The New England Journal of Medicine October 17, 2019

Ticagrelor or Prasugrel in Patients with Acute Coronary Syndromes

Stefanie Schüpke, Franz-Josef Neumann, Maurizio Menichelli, et al. (ISAR-REACT 5 Trial Investigators)

Bottom Line

In patients with acute coronary syndromes planned for an invasive strategy, prasugrel was superior to ticagrelor in reducing the composite rate of death, myocardial infarction, or stroke at 1 year without significantly increasing the risk of major bleeding.

Key Findings

1. The primary endpoint (composite of death, myocardial infarction, or stroke at 1 year) occurred in 9.3% (184 of 2012) of patients in the ticagrelor group compared to 6.9% (137 of 2006) in the prasugrel group (Hazard Ratio for ticagrelor vs. prasugrel, 1.36; 95% CI, 1.09 to 1.70; P = 0.006).

2. The reduction in the primary composite endpoint was driven primarily by a significant reduction in myocardial infarction, which occurred in 4.8% of the ticagrelor group vs. 3.0% of the prasugrel group.

3. Rates of all-cause death (4.5% vs. 3.7%) and stroke (1.1% vs. 1.0%) were numerically higher in the ticagrelor group compared to the prasugrel group, though these individual components were not independently statistically significant.

4. Major bleeding (Bleeding Academic Research Consortium [BARC] type 3, 4, or 5), the primary safety endpoint, was not significantly different between the two groups, occurring in 5.4% of the ticagrelor group and 4.8% of the prasugrel group (Hazard Ratio 1.12; 95% CI, 0.83 to 1.51; P = 0.46).

Study Design

Design

Randomized Controlled Trial

Open-Label

Sample

4,018

Patients

Duration

1 year

Median

Setting

Multicenter, Europe

Population Adult patients presenting with acute coronary syndromes (STEMI, NSTEMI, or unstable angina) for whom an invasive diagnostic evaluation was planned.

Intervention Ticagrelor (180 mg loading dose administered as early as possible after randomization, followed by a maintenance dose of 90 mg twice daily).

Comparator Prasugrel (60 mg loading dose administered as early as possible for STEMI, but delayed until coronary anatomy was known for NSTE-ACS, followed by a maintenance dose of 10 mg daily; reduced to 5 mg daily for patients ≥75 years of age or weighing <60 kg).

Outcome Composite of all-cause death, myocardial infarction, or stroke at 1 year after randomization.

Study Limitations

• The open-label design could have introduced reporting or management bias, although all primary and secondary clinical endpoints were evaluated by a blinded adjudication committee.

• There was an asymmetrical discontinuation of the assigned study drug, with significantly higher and earlier discontinuation in the ticagrelor group (15.2%) compared to the prasugrel group (12.5%).

• The study results specifically apply to patients intended for an invasive treatment strategy; the comparative efficacy in patients strictly managed medically without planned angiography remains unaddressed by this trial.

• Medication adherence could have played a role in real-world efficacy, as ticagrelor requires a twice-daily maintenance dosing regimen compared to the once-daily regimen for prasugrel.

Clinical Significance

The ISAR-REACT 5 trial generated a major paradigm shift in the management of Acute Coronary Syndromes (ACS). Prior to this head-to-head comparison, both ticagrelor and prasugrel were granted equal Class I recommendations over clopidogrel in international guidelines for ACS patients undergoing percutaneous coronary intervention. By demonstrating unambiguous superiority of prasugrel in reducing ischemic events without increasing major bleeding, this trial positioned prasugrel as the preferred first-line P2Y12 inhibitor for ACS patients planned for invasive evaluation.

Historical Context

The modern era of dual antiplatelet therapy for ACS was shaped by the landmark TRITON-TIMI 38 (2007) and PLATO (2009) trials, which respectively demonstrated the superiority of prasugrel and ticagrelor over the older standard, clopidogrel. However, because neither trial compared the newer potent agents directly, clinicians debated for a decade over which drug was optimal. Indirect cross-trial comparisons often favored ticagrelor due to a mortality benefit seen in PLATO and an elevated bleeding risk seen with prasugrel in TRITON. ISAR-REACT 5 was designed as an independent, investigator-initiated, pragmatic trial to settle this debate. The results profoundly surprised the cardiovascular community, reversing the prevalent assumption of ticagrelor's clinical superiority and heavily influencing subsequent clinical practice guidelines.

Guided Discussion

High-yield insights from every perspective

Med Student

Medical Student

Both ticagrelor and prasugrel are potent P2Y12 inhibitors used in acute coronary syndromes, but they belong to different chemical classes. How do their mechanisms of action and pharmacokinetic properties differ, and why are these differences relevant in the acute setting?

Key Response

Prasugrel is a thienopyridine prodrug that requires hepatic activation and binds irreversibly to the P2Y12 receptor. Ticagrelor is a cyclopentyltriazolopyrimidine, which is a direct-acting agent (no activation needed) that binds reversibly. Understanding these foundational differences is crucial for board exams, as they dictate the onset of action, offset time (which is critical if the patient needs urgent CABG), and drug-drug interaction profiles.

Resident

In the ISAR-REACT 5 trial, the timing of the loading dose for patients with NSTEMI differed between the two treatment arms. How did this timing differ, and how should this influence your clinical workflow when admitting a patient with NSTEMI who is planned for an early invasive strategy?

Key Response

In the trial, ticagrelor was given as a routine upstream loading dose at presentation, whereas prasugrel was only administered after coronary angiography once the anatomy was known. For residents managing CCU admissions, this highlights a shift in workflow: if prasugrel is the chosen agent based on this trial's outcomes, upstream P2Y12 loading in the emergency department for NSTEMI should be avoided, shifting the loading dose to the cath lab.

Fellow

The ISAR-REACT 5 trial demonstrated a significantly higher rate of recurrent myocardial infarction in the ticagrelor arm. Considering the known side effect profile of ticagrelor and the trial's open-label design, what real-world factors might explain this divergence in efficacy?

Key Response

Ticagrelor requires twice-daily dosing and is known to cause dyspnea in a notable percentage of patients, whereas prasugrel is dosed once daily. In an open-label trial primarily utilizing telephone follow-ups, higher non-adherence or early discontinuation in the ticagrelor arm due to pill burden or subjective side effects could disproportionately drive the higher MI rates. Fellows must consider how real-world adherence impacts comparative effectiveness.

Attending

While ISAR-REACT 5 demonstrated the overall superiority of prasugrel over ticagrelor in patients with ACS planned for invasive management, what specific patient populations or clinical scenarios dictate that ticagrelor must still be chosen over prasugrel?

Key Response

Attendings must balance trial data with strict contraindications. Prasugrel is absolutely contraindicated in patients with a history of prior stroke or TIA due to an unacceptable risk of intracranial hemorrhage, and carries a black box warning for patients over 75 years of age or weighing less than 60 kg. Furthermore, ticagrelor remains indicated for medically managed ACS patients, as prasugrel's benefit is strictly validated in those undergoing PCI.

Scholarly Review

Critical appraisal through the lens of expert reviewers and guideline development

PhD

ISAR-REACT 5 utilized a PROBE (Prospective Randomized Open, Blinded Endpoint) design rather than a double-blind, placebo-controlled design. How does the PROBE design introduce specific biases in a comparative trial involving a drug known to cause subjective adverse effects like dyspnea, and what statistical or methodological approaches are needed to account for this?

Key Response

Open-label designs are highly susceptible to performance and detection bias. Because ticagrelor causes dyspnea, patients and physicians were aware of the assignments, potentially leading to differential dropout or non-adherence (informative censoring). A PhD critique should evaluate whether the intention-to-treat analysis masks these effects, and whether causal inference models like inverse probability weighting for artificial censoring would provide a more accurate estimate of true biological efficacy.

Journal Editor

As a peer reviewer evaluating the ISAR-REACT 5 manuscript, how would you critique the fact that the trial tested two variables simultaneously in the NSTEMI cohort (the specific drug AND the timing of the loading dose), and how does this confound the trial's primary conclusion?

Key Response

A critical editor would flag that this is a pragmatic strategy trial rather than a pure pharmacological comparison. Because NSTEMI patients received upstream ticagrelor but delayed prasugrel (post-angiography), the superiority of the prasugrel arm could be attributed to the drug itself, the safer delayed-loading strategy (which avoids bleeding in those ultimately needing CABG), or a combination of both. This limits the ability to claim pure pharmacological superiority of prasugrel.

Guideline Committee

Following the publication of ISAR-REACT 5, the ESC updated its ACS guidelines to give prasugrel a Class IIa preference over ticagrelor for NSTEMI patients proceeding to PCI. Does the evidence from this single, open-label, strategy-confounded trial justify changing the previous guidelines where both drugs were given equal Class I recommendations, and how does this align with current ACC/AHA guidelines?

Key Response

The Guideline Committee must weigh whether a single, albeit large, investigator-initiated open-label trial provides sufficient evidence to establish a hierarchy between two potent P2Y12 inhibitors. While the ESC adopted a Class IIa recommendation favoring prasugrel based on this data, ACC/AHA guidelines have historically maintained an equal Class IIa recommendation for either agent over clopidogrel, reflecting hesitation to declare one strictly superior without confirmatory double-blind, perfectly matched data.

Clinical Landscape

Noteworthy Related Trials

2007

TRITON-TIMI 38 Trial

n = 13,608 · NEJM

Tested

Prasugrel 10 mg daily

Population

Patients with ACS scheduled for percutaneous coronary intervention

Comparator

Clopidogrel 75 mg daily

Endpoint

Composite of cardiovascular death, nonfatal MI, or nonfatal stroke

Key result: Prasugrel significantly reduced ischemic events including stent thrombosis compared to clopidogrel, but with an increased risk of major bleeding.

2009

PLATO Trial

n = 18,624 · NEJM

Tested

Ticagrelor 90 mg twice daily

Population

Patients with acute coronary syndromes

Comparator

Clopidogrel 75 mg daily

Endpoint

Composite of vascular death, MI, or stroke

Key result: Ticagrelor significantly reduced the primary composite endpoint compared to clopidogrel without a significant increase in overall major bleeding.

2016

PRAGUE-18 Trial

n = 1,230 · JACC

Tested

Prasugrel

Population

Patients with acute myocardial infarction treated with primary PCI

Comparator

Ticagrelor

Endpoint

Composite of death, reinfarction, target vessel revascularization, stroke, or severe bleeding

Key result: The trial was stopped early for futility, finding no significant efficacy or safety differences between prasugrel and ticagrelor in the acute phase.

Tailored to your role

Want this tailored to you?

Add your specialty or training stage to get role-specific takeaways and more questions.

Personalize this analysis