parallel-group, double-blind study—will compare two prasugrel loading dose schedules in patients with NSTEMI who are scheduled for coronary angiography or PCI. ACCOAST is currently recruiting and will enrol around 4100 patients at more than 100 sites in over 10 countries. Patients will randomly receive either45:

  • a loading dose of 30 mg of prasugrel after diagnosis followed by coronary angiography with an additional dose of 30 mg of prasugrel at the time of PCI. This constitutes the pre-treatment group and

  • the control group will receive a 60 mg loading dose of prasugrel at the time of PCI.

Figure 2 summarizes the study design. Patients in the pre-treatment arm will receive the initial prasugrel loading dose as soon as possible after randomization. The second blinded loading dose will be administered after completion of coronary angiography and when investigators decide to proceed with PCI. All patients will receive open-label maintenance prasugrel beginning 18–24 h after the PCI.45

Figure 2.

ACCOAST study design.45

Eligibility criteria include a diagnosis of NSTEMI (a history of chest discomfort or ischemic symptoms of at least 10 min duration at rest within 48 h before entry into the study and without persistent ST-segment elevation) and raised cardiac troponin levels (≥1.5 times the upper limit of normal). Patients should be scheduled to undergo coronary angiography or PCI within 2–24 h of randomization. If necessary, the procedure may be scheduled for the following day. However, all coronary angiographies or PCIs need to be performed within 48 h after randomization.45

The primary hypothesis is that prasugrel pre-treatment is superior to the 60 mg loading dose. The primary endpoint is the composite of cardiovascular death, MI, stroke, urgent revascularization or glycoprotein IIb/IIIa inhibitor bailout during the 7 days after randomization. The 7-day endpoint specifically evaluates the early effects associated with pre-treatment. The study will continue until 400 patients have experienced the primary efficacy endpoint event, which the statistical design estimates to occur when approximately 4100 patients (2050 in each arm) have received treatment. The design provides 80% power to establish the superiority of prasugrel pre-treatment over the loading dose given at time of PCI using the primary composite endpoint. Key safety endpoints include TIMI major and minor bleeding risks.45


Initial medical therapy for patients with suspected ACS should reduce ischemia and prevent further myocardial damage.46 However, many NSTE-ACS patients—even when at high risk of experiencing major adverse cardiovascular events—still do not receive revascularization and the residual mortality remains considerable, despite DAPT including clopidogrel. This underscores the need to optimize treatment further.

Improving antithrombotic therapy should form an important element in the multifaceted approach needed to optimize outcomes in NSTE-ACS. Although clopidogrel represented an important therapeutic advance in the management of NSTE-ACS, variations in platelet response and a relatively slow onset of action compromises outcomes in this setting compared with the newer antiplatelet drugs ticagrelor and prasugrel. Indeed, a growing body of evidence shows that in NSTE-ACS patients, ticagrelor and prasugrel are more effective than clopidogrel and are relatively well tolerated. Treatment guidelines should be updated to reflect the growing evidence base showing that ticagrelor and prasugrel are effective in patients presenting with NSTEMI and unstable angina, which is, after all, the most common ACS presentation.

In the meantime, differences between ticagrelor and prasugrel are beginning to emerge that should allow clinicians to better tailor treatment to the patient. For example, ticagrelor’s relatively short plasma half-life mandates twice daily dosing47; clopidogrel and prasugrel are administered once daily. Head-to-head comparisons are now needed to directly compare the relative risks and benefits of ticagrelor and prasugrel in NSTE-ACS, especially as differences in trial design means that PLATO and TRITON-TIMI 38 are not directly comparable. Finally, more data in NSTEMI patients are required. Notably, the ACCOAST and TRILOGY-ACS studies will provide important evidence with regard to the benefits and risks of prasugrel pre-treatment.

In the meantime, consensus management guidelines suggest that considering antithrombotic efficacy, the risk of bleeding adverse events and experience with a medication currently offer the best guide for decisions about the choice of thienopyridine for individual patients.6 The issues discussed in this review build on the foundations laid by the guidelines and should enhance clinicians’ ability to optimize and individualize NSTE-ACS treatment, thereby further reduce the morbidity and mortality associated with this common cardiovascular condition.


Daiichi Sankyo Europe and Eli Lilly and Company.

Conflict of interest: Pr G.M. discloses the following relationships: Research Grants (to the Institution) from Abbott Vascular, AstraZeneca, BMS, Boston Scientific, Cordis, Eli Lilly, Fédération Française de Cardiologie, Fondation de France, Guerbet Medical, INSERM, ITC Edison, Medtronic, Pfizer, Sanofi-Aventis, Servier, Société Française de Cardiologie, Stago; consulting or lecture fees from AstraZeneca, Bayer, Boehringer-Ingelheim, Cardiovascular Research Foundation, Cleveland Clinic Research Foundation, Daiichi-Sankyo, Duke Institute, Eli Lilly, Europa, Lead-Up, GSK, Institut de Cardiologie de Montreal, Menarini, Nanospheres, Novartis, Pfizer, Portola, Sanofi-Aventis, The Medicines Company, TIMI study group. Pr J.P.C. reports receiving research grants from Bristol-Myers Squibb, Sanofi-Aventis, Eli Lilly, Guerbet Medical, Medtronic, Boston Scientific, Cordis, Stago, Fondation de France, INSERM, Fédération Française de Cardiologie, and Société Française de Cardiologie; consulting fees from Sanofi-Aventis, Eli Lilly and Bristol-Myers Squibb; and lecture fees from Bristol-Myers Squibb, Sanofi-Aventis, Eli Lilly and AstraZeneca. Dr J.S. reports receiving research grants from Sanofi-Aventis, Daiichi-Sankyo, Eli Lilly, Brahms, INSERM, Fédération Française de Cardiologie and Société Française de Cardiologie; consulting fees from Daiichi-Sankyo, Eli Lilly and speaker honoraria from AstraZeneca, Daiichi Sankyo, Eli Lilly and Servier. Dr S.A.O’C. has received research grants from Menarini and the European Society of Cardiology. Dr G.C. reports receiving a research grant from la Fédération Française de Cardiologie; consultant fees from Abbott Vascular, AstraZeneca, CLS Behring, Daiichi Sankyo, Eli Lilly; Iroko Cardio lecture fees from Abbott Vascular, AstraZeneca, Biotronik, CLS Behring, Daiichi Sankyo, Eli Lilly and Iroko Cardio.


The authors would like to acknowledge the assistance of Mark Greener, medical writer and Springer Healthcare Ltd in the preparation of this manuscript. The authors are responsible for the final version of the review.


View Abstract