Fewer MIs With Prasugrel Than Ticagrelor in STEMI Patients: ISAR-REACT 5

Given the overall trial results and the MI benefit seen in STEMI patients, experts say they’re shifting to greater use of prasugrel.

Fewer MIs With Prasugrel Than Ticagrelor in STEMI Patients: ISAR-REACT 5

For STEMI patients undergoing primary PCI, there are no significant differences in the risks of death, MI, or stroke between those treated with prasugrel and those who received ticagrelor, nor are there any significant differences in bleeding, according to the results of a new subgroup analysis from ISAR-REACT 5.

However, ticagrelor use in STEMI patients undergoing revascularization was associated with a significantly increased risk of MI compared with prasugrel (HR 1.95; 95% CI 1.18-3.23).

Investigators led by Alp Aytekin, MD (Deutsches Herzzentrum München and Technische Universität München, Germany), say that while the overall efficacy of prasugrel and ticagrelor was not significantly different, “the advantage observed with prasugrel regarding the risk of recurrent myocardial infarction might be considered in the treatment of patients with STEMI undergoing primary PCI who present with a particularly higher risk for thrombotic complications.”

To TCTMD, senior investigator Adnan Kastrati, MD (Deutsches Herzzentrum München and Technische Universität München), said in an email that “both the primary result [from ISAR-REACT 5] and the significant difference in recurrent MI [in the STEMI subgroup] are sufficiently strong findings to guide our treatment decisions.” As such, “in patients with ACS planned for an invasive treatment strategy and without an indication for oral anticoagulation, prasugrel is our standard P2Y12 inhibitor drug,” he stated.

The new analysis, published October 29, 2020, in Circulation, included 1,653 patients with STEMI randomized as part of the 4,018-patient study of ACS patients. The trial, when it was first presented at the European Society of Cardiology Congress in 2019, caught a lot of people off guard since investigators had hypothesized that ticagrelor would be more effective than prasugrel for preventing the primary endpoint of death, MI, or stroke. Instead, clinical efficacy significantly favored prasugrel at 1 year, a benefit that did not come at a cost of increased bleeding.

Despite the results, some cardiologists had some issues with ISAR-REACT 5, particularly the open-label design and differences in loading strategies based on patient presentation (ie, STEMI vs unstable angina/NSTEMI) and drug-specific, guideline-recommended care. In an editorial accompanying the diabetic subgroup analysis published last month, one expert even suggests that ISAR-REACT 5 findings were statistical false positives, a suggestion the researchers strongly disputed.

Similar Hazard as Overall Trial

In the STEMI subgroup, the primary endpoint occurred in 10.1% of patients assigned to ticagrelor and 7.9% of patients randomized to prasugrel (HR 1.31; 95% CI 0.95-1.82). There was no significant difference in the risk of all-cause mortality—4.9% with ticagrelor versus 4.7% with prasugrel (P = 0.83)—and no difference in the risk of stroke.

The incidence of MI was 5.3% in the ticagrelor arm and 2.8% in the prasugrel group, a statistically significant difference (P = 0.01). The investigators point out that the difference in MI was mostly driven by an increase in spontaneous (2.9% vs 1.5%; P = 0.04) and PCI-related events with ticagrelor (2.3% vs 1.2%; P = 0.09). With respect to definite or probable stent thrombosis, rates were 2.1% and 1.3% in the ticagrelor- and prasugrel-treatment arms, respectively (HR 1.55; 95% CI 0.72-3.30).

The secondary safety endpoint of BARC type 3 to 5 bleeding did not differ between ticagrelor and prasugrel (6.1% vs 5.1%; P = 0.36).

Regarding the overall results, Kastrati pointed out that the hazard ratio in the STEMI subset indicating a higher risk with ticagrelor was similar to the hazard observed in the overall trial (1.36; 95% CI 1.09-1.70). “Statistical power is certainly the major contributor to the lack of formal significance for the STEMI subset,” said Kastrati. In two older trials—TRITON-TIMI 38 and PLATO—prasugrel and ticagrelor were superior to the less-potent clopidogrel for the prevention of ischemic events in ACS patients, and there was no significant interaction between ACS type and treatment effect, he noted.

Robert Byrne, MBBCh, PhD (Mater Private Hospital, Dublin, Ireland), who wasn’t involved in the study but has worked closely with the ISAR-REACT 5 investigators in the past, said the head-to-head comparison of prasugrel and ticagrelor is an “essential” clinical trial independent from industry and more studies of this kind are needed. He also pointed out that industry-independent trials tend to have more surprises than those sponsored by drug and device companies.

In terms of the results, Byrne, like Kastrati, said the results in the STEMI subgroup line up with the overall trial.

“I think the treatment effects seems to be the same in STEMI as in the non-STEMI patients,” he said. While there was no significant difference in the primary endpoint, “I don’t read too much into that because the trial was not powered for these subgroups, be it STEMI or non-STEMI.” With respect to the higher incidence of MI in the STEMI patients treated with ticagrelor, that also lines up with overall findings, he said.

Prasugrel Preferred

For Byrne, prasugrel should be the preferred agent for patients undergoing planned revascularization for ACS, including STEMI, but he noted that clinical practice tends to lag behind clinical trial results.

“If we look at patients coming through the hospital, there’s still a lot who receive ticagrelor for ACS and that’s because physicians have historically been more comfortable with ticagrelor,” said Byrne. “Maybe they’re more comfortable with the contraindications or [feel] that there are fewer contraindications in comparison with prasugrel. The loading and maintenance dose is more straightforward for the physician, although there are certainly advantages for the patient with prasugrel and its once-daily administration. That is not something to underestimate. We know in the real world, patients are not as compliant as we want them to be or as we think them to be.” 

Dharam Kumbhani, MD (UT Southwestern Medical Center, Dallas, TX), another cardiologist who wasn’t involved in ISAR-REACT 5, said that while prasugrel and ticagrelor were both shown to be superior to clopidogrel in ACS, the reduction in all-cause mortality with ticagrelor in PLATO swayed a lot of operators towards that agent. That benefit was possibly a chance finding, said Kumbhani, noting that it wasn’t documented in other ticagrelor studies.

“ISAR-REACT 5 is interesting, and although it has several flaws, it does directly compare the two drugs head-to-head,” Kumbhani said in email. “It has influenced our/my practice, and we have moved towards more prasugrel use than ticagrelor.” The advantages of prasugrel are that it’s generic, a once-daily dose, and there are no concerns with dyspnea, a well-known side-effect of ticagrelor, he added. “I still prefer ticagrelor for low-body-weight patients or those with prior stroke.”

As for the MI finding, Kumbhani said he considers it hypothesis-generating at this stage. While the MI differences appear to be driven by spontaneous and PCI-related events, given the small numbers, it’s possible this is a chance finding. Kumbhani pointed out that there were only 67 MIs in total and 29 patients were lost to follow-up, which is a limitation of the present analysis. While there are some potential mechanisms suggesting greater ischemic protection with prasugrel, this would need to be proven in a future study, he said.

“Discontinuation rates for ticagrelor were higher than for prasugrel so it is possible that medication noncompliance is driving of this,” said Kumbhani, referring to the higher MI rate with ticagrelor.

Michael O’Riordan is the Managing Editor for TCTMD. He completed his undergraduate degrees at Queen’s University in Kingston, ON, and…

Read Full Bio
  • Kastrati, Byrne, and Kumbhani report no relevant conflicts of interest.