Aspirin therapy in primary cardiovascular disease prevention A position paper of the ESC Working Group Thrombosis

Halvorsen S. et al., Online Supplemental Material, page 1 Aspirin therapy in primary cardiovascular disease prevention A position paper of the ESC Working Group Thrombosis Short title: aspirin in primary prevention Sigrun Halvorsen Department of Cardiology, Oslo University Hospital Ulleval, Oslo, Norway - sigrun.h@online.no Felicita Andreotti - Department of Cardiovascular Science, Catholic University, Rome, Italy felicita.andreotti@iol.it Jurriën M. ten Berg - Department of Cardiology, St Antonius Hospital, Nieuwegein, the Netherlands - jurtenberg@gmail.com Marco Cattaneo - Medicina 3, Ospedale San Paolo. Dipartimento di Scienze della Salute, Università degli Studi di Milano. Milan, Italy - marco.cattaneo@unimi.it Sergio Coccheri - Department of Cardiovascular Disease, University of Bologna, Italy coccheris.angio@libero.it Roberto Marchioli Quintiles Pescara, Italy - marchioli.roberto@gmail.com Joao Morais - Santo Andre s Hospital, Leiria, Portugal - jaraujomorais@gmail.com Freek W.A. Verheugt, Department of Cardiology, Heartcenter, Onze Lieve Vrouwe Gasthuis (OLVG), Amsterdam, Netherlands - f.w.a.verheugt@olvg.nl Raffaele de Caterina - Institute of Cardiology, G. D Annunzio University Chieti, Italy rdecater@unich.it Correspondence: Raffaele De Caterina, MD, PhD Institute of Cardiology C/o Ospedale SS. Annunziata Via dei Vestini 66013 Chieti, Italy Tel. +39-0871-41512 FAX: +39-0871-402817 E-mail: rdecater@unich.it ONLINE SUPPLEMENTAL MATERIAL 1

Halvorsen S. et al., Online Supplemental Material, page 2 Oral anticoagulants in primary prevention Only one proper randomized trial, the Thrombosis Prevention Trial (TPT) (1), has been carried out in primary prevention of associated with atherothrombosis with oral anticoagulants (vitamin K antagonists). In British general practice, 5,085 healthy males aged 45 to 69 and at the highest quintile of CV risk were randomised to double-blind warfarin (target INR 1.5), low-dose aspirin (75 mg daily), both or neither, and followed for nearly 7 years. Warfarin without aspirin reduced all ischemic heart disease from 13.3 to 10.3 per 1000 person-years compared to those receiving placebo (22% reduction, 95% confidence interval (CI) -4 to 41). This reduction was of the same magnitude as that obtained with aspirin. Major bleeding was seen in 9 men on warfarin and 4 on placebo. Because of this increase in bleeding, vitamin K antagonists are generally not recommended in primary prevention, and will not be discussed further in this review. The efficacy of aspirin for primary prevention of The Physicians Health Study (PHS) randomly allocated, in a double-blind fashion, 22,071 male physicians 40 years of age or older, to receive 325 mg of aspirin every other day or placebo (2). The average follow-up amounted to 60 months. The risk of MI was reduced by 44% in the aspirin group compared to placebo (relativ risk, 0.56; 95% CI, 0.45 to 0.70; P<0.00001). The risk of stroke was slightly increased numerically, but not significantly (P = 0.15). In the non-blinded British Doctors Trial (BDT), 5,139 male physicians under the age of 80 received aspirin 500 or 300 mg daily or no treatment during 6 years (3). Treatment with aspirin did not influence the incidence of MI: 5.0% for aspirin and 5.2% for no aspirin (relative risk reduction 4%, 95% CI 24 to 27%, P = N.S.). Total mortality, however, was 14% lower in the aspirin group (P = N.S.). In the Thrombosis Prevention Trial (TPT) aspirin reduced the incidence of new ischaemic heart disease from 13.3 to 10.2 per 1000 person-years compared to placebo (23% reduction, 95% CI -3 to 42%) (1). In the Hypertension Optimal Treatment (HOT) trial, 18,790 male and female individuals aged 50 to 80 years old and with well-managed hypertension were randomised to 75 mg aspirin daily or placebo (4). After nearly 4 years, MI was reduced by 36% (95% CI 15 to 51%) from 1.4 to 0.9% (4). 2

Halvorsen S. et al., Online Supplemental Material, page 3 The Primary Prevention Project (PPP) (5) was carried out in general practice in Italy to establish the role of aspirin in primary prevention, but was prematurely halted because of the publication of other positive studies (1,4). In 4,495 healthy males and females over the age of 65, aspirin 100 mg daily or placebo was given for 3 years. Aspirin reduced the rate of MI by 25% (95% CI -26 to 55%) (5). The Womens Health Study (WHS) randomly allocated, in a double-blind fashion, 39,876 women 45 years of age or older to 100 mg of aspirin every other day or placebo (6). The average follow-up amounted to 10 years. A reduction of the risk of MI was not found: 1.0 vs 1.0% (relative risk 1.02, 95% CI 0.84 to 1.25, P = 0.83). The risk of stroke was reduced by 17% (relative risk 0.83, 95% CI 0.69 to 0.99, P = 0.04) (6). In the Prevention of Progression of Arterial Disease and Diabetes (POPADAD) trial (7), 1276 individuals with type 1 or type 2 diabetes and an ankle brachial pressure index of 0.99 or less but no symptomatic CV disease were randomised to aspirin or placebo for a median follow-up time of 6.7 years. The authors used two hierarchial composite primary end points: death from coronary heart disease or stroke, non-fatal MI or stroke, or above-ankle amputation for critical limb ischaemia; and death from coronary heart disease or stroke. There was no significant difference between groups in the occurrence of either of the primary end points: 18.2% on aspirin vs 18.3% on placebo (P=0.86), and 6.7% on aspirin vs 5.5% on placebo (P=0.36), respectively (7). The Japanese Primary Prevention of Atherosclerosis with Aspirin for Diabetes (JPAD) (8) trial randomised 2539 patients with type 2 diabetes, without a history of atherosclerotic disease, to aspirin 81- or placebo. After a median follow-up of 4.4 years, 5.4% of patients in the aspirin group vs 6.7% in the non-aspirin group experienced an atherothrombotic event (HR 0.80, 95% CI 0.58 1.10, P=0.16). The combined end point of fatal coronary events and fatal cerebrovascular events was significantly reduced in the aspirin group (HR 0.10; 95% CI, 0.01-0.79; P=0.0037). Finally, in the Aspirin for Asymptomatic Atherosclerosis Trialists study (AAA) (9), 3350 individuals with a low ankle brachial pressure index ( 0.95) but without clinical CV disease, were randomly assigned to 100 mg aspirin once daily or placebo. The primary end point was a composite of initial fatal or non-fatal coronary event or stroke or revascularisation. Two secondary end points were (1) all initial vascular events defined as a composite of a primary end point event or angina, intermittent 3

Halvorsen S. et al., Online Supplemental Material, page 4 claudication, or transient ischemic attack; and (2) all-cause mortality. After a mean follow-up time of 8.2 years, no statistically significant difference in the occurrence of the primary end point events (13.7 events per 1000 person-years in the aspirin group vs 13.3 in the placebo group; HR 1.03; 95% CI 0.84-1.27) or all-cause mortality was found between the groups. An additional, now remote trial, the Early Treatment Diabetic Retinopathy Study (ETDRS) report 14 (10), is sometimes considered for aspirin in the setting of primary prevention. In this multicenter, randomised clinical trial of aspirin vs placebo, 3711 men and women between the ages of 18 and 70 years with a clinical diagnosis of diabetes mellitus were eligible (30% considered to have type 1 diabetes, 31% having type 2, and 39% a type which could not be determined definitely). Forty-nine % of subjects recruited in the aspirin arm and 48.5% in the placebo arm reported a history of, but this included the presence of hypertension or the use of diuretics and other anti-hypertensive agents, and no separate analysis for truly primary and secondary prevention (as defined above) was available. Patients were here randomly assigned to aspirin (two 325-mg tablets once per day) or placebo. The main outcome measure was mortality from all causes. Other outcome variables included cause-specific mortality and CV events. Here the estimate of relative risk for total mortality for aspirin patients compared with placebo- patients for the entire study period (an average followup of 5 years, range 4 to 9) was 0.91 (99% confidence interval, 0.75 to 1.11). Larger differences, but still not statistically significant, were noted for the occurrence of fatal and non-fatal MI, for which the estimate of relative risk was 0.83 for the entire follow-up period (99% confidence interval, 0.66 to 1.04). The authors concluded that the effects of aspirin on any of the cardiovascular events were not substantially different from the effects observed in other previous studies that included mainly nondiabetic persons, without evidence of harmful effects of aspirin, thus supporting, according to the authors, the use of aspirin for persons with diabetes at increased risk of (10). This study was not included in the present analysis (similarly as in other meta-analyses, listed below), because no separate data for primary and secondary prevention were here available. 4

Halvorsen S. et al., Online Supplemental Material, page 5 Extended Legend to Figure 1: Relationships between magnitude of antithrombotic benefit and of bleeding risk connected with the use of aspirin (on the ordinate) and absolute cardiovascular risk (on the abscissa), in various subsets of subjects in primary prevention. The magnitude of antithrombotic benefit and of bleeding risk connected with the use of aspirin are plotted on the ordinate as a function of the absolute cardiovascular risk (on the abscissa), in various subsets of subjects in primary prevention. The commonly adopted approach of computer-aided literature search was applied to retrieve all published RCTs. Nine trials have been included in the present analysis, with the following criteria for inclusion: (1) designed aspirin treatment in the primary prevention of cardiovascular disease, (2) exclusion of patients with prior myocardail infarction or cerebrovascular stroke, (2) randomized design, (3) recorded incidence of cardiovacular events, (4) at least 1,000 patients included in the trial and (5) at least 12 months of follow-up Trials included: Trials included: U.S. Physicians Health Study-PHS (2), British Doctors Trial-BDT (3), Thrombosis Prevention Trial-TPT (1), HOT study (4), Primary Prevention Project-PPP (5), Women s Health Study (6), POPAPAD Study (7), JPAD Study (8), and Aspirin for Asymptomatic Atherosclerosis (AAA) trial (9). Data from the Swedish Angina Pectoris Aspirin Trial (SAPAT) of aspirin in stable angina (11), are also included as reference to the lowest published cardiovascular (CV) risk category in secondary CV prevention. To examine the strength of the association between treatment effects of aspirin on CV events, major gastrointestinal bleeding, and total major bleeding with the level of CV risk per 100 person-years in the control arm of the trials, we fitted univariate inverse variance-weighted linear regressions of the risk difference for the outcome events per 100 person-years between the two experimental arms of each study as a dependent variable against the aforementioned explanatory variable. Results have been reported together with their 95% confidence intervals (CIs). Adjustment for daily aspirin dose tested in the trials did not materially change the results. Size of circles is proportional to the inverse of variance of the risk difference. All analyses were carried out with a specifically designed software written in SAS language. Notice that the risk of major gastro-intestinal bleeding events requiring transfusion (purple line) in the various studies, although increasing as a function of cardiovascular risk as pointed out in the ATT Collaboration meta-analysis (12), is less dependent on the baseline CV risk than the absolute CV benefit (red line) see the different slopes of the two regression lines. These estimates can be used to 5

Halvorsen S. et al., Online Supplemental Material, page 6 identify a threshold value of absolute CV risk below which the use of aspirin in primary CV prevention is unattractive, since associated with a quite uncertain benefit-risk ratio. Equally, an absolute CV risk above 2 major cardiovascular events / 100 is an area where the separation of the regression lines depicting the benefit (red line) and the risk (purple and green lines) separate considerably, defining an area of theoretical net clinical benefit despite the prescription uncertainty deriving from the lack of trials. Red arrow denotes the area where benefit likely equals risk, yellow area denotes area of prescription uncertainty, and green arrow denotes the area where benefit most likely exceeds risk. Abbreviations and notes: Absolute Risk Change is absolute risk reduction for CV events and absolute risk increase for major gastrointestinal bleeds and total major bleeds. Continuous Line = Linear Regression; Dotted line = Lower and Higher 95% CI. CV = Cardiovascular; GI = Gastrointestinal; CI = Confidence Interval; ASA = Acetylsalicylic Acid (aspirin). 6

Halvorsen S. et al., Online Supplemental Material, page 7 e 1: Primary relevant data of currently available primary prevention trials with aspirin in cardiovascular disease ^ Reference Mean duration of follow-up (years) ASPIRIN / year Major cardiovascular events * CONTROLᶲ / year Absolute risk reduction (placebo - aspirin) NNT (with decimals) NNT (integer number) 3 5,5 289 3429 0,0843 1,5324 0,0153238 147 1710 0,0860 1,5630 0,0156300 0,0003061 3266,46 3266 2 5 307 11037 0,0278 0,5563 0,0055631 370 11034 0,0335 0,6707 0,0067065 0,0011434 874,56 875 4 3,8 388 9399 0,0413 1,0863 0,0108634 425 9391 0,0453 1,1909 0,0119095 0,0010461 955,95 956 1 6,4 228 2545 0,0896 1,3998 0,0139980 260 2540 0,1024 1,5994 0,0159941 0,0019961 500,99 501 5 3,6 45 2226 0,0202 0,5615 0,0056155 64 2269 0,0282 0,7835 0,0078351 0,0022196 450,53 451 6 10,1 477 19934 0,0239 0,2369 0,0023692 522 19942 0,0262 0,2592 0,0025917 0,0002225 4494,99 4495 7 6,7 105 638 0,1646 2,4564 0,0245637 108 638 0,1693 2,5266 0,0252655 0,0007018 1424,87 1425 8 4,37 68 1262 0,0539 1,2330 0,0123301 86 1277 0,0673 1,5411 0,0154108 0,0030807 324,60 325 9 8,2 181 1675 0,1081 1,3178 0,0131780 176 1675 0,1051 1,2814 0,0128140-0,0003640-2747,00-2747 11 4,2 111 1009 0,1100 2,6193 0,0261928 159 1026 0,1550 3,6898 0,0368978 0,0107050 93,41 93 Reference Mean duration of follow-up (years) / year / year Absolute excess of major bleeding due to aspirin (aspirinplacebo) NNH (with decimals) NNH (integer numbers) 3 5,5 29 3429 0,0085 0,1538 0,0015377 7 1710 0,0041 0,0744 0,0007443 0,0007934 1260,40 1260 2 5 48 11037 0,0043 0,0870 0,0008698 28 11034 0,0025 0,0508 0,0005075 0,0003623 2760,30 2760 4 3,8 130 9399 0,0138 0,3640 0,0036398 75 9391 0,0080 0,2102 0,0021017 0,0015381 650,14 650 1 6,4 20 2545 0,0079 0,1228 0,0012279 13 2540 0,0051 0,0800 0,0007997 0,0004282 2335,39 2335 5 3,6 24 2226 0,0108 0,2995 0,0029949 6 2269 0,0026 0,0735 0,0007345 0,0022604 442,41 442 6 10,1 178 19934 0,0089 0,0884 0,0008841 132 19942 0,0066 0,0655 0,0006554 0,0002287 4371,78 4372 7 6,7 30 638 0,0470 0,7018 0,0070182 34 638 0,0533 0,7954 0,0079540-0,0009358-1068,65-1069 8 4,37 17 1262 0,0135 0,3083 0,0030825 7 1277 0,0055 0,1254 0,0012544 0,0018282 547,00 547 9 8,2 34 1675 0,0203 0,2475 0,0024754 20 1675 0,0119 0,1456 0,0014561 0,0010193 981,07 981 11 4,2 20 1009 0,0198 0,4719 0,0047194 13 1026 0,0127 0,3017 0,0030168 0,0017026 587,33 587 Reference Mean duration of follow-up (years) ASPIRIN ASPIRIN / year Total major bleeding CONTROLᶲ Major gastrointestinal bleeding CONTROLᶲ / year Absolute excess of major GI bleeding due to aspirin (aspirin-placebo) NNH (with decimals) NNH (integer numbers) 3 5,5 20 3429 0,0058 0,1060 0,0010605 7 1710 0,0041 0,0744 0,0007443 0,0003162 3162,67 3163 2 5 11037 0,0000 0,0000 0,0000000 11034 0,0000 0,0000 0,0000000 0,0000000 missing missing 4 3,8 77 9399 0,0082 0,2156 0,0021559 37 9391 0,0039 0,1037 0,0010368 0,0011191 893,61 894 1 6,4 15 2545 0,0059 0,0921 0,0009209 11 2540 0,0043 0,0677 0,0006767 0,0002443 4094,16 4094 5 3,6 17 2226 0,0076 0,2121 0,0021214 5 2269 0,0022 0,0612 0,0006121 0,0015093 662,57 663 6 10,1 127 19934 0,0064 0,0631 0,0006308 91 19942 0,0046 0,0452 0,0004518 0,0001790 5586,93 5587 7 6,7 28 638 0,0439 0,6550 0,0065503 31 638 0,0486 0,7252 0,0072521-0,0007018-1424,87-1425 8 4,37 4 1262 0,0032 0,0725 0,0007253 0 1277 0,0000 0,0000 0,0000000 0,0007253 1378,74 1379 9 8,2 9 1675 0,0054 0,0655 0,0006553 8 1675 0,0048 0,0582 0,0005825 0,0000728 13735,00 13735 11 4,2 11 1009 0,0109 0,2596 0,0025957 6 1026 0,0058 0,1392 0,0013924 0,0012033 831,04 831 Reference Mean duration of follow-up (years) ASPIRIN / year / year Absolute excess of hemorrhagic stroke due to aspirin (aspirin-placebo) NNH (with decimals) NNH (integer numbers) 3 5,5 10 3429 0,0029 0,0001 0,0000009 4 1710 0,0023 0,0001 0,0000007 0,0000002 5659126,76 5659127 2 5 23 11037 0,0021 0,0002 0,0000024 12 11034 0,0011 0,0001 0,0000012 0,0000011 878203,66 878204 4 3,8 NA 9399 - - - NA 9391 - - - - - - 1 6,4 14 2545 0,0055 0,0011 0,0000110 6 2540 0,0024 0,0005 0,0000047 0,0000063 159616,28 159616 5 3,6 2 2226 0,0009 0,0002 0,0000020 3 2269 0,0013 0,0003 0,0000029-0,0000009-1064443,03-1064443 6 10,1 51 19934 0,0026 0,0001 0,0000006 41 19942 0,0021 0,0000 0,0000005 0,0000001 8945619,51 8945620 7 6,7 0 638 0,0000 0,0000 0,0000000 2 638 0,0031 0,0002 0,0000022-0,0000022-454575,00-454575 8 4,37 NA 1262 - - - NA 1277 - - - - - - 9 8,2 5 1675 0,0030 0,0001 0,0000018 4 1675 0,0024 0,0001-0,0000009 0,0000027 377149,59 377150 11 4,2 NA 1009 - - - NA 1026 - - - - - - ainable: GI: gastrointestinal; NNT: number needed to treat; NNH: number needed to harm AT trial in secondary prevention as a reference myocardial infarction, non-fatal stroke exception of the British Doctors Trial where no drug given in the control arm t on any prespecified endpoint" min E nor treatment with beta carotene significantly modified the effect of Aspirin on the endpiont" 50 mounths = 4,2 years 60,2 mounths = 5 years Hemorrhagic stroke CONTROLᶲ 7

Halvorsen S. et al., Online Supplemental Material, page 8 Online Supplemental Table 2: Ongoing randomized controlled trials of aspirin in the primary prevention of CV events. Individuals without ARRIVE Aspirin to Reduce Risk of Initial Vascular Event ASPREE Aspirin in Reducing in the Elderly JPPP Japanese Primary Prevention Project ASCEND A Study of Cardiovascu lar in Diabetes ACCEPT-D Aspirin and simvastatin Combinatio n for CV Event Prevention Trial in Diabetes ENVIS-ion Elderly NeuroVascu lar Imaging Study (ASPREE Substudy) DESIGN 1:1 Double blind control 1:1 Double blind control Open label Controlled Blind adjudication Double blind control 2x2 Factorial Open label Controlled 1:1 Double blind control NUMBER of SUBJECTS ~ 12,000 No known Without diabetes ~ 15,000 No overt AGE RISK FACTORS Men > 55 y with 2-4 RFs Women > 60 y with > 3 RFs Men and women > 70 y Multiple RFs 14,466 Men and women 60-85 y with HT, diabetes or dyslipidemia 15,480 Type I and II diabetes No overt 5,170 Type I and II diabetes on, or eligible for, statin therapy No overt ~ 600 No overt Normal cognitive function Men and women > 40 y With diabetes Men and women > 50 y With diabetes on statin therapy Men and women > 70 y Multiple RFs ASA DOSE CONTROL No ASA No ASA OTHER THERAPIES Omega-3 fatty acids 1 g daily or olive oil placebo Simvastatin 20 mg daily in all for LDL-C > 100 mg/dl PRIMARY EFFICACY END POINT Nonfatal MI Nonfatal stroke CV death Major adverse CV events and dementia Nonfatal MI Nonfatal stroke CV death Nonfatal MI Nonfatal stroke* CV death Nonfatal MI Nonfatal stroke CV death Hospital admission for CV cause Brain lesions at magnetic resonance imaging FOLLOW-UP START / END ~ 5 years Event driven ~ 5 years ~ 4 years Enrollment 2005-2007 ~ 7.5 years Start 2005 End ~ 2017 Event driven ~ 3 years Start 2008 * excluding confirmed cerebral hemorrhage ASA=aspirin or acetylsalicylic acid; CV=cardiovascular; =cardiovascular disease; diabetes=diabetes mellitus; EP=endpoint; HT=hypertension; LDL-C=low density lipoprotein cholesterol; MI=myocardial infarction; N=number; po=per os; preop.=preoperatively; postop.=postoperatively; RFs= cardiovascular risk factors; y=years. 8

Halvorsen S. et al., Online Supplemental Material, page 9 REFERENCES 1. The Medical Research Council's General Practice Research Framework. Thrombosis Prevention Trial: randomised trial of low-intensity oral anticoagulation with warfarin and low-dose aspirin in the primary prevention of ischaemic heart disease in men at increased risk. Lancet 1998;351:233-41. 2. Steering Committee of the Physicians' Health Study Research Group. Final report on the aspirin component of the ongoing Physicians' Health Study. Steering Committee of the Physicians' Health Study Research Group. N Engl J Med 1989;321:129-35. 3. Peto R, Gray R, Collins R et al. trial of prophylactic daily aspirin in British male doctors. Br Med J (Clin Res Ed) 1988;296:313-6. 4. Hansson L, Zanchetti A, Carruthers SG et al. Effects of intensive blood-pressure lowering and low-dose aspirin in patients with hypertension: principal results of the Hypertension Optimal Treatment (HOT) randomised trial. HOT Study Group. Lancet 1998;351:1755-62. 5. Primary Prevention Project Investigators. Low-dose aspirin and vitamin E in people at cardiovascular risk: a randomised trial in general practice. Collaborative Group of the Primary Prevention Project. Lancet 2001;357:89-95. 6. Ridker PM, Cook NR, Lee IM et al. A randomized trial of low-dose aspirin in the primary prevention of cardiovascular disease in women. N Engl J Med 2005;352:1293-304. 7. Belch J, MacCuish A, Campbell I et al. The prevention of progression of arterial disease and diabetes (POPADAD) trial: factorial randomised placebo controlled trial of aspirin and antioxidants in patients with diabetes and asymptomatic peripheral arterial disease. BMJ 2008;337:a1840. 8. Ogawa H, Nakayama M, Morimoto T et al. Low-dose aspirin for primary prevention of atherosclerotic events in patients with type 2 diabetes: a randomized controlled trial. JAMA 2008;300:2134-41. 9. Fowkes FG, Price JF, Stewart MC et al. Aspirin for prevention of cardiovascular events in a general population screened for a low ankle brachial index: a randomized controlled trial. JAMA 2010;303:841-8. 10. ETDRS Investigators. Aspirin effects on mortality and morbidity in patients with diabetes mellitus. Early Treatment Diabetic Retinopathy Study report 14. JAMA 1992;268:1292-300. 11. Juul-Moller S, Edvardsson N, Jahnmatz B et al. Double-blind trial of aspirin in primary prevention of myocardial infarction in patients with stable chronic angina pectoris. Lancet 1992;340:1421-1425. 12. Baigent C, Blackwell L, Collins R et al. Aspirin in the primary and secondary prevention of vascular disease: collaborative meta-analysis of individual participant data from randomised trials. Lancet 2009;373:1849-60. 9