Author + information
- Received October 10, 2012
- Revision received April 11, 2013
- Accepted May 7, 2013
- Published online August 13, 2013.
- David Erlinge, MD, PhD∗∗ (, )
- Paul A. Gurbel, MD†,
- Stefan James, MD, PhD‡,
- Tomas L. Lindahl, MD, PhD§,
- Peter Svensson, MD, PhD⋮,
- Jurrien M. Ten Berg, MD, PhD¶,
- David P. Foley, MBBCh, PhD#,
- Henrik Wagner, MD∗,
- Patricia B. Brown, BSN, RN∗∗,
- Junxiang Luo, PhD∗∗,
- Chunmei Zhou, MS∗∗,
- Brian A. Moser, MS∗∗,
- Joseph A. Jakubowski, PhD∗∗,
- David S. Small, PhD∗∗,
- Kenneth J. Winters, MD∗∗ and
- Dominick J. Angiolillo, MD, PhD††
- ∗Department of Cardiology, Lund University, Lund, Sweden
- †Sinai Center for Thrombosis Research, Baltimore, Maryland
- ‡Department of Medical Sciences, Cardiology, and Uppsala Clinical Research Center, Uppsala University, Uppsala, Sweden
- §Department of Clinical and Experimental Medicine, Linköping University, Linköping, Sweden
- ⋮Department of Coagulation Medicine, Lund University, Malmö, Sweden
- ¶Department Cardiology, St. Antonius Hospital, Nieuwegein, the Netherlands
- #Clinical Research Centre, Beaumont Hospital, Dublin, Ireland
- ∗∗Eli Lilly and Company, Indianapolis, Indiana
- ††Cardiovascular Research Center at University of Florida College of Medicine-Jacksonville, Jacksonville, Florida
- ↵∗Reprint requests and correspondence:
Dr. David Erlinge, Department of Cardiology, Lund University, Skane University Hospital, 221 85 Lund, Sweden.
Objectives This study assessed pharmacodynamic (PD) response to the reduced prasugrel maintenance dose of 5 mg in very elderly (VE) patients (≥75 years of age).
Background In the TRITON–TIMI 38 (TRial to Assess Improvement in Therapeutic Outcomes by Optimizing Platelet InhibitioN with Prasugrel–Thrombolysis In Myocardial Infarction 38) study prasugrel 10 mg reduced ischemic events versus clopidogrel 75 mg, but increased bleeding in VE patients.
Methods We examined PD and active metabolite pharmacokinetics (PKs) with prasugrel 5 and 10 mg and clopidogrel 75 mg in a 3-period (12 days each) blinded, crossover study in VE (n = 73; mean: 79 ± 3 years of age) or (n = 82) nonelderly (NE) (≥45 to <65 years of age; mean: 56 ± 5 years of age) stable coronary artery disease (CAD) patients receiving background aspirin. Assays included light transmission aggregometry (LTA), VerifyNow P2Y12 (VN-P2Y12), and vasodilator-associated stimulated phosphoprotein (VASP). The primary comparison was noninferiority of maximum platelet aggregation (MPA) comparing the median for prasugrel 5 mg in VE versus the 75th percentile for prasugrel 10 mg in NE, using a pre-specified 1-sided 97.5% confidence interval for the difference <15%.
Results Prasugrel 5 mg in VE met the primary PD noninferiority criterion versus prasugrel 10 mg in NE. For prasugrel 5 mg, MPA was significantly lower (57 ± 14%) than clopidogrel (63 ± 14%; p < 0.001) in VE but higher than prasugrel 10 mg in NE (46 ± 12%; p < 0.001). PD response by LTA, VN-P2Y12, and VASP during all treatments appeared similar between age cohorts. Prasugrel 5 mg resulted in fewer VE poor responders than clopidogrel. Rates of mild bleeding were higher with prasugrel 10 mg but similar for prasugrel 5 mg versus clopidogrel 75 mg.
Conclusions In aspirin-treated stable CAD patients, prasugrel 5 mg in VE attenuated platelet inhibition while meeting pre-specified noninferiority criterion versus prasugrel 10 mg in NE, with significantly better PD response and fewer poor responders compared to clopidogrel 75 mg in VE. (Comparison of Prasugrel and Clopidogrel in Very Elderly and Non-Elderly Patients With Stable Coronary Artery Disease [GENERATIONS]; NCT01107912)
The TRITON–TIMI 38 (TRial to Assess Improvement in Therapeutic Outcomes by Optimizing Platelet InhibitioN with Prasugrel–Thrombolysis In Myocardial Infarction 38) study demonstrated significantly reduced rates of the composite efficacy endpoint of cardiovascular death, myocardial infarction, or stroke for prasugrel versus clopidogrel in patients with acute coronary syndrome (ACS) undergoing percutaneous coronary intervention (1). However, risk of major bleeding events was increased in very elderly (VE) patients receiving the prasugrel 10-mg maintenance dose (MD).
A prasugrel 5-mg MD should reduce bleeding risk, based on population pharmacokinetic (PK) substudy modeling in TRITON–TIMI 38 (2), which predicted active metabolite exposures with prasugrel 5 mg in VE patients, similar to the lowest exposure quartile in TRITON patients, where efficacy was maintained but bleeding risk was lower (2,3). However, direct clinical, pharmacodynamic (PD), and PK data to support the potential use of prasugrel 5-mg MD in VE patients are limited. We therefore designed the GENERATIONS (Comparison of Prasugrel and Clopidogrel in Very Elderly and Non-Elderly Patients With Stable Coronary Artery Disease) trial to test, in aspirin-treated patients with stable coronary artery disease (CAD), whether prasugrel 5 mg in VE patients would meet pre-specified antiplatelet noninferiority criteria versus prasugrel 10 mg in nonelderly (NE) (≥45 to <65 years of age) patients. The approved clopidogrel 75-mg MD was also assessed versus prasugrel 5 mg and 10 mg.
The GENERATIONS trial was an international Phase 1b trial. A detailed description of the trial design (Online Fig. 1; similar to the FEATHER study ) and methods is provided in the Online Appendix. Briefly, an aspirin run-in was followed by a baseline/randomization visit, then 3 crossover treatment periods (12 ± 2 days each). During treatment period 1 (single-blinded), all VE patients received prasugrel 5 mg and NE patients received prasugrel 10 mg. During periods 2 and 3 (double-blinded), VE patients received either prasugrel 10 mg (period 2) then clopidogrel 75 mg (period 3) or the reverse; NE patients received prasugrel 5 mg and then clopidogrel 75 mg or the reverse.
The protocol was approved by local investigational review boards and performed in compliance with good clinical practice and the Declaration of Helsinki. All patients provided signed informed consent prior to any study procedure.
The primary objective was a noninferiority analysis of maximum platelet aggregation (MPA) to 20 μM adenosine diphosphate (ADP) by light transmission aggregometry (LTA), comparing the median for prasugrel 5 mg in VE patients versus the 75th percentile for prasugrel 10 mg in NE patients (Online Fig. 2). This value reflected the goal of reducing PD effect in VE to the lower range of PD effect for prasugrel 10 mg in NE. The upper limit of the 1-sided 97.5% confidence interval (CI) of the difference was compared with the pre-specified, noninferiority margin of 15 percentage points (Online Appendix). The primary analysis had >80% power at a 1-sided 2.5% significance level.
Between-age cohort comparisons of PD response used analysis of covariance, and comparison of high on-treatment platelet reactivity (HPR) used Fisher exact test.
Among randomized patients (N = 155) (Online Fig. 3), most had a history of myocardial infarction (68.4%), hypertension (70.3%), and/or hyperlipidemia (85.2%); 22.6% had diabetes (Table 1). Mean ages were 78.9 (VE) and 56.3 (NE) years old. At screening, VE (versus NE) had significantly lower mean weight and body mass index, higher mean systolic blood pressure, and a lower percentage currently used tobacco (Table 1) (all p < 0.05). The 2 cohorts were otherwise well matched.
Platelet reactivity at baseline was comparable between age cohorts according to LTA measurements and vasodilator-associated stimulated phosphoprotein platelet reactivity index (VASP-PRI) but significantly greater in VE for VerifyNow P2Y12 (Accumetrics Inc., San Diego, California) (VN-P2Y12; p < 0.005) (Online Table 1).
Median MPA response to prasugrel 5 mg in VE (58.0%) was noninferior to the 75th percentile of MPA response to prasugrel 10 mg in NE (52.0%; difference, 6.0% [95% CI: 1.0% to 9.0%]), as the upper boundary of the CI was 9.0%, below the pre-specified 15% margin (Fig. 1). Nonetheless, antiplatelet effect was significantly lower with prasugrel 5 mg in VE than 10 mg in NE, based on higher mean MPA to 20 μM ADP (LS [least squares] mean difference, 91.4% [95% CI: 5.4% to 13.3%]) and other aggregation measures (all p < 0.001) (Fig. 2) and significantly greater HPR rates (all p <0.001) (Table 2). Comparing prasugrel 5 mg and clopidogrel 75 mg in VE, antiplatelet effect was greater with prasugrel 5 mg based on significantly lower MPA to 20 μM ADP (LSmean difference: −5.7% [95% CI: −8.4% to −3.0%]) and other measures (all p < 0.001) (Fig. 3) and lower HPR rates (all measures p < 0.05) (Fig. 4). In NE patients, prasugrel 10 mg further reduced aggregation response (MPA to 20 μM ADP LSmean difference: −13.3% [95% CI: −16.1% to −10.5%]) and HPR rates (Table 2) versus clopidogrel 75 mg.
In a comparison of age groups, PD responses during treatment were generally similar (Online Table 1, Fig. 1), except for greater platelet inhibition in VE patients by VASP-PRI, with both prasugrel doses (p < 0.05) and reduced inhibition in VE patients by VN-P2Y12 with clopidogrel 75 mg (p = 0.011).
For each treatment, active metabolite exposures (Fig. 5) were slightly greater in VE than in NE patients; the difference in area under the concentration-time curve, time 0 to last quantifiable concentration (AUC(0-tlast) for prasugrel 10 mg was 12% (41.2 vs. 36.7 ng/h/ml) and 17% for 5 mg (18.9 vs. 16.1 ng/h/ml, respectively). For VE receiving prasugrel 5 mg relative to NE receiving prasugrel 10 mg, the ratio of geometric means of AUC(0-tlast) was 51%. Active metabolite exposures were higher for prasugrel 5 mg than for clopidogrel 75 mg in each age cohort.
Safety and tolerability
Bleeding-related adverse event rates were similar between age groups (Table 3); most events were mild contusions and/or hematomas and were not clinically significant. Between treatments, rates of bleeding events with prasugrel 5 mg and clopidogrel 75 mg were similar and significantly lower than prasugrel 10 mg (see the Online Appendix for additional safety results).
In this PK/PD study in stable CAD patients receiving aspirin, MPA for prasugrel 5 mg in VE met noninferiority criterion versus prasugrel 10 mg in NE patients. Furthermore, in VE patients, platelet inhibition was significantly greater for prasugrel 5 mg than for clopidogrel 75 mg, while rates of bleeding events (mainly mild bruising) were similar between prasugrel 5 mg and clopidogrel 75 mg and were lower than prasugrel 10 mg.
Although the current study met the primary PD noninferiority endpoint, platelet inhibition was significantly lower and HPR rates were significantly higher for prasugrel 5 mg in VE than prasugrel 10 mg in NE patients. This agrees with prior modeling (2) and reflects an expected reduction in PD effect intended to reduce bleeding risk in VE patients (1). The PK of prasugrel active metabolite (Pras-AM) has been assessed in elderly individuals in 2 previous studies. One study found no significant difference in exposure between healthy elderly subjects (65 to 80 years of age) and younger (20 to 39 years of age) during prasugrel 5 mg and 10 mg (with background aspirin) (5). In the TRITON–TIMI 38 study, VE patients had higher exposure than patients <75 years of age, although differences were largely explained by body weight (3). In the current study, prasugrel 5 mg resulted in ∼50% lower Pras-AM concentrations than prasugrel 10 mg within age cohorts, with only slightly higher Pras-AM exposure in VE than in NE patients, likely reflecting body weight differences (4).
PD response did not generally differ between age groups. This contrasts somewhat with findings of reduced (clopidogrel) or borderline reduced (prasugrel) antiplatelet effect in VE patients in the SENIOR-PLATELET observational study (6); however, there were major baseline clinical differences between cohorts in that study. Our findings are consistent with a meta-analysis of randomized, controlled, phase I/II studies, which found no age-dependent differences (7).
If used in VE patients, the prasugrel 5-mg MD is recommended in European labeling (8). Across measurements in this study, prasugrel 5 mg reduced platelet reactivity significantly versus clopidogrel 75 mg, with lower HPR rates. This is consistent with the previously reported increase in platelet inhibition observed for VE clopidogrel poor responders following a switch to prasugrel 5 mg (9), greater antiplatelet effects in VE ACS/percutaneous coronary intervention (PCI) patients with HPR on standard clopidogrel therapy, when switched to prasugrel 5-mg versus double-dose clopidogrel (150 mg) (10); and greater platelet inhibition versus clopidogrel 75 mg with prasugrel 10 mg or 5 mg (in VE and patients <60 kg) in the TRILOGY ACS study in medically treated patients (11). However, in TRILOGY ACS, no significant differences were seen between treatments in the primary clinical endpoint (12). One interpretation has been that in medically treated, nonstented patients, further platelet inhibition does not provide additional benefit, which is in contrast to a previous study in PCI-treated patients (TRITON) (1).
This study was not designed to assess clinical outcomes; however, we observed no major drug-related bleeding events, but relatively frequent minor contusions, hematomas, or epistaxis-related bleeding events. Prasugrel 5-mg treatment resulted in a bleeding event rate lower versus prasugrel 10 mg and similar to clopidogrel-treated patients.
This study was limited in that, as a PK/PD study, duration was limited, and it was not intended to assess clinical outcomes. Dosing for some age and treatment comparisons occurred in separate periods due to the crossover design. In addition, VE patients had lower mean body weight and body mass index despite entry criteria requiring body weight >60 kg; also, fewer subjects were smokers versus NE patients, factors which could potentially have impacted platelet responses somewhat.
In summary, PD response to prasugrel 5 mg in VE patients met the noninferiority criterion of the primary comparison to prasugrel 10 mg in NE patients, but platelet inhibition was significantly lower. Prasugrel 5 mg generated greater active metabolite exposure and antiplatelet PD effect than clopidogrel 75 mg in VE patients, with lower HPR rates, while minor bleeding rates for prasugrel 5 mg were lower than prasugrel 10 mg and similar to clopidogrel-treated patients. These results suggest that maintenance dosing with prasugrel 5 mg compared to prasugrel 10 mg could provide better risk-benefit balance for VE ACS patients with lower rates of HPR and a similar bleeding risk compared to clopidogrel 75 mg. Additional randomized, controlled clinical trials are necessary to verify the clinical relevance of these results, particularly in ACS patients undergoing PCI.
This study was funded by Daiichi Sankyo Company Ltd. and Eli Lilly and Company. Dr. Erlinge has received speaker fees from Daiichi Sankyo Company, Ltd., Eli Lilly and Co., AstraZeneca, Sanofi-Aventis, and Accumetrics; and for membership on advisory boards of AstraZeneca, Eli Lilly and Co., and Merck. Dr. Gurbel has received research funding, consultation fees, and/or honoraria from AstraZeneca, Daiichi Sankyo Company, Ltd., Eli Lilly and Co., Pozen, Bayer Healthcare, Sanofi-Aventis, CSL Pharmaceuticals, Accumetrics, Nanosphere, and Haemoscope; and honoraria from Merck, Daiichi Sankyo Company, Ltd., Eli Lilly and Co., Boerhinger-Ingleheim, Johnson & Johnson, AstraZeneca, and the Discovery Channel. Dr. James has received institutional research grants and honoraria from AstraZeneca, Eli Lilly and Co., Merck, and Bristol-Myers Squibb; fees for being an advisory board member for AstraZeneca, Eli Lilly and Company, and Merck; and honoraria from The Medicines Company. Dr. Lindahl has received speaker fees from Boehringer-Ingelheim, Octapharma, Leo Pharma, and Roche Diagnostics; has served as an advisory board member for Boehringer-Ingelheim and Bristol-Myers Squibb; and has served as a member of the board of Medirox and Nordic Haemostasis. Dr. Svensson has received speaker fees from Bayer, Boehringer-Ingelheim, Octapharma, and Roche Diagnostics; and has served as an advisory board member for Boehringer-Ingelheim, Bristol-Myers Squibb, and Pfizer. Dr. Angiolillo has received honoraria for lectures from AstraZeneca, Bristol-Myers Squibb, Daiichi Sankyo Company, Ltd., Eli Lilly and Co., and Sanofi-Aventis; has received consulting fees from Abbott Vascular, Accumetrics, Arena Pharmaceuticals, AstraZeneca, Bristol-Myers Squibb, Daiichi Sankyo Company, Ltd., Eli Lilly and Co., Medicure, Novartis, Portola, Sanofi-Aventis, and The Medicines Company; and has received research grants from Accumetrics, AstraZeneca, Bristol-Myers Squibb, Daiichi Sankyo Company, Ltd., Eli Lilly and Co., Eisai, GlaxoSmithKline, Otsuka, Portola, Sanofi-Aventis, Schering-Plough, and The Medicines Company. Drs. Brown, Luo, Zhou, Moser, Jakubowski, Small, and Winters are employees and shareholders of Eli Lilly and Co.
- Abbreviations and Acronyms
- area under the concentration-time curve, time 0 to last quantifiable concentration ≤4 h post-dose
- coronary artery disease
- high on-treatment platelet reactivity
- light transmission aggregometry
- maintenance dose
- maximum platelet aggregation
- nonelderly (≥45 to <65 years of age)
- residual platelet aggregation
- vasodilator-associated stimulated phosphoprotein platelet reactivity index
- very elderly (≥75 years of age)
- Received October 10, 2012.
- Revision received April 11, 2013.
- Accepted May 7, 2013.
- American College of Cardiology Foundation
- Erlinge D.,
- ten B.J.,
- Foley D.,
- et al.
- Silvain J.,
- Cayla G.,
- Hulot J.S.,
- et al.
- ↵(2012) Efient (prasugrel) European Union summary of product characteristics (Eli Lilly and Company, Indianapolis, IN).
- Capranzano P.,
- Tamburino C.,
- Capodanno D.,
- et al.