ISAR-REACT 5: Subanalysis of PCI Patients Confirms Prasugrel Superiority

Among patients with ACS who undergo PCI, a prasugrel-based strategy is superior to one using ticagrelor in terms of all-cause death, MI, and stroke at 12 months, according to a prespecified subanalysis of the ISAR-REACT 5 trial.

Researchers here had narrowed their focus to only patients who actually were treated with PCI after being randomized to prasugrel versus ticagrelor—representing 84.4% of study participants.

While the initial findings of ISAR-REACT 5, as reported by TCTMD in 2019, were a surprise to the investigators, who had hypothesized that ticagrelor would hold an advantage. Prior evidence had shown increased bleeding associated with prasugrel, but this was not shown to be the case in ISAR-REACT 5, and, subsequently, the 2020 European Society of Cardiology guidelines for NSTE ACS were updated to back a prasugrel-based strategy.

This analysis “confirms the results of that trial and highlights in particular the subgroup of patients who are treated with PCI, which were the dominant group of patients,” lead author J. J. Coughlan, MBBCh (Deutsches Herzzentrum München, Munich, Germany), told TCTMD.

“The difficulty with performing these kinds of trials is that often at the time of randomization, you don't know which patients are going to end up having PCI performed. So that is obviously a challenge,” he added, explaining the rationale for the current paper, published online yesterday in JAMA Cardiology.

In an editor’s note, Ajay J. Kirtane, MD (NewYork-Presbyterian/Columbia University Irving Medical Center, New York, NY), writes: “Although surprising, the editors felt that these findings were important as they support the main trial results and once again emphasize the essential importance of conducting head-to-head trials of pharmacologic strategies for patients with ACS.”

Prasugrel Superior to Ticagrelor

For this analysis, Coughlan and colleagues compared the 1,676 patients randomized to ticagrelor and the 1,701 randomized to prasugrel within the PCI subset. The incidence of the primary endpoint—all-cause death, MI, and stroke at 12 months—was higher in the former group, confirming the superiority of prasugrel (9.8% vs 7.1%; HR 1.41; 95% CI 1.11-1.78). This finding was driven by a higher rate of MI in the ticagrelor arm (5.3% vs 3.8%; HR 1.67; 95% CI 1.19-2.34). There were no differences with regard to definite stent thrombosis.

The safety endpoint of BARC type 3 to 5 bleeding was comparable between the ticagrelor and prasugrel arms (5.3% vs 4.9%; HR 1.10; 95% CI 0.81-1.50).

The findings were confirmed in all subgroups, with the exception of diabetics, in whom there was no significant difference in the primary endpoint between the ticagrelor and prasugrel groups (10.7% vs 13.8%; HR 0.77; 95% CI 0.50-1.16). Additionally, a landmark analysis showed no significant difference between the study arms for the primary endpoint within the first 30 days (4.5% vs 3.5%; HR 1.29; 95% CI 0.92-1.82) but a clear benefit with prasugrel from 30 days through 1 year (5.2% vs 3.5%; HR 1.52; 95% CI 1.09-2.11).

Notably, at 12 months, drug discontinuation was higher in the ticagrelor arm than in those assigned to prasugrel (13.0% vs 10.5%; P = 0.02). Patients who stopped taking their drug did so sooner if they were assigned to ticagrelor (median 90 vs 113 days; P = 0.05) and more often due to dyspnea which was reported more often in the ticagrelor arm (2.3% vs < 0.01%; P < 0.001).

Pretreatment Considerations

“Obviously what the ISAR-REACT 5 trial showed was the importance of comparing the two medications head-to-head, and based on that trial the European guidelines recommended that for non-ST elevation ACS prasugrel should be considered in preference to ticagrelor in patients who are proceeding to PCI,” Coughlan said. “This study reiterates that point and highlights that these are the patients that perhaps benefit from this prasugrel-based strategy.”

This strategy, though, is “dependent on patients being able to get their angiogram within an early time frame, and that is one thing that is a limitation of applying this study in some healthcare systems,” he continued, noting that the evidence for pretreatment is “quite limited” and more study in that field is needed.

“People are often uneasy delaying loading a patient with a second antiplatelet agent, but I do think that we do have good evidence from other trials—for example, the ACCOAST trial for prasugrel—indicating that there was actually no benefit to preloading and in fact it actually may result in harm in terms of increased bleeding outcomes,” Coughlan explained. “Even though there are two different agents being used in this study, [it] does highlight that you don't gain an advantage necessarily by pretreating with ticagrelor compared to waiting and using prasugrel. I think that is a very important message to get across to other doctors who might be uneasy waiting. They might decide to go with a ticagrelor-based pretreatment strategy, but our findings show that that probably isn't a logical thing to do based on the data that we have at the moment.”

Limitations of a Postrandomization Analysis

In his editorial, Kirtane notes some caveats to the analysis. “Although informative in the sense that an analysis of the PCI subgroup addresses the treatment effect among patients undergoing a very different treatment strategy compared with medical therapy or surgical revascularization, the fact that this subgroup by definition is created and occurs after randomization renders treatment comparisons more challenging to interpret,” he writes. “The methodological issues inherent to postrandomization subgroup analyses in the ACS space have been explored in the literature and, at the very least, suggest that one ought to be circumspect in the interpretation of these analyses.”

Commenting on the study for TCTMD, Johanne Silvain, MD (Sorbonne University/Hôpital Pitié-Salpêtrière, Paris, France), said although ISAR-REACT 5 did show a difference between prasugrel and ticagrelor, at the end of the day the higher cost of the latter plays a bigger role in his drug choice. “In France, ticagrelor is still three to four times the price of prasugrel,” he explained. “I would say that [this] trumps the slight difference, clinical or biological, to a finding like [that seen in] ISAR- REACT 5 or previous biological studies. That's why these days I suggest that we [use] prasugrel mostly.”

Prior to the initial presentation of ISAR-REACT 5, Silvain explained that it was common practice in France to pretreat patients with ticagrelor and then keep them on that agent through hospital discharge, as was supported by the ACCOAST trial.

“ISAR-REACT 5 changed that a little bit because some people believed the results are accurate and there is a slight superiority of prasugrel over ticagrelor or at least the strategy, and so we looked at our prescriptions and we did realize that we were prescribing mostly ticagrelor,” he said. “At least there was no doubt on the difference at least on the price, . . . so we started to change the practice. We still use ticagrelor of course because the difference is not major, but we try to be more conscious about the price because it is at least a 12-month prescription, so we try to prescribe more prasugrel than ticagrelor.”

The main reasons why he would choose ticagrelor over prasugrel are that it can be orally administered and is sometimes better tolerated by patients, Silvain said.

What remains to be discovered is a fast-acting P2Y12 inhibitor that can be delivered intravenously for emergency use. “Although there is cangrelor, it is a little too expensive and not as strong as ticagrelor and prasugrel,” he concluded.