Cardiology · Medical Literature
Antithrombotic strategy according to age in patients with atrial fibrillation and drug-eluting stents: a post hoc analysis of the ADAPT AF-DES randomised trial.
Abstract
Evidence for non-vitamin K antagonist oral anticoagulant (NOAC) monotherapy for older patients with atrial fibrillation (AF) and drug-eluting stents is lacking. We aimed to evaluate the safety and efficacy of NOAC monotherapy in older AF patients with drug-eluting stents. This is a secondary analysis of the ADAPT AF-DES randomised trial comparing NOAC monotherapy with combination therapy with NOAC plus clopidogrel in patients with AF and drug-eluting stents. Patients were stratified by age (≥75 years and <75 years). Apixaban or rivaroxaban was used as the NOAC. The primary endpoint was a net adverse clinical event at 1 year after randomisation, defined as a composite of all-cause death, myocardial infarction, stent thrombosis, stroke, systemic embolism or major or clinically relevant non-major bleeding defined by the International Society on Thrombosis and Haemostasis criteria. Among 960 patients included in the ADAPT AF-DES trial, 376 (39.2%) patients were aged ≥75 years. In patients aged ≥75 years, the incidence of the primary endpoint was lower in the NOAC monotherapy group (9.8% vs. 22.3%; adjusted hazard ratio, 0.37; 95% confidence interval, 0.21-0.66; P < .001) than in the combination therapy group. In contrast, the incidences did not differ between treatment strategies in patients aged <75 years. The interaction between age groups and strategy groups did not reach statistical significance (P for interaction = .095). The reduction in major or clinically relevant non-major bleeding with NOAC monotherapy was consistent across age groups. In contrast, the benefit of NOAC monotherapy in major adverse cardiac and cerebrovascular events was more pronounced in patients aged ≥75 years, with a statistically significant interaction. In this post hoc analysis, NOAC monotherapy was associated with a lower risk of net adverse clinical events, specifically within the older age subgroup. Tailored approach regarding age and individual ischaemic risk may be needed.
Clinical relevance
In patients with atrial fibrillation (AF) who have undergone drug-eluting stent (DES) implantation, does NOAC monotherapy provide superior net clinical benefit compared with NOAC plus clopidogrel combination therapy, and does this effect differ by age (≥75 years vs. <75 years)? __SNAPSHOT__ {"studyDesign":"Post hoc secondary analysis of a multicenter, randomized, open-label, non-inferiority trial (ADAPT AF-DES)","setting":"32 centers in South Korea (single-country East Asian academic and clinical hospital network)","sampleSize":"n=960 total (parent trial); age-stratified subgroups: ≥75 years n=376 (39.2%), <75 years n=584 (60.8%); subgroup-level arm allocation not reported in abstract","population":"Adults with non-valvular AF (CHA₂DS₂-VASc ≥2) and stable coronary artery disease who had undergone DES implantation at least 12 months prior; mean age of parent cohort 71.1 years; 21.4% women; ~65% had ACS at index PCI; mean CHA₂DS₂-VASc score 4.1; ~60% had persistent or permanent AF","intervention":"NOAC monotherapy: apixaban 5 mg twice daily or rivaroxaban 20 mg once daily (dose-adjusted per labeling); clopidogrel discontinued","comparator":"Combination therapy: same NOAC plus clopidogrel 75 mg once daily (rivaroxaban 15 mg once daily used in combination arm)","primaryOutcome":"Net adverse clinical events (NACE) at 1 year: composite of all-cause death, myocardial infarction, stent thrombosis, stroke, systemic embolism, or major or clinically relevant non-major bleeding (ISTH criteria)","followUp":"12 months from randomization","fundingSource":"Cardiovascular Research Center and Samjin Pharmaceutical (mixed non-profit/industry; Korean pharma company)"}
Key findings
This post hoc analysis stratified 960 ADAPT AF-DES participants by age ≥75 years (n=376, 39.2%) and <75 years (n=584, 60.8%). In the ≥75-year cohort, NOAC monotherapy was associated with a substantially lower incidence of the primary NACE composite compared with NOAC plus clopidogrel combination therapy: 9.8% vs. 22.3% (adjusted hazard ratio [aHR] 0.37; 95% CI truncated in abstract — lower bound and p-value not fully reported in the available abstract). The absolute risk difference in the ≥75 group was 12.5 percentage points (calculated), favoring monotherapy. In the parent ADAPT AF-DES trial overall (n=960), NACE occurred in 9.6% of the monotherapy group vs. 17.2% in the combination therapy group (HR 0.54; 95.2% CI 0.37–0.77; absolute difference −7.6 percentage points; P<0.001 for both non-inferiority and superiority). The component driving benefit across the full trial was major or clinically relevant non-major bleeding (5.2% vs. 13.2%; HR 0.38; 95% CI 0.24–0.60), while ischemic outcomes — MI (0.8% vs. 1.3%), stroke (1.0% vs. 0.8%), stent thrombosis (0.4% vs. 0.4%), and all-cause death (2.9% vs. 4.0%) — were numerically similar between arms. Whether formal interaction testing between age group and treatment effect was statistically significant, and the complete NACE results for the <75 subgroup, are not reported in the available abstract. An NNT for the ≥75 subgroup cannot be calculated without the full confidence interval and event-level data per arm. Notably, 11.3% of patients in the monotherapy group and 29% in the combination therapy group received an inappropriate NOAC underdose, raising a clinically important protocol adherence concern.
Practice implications
This post hoc analysis provides the first randomized data specifically addressing NOAC monotherapy in AF patients aged ≥75 years with prior DES implantation — a population with disproportionately high bleeding risk from antiplatelet co-administration and high thromboembolic risk from AF. The observed aHR of 0.37 in older patients suggests a numerically larger relative benefit with monotherapy in this subgroup than seen in the overall trial (HR 0.54), though formal interaction testing results are not reported in the available abstract, preventing definitive conclusions about age as an effect modifier. What the study established: NOAC monotherapy was associated with markedly lower NACE rates in patients ≥75 years in a post hoc analysis, largely consistent with — and potentially amplifying — the overall trial finding. What the study did NOT establish: whether age significantly modifies treatment effect (interaction p-value not reported); complete efficacy and safety component data by age subgroup; generalizability beyond East Asian patients with DES implanted ≥12 months prior; outcomes beyond 12 months; or applicability to patients with recent ACS (<1 year), bioprosthetic valves, or those requiring aspirin rather than clopidogrel. Patients not adequately represented include women (21.4%), non-East Asian ethnicities, patients with complex frailty, dementia, or severe renal impairment, and those on NOACs other than apixaban or rivaroxaban. The apparent magnitude of benefit in the elderly subgroup is clinically meaningful if confirmed, given the high background bleeding risk of this population with dual antithrombotic exposure. However, as a post hoc analysis — not a pre-specified primary subgroup in the parent trial — these findings are hypothesis-generating and should inform clinical judgment rather than replace individualized risk–benefit assessment. Note: Only an abstract was available for this post hoc publication; complete data, interaction statistics, and subgroup-level component outcomes may differ from what is summarized here. __LIMITATIONS__ Post hoc secondary analysis: not powered or pre-specified as a primary subgroup analysis; results are hypothesis-generating and subject to type I and type II error,Abstract truncated: the full 95% CI for the primary aHR in patients ≥75 years and the p-value are not available; interaction p-value between age and treatment not reported,Open-label design: lack of blinding may introduce ascertainment bias, particularly for subjectively adjudicated endpoints such as clinically relevant non-major bleeding,Single-country (South Korean) East Asian cohort: pharmacogenomic differences (e.g., CYP2C19 metabolism relevant to clopidogrel) and baseline cardiovascular risk profiles may limit generalizability to Western or other populations,Short follow-up (12 months): insufficient to detect differences in late stent thrombosis, long-term ischemic events, or mortality trends,Significant NOAC underdosing: 29% in combination arm and 11.3% in monotherapy arm received subtherapeutic doses, potentially confounding safety and efficacy comparisons,DES implantation ≥12 months prior only: findings do not apply to the early post-PCI period or to patients with bare-metal stents,Parent trial non-inferiority design: superiority and subgroup results should be interpreted cautiously as they were secondary analyses within a non-inferiority framework,Women underrepresented (21.4%): limits generalizability to female patients, who may have different bleeding and thromboembolic risk profiles,Partial industry sponsorship (Samjin Pharmaceutical): potential for influence on trial conduct or reporting, though the magnitude of effect is unlikely to be entirely sponsor-driven __BOTTOM_LINE_CATEGORY__ potentially_practice_informing __BOTTOM_LINE__ In this post hoc analysis of the ADAPT AF-DES trial, NOAC monotherapy was associated with markedly fewer net adverse clinical events compared with NOAC plus clopidogrel in AF patients aged ≥75 years with prior DES implantation (9.8% vs. 22.3%; aHR 0.37), suggesting older patients may derive an even greater absolute benefit from dropping antiplatelet therapy than the overall trial cohort. These findings are hypothesis-generating given the post hoc design, incomplete abstract data, and absence of reported interaction statistics, and should reinforce rather than replace individualized clinical decision-making in this high-risk, bleeding-prone population. Clinicians managing older AF patients with stable coronary disease beyond 1 year post-DES should consider NOAC monotherapy as the preferred long-term strategy, consistent with current ACC/AHA and ESC guideline recommendations, while ensuring appropriate NOAC dosing.