Prasugrel vs. ticagrelor after acute coronary syndrome: a critical appraisal of the ISAR-REACT 5 trial

In the era of invasive coronary angiography and stent implantation in patients presenting with acute coronary syndrome, antiplatelet treatment strategies occupy a central position in the pharmacotherapeutic approach to prevent recurrent myocardial ischaemia.1,2 In addition to standard acetylsalicylic acid therapy, the faster acting and more potent antiplatelet agents ticagrelor and prasugrel were recently investigated in direct comparison with clopidogrel in terms of efficacy and safety. Within the Trial to Assess Improvement in Therapeutic Outcomes by Optimizing Platelet Inhibition with Prasugrel–Thrombolysis in Myocardial Infarction (TRITON-TIMI 38), prasugrel was found to be superior to clopidogrel in patients presenting with acute coronary syndrome via reducing the primary composite cardiovascular endpoint defined by death from cardiovascular causes, non-fatal myocardial infarction. and non-fatal stroke.3 Notably this effect was mainly driven by a reduction of recurrent myocardial ischaemic events. Similarly, in the Platelet Inhibition and Patient Outcomes (PLATO) trial, ticagrelor proved to be superior to clopidogrel via reducing the frequency of the composite cardiovascular outcome defined as death from cardiovascular causes, any myocardial infarction, or stroke.4 While patients in the prasugrel arm showed an elevated risk of major bleeding events during TRITON-TIMI 38, PLATO showed no increase in the safety endpoint for any major bleeding event for ticagrelor, but an increased rate of non-procedure-related bleeding events. However, valid data on the relative merits of treatment for prasugrel as compared with ticagrelor in individuals that underwent coronary angiography are scarce. To fill this gap of knowledge, Schüpke et al. compared the efficacy and safety of prasugrel and ticagrelor, both in addition to standard acetylsalicylic acid, in patients with acute coronary syndrome within the Intracoronary Stenting and Antithrombotic Regimen: Rapid Early Action for Coronary Treatment (ISAR-REACT 5) trial.5 The authors of this multicentre randomized controlled trial enrolled 4018 patients with acute coronary syndrome that underwent coronary angiography, covering a representative fraction of individuals presenting with ST-segment elevation myocardial infarction (STEMI; 41.1%). The data of ISAR-REACT 5 highlighted that prasugrel-based dual antithrombotic treatment was superior to the ticagrelor-based strategy in reducing the incidence of the composite endpoint of death, myocardial infarction, or ischaemic stroke (prasugrel, 6.9% vs. ticagrelor, 9.3%; P = 0.006; number needed to treat = 42) which was mainly attributed to a strong reduction in the incidence of recurrent myocardial ischaemia. Notably, the observed superiority of prasugrel needs to be emphasized when considering the lower than expected rate of observed endpoints, a relevant number of individuals receiving conservative treatment (14.2%), and the delay of prasugrel loading in individuals presenting with non-ST-segment elevation myocardial infarction (NSTEMI). Moreover, the authors observed no difference in the safety endpoint for any major bleeding event for both treatment arms. The detected treatment benefits were evident for almost all relevant subgroups and on-treatment analyses. The lack of a double-dummy design may have introduced a reporting bias. However, since the provided results are conflicting with the primary hypothesis of the study, i.e. that ticagrelor holds a more potent antithrombotic potential, this bias appears to be unlikely. The fact that endpoint events were predominantly ascertained via telephone follow-up contacts and not personal patient visits confers an increased risk for accompanying detection bias. Further, it needs to be considered that ISAR-REACT 5 excluded patients that presented with a history of cerebrovascular events. This subgroup mirrors a subset of participants that showed a net clinical harm for prasugrel within the TRITON-TIMI 38 trial, but not for ticagrelor in PLATO, which might alter the overall picture of the results in ISAR-REACT 5 via introducing a selection bias in terms of efficacy and safety in the prasugrel arm. It is noteworthy that the authors reported that 19% of patients did not receive any trial drug at the time of discharge. Unfortunately, there is no clear explanation provided for this particularly high rate of study drop-outs during the index stay, but the majority of early drop-outs might be deductible via patients receiving a loading of trial drug without confirmation of acute coronary syndrome during coronary angiography, or a contraindication for prasugrel and ticagrelor due to the need for triple antithrombotic therapy. Of note, patients in the ticagrelor arm discontinued the trial drug more frequently and sooner after discharge, as compared with individuals receiving prasugrel (prasugrel, 12.5% vs. ticagrelor, 15.2%; P = 0.03). The most likely reasons for non-adherence to ticagrelor were the typical side effect of dyspnoea as well as its twice-daily administration. Consequently, the presented data give the overall impression of a more favourable side effect profile of prasugrel in this head to head comparison with ticagrelor. Of utmost importance, the fraction of participants that were excluded for investigation of the primary safety endpoint in the prasugrel treatment arm was 10 times higher as compared with patients receiving ticagrelor, which is mainly attributed to the fact that patients that underwent coronary angiography during NSTEMI received a loading dose for prasugrel not before coronary angiography. As a consequence, within the safety analysis—only assessing patients who received at least one dosage of the study drug—early drop-outs were analysed in the ticagrelor arm but not in the prasugrel arm. This observed discrepancy subsequently acts as a potential confounder of detected incidence rates for bleeding events in the prasugrel group, but does not explain the favourable efficacy of prasugrel. A subgroup analysis for STEMI and NSTEMI patients with regard to bleeding may elucidate this discrepancy in the safety analysis. Within the study protocol, the reduced maintenance dose of prasugrel was used appropriately in patients ≥75 years and a body weight of ≤60 kg. Patients meeting the reduction criteria also covered a representative fraction of the prasugrel treatment arm with approximately one-third of participants. It is noteworthy that the subgroup analysis for patients ≥75 years and body weight ≤60 kg did not show a significant effect on the primary efficacy endpoint; consequently, dose adjustment of prasugrel does not mirror an explanation for the observed benefit. Despite the consistent efficacy of the prasugrel treatment strategy within tested subgroups, the strata of participants with type 2 diabetes mellitus deserve further attention. With 22% of all enrolled individuals within ISAR-REACT 5, type 2 diabetes mellitus constitutes a relevant subgroup, as already observed in TRITON-TIMI 38 and PLATO with 23% and 25%, respectively. The hazard ratio for patients presenting without type 2 diabetes mellitus highlighted the most prominent efficacy benefit for prasugrel; while there seems to be a comparable benefit of both P2Y12 inhibitors in participants presenting with type 2 diabetes mellitus, unfortunately the authors did not provide a P-value for interaction in this regard. Based on the potential benefit of ticagrelor in this subgroup, it needs to be considered that the general efficacy of ticagrelor after acute coronary syndrome mainly relies on its beneficial effect in patients presenting with type 2 diabetes mellitus, that has already been addressed in previous investigations.6 Additionally, there was no benefit for prasugrel observed within the subgroup analysis for patients that were treated conservatively after acute coronary syndrome. In conclusion, ISAR-REACT 5 clearly shows a beneficial effect of prasugrel over ticagrelor in terms of efficacy and comparable safety characteristics in the general acute coronary syndrome population. Therefore, a dual antiplatelet therapy of prasugrel in addition to standard acetylsalicylic acid seems the treatment approach of choice in patients presenting with acute coronary syndrome undergoing percutaneous coronary intervention (PCI). However, the clinical implementation needs to be undertaken with caution in selected patients—especially considering individuals presenting with type 2 diabetes mellitus, those who received conservative therapy, or patients with prior cerebrovascular events. Similarly, the effects of prasugrel compared with ticagrelor from a long-term perspective beyond 1 year of follow-up deserve further attention as do subgroup analyses for bleeding. Conflict of interest: P.