Apixaban Plus Mono vs Dual Antiplatelet Therapy in ACS
Apixaban Plus Mono vs Dual Antiplatelet Therapy in ACS
Using data from the APPRAISE-2 trial, we found that approximately 20% of patients changed antiplatelet therapy regimen at least once after randomization. There was no effect on efficacy and increased bleeding risk for apixaban versus placebo among patients on aspirin alone or aspirin plus clopidogrel. We obtained similar results with MSM analyses to account for antiplatelet therapy switching during follow-up. These data suggest that use of novel oral anticoagulants with a strategy of concomitant aspirin alone compared with aspirin plus clopidogrel might not improve the safety of these agents. Further investigation with prospective studies evaluating other alterations in antithrombotic regimens, such as monotherapy with P2Y12 receptor inhibitors, should be pursued.
There is great interest in improving the safety of novel oral anticoagulant use in ACS, but there are limited data exploring the issue. A recent meta-analysis of 7 placebo-controlled randomized trials found that the addition of a novel oral anticoagulant to aspirin reduced the incidence of cardiovascular events (HR: 0.70; 95% CI: 0.59 to 0.84) but increased bleeding (HR: 1.79; 95% CI: 1.54 to 2.09). Adding a novel oral anticoagulant to aspirin plus clopidogrel resulted in a modest decrease in cardiovascular events (HR: 0.87; 95% CI: 0.80 to 0.95) but also more than doubled the risk of bleeding (HR: 2.34; 95% CI: 2.06 to 2.66). Although intriguing, the study was limited by the inclusion of anticoagulants such as ximelagatran that are no longer commercially available, as well as phase II clinical trials (including drug doses not evaluated in phase III studies). The meta-analysis was also not performed using patient-level data. The phase III clinical trials included in this analysis were ATLAS ACS-2 TIMI-51 and APPRAISE-2. On the basis of the results of ATLAS ACS-2 TIMI-51, rivaroxaban was approved for use in ACS patients in Europe, but this trial included a small percentage of patients on single antiplatelet therapy, precluding examination of concomitant aspirin-only use in the study.
We therefore examined the use of apixaban for ACS on the background of aspirin alone versus aspirin plus clopidogrel. In contrast to rivaroxaban, in APPRAISE-2, apixaban caused excess bleeding without ischemic benefit, which led to early trial termination. It should be noted that APPRAISE-2 tested the same dose of apixaban used for stroke prevention in nonvalvular atrial fibrillation and included higher risk patients than those patients in ATLAS ACS-2 TIMI-51, which studied one-quarter or one-half of the drug dose used for atrial fibrillation. These factors, as well as the shorter trial duration of APPRAISE-2 or chance, could potentially account for the differences in the trial results. Like rivaroxaban, apixaban is a factor Xa inhibitor. There appears to be a class effect on bleeding with the same 3- to 4-fold relative increase in bleeding relative to the baseline observed in phase II and phase III trials of these oral anticoagulants in this population. Thus, despite the early termination of APPRAISE-2, this analysis might provide hypothesis-generating insight into optimizing the use of novel oral anticoagulants such as rivaroxaban, especially because an analogous study from ATLAS ACS-2 TIMI-51 cannot be performed.
In our study, randomization of patients occurred at a median of 6 days after the index ACS event. In accordance with ACS treatment guidelines, almost all patients (97%) were on aspirin at baseline. However, 19% of patients were on a single antiplatelet agent at randomization, despite guidelines recommendations for dual antiplatelet therapy for ACS. This might be partially due to regional variation in practice patterns, because the majority of patients on baseline monotherapy were enrolled in Eastern Europe. Although we did not observe lower rates of ischemic events in patients treated with dual antiplatelet therapy versus monotherapy, as was demonstrated in randomized clinical trials, this was likely due to the nonrandomized use of antiplatelet medications and residual confounding in our study. These findings of underuse and regional variation in use of dual antiplatelet therapy in ACS deserve further exploration.
We also found that approximately 1 in 5 patients switched antiplatelet therapies after randomization. Because of these changes, we used statistical methodology to account for actual, not just baseline, use of aspirin versus aspirin plus clopidogrel in our analyses. We adjusted for not only baseline features, but we also weighted time-dependent covariates associated with antiplatelet therapy use and discontinuation. Using 2 statistical approaches and in subgroup analyses, we consistently found that apixaban versus placebo increased bleeding when added to either aspirin alone or aspirin plus clopidogrel. These findings suggested that this strategy might not improve the safety of novel oral anticoagulants for post-ACS treatment, although prospective studies are needed to confirm this.
Efforts to improve the safety of oral anticoagulation should continue to assess different combinations of antithrombotic agents. One potential approach might be to use concomitant single antiplatelet therapy. A previous study found that monotherapy with clopidogrel plus warfarin caused less bleeding and had no increase in thrombotic complications compared with clopidogrel plus aspirin in patients who underwent PCI. However, less than one-third of patients in this trial presented with ACS, and neither newer P2Y12 receptor antagonists, such as prasugrel or ticagrelor, nor novel oral anticoagulants, were studied. Additional insight into this area will come from the COMPASS (Rivaroxaban for the Prevention of Major Cardiovascular Events in Coronary or Peripheral Artery Disease; NCT01776424) trial, which will test rivaroxaban with versus without aspirin in patients with remote ACS. These results will have to be reconciled with those from the recent DAPT (Dual Antiplatelet Therapy) trial, which suggested benefit of prolonged dual antiplatelet therapy after PCI with a drug-eluting stent, and the PEGASUS-TIMI 54 (Prevention of Cardiovascular Events in Patients with Prior Heart Attack Using Ticagrelor Compared to Placebo on a Background of Aspirin-Thrombolysis In Myocardial Infarction 54) trial, which showed benefit of ticagrelor for patients with previous remote MI, although patients on long-term anticoagulation were excluded from these trials. In addition, because of the increasing number of patients with multiple indications for anticoagulant and antiplatelet therapies, important data to help define optimal regimens for secondary prevention of ACS might also come from trials such as PIONEER AF-PCI (A Study Exploring Two Strategies of Rivaroxaban and One of Oral Vitamin K Antagonist in Patients With Atrial Fibrillation Who Undergo Percutaneous Coronary Intervention; NCT01830543) and REDUAL PCI (Evaluation of Dual Therapy With Dabigatran vs. Triple Therapy With Warfarin in Patients With AF That Undergo a PCI With Stenting; NCT02164864), which are studying the use of novel oral anticoagulants in patients with atrial fibrillation who are undergoing PCI.
The use of apixaban versus placebo was not randomized within antiplatelet therapy groups, nor was the use of aspirin alone or aspirin plus clopidogrel randomized. Despite adjustment for baseline and post-randomization variables, residual confounding likely exists. The use of concomitant antiplatelet therapy was not blinded, and there might have been differential post-baseline use of these therapies among patients on apixaban versus placebo. We did not have data regarding aspirin dosing, nor did we have a sufficient sample size to analyze patients on clopidogrel monotherapy. This analysis had limited power to detect modest, but still potentially clinically important differences, in the overall population and particularly in subgroups and for interactions between subgroups. Furthermore, this study was a subgroup analysis from a randomized clinical trial, and these results might not be applicable to the general population. Finally, APPRAISE-2 was terminated early, which would lead to an overestimation not only of the event rate leading to termination (bleeding) but also to the lack of efficacy seen with apixaban versus placebo.
Discussion
Using data from the APPRAISE-2 trial, we found that approximately 20% of patients changed antiplatelet therapy regimen at least once after randomization. There was no effect on efficacy and increased bleeding risk for apixaban versus placebo among patients on aspirin alone or aspirin plus clopidogrel. We obtained similar results with MSM analyses to account for antiplatelet therapy switching during follow-up. These data suggest that use of novel oral anticoagulants with a strategy of concomitant aspirin alone compared with aspirin plus clopidogrel might not improve the safety of these agents. Further investigation with prospective studies evaluating other alterations in antithrombotic regimens, such as monotherapy with P2Y12 receptor inhibitors, should be pursued.
There is great interest in improving the safety of novel oral anticoagulant use in ACS, but there are limited data exploring the issue. A recent meta-analysis of 7 placebo-controlled randomized trials found that the addition of a novel oral anticoagulant to aspirin reduced the incidence of cardiovascular events (HR: 0.70; 95% CI: 0.59 to 0.84) but increased bleeding (HR: 1.79; 95% CI: 1.54 to 2.09). Adding a novel oral anticoagulant to aspirin plus clopidogrel resulted in a modest decrease in cardiovascular events (HR: 0.87; 95% CI: 0.80 to 0.95) but also more than doubled the risk of bleeding (HR: 2.34; 95% CI: 2.06 to 2.66). Although intriguing, the study was limited by the inclusion of anticoagulants such as ximelagatran that are no longer commercially available, as well as phase II clinical trials (including drug doses not evaluated in phase III studies). The meta-analysis was also not performed using patient-level data. The phase III clinical trials included in this analysis were ATLAS ACS-2 TIMI-51 and APPRAISE-2. On the basis of the results of ATLAS ACS-2 TIMI-51, rivaroxaban was approved for use in ACS patients in Europe, but this trial included a small percentage of patients on single antiplatelet therapy, precluding examination of concomitant aspirin-only use in the study.
We therefore examined the use of apixaban for ACS on the background of aspirin alone versus aspirin plus clopidogrel. In contrast to rivaroxaban, in APPRAISE-2, apixaban caused excess bleeding without ischemic benefit, which led to early trial termination. It should be noted that APPRAISE-2 tested the same dose of apixaban used for stroke prevention in nonvalvular atrial fibrillation and included higher risk patients than those patients in ATLAS ACS-2 TIMI-51, which studied one-quarter or one-half of the drug dose used for atrial fibrillation. These factors, as well as the shorter trial duration of APPRAISE-2 or chance, could potentially account for the differences in the trial results. Like rivaroxaban, apixaban is a factor Xa inhibitor. There appears to be a class effect on bleeding with the same 3- to 4-fold relative increase in bleeding relative to the baseline observed in phase II and phase III trials of these oral anticoagulants in this population. Thus, despite the early termination of APPRAISE-2, this analysis might provide hypothesis-generating insight into optimizing the use of novel oral anticoagulants such as rivaroxaban, especially because an analogous study from ATLAS ACS-2 TIMI-51 cannot be performed.
In our study, randomization of patients occurred at a median of 6 days after the index ACS event. In accordance with ACS treatment guidelines, almost all patients (97%) were on aspirin at baseline. However, 19% of patients were on a single antiplatelet agent at randomization, despite guidelines recommendations for dual antiplatelet therapy for ACS. This might be partially due to regional variation in practice patterns, because the majority of patients on baseline monotherapy were enrolled in Eastern Europe. Although we did not observe lower rates of ischemic events in patients treated with dual antiplatelet therapy versus monotherapy, as was demonstrated in randomized clinical trials, this was likely due to the nonrandomized use of antiplatelet medications and residual confounding in our study. These findings of underuse and regional variation in use of dual antiplatelet therapy in ACS deserve further exploration.
We also found that approximately 1 in 5 patients switched antiplatelet therapies after randomization. Because of these changes, we used statistical methodology to account for actual, not just baseline, use of aspirin versus aspirin plus clopidogrel in our analyses. We adjusted for not only baseline features, but we also weighted time-dependent covariates associated with antiplatelet therapy use and discontinuation. Using 2 statistical approaches and in subgroup analyses, we consistently found that apixaban versus placebo increased bleeding when added to either aspirin alone or aspirin plus clopidogrel. These findings suggested that this strategy might not improve the safety of novel oral anticoagulants for post-ACS treatment, although prospective studies are needed to confirm this.
Efforts to improve the safety of oral anticoagulation should continue to assess different combinations of antithrombotic agents. One potential approach might be to use concomitant single antiplatelet therapy. A previous study found that monotherapy with clopidogrel plus warfarin caused less bleeding and had no increase in thrombotic complications compared with clopidogrel plus aspirin in patients who underwent PCI. However, less than one-third of patients in this trial presented with ACS, and neither newer P2Y12 receptor antagonists, such as prasugrel or ticagrelor, nor novel oral anticoagulants, were studied. Additional insight into this area will come from the COMPASS (Rivaroxaban for the Prevention of Major Cardiovascular Events in Coronary or Peripheral Artery Disease; NCT01776424) trial, which will test rivaroxaban with versus without aspirin in patients with remote ACS. These results will have to be reconciled with those from the recent DAPT (Dual Antiplatelet Therapy) trial, which suggested benefit of prolonged dual antiplatelet therapy after PCI with a drug-eluting stent, and the PEGASUS-TIMI 54 (Prevention of Cardiovascular Events in Patients with Prior Heart Attack Using Ticagrelor Compared to Placebo on a Background of Aspirin-Thrombolysis In Myocardial Infarction 54) trial, which showed benefit of ticagrelor for patients with previous remote MI, although patients on long-term anticoagulation were excluded from these trials. In addition, because of the increasing number of patients with multiple indications for anticoagulant and antiplatelet therapies, important data to help define optimal regimens for secondary prevention of ACS might also come from trials such as PIONEER AF-PCI (A Study Exploring Two Strategies of Rivaroxaban and One of Oral Vitamin K Antagonist in Patients With Atrial Fibrillation Who Undergo Percutaneous Coronary Intervention; NCT01830543) and REDUAL PCI (Evaluation of Dual Therapy With Dabigatran vs. Triple Therapy With Warfarin in Patients With AF That Undergo a PCI With Stenting; NCT02164864), which are studying the use of novel oral anticoagulants in patients with atrial fibrillation who are undergoing PCI.
Study Limitations
The use of apixaban versus placebo was not randomized within antiplatelet therapy groups, nor was the use of aspirin alone or aspirin plus clopidogrel randomized. Despite adjustment for baseline and post-randomization variables, residual confounding likely exists. The use of concomitant antiplatelet therapy was not blinded, and there might have been differential post-baseline use of these therapies among patients on apixaban versus placebo. We did not have data regarding aspirin dosing, nor did we have a sufficient sample size to analyze patients on clopidogrel monotherapy. This analysis had limited power to detect modest, but still potentially clinically important differences, in the overall population and particularly in subgroups and for interactions between subgroups. Furthermore, this study was a subgroup analysis from a randomized clinical trial, and these results might not be applicable to the general population. Finally, APPRAISE-2 was terminated early, which would lead to an overestimation not only of the event rate leading to termination (bleeding) but also to the lack of efficacy seen with apixaban versus placebo.
Source...