Author + information
- †Departments of Medicine and Epidemiology and Biostatistics, University of California San Francisco, San Francisco, California
- ‡Departments of Health Research and Policy and Medicine, Stanford University School of Medicine, Stanford, California
- ↵∗Reprint requests and correspondence:
Dr. Dhruv S. Kazi, Division of Cardiology, San Francisco General Hospital, 1001 Potrero Avenue, Room 5G1, San Francisco, California 94110.
For more than a decade, dual antiplatelet therapy with aspirin and clopidogrel has been the standard of care for patients presenting with acute coronary syndrome (ACS) (1). However, the therapeutic landscape has changed dramatically with the introduction of prasugrel and ticagrelor, which improve clinical outcomes relative to that of clopidogrel (2,3), development of generic formulations of clopidogrel, and availability of point-of-care platelet reactivity and genotype tests. The possible combinations and permutations of these new options have resulted in many therapeutic strategies, and their relative effects on long-term outcomes merit careful evaluation.
These innovations have enormous clinical and economic implications in light of the 625,000 patients hospitalized annually for ACS in the United States alone (4) and the 60- to 70-fold higher cost of the novel agents relative to that of generic clopidogrel (5). Nevertheless, novel dual antiplatelet strategies may provide good value for the additional cost if they improve long-term clinical outcomes and reduce subsequent health care costs.
In this issue of the Journal, Cowper et al. (6) present an economic analysis of ticagrelor therapy from the U.S. health care perspective, based on patient-level cost and outcomes data collected prospectively alongside the PLATO (Platelet Inhibition and Patient Outcomes) trial (3), which randomized 18,624 patients with ACS from 43 countries (1,413 from the United States) to receive either ticagrelor plus aspirin or clopidogrel plus aspirin for 12 months. Patients assigned to the ticagrelor arm exhibited a 16% relative reduction in the primary endpoint of vascular death, myocardial infarction, or stroke relative to those who received clopidogrel (9.8% vs. 11.7%, respectively; p < 0.001). However, U.S. patients appeared to have less benefit from ticagrelor than patients in the rest of the world; post hoc analyses suggested this may have been due to the higher dose of aspirin used in the United States (7). The U.S. Food and Drug Administration ultimately approved ticagrelor but warned against concomitant aspirin doses exceeding 100 mg/day.
In order to assess the approved regimen’s cost effectiveness, Cowper et al. (6) assumed that use of ticagrelor with low-dose aspirin would produce the same reduction in clinical events among American patients as in the overall PLATO trial and that the resulting resource utilization could be estimated based on the subset of U.S. patients treated with low-dose aspirin. They then applied several analytical methods to project long-term survival, cost, and cost effectiveness of using ticagrelor rather than clopidogrel to treat patients with ACS.
The authors project that using ticagrelor would improve life expectancy relative to clopidogrel by 2.6 months (0.14 quality-adjusted life-years [QALYs]). Because they found no differences in other medical costs among the 547 U.S. patients treated with low-dose aspirin, they estimated the net cost of using ticagrelor to be $2,172, the difference in the prices of ticagrelor and generic clopidogrel for 12 months. Based on these data, they calculated the incremental cost-effectiveness ratio (ICER) for ticagrelor compared with clopidogrel of $29,700/QALY, which was robust in sensitivity analyses and bootstrap replications. They found moderate variations in the ICER across clinical subgroups, primarily due to variations in the size of the absolute clinical benefits of using ticagrelor. For instance, ticagrelor appeared to be more cost effective in patients who underwent invasive treatment ($27,300/QALY) than among patients who were managed medically ($47,000/QALY), yet all subgroup estimates were within the range of generally acceptable treatments.
Economic evaluations incorporate many assumptions, so sensitivity analyses that evaluate the impact of variations in these assumptions remain an essential facet of any cost-effectiveness analysis. In a sensitivity analysis, investigators vary one or more input parameters while holding others constant and recompute the ICER. Two of the study’s sensitivity analyses deserve closer attention. First, because incremental costs in this analysis were driven almost entirely by drug costs, the cost effectiveness of ticagrelor varied substantially based on the price difference between ticagrelor and clopidogrel: the smaller the difference, the more economically attractive ticagrelor became. Second, using outcomes from the entire study cohort might have overestimated the benefit of ticagrelor in the United States (as mortality in U.S. patients was lower and absolute risk reductions drove the cost-effectiveness ratio). Recalibrating the model using outcomes observed in the U.S. cohort increased the ICER slightly, from $29,700/QALY to $36,500/QALY, which is similar to the ICER of $40,300/QALY that we found by using an independent simulation model (5).
Despite the strengths of this trial-based economic analysis by Cowper et al. (6), it has several limitations. Incremental cost effectiveness is always defined relative to an alternative, and the PLATO study examined only 1 of the many strategies for post-ACS antiplatelet use. For instance, prasugrel use was not considered as an alternative, nor was individualized therapy using platelet testing or genotyping. Trial-based economic evaluations are necessarily tied to the design of the parent clinical trial and cannot address alternatives other than the randomized strategies. Also, the authors assumed that heterogeneity of treatment effectiveness for ticagrelor comparing the United States with the rest of the world was explained by the aspirin dose; they did not test the possibility of systematic variations in relative risk reduction from ticagrelor, which would have increased the uncertainty in the results beyond the bootstrap confidence intervals. Finally, the effectiveness of treatments in real-world practice may be substantially different from the efficacy found in randomized trials. For instance, adherence to ticagrelor in the community may be adversely affected by the drug’s twice-daily dosing (compared with the once-daily administration of clopidogrel and prasugrel) or the occurrence of ticagrelor-associated dyspnea. These cost-effectiveness estimates should be updated when real-world, post-marketing data for ticagrelor and prasugrel become available.
Policy makers need to consider the total budgetary impact of new drugs, not just their incremental cost effectiveness. In the fragmented U.S. health care system, the entity that pays for initial treatments (e.g., the prescription insurance plan) does not always benefit from downstream savings (reduced hospitalizations or disability costs). The rise of accountable care organizations that bear the complete economic risk of downstream health care costs may properly align incentives to promote the long-term health of patients. Nevertheless, policy makers managing total budgets face tough decisions about whether they can bear the substantial additional expenditures for new, costly treatments: at an incremental cost of $2,172, using ticagrelor instead of generic clopidogrel for each of 650,000 ACS episodes in the United States would lead to $1.4 billion in additional spending.
The cardiovascular community is justifiably proud of its commitment to evidence-based medicine and is now starting to consider value as well as efficacy in developing clinical guidelines (8). Evaluation of long-term effectiveness alongside lifetime cost in systematic cost-effectiveness studies can enhance the quality of health care and health policy decisions, as well as identify the critical gaps in knowledge that need to be addressed. The analysis by Cowper et al. (6) suggests that a 1-year course of ticagrelor and aspirin among PLATO-eligible patients with ACS appears to provide a reasonable value for the money spent, at least from the U.S. health care perspective. With time, a better understanding of the effectiveness of ticagrelor in the real world, particularly compared with contemporary alternatives such as prasugrel or genotype-tailored strategies, will help identify the most cost-effective strategy for managing antiplatelet therapy after ACS.
↵∗ Editorials published in the Journal of the American College of Cardiology reflect the views of the authors and do not necessarily represent the views of JACC or the American College of Cardiology.
Both authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- American College of Cardiology Foundation
- Go A.S.,
- Mozaffarian D.,
- Roger V.L.,
- et al.
- Cowper P.A.,
- Pan W.,
- Anstrom K.J.,
- et al.
- Mahaffey K.W.,
- Wojdyla D.M.,
- Carroll K.,
- et al.
- Anderson J.L.,
- Heidenreich P.A.,
- Barnett P.G.,
- et al.