Review of the TRITON-TIMI 38 Trial

Prasugrel versus Clopidogrel in Patients with Acute Coronary Syndromes

Note to readers: Since going live with Cardiology Trials Substack in January of 2024 we have been exclusively covering trials that we have categorized as belonging to the major subject heading “Acute Coronary Syndrome” belonging to the subsection “Medicines”.

Our indexing scheme was described in one of our original posts and we encourage our audience to read it if you have not already. This is pertinent because the next several trials being presented may seem to come out of the blue but we assure you there is a method.

---

N Engl J Med 2007;357:2001-15.

Background Up to now we have presented trials involving major foundational medical therapies for acute coronary syndrome which include aspirin, thrombolytic agents and anticoagulation, but not those involving percutaneous coronary intervention (PCI) as they are reserved for another section.

But, by the turn of the 21^st century, PCI had become the dominant up-front strategy for revascularization in many countries around the world. Clinical trials demonstrated it improved outcomes, the main one being re-infarction, compared to thrombolysis in patients with STEMIs, and there was an evolving evidence for it in non-ST-segment elevation acute coronary syndrome (STEACS) as well, where thrombolysis had not demonstrated any significant benefits.

As PCI became dominant, antithrombotic strategies for optimizing outcomes following PCI evolved along with it. These early trials generally involved a mixture of patient phenotypes (acute vs elective PCI) and were relatively small and of limited quality by comparison to many of the seminal trials presented thus far. Instead of presenting each of these smaller studies, we direct readers to a narrative review that nicely describes the evolution of dual-antiplatelet therapy for PCI and other indications.

Briefly: dual-antiplatelet therapy with aspirin and ticlopidine, an antiplatelet agent belonging to the drug class of thienopyridines, which inhibits platelet aggregation induced by ADP, was found superior to aspirin alone or aspirin plus anticoagulation when PCI was performed; however, there were concerns about its safety.

Clopidogrel was developed after ticlopidine; it had a similar mechanism of action but less safety concerns and could be given as a loading dose to produce more rapid effects. Despite limited evidence from clinical trials comparing it head-to-head with ticlopidine it became the dominant thienopyridine agent on the market and still has a prominent role in the management of cardiovascular diseases today.

Following PCI and dual-antiplatelet therapy with aspirin and clopidogrel, patients continue to have an elevated risk of coronary events, in general, and in-stent related coronary events, in particular. Some of this risk has been attributed to limitations of clopidogrel itself. Clopidogrel has modest antiplatelet effects (compared to other thienopyridines) with substantial interpatient variability due to genetic polymorphisms that impact clopidogrel metabolism and antiplatelet efficacy. Clopidogrel also has a delayed onset of action, which is especially relevant regarding its ability to protect against the dreaded adverse event of early in-stent thrombosis.

Prasugrel is a thienopyridine—developed after clopidogrel—that inhibits platelet aggregation more rapidly, consistently and to a greater extent. The Therapeutic Outcomes by Optimizing Platelet Inhibition with Prasugrel–Thrombolysis in Myocardial Infarction (TRITON–TIMI) 38 trial sought to test the hypothesis that prasugrel would reduce major cardiovascular events compared to clopidogrel in patients with acute coronary syndrome undergoing PCI.

Patients Eligible patients had either moderate-to-high risk unstable angina (UA) or NSTEMI or STEMI. UA and NSTEMI were defined by ischemic symptoms lasting 10 minutes or more and occurring within 72 hours before randomization, a TIMI risk score of 3 or more, and either ST-segment deviation of 1 mm or more or elevated levels of a cardiac biomarker of necrosis. STEMI was traditionally defined. Key exclusion criteria included an increased risk of bleeding, anemia, thrombocytopenia, a history of pathologic intracranial findings, or the use of any thienopyridine within 5 days before enrollment.

Baseline characteristics The median age of patients was 61 years with 13% being ≥75 years and 74% were men; over 90% were white. The index event was UA or NSTEMI in 74% and STEMI in 26%.  PCI was performed in 99% of patients and split evenly between those receiving bare metal or drug eluting stent(s).  18% of patients had a prior MI, 23% had diabetes, 64% had hypertension and 38% were tobacco users. Only 11% of patients had CKD defined as a creatinine clearance ≤60 ml/min.

Procedures A loading dose of prasugrel 60 mg or clopidogrel 300 mg was given in a double blind manner anytime between randomization up to 1 hour after leaving the catheterization laboratory. In order to be randomized, the plan for PCI had to be known. This could occur before going to the cath lab for planned PCI, if the anatomy was already known or occur in the cath lab during the case where anatomy was determined and PCI was performed. If PCI was planned, patients were eligible to undergo pretreatment with the study drug for up to 24 hours prior to PCI.

Treating physicians determined the vessels treated, devices used, and adjunctive medication administered to support PCI. After PCI, patients received maintenance doses of either prasugrel 10 mg daily or clopidogrel 75 mg daily. Use of aspirin at a dose of 75 to 162 mg daily was recommended. Study visits were conducted at hospital discharge, 30 days, 90 days, and 3-month intervals thereafter, for a total of 6 to 15 months.

Endpoints The primary efficacy endpoint was a composite of cardiovascular death, nonfatal MI or stroke during the follow up period. A prespecified “landmark” analysis was undertaken to compare the primary endpoint event rate up to 3 days following randomization and from day 3 to the end of the study. The sample size calculation was event-driven and it was determined that 875 primary endpoint events would provide 90% power to detect a relative risk reduction of 20%. A prespecified analysis performed after 650 events revealed a lower than anticipated event rate and the investigators increased the sample size accordingly.

Results A total of 13,608 patients (10,074 with UA or NSTEMI and 3534 with STEMI), from 707 sites in 30 countries were enrolled. There were 6,813 patients assigned to the prasugrel group and 6,795 assigned to clopidogrel. The median duration of therapy was 14.5 months. Prasugrel significantly reduced the primary composite endpoint compared to clopidogrel (9.9% vs 12.1%; HR 0.81; 95% CI 0.73-0.90; P<0.001). This reduction was driven by significant differences in the rates of nonfatal MI between groups (7.3 vs 9.5%; HR 0.76; 95% CI 0.67–0.85; P<0.001) but not cardiovascular death (2.1% vs 2.4%; P=NS) or stroke (1.0% vs 1.0%; P=NS). There was also no difference in all-cause death (3.0% vs 3.2%; P=NS).

In the prespecified landmark analysis, benefit from prasugrel was observed before day 3 (4.7% vs 5.6%; P=0.01) and from 3 days onward (5.6% vs 6.9%; P=0.003).

Treatment response variation in the trial was observed based on age, susceptibility to bleeding and history of stroke or TIA. Some of this data is summarized in the primary Forest plot of major subgroups and also in a table that details patient risk based on combinations of traits.

Subgroups Patients with a history of stroke or TIA accounted for a small minority of patients; however, prasugrel numerically increased the PEP in these patients compared to clopidogrel (19.1% vs 14.4%; P=NS) but was underpowered to achieve statistical significance. There was a statistically significant interaction in this subgroup compared to prasugrel use in all other patients without a stroke or TIA (9.5% vs 12.3%; P for interaction = 0.02).

Patients <65 composed the majority of patients (n=8,322) and the difference in the PEP between prasugrel and clopidogrel was 8.1% vs 10.6% - a relative difference of 24%. Patients 65-74 years of age accounted for 3,477 patients and the relative difference in the PEP was 14% (10.7% vs 12.3%). There were 1,809 patients ≥75 years of age whose risk of a PEP event was nearly twice as high as those <65 but the relative risk reduction was only 6% (17.2% vs 18.3%) with the 95% CI crossing 1.0. Treatment response variation is also suggested for women compared to men. Importantly, there was no signal of treatment response heterogeneity based on ACS index event (UA or NSTEMI vs STEMI).

Patients with any of the following traits (age ≥75, body weight <60 kg or history of stroke or TIA) accounted for 20% of patients in the trial and had an elevated risk of the PEP compared to patients without this combination; however, they did not benefit from prasugrel (patients with combination of traits; 16.1% vs 16.0%; patients without any traits; 8.3% vs 11.0%) and there was a strong interaction between these subgroups (P for interaction = 0.008).

Safety: In regard to safety, prasugrel significantly increased the composite endpoint of non-CABG-related TIMI major bleeding (2.4% vs 1.8%; HR 1.32; 95% CI 1.03–1.68; P=0.03). Individual components of this key safety endpoint that were significantly different between groups included spontaneous bleeding, life-threatening bleeding and fatal bleeding. There were also highly statistically significant differences in the individual endpoints of bleeding requiring transfusion and CABG-related TIMI major bleeding.

Conclusions In patients with ACS undergoing PCI, prasugrel (vs clopidogrel) reduces the rate of cardiovascular events; mainly driven by nonfatal MI. For patients similar to those enrolled in TRITON-TIMI 38, approximately 50 patients would need to be treated with prasugrel instead of clopidogrel to prevent 1 person from having a nonfatal MI; approximately 170 patients would need to be treated to cause 1 person to have a non-CABG-related major bleeding event.

There is a significant benefit of prasugrel that is realized very early in treatment as demonstrated by the landmark analysis in this study. This is likely due to the faster onset of action of prasugrel compared to clopidogrel conferring a reduction in very early in-stent events. It should be noted that continued differences in treatment are observed over the entire course of the study.

Caution is advised in treating patients with a history of stroke or TIA, who are over 75 years of age and below a body weight 60 kg as such patients derived less benefit and would be expected to garner more treatment related harm. Caution is also advised in treating any patients with increased bleeding risks. These patients were generally excluded from the trial but the specifics regarding how this exclusion criteria was applied in the main manuscript of the trial are lacking.