DATE: 06 June 2012 CONTEXT AND POLICY ISSUES

Transcription

1 TITLE: Clopidogrel, Prasugrel and Ticagrelor in Adults with Acute Coronary Syndrome: A Review of the Clinical Effectiveness, Cost Effectiveness and Guidelines DATE: 06 June 2012 CONTEXT AND POLICY ISSUES Acute coronary syndrome (ACS) is a collective term used to describe the acute onset of myocardial ischemia resulting from the occlusion of coronary arteries and includes the following designations: ST segment elevation myocardial infarction (STEMI), non-st segment elevation myocardial infarction (NSTEMI) and unstable angina (UA). 1 When patients present with signs and symptoms of ACS, they receive an immediate oral administration of dual antiplatelet therapy consisting of acetylsalicylic acid (aspirin; ASA) and a thienopyridine. 2 Thienopyridines are a class of P2Y 12 platelet adenosine diphosophate (ADP) receptor antagonists and the one that is most commonly administered to ACS patients is clopidogrel. 3 ACS (STEMI, or NSTEMI/UA) is classified based on the characteristics of electrocardiograms (ECG) and the level of cardiac enzymes circulating in the blood. 4 STEMI patients receive immediate care, which may include percutaneous coronary intervention (PCI), coronary artery bypass grafting (CABG), or management by drug alone. 5,6 NSTEMI patients receive similar treatment options, but the administration of these treatments are less time-critical than in STEMI patients. 6,7 Currently, dual therapy with clopidogrel and ASA is the standard treatment for ACS patients in Canada. 6 Clopidogrel has many limitations including a slow onset of action, incomplete platelet inhibition, poor antiplatelet response in some patients, and an irreversible antiplatelet effect. 8 A new thienopyridine, prasugrel, has been observed to provide a faster response and better clinical efficacy than clopidogrel, at the expense of an increased risk of major bleeding. 9 Like clopidogrel, prasugrel irreversibly inhibits platelet aggregation by binding to the P2Y 12 platelet ADP receptor for the entire 8 to 9 day life span of platelets. ACS patients who are on clopidogrel or prasugrel are at a higher risk of bleeding during emergency CABG procedures or other surgical interventions. 10,11 Ticagrelor, a cyclopentyltriazolopyridine, has different properties than thienopyridines. 3 Since it is not a prodrug requiring metabolic activation, ticagrelor has a faster onset response than clopidogrel. 12 In addition, it binds reversibly to P2Y 12 platelet ADP receptor, which allows for the restoration of platelet aggregation upon termination of therapy. 12 The rapid and reversible effect of ticagrelor makes it a promising option for the treatment of ACS patients including those managed medically or with PCI and/or CABG. 13 Disclaimer: The Rapid Response Service is an information service for those involved in planning and providing health care in Canada. Rapid responses are based on a limited literature search and are not comprehensive, systematic reviews. The intent is to provide a list of sources and a summary of the best evidence on the topic that CADTH could identify using all reasonable efforts within the time allowed. Rapid responses should be considered along with other types of information and health care considerations. The information included in this response is not intended to replace professional medical advice, nor should it be construed as a recommendation for or against the use of a particular health technology. Readers are also cautioned that a lack of good quality evidence does not necessarily mean a lack of effectiveness particularly in the case of new and emerging health technologies, for which little information can be found, but which may in future prove to be effective. While CADTH has taken care in the preparation of the report to ensure that its contents are accurate, complete and up to date, CADTH does not make any guarantee to that effect. CADTH is not liable for any loss or damages resulting from use of the information in the report. Copyright: This report contains CADTH copyright material. It may be copied and used for non-commercial purposes, provided that attribution is given to CADTH. Links: This report may contain links to other information available on the websites of third parties on the Internet. CADTH does not have control over the content of such sites. Use of third party sites is governed by the owners own terms and conditions.

2 Prasugrel was approved by FDA in 2009 to be co-administered with ASA to prevent ischemic events and stent thrombosis in patients undergoing PCI. 14 The newer antiplatelet agent, tricagrelor was recently approved by FDA in 2010 to be used to reduce the risk of thrombotic events in ACS patients. 15 This review provides an update to the recent two CADTH reports 16,17 on the clinical effectiveness and the clinical guidelines on the use of clopidogrel, prasugrel and ticagrelor for the treatment of adult patients with ACS. This review also provides evidence on the costeffectiveness of clopidogrel, prasugrel and ticagrelor for treatment of ACS. Relevant FDA reports on prasugrel and ticagrelor are also summarized. RESEARCH QUESTIONS 1. What is the clinical effectiveness of clopidogrel, prasugrel or ticagrelor in adults with ST elevation or non-st elevation acute coronary syndrome (ACS)? 2. What is the comparative clinical effectiveness of clopidogrel, prasugrel or ticagrelor in adults with ST-elevation or non-st elevation ACS? 3. What is the cost-effectiveness of clopidogrel, prasugrel or ticagrelor in adults with ST elevation or non-st elevation ACS? 4. What are the evidence-based guidelines and recommendations on the use of clopidogrel, prasugrel or ticagrelor for ACS? KEY MESSAGE While combination therapy with clopidogrel and ASA remains the basis of antiplatelet therapy in ACS patients, each combined therapy of P2Y 12 inhibitors (clopidogrel, prasugrel, or ticagrelor) and ASA can be considered for some groups depending on the clinical presentation and subsequent management for patients with ACS. METHODS Literature Search Strategy A limited literature search was conducted on key resources including PubMed, Ovid Embase, The Cochrane Library (2012, Issue 5), University of York Centre for Reviews and Dissemination (CRD) databases, ECRI (Health Devices Gold), Canadian and major international health technology agencies, as well as a focused Internet search. Methodological filters were applied to limit retrieval to health technology assessments, systematic reviews, meta-analyses, randomized controlled trials and guidelines for questions 1, 2 and 4 and to economic studies for question 3. Where possible, retrieval was limited to the human population. The search was also limited to English language documents published between May 1, 2011 and May 7, 2012 for questions 1, 2 and 4 and between January 1, 2007 and May 7, 2012 for question 3. Antiplatelets for ACS 2

3 Selection Criteria and Methods One reviewer screened the titles and abstracts of the retrieved publications and evaluated the full-text publications for the final article selection, according to selection criteria presented in Table 1. Table 1: Selection Criteria Population Adults (18 or older) with ACS [including NSTEMI and STEMI] Intervention Comparator Outcomes Study Designs Clopidogrel, prasugrel, ticagrelor [with or without acetylsalicylic acid (ASA)] Q1: placebo or ASA Q2: clopidogrel, prasugrel, ticagrelor [with or without ASA] Q3, Q4: ASA, clopidogrel, prasugrel, ticagrelor [the latter three are used with or without ASA] Q1, Q2: Primary outcomes: composite of death, MI, and stroke Secondary (individual) outcomes: all-cause mortality, cardiovascular mortality, MI, stroke Harms: major and minor bleeding (as defined in individual studies) Q3: ICER, cost per QALY Q4: Guidelines Health technology assessments, systematic reviews, meta-analyses, randomized controlled trials, economic evaluations, guidelines Exclusion Criteria Studies were excluded if they did not satisfy the selection criteria in Table 1, if they were published prior to 2007, duplicate publications of the same study, or included in a selected health technology assessment or systematic review. Critical Appraisal of Individual Studies Critical appraisal of the included studies was based on study design. The methodological quality of the included randomized controlled trials (RCT) was evaluated using the SIGN50 quality assessment tool. 18 The internal validity was assessed with the following components: the adequacy of randomization, allocation concealment, degree of blinding, and use of intention to treat approach for analysis. The Appraisal of Guidelines Research & Evaluation (AGREE) instrument 19 was used by two independent reviewers to evaluate the quality of the included guidelines. The domains of scope and purpose, stakeholder involvement, rigour of development, clarity and presentation, applicability, and editorial independence were assessed using 23 key items. Finally, the methodological quality of the included cost-effectiveness studies were assessed using the guidelines for appraisal of economic studies by Drummond et al. 20 For the critical appraisal of studies, a numeric score was not calculated. Instead, the strength and limitations of the study were described. Antiplatelets for ACS 3

4 SUMMARY OF EVIDENCE Quantity of Research Available The literature search yielded 221 citations. Upon screening titles and abstracts, 192 citations were excluded and 29 potential relevant articles were retrieved for full-text review. Five additional relevant reports were retrieved from other sources. Of the 34 potentially relevant articles, 20 were included in this review. They are three RCTs, economic evaluations and five guidelines The study selection process is outlined in a PRISMA flowchart (Appendix 1). Summary of Study Characteristics Randomized controlled trials: A summary of study characteristics can be found in Appendix 2. One RCT (CIPAMI trial) 21 conducted in Germany compared a loading dose of 600 mg clopidogrel given in the prehospital phase (n=164) versus clopidogrel administered only after the diagnostic angiogram in patients with STEMI scheduled for primary PCI (n=171). Duration of follow-up was until hospital discharge or day 7, whichever happened first. Clinical outcomes included single and composite endpoint of death, re-myocardial infarction or urgent revascularization, thrombolysis in myocardial infarction (TIMI) major bleeding, and TIMI 2/3 patency of the infarct-related artery in the first diagnostic angiogram immediately prior to PCI. One RCT (ARMYDA-6 MI) 22 compared 600 mg (n=103) and 300 mg clopidogrel (n=98) loading doses in patients with STEMI undergoing PCI. Duration of follow-up was 30 days for major adverse cardiovascular events (MACE) and TIMI major bleeding, and 72 hours for infarct size, TIMI flow grade and left ventricular ejection fraction. A sub-study 23 of the PLATO trial compared ticagrelor (n=2601) versus clopidogrel (n=2615) in patients with ACS intended for non-invasive management. The PLATO trial had been reviewed in the previous CADTH report. 16 Economic evaluations: Twelve economic evaluations were identified The characteristics are summarized below and described in Appendix 3. All were cost-effectiveness studies, which can be arranged into five groups based on interventions and patient clinical conditions. Ticagrelor versus clopidogrel (genotype driven) in ACS patients: One US economic evaluation 24 determined the cost-effectiveness of ticagrelor compared with a genotype-driven selection of antiplatelet agents (clopidogrel therapy with no CYP2C19*2 mutation; ticagrelor therapy with mutation). The analysis was conducted from the perspective of US Medicare. Efficacy data comparing the clinical performance of ticagrelor and clopidogrel were taken from the PLATO trial, while the data of the CURE trial were used to adjust the hazard rates in patients with CYP2C19*2 mutation, who received no benefit of clopidogrel. The assumed generic price of clopidogrel was $30/month and the price of ticagrelor was $164/month. All costs were in 2009 US dollars. The study was supported by public funding. Prasugrel versus clopidogrel in ACS patients undergoing PCI: Two US economic evaluations 25,26 estimated the cost-effectiveness of prasugrel compared with clopidogrel using data from TRITON-TIMI 38. One analysis was developed from a managed care organization perspective, 25 while the other was evaluated from the perspective of the US health care Antiplatelets for ACS 4

5 system; 26 both used life time horizon. Drug costs were similar between studies, clopidogrel was $6.08/day 25 and $4.62/day, 26 prasugrel was $6.07/day 25 and $5.45/day. 26 One study estimated the cost of generic clopidogrel to be $1/day. 26 Both studies received funding from industry. Clopidogrel + ASA versus ASA in patients with STEMI: Five economic evaluations assessed the cost-effectiveness of clopidogrel + ASA compared with standard therapy (including ASA) in patients with STEMI. The analyses were developed from US medicare perspective, 27,30 from the perspective of UK National Health Service, 28 from the societal perspective in the Netherlands, 29 and from the health care payer perspective for France and Germany and a societal perspective for Sweden. 31 Source of efficacy estimate of these economic evaluations was from CLARITY 27-29,31 and COMMIT. 28,30,31 All studies used branded cost of clopidogrel and received funding from industries. Clopidogrel + ASA versus ASA in patients with NSTEMI: Two economic evaluations assessed the cost-effectiveness of clopidogrel + ASA compared with ASA in patients with NSTEMI from the payer s perspective within the German health care system 32 and from the Canadian health care perspective. 33 Both used efficacy data from the CURE trial. Only direct costs were incorporated in the analyses of both studies. Daily cost of clopidogrel was 2,21 32 or $2.40 (CAD). 33 Both studies were funded by industries. Clopidogrel + ASA versus ASA in ACS patients undergoing PCI: Two economic evaluations assessed the cost-effectiveness of pre-treatment and long-term treatment of clopidogrel in patients undergoing PCI from the health care payer perspective for France and Germany and a societal perspective for Sweden, 34 and from the Dutch health care perspective (direct costs only). 35 The Dutch study 35 compared pre-treatment followed by long-term therapy with only 4 weeks of treatment. The sources of efficacy estimate were PCI-CURE, 34,35 CREDO 34,35 and PCI- CLARITY. 34 All cost were reported for the price year and 2004, 35 where clopidogrel cost was of branded price. Both studies were funded by industries. Guidelines: Five documents providing recommendations were included in this review, one published in and four in Of the five recommendation statements, one was from Canada, 36 one was from Europe, 38 one was from US, 37 and two were from UK. 39,40 The recommendations are from Atlantic Cardiovascular Society (ACS), 36 European Society of Cardiology (ESC), 38 Institute for Clinical System Improvement (ICSI), 37 and National Institute for Health and Clinical Excellence (NICE). 39,40 Three guidelines provided recommendations on the use of antiplatelets (clopidogrel, prasugrel, ticagrelor) for ACS patients with UA/NSTEMI and STEMI, one on the use of ticagrelor for treatment of ACS, 39 and one on the use of prasugrel for treatment of ACS with PCI. 40 NICE recommendations 39,40 were not true evidence-based guidelines, but were based on single technology appraisal. The characteristics on the grading of recommendations and levels of evidence used to develop the corresponding guidelines are summarized in Appendix 7. Antiplatelets for ACS 5

6 Summary of Critical Appraisal The strengths and limitations of included studies are summarized in Appendix 4. The investigators of the CIPAMI trial 21 and the randomized patients were not blinded to the treatment allocation; in trial ARMYDA-6, only investigators blinding was specified, and in PLATO trial, both investigators and patient were blinded. The participating investigators had perceived beneficial effects of the pre-hospital administration of clopidogrel, and they were unwilling to randomize more patients; therefore the trial was terminated prematurely. In the ARMYDA-6 MI trial, 22 the primary outcome (infarct size) was estimated using a mathematical model based on surrogate measures (blood creatine kinase-myocardial band, troponin I and blood hemoglobin). The estimation method might not reflect a real infarct size, and the relation with the clinical outcomes was discussed. Finally in the PLATO trial, 23 the article reported a subgroup analysis from the PLATO population; however, it was not reported if this analysis was powered enough to detect the trial outcomes in the analyzed stratum. The main limitation of all included economic studies was the source of data used to feed the estimation models. The data used was either from historical anterior periods or from countries other than the location where the studies were looking to implement their results. The standards of care and use of resources might differ between countries and through the years; however, these factors were not fully considered in these studies. Three included guidelines did not report if patients preferences and views were taken into consideration at the conception of the guidelines Furthermore, the methods used to search and synthesize the evidence were not reported in two guidelines. 38,40 Summary of Findings Randomized controlled trials: The main study findings and authors conclusions from the clinical studies can be found in Appendix 5. Clopidogrel plus ASA versus placebo plus ASA New information regarding this topic was not identified in this review. This topic had been assessed in the previous CADTH report, 16 and the findings from two health technology assessments and four systematic reviews suggest that clopidogrel plus ASA is more effective than placebo plus ASA in ACS patients with UA/NSTEMI, or STEMI, who were clinically managed or underwent PCI. Clopidogrel pretreatment The evidence of clopidogrel pre-treatment has not been assessed in the previous CADTH report. 16 In the CIPAMI trial, mg loading dose clopidogrel given in the pre-hospital phase was found to be associated with a trend towards reduction in clinical events compared to clopidogrel given after diagnostic angiogram in patients with STEMI scheduled for primary PCI. The overall combined endpoint of death, re-infarction, and urgent revascularization occurred in 3.0% and 7.0% (P = 0.09; OR 0.42, 95% CI 0.14 to 1.20). Upon removal of two patients in the pre-hospital group, who did not receive clopidogrel before angiography, and five patients in the control group, who received clopidogrel in the pre-hospital phase, the difference of the combined endpoint was statistically significant (2.5% vs. 7.5%, P < 0.05; OR 0.38, 95% CI 0.12 to 0.98). There were no significant differences between treatment groups in TIMI major bleeding (9.1% Antiplatelets for ACS 6

7 vs. 8.2%, P = 0.80; OR 1.1, 95% CI 0.5 to 2.4), and TIMI 2/3 patency of infarct-related before PCI (49.3% vs. 45.1%, P = 0.5; OR 1.18, 95% CI 0.75 to 1.87). It was concluded that early inhibition of platelet ADP-receptor with a high loading dose of 600 mg clopidogrel given in the pre-hospital phase in patients with STEMI scheduled for primary PCI is safe, did not increase pre-pci patency of the infarct vessel, but was associated with a trend towards a reduction in clinical events. Clopidogrel loading dose (600 mg versus 300 mg) In the previous CADTH report, 16 the systematic review by Siller-Matula (2010) was included to assess the efficacy and safety of two clopidogrel loading doses (600 mg vs. 300 mg) in ACS patients (STEMI, NSTEMI) undergoing PCI. The results of the meta-analysis suggest that intensified loading dose of clopidogrel reduced the incidence of major cardiovascular events without an increase in major bleeding. The ARMYDA-6 MI trial, conducted in Italy, Hungary, Serbia, and Belgium, evaluated the impact of pre-treatment with a 600-mg versus a 300-mg clopidogrel loading dose on efficacy and safety outcomes in the setting of urgent PCI for ACS patients with STEMI. Both arms had similar symptoms-to-balloon time (600 mg: 283 ± 183 min vs. 300 mg: 288 ± 223 min; P = 0.86) and clopidogrel load-to-balloon time (35 ± 34 min vs. 39 ± 41 min; P = 0.45). The incidence of 30- day MACE was lower in the 600-mg group (5.8%, 6/103) compared with 300-mg group (15.0%, 15/98). The difference was statistically significant (p=0.049). There was a trend in the reduction of death (3.9% vs. 7.1%), re-infarction (0.98% vs. 5.1%), and target vessel re-vascularization (0.98% vs. 7.1%) in the 600-mg group compared with the 300-mg group, although the differences were not statistically significant. The safety endpoint in terms major bleeding (1.9% vs. 2.0%) minor bleeding (7.8% vs. 6.1%) and entry-site complications (2.9% vs. 3.1%) did not differ in the 600-mg and 300-mg groups. Infarct size was significantly lower in the high loading dose regimen, measured by area under curve of cardiac markers: median creatine kinase myocardial band (2,070 ng/ml vs. 3,049 ng/ml, P = ); troponin I (255 ng/ml vs. 380 ng/ml, P < ). Overall, TIMI flow grade <3 after PCI was less frequent in the 600-mg arm (6% vs. 16%, P = 0.031), whereas left ventricular fraction at discharge was improved (52.1 ± 9.5% vs ± 11.3%, P = 0.026). It was concluded that 600-mg clopidogrel loading dose is safe and more effective than 300-mg in ACS patients with STEMI undergoing PCI. New P2Y 12 inhibitors (prasugrel or ticagrelor) versus clopidogrel The overall conclusion derived from three systematic reviews assessed in the previous CADTH report 16 is that both prasugrel and ticagrelor are more efficacious than clopidogrel in ACS patients with STEMI and UA/NSTEMI. Indirect comparison suggests that prasugrel and ticagrelor have similar efficacy, except prasugrel is more protective for stent thrombosis, but causes more bleeding. A sub-study of the PLATO trial was identified in this review, and was not available in the previous CADTH report. This study evaluated the efficacy and safety outcomes in patients in the PLATO trial who at randomization were planned for a non-invasive treatment strategy. All patients received ASA, unless intolerant of it, though no range of ASA dose received was indicated for this sub-study group. The incidence of primary endpoint (CV death, MI, stroke) was significantly lower with ticagrelor than with clopidogrel (12.0% vs. 14.3%; HR 0.85, 95% CI 0.73, 1.00; P = 0.04). The result was mainly driven by the reduction in CV death (5.5% vs. 7.2%; HR 0.76, 95% CI 0.61, 0.96; P = 0.019) without any significant difference in MI or stroke. Overall mortality was also lower with ticagrelor (6.1% vs. 8.2%; HR 0.75, 95% CI 0.61, 0.93; P = 0.010). There was numerical increase in major bleeding (11.9% vs. 10.3%; HR 1.17, 95% CI 0.89 to 1.39; P = 0.079), non-cabg related major bleeding (4.0% vs. 3.1%; HR 1.30, 95% CI 0.95 to Antiplatelets for ACS 7

8 1.77; P = 0.103), and intracranial bleeding (0.5% vs. 0.2%; HR 2.83, 95% CI 0.90 to 8.90; P = 0.075), but these differences were not statistically significant. It was concluded that, in ACS patients initially intended for non-invasive management, the benefits of ticagrelor over clopidogrel were consistent with those derived from the overall population of PLATO. Economic evaluations: The main economic study findings and authors conclusions can be found in Appendix 6. Ticagrelor versus clopidogrel (genotype driven) in ACS patients A US economic evaluation (2011) 24 determined the cost-effectiveness of universal ticagrelor compared with genotype driven selection of either clopidogrel or ticagrelor based the presence or absence of CYP2C19*2 mutation. The 5-year medical costs (in 2009 US$) was estimated using a hybrid decision tree/markov model. Under a base-case scenario, the incremental costeffectiveness ratio (ICER) for universal ticagrelor after 5 year was $10,059 per QALY compared to genotype driven treatment. One-way sensitivity analysis showed that the result was most sensitive to the price of ticagrelor and hazard ratio for death; ICERs remained below the common threshold of $50,000 per QALY until a monthly ticagrelor price of $693 or a 0.93 hazard ratio for death. Probabilistic sensitivity analysis using Monte Carlo simulation estimated that the ICER for universal ticagrelor was below $50,000 per QALY in 97.7% of simulations. It was concluded that universal ticagrelor increased QALY for ACS patients at a cost below a typically accepted threshold. Prasugrel versus clopidogrel in ACS patients undergoing PCI A US economic evaluation (2012), 25 using a disease-progression model, estimated the costeffectiveness of prasugrel in relation to clopidogrel from a US payer perspective for treatment of ACS patients undergoing PCI. All costs were adjusted to 2009 US dollars. The average total cost (median follow-up 15 months) was $97,090 per 100 patients lower with prasugrel due to reduced hospitalization rate. Under a base-case scenario, prasugrel-based therapy was associated with cost-savings and fewer clinical events in patients subgroups, in alternative time horizons (30 days, 1 year), in a US cohort and a North American cohort, and with generic clopidogrel at $3 per day less than prasugrel. The cost was higher for prasugrel when clopidogrel cost was at $4 per day less than prasugrel. In lifetime analysis based on clopidogrel generic prices ($3 or $4 per day less than prasugrel), the ICERs were $6,643 or $13,906 per life year gained (LYG), respectively. One-way sensitivity analyses showed that the results were most sensitive to the relative costs of the two treatments and the cost for hospital stays. Probabilistic sensitivity analyses showed that, at the clopidogrel branded price, prasugrel-based therapy was dominant in 97.6% of the 10,000 simulations. With clopidogrel prices of $3 and $4 less per day than prasugrel, the probability that prasugrel is cost-effective using a threshold value of $50,000 per LYG was 99.5% and 98.2%, respectively. It was concluded that the use of prasugrel-based therapy in ACS patients undergoing PCI resulted in cost-savings under branded price of clopidogrel and favorable cost-effective ratios at generic prices due to offsetting savings in the cost of rehospitalization. A US economic evaluation (2010), 26 evaluated the cost-effectiveness of prasugrel versus clopidogrel from the perspective of US healthcare system. All costs were in 2005 US dollars. Cumulative medical care costs (median follow-up 14.7 months), including study drug and the initial and follow-up hospitalizations were $221 per patient lower with prasugrel (95% CI -759 to 299) due to lower rate of re-hospitalization. Prasugrel was associated with years of life expectancy gained (95% CI to 0.180), due to decreased in non-fatal MI. Under base-case assumptions, treatment with prasugrel was a dominant strategy in 79.7% of bootstrap Antiplatelets for ACS 8

9 replicates, and the ICER was <$50,000 per life-year gained in 99.8%. If the generic clopidogrel cost was $1/day, the incremental cost with prasugrel was $996 per patient with an associated ICER of $9,727 per life-year gained, and 98.2% of bootstrap estimates were <$50,000 per lifeyear gained. It was concluded that treatment of ACS patients undergoing PCI with prasugrel for up to 15 months was an attractive strategy compared with clopidogrel. Clopidogrel + ASA versus ASA in patients with STEMI A US economic evaluation (2010), 27 calculated the cost-effectiveness of short-term clopidogrel therapy (30 days) in STEMI patients treated with fibrinolysis. Total costs and resource use were not significantly different for the clopidogrel and placebo groups ($8,128 vs. $8,134). All costs were in 2008 US dollars. Clopidogrel remained economically dominant in terms of cost per LYG, or cost per event prevented. In a sensitivity analysis accounting for higher long-term medical costs due to greater life expectancy, clopidogrel remained under $6000 per LYG. Clopidogrel price changes had little effect on cost-effectiveness ratios. Clopidogrel therapy was dominant in 35% of the bootstrap simulations and cost less than $50,000 per LYG in 67% of simulations. It was concluded that short-term clopidogrel therapy was a highly economically attractive strategy, because it improved clinical outcomes at no increase in costs. An economic evaluation (2010) 28 carried out from the perspective of the UK NHS assessed the long-term cost effectiveness of treatment for 1 month, and for 1 year with clopidogrel in addition to ASA compared with ASA in patients with STEMI. Costs were reported as 2006 values. Under a base-case scenario, the ICERs after 1 month and 1 year therapy were 2284 and 3891 per QALY, respectively. Univariate sensitivity analysis showed no variation of the included parameters resulted in an ICER greater than 10,000 (COMMIT/CCS-2 data) or 6000 (CLARITY-TIMI 28). Probabilistic sensitivity analyses, presented as cost-effectiveness acceptability curves, demonstrated the probability that treatment with clopidogrel has higher net benefits than treatment with standard therapy alone for a range of monetary values that a decision-maker may consider the maximum acceptable to gain one QALY. Both costeffectiveness acceptability curves for 1 month and 1 year showed that the probability of clopidogrel plus standard therapy being cost-effective was around 0.95 at current willingness to pay thresholds. It was concluded that clopidogrel plus standard therapy treatment of STEMI patients undergoing fibrinolytic intervention was cost-effective in the UK setting. An economic evaluation (2010) 29 assessed the costs and effects of clopidogrel plus ASA compared with ASA in patients with STEMI in a societal perspective in the Netherlands. All costs were in euros at 2006 values. For a cohort similar to that of CLARITY protocol and with the 28-day efficacy, treatment with clopidogrel plus ASA resulted in 0.05 LYG and QALYs gained for a cost that was 1929 lower than ASA therapy, indicating that combined treatment was economically dominant. Sensitivity analyses showed that ICERs were insensitive to variation in all listed factors except the upper margin of the 95% CI of risk reduction of CV mortality. Continuation of treatment outside the trial period (2 to 12 months) was expected to result in ICERs of below 20,000 per QALY as long as the real risk reduction (product of risk and risk reduction) of combination treatment is greater than 0.487% per year. It was concluded that combined therapy of clopidogrel and ASA was cost-effective according to the regimen as in CLARITY trial. A US economic evaluation (2009) 30 used a decision tree model to estimate the long-term costeffectiveness of clopidogrel in STEMI patients in the COMMIT trial from the US Medicare perspective. All costs were in 2003 US dollars. Within 28 days, clopidogrel plus ASA was a dominant strategy because total event rate was decreased (9.2% vs. 10.1%) without an increase in cost ($7791 vs. $7797). Over life-year time, 1-year treatment of clopidogrel plus ASA produced an ICER of $10,691 per LYG. Sensitivity analyses, using event rate and LYG as Antiplatelets for ACS 9

10 parameters, showed that ICERs for clopidogrel plus ASA were well below the common benchmark ceiling ratio of $50,000 per LYG. It was concluded that addition of clopidogrel to ASA, up to 1 year of therapy, was a highly cost-effective strategy in STEMI patients. An economic evaluation (2007) 31 assessed the cost-effectiveness of clopidogrel in short- and long-term treatment of STEMI with the use of data from two trials in Sweden, Germany and France (CLARITY and COMMIT). The analyses were conducted from a healthcare perspective for France and Germany, and a societal perspective for Sweden. All costs were reported for the price year Under base-case scenario and as for CLARITY and COMMIT populations, the ICERs in Sweden were 2100/LYG and 2772/LYG, respectively; the ICERs in Germany were 92/LYG and 4144/LYG, respectively; and the ICERs in France were dominant and 2786/LYG, respectively. One-way sensitivity analyses regarding variations in treatment effects, costs and discount factors for the two studies showed that the results in all three countries were most sensitive to the cost of added LYGs and the exclusion of costs after the first year from the analysis. For Sweden at the commonly quoted willingness-to-pay threshold of 50,000/QALY, the probabilistic sensitivity analyses showed that the clopidogrel strategy was cost-effective in around 90% of the simulations (CLARITY cohort) and 84% when applying COMMIT results. It was concluded that treatment of STEMI patients with clopidogrel was cost-effective in all three European countries, with the predicted ICERs well below the accepted threshold values. Clopidogrel + ASA versus ASA in patients with NSTEMI An economic evaluation (2007) 32 conducted in Germany assessed the long-term costeffectiveness of clopidogrel in ACS patients with NSTEMI from the payer s perspective within the German healthcare system. A Markov model was used to evaluate the incremental costeffectiveness of clopidogrel in addition to ASA using data from the CURE trial. Costs were adjusted to 2004 value. Direct medical costs per patient in 1 year of treatment with clopidogrel plus ASA and ASA alone were 8,953 and 8,548, respectively. The model predicted a survival of 9.02 years in clopidogrel plus ASA group and 8.89 years in ASA group, yielding an ICER of 3,113 per LYG. In sensitivity analyses, variations of key parameters did not alter the overall results of the base-case scenario, resulting in a range of ICERs from 1,338 to 9,322. It was concluded that adding clopidogrel to ASA for ACS patients with NSTEMI generated an additional life year saved at a relatively low cost of 3,113. An economic evaluation (2007) 33 assessed the long-term cost-effectiveness of clopidogrel in ACS patients with NSTEMI in the Canadian health care system. The analysis was based on the CURE trial. All costs were reported for the price year 2004 Canadian dollars. Clopidogrel was shown to be cost-effective, with ICERs less than $10,000 per event prevented and less than $4,000 per LYG. The probability of clopidogrel resulting in cost/lyg of less than $20,000 was for CURE patients and for PCI-CURE patients. It was concluded that combination therapy of clopidogrel and ASA for ACS patients with NSTEMI was cost-effective. Clopidogrel + ASA versus ASA in ACS patients undergoing PCI An economic evaluation (2008) 34 conducted a cost-effectiveness analysis of pre-treatment and long-term treatment with clopidogrel in ACS patients undergoing PCI in three European countries based on meta-analysis of PCI-CURE, CREDO and CLARITY trials. The analyses were conducted from a healthcare perspective for France and German, and a societal perspective for Sweden. All costs were reported for the price year Under base-case scenario, the ICER in Sweden was 4,225 per QALY; the ICER in Germany was 7,871 per QALY; and the ICER in France was 5,226 per QALY. The one-way sensitivity analyses showed that costs of added years of life had highest impact on the overall results. Clopidogrel is cost saving across all countries when assessing pre-treatment only. The results were relatively Antiplatelets for ACS 10

11 stable when key parameters were varied. In probabilistic sensitivity analyses, using the commonly quoted willingness-to-pay threshold of 50,000 per QALY, at least 99% of simulations can be considered cost-effective in three countries. At a lower threshold of 10,000 per QALY, 91% of simulations would be cost-effective in Sweden, 87% in France and 59% in Germany (mainly due to higher unit cost of clopidogrel). It was concluded that pre-treatment and long-term treatment with clopidogrel for up to one year in a PCI setting were cost-effective. An economic evaluation (2007) 35 examined the cost-effectiveness, from the Dutch healthcare perspective, of clopidogrel in patients undergoing PCI. The analysis compared pre-treatment followed by long-term therapy (9-12 months) with only 4 weeks of treatment based on data of PCI-CURE and CREDO trials. All costs were reported for the price year Based on PCI- CURE or CREDO population, long-term therapy was found to be dominant (more effective, less costly) over short-term therapy. The results were robust over the entire range of the univariate sensitivity analyses. With PCI-CURE population, long-term clopidogrel had 0.98 probability of being cost-saving and a 0.65 probability of being more effective, with additional life-years saved and QALYs gained, and the probability that long-term clopidogrel was cost-effective was 87% at 20,000 per LYG threshold. With CREDO population, long-term clopidogrel therapy was more effective in 99% and cost saving in 99% of the 1000 simulations; and the probability that longterm clopidogrel was cost-effective was 99% at 20,000 per LYG threshold. It was concluded that in the Netherlands a loading dose of clopidogrel before PCI followed by long-term therapy is dominant for the prevention of subsequent ischemic events in patients undergoing either elective procedures or in patients undergoing primary PCI compared with short-term treatment with clopidogrel without a pre-treatment dose. Guidelines: The previous CADTH report 17 reviewed the recommendations from eleven guidelines on the use of antiplatelet agents for ACS. The conclusions drawn at that time were that: 1) clopidogrel in combination with ASA is used to prevent ischemic events and stent thrombosis in NSTEMI and STEMI patients; 2) prasugrel in combination with ASA is used to prevent stent thrombosis in patients undergoing PCI; 3) ticagrelor may be used to reduce risk of thrombotic events in ACS patients, though its place in therapy with respect to other antiplatelet agents is not clear. Of the five additional guidelines identified in this review, the Atlantic Anti-platelet Initiative (AAPI) guidelines by Love et al., developed in collaboration with the Atlantic Cardiovascular Society was the most up-to-date and was based on the Canadian setting. Details of the recommendations of five guidelines are presented in Appendix 8. An overview of recommendations of AAPI is provided in Table 2. ASA was recommended for all patients with definite or suspected ACS who do not have contraindications to therapy. Table 2: Summary of recommendations of AAPI Clinical Presentations Recommendations STEMI receiving 1. Clopidogrel should continue to be the preferred P2Y 12 inhibitor (300 thrombolytic therapy mg LD, 75 mg od; omit LD in patients > 75 years) [Strong STEMI undergoing primary PCI recommendation, high-quality evidence] 1. Clopidogrel should be the preferred P2Y 12 inhibitor administered prior to cardiac catheterization laboratory arrival ( mg LD, 75 mg od) [Strong recommendation, high-quality evidence] 2. Prasugrel or ticagrelor is not recommended for pre-hospital of emergency department administration [Conditional recommendation, very low-quality evidence] 3. Prasugrel (60 mg LD, 10 mg od) or ticagrelor (180 mg LD, 90 mg bid) is recommended if first administered in the cardiac Antiplatelets for ACS 11

12 Clinical Presentations STEMI with planned medical management NSTE-ACS requiring urgent invasive assessment NSTE-ACS with planned invasive management NSTE-ACS with planned medical management Patients with ACS undergoing early CABG Recommendations catheterization laboratory and there are no contraindication [Strong recommendation, moderate-quality evidence] 4. Switching from clopidogrel to prasugrel [Conditional recommendation, very low-quality evidence] or ticagrelor [Strong recommendation, moderate-quality evidence] during or after cardiac catheterization can be considered if higher degree of platelet inhibition is required. 1. Clopidogrel should be the preferred P2Y 12 inhibitor (300 mg LD, 75 mg od; omit LD in patients > 75 years) [Strong recommendation, high-quality evidence] 1. Clopidogrel should be the preferred agent ( mg LD, 75 mg od) [Strong recommendation, high-quality evidence] 2. Prasugrel or ticagrelor should not administered prior to cardiac catheterization [Conditional recommendation, very low-quality evidence] 3. Switching from clopidogrel to prasugrel [Conditional recommendation, very low-quality evidence] or ticagrelor [Strong recommendation, moderate-quality evidence] during or after cardiac catheterization can be considered in patients with high risk of NSTE-ACS in the absence of contraindications 1. Clopidogrel should be the preferred P2Y 12 inhibitor (300 mg LD, 75 mg od) [Strong recommendation, high-quality evidence] 2. Ticagrelor (180 mg LD, 90 mg bid) can be considered for high risk patients in the absence of contraindications [Conditional recommendation, moderate-quality evidence] 3. Switching from initially treated clopidogrel to ticagrelor in high risk patients should be considered once more is known about patient characteristics and coronary anatomy [Strong recommendation, moderate-quality evidence] 4. Switching from initially treated clopidogrel to prasugrel is generally not recommended [Conditional recommendation, very lowquality evidence] 1. Clopidogrel should be the preferred P2Y 12 inhibitor (300 mg LD, 75 mg od) [Strong recommendation, high-quality evidence] 2. Ticagrelor (180 mg LD, 90 mg bid) can be considered for high risk patients in the absence of contraindications [Conditional recommendation, moderate-quality evidence] 3. Switching from initially treated clopidogrel to ticagrelor in high risk patients should be considered once more is known about patient characteristics and coronary anatomy (if cardiac catheterization performed) [Strong recommendation, moderate-quality evidence] 4. Prasugrel is currently not recommended [Conditional recommendation, very low-quality evidence] 1. P2Y 12 inhibitor should generally not be administered prior to cardiac catheterization [Conditional recommendation, low-quality evidence] 2. In patients who received a P2Y 12 inhibitor, clopidogrel and ticagrelor should be discontinued 5 days before surgery, or prasugrel 7 days before surgery [Strong recommendation, moderate-quality evidence] 3. Same P2Y 12 inhibitor administered pre-operatively should be restarted after surgery [Conditional recommendation, low-quality Antiplatelets for ACS 12

13 Clinical Presentations Recommendations evidence] ACS=acute coronary syndrome; bid=twice daily; CABG=coronary artery bypass grafting; LD=loading dose; NSTE=non ST-segment elevation; od=once daily; PCI=percutaneous coronary intervention; STEMI=ST-segment elevation myocardial infarction The European Society of Cardiology (ESC) provided recommendations on the use oral antiplatelet agents for the management of NSTE-ACS. 38 Ticagrelor (180 mg LD, 90 mg bid) is recommended for all patients at moderate-to-high risk of ischemic events [Class I, Level B] Prasugrel (60 mg LD, 10 mg od) is recommended for P2Y 12 inhibitor-naïve patients (especially diabetics) undergoing PCI with known coronary anatomy and no contraindications [Class I, Level B] Clopidogrel (300 mg LD, 75 mg od) is recommended for patients who cannot receive ticagrelor or prasugrel [Class I, Level A] A 600-mg LD clopidogrel is recommended for patients undergoing invasive strategy when ticagrelor or prasugrel is not an option [Class I, Level B] A 150-mg maintenance dose of clopidogrel should be considered for the first 7 days in patients managed with PCI [Class IIa, Level B] In patients pretreated with P2Y 12 inhibitors who need to undergo non-urgent major surgery (including CABG), clopidogrel and ticagrelor should be discontinued 5 days before surgery, or prasugrel 7 days before surgery [Class IIa, Level C] Ticagrelor or clopidogrel should be re-started after CABG surgery [Class IIa, Level B] The Institute for Clinical System Improvement (ICSI) had recommendations on the use of P2Y 12 inhibitors for invasive and non-invasive management of ACS. 37 For patients with planned non-invasive management strategy, a loading dose of clopidogrel of ticagrelor should be given in addition to ASA as soon as possible. treatment should be continued for at least 1 month and up to 12 months. For STEMI and primary PCI, a loading dose of a P2Y 12 inhibitor should be given as soon as possible or before PCI. Prasugrel is not recommended in patients with history of stroke or transient ischemic attack, or who are >75 years, except high risk situations (diabetes or prior MI history). Clopidogrel is recommended in STEMI patients receiving fibrinolytic therapy and undergoing non-primary PCI. Clopidogrel or ticagrelor is recommended in STEMI patients receiving no fibrinolytic therapy and undergoing non-primary PCI. Prasugrel should be given to STEMI patients once the coronary anatomy is known and PCI is planned. When revascularization is required and CABG is planned, clopidogrel or ticagrelor should be discontinued for 5 days and prasugrel for 7 days before surgery. The National Institute for Health and Clinical Excellence (NICE) recommended ticagrelor in combination with low-dose ASA for up to 12 months as treatment option in ACS patients with STEMI, NSTEMI or UA. 39 NICE also recommended prasugrel in combination with ASA as treatment option in ACS patients with STEMI undergoing primary PCI, when stent thrombosis has occurred during clopidogrel treatment, or in ACS patients having diabetes. 40 Antiplatelets for ACS 13

14 FDA approval: Prasugrel The FDA Action Package for prasugrel revealed the timing of loading in TRITON-TIMI 38 trial. 41 The FDA found that the loading dose of study drugs was not administered immediately in TRITON, as was the case in previous clopidogrel trials and in recommended guidelines, but was delayed by protocol until after angiography was performed for UA/NSTEMI patients and not specified as immediate for STEMI patients. This aspect, together with excess bleeding with prasugrel, precluded the FDA from deeming prasugrel superior to clopidogrel. Prasugrel was approved by FDA on July 10, for the reduction of thrombotic cardiovascular events (including stent thrombosis) in patients with ACS who are to be managed with PCI as followed: Patients with unstable UA or NSTEMI Patients with STEMI when managed with either primary or delayed PCI Latest approval of the Full Prescription Information was on September 27, 2011, 43 with box warnings for bleeding risk. A Risk Evaluation and Mitigation Strategy 44 has been put in place and a Medication Guide 45 is now available for distribution with each prescription to inform patients about the serious risks associated with prasugrel, particularly risk of bleeding. The manufacturer has also implemented a communication plan to healthcare providers to convey information regarding serious risk of bleeding associated with prasugrel, and appropriate patient selection. 46 Prasugrel can cause significant, sometimes fatal, bleeding. Risk factors for bleeding include age 75 years, body weight <60 kg, propensity to bleed, and concomitant use of medications that increase the risk of bleeding. Prasugrel should not be started in patients likely to undergo urgent CABG and when possible, prasugrel should be discontinued at least 7 days prior to any surgery. Ticagrelor In the Complete Response Letter, 47 the FDA expressed concern that the North American results are not a chance finding. In the US population, there was a statistically non-significant trend towards a higher incidence the composite outcome with ticagrelor compared to clopidogrel. The effect of ASA dose is an alternative explanation for the difference in treatment effect between US and outside of US (OUS). US patients received higher ASA dose (median 325 mg/day) than OUS patients (median 100 mg/day). In light of this aspect, the FDA recommended the manufacturer provide further detailed analyses of ASA dose on US/OUS findings based on a number of methods to categorize ASA dose. The effect of ASA dose should be analyzed on the primary composite endpoint and its components, as well as on major subgroups. FDA also required the manufacturer to submit draft labeling and A Risk Evaluation and Mitigation Strategy (REMS), which contains a Medication Guide, Communication Plan and a timetable for submission of assessments of the REMS. FDA also required a safety update. Ticagrelor received approval from FDA on July 20, for the reduction of thrombotic cardiovascular events in patients with ACS (UA, NSTEMI or STEMI). The FDA indicated that maintenance doses of ASA above 100 mg daily should be avoided. The Full Prescription Information was approved in July, 2011, 49 which contains a box warning for bleeding risk and ASA dose. The REMS 50 has been approved by FDA, and the Medication Guide 51 is now available for distribution with each prescription. Antiplatelets for ACS 14

15 Limitations Clinical The CIPAMI trial 21 was not powered to show significant differences for clinical endpoints. The duration of follow-up was relatively short, until hospital discharge or day seven, whichever happened first. Treatment allocation was not blinded to patients or investigators. The ARMYDA-6 MI trial 22 was also not powered for the evaluation of clinical endpoints. Surrogate outcome measures of infarct size were chosen as primary outcome. The method used to measure infarct size was mathematical modeling based on surrogate outcome measures and may not be an true measurement of actual infarct size. In the PLATO sub-study, 23 despite an intended non-invasive strategy at the time of randomization, about half of patients had coronary angiography, one third had PCI, and one tenth had CABG during follow-up. The results could be therefore confounded by invasive treatment when intention-to-treat analysis was applied. The CIPAMI and the ARMYDA-6 trials were not conducted in Canada. Although PLATO was multi-national trial, patients from Canada represented only 401 out of 18,624 randomized patients (2.15%). Hence the applicability of the findings from those trials to Canadian setting is uncertain. There were no head-to-head comparison of prasugrel and ticagrelor for treatment efficacy in ACS, limiting the ability to draw conclusions about their relative effectiveness. Economic There were no economic evaluations which evaluated the cost-effectiveness of prasugrel or ticagrelor from the Canadian healthcare perspective. Among the economic evaluations for clopidogrel, there was only one cost-effectiveness analysis of clopidogrel for NSTEMI in the Canadian healthcare system. Most of the economic evaluations (11 out of 12) were financially sponsored by industry. The branded prices of clopidogrel, instead of the generic price, were used in 10 out of 12 economic evaluations. These evaluations may be losing relevance as new generic products are introduced in Canada. Guidelines Of the five included guidelines, only one was published by a Canadian professional society. 36 The ICSI guidelines 37 did not provide grading for the recommendations and levels of evidence. The NICE guidelines for ticagrelor 39 and prasugrel 40 used the single technology appraisal process developed by NICE, in which the grading of recommendations and levels of evidence were not clearly specified. They may not be classified as complete guidelines. CONCLUSIONS AND IMPLICATIONS FOR DECISION OR POLICY MAKING Both clinical and economic evaluations showed that clopidogrel plus ASA was associated with a reduced risk of cardiovascular events and was cost-effective compared with ASA alone in ACS patients with UA/NSTEMI or STEMI, who were clinically managed or underwent PCI. In STEMI patients scheduled for primary PCI, a 600-mg loading dose of clopidogrel given to patients before arrival to cardiac catheterization laboratory was associated with a trend towards reduction in clinical events than a loading dose given after diagnostic angiography. Moreover, an intensified loading dose of clopidogrel (600 mg) was found to be safe and more effective than 300-mg in ACS patients with STEMI or NSTEMI undergoing PCI. Antiplatelets for ACS 15