JACC: CARDIOVASCULAR INTERVENTIONS VOL. 14, NO.

13, 2021

ª 2021 BY THE AMERICAN COLLEGE OF CARDIOLOGY FOUNDATION

PUBLISHED BY ELSEVIER

Impact of Age on the Safety and Efficacy

of Ticagrelor Monotherapy in Patients
Undergoing PCI
Dominick J. Angiolillo, MD, PHD,a Davide Cao, MD,b Usman Baber, MD, MS,c Samantha Sartori, PHD,b
Zhongjie Zhang, MPH,b George Dangas, MD, PHD,b Shamir Mehta, MD, MSC,d Carlo Briguori, MD, PHD,e
David J. Cohen, MD, MSC,f Timothy Collier, MSC,g Dariusz Dudek, MD, PHD,h Javier Escaned, MD, PHD,i
C. Michael Gibson, MD, MS,j Robert Gil, MD, PHD,k Kurt Huber, MD,l Upendra Kaul, MD,m Ran Kornowski, MD,n
Mitchell W. Krucoff, MD,o Vijay Kunadian, MB, BS, MD,p David J. Moliterno, MD,q E. Magnus Ohman, MD,o
Keith Oldroyd, MB, CHB, MD,r Gennaro Sardella, MD,s Samin K. Sharma, MD,b Richard Shlofmitz, MD,t
Giora Weisz, MD,u Bernhard Witzenbichler, MD,v Stuart Pocock, PHD,g Roxana Mehran, MDb

ABSTRACT

OBJECTIVES The aim of this study was to assess the impact of age on the safety and efficacy of ticagrelor monotherapy
after percutaneous coronary intervention (PCI).

BACKGROUND As the risk for bleeding and ischemic complications after PCI increases with age, the authors conducted
a pre-specified analysis of the TWILIGHT (Ticagrelor With Aspirin or Alone in High-Risk Patients After Coronary Inter-
vention) trial to evaluate the possible benefits of ticagrelor monotherapy according to age.

METHODS The TWILIGHT trial enrolled patients undergoing PCI with drug-eluting stents who fulfilled at least 1 clinical
and 1 angiographic high-risk criterion. Age $65 years was a clinical entry criterion. After 3 months of dual-antiplatelet
therapy with ticagrelor, event-free patients were randomized to ticagrelor plus placebo or ticagrelor plus aspirin for an
additional 12 months. The primary endpoint was Bleeding Academic Research Consortium type 2, 3, or 5 bleeding. The key
secondary endpoint was the composite of all-cause death, myocardial infarction, or stroke.

RESULTS A total of 3,113 patients (47.7%) were $65 years of age. At 1 year after randomization, ticagrelor mono-
therapy significantly reduced BARC type 2, 3, or 5 bleeding (4.5% vs. 8.2%; hazard ratio: 0.53; 95% confidence interval:
0.40 to 0.71) without increasing ischemic events (4.2% vs. 4.4%; hazard ratio: 0.96; 95% confidence interval: 0.68 to
1.35) compared with ticagrelor plus aspirin among patients $65 years of age. These findings were consistent in
patients <65 years of age with respect to the primary (pinteraction ¼ 0.62) and key secondary (pinteraction ¼ 0.77) endpoints
and across different age categories.

CONCLUSIONS A strategy of ticagrelor monotherapy following 3 months of dual-antiplatelet therapy significantly

reduced clinically relevant bleeding compared with ticagrelor plus aspirin without an increase in ischemic events, irre-
spective of age. (J Am Coll Cardiol Intv 2021;14:1434–46) © 2021 by the American College of Cardiology Foundation.

From the aUniversity of Florida College of Medicine, Jacksonville, Florida, USA; bThe Zena and Michael A. Wiener Cardiovascular
Institute, Mount Sinai Hospital, New York, New York, USA; cThe University of Oklahoma Health Sciences Center, Oklahoma City,
Oklahoma, USA; dHamilton Health Sciences, Hamilton, Ontario, Canada; eMediterranea Cardiocentro, Naples, Italy; fUniversity of
Missouri–Kansas City, Kansas City, Missouri, USA; gLondon School of Hygiene and Tropical Medicine, London, United Kingdom;
h
Jagiellonian University Medical College, Krakow, Poland; iInstituto de Investigacion Sanitaria del Hospital Clinico San Carlos and
Complutense University, Madrid, Spain; jBeth Israel Deaconess Medical Center, Boston, Massachusetts, USA; kCenter of Post-
graduate Medical Education, Central Clinical Hospital of the Ministry of Interior and Administration, Warsaw, Poland; lWilhel-
minenspital, Wien, Austria; mBatra Hospital and Medical Research Centre, New Delhi, India; nRabin Medical Center, Petach Tikva,
Israel; oDuke University Medical Center–Duke Clinical Research Institute, Durham, North Carolina, USA; pTranslational and
Clinical Research Institute, Faculty of Medical Sciences, Newcastle University and Freeman Hospital, Newcastle upon Tyne
Hospitals NHS Foundation Trust, Newcastle upon Tyne, United Kingdom; qUniversity of Kentucky, Lexington, Kentucky, USA;
r
The West of Scotland Heart and Lung Centre, Golden Jubilee National Hospital, Clydebank, United Kingdom; sPoliclinico Umberto
I University, Rome, Italy; tSt. Francis Hospital, Roslyn, Roslyn, New York, USA; uNewYork-Presbyterian Hospital, Columbia
University Medical Center, New York, USA; and vHelios Amper-Klinikum, Dachau, Germany.
Eric Bates, MD, served as the Guest Editor for this paper.

ISSN 1936-8798/$36.00 https://doi.org/10.1016/j.jcin.2021.04.043

JACC: CARDIOVASCULAR INTERVENTIONS VOL. 14, NO. 13, 2021 Angiolillo et al. 1435
JULY 12, 2021:1434–46 Age Substudy of the TWILIGHT Trial

D ual-antiplatelet therapy (DAPT) with independent data and safety monitoring ABBREVIATIONS

aspirin and a P2Y 12 inhibitor is the standard board provided external oversight to ensure AND ACRONYMS

of care for the prevention of ischemic com- safety of trial participants.

ACS = acute coronary
plications, including stent thrombosis, in patients un- syndrome
STUDY POPULATION. Patients undergoing
dergoing percutaneous coronary intervention (PCI)
successful PCI with at least 1 commercially ARC-HBR = Academic
(1,2). However, such ischemic benefit occurs at the Research Consortium for High
available drug-eluting stent whom the treat-
expense of increased bleeding, which negatively af- Bleeding Risk
ing clinician intended to discharge on tica-
fects prognosis (1–4). Importantly, the risk for both BARC = Bleeding Academic
grelor plus aspirin were eligible to Research Consortium
ischemic and bleeding events post-PCI increase with
participate. Trial inclusion required the CI = confidence interval
age, underscoring the need to identify antiplatelet
presence of at least 1 clinical and 1 angio-
treatment regimens that reduce bleeding without DAPT = dual-antiplatelet
graphic feature associated with a high risk for therapy
any trade-off in antithrombotic efficacy (5,6). A strat-
ischemic or bleeding events (8,10). Age $65 HBR = high bleeding risk
egy of P2Y12 inhibitor monotherapy, after a brief
years represented a clinical study entry cri- HR = hazard ratio
period of DAPT, has recently been proposed for this
terion; other clinical criteria included female
purpose (7). In particular, the TWILIGHT (Ticagrelor MI = myocardial infarction
sex, troponin-positive acute coronary syn-
With Aspirin or Alone in High-Risk Patients After Cor- PCI = percutaneous coronary
drome (ACS), atherosclerotic vascular disease intervention
onary Intervention) trial showed that among high-
(prior myocardial infarction [MI], coronary
risk PCI patients, discontinuation of aspirin after
revascularization, or peripheral arterial disease),
3 months of DAPT while maintaining P2Y 12 inhibitor
diabetes mellitus requiring medication, and chronic
monotherapy with ticagrelor reduced bleeding
kidney disease (estimated glomerular filtration
without increasing ischemic harm (8). The demo-
rate <60 ml/min/1.73 m2 or creatinine
graphic trends of increased life expectancy have
clearance <60 ml/min). Angiographic criteria
raised interest in the safety and efficacy of antiplate-
included multivessel coronary artery disease, total
let regimens in the elderly population undergoing PCI
stent length >30 mm, thrombotic target lesion,
(5,6,9). Therefore, we conducted a pre-specified anal-
bifurcation lesion requiring 2 stents, obstructive left
ysis of the TWILIGHT trial to assess the impact of age
main or proximal left anterior descending coronary
on the effects of ticagrelor monotherapy versus tica-
artery lesion, and calcified target lesion requiring
grelor plus aspirin in high-risk patients undergoing
debulking devices. Key exclusion criteria included
PCI.
presentation with an ST-segment elevation MI,
SEE PAGE 1447
cardiogenic shock, prior stroke, need for oral anti-
METHODS coagulation, or contraindication to aspirin or tica-
grelor. Furthermore, 587 patients were excluded from
TRIAL DESIGN AND OVERSIGHT. TWILIGHT was a the present analysis because of lack of approval to
randomized, placebo-controlled trial conducted at share the data from national legal and regulato-
187 sites in 11 countries. The trial rationale, design, ry authorities.
and principal results have been reported previously All enrolled patients received open-label ticagrelor
(10). TWILIGHT was an investigator-initiated trial (90 mg twice daily) and enteric-coated aspirin (81 to
designed, coordinated, and sponsored by the Icahn 100 mg/day) after index PCI. At 3 months, patients
School of Medicine at Mount Sinai. AstraZeneca pro- without major bleeding or ischemic events were
vided an investigator-initiated grant and supplied randomized 1:1 in a double-blind fashion to aspirin or
ticagrelor for the trial but had no role in the design, matching placebo for an additional 12 months as an
collection, analysis, or interpretation of the data. The adjunct to open-label ticagrelor (8,10). Patients
executive and steering committees were responsible experiencing Bleeding Academic Research Con-
for trial conduct, integrity of data analysis, and sortium (BARC) type 3b or higher bleeding events or
reporting of results. National regulatory agencies and ischemic events (stroke, MI, or coronary revasculari-
Institutional Review Boards or ethics committees of zation) between index PCI and 3 months were not
participating centers approved the trial protocol. An eligible for randomization (11). Moreover, patients

The authors attest they are in compliance with human studies committees and animal welfare regulations of the authors’
institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information,
visit the Author Center.

Manuscript received February 23, 2021; revised manuscript received April 13, 2021, accepted April 19, 2021.
1436 Angiolillo et al. JACC: CARDIOVASCULAR INTERVENTIONS VOL. 14, NO. 13, 2021

Age Substudy of the TWILIGHT Trial JULY 12, 2021:1434–46

T A B L E 1 Baseline Clinical Characteristics

Age $65 yrs (n ¼ 3,113) Age <65 yrs (n ¼ 3,419)

Ticagrelor þ Placebo Ticagrelor þ Aspirin Ticagrelor þ Placebo Ticagrelor þ Aspirin

(n ¼ 1,567 [50.3%]) (n ¼ 1,546 [49.7%]) p Value (n ¼ 1,698 [49.7%]) (n ¼ 1,721 [50.3%]) p Value

Age, yrs 71.9  5.5 72.1  5.4 0.521 55.7  6.5 55.7  6.7 0.871
Female 404 (25.8) 400 (25.9) 0.954 367 (21.6) 363 (21.1) 0.710
Non-White race 288 (18.4) 271 (17.5) 0.537 532 (31.3) 518 (30.1) 0.435
BMI, kg/m2 28.3  5.3 28.3  5.2 0.955 29.4  5.8 29.4  5.9 0.886
Enrolling region 0.997 0.859
North America 743 (47.4) 733 (47.4) 741 (43.6) 755 (43.9)
Europe 649 (41.4) 639 (41.3) 602 (35.5) 619 (36.0)
Asia 175 (11.2) 174 (11.3) 355 (20.9) 347 (20.2)
Diabetes 567 (36.2) 544 (35.2) 0.562 646 (38.0) 648 (37.7) 0.813
Diabetes treated with insulin 132 (23.3) 148 (27.2) 0.132 176 (27.2) 194 (29.9) 0.284
Chronic kidney disease 410 (27.4) 426 (28.7) 0.423 136 (8.3) 121 (7.3) 0.273
Anemia 368 (24.6) 366 (24.7) 0.967 240 (14.7) 233 (14.1) 0.624
Current smoking 161 (10.3) 185 (12.0) 0.131 495 (29.2) 559 (32.5) 0.036
Hypercholesterolemia 1,095 (69.9) 1,096 (70.9) 0.536 1,031 (60.7) 1,017 (59.1) 0.332
Hypertension 1,231 (78.6) 1,233 (79.8) 0.411 1,171 (69.0) 1,165 (67.7) 0.439
Peripheral arterial disease 135 (8.6) 133 (8.6) 0.990 103 (6.1) 107 (6.2) 0.854
Previous MI 400 (25.5) 411 (26.6) 0.501 572 (33.7) 554 (32.2) 0.352
Previous PCI 662 (42.2) 696 (45.0) 0.119 760 (44.8) 720 (41.8) 0.085
Previous CABG 223 (14.2) 213 (13.8) 0.710 136 (8.0) 131 (7.6) 0.665
Multivessel CAD 1,040 (66.4) 958 (62.0) 0.010 1,044 (61.5) 1,027 (59.7) 0.279
Previous major bleeding 14 (0.9) 20 (1.3) 0.283 13 (0.8) 7 (0.4) 0.169
Indication for PCI 0.156 0.709
Stable CAD 704 (45.0) 656 (42.4) 527 (31.0) 524 (30.4)
ACS 862 (55.0) 890 (57.6) 1,171 (69.0) 1,197 (69.6)

Values are mean  SD or n (%).

ACS ¼ acute coronary syndrome; BMI ¼ body mass index; CABG ¼ coronary artery bypass graft; CAD ¼ coronary artery disease; MI ¼ myocardial infarction;
PCI ¼ percutaneous coronary intervention.

were ineligible for randomization if they were non- moderate, severe, or life-threatening bleeding; or
adherent to ticagrelor or aspirin. Randomization was major bleeding as defined by the International Society
performed using a secure web-based system; an in- on Thrombosis and Haemostasis (8,10). Other sec-
dependent statistician not involved with the trial ondary endpoints included cardiovascular death,
generated the allocation sequence, which was strati- nonfatal MI, ischemic stroke, and definite or probable
fied by site with randomly varying block sizes of 4, 6, stent thrombosis. MI was defined according to the
or 8. Follow-up occurred 1 month after randomization third universal definition, and stent thrombosis was
via telephone and in person at 6 and 12 months after classified according to the Academic Research Con-
randomization. After 12 months of protocol- sortium (8,10). All clinical events were adjudicated by
mandated therapy, patients were switched to a an independent committee, blinded to treat-
standard-of-care antiplatelet regimen at the discre- ment assignment.
tion of their treating physicians, followed by final
STATISTICAL ANALYSES. For the purpose of the
telephone follow-up 3 months later.
present analysis, the primary bleeding and secondary
OUTCOMES. The primary endpoint was the compos- ischemic endpoints were evaluated according to pa-
ite of BARC type 2, 3, or 5 bleeding up to 1 year after tient age. In particular, patients $65 years of age, a
randomization (8,10). The key secondary endpoint key clinical entry criterion, were compared with
was the composite of all-cause death, MI, or stroke. those <65 years of age. To evaluate the study treat-
Secondary bleeding endpoints included BARC type 3 ment effects across different age categories, we
or 5 bleeding; TIMI (Thrombolysis In Myocardial further stratified the overall population into 10-year
Infarction) major or minor bleeding; GUSTO (Global intervals of age (i.e., <55, 55 to 64, 65 to 74,
Use of Strategies to Open Occluded Arteries) and $75 years). Moreover, in line with the consensus
JACC: CARDIOVASCULAR INTERVENTIONS VOL. 14, NO. 13, 2021 Angiolillo et al. 1437
JULY 12, 2021:1434–46 Age Substudy of the TWILIGHT Trial

T A B L E 2 Baseline Procedural Characteristics

Age $65 yrs (n ¼ 3,113) Age <65 yrs (n ¼ 3,419)

Ticagrelor þ Placebo Ticagrelor þ Aspirin Ticagrelor þ Placebo Ticagrelor þ Aspirin

(n ¼ 1,567 [50.3%]) (n ¼ 1,546 [49.7%]) p Value (n ¼ 1,698 [49.7%]) (n ¼ 1,721 [50.3%]) p Value

Radial artery access 1,068 (68.2) 1,038 (67.1) 0.545 1,255 (73.9) 1,267 (73.6) 0.847
Multivessel CAD 1,040 (66.4) 958 (62.0) 0.010 1,044 (61.5) 1,027 (59.7) 0.279
Target vessel
Left main coronary artery 77 (4.9) 79 (5.1) 0.802 60 (3.5) 73 (4.2) 0.284
LAD 880 (56.2) 869 (56.2) 0.977 939 (55.3) 951 (55.3) 0.980
LCx 530 (33.8) 489 (31.6) 0.192 524 (30.9) 556 (32.3) 0.363
RCA 520 (33.2) 536 (34.7) 0.381 620 (36.5) 621 (36.1) 0.794
Number of vessels treated 1.3  0.5 1.3  0.5 0.830 1.3  0.5 1.3  0.5 0.341
Number of lesions treated 1.5  0.8 1.5  0.8 0.714 1.5  0.7 1.5  0.7 0.858
Lesion morphology*
Moderate/severe calcification 277 (17.7) 273 (17.7) 0.989 200 (11.8) 190 (11.0) 0.497
Bifurcation 207 (13.2) 173 (11.2) 0.085 195 (11.5) 215 (12.5) 0.364
Total occlusion 80 (5.1) 85 (5.5) 0.625 105 (6.2) 97 (5.6) 0.497
Thrombotic 136 (8.7) 129 (8.3) 0.738 217 (12.8) 242 (14.1) 0.272
Total stent length, mm† 38.6  23.6 38.4  22.6 0.824 39.5  23.2 39.1  24.4 0.628
Minimum stent diameter, mm 2.8  0.5 2.9  0.5 0.548 2.8  0.5 2.9  0.5 0.448
Complex PCI‡ 540 (34.5) 532 (34.4) 0.977 490 (28.9) 510 (29.6) 0.618

Values are n (%) or mean  SD. *Lesion morphology assessed by operators. †Stent length calculated by operators. ‡Complex PCI was defined as any of the following: 3 vessels
treated, $3 lesions treated, total stent length >60 mm, bifurcation with 2 stents implanted, atherectomy device use, left main PCI, surgical bypass graft, or chronic total
occlusion as target lesion.
LAD ¼ left anterior descending coronary artery; LCx ¼ left circumflex coronary artery; RCA ¼ right coronary artery; other abbreviations as in Table 1.

definitions from the Academic Research Consortium major protocol deviations) (8,10). Treatment effects
for High Bleeding Risk (ARC-HBR), exploratory ana- were estimated according to patient age as defined
lyses were performed among patients $75 years of earlier with formal interaction testing to assess for
age. According to the ARC-HBR, patients are consid- effect modification. Finally, the association between
ered to be at high bleeding risk (HBR) if at least 1 age (as a continuous variable) and the 1-year Kaplan-
major or 2 minor criteria are met (12). Given that Meier rates of ischemic and bleeding events was also
age $75 years is considered a minor criterion, we evaluated fitting a smoothing spline curve with 4
evaluated treatment effects among HBR and non-HBR knots whose locations were placed at equally spaced
patients within this age category. percentiles across the range of age. All analyses were
Clinical and procedural characteristics are sum- performed using Stata version 16.0 (StataCorp, Col-
marized by randomized group as mean  SD and lege Station, Texas).
frequencies for continuous and categorical variables,
respectively. The cumulative incidence of both pri- RESULTS
mary and secondary endpoints was estimated using
the Kaplan-Meier method. Patients without primary PATIENT CHARACTERISTICS. Among the 6,532 pa-
endpoints between randomization and 1 year were tients randomized in the main TWILIGHT trial and
censored at the time of death, last known contact, or available for analysis, 47.7% (n ¼ 3,113) were $65
365 days, whichever came first. Hazard ratios (HRs) years of age. Of these, 50.3% were randomized to
and 95% confidence intervals (CIs) were generated ticagrelor plus placebo and 49.7% to ticagrelor plus
using unadjusted Cox proportional hazards models. aspirin. Patients $65 years of age were more likely to
Analyses of bleeding were performed using the be women, from North America or Europe, and of
intention-to-treat cohort (i.e., all participants who White race than patients <65 years of age. They had
were successfully randomized at the 3-month visit), more cardiovascular risk factors and comorbidities
while ischemic outcomes were analyzed using the per but presented less often with ACS as the indication
protocol cohort (i.e., randomized participants who for PCI (Supplemental Table 1). With respect
completed all study-related contacts without any to angiographic and procedural characteristics,
1438 Angiolillo et al. JACC: CARDIOVASCULAR INTERVENTIONS VOL. 14, NO. 13, 2021

Age Substudy of the TWILIGHT Trial JULY 12, 2021:1434–46

C E N T R A L I LL U ST R A T I ON Ticagrelor With or Without Aspirin After PCI in Patients Aged 65 Years or Older

Prespecified subgroup analysis from the multicenter, double-blind, randomized TWILIGHT trial

3,113 Patients ≥65 years Ticagrelor Ticagrelor

(90 mg twice daily) (90 mg twice daily)
Randomized at 3 months after
+ +
PCI with a DES if free from major
Placebo Aspirin
ischemic and bleeding events
(81-100 mg daily)

(N = 1,567) (N = 1,546)

BARC 2, 3, or 5 bleeding 4.5% 8.2%

At 12 months after randomization
HR: 0.53, 95% CI: 0.40-0.71; pinteraction = 0.62

Death, MI, or stroke 4.2% 4.4%

At 12 months after randomization
HR: 0.96, 95% CI: 0.68-1.35; pinteraction = 0.77

Subgroup analyses by 10-year age intervals showed consistent risk reduction in BARC 2, 3, or 5
bleeding (pinteraction = 0.09) and death, MI, or stroke (pinteraction = 0.33) across all age categories

Angiolillo, D.J. et al. J Am Coll Cardiol Intv. 2021;14(13):1434–46.

Among patients 65 years of age or older who were free from major ischemic and bleeding events and adherent to dual-antiplatelet therapy with ticagrelor for
3 months after percutaneous coronary intervention with drug-eluting stents, discontinuation of aspirin and continuation of ticagrelor monotherapy, compared
with ticagrelor plus aspirin for an additional 12 months, reduced the incidence of clinically relevant Bleeding Academic Research Consortium (BARC) type 2, 3, or 5
bleeding by 47%, without increasing the rate of all-cause death, myocardial infarction, or stroke. There were no significant differences between ticagrelor
monotherapy and ticagrelor plus aspirin with respect to the primary bleeding and key secondary ischemic endpoints across all age subgroups. CI ¼ confidence
interval; DES ¼ drug-eluting stent(s); HR ¼ hazard ratio; MI ¼ myocardial infarction; PCI ¼ percutaneous coronary intervention.

patients $65 years of age were less likely to undergo difference with respect to study drug (15.1% vs.
PCI via radial access and were more likely to have 15.7%; p ¼ 0.63).
multivessel disease and calcific lesions
(Supplemental Table 2). Tables 1 and 2 show the de- BLEEDING EVENTS. Bleeding events increased with
mographic, clinical, and procedural characteristics, age (Figure 1A). As shown in Figure 2, in the cohort of
which were well balanced between treatment arms, patients $65 years of age, the primary outcome of
except for a higher prevalence of multivessel disease BARC type 2, 3, or 5 bleeding occurred in 69 patients
(66.4% vs. 62.0%; p ¼ 0.010) in the ticagrelor-plus- (4.5%) randomized to ticagrelor plus placebo versus
placebo group among patients $65 years of age. 125 patients (8.2%) randomized to ticagrelor plus
Rates of permanent ticagrelor discontinuation at 1 aspirin (HR: 0.53; 95% CI: 0.40 to 0.71; p < 0.001).
year were similar among those $65 years of age One-year BARC type 3 or 5 bleeding rates were 1.3%
randomized to ticagrelor plus placebo versus tica- and 2.4%, respectively (HR: 0.55; 95% CI 0.32 to 0.94;
grelor plus aspirin (16.7% vs. 16.8%; p ¼ 0.91). Anal- p ¼ 0.030). This treatment effect was consistent
ogous results for blinded study drug discontinuation across different bleeding scales, including TIMI,
were 20.7% and 21.3%; respectively (p ¼ 0.63). GUSTO, and International Society on Thrombosis and
Among patients <65 years of age, ticagrelor discon- Haemostasis (Figure 3). There was no significant
tinuation rates were lower in the placebo group interaction between age and treatment group with
(10.2% vs. 12.6%; p ¼ 0.024), while there was no respect to the bleeding endpoints.
JACC: CARDIOVASCULAR INTERVENTIONS VOL. 14, NO. 13, 2021 Angiolillo et al. 1439
JULY 12, 2021:1434–46 Age Substudy of the TWILIGHT Trial

F I G U R E 1 Rates of Bleeding and Ischemic Events According to Age

A smoothing spline curve was used to plot the rates of Bleeding Academic Research Consortium (BARC) type 2, 3, or 5 bleeding (A) and all-
cause death, myocardial infarction (MI), or stroke (B) at 1 year after randomization in the overall trial population. The colored lines indicate
Kaplan-Meier estimated event rates, while the shaded areas indicate 95% confidence intervals.

ISCHEMIC EVENTS. Ischemic events increased with 67 patients (4.4%) randomized to ticagrelor plus
age (Figure 1B). As shown in Figure 4, in the cohort of aspirin (HR: 0.96; 95% CI: 0.68 to 1.35; p ¼ 0.798).
patients $65 years of age, the composite outcome of Rates of all-cause death (1.3% vs. 1.9%), MI (2.8% vs.
all-cause death, MI, or stroke occurred in 65 patients 2.8%), ischemic stroke (0.5% vs. 0.2%), and definite or
(4.2%) randomized to ticagrelor plus placebo versus probable stent thrombosis (0.3% vs. 0.4%) were
1440 Angiolillo et al. JACC: CARDIOVASCULAR INTERVENTIONS VOL. 14, NO. 13, 2021

Age Substudy of the TWILIGHT Trial JULY 12, 2021:1434–46

F I G U R E 2 Rates of BARC Type 2, 3, or 5 Bleeding at 1 Year After Randomization

Kaplan-Meier estimates of Bleeding Academic Research Consortium (BARC) type 2, 3, or 5 bleeding with ticagrelor (tica) plus placebo versus
ticagrelor plus aspirin in relation to age cutoff of 65 years in the intention-to-treat cohort. CI ¼ confidence interval; HR ¼ hazard ratio.

similar between treatment groups (p > 0.10 for all) monotherapy was significantly greater among
(Figure 5). There was no significant interaction be- patients $75 years of age at HBR (5.8% vs. 12.0%; HR:
tween age and treatment group with respect to the 0.46; 95% CI: 0.25 to 0.85; p ¼ 0.014) compared with
ischemic endpoints. those without HBR status (8.6% vs. 6.9%; HR: 1.25;
95% CI: 0.60 to 2.62; p ¼ 0.555; p interaction ¼ 0.042)
ADDITIONAL ANALYSES. The subgroup analysis by
(Supplemental Table 5). The outcome of all-cause
10-year age intervals showed a consistent risk
death, MI, or stroke was similar in patients $75
reduction in BARC type 2, 3, or 5 bleeding across all
years of age randomized to ticagrelor plus placebo
age categories (p interaction ¼ 0.090) (Figure 6). Simi-
versus ticagrelor plus aspirin, regardless of the pres-
larly, there were no differences between ticagrelor
ence of HBR (p interaction ¼ 0.474) (Supplemental
monotherapy and ticagrelor plus aspirin with respect
Table 6).
to the key secondary endpoint of death, MI, or stroke
across all age strata (p interaction ¼ 0.333).
Of the 905 patients who were $75 years of age DISCUSSION
(Supplemental Tables 3 and 4), 58.6% (n ¼ 530) met
the Academic Research Consortium definition of HBR. The key findings of our pre-specified analysis evalu-
Patients at HBR $75 years of age had the highest rates ating the impact of age on the safety and efficacy
of both bleeding and ischemic events at 1 year after outcomes of patients randomized in the TWILIGHT
randomization (Supplemental Figure 1) compared trial include the following: 1) the rate of adverse
with all other age groups. The risk reduction in BARC events increases with age, with a sharp rise after the
type 2, 3, or 5 bleeding associated with ticagrelor age of 65 years for BARC type 2, 3, or 5 bleeding and
JACC: CARDIOVASCULAR INTERVENTIONS VOL. 14, NO. 13, 2021 Angiolillo et al. 1441
JULY 12, 2021:1434–46 Age Substudy of the TWILIGHT Trial

F I G U R E 3 Risk for Bleeding Events at 1 Year After Randomization

Forest plots showing the effect of ticagrelor plus placebo versus ticagrelor plus aspirin on the bleeding endpoints in relation to age cutoff of 65 years. Event rates at 1
year were estimated using the Kaplan-Meier method. HR and 95% CIs with interaction p values generated using Cox regression. Bleeding outcomes were performed in
the intention-to-treat cohort. GUSTO ¼ Global Use of Strategies to Open Occluded Arteries; ISTH ¼ International Society on Thrombosis and Haemostasis,
TIMI ¼ Thrombolysis In Myocardial Infarction; other abbreviations as in Figure 2.

after 70 years of age for death, MI, or stroke; 2) tica- inhibitor monotherapy, only GLOBAL LEADERS (A
grelor monotherapy, compared with ticagrelor plus Clinical Study Comparing Two Forms of Anti-Platelet
aspirin, reduced the incidence of clinically relevant Therapy After Stent Implantation) had a number of
BARC type 2, 3, or 5 bleeding as well as major BARC elderly patients, defined as $75 years of age, higher
type 3 or 5 bleeding over 1 year of follow-up by almost than TWILIGHT. In the pre-specified analysis of
50% in patients $65 years of age; 3) ticagrelor mon- elderly patients from GLOBAL LEADERS (n ¼ 2,565),
otherapy was not associated with significant differ- there was no differential treatment effect of ticagrelor
ences in the rate of all-cause death, MI, or stroke in monotherapy (after 1 month of DAPT) found in an all-
patients $65 years of age; and 4) the treatment effect comers population undergoing PCI with respect to the
of ticagrelor monotherapy on ischemic and bleeding primary endpoint of all-cause mortality or new Q-
outcomes were preserved irrespective of age (Central wave MI and BARC type 3 or 5 bleeding at 2 years (17).
Illustration). These findings, however, need to be interpreted in
A number of trials have investigated the safety and the context of a trial that failed to meet its primary
efficacy of P2Y12 inhibitor monotherapy after a mini- endpoint (13).
mal duration (1 to 3 months) of DAPT following PCI Prior studies on the potent P2Y12 inhibitors prasu-
(13–16). However, TWILIGHT was the only trial to be grel and ticagrelor compared with clopidogrel in pa-
placebo controlled and enroll patients with both tients with ACS undergoing PCI showed reduced
clinical and angiographic features associated with ischemic events at the expense of increased bleeding
increased risk for ischemic or bleeding complications (18). In particular, in the TRITON-TIMI 38 (Trial to
post-PCI (10). Age $65 years represented a key clin- Assess Improvement in Therapeutic Outcomes by
ical entry criterion in TWILIGHT, thereby making this Optimizing Platelet Inhibition With Prasugrel–
important cohort of patients a large subset (47.7% Thrombolysis In Myocardial Infarction 38) trial, such
[n ¼ 3,113]) of the trial population. Of note, 13.9% of excess in bleeding, including fatal bleeding, resulted
the trial population (n ¼ 905) were $75 years of age. in a neutral net clinical benefit in the elderly sub-
Patients were also required to have high-risk angio- group (19). On the basis of these findings, prasugrel is
graphic features to be enrolled in the study, and generally not recommended in patients $75 years of
collectively this enabled us to study a patient popu- age. The increased risk for bleeding among the elderly
lation enriched with ischemic and bleeding compli- can be attributed to increased active metabolite levels
cations. Of the available trials assessing P2Y 12 with prasugrel 10 mg, suggesting the need to reduce
1442 Angiolillo et al. JACC: CARDIOVASCULAR INTERVENTIONS VOL. 14, NO. 13, 2021

Age Substudy of the TWILIGHT Trial JULY 12, 2021:1434–46

F I G U R E 4 Rates of Death, MI, or Stroke at 1 Year After Randomization

Kaplan-Meier estimates of all-cause death, myocardial infarction, or stroke with ticagrelor plus placebo versus ticagrelor plus aspirin in
relation to age cutoff of 65 years in the per protocol cohort. Abbreviations as in Figures 1 and 2.

the maintenance dose to 5 mg (20). Although prasu- more contemporary POPular AGE (Ticagrelor or Pra-
grel 5 mg provides more potent platelet inhibition sugrel Versus Clopidogrel in Elderly Patients With an
compared with clopidogrel among elderly patients, Acute Coronary Syndrome and a High Bleeding Risk:
the differences are small and have not been shown to Optimization of Antiplatelet Treatment in High-risk
translate into clinical benefits (21–25). However, in Elderly) trial, clopidogrel significantly reduced net
the subgroup of patients $75 years of age from the clinical outcomes due to decreased bleeding without
ISAR-REACT 5 (Intracoronary Stenting and Antith- differences in ischemic events compared with potent
rombotic Regimen: Rapid Early Action for Coronary P2Y12 inhibitors (mostly ticagrelor) among patients
Treatment 5) trial, prasugrel 5 mg reduced 12-month >70 years of age (29). In the Bremen STEMI Registry,
major bleeding and had similar efficacy compared ticagrelor was associated with decreased ischemic
with standard-dose ticagrelor (26). events and no significant increase in bleeding in
There is more evidence on the safety and efficacy elderly patients (30). Conversely, in the SWEDE-
of ticagrelor in elderly patients. In the PLATO HEART (Swedish Web-System for Enhancement and
(Platelet Inhibition and Patient Outcomes) trial, Development of Evidence-Based Care in Heart Dis-
although bleeding events increased with age, they ease Evaluated According to Recommended Thera-
were not significantly increased in patients treated pies) registry, ticagrelor provided similar efficacy to
with ticagrelor versus clopidogrel across age sub- clopidogrel but increased bleeding and mortality
groups (27,28). Accordingly, use of ticagrelor 90 mg among those $80 years of age (31).
twice daily is recommended after ACS, with no spe- It is important to note that the aforementioned
cific age-related recommendations. However, in the studies with potent P2Y 12 inhibitors were conducted
JACC: CARDIOVASCULAR INTERVENTIONS VOL. 14, NO. 13, 2021 Angiolillo et al. 1443
JULY 12, 2021:1434–46 Age Substudy of the TWILIGHT Trial

F I G U R E 5 Risk for Ischemic Events at 1 Year After Randomization

Forest plots showing the effect of ticagrelor plus placebo versus ticagrelor plus aspirin on the ischemic endpoints in relation to age cutoff of 65 years. Event rates at 1
year were estimated using the Kaplan-Meier method. HRs and 95% CIs with interaction p values generated using Cox regression. Ischemic outcomes were performed in
the per protocol cohort. CV ¼ cardiovascular; ST ¼ stent thrombosis; other abbreviations as in Figures 1, 2, and 3.

on a background of aspirin therapy. Aspirin is well bleeding by HBR status among those $75 years of age
established to be associated with bleeding complica- highlights the importance of bleeding-avoidance
tions due to its gastrointestinal side effects (32). The strategies especially in older and otherwise vulner-
physical disruption of the protective gastric phos- able patients. Collectively, these observations make
pholipid barrier induced by acetylsalicylic acid (i.e., the elderly population ideal for considering aspirin-
aspirin) leads to impaired gastrointestinal protection, free approaches if alternative and effective second-
which enables (i.e., promoting new mucosal lesions) ary prevention antiplatelet treatment regimens are
and propagates (i.e., worsening existing lesions) available.
direct acid injury (32). In the presence of impaired The present analysis is in line with the overall
hemostasis such as with DAPT, the risk for bleeding findings from TWILIGHT and other trials that with-
complications is significantly enhanced. Moreover, drawal of aspirin after a brief period of DAPT does not
the gastrointestinal toxicity induced by aspirin and incur any increase in ischemic complications
the subsequent risk for bleeding can be exacerbated (10,13–16). These efficacy observations are supported
with increasing age (33). by a number of in vitro and ex vivo pharmacodynamic
Finally, it is noteworthy that a large portion of studies. In particular, in vitro investigations con-
elderly patients with coronary artery disease exhibit ducted in platelets from healthy volunteers treated
clinical conditions, such as chronic medications and with potent P2Y 12 inhibitors suggested that aspirin
comorbidities, which further increase their bleeding may provide limited additional platelet inhibition
risk (34). The results of our study showed that among (35,36). Similar findings assessing thrombus forma-
patients $75 years of age, more than half fulfilled the tion were observed in animal studies (37). Studies
ARC-HBR definition. In this subgroup, the incidence conducted in patients with coronary artery disease,
of BARC type 2, 3, or 5 bleeding between 3 and including a substudy from the TWILIGHT trial,
15 months post-PCI was 12.0% with ticagrelor plus showed that although aspirin withdrawal is associ-
aspirin compared with 5.8% with ticagrelor mono- ated with an increase in markers sensitive to
therapy. Although the benefits of ticagrelor mono- cyclooxygenase-1 blockade, this did not affect
therapy were consistent across all age categories, the markers of P2Y12 signaling or ex vivo platelet-
significant interaction for BARC type 2, 3, or 5 dependent thrombus formation (38,39). It is
1444 Angiolillo et al. JACC: CARDIOVASCULAR INTERVENTIONS VOL. 14, NO. 13, 2021

Age Substudy of the TWILIGHT Trial JULY 12, 2021:1434–46

F I G U R E 6 Risk for the Primary and Key Secondary Endpoints by 10-Year Age Intervals

Forest plots showing the effect of ticagrelor plus placebo versus ticagrelor plus aspirin on the bleeding and ischemic endpoints according to 10-year intervals of age.
Event rates at 1 year were estimated using the Kaplan-Meier method. HRs and 95% CIs with interaction p values generated using Cox regression. Abbreviations as in
Figures 1, 2, and 3.

important to note that these findings of pharmaco- P2Y12 inhibitor monotherapy with ST-segment eleva-
dynamic efficacy occur in the presence of reliable and tion MI was not addressed, as these patients were
effective P2Y 12 blockade such as that achieved by excluded from participation in TWILIGHT. Ulti-
ticagrelor and do not apply to clopidogrel, which is mately, the power was limited to detect rare, yet
characterized by less predictable and potent P2Y 12 clinically important differences in ischemic events,
inhibition (40). It is furthermore important to note including stent thrombosis or stroke.
that elderly patients have been consistently shown to
be at increased risk for having high platelet reactivity, CONCLUSIONS
a marker of thrombotic risk, while on treatment with
clopidogrel (41). Age-related platelet dysfunction as Among high-risk patients undergoing PCI, a strategy
well as impaired drug metabolism may contribute to of ticagrelor monotherapy following 3 months of
these findings (5,6). Therefore, these considerations DAPT significantly reduced clinically relevant
warrant caution when considering a strategy of early bleeding compared with ticagrelor plus aspirin
aspirin withdrawal followed by clopidogrel mono- without an increase in ischemic events irrespective of
therapy as bleeding reduction strategy in elderly age. These findings support such a bleeding avoid-
patients. ance strategy, which can be implemented without
any signals for harm in elderly patients at increased
STUDY LIMITATIONS. Although our analysis was pre- risk for bleeding and ischemic complications.
specified, randomization was not stratified by age.
Therefore, our results must be considered hypothesis FUNDING SUPPORT AND AUTHOR DISCLOSURES
generating and warrant dedicated, prospective
This work was supported by an investigator-initiated grant from
confirmation. Our findings may not generalize to pa- AstraZeneca. Dr. Angiolillo has received consulting fees or honoraria
tients treated with other oral P2Y 12 inhibitors, from Abbott, Amgen, Aralez, AstraZeneca, Bayer, Biosensors, Boeh-
including prasugrel and clopidogrel. The representa- ringer Ingelheim, Bristol Myers Squibb, Chiesi, Daiichi-Sankyo, Eli
Lilly, Haemonetics, Janssen, Merck, PhaseBio, PLx Pharma, Pfizer,
tion of female sex and non-White race in the trial was
Sanofi, and The Medicines Company; has participated in review ac-
limited, although the former was a clinical study en- tivities for CeloNova and St. Jude Medical; has received institutional
try criterion. Moreover, elderly patients commonly payments for grants from Amgen, AstraZeneca, Bayer, Biosensors,
affected by functional, social, and cognitive impair- CeloNova, CSL Behring, Daiichi-Sankyo, Eisai, Eli Lilly, Gilead, Idor-
sia, Janssen, Matsutani Chemical Industry, Merck, Novartis, Osprey
ment would have not been enrolled in the TWILIGHT
Medical, Renal Guard Solutions, and the Scott R. MacKenzie Foun-
trial, and thus our findings cannot be generalized to dation. Dr. Baber has received speaking honoraria from AstraZeneca
all elderly subjects (42,43). The safety and efficacy of and Boston Scientific. Dr. Dangas has received consulting fees and
JACC: CARDIOVASCULAR INTERVENTIONS VOL. 14, NO. 13, 2021 Angiolillo et al. 1445
JULY 12, 2021:1434–46 Age Substudy of the TWILIGHT Trial

advisory board fees from AstraZeneca; has received consulting fees Medicines Company, Spectranetics/Philips/Volcano, and Watermark
from Biosensors; and previously held stock in Medtronic. Dr. Mehta Research Partners; and has received nonfinancial support and other
has received grant support from and serving on an executive com- compensation from Regeneron Pharmaceuticals and Idorsia Pharma-
mittee and as site investigator for AstraZeneca. Dr. Cohen has ceuticals. All other authors have reported that they have no re-
received research grant support from Edwards Lifesciences, Med- lationships relevant to the contents of this paper to disclose.
tronic, Boston Scientific, Abbott, Svelte, Corvia, and Ancora; has
received consulting fees from Edwards Lifesciences, Medtronic,
ADDRESS FOR CORRESPONDENCE: Dr. Roxana
Boston Scientific, Abbott, Svelte, Corvia, and Ancora. Dr. Krucoff has
received consulting fees or honoraria and research grants from Abbott Mehran, Center for Interventional Cardiovascular
Vascular, Biosensors, Boston Scientific, CeloNova, Medtronic, and Research and Clinical Trials, The Zena and Michael A.
OrbusNeich. Dr. Gibson has received research grant support from
Wiener Cardiovascular Institute, Icahn School of
AstraZeneca. Dr. Moliterno has received grants from AstraZeneca,
Medicine at Mount Sinai, One Gustave L. Levy Place,
during the conduct of the study. Dr. Gibson has received grant sup-
port and consulting fees from Angel Medical, Bayer, CSL Behring, Box 1030, New York, New York 10029-6574, USA.
Janssen Pharmaceuticals, Johnson & Johnson, and Portola Pharma- E-mail: [email protected].
ceuticals; has received consulting fees from The Medicines Company,
Eli Lilly, Gilead Sciences, Novo Nordisk, WebMD, UpToDate Cardio-
vascular Medicine, Amarin Pharma, Amgen, Boehringer Ingelheim,
PERSPECTIVES
Chiesi, Merck, PharmaMar, Sanofi, Somahlution, Verreseon, Boston
Scientific, Impact Bio, MedImmume, Medtelligence, MicroPort, PERT
Consortium, and GE Healthcare; holds equity in nference; serves as
WHAT IS KNOWN? In patients with coronary artery disease,
chief executive officer of the Baim Institute; and has received grant
support, paid to the Baim Institute, from Bristol Myers Squibb. Dr. ischemic and bleeding events increase with age warranting
Huber has received lecture fees from AstraZeneca and Bayer. Dr. careful evaluation of the risks and benefits associated with
Weisz has received grant support and advisory board fees from and different antiplatelet therapies.
holds equity in Corindus; has received advisory board fees from and
holds equity in Filterlex; serves on an advisory board for and holds
options in Trisol; and has received grant support from Abbott, Car-
WHAT IS NEW? Among PCI patients $65 years of age, a
diovascular Systems Inc, and RenalGuard. Dr. Oldroyd has received strategy of ticagrelor monotherapy, after a short DAPT course,
grant support and lecture fees from AstraZeneca. Dr. Escaned has significantly reduced clinically relevant and major bleeding
received consulting and lecture fees from Abbott, Philips, Boston
compared with ticagrelor plus aspirin without an increase in
Scientific, and Medtronic; and has received lecture fees from
Abiomed, Terumo, and Biosensors. Dr. Sharma has received consul-
ischemic events. These benefits were preserved across different
ting fees or honoraria from Abbott Vascular, Boston Scientific, and age categories and enhanced among HBR patients $75 years.
Cardiovascular Systems. Dr. Mehran reports grants from Abbott
Laboratories, AstraZeneca, Bayer, Beth Israel Deaconess, Bristol WHAT IS NEXT? In high-risk patients undergoing PCI, after a
Myers Squibb, CSL Behring, Daiichi-Sankyo, Medtronic, Novartis
short DAPT course, P2Y12 inhibitor monotherapy with ticagrelor
Pharmaceuticals, OrbusNeich; has received personal fees from Abbott
Laboratories, Boston Scientific, Medscape/WebMD, Siemens Medical is a safe bleeding-avoidance strategy, irrespective of age. Dedi-
Solutions, PLx Opco dba PLx Pharma, Roivant Sciences, Sanofi, cated and adequately powered studies are warranted to confirm
Medtelligence (Janssen Scientific Affairs), and Janssen Scientific Af-
these findings particularly in the subgroup of elderly patients.
fairs; has received other compensation from Abbott Laboratories,
Abiomed, Bristol Myers Squibb, Claret Medical, Elixir Medical, The

REFERENCES

1. Capodanno D, Alfonso F, Levine GN, review of the evidence and evolving paradigms. report from the American Heart Association. Cir-
Valgimigli M, Angiolillo DJ. ACC/AHA versus ESC Expert Opin Drug Safety 2019;18:1171–89. culation 2019;139:e56–66.
guidelines on dual antiplatelet therapy: JACC
5. Capodanno D, Angiolillo DJ. Antithrombotic 10. Baber U, Dangas G, Cohen DJ, et al. Ticagrelor
guideline comparison. J Am Coll Cardiol 2018;72:
therapy in the elderly. J Am Coll Cardiol 2010;56: with aspirin or alone in high-risk patients after
2915–31.
1683–92. coronary intervention: rationale and design of the
2. Bittl JA, Baber U, Bradley SM, Wijeysundera DN. TWILIGHT study. Am Heart J 2016;182:125–34.
6. Andreotti F, Rocca B, Husted S, et al. Antith-
Duration of dual antiplatelet therapy: a systematic 11. Mehran R, Rao SV, Bhatt DL, et al. Standard-
rombotic therapy in the elderly: expert position
review for the 2016 ACC/AHA guideline focused ized bleeding definitions for cardiovascular clinical
paper of the European Society of Cardiology
update on duration of dual antiplatelet therapy in trials: a consensus report from the Bleeding Aca-
Working Group on Thrombosis. Eur Heart J 2015;
patients with coronary artery disease: a report of demic Research Consortium. Circulation 2011;123:
36:3238–49.
the American College of Cardiology/American 2736–47.
Heart Association Task Force on Clinical Practice 7. Capodanno D, Mehran R, Valgimigli M, et al.
12. Urban P, Mehran R, Colleran R, et al. Defining
Guidelines. J Am Coll Cardiol 2016;68:1116–39. Aspirin-free strategies in cardiovascular disease
high bleeding risk in patients undergoing percu-
and cardioembolic stroke prevention. Nat Rev
3. Genereux P, Giustino G, Witzenbichler B, et al. taneous coronary intervention. Circulation 2019;
Cardiol 2018;15:480–96.
Incidence, predictors, and impact of post- 140:240–61.
discharge bleeding after percutaneous coronary 8. Mehran R, Baber U, Sharma SK, et al. Ticagrelor
13. Vranckx P, Valgimigli M, Juni P, et al. Tica-
intervention. J Am Coll Cardiol 2015;66:1036–45. with or without aspirin in high-risk patients after
grelor plus aspirin for 1 month, followed by tica-
PCI. N Engl J Med 2019;381:2032–42.
4. Buccheri S, Capodanno D, James S, grelor monotherapy for 23 months vs aspirin plus
Angiolillo DJ. Bleeding after antiplatelet therapy 9. Benjamin EJ, Muntner P, Alonso A, et al. Heart clopidogrel or ticagrelor for 12 months, followed
for the treatment of acute coronary syndromes: a disease and stroke statistics—2019 update: a by aspirin monotherapy for 12 months after
1446 Angiolillo et al. JACC: CARDIOVASCULAR INTERVENTIONS VOL. 14, NO. 13, 2021

Age Substudy of the TWILIGHT Trial JULY 12, 2021:1434–46

implantation of a drug-eluting stent: a multi- the safety of long-term dual antiplatelet therapy 34. Capranzano P, Angiolillo DJ. Antithrombotic
centre, open-label, randomised superiority trial. with reduced-dose prasugrel versus standard- management of elderly patients with coronary
Lancet 2018;392:940–9. dose clopidogrel. Circulation 2013;128:823–33. artery disease. J Am Coll Cardiol Intv 2021;14:
723–38.
14. Hahn JY, Song YB, Oh JH, et al. Effect of P2Y12 24. Savonitto S, Ferri LA, Piatti L, et al. Com-
inhibitor monotherapy vs dual antiplatelet therapy parison of reduced-dose prasugrel and standard- 35. Armstrong PC, Leadbeater PD, Chan MV, et al. In
on cardiovascular events in patients undergoing dose clopidogrel in elderly patients with acute the presence of strong P2Y12 receptor blockade,
percutaneous coronary intervention: the SMART- coronary syndromes undergoing early percuta- aspirin provides little additional inhibition of platelet
CHOICE randomized clinical trial. JAMA 2019;321: neous revascularization. Circulation 2018;137: aggregation. J Thromb Haemost 2011;9:552–61.
2428–37. 2435–45. 36. Kirkby NS, Leadbeater PD, Chan MV,
15. Watanabe H, Domei T, Morimoto T, et al. Ef- 25. Cayla G, Cuisset T, Silvain J, et al. Platelet Nylander S, Mitchell JA, Warner TD. Antiplatelet
fect of 1-month dual antiplatelet therapy followed function monitoring to adjust antiplatelet therapy effects of aspirin vary with level of P2Y12 receptor
by clopidogrel vs 12-month dual antiplatelet in elderly patients stented for an acute coronary blockade supplied by either ticagrelor or prasu-
therapy on cardiovascular and bleeding events in syndrome (ANTARCTIC): an open-label, blinded- grel. J Thromb Haemost 2011;9:2103–5.
patients receiving PCI: the STOPDAPT-2 random- endpoint, randomised controlled superiority trial. 37. Vilahur G, Gutiérrez M, Casani L, et al. P2Y12
ized clinical trial. JAMA 2019;321:2414–27. Lancet 2016;388:2015–22. antagonists and cardiac repair post-myocardial
16. Kim BK, Hong SJ, Cho YH, et al. Effect of 26. Schüpke S, Neumann FJ, Menichelli M, et al. infarction: global and regional heart function
ticagrelor monotherapy vs ticagrelor with aspirin Ticagrelor or prasugrel in patients with acute analysis and molecular assessments in pigs. Car-
on major bleeding and cardiovascular events in coronary syndromes. N Engl J Med 2019;381: diovasc Res 2018;114:1860–70.
patients with acute coronary syndrome: the TICO 1524–34. 38. Baber U, Zafar MU, Dangas G, et al. Ticagrelor
randomized clinical trial. JAMA 2020;323: with or without aspirin after PCI: the TWILIGHT
27. Wallentin L, Becker RC, Budaj A, et al. Tica-
2407–16. platelet substudy. J Am Coll Cardiol 2020;75:
grelor versus clopidogrel in patients with acute
17. Tomaniak M, Chichareon P, Modolo R, et al. coronary syndromes. N Engl J Med 2009;361: 578–86.
Ticagrelor monotherapy beyond one month after 1045–57. 39. Franchi F, Rollini F, Faz G, et al. Pharmaco-
PCI in ACS or stable CAD in elderly patients: a pre- dynamic effects of vorapaxar in prior myocardial
28. Husted S, James S, Becker RC, et al. Ticagrelor
specified analysis of the GLOBAL LEADERS trial. infarction patients treated with potent oral P2Y12
versus clopidogrel in elderly patients with acute
EuroIntervention 2020;15:e1605–14. receptor inhibitors with and without aspirin: re-
coronary syndromes: a substudy from the pro-
18. Franchi F, Angiolillo DJ. Novel antiplatelet spective randomized Platelet Inhibition and Pa- sults of the VORA-PRATIC study. J Am Heart Assoc
agents in acute coronary syndrome. Nat Rev Car- tient Outcomes (PLATO) trial. Circ Cardiovasc Qual 2020;9:e015865.
diol 2015;12:30–47. Outcomes 2012;5:680–8. 40. Franchi F, Rollini F, Kairouz V, et al. Pharma-
codynamic effects of vorapaxar in patients with
19. Wiviott SD, Braunwald E, McCabe CH, et al. 29. Gimbel M, Qaderdan K, Willemsen L, et al.
and without diabetes mellitus: results of the
Prasugrel versus clopidogrel in patients with acute Clopidogrel versus ticagrelor or prasugrel in pa-
OPTIMUS-5 study. J Am Coll Cardiol Basic Trans
coronary syndromes. N Engl J Med 2007;357: tients aged 70 years or older with non-ST-
Science 2019;4:763–75.
2001–15. elevation acute coronary syndrome (POPular
AGE): the randomised, open-label, non-inferiority 41. Angiolillo DJ, Capodanno D, Danchin N, et al.
20. Riesmeyer JS, Salazar DE, Weerakkody GJ,
trial. Lancet 2020;395:1374–81. Derivation, validation, and prognostic utility of a
et al. Relationship between exposure to prasugrel
prediction rule for nonresponse to clopidogrel: the
active metabolite and clinical outcomes in the 30. Schmucker J, Fach A, Mata Marin LA, et al.
ABCD-GENE score. J Am Coll Cardiol Intv 2020;13:
TRITON-TIMI 38 substudy. J Clin Pharmacol 2012; Efficacy and safety of ticagrelor in comparison to
606–17.
52:789–97. clopidogrel in elderly patients with ST-segment-
elevation myocardial infarctions. J Am Heart 42. Capranzano P, Angiolillo DJ. Tailoring P2Y12
21. Erlinge D, Gurbel PA, James S, et al. Prasugrel
Assoc 2019;8:e012530. inhibiting therapy in elderly patients with
5 mg in the very elderly attenuates platelet inhi-
myocardial infarction undergoing primary percu-
bition but maintains noninferiority to prasugrel 31. Szummer K, Montez-Rath ME, Alfredsson J,
taneous coronary intervention. J Am Heart Assoc
10 mg in nonelderly patients: the GENERATIONS et al. Comparison between ticagrelor and clopi-
2019;8:e014000.
trial, a pharmacodynamic and pharmacokinetic dogrel in elderly patients with an acute coronary
study in stable coronary artery disease patients. syndrome: insights from the SWEDEHEART regis- 43. Capranzano P, Angiolillo DJ. Ticagrelor or
J Am Coll Cardiol 2013;62:577–83. try. Circulation 2020;142:1700–8. clopidogrel in elderly patients with myocardial
infarction: when the choice makes the difference.
22. Capranzano P, Tamburino C, Capodanno D, 32. García Rodríguez LA, Hernández-Díaz S, de
Circulation 2020;142:1709–12.
et al. Platelet function profiles in the elderly: re- Abajo FJ. Association between aspirin and upper
sults of a pharmacodynamic study in patients on gastrointestinal complications: systematic review
clopidogrel therapy and effects of switching to of epidemiologic studies. Br J Clin Pharmacol KEY WORDS age, bleeding, PCI,
prasugrel 5 mg in patients with high platelet 2001;52:563–71. thrombosis, ticagrelor monotherapy
reactivity. Thromb Haemost 2011;106:1149–57.
33. Capodanno D, Ingala S, Calderone D,
23. Roe MT, Goodman SG, Ohman EM, et al. Angiolillo DJ. Aspirin for the primary prevention of A PPE NDI X For supplemental tables and a
Elderly patients with acute coronary syndromes cardiovascular disease: latest evidence. Expert Rev figure, please see the online version of this
managed without revascularization: insights into Cardiovasc Ther 2019;17:633–43. paper.