The Long Journey of Individualizing Antiplatelet Therapy After Acute Coronary Syndromes

Robert F. Storey


Eur Heart J. 2020;41(37):3546-3548. 

Acute coronary syndromes (ACS) are most commonly, but not exclusively, precipitated by atherosclerotic plaque rupture or erosion, leading to a platelet-driven thrombotic response with occlusion at the site of the culprit plaque and/or distal embolism into the microvasculature. The discovery that acetylsalicylic acid irreversibly inactivates platelet cyclo-oxygenase-1, a key component of the enzymatic pathway converting arachidonic acid into the potent platelet activator thromboxane A2, was the foundation for clinical trials that established aspirin as the standard treatment for myocardial infarction (MI).[1] Ongoing atherothrombotic events despite aspirin therapy led to the development of increasingly potent antagonists of the platelet P2Y12 receptor (P2Y12), which amplifies platelet activation induced by multiple platelet agonists through the action of ADP that is released from the dense granules of activated platelets.[2] Ticagrelor is a reversibly-binding P2Y12 inhibitor that offers highly consistent and potent P2Y12 inhibition during maintenance therapy, reflecting the lack of requirement for metabolism to achieve its inhibitory effect and continuous presence in the plasma during twice-daily dosing.[3,4] The combination of aspirin and ticagrelor, both reliable and potent inhibitors of distinct pathways of platelet activation, has proven amazingly successful at preventing stent thrombosis, particularly following improvements in this outcome that have arisen from recognition of delayed absorption of P2Y12 inhibitors in opiate-treated patients.[5] However, this has come at the expense of more spontaneous bleeding, largely reflecting the compromise to haemostasis from potent dual antiplatelet therapy (DAPT).

Contemporaneously with the development of potent DAPT strategies in the management of ACS have been improvements in stent design. Early drug-eluting stents (DESs) had relatively thick struts with drug regimens that promoted incomplete strut coverage as well as polymer coatings that seemed to trigger late neoatherosclerotic complications, leading to late and very late stent thrombosis events that undermined the advantage over bare-metal stents of reduced symptomatic restenosis rates. This phenomenon drove some of the benefit of prolonged DAPT regimens seen in trials such as the DAPT study, where continuing DAPT with clopidogrel or prasugrel reduced very late thrombotic events.[6] However, times have moved on with thin-strut DESs that have safer elution systems and minimize the risk of late/very late thrombotic events, helped also by the recognition that optimal stent deployment with adequate stent expansion is critical in reducing thrombotic risk. This revolution in the safety of percutaneous coronary intervention (PCI) has also occurred at the same time as improvements in the management of modifiable risk factors for recurrent atherothrombotic events, including the standard use of high-dose potent statins and improved blood pressure targeting, as well as lifestyle interventions that include supported smoking cessation.

So now the question arises: does an ACS patient who has had their culprit lesion optimally treated with PCI, and whose ACS event was predominantly driven by modifiable risk factors, need prolonged potent DAPT or just a sufficient duration to cover the period of stent endothelialization, which seems to be complete within 3 months in the overwhelming majority of contemporary cases? The TWILIGHT-ACS substudy data presented in this issue of the European Heart Journal[7] go some way to providing an answer to this question: discontinuing aspirin at 3 months after PCI in ticagrelor-treated ACS patients, as well as leading to a predictable reduction in bleeding risk, did not seem to increase the risk of thrombotic events despite the loss of platelet inhibition attributable to aspirin. On the face of it, this seems to contradict the evidence of the PEGASUS-TIMI 54 study, which showed that long-term DAPT with aspirin and ticagrelor in patients with prior MI reduced major adverse cardiovascular events (MACE) compared with aspirin alone, including signals of cardiovascular mortality reduction in patients who had the highest ischaemic risk.[8–11] The argument that aspirin adds little to antiplatelet efficacy in the presence of potent P2Y12 inhibition is not consistent with the substantially additive platelet inhibition achieved with aspirin (Figure 1) or with the adverse effects on haemostasis consequent to this. An alternative explanation could be that ticagrelor monotherapy is markedly superior to aspirin monotherapy, but this is not well supported by evidence from the GLOBAL LEADERS and SOCRATES studies, which showed similar outcomes during the phases of aspirin vs. ticagrelor monotherapy.[12,13] Despite this, it is possible that potent P2Y12 inhibition is rather more effective than aspirin monotherapy whilst there is a persisting, albeit small, risk of late stent thrombosis since P2Y12 plays a more central role in platelet activation than thromboxane A2 (Figure 1), and pre-clinical thrombosis models support this hypothesis of greater antithrombotic efficacy.

Figure 1.

Algorithm illustrating the potential dichotomization of ticagrelor-treated patients at 3 months post-ACS according to presence or absence of risk factors for ischaemic and bleeding events.

A major issue with translating the TWILIGHT study results into clinical practice is the uncertainty relating to its insufficient power to assess the efficacy of ticagrelor monotherapy vs. ticagrelor-based DAPT in preventing MACE, further exaggerated when assessing the smaller ACS subgroup. With only 4614 randomized ACS patients in the TWILIGHT-ACS study—and even fewer MI patients, who benefit most from potent DAPT—compared with 21 162 patients in the PEGASUS-TIMI 54 study, there is more certainty about the efficacy of long-term ticagrelor-based DAPT post-MI compared with the efficacy of long-term ticagrelor monotherapy in an MI population. The hazard ratio for MACE in the ACS patients in TWILIGHT-ACS was 0.96 for placebo vs. aspirin, with an upper confidence interval of 1.28, meaning that the level of risk reduction seen with ticagrelor-based DAPT in PEGASUS-TIMI 54 (15–16% relative risk reduction) could not be confidently excluded, particularly in those with more risk factors in whom the hazard ratios and upper confidence intervals were higher.[7] This argues for a cautious and pragmatic approach pending the accumulation of more evidence and new guideline recommendations. Some of the greatest absolute risk reductions in the PEGASUS-TIMI 54 population were seen in the subgroups with multivessel coronary artery disease,[11] chronic kidney disease (CKD),[10] or peripheral arterial disease,[9] whereas the hazard ratios numerically favoured DAPT vs. ticagrelor monotherapy in multivessel disease and CKD patients in the overall TWILIGHT population.[14] This paves the way for a dichotomization of ticagrelor-treated MI patients, with long-term DAPT for those with these unmodifiable risk factors, as recommended by current ESC guidelines,[15] vs. 3 months DAPT followed by ticagrelor monotherapy for those without them. Furthermore, those with ongoing bleeding or a high bleeding risk may also be reasonably managed with ticagrelor monotherapy from 3 months post-ACS, given the clear benefit of improved haemostasis with this strategy and lack of evidence supporting long-term DAPT in these patients (Figure 1). The difficulty for clinicians in adopting such a strategy is that many patients cannot be comfortably placed in one or other of these categories, and complications of plaque progression cannot be reliably predicted on the basis of clinical characteristics and serum creatinine alone. Plaque imaging, additional biomarkers, and safe and effective anti-inflammatory therapies will therefore probably be required in the future in order to increase confidence in abandoning DAPT at 3 months post-ACS in a larger proportion of the ACS population.

What about monotherapy with oral P2Y12 inhibitors other than ticagrelor? The data from the TWILIGHT-ACS substudy cannot be extrapolated to either clopidogrel or prasugrel as these have distinct pharmacological properties compared with ticagrelor.[2] Clopidogrel produces widely variable levels of P2Y12 inhibition between individuals, making it an unreliable option as monotherapy, and prasugrel, though a more reliable prodrug than clopidogrel, has licensed maintenance regimens that are not intended to produce the consistently high levels of platelet P2Y12 inhibition seen with ticagrelor. Consequently, further work is required before prasugrel monotherapy can be embraced as a post-ACS strategy.

Switch from ticagrelor monotherapy to aspirin monotherapy at 12–15 months or continue ticagrelor monotherapy long-term? That part of the journey remains unsignposted.