Background: Dual antiplatelet therapy (DAPT) for 12 months is the standard of care after coronary stenting in patients with acute coronary syndrome (ACS). The aim of this individual patient-level meta-analysis was to summarise the evidence comparing DAPT de-escalation to ticagrelor monotherapy versus continuing DAPT for 12 months after coronary drug-eluting stent implantation.

Methods: A systematic review and individual patient data (IPD)-level meta-analysis of randomised trials with centrally adjudicated endpoints was performed to evaluate the comparative efficacy and safety of ticagrelor monotherapy (90 mg twice a day) after short-term DAPT (from 2 weeks to 3 months) versus 12-month DAPT in patients undergoing percutaneous coronary intervention with a coronary drug-eluting stent. Randomised trials comparing P2Y12 inhibitor monotherapy with DAPT after coronary revascularisation were searched in Ovid MEDLINE, Embase, and two websites (www.tctmd.com and www.escardio.org) from database inception up to May 20, 2024. Trials that included patients with an indication for long-term oral anticoagulants were excluded. The risk of bias was assessed using the revised Cochrane risk-of-bias tool. The principal investigators of the eligible trials provided IPD by means of an anonymised electronic dataset. The three ranked coprimary endpoints were major adverse cardiovascular or cerebrovascular events (MACCE; a composite of all-cause death, myocardial infarction, or stroke) tested for non-inferiority in the per-protocol population; and Bleeding Academic Research Consortium (BARC) 3 or 5 bleeding and all-cause death tested for superiority in the intention-to-treat population. All outcomes are reported as Kaplan-Meier estimates. The non-inferiority was tested using a one-sided α of 0·025 with the prespecified non-inferiority margin of 1·15 (hazard ratio [HR] scale), followed by the ranked superiority testing at a two-sided α of 0·05. This study is registered with PROSPERO (CRD42024506083).

Findings: A total of 8361 unique citations were screened, of which 610 records were considered potentially eligible during the screening of titles and abstracts. Of these, six trials that randomly assigned patients to ticagrelor monotherapy or DAPT were identified. De-escalation took place a median of 78 days (IQR 31-92) after intervention, with a median duration of treatment of 334 days (329-365). Among 23 256 patients in the per-protocol population, MACCE occurred in 297 (Kaplan-Meier estimate 2·8%) with ticagrelor monotherapy and 332 (Kaplan-Meier estimate 3·2%) with DAPT (HR 0·91 [95% CI 0·78-1·07]; p=0·0039 for non-inferiority; τ2<0·0001). Among 24 407 patients in the intention-to-treat population, the risks of BARC 3 or 5 bleeding (Kaplan-Meier estimate 0·9% vs 2·1%; HR 0·43 [95% CI 0·34-0·54]; p<0·0001 for superiority; τ2=0·079) and all-cause death (Kaplan-Meier estimate 0·9% vs 1·2%; 0·76 [0·59-0·98]; p=0·034 for superiority; τ2<0·0001) were lower with ticagrelor monotherapy. Trial sequential analysis showed strong evidence of non-inferiority for MACCE and superiority for bleeding among the overall and ACS populations (the z-curve crossed the monitoring boundaries or the required information size without crossing the futility boundaries or approaching the null). The treatment effects were heterogeneous by sex for MACCE (p interaction=0·041) and all-cause death (p interaction=0·050), indicating a possible benefit in women with ticagrelor monotherapy, and by clinical presentation for bleeding (p interaction=0·022), indicating a benefit in ACS with ticagrelor monotherapy.

Interpretation: Our study found robust evidence that, compared with 12 months of DAPT, de-escalation to ticagrelor monotherapy does not increase ischaemic risk and reduces the risk of major bleeding, especially in patients with ACS. Ticagrelor monotherapy might also be associated with a mortality benefit, particularly among women, which warrants further investigation.