Unguided de-escalation from ticagrelor to clopidogrel in stabilised patients with acute myocardial infarction undergoing percutaneous coronary intervention (TALOS-AMI): an investigator-initiated, open-label, multicentre, non-inferiority, randomised trial

Abstract

In patients with acute myocardial infarction receiving potent antiplatelet therapy, the bleeding risk remains high during the maintenance phase. We sought data on a uniform unguided de-escalation strategy of dual antiplatelet therapy (DAPT) from ticagrelor to clopidogrel after acute myocardial infarction. In this open-label, assessor-masked, multicentre, non-inferiority, randomised trial (TALOS-AMI), patients at 32 institutes in South Korea with acute myocardial infarction receiving aspirin and ticagrelor without major ischaemic or bleeding events during the first month after index percutaneous coronary intervention (PCI) were randomly assigned in a 1:1 ratio to a de-escalation (clopidogrel plus aspirin) or active control (ticagrelor plus aspirin) group. Unguided de-escalation without a loading dose of clopidogrel was adopted when switching from ticagrelor to clopidogrel. The primary endpoint was a composite of cardiovascular death, myocardial infarction, stroke, or bleeding type 2, 3, or 5 according to Bleeding Academic Research Consortium (BARC) criteria from 1 to 12 months. A non-inferiority test was done to assess the safety and efficacy of de-escalation DAPT compared with standard treatment. The hazard ratio (HR) for de-escalation versus active control group in a stratified Cox proportional hazards model was assessed for non-inferiority by means of an HR margin of 1·34, which equates to an absolute difference of 3·0% in the intention-to-treat population and, if significant, a superiority test was done subsequently. To ensure statistical robustness, additional analyses were also done in the per-protocol population. This trial is registered at ClinicalTrials.gov, NCT02018055. From Feb 26, 2014, to Dec 31, 2018, from 2901 patients screened, 2697 patients were randomly assigned: 1349 patients to de-escalation and 1348 to active control groups. At 12 months, the primary endpoints occurred in 59 (4·6%) in the de-escalation group and 104 (8·2%) patients in the active control group (pnon-inferiority<0·001; HR 0·55 [95% CI 0·40-0·76], psuperiority=0·0001). There was no significant difference in composite of cardiovascular death, myocardial infarction, or stroke between de-escalation (2·1%) and the active control group (3·1%; HR 0·69; 95% CI 0·42-1·14, p=0·15). Composite of BARC 2, 3, or 5 bleeding occurred less frequently in the de-escalation group (3·0% vs 5·6%, HR 0·52; 95% CI 0·35-0·77, p=0·0012). In stabilised patients with acute myocardial infarction after index PCI, a uniform unguided de-escalation strategy significantly reduced the risk of net clinical events up to 12 months, mainly by reducing the bleeding events. ChongKunDang Pharm, Medtronic, Abbott, and Boston Scientific.