Articles

Ticagrelor alone versus ticagrelor plus aspirin from month 1

to month 12 after percutaneous coronary intervention in
patients with acute coronary syndromes (ULTIMATE-DAPT):
a randomised, placebo-controlled, double-blind clinical trial
Zhen Ge*, Jing Kan*, Xiaofei Gao*, Afsar Raza, Jun-Jie Zhang, Bilal S Mohydin, Fentang Gao, Yibing Shao, Yan Wang, Hesong Zeng, Feng Li,
Hamid Sharif Khan, Naeem Mengal, Hongliang Cong, Mingliang Wang, Lianglong Chen, Yongyue Wei, Feng Chen, Gregg W Stone,
Shao-Liang Chen, for the ULTIMATE-DAPT investigators†

Summary
Background Following percutaneous coronary intervention with stent placement to treat acute coronary syndromes, Published Online
international clinical guidelines generally recommend dual antiplatelet therapy with aspirin plus a P2Y12 receptor April 7, 2024
https://doi.org/10.1016/
inhibitor for 12 months to prevent myocardial infarction and stent thrombosis. However, data on single antiplatelet S0140-6736(24)00473-2
therapy with a potent P2Y12 inhibitor earlier than 12 months after percutaneous coronary intervention for patients
See Online/Comment
with an acute coronary syndrome are scarce. The aim of this trial was to assess whether the use of ticagrelor alone, https://doi.org/10.1016/
compared with ticagrelor plus aspirin, could reduce the incidence of clinically relevant bleeding events without an S0140-6736(24)00586-5
accompanying increase in major adverse cardiovascular or cerebrovascular events (MACCE). For the Mandarin translation of
the abstract see Online for
appendix 1
Methods In this randomised, placebo-controlled, double-blind clinical trial, patients aged 18 years or older with an acute
coronary syndrome who completed the IVUS-ACS study and who had no major ischaemic or bleeding events *Contributed equally

after 1-month treatment with dual antiplatelet therapy were randomly assigned to receive oral ticagrelor (90 mg twice †Study group members are listed
in appendix 2 (pp 5–6)
daily) plus oral aspirin (100 mg once daily) or oral ticagrelor (90 mg twice daily) plus a matching oral placebo, beginning
Nanjing First Hospital, Nanjing
1 month and ending at 12 months after percutaneous coronary intervention (11 months in total). Recruitment took place
Medical University, Nanjing,
at 58 centres in China, Italy, Pakistan, and the UK. Patients were required to remain event-free for 1 month on dual China (Z Ge MD, J Kan MD,
antiplatelet therapy following percutaneous coronary intervention with contemporary drug-eluting stents. Randomisation X Gao MD, J-J Zhang MD,
was done using a web-based system, stratified by acute coronary syndrome type, diabetes, IVUS-ACS randomisation, Prof S-L Chen MD); Airdale
General Hospital,
and site, using dynamic minimisation. The primary superiority endpoint was clinically relevant bleeding (Bleeding West Yorkshire, UK (A Raza MD);
Academic Research Consortium [known as BARC] types 2, 3, or 5). The primary non-inferiority endpoint was MACCE Punjab Institute of Cardiology,
(defined as the composite of cardiac death, myocardial infarction, ischaemic stroke, definite stent thrombosis, or Lahore, Pakistan
clinically driven target vessel revascularisation), with an expected event rate of 6·2% in the ticagrelor plus aspirin group (B S Mohydin MD); Gansu
Provincial People’s Hospital,
and an absolute non-inferiority margin of 2·5 percentage points between 1 month and 12 months after percutaneous Lanzhou, China (F Gao MD);
coronary intervention. The two co-primary endpoints were tested sequentially; the primary superiority endpoint had to Qingdao Municipal Hospital,
be met for hypothesis testing of the MACCE outcome to proceed. All principal analyses were assessed in the intention- Qingdao, China (Y Shao MD);
to-treat population. This trial is registered with ClinicalTrials.gov, NCT03971500, and is completed. Xiamen Heart Center, Xiamen
University, Xiamen, China
(Y Wang MD); Tongji Hospital,
Findings Between Sept 21, 2019, and Oct 27, 2022, 3400 (97·0%) of the 3505 participants in the IVUS-ACS study were Tongji Medical College,
randomly assigned (1700 patients to ticagrelor plus aspirin and 1700 patients to ticagrelor plus placebo). 12-month follow- Huazhong University of Science
up was completed by 3399 (>99·9%) patients. Between month 1 and month 12 after percutaneous coronary intervention, and Technology, Wuhan, China
(H Zeng MD); Affiliated Oriental
clinically relevant bleeding occurred in 35 patients (2·1%) in the ticagrelor plus placebo group and in 78 patients (4·6%) Huainan General Hospital,
in the ticagrelor plus aspirin group (hazard ratio [HR] 0·45 [95% CI 0·30 to 0·66]; p<0·0001). MACCE occurred in Anhui University of Science and
61 patients (3·6%) in the ticagrelor plus placebo group and in 63 patients (3·7%) in the ticagrelor plus aspirin group Technology, Huainan, China
(absolute difference –0·1% [95% CI –1·4% to 1·2%]; HR 0·98 [95% CI 0·69 to 1·39]; pnon-inferiority<0·0001, psuperiority=0·89). (F Li MD); Rawalpindi Institute
of Cardiology, Rawalpindi,
Pakistan (H S Khan MD);
Interpretation In patients with an acute coronary syndrome who had percutaneous coronary intervention with National Institute of
contemporary drug-eluting stents and remained event-free for 1 month on dual antiplatelet therapy, treatment with Cardiovascular Diseases of
Pakistan, Karaqi, Pakistan
ticagrelor alone between month 1 and month 12 after the intervention resulted in a lower rate of clinically relevant
(N Mengal MD); Tianjin Chest
bleeding and a similar rate of MACCE compared with ticagrelor plus aspirin. Along with the results from previous Hospital, Tianjin University,
studies, these findings show that most patients in this population can benefit from superior clinical outcomes with Tianjin, China (H Cong MD);
aspirin discontinuation and maintenance on ticagrelor monotherapy after 1 month of dual antiplatelet therapy. Puto People’s Hospital, Tongji
University, Shanghai, China
(M Wang MD); Fujian Medical
Funding The Chinese Society of Cardiology, the National Natural Scientific Foundation of China, and the Jiangsu University Union Hospital,
Provincial & Nanjing Municipal Clinical Trial Project. Fuzhou, China (L Chen MD);
Center for Public Health and
Epidemic Preparedness &
Copyright © 2024 Elsevier Ltd. All rights reserved.

www.thelancet.com Published online April 7, 2024 https://doi.org/10.1016/S0140-6736(24)00473-2 1

 Articles

Response, Peking University,

Beijing, China (Prof Y Wei PhD); Research in context
School of Public Health, Center
of Global Health, Nanjing Evidence before this study The results of this study show that patients with an acute
Medical University, Nanjing, We searched PubMed on March 11, 2019, using the search coronary syndrome treated with contemporary drug-eluting
China (Prof F Chen PhD); The terms “intravascular ultrasound”, “angiography”, “percutaneous stents and who are event-free after a 1-month course of dual
Zena and Michael A Wiener
Cardiovascular Institute, Icahn
coronary intervention”, “antiplatelet therapy”, and “acute antiplatelet therapy can be safely maintained on ticagrelor
School of Medicine at Mount coronary syndrome (ACS)” using MeSH terms and appropriate monotherapy from 1 month after to 12 months after
Sinai, New York, NY, USA variations, with no language or date restrictions, before percutaneous coronary intervention, and that this regimen will
(Prof G W Stone MD) designing our study. Subgroup analyses from several substantially reduce major and minor bleeding compared with
Correspondence to: randomised trials suggested that a course of dual antiplatelet standard ongoing treatment with ticagrelor plus aspirin. In
Prof Shao-Liang Chen, Nanjing
therapy lasting 1–3 months using different P2Y12 inhibitors addition, no increase in ischaemic events such as myocardial
First Hospital, Nanjing Medical
University, might safely reduce bleeding events in patients with an acute infarction, stent thrombosis, or cardiac death was observed
Nanjing 210006, China coronary syndrome after implantation of drug-eluting stents. with the ticagrelor monotherapy regimen in patients with a
[email protected]; We found no previous randomised, placebo-controlled trials in high-risk acute coronary syndrome, and net adverse clinical
@shao_liangchen
patients with an acute coronary syndrome who remained events occurred less frequently with ticagrelor monotherapy
or
event-free after 1 month of dual antiplatelet therapy that than with ticagrelor plus aspirin.
Prof Gregg W Stone,
compared ticagrelor alone with ticagrelor plus aspirin between
The Zena and Michael A Wiener Implications of all the available evidence
Cardiovascular Institute, month 1 and month 12 of follow-up.
The data from ULTIMATE-DAPT (along with that of previous
Icahn School of Medicine at
Mount Sinai,
Added value of this study studies) provide strong evidence that, in patients with an acute
New York, NY 10029, USA ULTIMATE-DAPT is the first randomised controlled trial to coronary syndrome treated with percutaneous coronary
[email protected]; compare dual antiplatelet therapy with a potent platelet intervention with contemporary drug-eluting stents and who
@greggwstone P2Y12 inhibitor plus aspirin versus a potent P2Y12 inhibitor plus are event-free after 1 month of post-procedural dual
See Online for appendix 2 placebo in patients with an acute coronary syndrome treated antiplatelet therapy, ticagrelor monotherapy can safely reduce
with percutaneous coronary intervention and who remained major and minor bleeding during follow-up.
event-free after only 1 month of dual antiplatelet therapy.

Introduction all-cause death in patients with an acute coronary

Despite advances in prevention and treatment, syndrome. However, P2Y12 inhibitor monotherapy after
cardiovascular disease remains the leading cause of percutaneous coronary intervention and only 1 month of
mortality worldwide, with ischaemic heart disease among dual antiplatelet therapy in acute coronary syndromes
the top contributors to burden of disease after the age of has not been widely studied, and most of the included
25 years.1 Patients with an acute coronary syndrome are at trials were open label.16 Therefore, we performed a
increased risk of death and myocardial infarction placebo-controlled randomised trial in patients with an
compared with patients with chronic coronary syndrome. acute coronary syndrome who remained event-free after
The prognosis of these patients can be improved by early percutaneous coronary intervention and 1 month of dual
revascularisation.2,3 Following percutaneous coronary antiplatelet therapy. The aim of the study was to assess
intervention for acute coronary syndromes, international whether the use of ticagrelor alone, compared with
guidelines currently recommend that most patients ticagrelor plus aspirin, could reduce the incidence of
receive dual antiplatelet therapy with aspirin plus a clinically relevant bleeding events without an
P2Y12 receptor inhibitor for 12 months to reduce the risk accompanying increase in major adverse cardiovascular
of myocardial infarction and stent thrombosis.2,3 or cerebrovascular events (MACCE).
Ticagrelor or prasugrel are preferred to clopidogrel in this
setting.4,5 However, the benefit of these potent Methods
P2Y12 receptor inhibitors is obtained at the cost of an Study design and participants
increased risk of bleeding events, which can be life- This randomised, placebo-controlled, double-blind
threatening. clinical trial was part of the IVUS-ACS and ULTIMATE-
The effects of early cessation of antiplatelet therapy DAPT integrated study programme (comparison of
with aspirin after percutaneous coronary intervention in 1-month vs 12-month dual antiplatelet therapy after
acute and chronic coronary syndromes have been studied implantation of drug-eluting stents guided by either
in various settings.4–15 A 2023 meta-analysis16 of intravascular ultrasound or angiography in patients with
32 randomised trials testing numerous pharmacological an acute coronary syndrome), an investigator-initiated
strategies reported that P2Y12 inhibitor monotherapy international study. The programme incorporated two
after percutaneous coronary intervention and a 6-month integrated ran­ dom­ised trials that were conducted at
or less treatment period with dual antiplatelet therapy 58 centres (ie, teaching hospitals; appendix 2 pp 5–6) in
was associated with the lowest risk of major bleeding and China (n=52), Pakistan (n=4), the UK (n=1), and

2 www.thelancet.com Published online April 7, 2024 https://doi.org/10.1016/S0140-6736(24)00473-2

 Articles

Italy (n=1). The study complied with the Declaration initiated before percutaneous coronary intervention
of Helsinki, and the protocol was approved by the (timing was dependent upon whether the patient had
institutional review board or ethics committee at each angina, NSTEMI, or STEMI).
centre. All patients provided written informed consent. A All patients or their family members provided written
data safety and monitoring board oversaw the trial informed consent before random assignment; for
(appendix 2 pp 2–4). patients with unstable angina or NSTEMI, there was
The protocol integrated two trials in a two-stage sufficient time to introduce the percutaneous coronary
factorial design to assess the utility of intravascular intervention procedure and this study to patients; for
ultrasound guidance of percutaneous coronary inter­ patients with STEMI, research staff briefly explained the
vention (IVUS-ACS trial) and ticagrelor monotherapy details to patients and their families immediately after
after 1 month of dual antiplatelet therapy following wiring and thrombolysis in myocardial infarction (TIMI)
percutaneous coronary intervention in patients with flow restoration. Sex data were collected according to
an acute coronary syndrome (ULTIMATE-DAPT). physical examination.
The present report describes the outcomes of the
ULTIMATE-DAPT trial. The principal results from the Randomisation and masking
IVUS-ACS trial are discussed in a separate publication.17 After 1 month, surviving patients who remained free
This trial is registered with ClinicalTrials.gov, from major bleeding events (Bleeding Academic
NCT03971500, and is completed. Research Consortium [BARC] types 3 or 5),19 adverse
Patients were eligible for inclusion in the trial if ischaemic events (myocardial infarction, ischaemic
they were aged 18 years or older; had an acute stroke, definite stent thrombosis, or clinically driven
coronary syndrome (ie, unstable angina [angiography target vessel revascularisation), and who had not
showing a severely narrowed or ruptured plaque or discontinued dual antiplatelet therapy for more than 48 h
thrombotic lesion without cardiac biomarker elevation], continued treatment with open-label ticagrelor and were
non-ST-segment elevation myocardial infarction randomly assigned for a second time, in a 1:1 ratio, to
[NSTEMI], or ST-segment elevation myocardial infarction receive dual antiplatelet therapy (90 mg ticagrelor twice
[STEMI]) caused by a culprit lesion in an untreated daily plus 100 mg enteric-coated aspirin once daily) or
coronary artery segment, up to 30 days before monotherapy (ie, 90 mg ticagrelor twice daily plus
randomisation;18 and had an indication for percutaneous a matching enteric-coated oral placebo) for an
coronary intervention with a second-generation drug- additional 11 months (second randomisation for the
eluting stent (appendix 2 p 8). Exclusion criteria were ULTIMATE-DAPT trial). 30 days was selected as the
stroke within 3 months or any permanent neurological timepoint for the second randomisation because this
deficit; any previous intracranial bleed or intracranial corresponds to when a clinician would decide whether
disease (eg, aneurysm or fistula); previous coronary the patient is sufficiently stable (ie, free from ischaemic
artery bypass graft surgery; any planned surgery within and bleeding events) to undergo aspirin discontinuation.6,11
12 months; any reason for which antiplatelet therapy Moreover, if randomisation for this decision was made
might need to be discontinued within 12 months; before 30 days while the patient was in hospital, inclusion
severe chronic kidney disease (defined as an estimated of late event data (>30 days) from patients who would not
glomerular filtration rate <20 mL/min per 1·73 m²); need have been eligible at 30 days might confound the results,
for chronic oral anticoagulation (ie, warfarin or coumadin and inclusion of early event data (<30 days) that would be
or direct oral anticoagulants); a platelet count of less than expected to be the same in both randomised groups still
100 000 mm³; contraindication to aspirin or ticagrelor; on dual antiplatelet therapy could bias the results toward
liver cirrhosis; people intending to become pregnant; a the null.
life expectancy of less than 1 year; and any condition Research staff were responsible for enrolling and
likely to interfere with study processes, including randomly assigning participants. Randomisation was
medication compliance or follow-up visits (eg, dementia, done using a web-based system, stratified by acute
alcohol abuse, severe frailty, or required to travel a long coronary syndrome type (unstable angina vs NSTEMI vs
distance for follow-up visits). STEMI), diabetes (yes vs no), the first-stage randomisation
Eligible patients were initially randomly assigned to (percutaneous coronary inter­vention with intravascular
receive percutaneous coronary intervention, guided by ultrasound or angiography guidance), and site, using
either intravascular ultrasound or angiography (first dynamic minimisation.20 The nurses and research staff
randomisation [IVUS-ACS trial]), of all lesions that each did not have any involvement in the rest of the trial. The
site operator believed were responsible for the acute allocation probability for each eligible patient was
coronary syndrome, as well as of other angiographically generated using the Pocock and Simon method.21 Drugs
severe lesions that warranted treatment per standard in both groups were labelled with pre-generated,
clinical practice. After the procedure, all patients were randomly shuffled numbers. When the algorithm
treated with oral ticagrelor (90 mg twice daily) plus oral generated allocation results for a participant, the tool
enteric-coated aspirin (100 mg daily) with the first doses would map the allocation result to one of the drug

www.thelancet.com Published online April 7, 2024 https://doi.org/10.1016/S0140-6736(24)00473-2 3

 Articles

numbers of the same group without exposing the and Tissue Plasminogen Activator for Occluded
allocation to the investigators. All patients and their Arteries (GUSTO);23 major bleeding according to
families, researchers, and treating physicians and staff definitions from International Society of Thrombosis
remained masked to random assignment during the and Haemostasis (ISTH);24 individual components of the
1-year follow-up. primary endpoints; and the composite of cardiac death,
non-fatal myocardial infarction, or ischaemic stroke.
Procedures Other secondary endpoints were target vessel failure
Post-discharge visits after the initial percutaneous without procedural myocardial infarction, non-fatal
coronary intervention were scheduled at 1, 4, 6, myocardial infarction, and probable stent thrombosis
and 12 months (with the second random assignment for (protocol version 2.0, appendix 2 p 97).
eligible patients done at the 1-month visit after physician
interview and physical examination demonstrated that Statistical analysis
they met eligibility criteria). A window of 23–37 days after Historical data from previous clinical trials used to
percutaneous coronary intervention was allowed for the estimate control arm event rates are shown in appendix 2
1-month visit and second randomisation. Angiographical (p 15). Assuming a 3·0% rate of clinically relevant
follow-up was done only for clinical indications. If bleeding between month 1 and month 12 of ticagrelor
dyspnoea developed on ticagrelor, the dosage was plus aspirin, 3400 patients provided 80% power to detect
reduced from 90 mg twice daily to 60 mg twice daily; if a 50% reduction in clinically relevant bleeding with
dyspnoea persisted, ticagrelor was replaced with oral ticagrelor plus placebo with a two-sided α of 0·05.
clopidogrel (75 mg daily, based on physician judgement). Assuming a 6·2% rate of MACCE between month 1 and
Ticagrelor plus open-label aspirin was prescribed for any month 12 of ticagrelor plus aspirin, inclusion of
patient who had a revascularisation procedure during the 3068 patients provided 80% power to show non-inferiority
follow-up period. Management of antiplatelet therapy in of ticagrelor plus placebo with an absolute margin of
patients who had a bleeding or ischaemic event was left 2·5 percentage points with a one-sided α of 0·025.
to the physician’s discretion. Unmasking was allowed Acknowledging the inherent uncertainty in selecting a
by the data safety monitoring board if knowledge of non-inferiority margin25 and the likelihood that the
the antiplatelet agent allocation was necessary for majority of MACCE between month 1 and month 12
management of a major adverse event (appendix 2 p 23) would consist of repeat revascularisations, the steering
and did not lead to data censoring. Baseline angiograms committee selected a margin of 2·5% to represent a
of the target lesion or lesions responsible for the acute reasonable estimate for therapeutic interchangeability in
coronary syndrome, as identified by the operator, were the estimated difference in MACCE between groups,
analysed at an independent core laboratory. A masked given the expected magnitude for the reduction in major
independent clinical events committee (appendix 2 p 2) and minor bleeding in this period with ticagrelor alone.
adjudicated all outcomes to prespecified endpoint The two primary endpoints were tested sequentially; the
definitions upon review of hospital documents and clinically relevant bleeding (superiority) endpoint had to
reports. pass for hypothesis testing of the MACCE (non-
inferiority) endpoint to proceed. The protocol prespecified
Outcomes that, if non-inferiority of MACCE was demonstrated, the
The trial had two primary endpoints (appendix 2 p 8). hypothesis of superiority on MACCE of monotherapy
The primary superiority endpoint was the occurrence of over dual antiplatelet therapy would also be tested at
clinically relevant bleeding (BARC types 2, 3, or 5).19 The a one-sided significance level of 0·025. Protocol
primary non-inferiority endpoint was the occurrence of amendments were minor, did not affect study processes,
MACCE (ie, the composite of cardiac death, myocardial and were approved by the Institutional Research Board
infarction, ischaemic stroke, definite stent thrombosis, on March 28, 2021 (protocol version 2.0, appendix 2
and clinically driven target vessel revascularisation). The pp 121–24).
key secondary endpoints were the time to first occurrence Continuous data were reported as mean (SD) and
of composite net adverse clinical events (defined as compared using t test if normally distributed, or as
any MACCE or any BARC bleeding [types 1, 2, 3, or 5]) median (IQR) and compared using Mann–Whitney U test
and percentage of crossover from ticagrelor to if not normally distributed. Binary data were compared
clopidogrel in each group. BARC type 1 bleeding was using the χ² test or Fisher’s exact test. Event rates were
included in the net clinical adverse event endpoint to estimated using the Kaplan–Meier method and compared
fully reflect the totality of bleeding events that patients using the log-rank test. Treatment effects were estimated
might have. Additional endpoints (appendix 2 pp 8–14) using Cox proportional hazards regression, with results
were BARC types 3 or 5 bleeding; major or minor presented as hazard ratios (HRs) and 95% CIs. Absolute
bleeding according to the TIMI risk assessment score;22 differences and 95% CIs for the primary and secondary
moderate, severe, or life-threatening bleeding according endpoints were calculated with standard errors estimated
to definitions from Global Utilization of Streptokinase by Greenwood’s method.26 MACCE were analysed using

4 www.thelancet.com Published online April 7, 2024 https://doi.org/10.1016/S0140-6736(24)00473-2

 Articles

the subdistribution method of Fine and Gray to account count at the 4-month and 12-month office visits).
for the competing risk of non-cardiac death. The Adjustment for multiplicity was not done for the
treatment effects for both primary analyses were adjusted secondary endpoints, and these results should not be
in the multivariable models for acute coronary syndrome used to infer definitive treatment effects.
type, the first randomisation, diabetes, and geographical Patients were censored if they withdrew, were lost to
region (ie, Italy, Pakistan, the UK, east China, follow-up, or died. The latest available follow-up data were
north China, south China, or west China). used for each patient. Missing data were not replaced; all
The principal analyses were assessed in the intention- outcomes are reported as a complete-case analysis. All
to-treat population, including all randomly assigned superiority tests were two-sided and p<0·05 was
patients, regardless of medication adherence. Sensitivity considered significant for superiority testing. All analyses
analyses were done in the per-protocol population, were done using SAS (version 9.4). This study is registered
excluding patients who did not take their assigned with ClinicalTrials.gov, NCT03971500, and is completed.
treatment for at least 7 days (unless medication
interruption or discontinuation was required due to Role of the funding source
adverse events), or who did not comply with the The funders of the study had no role in study design,
medication schedule (defined as a medication possession data collection, data analysis, data interpretation, or
ratio of >80% of dispensed tablets based on manual pill writing of the report.

3505 patients randomly assigned

1753 assigned to IVUS-guided PCI and ticagrelor 1752 assigned to angiography-guided PCI and
plus aspirin for 1 month ticagrelor plus aspirin for 1 month

40 did not undergo second randomisation 65 did not undergo second randomisation
17 had dyspnoea from ticagrelor 23 had dyspnoea from ticagrelor
9 had clinical events within 30 days* 12 patients refused
5 had BARC 3 or 5 bleeding 10 had clinical events within 30 days*
5 patients refused 9 had BARC 3 or 5 bleeding
2 discontinued treatment for ≥48 h 6 discontinued treatment for ≥48 h
1 had an allergy to ticagrelor 4 needed chronic oral anticoagulation
1 lost to follow-up 1 had an allergy to ticagrelor

1713 randomly assigned for a second time 1687 randomly assigned for a second time

857 assigned to ticagrelor plus 856 assigned to ticagrelor plus 843 assigned to ticagrelor plus 844 assigned to ticagrelor plus
aspirin placebo aspirin placebo

1700 in the ticagrelor plus placebo group 1700 in the ticagrelor plus aspirin group

1 lost to follow-up

1699 completed 11-month follow-up 1700 completed 11-month follow-up

1700 in the intention-to-treat analysis 1700 in the intention-to-treat analysis

1696 in the per-protocol analysis 1699 in the per-protocol analysis

Figure 1: Trial profile

The intention-to-treat population included all randomly assigned patients, regardless of therapy received. The per-protocol population consists of all patients as
randomised, excluding those who did not take their assigned treatment for at least 7 days (unless medication interruption or discontinuation was required for
adverse events), or who did not comply with the medication schedule. BARC=Bleeding Academic Research Consortium. IVUS=intravascular ultrasound.
PCI=percutaneous coronary intervention. *Clinically driven target vessel revascularisation, definite stent thrombosis, stroke, ST-segment elevation myocardial
infarction, BARC 3 or 5 major bleeding, or death.

www.thelancet.com Published online April 7, 2024 https://doi.org/10.1016/S0140-6736(24)00473-2 5

 Articles

Results
Ticagrelor plus Ticagrelor plus Between Sept 21, 2019, and Oct 27, 2022, 3710 patients
placebo (n=1700) aspirin (n=1700) with an acute coronary syndrome were screened and
Age, years 3505 (94·5%) were randomly assigned to intravascular
Median (IQR) 62 (54–70) 63 (54–69) ultrasound-guided or angiography-guided percutaneous
Sex coronary intervention and subsequently discharged on
Male 1264 (74·4%) 1257 (73·9%) dual antiplatelet therapy (figure 1). 3400 (97·0%) of the
Female 436 (25·7%) 443 (26·1%) 3505 patients underwent the second random assignment
Race
at 1 month (median 31 days [IQR 30–34] for both groups)
Chinese 1476 (86·8%) 1519 (89·4%)
to receive either ticagrelor plus placebo (n=1700) or
Other 224 (13·2%) 181 (10·7%)
ticagrelor plus aspirin (n=1700); these 3400 patients
Country of enrolment
comprise the study population of ULTIMATE-DAPT
(appendix 2 p 16).
China 1476 (86·8%) 1519 (89·4%)
Baseline clinical characteristics were well matched
Pakistan 202 (11·9%) 159 (9·4%)
between groups (table 1), as were index procedural data
UK 12 (0·7%) 11 (0·7%)
and angiographical outcomes (appendix 2 pp 17–19). The
Italy 10 (0·6%) 11 (0·7%)
median patient age was 63 years (IQR 54–70),
Initial presentation
2521 (74·1%) patients were men and 879 (25·6%) were
Unstable angina 668 (39·3%) 708 (41·7%)
women, 1075 (31·6%) had diabetes, and 2024 (59·5%)
With ischaemic changes on 650/668 (97·3%) 685/708 (96·8%)
electrocardiogram
presented with an acute myocardial infarction.
Acute myocardial infarction 1032 (60·7%) 992 (58·4%)
835 (24·6%) of 3400 culprit vessels before the index
Non-STEMI 545 (32·1%) 531 (31·2%)
procedure had a TIMI flow score of less than 3. The
median dual antiplatelet therapy duration was 28 days
STEMI 487 (28·7%) 461 (27·1%)
(IQR 25–33) in the ticagrelor plus placebo group and
Killip class
365 days (365–365) in the ticagrelor and aspirin group.
1 475/1032 (46·0%) 442/992 (44·6%)
During follow-up, 12 (0·7%) patients required a reduction
2 457/1032 (44·3%) 468/992 (47·2%)
in ticagrelor dosage from 90 mg to 60 mg twice daily in
3 100/1032 (9·7%) 82/992 (8·3%)
the ticagrelor plus placebo group, as did 16 (0·9%)
Medical history
patients in the ticagrelor plus aspirin group. Conversion
Hypertension 1058 (62·2%) 1063 (62·5%)
from ticagrelor to clopidogrel was required in 22 (1·3%)
Diabetes 540 (31·8%) 535 (31·5%)
patients in the ticagrelor plus placebo group and
Insulin-treated diabetes 136 (8·0%) 144 (8·5%)
19 (1·1%) patients in the ticagrelor plus aspirin group.
Dyslipidaemia 1178 (69·3%) 1157 (68·1%)
Unmasking during follow-up was required for 11 (0·7%)
Current smoking* 486 (28·6%) 482 (28·4%) patients in the ticagrelor plus placebo group and
Chronic renal insufficiency† 119 (7·0%) 129 (7·6%) 28 (1·7%) in the ticagrelor plus aspirin group due to
Previous percutaneous coronary intervention 171 (10·1%) 174 (10·2%) BARC 3 or 5 bleeding events, and for three (0·2%)
Previous coronary artery bypass graft surgery 2 (0·1%) 4 (0·2%) patients in the ticagrelor plus placebo group and
Previous myocardial infarction 143 (8·4%) 156 (9·2%) five (0·3%) in the ticagrelor plus aspirin group due to
Previous stroke 154 (9·1%) 147 (8·7%) definite stent thrombosis. Pill counts at 4 months and
Peripheral arterial disease 76 (4·5%) 79 (4·7%) 12 months and patient reporting of medication use at
Heart failure 109 (6·4%) 101 (5·9%) other time intervals showed acceptable antiplatelet agent
Left ventricular ejection fraction (%)‡ adherence during follow-up in both groups
Median (IQR) 62% (55–65) 63% (56–65) (appendix 2 p 22). 12-month clinical follow-up was
Medication use at 1 month (the time of the second randomisation) completed by 3399 (>99·9%) of 3400 participants.
Aspirin 1700 (100%) 1700 (100%) Angiographical follow-up was done for clinical
Ticagrelor 1700 (100%) 1700 (100%) indications in 256 (15·1%) patients assigned to ticagrelor
β blocker 838 (49·3%) 802 (47·2%) plus placebo at a median of 347 days (IQR 329–368) after
Angiotensin converting enzyme inhibitor or 774 (45·5%) 788 (46·4%) percutaneous coronary intervention, and in 278 (16·4%)
angiotensin receptor blocker patients assigned to ticagrelor plus aspirin at a median of
Calcium antagonist 430 (25·3%) 463 (27·2%) 347 days (IQR 331–370) after percutaneous coronary
Statin 1417 (83·4%) 1404 (82·6%) intervention.
Data are n (%) unless otherwise specified. STEMI=ST-segment elevation myocardial infarction. *Defined as Between month 1 and month 12 after percuta­
≥100 lifetime cigarettes and still smoking at the time of enrolment; other tobacco products were not included. neous coronary intervention, the primary superiority
†Defined as an estimated glomerular filtration rate of <60 mL/min per 1·73 m². ‡Includes baseline transthoracic
endpoint of clinically relevant bleeding occurred in
echocardiographic measurements from 1496 patients in the ticagrelor plus placebo group and 1504 patients in the
ticagrelor plus aspirin group. 35 patients (2·1%) treated with ticagrelor plus placebo
and in 78 patients (4·6%) treated with ticagrelor plus
Table 1: Baseline characteristics and medication use
aspirin (HR 0·45 [95% CI 0·30–0·66]; p<0·0001) in the

6 www.thelancet.com Published online April 7, 2024 https://doi.org/10.1016/S0140-6736(24)00473-2

 Articles

intention-to-treat population (table 2; figure 2); similar

Ticagrelor plus Ticagrelor plus Hazard ratio p value
results were observed in the per-protocol population placebo (n=1700) aspirin (n=1700) (95% CI)
(appendix 2 p 20). Major bleeding (BARC types 3 or 5)
Primary endpoints
also occurred less frequently with ticagrelor plus placebo
Clinically relevant bleeding 35 (2·1%) 78 (4·6%) 0·45 (0·30 to 0·66) <0·0001
compared with ticagrelor plus aspirin, as did other (BARC types 2, 3, or 5)
prespecified measures of bleeding (table 2; Major adverse cardiovascular 61 (3·6%) 63 (3·7%) 0·98 (0·69 to 1·39) 0·89†
appendix 2 p 21). The treatment effects for clinically or cerebrovascular events*
relevant bleeding were consistent across prespecified Key secondary endpoint
subgroups (figure 3). Numbers needed-to-treat were 48 to Net adverse clinical events 97 (5·7%) 140 (8·2%) 0·68 (0·53 to 0·88) 0·0066
prevent one clinically relevant bleeding event and 100 to Secondary bleeding endpoints
prevent one BARC type 3 or 5 bleeding event. Major bleeding (BARC types 3 11 (0·7%) 28 (1·7%) 0·39 (0·19 to 0·79) 0·0087
Between month 1 and month 12 after percutaneous or 5)
coronary intervention, the primary non-inferiority TIMI major or minor bleeding 11 (0·7%) 27 (1·6%) 0·41 (0·20 to 0·82) 0·012
endpoint of MACCE occurred in 61 participants (3·6%) GUSTO moderate, severe, or 8 (0·5%) 19 (1·1%) 0·42 (0·18 to 0·96) 0·041
treated with ticagrelor plus placebo and 63 (3·7%) treated life-threatening bleeding
with ticagrelor plus aspirin (absolute difference –0·1% ISTH major bleeding 8 (0·5%) 21 (1·2%) 0·38 (0·17 to 0·86) 0·020
[95% CI –1·4% to 1·2%]; HR 0·98 [95% CI 0·69 to 1·39]; Patients with any of the four 11 (0·7%) 28 (1·7%) 0·39 (0·19 to 0·79) 0·0087
secondary bleeding endpoints
pnon-inferiority<0·0001, psuperiority=0·89) in the intention-to-treat
Patients with all four 8 (0·5%) 19 (1·1%) 0·42 (0·18 to 0·96) 0·041
population (table 2, figure 2); similar results were observed secondary bleeding endpoints
in the per-protocol population (appendix 2 p 20). The rates
Secondary ischaemic events and mortality
of the individual components of the primary non-
All-cause death 12 (0·7%) 13 (0·8%) 0·93 (0·42 to 2·03) 0·84
inferiority endpoint were also similar between the two
Cardiac death 8 (0·5%) 7 (0·4%) 1·15 (0·42 to 3·18) 0·46
groups, as were other composite ischaemic outcomes
Any stroke 20 (1·2%) 24 (1·4%) 0·83 (0·46 to 1·50) 0·54
(table 2). The treatment effects for MACCE in the
Ischaemic 11 (0·7%) 15 (0·9%) 0·74 (0·34 to 1·61) 0·58
prespecified subgroups are shown in figure 4.
Haemorrhagic 4 (0·2%) 3 (0·2%) 1·33 (0·29 to 3·71) 0·71
Between month 1 and month 12 after percutaneous
Uncertain 5 (0·3%) 6 (0·4%) 0·97 (0·49 to 1·92) 0·94
coronary intervention, the key secondary endpoint of net
Myocardial infarction 17 (1·0%) 11 (0·7%) 1·45 (0·67 to 3·23) 0·27
clinical adverse events occurred less frequently in the
Procedural myocardial 1 (0·1%) 1 (0·1%) 0·00 (–0·28 to 0·28) 0·88
ticagrelor plus placebo group than in the ticagrelor plus
infarction
aspirin group (table 2; appendix 2 p 23).
Non-procedural myocardial 16 (0·9%) 11 (0·7%) 1·42 (0·66 to 3·03) 0·29
infarction
Discussion Stent thrombosis 5 (0·3%) 5 (0·3%) 0·97 (0·28 to 3·40) 0·96
The ULTIMATE-DAPT placebo-controlled, double-blind Definite 3 (0·2%) 5 (0·3%) 0·59 (0·14 to 2·51) 0·47
trial showed that, in patients with an acute coronary Probable 2 (0·1%) 0 ·· ··
syndrome treated with percutaneous coronary inte­ Clinically driven 40 (2·4%) 41 (2·4%) 0·99 (0·64 to 1·53) 0·95
rvention followed by 1 month of aspirin plus ticagrelor, revascularisation‡
follow-up treatment with ticagrelor monotherapy for an Target vessel 33 (2·0%) 36 (2·1%) 0·93 (0·58 to 1·49) 0·75
additional 11 months reduced clinically relevant bleeding revascularisation
without an accompanying increase in MACCE, compared Target lesion 27 (1·6%) 28 (1·7%) 0·97 (0·57 to 1·65) 0·92
with follow-up treatment with ticagrelor and aspirin for revascularisation

an additional 11 months. Cardiac death, non-fatal 31 (1·8%) 32 (1·9%) 0·98 (0·63 to 1·67) 0·91
myocardial infarction,
Compared with patients with chronic coronary or ischaemic stroke
syndrome, patients with an acute coronary syndrome have
Data are n (%) unless otherwise specified. Rates are number of events occurring between month 1 and month 12 after
increased 1-year rates of cardiac death and myocardial percutaneous coronary intervention (the second randomisation period), with Kaplan–Meier estimated percentages.
infarction arising from both treated culprit lesions and BARC=Bleeding Academic Research Consortium. GUSTO=Global Utilization of Streptokinase and Tissue Plasminogen
untreated non-culprit lesions.27 More than two decades Activator for Occluded Arteries. ISTH=International Society on Thrombosis and Haemostasis. TIMI=Thrombolysis in
Myocardial Infarction flow grading system. *The composite of cardiac death, myocardial infarction, ischaemic stroke,
ago, the CURE trial showed that inhibition of the platelet definite stent thrombosis, and clinically driven target vessel revascularisation. †p<0·0001 for non-inferiority for major
P2Y12 receptor with clopidogrel for 1 year improved adverse cardiovascular or cerebrovascular events. ‡For patients who had revascularisation, oral ticagrelor (90 mg twice
cardiovascular outcomes in patients with an acute coronary daily) plus oral open-labelled aspirin (100 mg once daily) were prescribed after procedure.
syndrome treated with aspirin after percutaneous coronary Table 2: Primary and secondary endpoints
intervention.28 The more potent agents ticagrelor and
prasugrel subsequently replaced clopidogrel as the
standard of care in this setting,29 on the basis of randomised coronary intervention with contemporary drug-eluting
trials showing further reductions in myocardial infarction, stents in acute (and chronic) coronary syndromes, most of
stent thrombosis and, in the case of ticagrelor, mortality, the ischaemic protection from dual antiplatelet therapy
albeit with increased non-surgery-related bleeding.4,5 The derives from the P2Y12 inhibitor, with aspirin responsible
current hypothesis is that, following percutaneous for many of the haemorrhagic complications.5–11

www.thelancet.com Published online April 7, 2024 https://doi.org/10.1016/S0140-6736(24)00473-2 7

 Articles

prasugrel monotherapy; used variable definitions to define

A
100 Ticagrelor plus aspirin bleeding; had the acute coronary syndrome cohort as a
Ticagrelor plus placebo subgroup of a broader population; and included the pre-
10
randomisation period of dual antiplatelet therapy in both
HR 0·45 (95% CI 0·30–0·66); p<0·0001
arms in the primary outcome. The consequent lack of
Clinically relevant bleeding (%)

8 robust data from previous trials might have hindered

widespread adoption of monotherapy with a potent
6 P2Y12 inhibitor after percutaneous coronary intervention
4·6% and only 1 month of dual antiplatelet therapy in high-risk
4 patients with an acute coronary syndrome.
Compared with ticagrelor plus aspirin, continued
2·1%
2 treatment with ticagrelor plus placebo for an additional
11 months resulted in a significant reduction in clinically
0 relevant bleeding events. Major bleeding (BARC types 3
0 60 120 180 240 300 330 or 5) was also reduced with ticagrelor monotherapy, as
Number at risk was bleeding according to the TIMI, GUSTO, and ISTH
(number censored)
Ticagrelor plus aspirin 1700 1681 1664 1652 1634 1622 1615 scales. These results are consistent with previous studies
(0) (1) (3) (4) (7) (9) (11) reporting that discontinuing aspirin while maintaining
Ticagrelor plus placebo 1700 1688 1684 1676 1665 1657 1654
(0) (2) (4) (6) (8) (11) (14)
P2Y12 receptor inhibition after percutaneous coronary
intervention and 1–6 months of dual antiplatelet therapy
B reduces the risk of subsequent major and minor bleeding
100
events in patients with an acute coronary syndrome.7–11,16
10 However, the present study extends these findings by
HR 0·98 (95% CI 0·69–1·39); pnon-inferiority<0·0001; psuperiority=0·89 showing that, in a double-blind, placebo-controlled trial,
Major adverse cardiovascular or

treatment of such patients with ticagrelor monotherapy

cerebrovascular events (%)

8
after 1 month and up to 12 months is safe, with no
6
apparent increased risk of thrombotic or ischaemic events.
Results were similar in the intention-to-treat and per-
protocol populations. Bleeding events were consistently
4 3·7%
less frequent with ticagrelor monotherapy in all examined
3·6% subgroups, and MACCE was not increased in any
2
subgroup, although an interaction was identified
according to age. However, the examined subgroups were
0 not adjusted for multiple comparisons, and these findings
0 60 120 180 240 300 330
Time since randomisation (days) should be viewed accordingly.
Number at risk
(number censored)
The ULTIMATE-DAPT results extend the findings from
Ticagrelor plus aspirin 1700 1693 1684 1669 1659 1648 1636 the unblinded TICO trial, in which ticagrelor alone,
(0) (0) (1) (3) (5) (7) (9) compared with ticagrelor plus aspirin, reduced TIMI
Ticagrelor plus placebo 1700 1690 1684 1673 1664 1652 1640
(0) (0) (0) (1) (2) (4) (6) major or minor bleeding events without an increase
in MACCE between 3 months and 12 months after
Figure 2: Primary efficacy and safety outcomes during follow-up between 1 month and 12 months after percutaneous coronary intervention in patients with an
percutaneous coronary intervention
acute coronary syndrome.10 The results of the present trial
(A) The primary efficacy endpoint of clinically relevant bleeding, defined as BARC types 2, 3, or 5 bleeding, was
assessed in the intention-to-treat population between 1 month and 12 months after percutaneous coronary are also consistent with those from the recent T-PASS trial,
intervention in patients who were event-free after 1 month of ticagrelor and aspirin. (B) The primary safety in which 2850 patients with an acute coronary syndrome
endpoint of major adverse cardiovascular or cerebrovascular events, which comprised cardiac death, myocardial in South Korea were randomly assigned to ticagrelor alone
infarction, ischaemic stroke, definite stent thrombosis, and clinically driven target vessel revascularisation,
was assessed in the intention-to-treat population between 1 month and 12 months after percutaneous coronary
or ticagrelor plus aspirin at a median of 16 days after
intervention in patients who were event-free after 1 month of ticagrelor and aspirin. Patients treated with percutaneous coronary intervention.11 However, like TICO,
ticagrelor monotherapy from month 1 had similar rates of adverse ischaemic events as patients who were T-PASS was open label and not placebo controlled;
maintained on ticagrelor plus aspirin. BARC=Bleeding Academic Research Consortium. HR=hazard ratio. random assignment occurred at the time of discharge,
confounding the results by including similar rates of early
Six previous trials have investigated P2Y12 inhibitor events in both groups before dichotomisation of the
monotherapy after percuta­ neous coronary intervention treatment regimens; and was powered for non-inferiority
and an obligate duration of dual antiplatelet therapy in testing for a primary endpoint of net clinical adverse
acute coronary syndromes.7–11 However, most of these events, rather than separately for safety and efficacy.
studies were open label and not placebo controlled; tested We found a relative risk reduction for the key secondary
P2Y12 inhibitor monotherapy after 1–3 months of dual endpoint of net clinical adverse events with ticagrelor
antiplatelet therapy; did not always test ticagrelor or monotherapy (HR 0·68 [95% CI 0·53–0·88]), indicating

8 www.thelancet.com Published online April 7, 2024 https://doi.org/10.1016/S0140-6736(24)00473-2

 Articles

Number of events/patients Hazard ratio (95% CI) pinteraction

Ticagrelor plus placebo Ticagrelor plus aspirin

Age (years) 0·30

≥65 23/729 40/703 0·55 (0·33–0·93)
<65 12/971 38/997 0·32 (0·17–0·62)
Sex 0·73
Male 24/1264 56/1257 0·42 (0·26–0·68)
Female 11/436 22/443 0·50 (0·24–1·03)
Race 0·69
Chinese 32/1476 74/1519 0·44 (0·29–0·67)
Other 3/224 4/181 0·59 (0·13–2·64)
Diabetes 0·92
Yes 11/540 25/535 0·43 (0·21–0·87)
No 24/1160 53/1165 0·45 (0·28–0·73)
Renal dysfunction 0·82
Yes* 4/119 8/129 0·48 (0·15–1·61)
No 31/1581 70/1571 0·44 (0·29–0·67)
BMI (kg/m2) 0·28
≤25 16/845 43/842 0·36 (0·20–0·64)
>25 19/855 35/858 0·54 (0·31–0·95)
Initial presentation 0·59
Unstable angina 14/668 29/708 0·51 (0·27–0·96)
Non-STEMI 14/545 32/531 0·42 (0·22–0·79)
STEMI 7/487 17/461 0·38 (0·16–0·92)
History of myocardial infarction 0·85
Yes 2/143 8/156 0·33 (0·07–1·59)
No 33/1557 70/1544 0·47 (0·31–0·70)
Biomarker positive at time of procedure 0·60
Yes 21/1032 49/992 0·41 (0·24–0·68)
No 14/668 29/708 0·51 (0·27–0·96)
Stent length (mm) 0·69
≥60 6/258 15/245 0·38 (0·15–0·98)
<60 29/1441 63/1454 0·46 (0·30–0·71)
Multivessel disease 0·32
Yes 10/534 30/545 0·33 (0·16–0·69)
No 25/1166 48/1155 0·52 (0·32–0·84)
IVUS-guided PCI 0·27
Yes 12/856 35/857 0·34 (0·18–0·66)
No 23/844 43/843 0·53 (0·32–0·88)
Overall 35/1700 78/1700 0·45 (0·30–0·66)

0·12 0·25 0·5 1 2 4

Favours placebo Favours aspirin

Figure 3: Clinically relevant bleeding events in prespecified subgroups

IVUS=intravascular ultrasound. PCI=percutaneous coronary intervention. STEMI=ST-segment elevation myocardial infarction. *Defined as an estimated glomerular
filtration rate <60 mL/min per 1·73 m².

improved net clinical outcomes with this regimen. Along intervention.29 Patients who had an ischaemic or bleeding
with previous studies, the evidence from this trial event or were non-adherent with aspirin or ticagrelor for
supports the use of ticagrelor monotherapy in event-free 48 h or more within the first month after percutaneous
patients to safely reduce major and minor bleeding coronary intervention were excluded at 30 days. The
events during follow-up. This regimen should be population was therefore identified as compliant at
particularly considered for patients at high risk of 1 month and likely to tolerate treatment with ticagrelor
haemorrhagic complications (eg, patients with frailty, alone, and few patients required ticagrelor down-titration
those older than 65 years, those with anaemia or previous or conversion to clopidogrel during the 1-year of follow-
bleeding history, and those with chronic kidney disease).30 up. The use of contemporary drug-eluting stents also
Several elements of the trial design contributed to the increased the safety of the ticagrelor monotherapy
safety of the ticagrelor monotherapy regimen in our regimen after 30 days.31,32 As seen in the subgroup
study. Thrombotic events occur most frequently results from the first randomisation in the IVUS-ACS
during the first month after percutaneous coronary trial,17 ischaemic event rates can be further reduced

www.thelancet.com Published online April 7, 2024 https://doi.org/10.1016/S0140-6736(24)00473-2 9

 Articles

Number of events/patients Hazard ratio (95% CI) pinteraction

Ticagrelor plus placebo Ticagrelor plus aspirin

Age (years) 0·016

≥65 39/729 26/703 1·48 (0·90–2·43)
<65 22/971 37/997 0·61 (0·36–1·03)
Sex 0·48
Male 48/1264 46/1257 1·05 (0·70–1·57)
Female 13/436 17/443 0·77 (0·37–1·58)
Race 0·28
Chinese 59/1476 59/1519 1·03 (0·72–1·48)
Other 2/224 4/181 0·38 (0·07–2·06)
Diabetes 0·091
Yes 28/540 20/535 1·43 (0·80–2·53)
No 33/1160 43/1165 0·77 (0·49–1·21)
Renal dysfunction 0·43
Yes* 6/119 4/129 1·64 (0·45–5·95)
No 55/1581 59/1571 0·93 (0·65–1·35)
BMI (kg/m2) 0·073
≤25 32/845 23/842 1·42 (0·83–2·43)
>25 29/855 40/858 0·73 (0·45–1·18)
Initial presentation 0·39
Unstable angina 28/668 29/708 1·03 (0·61–1·74)
Non-STEMI 12/545 23/531 0·51 (0·25–1·03)
STEMI 21/487 11/461 1·76 (0·85–3·68)
History of myocardial infarction 0·16
Yes 9/143 5/156 2·11 (0·70–6·34)
No 52/1557 58/1544 0·89 (0·61–1·30)
Biomarker positive at time of procedure 0·78
Yes 33/1032 34/992 0·93 (0·58–1·50)
No 28/668 29/708 1·03 (0·61–1·74)
Stent length (mm) 0·61
≥60 12/258 10/245 1·21 (0·52–2·81)
<60 49/1441 53/1454 0·94 (0·64–1·39)
Multivessel disease 0·50
Yes 25/534 30/545 0·84 (0·49–1·42)
No 36/1166 33/1155 1·09 (0·68–1·74)
IVUS-guided PCI 0·10
Yes 16/856 25/857 0·65 (0·35–1·22)
No 45/844 38/843 1·20 (0·78–1·85)
Overall 61/1700 63/1700 0·98 (0·69–1·39)

0·12 0·25 0·5 1 2 4

Favours placebo Favours aspirin

Figure 4: Composite major adverse cardiovascular or cerebrovascular events in prespecified subgroups

IVUS=intravascular ultrasound. PCI=percutaneous coronary intervention. STEMI=ST-segment elevation myocardial infarction. *Defined as an estimated glomerular
filtration rate <60 mL/min per 1·73 m².

by the use of intravascular imaging guidance, as associated with subsequent mortality.19,22–24 Second, we
compared with angiography guidance, during the tested non-inferiority for MACCE with an absolute
index percutaneous coronary intervention procedure in margin of 2·5%. Given the lower than anticipated
patients with an acute coronary syndrome. Detailed observed ischaemic event rate in the ticagrelor plus
outcomes from all four groups of the trial will be reported aspirin group (3·7% observed vs 6·2% anticipated), this
subsequently. relative margin for non-inferiority testing is wide.
Our study has some limitations. First, the primary However, considering the large sample size and nearly
efficacy endpoint included minor bleeding (BARC type 2). identical event rates in both groups, the rate of MACCE
However, we also observed lower rates of major bleeding with ticagrelor alone is unlikely to be much higher than
(BARC types 3 or 5; ie, severe or fatal bleeding); with ticagrelor plus aspirin. Third, approximately 40% of
TIMI major or minor bleeding; GUSTO moderate, patients had biomarker-negative unstable angina.
severe, or life-threatening bleeding; and ISTH bleeding However, high-sensitivity troponin assays were not widely
with ticagrelor monotherapy, all of which have been available in China and Pakistan during the enrolment

10 www.thelancet.com Published online April 7, 2024 https://doi.org/10.1016/S0140-6736(24)00473-2

 Articles

period of this trial, and it is likely that many of these patients. GWS and S-LC drafted the manuscript with input from ZG,
patients had NSTEMI. Fourth, this study did not assess XG, and AR. JK provided crucial input into the conduct of the trial and
the draft of the manuscript. J-JZ, BSM, FG, YS, YWa, HZ, FL, HSK,
the safety of ticagrelor monotherapy initiated less than NM, HC, MW, and LC contributed to manuscript revisions. YWe and
1 month after percutaneous coronary intervention. In the FC were the primary biostatisticians. GWS and SLC accessed and
T-PASS trial, very early cessation of aspirin (median verified the data in the study. All authors had unrestricted access to the
16 days, IQR 12–25) followed by ticagrelor monotherapy data, vouch for the accuracy of the data and the fidelity of the trial to the
protocol, and had final responsibility for the decision to submit for
for 1 year after percutaneous coronary intervention publication.
decreased bleeding events without an increase in adverse
Declaration of interests
ischaemic outcomes in patients with an acute coronary GWS reports speaker honoraria from Medtronic, Pulnovo, Infraredx,
syndrome.11 Further placebo-controlled randomised trials Abiomed, Abbott, Amgen, and Boehringer Ingelheim; has served as a
testing very early aspirin cessation (ie, less than 1 month consultant to Daiichi Sankyo, Ablative Solutions, CorFlow, Apollo
after percutaneous coronary intervention) are necessary Therapeutics, Cardiomech, Gore, Robocath, Miracor, Vectorious,
Abiomed, Valfix, TherOx, HeartFlow, Neovasc, Ancora, Elucid Bio,
before any recommendations for a course of dual Occlutech, Impulse Dynamics, Adona Medical, Millennia Biopharma,
antiplatelet therapy of less than 1 month can be made for and Oxitope; has equity or options from Ancora, Cagent, Applied
this population. Fifth, although we saw no interaction Therapeutics, Biostar family of funds, SpectraWave, Orchestra Biomed,
between the geography of the enrolling centres and Aria, Cardiac Success, Valfix, and Xenter; reports research support from
Abbott, Abiomed, Bioventrix, Cardiovascular Systems, Phillips, Biosense-
randomisation on the clinical outcomes, 2995 (88·1%) of Webster, Shockwave, Vascular Dynamics, Pulnovo, and V-wave paid
the 3400 enrolled patients were from China, which can directly to institution; and his daughter is an employee at IQVIA.
reduce the generalisability of the results. Sixth, although S-LC reports speaker honoraria from Microport, Pulnovo, Boston
the results were consistent in patients presenting with International Scientific, Medtronic, Sanofi, and BioMed; grants from the
National Scientific Foundation of China; and is a Fellow at the
unstable angina (1376 [40·5%] of 3400 patients) and Collaborative Innovation Center for Cardiovascular Disease Translational
myocardial infarction (2024 [59·5%] of 3400 patients), Medicine, Nanjing Medical University, Nanjing, China. All other authors
additional trials exclusively enrolling patients with report no competing interests.
myocardial infarction would be warranted to provide Data sharing
confirmatory data of non-inferiority for safety (especially Patient-level data collected for this study will not be made publicly
available. However, the investigators will consider collaboration and data
in STEMI). Seventh, these findings might not apply to
sharing for specific projects; requests should be addressed to
patients with previous coronary artery bypass graft Shao-Liang Chen.
surgery or with other conditions constituting exclusion
Acknowledgments
criteria for this trial. Eighth, BARC type 1 bleeding was This study was funded by the Chinese Society of Cardiology (grant
also included in the net clinical adverse events endpoint. number CSCF 2019-A0003), the National Natural Scientific Foundation
However, very few such minor bleeds were reported of China (grant numbers 91639303, 81770441, and 82121001), and Jiangsu
Provincial & Nanjing Municipal Clinical Trial Project (grant number
(appendix 2 p 21), and we believe this inclusion did not BE 2019615). Stents were supplied by Medtronic and Microport Medical,
have any major influence on the results. Ninth, the and antiplatelet agents were supplied by Yung Shin Pharmaceutical
present results are specific to a ticagrelor monotherapy Industrial and Shenzhen Salubris Pharmaceuticals. The study was led by
regimen and do not apply to treatment with prasugrel or Shao-Liang Chen at the Nanjing Medical University and Nanjing First
Hospital, Nanjing, China.
clopidogrel alone. Finally, this study does not inform the
optimal pharmacotherapy regimen after 12 months. References
1 Vos T, Lim SS, Abbafati C, et al. Global burden of 369 diseases and
In summary, to our knowledge, ULTIMATE-DAPT is injuries in 204 countries and territories, 1990–2019: a systematic
the first large-scale randomised, placebo-controlled, analysis for the Global Burden of Disease Study 2019. Lancet 2020;
396: 1204–22.
double-blinded trial comparing the efficacy and safety of
2 Byrne RA, Rossello X, Coughlan JJ, et al. 2023 ESC Guidelines for
treatment with ticagrelor alone versus ticagrelor plus the management of acute coronary syndromes. Eur Heart J 2023;
aspirin between month 1 and month 12 in patients with 44: 3720–826.
an acute coronary syndrome who remained event-free 3 Amsterdam EA, Wenger NK, Brindis RG, et al. 2014 AHA/ACC
guideline for the management of patients with non–ST-elevation
on dual antiplatelet therapy for 1 month after ACS: executive summary. Circulation 2014; 130: 2354–94.
percutaneous coronary intervention with contemporary 4 Wallentin L, Becker RC, Budaj A, et al. Ticagrelor versus clopidogrel
drug-eluting stents. The present results show that, in in patients with acute coronary syndromes. N Engl J Med 2009;
361: 1045–57.
these patients, treatment with a ticagrelor monotherapy
5 Wiviott SD, Braunwald E, McCabe CH, et al. Prasugrel versus
regimen between month 1 and month 12 after the clopidogrel in patients with acute coronary syndromes.
intervention decreases the risk of clinically relevant N Engl J Med Overseas Ed 2007; 357: 2001–15.
bleeding events while providing similar protection 6 Tomaniak M, Chichareon P, Onuma Y, et al. Benefit and risks of
aspirin in addition to ticagrelor in acute coronary syndromes: a post
from MACCE as compared with ticagrelor plus aspirin. hoc analysis of the randomized GLOBAL LEADERS trial.
Further analyses are warranted to identify which patient JAMA Cardiol 2019; 4: 1092–101.
subgroups, according to varying bleeding and ischaemic 7 Watanabe H, Morimoto T, Natsuaki M, et al. Comparison of
clopidogrel monotherapy after 1 to 2 months of dual antiplatelet
risk, can benefit the most from ticagrelor monotherapy. therapy with 12 months of dual antiplatelet therapy in patients with
Contributors acute coronary syndromes: the STOPDAPT-2 ACS randomized
GWS and S-LC designed the study and analysed the data. ZG, XG, AR, clinical trial. JAMA Cardiol 2022; 7: 407–17.
J-JZ, BSM, FG, YS, YWa, HZ, FL, HSK, NM, HC, MW, and LC enrolled

www.thelancet.com Published online April 7, 2024 https://doi.org/10.1016/S0140-6736(24)00473-2 11

 Articles

8 Song PS, Park YH, Oh JH, et al. P2Y12 inhibitor monotherapy 19 Mehran R, Rao SV, Bhatt DL, et al. Standardized bleeding
versus conventional dual antiplatelet therapy or aspirin definitions for cardiovascular clinical trials: a consensus report
monotherapy in acute coronary syndromes: a pooled analysis of the from the Bleeding Academic Research Consortium. Circulation
SMART-DATE and SMART-CHOICE trials. Am J Cardiol 2021; 2011; 123: 2736–47.
150: 47–54. 20 Kaiser LD. Dynamic randomization and a randomization model for
9 Baber U, Dangas G, Angiolillo DJ, et al. Ticagrelor alone vs clinical trials data. Stat Med 2012; 31: 3858–73.
ticagrelor plus aspirin following percutaneous coronary 21 Pocock SJ, Simon R. Sequential treatment assignment with
intervention in patients with non-ST-segment elevation acute balancing for prognostic factors in the controlled clinical trial.
coronary syndromes: TWILIGHT-ACS. Eur Heart J 2020; Biometrics 1975; 31: 103–15.
41: 3533–45. 22 Bovill EG, Terrin ML, Stump DC, et al. Hemorrhagic events during
10 Kim BK, Hong SJ, Cho YH, et al. Effect of ticagrelor monotherapy therapy with recombinant tissue-type plasminogen activator,
vs ticagrelor with aspirin on major bleeding and cardiovascular heparin, and aspirin for acute myocardial infarction. Results of the
events in patients with acute coronary syndromes: the TICO Thrombolysis in Myocardial Infarction (TIMI), phase II trial.
randomized clinical trial. JAMA 2020; 323: 2407–16. Ann Intern Med 1991; 115: 256–65.
11 Hong SJ, Lee SJ, Suh Y, et al. Stopping aspirin within 1 month after 23 GUSTO investigators. An international randomized trial comparing
stenting for ticagrelor monotherapy in acute coronary syndromes: four thrombolytic strategies for acute myocardial infarction.
the T-PASS randomized noninferiority trial. Circulation 2024; N Engl J Med 1993; 329: 673–82.
149: 562–73. 24 Kaatz S, Ahmad D, Spyropoulos AC, Schulman S. Definition of
12 Hahn JY, Song YB, Oh JH, et al. Effect of P2Y12 inhibitor clinically relevant non-major bleeding in studies of anticoagulants
monotherapy vs dual antiplatelet therapy on cardiovascular events in atrial fibrillation and venous thromboembolic disease in non-
in patients undergoing percutaneous coronary intervention: the surgical patients: communication from the SSC of the ISTH.
SMART-CHOICE randomized clinical trial. JAMA 2019; J Thromb Haemost 2015; 13: 2119–26.
321: 2428–37. 25 US Food and Drug Administration. Non-inferiority clinical trials to
13 Vranckx P, Valgimigli M, Jüni P, et al. Ticagrelor plus aspirin for establish effectiveness—guidance for industry. 2016. https://www.
1 month, followed by ticagrelor monotherapy for 23 months vs fda.gov/media/78504/download (accessed Feb 28, 2024).
aspirin plus clopidogrel or ticagrelor for 12 months, followed by 26 Com-Nougue C, Rodary C, Patte C. How to establish equivalence
aspirin monotherapy for 12 months after implantation of a drug- when data are censored: a randomized trial of treatments for
eluting stent: a multicentre, open-label, randomised superiority B non-Hodgkin lymphoma. Stat Med 1993; 12: 1353–64.
trial. Lancet 2018; 392: 940–49.
27 Chau KH, Kirtane AJ, Easterwood RM, et al. Stent thrombosis risk
14 Watanabe H, Domei T, Morimoto T, et al. Effect of 1-month dual over time on the basis of clinical presentation and platelet reactivity:
antiplatelet therapy followed by clopidogrel vs 12-month dual analysis from ADAPT-DES. JACC Cardiovasc Interv 2021; 14: 417–27.
antiplatelet therapy on cardiovascular and bleeding events in
28 Mehta SR, Yusuf S, Peters RJ, et al. Effects of pretreatment with
patients receiving PCI: the STOPDAPT-2 randomized clinical trial.
clopidogrel and aspirin followed by long-term therapy in patients
JAMA 2019; 321: 2414–27.
undergoing percutaneous coronary intervention: the PCI-CURE
15 Mehran R, Baber U, Sharma SK, et al. Ticagrelor with or without study. Lancet 2001; 358: 527–33.
aspirin in high-risk patients after PCI. N Engl J Med 2019;
29 Angiolillo DJ, Galli M, Collet JP, Kastrati A, O’Donoghue ML.
381: 2032–42.
Antiplatelet therapy after percutaneous coronary intervention.
16 Kuno T, Watanabe A, Shoji S, et al. Short-term DAPT and DAPT EuroIntervention 2022; 17: e1371–96.
de-escalation strategies for patients with acute coronary syndromes:
30 Costa F, van Klaveren D, James S, et al. Derivation and validation of
a systematic review and network meta-analysis.
the predicting bleeding complications in patients undergoing stent
Circ Cardiovasc Interv 2023; 16: e013242.
implantation and subsequent dual antiplatelet therapy
17 Li X, Ge Z, Kan J, et al. Intravascular ultrasound-guided versus (PRECISE-DAPT) score: a pooled analysis of individual-patient
angiography-guided percutaneous coronary intervention in acute datasets from clinical trials. Lancet 2017; 389: 1025–34.
coronary syndromes (IVUS-ACS): a two-stage, multicentre,
31 Stone GW, Rizvi A, Newman W, et al. Everolimus-eluting versus
randomised trial. Lancet 2024; published online April 8. https://doi.
paclitaxel-eluting stents in coronary artery disease. N Engl J Med
org/10.1016/S0140-6736(24)00282-4.
2010; 362: 1663–74.
18 Ge Z, Gao XF, Kan J, et al. Comparison of one-month versus
32 Kandzari DE, Mauri L, Koolen JJ, et al. Ultrathin, bioresorbable
twelve-month dual antiplatelet therapy after implantation of drug-
polymer sirolimus eluting stents versus thin, durable polymer
eluting stents guided by either intravascular ultrasound or
everolimus-eluting stents in patients undergoing coronary
angiography in patients with acute coronary syndrome: rationale
revascularization (BIOFLOW V): a randomized trial. Lancet 2017;
and design of prospective, multicenter, randomized, controlled
390: 1843–52.
IVUS-ACS and ULTIMATE-DAPT trial. Am Heart J 2021;
236: 49–58.

12 www.thelancet.com Published online April 7, 2024 https://doi.org/10.1016/S0140-6736(24)00473-2