European Heart Journal (2016) 37, 1133–1142 doi:10.1093/eurheartj/ehv531

ESC Clinical Trial Update

Thrombosis and antithrombotic therapy

Ischaemic risk and efficacy of ticagrelor in relation to time from P2Y12 inhibitor withdrawal in patients with prior myocardial infarction: insights from PEGASUS-TIMI 54 Marc P. Bonaca1*, Deepak L. Bhatt 1, P. Gabriel Steg 2, Robert F. Storey3, Marc Cohen 4, Kyungah Im 1, Ton Oude Ophuis 5, Andrej Budaj 6, Shinya Goto 7, Jose´ Lo´pez-Sendo´n 8, Rafael Diaz 9, Anthony Dalby 10, Frans Van de Werf 11, Diego Ardissino 12, Gilles Montalescot 13, Philip Aylward 14, Giulia Magnani 1, Eva C. Jensen 15, Peter Held 15, Eugene Braunwald1, and Marc S. Sabatine 1 1 TIMI Study Group, Brigham and Women’s Hospital, Heart & Vascular Center, 75 Francis Street, Boston, MA 02115, USA; 2De´partement de Cardiologie, Hoˆpital Bichat, Assistance Publique, Paris, France; 3University of Sheffield, Sheffield, UK; 4Cardiovascular Division, Newark Beth Israel Medical Center, Mount Sinai School of Medicine, New York, USA; 5CWZ Hospital, Nijmegen, The Netherlands; 6Postgraduate Medical School, Grochowski Hospital, Warsaw, Poland; 7Tokai University School of Medicine, Institute of Medical Science, Tokyo, Japan; 8Hospital Universitario La Paz, Instituto de Investigacio´n La PAZ, Madrid, Spain; 9ECLA (Estudios Clı´nicos Latino Ame´rica), Rosario, Argentina; 10Milpark Hospital, Johannesburg, South Africa; 11Department of Cardiovascular Sciences, University of Leuven, Leuven, Belgium; 12Cardiovascular Division, Azienda Ospedaliero-Universitaria di Parma, Parma, Italy; 13 Institut de Cardiologie, Pitie´-Salpeˆtrie`re Hospital, 47 boul de l’Hoˆpital, Paris, France; 14Division of Medicine, Cardiac & Critical Care Services, Flinders Medical Centre, Adelaide, South Australia, Australia; and 15AstraZenecaAZ R&D, Molndal, Sweden

Received 31 July 2015; revised 13 September 2015; accepted 18 September 2015; online publish-ahead-of-print 21 October 2015

See page 1143 for the editorial comment on this article (doi:10.1093/eurheartj/ehv573)

Aims

Ticagrelor reduced major adverse cardiovascular event (MACE) by 15– 16% in patients with prior myocardial infarction (MI) in PEGASUS-TIMI 54. We hypothesized that patients who recently discontinued P2Y12 inhibition, even years after MI, may be at particular risk of MACE and may derive particular benefit from continuation or reinitiation of therapy. ..................................................................................................................................................................................... Methods Patients in PEGASUS-TIMI 54 were categorized by time from last P2Y12 inhibitor (days: ≤30, .30 –360, .360). The and results risk of MACE and the efficacy of ticagrelor were compared across categories. In the placebo arm, patients who more recently stopped P2Y12 inhibitor therapy had a greater number of risk factors but still had a higher risk of MACE after multivariable adjustment [≤30 days, hazard ratio (HR)adj 1.47, 95% confidence interval (CI) 1.12 –1.93, P ¼ 0.0051; 30 days – 1 year, HRadj 1.28, 95% CI 0.98 – 1.67, P ¼ 0.073] compared with those who stopped .1 year prior (P-trend ¼ 0.0097). The benefit of ticagrelor depended on the time from last dose, with HRs (95% CI) for ticagrelor (pooled doses) vs. placebo of 0.73 (0.61 – 0.87), 0.86 (0.71 – 1.04), and 1.01 (0.80 – 1.27), respectively, by category (P-trend for interaction , 0.001). The benefit in those ≤30 days of stopping was similar regardless of time from MI (,2 years, HR 0.73, 95% CI 0.60–0.89 vs. ≥2 years, HR 0.71, 95% CI 0.50 –1.00). ..................................................................................................................................................................................... Conclusion The benefit of ticagrelor for long-term secondary prevention in patients with prior MI and at least one additional risk factor appeared more marked in patients continuing on or re-starting after only a brief interruption of P2Y12 inhibition, when compared with patients who had proved themselves stable more than 2 years from their MI and off P2Y12 inhibitor therapy for more than a year. The increase in bleeding events with ticagrelor was similar regardless of this time interval. For clinicians considering a strategy of prolonged P2Y12 inhibitor therapy in high-risk patients, these data suggest greater benefit in the continuation of such therapy without interruption after MI, rather than re-initiating such therapy in patients who have remained stable for an extended period. Future analyses may help to clarify further the profile of post-MI patients most likely to benefit from uninterrupted dual antiplatelet therapy. ..................................................................................................................................................................................... * Corresponding author. Tel: +1 617 278 0071, Fax: +1 617 734 7329, Email: [email protected] Published on behalf of the European Society of Cardiology. All rights reserved. & The Author 2015. For permissions please email: [email protected].

1134

M.P. Bonaca et al.

Clinical Trial Registration Information

http://www.clinicaltrials.gov NCT01225562.

Keywords

Coronary artery disease † Platelet inhibition † P2Y12 inhibition † Ischaemic risk † Ticagrelor † Antiplatelet therapy † Secondary prevention

-----------------------------------------------------------------------------------------------------------------------------------------------------------

Introduction In the PEGASUS-TIMI 54 trial, long-term secondary prevention with ticagrelor reduced the risk of major adverse cardiovascular events (MACE) in stable outpatients with a history of myocardial infarction (MI).1 The population studied included both patients who had just completed P2Y12 inhibition after their index MI as well as patients who were stable on aspirin monotherapy having completed therapy with P2Y12 inhibition months or even more than a year prior. For clinicians, these different populations represent distinct clinical scenarios in practice. Recently, the dual antiplatelet therapy (DAPT) trial investigated the optimal duration of P2Y12 inhibition after coronary stenting for the prevention of stent thrombosis and MACE, a different but related question. A notable observation from the DAPT trial was an apparent increased risk of ischaemic events during the 3 months after P2Y12 inhibition withdrawal, both for patients randomized to placebo at 12 months from their qualifying percutaneous coronary intervention (PCI) as well as those withdrawn at the end of the 30-month treatment period.2 These findings suggest that withdrawal of P2Y12 inhibition may be associated with heightened risk by unmasking patients vulnerable to ischaemic events, even years after their qualifying event.2 In a pre-specified analysis, we hypothesized that in PEGASUSTIMI 54 the timing of P2Y12 inhibitor withdrawal prior to randomization would be associated with ischaemic risk and the benefit of ticagrelor. Specifically, we hypothesized that patients enrolled after recent P2Y12 inhibitor withdrawal would be at heightened ischaemic risk and therefore derive greater benefit from long-term ticagrelor therapy, whereas patients who had remained event-free on aspirin therapy alone for an extended period would be at relatively lower risk and derive a more modest benefit.

Methods Study population The PEGASUS-TIMI 54 trial randomized 21 162 patients with a history of a spontaneous MI and one additional high-risk feature to receive ticagrelor 90 mg orally twice daily, ticagrelor 60 mg orally twice daily, or placebo. Full inclusion and exclusion criteria have been published previously.3 Patients were recruited in the stable setting a median of 1.7 years, interquartile range (IQR) (1.2 – 2.3) from their qualifying MI. Patients requiring or expected to need a P2Y12 inhibitor (e.g. for planned coronary revascularization) at the time of enrolment were excluded. Patients on a P2Y12 inhibitor that was planned to be stopped could be enrolled upon or soon after stopping the P2Y12 inhibitor.3 Sites were to report the type and last dose date of P2Y12 inhibitor

therapy prior to randomization if known. All patients were to take aspirin at a dose of 75 – 150 mg daily.

Endpoints The pre-specified primary efficacy endpoint was the composite of cardiovascular death, MI, or stroke. Additional efficacy endpoints included the individual components of the composite as well as coronary heart disease-related death. The primary safety endpoint was TIMI Major Bleeding. Other safety endpoints included intracranial haemorrhage and fatal bleeding. All potential events were adjudicated by a clinical events committee which was blinded to treatment allocation.

Statistical methods Patients were categorized according to the time from last dose of a P2Y 12 inhibitor to randomization into three pre-specified groups (≤30 days, n ¼ 7181; .30 – 360 days, n ¼ 6501; and .360 days, n ¼ 5079). Patients with missing information about pre-randomization P2Y12 inhibitor utilization were excluded from the primary analysis (n ¼ 2401), leaving 18 761 for analysis. Baseline characteristics among groups were compared using the x 2 test for categorical variables and the Kruskal – Wallis test for continuous variables. The risk of the primary endpoint in the placebo group through 90 days and 3 years was compared between categories both unadjusted and subsequently adjusted for baseline characteristics that differed between groups including age, sex, race, region, time from qualifying MI, diabetes, multivessel disease, hypertension, hypercholesterolaemia, and history of PCI/stent. A trend-test was examined to assess for a significant increasing or decreasing risk of the primary efficacy and safety endpoints across P2Y12 inhibitor withdrawal time category. The efficacy and safety of ticagrelor relative to placebo was compared within each of the three P2Y 12 inhibitor timing categories with testing for formal treatment interaction for between treatment and time category. Because the two doses of ticagrelor (60 and 90 mg twice daily) showed similar efficacy overall for the primary endpoint, the main approach for the current analysis was to evaluate both doses of ticagrelor pooled against placebo, with additional analyses performed to assess the consistency of findings with each dose vs. placebo. Cumulative event rates were calculated by the Kaplan – Meier method. Efficacy data and allcause mortality were analysed on an intention-to-treat basis. Safety analyses were performed among patients who received one or more doses of study drug and included events through 7 days after permanent cessation of study therapy. Hazard ratios (HRs) and 95% confidence intervals (CIs) were generated using a Cox proportional hazards model and all reported P values are two-sided. Deviation from proportionality assumption was visually examined by Schonfeld residuals and complementary log – log plots. Analyses were performed using SAS software v9.3 (SAS Institute Inc., Cary, NC, USA).

1135

P2Y12 inhibitor withdrawal and ischaemic risk

Results Baseline characteristics

Ischaemic and bleeding risk by timing of P2Y12 inhibitor withdrawal in patients randomized to placebo

A total of 18 761 patients had a recorded history of the timing of their last P2Y12 inhibitor dose prior to randomization: 7181 (38%) had their last dose within 30 days from randomization, 6501 (35%) had their last dose between 30 days and 1 year prior to randomization, and 5079 (27%) had their last dose more than a year from randomization. Baseline characteristics by timing from last dose of P2Y12 inhibitor are presented in Table 1 and baseline characteristics by treatment allocation within each time bin are shown in Supplementary material online, Table S1. Patients with P2Y12 inhibitor withdrawal ≤30 days were, temporally closer to their most recent MI (median 16 vs. 19 vs. 29 months, P , 0.001). They also more frequently had diabetes, multivessel coronary artery disease (CAD), and prior PCI, and were more likely to be randomized in North America. In those in whom prior P2Y12 therapy was reported, clopidogrel had been used in 94%, prasugrel in 5%, ticagrelor in ,1%, and ticlopidine in ,1%. Almost all patients stopped P2Y12 inhibition prior to randomization at the recommendation of their treating physician (94% in all three time bins; Supplementary material online, Table S2).

Patients in the placebo arm who had discontinued their P2Y12 inhibitor within the past 30 days were at higher risk for cardiovascular death, MI, or stroke (9.91% at 3 years) when compared with those who had discontinued 30 days to 1 year prior (8.70%) and those who stopped more than 1 year before randomization (6.91%, P-trend 0.0097). Examining the cumulative event curves over time, this relative excess was particularly pronounced early, within the first 90 days from randomization, with event rates of 1.46, 0.55, and 0.60%, respectively, in the three groups (Figure 1). The heightened risk was directionally similar for each of the individual endpoints of cardiovascular death, MI, and stroke (Supplementary material online, Figure S1). Within each time category from last dose of P2Y12 inhibitor, the risk of cardiovascular death, MI, or stroke tended to be slightly greater in patients who were closer to their qualifying MI (Supplementary material online, Figure S2). Nonetheless, after adjusting for time from qualifying MI and the differences in other baseline characteristics, a highly significant relationship between time from last dose of P2Y12 inhibitor and cardiovascular risk remained

Table 1

Baseline characteristics (randomized treatment allocation)

Variable

≤30 days (n 5 7181)

>30 days to 1 year (n 5 6501)

>1 year (n 5 5079)

P-value

,0.001

............................................................................................................................................................................... Demographics Age, median (IQR)

65 (58– 71)

65 (59– 71)

66 (59– 71)

Female (%)

22

23

23

Caucasiana (%) BMI, median (IQR)

82 28 (25– 31)

91 28 (25– 31)

91 28 (25– 31)

Hypertension (%)

78

76

76

0.012

Hyperlipidaemia (%) Current smoker (%)

80 17

76 16

77 17

,0.001 0.27

0.054 ,0.001 0.070

............................................................................................................................................................................... Clinical characteristics

Diabetes mellitus (%)

33

30

31

,0.001

Multivessel CAD (%) PCI for index MI (%)

68 92

60 87

58 87

,0.001 ,0.001

History of more than 1 prior MI (%)

16

16

16

0.84

eGFR at baseline ,60 mL/min (%)b

22

22

23

0.41

Western Europe (%) Eastern Europe (%)

24 25

34 30

38 29

North America (%)

28

16

13

South America (%) Asia/Pacific (%)

7 17

12 8

12 7

............................................................................................................................................................................... Region

}

,0.001

............................................................................................................................................................................... Qualifying event Months from MI—median (IQR)

a

16 (13– 24)

19 (15– 23)

29 (25– 32)

STEMI (%)

55

55

56

NSTEMI (%) MI type unknown (%)

41 4

42 4

40 4

Self reported. The estimated glomerular filtration rate was calculated with the use of the Modification of Diet in Renal Disease equation.

b

}

,0.001 0.77

1136 (P for trend ¼ 0.0097), with HRs of 1.47 (95% CI 1.12 – 1.93) and 1.28 (95% CI 0.98 –1.67) for cardiovascular death, MI, or stroke in patients who had stopped within 30 days and those who stopped between 30 days and 1 year, respectively, when compared with those who stopped .1 year before randomization adjusted for differences in baseline characteristics as shown in Table 1. TIMI Major Bleeding in patients randomized to placebo was similar across time bins from P2Y12 inhibitor withdrawal with 3 year-rates in those ≤30 days, 30 days to 1 year, and more than 1 year of 0.7, 1.2, and 0.9%, respectively. After adjustment for baseline differences, the HR for TIMI Major Bleeding was 0.51 (95% CI 0.20–1.33, P ¼ 0.17) for patients stopping ≤30 days and 0.69 (0.29–1.66, P ¼ 0.41) for those stopping 30 days to 1 year both compared with those stopping .1 year from randomization (P-trend across groups 0.18).

Efficacy and safety of ticagrelor by time from P2Y12 inhibitor withdrawal relative to randomization There was a highly statistically significant interaction between the efficacy of ticagrelor in reducing the risk of cardiovascular death, MI, or stroke and the time from last dose of P2Y12 inhibitor (P for trend for interaction ,0.001, Figure 2). The greatest benefit with ticagrelor was seen in patients who had discontinued P2Y12 inhibition within 30 days, with HR of 0.73 (95% CI 0.61–0.87, P ¼ 0.0003) for the pooled ticagrelor doses vs. placebo, and with HRs of 0.75 (95% CI 0.61–0.92, P ¼ 0.0064) and 0.70 (95% CI 0.57 –0.87, P ¼ 0.0009) for the 60 and 90 mg doses, respectively (Figure 3). The magnitude of benefit in this subgroup was similar regardless of whether the patient’s qualifying MI was ,2 years prior to randomization (HR 0.73, 95% CI 0.60 –0.89) or 2 years or more (HR 0.71, 95% CI 0.50–1.00, Figure 4). The benefit of ticagrelor (doses pooled) was notable both in the first 90 days after randomization, with an HR of 0.54 (95% CI

M.P. Bonaca et al.

0.34–0.86), and also beyond 90 days (HR 0.76, 95% CI 0.63–0.92) for both doses pooled (Figure 5). Similar effect sizes were seen for the individual components of the primary endpoint, with HRs for pooled ticagrelor doses vs. placebo for cardiovascular death of 0.78 (95% CI 0.57 – 1.07), for MI of 0.72 (95% CI 0.58 – 0.90), and for stroke of 0.64 (95% CI 0.43–0.95) (Table 2). Patients who had discontinued P2Y12 inhibition 30 days to 1 year prior had HRs of 0.86 (95% CI 0.71– 1.04, P ¼ 0.11), 0.82 (95% CI 0.65 – 1.02, P ¼ 0.077), and 0.90 (95% CI 0.72 –1.12, P ¼ 0.35), respectively, for the pooled, 60, and 90 mg doses. No benefit was seen in patients who had discontinued P2Y12 inhibition .1 year prior, with HRs of 1.01 (95% CI 0.80 – 1.27, P ¼ 0.96), 1.05 (95% CI 0.81 – 1.38, P ¼ 0.70), and 0.96 (95% CI 0.73 – 1.26, P ¼ 0.75), respectively. Additional individual endpoints are presented in Table 2. Overall efficacy results were similar when excluding patients (n ¼ 876) who had had an MI or PCI within 12 months prior to randomization (Supplementary material online, Table S3). The rates of ticagrelor study drug discontinuation were somewhat higher in patients who were further removed from their last P2Y12 inhibitor dose at randomization, with rates for the pooled ticagrelor doses of 31, 32, and 36% in patients who had stopped within 30 days, between 30 days and 1 year, and .1 year prior to randomization, respectively. Consequently, on-treatment analyses showed a larger effect for ticagrelor in all groups but particularly in the latter, with HRs (95% CI) for cardiovascular death, MI, or stroke with ticagrelor of 0.68 (0.56–0.83), 0.81 (0.65–1.01), and 0.88 (0.67–1.17), respectively, for patients who had stopped within 30 days, between 30 days and 1 year, and .1 year prior to randomization. Both doses of ticagrelor increased TIMI Major Bleeding compared with placebo, with similar HRs and absolute risk differences regardless of time from last dose of P2Y12 inhibitor (P-interaction for 90 mg ¼ 0.90, 60 mg ¼ 0.62, Table 2, Supplementary material online, Tables S3 and S4). There were no statistically significant

Figure 1 Cardiovascular death, myocardial infarction, or stroke in patients randomized to placebo at 90 days and 3 years by time from P2Y12 inhibitor withdrawal.

P2Y12 inhibitor withdrawal and ischaemic risk

1137

Figure 2 Cardiovascular death, myocardial infarction, or stroke at 3 years with ticagrelor 90 mg twice daily (red), ticagrelor 60 mg twice daily (blue), and both doses pooled (purple) compared with placebo by time from P2Y12 inhibitor withdrawal.

Figure 3 Cardiovascular death, myocardial infarction, or stroke at 3 years with ticagrelor 90 mg twice daily (red), ticagrelor 60 mg twice daily (blue) compared with placebo (green) in patients withdrawn from P2Y12 inhibitor therapy within 30 days (A), between 30 days and 1 year (B), and more than 1 year of randomization (C).

differences in the rates of either intracranial haemorrhage or fatal bleeding although events were infrequent (Table 2).

Data for the 2401 patients with missing data on the timing of P2Y12 inhibitor withdrawal are shown in Supplementary material online, Table S4.

1138

M.P. Bonaca et al.

Figure 4 Cardiovascular death, myocardial infarction, or stroke at 3 years with ticagrelor doses pooled (purple) compared with placebo in patients withdrawn from P2Y12 inhibitor therapy within 30 days of randomization stratified by a qualifying myocardial infarction ,24 months prior to randomization (left) and ≥24 months prior to randomization (right).

Figure 5 Cardiovascular death, myocardial infarction, or stroke for the first 90 days (left) and as a landmark from 90 days to 3 years (right) in patients withdrawn from P2Y12 inhibition ≤30 days from randomization by treatment (placebo vs. ticagrelor doses pooled).

Discussion The results of the current analysis demonstrate that patients with prior MI and at least one additional risk factor who recently discontinued P2Y12 inhibition appeared to derive a greater ischaemic risk reduction from long-term ticagrelor relative to patients who

initiated therapy after remaining event free on aspirin monotherapy for a prolonged period. Importantly, ticagrelor increased bleeding to a similar extent regardless of time from P2Y12 inhibitor withdrawal. For clinicians considering a strategy of prolonged P2Y12 inhibitor therapy in high-risk patients, our data suggest greater benefit in the continuation of such therapy without interruption after MI,

Outcomes at 3 years by time from P2Y12 inhibitor withdrawal to randomization

Timing from P2Y12 inhibitor withdrawal

Endpoint

Placebo (n 5 6258) n (%)

Ticagrelor pooled (n 5 12 503) n (%)

Ticagrelor 90 mg bid (n 5 6245) n (%)

Ticagrelor 60 mg bid (n 5 6258) n (%)

HR (95% CI) P-value

.................................................................................

Ticagrelor pooled vs. placebo

Ticagrelor 90 mg bid vs. placebo

Ticagrelor 60 mg bid vs. placebo

0.73 (0.61 –0.87) 0.0003 0.78 (0.57 –1.07) 0.13 0.72 (0.47 –1.11) 0.14 0.72 (0.58 –0.90) 0.0038 0.64 (0.43 –0.95) 0.025

0.70 (0.57– 0.87) 0.0009 0.83 (0.58– 1.19) 0.31 0.67 (0.40– 1.12) 0.12 0.67 (0.51– 0.87) 0.0028 0.64 (0.4– 1.02) 0.061

0.75 (0.61 –0.92) 0.0064 0.74 (0.51 –1.08) 0.12 0.78 (0.48 –1.28) 0.33 0.78 (0.61 –1.01) 0.061 0.64 (0.4– 1.03) 0.064

3.36 (1.91 –5.92) ,0.0001 3.50 (2.21 –5.54) ,0.0001 1.79 (0.77 –4.18) 0.18 2.2 (0.47– 10.34) 0.32 0.91 (0.72 –1.15) 0.43

3.44 (1.88– 6.28) ,0.0001 3.57 (2.19– 5.83) ,0.0001 2.07 (0.82– 5.18) 0.12 2.25 (0.41– 12.28) 0.35 1.02 (0.78– 1.33) 0.90

3.3 (1.8–6.03) 0.0001 3.43 (2.1– 5.60) ,0.0001 1.53 (0.58 –4.03) 0.39 2.15 (0.4– 11.76) 0.3 0.80 (0.60 –1.06) 0.12

0.86 (0.71 –1.04) 0.11 0.76 (0.55 –1.06) 0.10 0.70 (0.46 –1.07) 0.099 0.85 (0.66 –1.08) 0.19 0.91 (0.6– 1.39) 0.68

0.90 (0.72– 1.12) 0.35 0.84 (0.57– 1.22) 0.35 0.71 (0.43– 1.17) 0.18 0.82 (0.61– 1.09) 0.17 1.02 (0.63– 1.64) 0.93

0.82 (0.65 –1.02) 0.08 0.69 (0.46 –1.02) 0.065 0.70 (0.42 –1.15) 0.16 0.88 (0.66 –1.16) 0.37 0.82 (0.49 –1.35) 0.42

2.89 (1.69 –4.94) 0.0001 3.11 (1.96 –4.96) ,0.0001

2.85 (1.59– 5.1) 0.0004 3.17 (1.92– 5.23) ,0.0001

2.94 (1.66 –5.22) 0.0002 3.07 (1.86 –5.05) ,0.0001

............................................................................................................................................................................................................................................. ≤30 days

Efficacy CV death, myocardial infarction, stroke CV death

213 (9.9)

313 (7.7)

152 (7.4)

161 (8.0)

64 (3)

100 (2.5)

53 (2.5)

47 (2.4)

Cor. heart dis. death

36 (1.8)

52 (1.3)

24 (1.1)

28 (1.5)

Myocardial infarction

134 (6.2)

195 (4.9)

90 (4.6)

105 (5.2)

Stroke

44 (2.1)

56 (1.4)

28 (1.3)

28 (1.5)

Safety TIMI Major Bleeding

14 (0.7)

86 (2.5)

43 (2.4)

43 (2.6)

21 (1.1)

134 (4.0)

67 (3.8)

67 (4.1)

7 (0.4)

23 (0.7)

13 (0.7)

10 (0.8)

2 (0.1)

8 (0.3)

4 (0.2)

4 (0.4)

106 (4.9)

192 (4.7)

108 (5.1)

84 (4.3)

TIMI Major or Minor Bleeding Intracranial bleeding Fatal bleeding All-cause mortality

P2Y12 inhibitor withdrawal and ischaemic risk

Table 2

............................................................................................................................................................................................................................................. .30 days to ≤1 year

Efficacy CV death, myocardial infarction, stroke CV death

171 (8.7)

293 (7.6)

152 (8.1)

141 (7.2)

59 (3.1)

90 (2.5)

49 (2.8)

41 (2.2)

Cor. heart dis. death

37 (1.8)

52 (1.4)

26 (1.4)

26 (1.4)

Myocardial infarction

103 (5.2)

174 (4.4)

83 (4.2)

91 (4.6)

Stroke

34 (1.9)

62 (1.7)

34 (1.9)

28 (1.4)

Safety TIMI Major Bleeding

16 (1.2)

82 (2.7)

39 (2.8)

43 (2.7)

21 (1.6)

116 (3.8)

57 (3.9)

59 (3.8)

TIMI Major or Minor Bleeding

1139

Continued

1140

Table 2

Continued

Timing from P2Y12 inhibitor withdrawal

Endpoint

Placebo (n 5 6258) n (%)

Ticagrelor pooled (n 5 12 503) n (%)

Ticagrelor 90 mg bid (n 5 6245) n (%)

Ticagrelor 60 mg bid (n 5 6258) n (%)

HR (95% CI) P-value

.................................................................................

Ticagrelor pooled vs. placebo

Ticagrelor 90 mg bid vs. placebo

Ticagrelor 60 mg bid vs. placebo

1.81 (0.66 –4.95) 0.25 0.57 (0.16 –1.97) 0.37 0.88 (0.68 –1.14) 0.34

1.63 (0.52– 5.14) 0.40 0.24 (0.03– 2.04) 0.19 1.01 (0.75– 1.35) 0.96

1.97 (0.66 –5.88) 0.22 0.87 (0.23 –3.25) 0.84 0.76 (0.56 –1.04) 0.086

1.01 (0.80 –1.27) 0.96 1.3 (0.86– 1.97) 0.22 0.86 (0.52 –1.44) 0.58 0.9 (0.66– 1.21) 0.47 1.04 (0.62 –1.75) 0.88

0.96 (0.73– 1.26) 0.75 1.15 (0.71– 1.86) 0.57 0.77 (0.42– 1.43) 0.42 0.84 (0.59– 1.19) 0.33 1.13 (0.63– 2.03) 0.69

1.05 (0.81 –1.38) 0.70 1.45 (0.92 –2.29) 0.11 0.96 (0.53 –1.71) 0.88 0.95 (0.68 –1.34) 0.78 0.95 (0.51 –1.75) 0.87

2.67 (1.43 –4.98) 0.0020 2.94 (1.72 –5.02) ,0.0001 1.48 (0.53 –4.14) 0.46 0.57 (0.04 –9.11) 0.69 1.32 (0.97 –1.80) 0.076

3.28 (1.7– 6.33) 0.0004 3.82 (2.18– 6.68) ,0.0001 1.42 (0.43– 4.64) 0.57 NA

2.1 (1.05– 4.22) 0.037 2.11 (1.16 –3.87) 0.015 1.56 (0.49 –4.9) 0.45 1.11 (0.07 –17.7) 0.94 1.34 (0.94 –1.9) 0.10

............................................................................................................................................................................................................................................. Intracranial bleeding

5 (0.4)

16 (0.5)

7 (0.5)

9 (0.6)

Fatal bleeding

5 (0.5)

5 (0.2)

1 (0.1)

4 (0.3)

91 (4.9)

161 (4.3)

91 (4.8)

70 (3.8)

All-cause mortality

............................................................................................................................................................................................................................................. .1 year

Efficacy CV death, myocardial infarction, stroke CV death

103 (6.9)

209 (6.6)

100 (6.3)

109 (6.9)

31 (2.2)

81 (2.6)

36 (2.3)

45 (2.8)

Cor. heart dis. death

23 (1.6)

40 (1.2)

18 (1.1)

22 (1.2)

Myocardial infarction

67 (4.4)

121 (3.9)

57 (3.6)

64 (4.1)

Stroke

21 (1.4)

44 (1.3)

24 (1.5)

20 (1.2)

Safety TIMI Major Bleeding

12 (0.9)

57 (2.4)

34 (3)

23 (1.9)

16 (1.2)

84 (3.5)

53 (4.4)

31 (2.3)

5 (0.4)

13 (0.6)

6 (0.5)

7 (0.6)

TIMI Major or Minor Bleeding Intracranial bleeding Fatal bleeding All-cause mortality

1 (0)

1 (0)

0 (0)

1 (0.1)

55 (3.8)

147 (4.6)

73 (4.7)

74 (4.6)

M.P. Bonaca et al.

Bleeding analyses performed as on-treatment including patients who received at least one dose and events occurring within 7 days of the last dose.

1.31 (0.92– 1.86) 0.13

1141

P2Y12 inhibitor withdrawal and ischaemic risk

rather than re-initiating such therapy in patients who have remained stable off P2Y12 inhibitor therapy for an extended period. We also demonstrate a nearly 50% increased risk of MACE in stable outpatients with prior MI who have recently discontinued P2Y12 inhibition. The excess risk was apparent both early and late. Conversely, a subgroup of patients with prior MI who have survived event free on aspirin monotherapy for an extended period have de facto demonstrated themselves to be at relatively lower ischaemic risk. Several trials evaluating shorter durations of P2Y12 inhibitor therapy after coronary stenting have suggested that stopping earlier than 12 months may not be associated with increased ischaemic risk. However, the study populations were generally low-risk, with only a minority of patients presenting with an MI, and there were very few major cardiovascular events.4 – 8 In contrast, the far larger DAPT trial showed, akin to what we observed in PEGASUS-TIMI 54, that there was a particularly high risk of ischaemic events shortly after P2Y12 withdrawal.2 Taken together, these findings may be informative for clinicians when asked, for example, to consider interrupting P2Y12 inhibitor therapy for an elective procedure in patients with prior MI. In the current study, we found that patients treated with ticagrelor within 30 days of withdrawal from P2Y12 inhibitor therapy derived the greatest benefit with a 27% relative reduction in risk, a 2.2% absolute risk reduction, and therefore a number needed to treat over 3 years to prevent one CVD/MI/stroke event of 46. This group of patients may also be a population in which clinicians can most readily assess tolerability and adherence since they are currently or were recently exposed to P2Y12 inhibitor therapy. The findings from this group of patients are supported by observations in the subset of patients with MI in the DAPT trial, which randomized patients who had completed 1 year of dual antiplatelet therapy after stenting to continued therapy for another 18 months or placebo.9 In those patients treated with ticagrelor after a longer interruption (30 days to 1 year), the point estimate for the HR for MACE (both doses pooled) was 0.86, with 95% CIs ranging from 0.71 to 1.04. Interestingly, the magnitude of the relative risk reduction was numerically greater with the 60 mg dose relative to the 90 mg dose (18 vs. 10%) suggesting that perhaps the better tolerability of the lower dose in patients more temporally distant from P2Y12 inhibitor exposure may have resulted in greater adherence to therapy. Finally, those patients who re-initiated therapy after surviving event-free more than a year on aspirin alone derived no clear benefit from the addition of either dose, suggesting that this group, by nature of their demonstrated stability, is selected to have lower future risk. Thus, patients who have survived, event-free on aspirin monotherapy without an ischaemic complication more than 1 year from stopping P2Y12 inhibition and in general over 2 years from their MI appear to be a group at lower risk where re-initiation of therapy increases the risk of bleeding without an appreciable ischaemic risk reduction. There was no excess or reduction in all-cause mortality with ticagrelor overall or in any of the time groups presented. This observation is in contrast to the mortality reduction seen in the PLATO trial which randomized patients with ACS to ticagrelor or clopidogrel and observed a reduction in all-cause mortality. Therefore, clinicians considering a strategy of pronged ticagrelor therapy must weigh the reductions in ischaemic risk against increased bleeding risk in determining the optimal strategy for an individual patient.

There are limitations to the current analysis. The time from P2Y12 inhibitor withdrawal was not randomized and was associated with differences in clinical characteristics. Therefore, the differences in risk observed were may have been driven by differences in clinical characteristics rather than P2Y12 inhibitor withdrawal timing per se. Statistical adjustment was used to account for all key differences at baseline; however, a risk of residual confounding due to unmeasured variables remains. Of note though, our observations are consistent with the findings from the DAPT trial in which patients were at an increased risk of ischaemic events soon after protocolmandated discontinuation of DAPT, both in the 12-month and the 30-month treatment arms.2 It should be noted, however, that the comparisons of ticagrelor to placebo were randomized and therefore should be free of confounding. Nonetheless, in determining the optimal duration of DAPT, clinicians optimally should consider all risk factors both for ischaemic and bleeding events to determine which patients will receive the greatest net benefit. An additional limitation to this analysis was that 15% of patients had missing data on the timing of P2Y12 inhibitor discontinuation and these missing data add uncertainty to the findings. In addition, in this group, there was no significant benefit of ticagrelor and therefore the patients included in the manuscript may not be representative and could over-estimate the treatment effect observed. The current analyses are limited to evaluation of the safety and efficacy of ticagrelor based on timing from P2Y12 inhibition withdrawal primarily with clopidogrel. Therefore, these analyses are not able to provide information with regard to other agents or potential differences in the withdrawal or continuation of other agents, although we believe the principles to be generalizable based on observations from the DAPT trial.

Conclusions The benefit of ticagrelor for long-term secondary prevention in patients with prior MI and at least one additional risk factor appeared more marked in patients continuing on or re-starting after only a brief interruption of P2Y12 inhibition, when compared with patients who have proved themselves stable more than 2 years from their MI and off P2Y12 inhibitor therapy for more than a year. The increase in bleeding events with ticagrelor was similar regardless of this time interval. For clinicians considering a strategy of prolonged P2Y12 inhibitor therapy in high-risk patients, these data suggest greater benefit in the continuation of such therapy without interruption after MI, rather than re-initiating such therapy in patients who have remained stable for an extended period. Future analyses may help to clarify further the profile of post-MI patients most likely to benefit from uninterrupted DAPT, and those in whom bleeding might be increased without a reduction in ischaemic events.

Supplementary material Supplementary material is available at European Heart Journal online.

Authors’ contributions K.I.: performed statistical analysis. E.B. and M.S.S.: handled funding and supervision. M.P.B., D.L.B., P.G.S., R.F.S., M.C., K.I., T.O.O.,

1142 A.B., S.G., J.L.-S., R.D., A.D., F.V.D.W., D.A., G.M., P.A., G.M., E.C.J., P.H., E.B., and M.S.S.: acquired the data, and conceived and designed the research. M.B. and M.S.: drafted the manuscript. D.L.B., P.G.S., R.F.S., M.C., K.I., T.O.O., A.B., S.G., J.L.-S., R.D., A.D., F.V.D.W., D.A., G.M., P.A., G.M., E.C.J., P.H., and E.B.: made critical revision of the manuscript for key intellectual content.

Funding This study was supported by a grant from AstraZeneca. Conflict of interest: The TIMI Study Group has received significant research grant support from AstraZeneca. M.P.B. reports consulting fees from AstraZeneca, Merck, Bayer, and Roche Diagnostics. D.L.B. reports the following relationships—Advisory Board: Cardax, Elsevier Practice Update Cardiology, Medscape Cardiology, Regado Biosciences; Board of Directors: Boston VA Research Institute, Society of Cardiovascular Patient Care; Chair: American Heart Association Get With The Guidelines Steering Committee; Data Monitoring Committees: Duke Clinical Research Institute, Harvard Clinical Research Institute, Mayo Clinic, Population Health Research Institute; Honoraria: American College of Cardiology (Senior Associate Editor, Clinical Trials and News, ACC.org), Belvoir Publications (Editor in Chief, Harvard Heart Letter), Duke Clinical Research Institute (clinical trial steering committees), Harvard Clinical Research Institute (clinical trial steering committee), HMP Communications (Editor in Chief, Journal of Invasive Cardiology), Journal of the American College of Cardiology (Guest Editor; Associate Editor), Population Health Research Institute (clinical trial steering committee), Slack Publications (Chief Medical Editor, Cardiology Today’s Intervention), WebMD (CME steering committees); Other: Clinical Cardiology (Deputy Editor); Research Funding: Amarin, AstraZeneca, Bristol-Myers Squibb, Eisai, Ethicon, Forest Laboratories, Ischemix, Medtronic, Pfizer, Roche, Sanofi-Aventis, The Medicines Company; Site Co-Investigator: Biotronik, St. Jude Medical; Trustee: American College of Cardiology; Unfunded Research: FlowCo, PLx Pharma, Takeda. P.G.S. reports significant research grants from Sanofi and Servier; other personal fees and non-financial support from AstraZeneca, Sanofi, Servier; personal fees from Amarin, Bayer, Boehringer Ingelheim, Bristol-Myers Squibb, CSL-Behring, Daiichi-Sankyo, Lilly, Merck, Janssen, Novartis, Medtronic, Pfizer, The Medicines Company, and GSK. R.F.S. reports grants and personal fees from AstraZeneca, during the conduct of the study; grants, personal fees, and other from AstraZeneca, personal fees from Aspen, personal fees from PlaqueTec, personal fees from The Medicines Company. M.C. reports grants and personal fees from AstraZeneca, during the conduct of the study; personal fees from Merck, personal fees from Janssen, personal fees from Maquet, personal fees from malpractice attorneys, grants from Janssen, grants from Edwards, personal fees from Merck, personal fees from BMS/Pfizer, personal fees from Janssen, personal fees from BI, personal fees from Lilly, outside the submitted work. T.O.O. reports speaker fees and consultancy fees from Astra Zeneca. A.B. reports grants and personal fees from AstraZeneca, during the conduct of the study; grants and personal fees from AstraZeneca, grants from Sanofi-Aventis, grants and personal fees from Bristol-Myers Squibb/Pfizer, grants from Boehringer Ingelheim, grants from Novartis, grants and personal fees from GlaxoSmithKline, grants from Eisai, outside the submitted work. S.G. reports Research grant modest Pfizer, significant Sanofi, speakers bureau BMS/Pfeizer, honoraria significant: Sanofi, AstraZeneca, modest: Eisai, Otsuka, Bayer, Novartis, Asteras, Pfizer, Medtronics-Japan, TanabeMitsubishi, Takeda, Daiichi-Sankyo, Mochida, MSD, consult: BMS, Armethron, Bayer. J.L.-S. reports research grants, honoraia: Astra Zeneca, Lilly, Daichi Sankio. A.D. reports serving on South African advisory

M.P. Bonaca et al.

boards for the following: Aspen, AstraZeneca, Bayer, Boehringer Ingelheim, Novartis, Sanofi, and Servier. He also reports receiving honoraria from AstraZeneca, and Servier and travel sponsorship from Bayer, Boehringer Ingelheim, Novartis, and Sanofi. G.M. reports consulting fees from AstraZeneca. E.C.J. is an employee of AstraZeneca. P.H. is an employee of AstraZeneca. E.B. reports grant support to institution from AstraZeneca. M.S.S. reports research grant support through Brigham and Women’s Hospital from: Abbott Laboratories; Accumetrics; Amgen; AstraZeneca; Bristol-Myers Squibb; Critical Diagnostics; Daiichi-Sankyo; Eisai; Genzyme; Gilead; GlaxoSmithKline; Intarcia; Merck; Nanosphere; Roche Diagnostics; Sanofi-aventis; and Takeda, as well as consulting for: Aegerion; Alnylam; Amgen; AstraZeneca; BristolMyers Squibb; Cubist; CVS Caremark; Intarcia; Merck; MyoKardia; Pfizer; Quest Diagnostics; Sanofi-Aventis; Vertex; and Zeus Scientific.

References 1. Bonaca MP, Bhatt DL, Cohen M, Steg PG, Storey RF, Jensen EC, Magnani G, Bansilal S, Fish MP, Im K, Bengtsson O, Oude Ophuis T, Budaj A, Theroux P, Ruda M, Hamm C, Goto S, Spinar J, Nicolau JC, Kiss RG, Murphy SA, Wiviott SD, Held P, Braunwald E, Sabatine MS, PEGASUS-TIMI 54 Steering Committee and Investigators. Long-term use of ticagrelor in patients with prior myocardial infarction. N Engl J Med 2015;372:1791 –1800. 2. Mauri L, Kereiakes DJ, Yeh RW, Driscoll-Shempp P, Cutlip DE, Steg PG, Normand SL, Braunwald E, Wiviott SD, Cohen DJ, Holmes DR Jr, Krucoff MW, Hermiller J, Dauerman HL, Simon DI, Kandzari DE, Garratt KN, Lee DP, Pow TK, Ver Lee P, Rinaldi MJ, Massaro JM, DAPT Study Investigators. Twelve or 30 months of dual antiplatelet therapy after drug-eluting stents. N Engl J Med 2014;371: 2155 –2166. 3. Bonaca MP, Bhatt DL, Braunwald E, Cohen M, Steg PG, Storey RF, Held P, Jensen EC, Sabatine MS. Design and rationale for the prevention of cardiovascular events in patients with prior heart attack using ticagrelor compared to placebo on a background of aspirin-thrombolysis in myocardial infarction 54 (PEGASUS-TIMI 54) trial. Am Heart J 2014;167:437 –444.e5. 4. Feres F, Costa RA, Abizaid A, Leon MB, Marin-Neto JA, Botelho RV, King SB III, Negoita M, Liu M, de Paula JE, Mangione JA, Meireles GX, Castello HJ Jr, Nicolela EL Jr, Perin MA, Devito FS, Labrunie A, Salvadori D Jr, Gusmao M, Staico R, Costa JR Jr, de Castro JP, Abizaid AS, Bhatt DL, OPTIMIZE Trial Investigators. Three vs twelve months of dual antiplatelet therapy after zotarolimus-eluting stents: the OPTIMIZE randomized trial. JAMA 2013;310:2510 –2522. 5. Kim BK, Hong MK, Shin DH, Nam CM, Kim JS, Ko YG, Choi D, Kang TS, Park BE, Kang WC, Lee SH, Yoon JH, Hong BK, Kwon HM, Jang Y, RESET Investigators. A new strategy for discontinuation of dual antiplatelet therapy: the RESET Trial (REal Safety and Efficacy of 3-month dual antiplatelet Therapy following Endeavor zotarolimus-eluting stent implantation). J Am Coll Cardiol 2012;60:1340 –1348. 6. Gwon HC, Hahn JY, Park KW, Song YB, Chae IH, Lim DS, Han KR, Choi JH, Choi SH, Kang HJ, Koo BK, Ahn T, Yoon JH, Jeong MH, Hong TJ, Chung WY, Choi YJ, Hur SH, Kwon HM, Jeon DW, Kim BO, Park SH, Lee NH, Jeon HK, Jang Y, Kim HS. Six-month versus 12-month dual antiplatelet therapy after implantation of drug-eluting stents: the Efficacy of Xience/Promus Versus Cypher to Reduce Late Loss After Stenting (EXCELLENT) randomized, multicenter study. Circulation 2012;125:505 –513. 7. Colombo A, Chieffo A, Frasheri A, Garbo R, Masotti-Centol M, Salvatella N, Oteo Dominguez JF, Steffanon L, Tarantini G, Presbitero P, Menozzi A, Pucci E, Mauri J, Cesana BM, Giustino G, Sardella G. Second-generation drug-eluting stent implantation followed by 6- versus 12-month dual antiplatelet therapy: the SECURITY randomized clinical trial. J Am Coll Cardiol 2014;64:2086 –2097. 8. Schulz-Schupke S, Byrne RA, Ten Berg JM, Neumann FJ, Han Y, Adriaenssens T, Tolg R, Seyfarth M, Maeng M, Zrenner B, Jacobshagen C, Mudra H, von Hodenberg E, Wohrle J, Angiolillo DJ, von Merzljak B, Rifatov N, Kufner S, Morath T, Feuchtenberger A, Ibrahim T, Janssen PW, Valina C, Li Y, Desmet W, Abdel-Wahab M, Tiroch K, Hengstenberg C, Bernlochner I, Fischer M, Schunkert H, Laugwitz KL, Schomig A, Mehilli J, Kastrati A, Intracoronary Stenting and Antithrombotic Regimen: Safety And EFficacy of 6 Months Dual Antiplatelet Therapy After Drug-Eluting Stenting (ISAR-SAFE) Trial Investigators. ISAR-SAFE: a randomized, double-blind, placebo-controlled trial of 6 vs. 12 months of clopidogrel therapy after drug-eluting stenting. Eur Heart J 2015;36:1252 –1263. 9. Yeh RW, Kereiakes DJ, Steg PG, Windecker S, Rinaldi MJ, Gershlick AH, Cutlip DE, Cohen DJ, Tanguay JF, Jacobs A, Wiviott SD, Massaro JM, Iancu AC, Mauri L, DAPT Study Investigators. Benefits and risks of extended duration dual antiplatelet therapy after PCI in patients with and without acute myocardial infarction. J Am Coll Cardiol 2015;65:2211 –2221.