Author + information
- Received May 10, 2012
- Revision received July 24, 2012
- Accepted August 1, 2012
- Published online November 13, 2012.
- David Erlinge, MD, PhD⁎,⁎ (, )
- Jurrien ten Berg, MD, PhD†,
- David Foley, MD‡,
- Dominick J. Angiolillo, MD, PhD§,
- Henrik Wagner, MD⁎,
- Patricia B. Brown, BSN, RN∥,
- Chunmei Zhou, MS∥,
- Junxiang Luo, PhD∥,
- Joseph A. Jakubowski, PhD∥,
- Brian Moser, MS∥,
- David S. Small, PhD∥,
- Thomas Bergmeijer, MD†,
- Stefan James, MD, PhD¶ and
- Kenneth J. Winters, MD∥
- ↵⁎Reprint requests and correspondence:
Dr. David Erlinge, Department of Cardiology, Lund University, Skane University Hospital, 221 85 Lund, Sweden
Objectives The aim of this study was to confirm prior modeling data suggesting that prasugrel 5 mg in low-body-weight (LBW) patients would be noninferior to prasugrel 10 mg in higher-body-weight (HBW) patients as assessed by maximal platelet aggregation (MPA).
Background Prasugrel 10 mg reduced ischemic events compared with clopidogrel 75 mg but increased bleeding, particularly in LBW patients.
Methods In this blinded, 3-period, crossover study in stable patients with coronary artery disease (CAD) taking aspirin, prasugrel 5 and 10 mg and clopidogrel 75 mg were administered to LBW (56.4 ± 3.7 kg; n = 34) and HBW patients (84.7 ± 14.9 kg; n = 38). Assays included light transmission aggregometry (LTA), VerifyNow P2Y12 (VN), and vasodilator-associated stimulated phosphoprotein (VASP) level measured predose and after each 12-day treatment.
Results Median MPA by LTA for prasugrel 5 mg in LBW patients was noninferior to the 75th percentile for prasugrel 10 mg in HBW patients (primary endpoint) and mean MPA was similar, but active metabolite exposure was lowered by 38%. Within LBW patients, prasugrel 5 mg lowered MPA more than clopidogrel (least squares mean difference [95% confidence interval]: −3.7% [−6.72%, −0.69%]) and resulted in lower rates of high on-treatment platelet reactivity (HPR). Within HBW patients, prasugrel 10 mg lowered MPA more than clopidogrel (−16.9% [−22.3%, −11.5%]). Similar results were observed by VN and VASP. Prasugrel 10 mg in LBW patients was associated with more mild to moderate bleeding (mainly bruising) compared with prasugrel 5 mg and clopidogrel.
Conclusions In aspirin-treated patients with CAD, prasugrel 5 mg in LBW patients reduced platelet reactivity to a similar extent as prasugrel 10 mg in HBW patients and resulted in greater platelet inhibition, lower HPR, and similar bleeding rates compared with clopidogrel. (Comparison of Prasugrel and Clopidogrel in Low Body Weight Versus Higher Body Weight With Coronary Artery Disease [FEATHER]; NCT01107925)
The TRITON–TIMI 38 (Trial to Assess Improvement in Therapeutic Outcomes by Optimizing Platelet Inhibition With Prasugrel–Thrombolysis in Myocardial Infarction) study demonstrated significantly reduced rates of ischemic events for prasugrel compared with clopidogrel for the composite efficacy endpoint of cardiovascular death, myocardial infarction, or stroke in patients with acute coronary syndrome (ACS) undergoing percutaneous coronary intervention (1). However, there was an increased risk of major bleeding events in patients treated with the 10-mg maintenance dose (MD) of prasugrel.
Analysis of TRITON–TIMI 38 data identified patients with low body weight (LBW) (<60 kg) as being at increased risk for bleeding on the 10-mg MD (2). As a result, a prasugrel 5-mg MD is recommended for use in LBW patients (3,4). Additional analyses modeling prasugrel's active metabolite concentrations by exposure quartiles during 10-mg MD in a population pharmacokinetic (PK) substudy (5) showed that higher exposure was associated with an increased risk of bleeding, and bleeding in LBW patients was shown to occur predominantly within the highest 2 exposure quartiles. Furthermore, LBW patients in the lowest 2 exposure quartiles demonstrated similar efficacy and risk of bleeding as higher-body-weight (HBW; ≥60 kg) patients. On the basis of this model, it was predicted that the prasugrel 5-mg MD in LBW patients would result in active metabolite exposure similar to that of the lowest quartile for the 10-mg MD—for which efficacy was maintained, but risk of bleeding was lowered (5).
We designed the FEATHER (Comparison of Prasugrel and Clopidogrel in Low Body Weight Versus Higher Body Weight With Coronary Artery Disease) trial to test the hypothesis that in aspirin-treated patients with stable coronary artery disease (CAD), the antiplatelet effect of a prasugrel 5-mg MD in LBW patients is noninferior to that of a prasugrel 10-mg MD in HBW patients. The approved clopidogrel 75-mg MD was also compared with prasugrel 5-mg and 10-mg MD by body weight cohort.
A detailed description of the trial design, participants, objectives and measures, and statistical analyses is provided in the supplemental methods.
FEATHER was an international, multicenter, blinded, active comparator phase Ib trial conducted in 72 randomized patients with stable CAD. The study design is illustrated in Figure 1. The study protocol was approved by local investigational review boards and was performed in compliance with the principles of good clinical practice and in accordance with the provision of the Declaration of Helsinki.
Patients enrolled in the FEATHER trial were men (n = 31) and women (n = 41) age 18 to 75 years receiving aspirin with a history of stable CAD. All patients provided signed informed consent.
Study objectives and measures
The primary objective of FEATHER was to compare maximal platelet aggregation (MPA) responses to 20 μM adenosine diphosphate (ADP) by light transmission aggregometry in LBW patients receiving prasugrel 5 mg versus HBW patients receiving prasugrel 10 mg (Fig. 2).
Baseline characteristics were compared between body weight (BW) cohorts using a 2-sample t test for continuous variables and a Pearson Chi-square test for categorical variables.
The primary noninferiority analysis compared the median of MPA to 20 μM ADP for prasugrel 5 mg in LBW patients with the 75th percentile of the MPA for prasugrel 10 mg in HBW patients at the end of period 1. The upper limit of the 1-sided 97.5% confidence interval (CI) of the difference was compared with the prespecified margin of 15 percentage points. The study had >80% power for the noninferiority test of the primary endpoint at a 1-sided 2.5% significance level.
Between BW cohorts, the mean comparison for pharmacodynamic (PD) parameters was conducted using analysis of covariance. The rate of high on-treatment platelet reactivity (HPR) was compared using the Fisher exact test. Within each BW cohort, PD response was compared using a mixed-effect model with repeated-measure analysis, whereas HPR was compared using a generalized linear model with repeated measures.
Patient disposition and baseline characteristics
Of 76 patients entering the study, 72 were enrolled and received study drug in period 1 (Fig. 2). The majority of participants had a history of prior myocardial infarction (58.3%), hypertension (63.9%), and/or hyperlipidemia (79.2%) (Table 1). Compared with HBW, LBW patients were more likely to be women, to be smokers, and to have a prior history of coronary artery bypass graft, but overall the 2 cohorts were well matched (Table 1).
Responses for MPA to 20 μM ADP (Fig. 3) were not significantly different between BW cohorts at baseline (LBW 76.3% ± 9.5%; HBW 77.9 ± 12.3%; p = 0.53). Relative to baseline, each treatment regimen substantially reduced aggregation response in both BW cohorts. The primary endpoint of the trial demonstrated that prasugrel 5-mg treatment in LBW (median 47.0%) patients was noninferior to prasugrel 10 mg (75th percentile 57.1%) in HBW patients (difference [95% CI]: −10.1% [−23.4%, 0.2%]). Specifically, the upper bound of the 97.5% CI for the primary comparison was 0.20%, well below the prespecified margin of 15% for noninferiority. During period 1, the mean MPA to 20 μM ADP for LBW patients receiving prasugrel 5 mg was not significantly different than the mean MPA in HBW patients receiving prasugrel 10 mg (least squares mean difference [95% CI]: 5.5% [−0.8%, 11.9%]) (Online Table 1). In addition, in the LBW cohort, prasugrel 5 mg was associated with a modestly but significantly lower MPA to 20 μM ADP versus clopidogrel 75 mg (difference −3.7% [−6.72%, −0.69%]). Prasugrel 10-mg treatment was associated with a more substantial reduction in aggregation response versus clopidogrel 75 mg in HBW patients (−16.9% [−22.3%, −11.5%]).
Aggregation responses for prasugrel 5 mg in LBW versus prasugrel 10 mg in HBW patients (Fig. 4) were not significantly different as measured by MPA to 5 μM ADP, residual platelet aggregation to 20 μM ADP, inhibition of platelet aggregation to 20 μM ADP, P2Y12 reaction unit values, and vasodilator-associated stimulated phosphoprotein platelet reactivity index (VASP-PRI). Mean data for these and other aggregation parameters are provided in Online Table 1. As shown in Figure 5, compared with clopidogrel 75 mg, prasugrel 5 mg was associated with numerically greater antiplatelet effects across the measures assessed, with some differences achieving statistical significance.
The proportion of patients exhibiting HPR was not significantly different between LBW patients receiving prasugrel 5 mg and HBW patients receiving 10 mg (Online Table 2). In the LBW cohort, a smaller proportion of prasugrel 5-mg patients exhibited MPA to 20 μM ADP >50%, residual platelet aggregation to 5 μM >14%, VerifyNow P2Y12 >235 P2Y12 reaction units, and VASP-PRI ≥50% compared with clopidogrel 75-mg patients, and this difference reached statistical significance for VASP-PRI ≥50% (Table 2). In the HBW cohort, prasugrel 10 mg was associated with a significantly smaller proportion of HPR versus clopidogrel in each measure assessed, except VerifyNow P2Y12 <15%.
PK of prasugrel's and clopidogrel's active metabolites
In general, exposures were higher for a given regimen in LBW patients relative to their HBW counterparts (Fig. 6). In patients receiving prasugrel 10 mg, the geometric mean area under the concentration curve (AUC0–tlast) was 27% higher in LBW versus HBW patients (59.3 vs. 46.7 ng*h/ml, respectively). In patients receiving prasugrel 5 mg, the mean AUC0–tlast was 45% higher in LBW versus HBW patients (28.9 vs. 19.9 ng*h/ml, respectively). For the comparison most germane to the study's primary objective, the geometric least squares mean AUC0–tlast for LBW patients receiving prasugrel 5 mg was 62% of that observed for HBW patients receiving prasugrel 10 mg. As shown in Figure 6, ≥90% of the prasugrel active metabolite AUCs in LBW patients receiving 5 mg fell below the median AUC in HBW patients receiving 10 mg, indicating that the reduced dosing regimen in LBW patients generally achieved active metabolite exposures that did not exceed those in the lower 2 quartiles of the 10-mg MD in HBW patients. Active metabolite exposures were also higher for prasugrel 5 mg versus clopidogrel 75 mg in the LBW (28.9 vs. 17.9 ng*h/ml) and HBW cohorts (19.9 vs. 12.8 ng*h/ml).
Safety and tolerability
Overall, for any given treatment, LBW patients experienced a higher incidence of treatment-related bleeding adverse events relative to HBW patients (Table 3). The majority of these events were minor contusions and/or hematomas that were not clinically significant. No drug-related bleeding events were considered severe or resulted in study drug discontinuation or withdrawal from the trial. Bleeding-related adverse event rates were similar following prasugrel 5-mg and clopidogrel 75-mg treatment in LBW patients (29.4% vs. 25.0%, respectively; p = 0.62) and across all treatment groups in HBW patients (prasugrel 5 mg 16.7%; prasugrel 10 mg 15.8%; clopidogrel 75 mg 16.2%).
The study met the primary endpoint for noninferiority between prasugrel 5 mg in LBW and prasugrel 10 mg in HBW patients, with the relevant comparison falling well below the predefined noninferiority margin; therefore, reducing the prasugrel MD from 10 mg to 5 mg did not result in an unacceptable loss of antiplatelet effect in LBW patients. Interestingly, we found that the 5-mg MD of prasugrel in LBW patients resulted in similar levels of platelet inhibition compared with the 10-mg MD in HBW patients. In addition, the 5-mg prasugrel MD in LBW patients resulted in similar rates of HPR as prasugrel 10 mg in HBW patients. Thus, in aspirin-treated patients with stable CAD, the prasugrel 5-mg MD in LBW patients had a similar platelet inhibitory effect as the 10-mg MD in HBW patients.
As predicted by previous PK/PD modeling (2), prasugrel 5 mg in LBW patients resulted in lower active metabolite exposure relative to prasugrel 10 mg in HBW patients; however, the PD responses were not significantly different. The similar PD responses observed were likely explained by the previously reported nonlinear relationship between active metabolite exposure and antiplatelet effect (6,7). The present data thus suggest that the lower active metabolite exposure generated by the prasugrel 5-mg MD is sufficient for LBW patients to achieve adequate platelet inhibition while potentially avoiding the risk of excessive bleeding reported in the TRITON–TIMI 38 trial in LBW patients receiving prasugrel 10-mg MD (2,5).
The prasugrel 5-mg MD approved for LBW patients was also compared with the clopidogrel 75-mg MD in LBW patients. Across platelet function measurements, prasugrel 5 mg reduced platelet activation more than clopidogrel 75 mg in LBW patients at or near significant levels. Prasugrel 5 mg also resulted in generally lower HPR compared with clopidogrel in LBW patients, consistent with the higher active metabolite exposures produced by prasugrel 5 mg compared with clopidogrel 75 mg. Thus, the prasugrel 5-mg MD offers the potential for greater clinical benefit compared with clopidogrel 75-mg MD in LBW patients. This question is being assessed directly in the TRILOGY-ACS (Targeted Platelet Inhibition to Clarify the Optimal Strategy to Medically Manage Acute Coronary Syndromes) trial comparing the use of prasugrel versus clopidogrel in patients with non–ST-segment elevation ACS (8).
Although the trial was not powered for evaluation of any clinical events, bleeding complications were closely monitored. There were no major drug-related bleeding events. Across treatments, LBW patients experienced more minor bleeding events, such as contusions and hematomas, compared with HBW patients. Prasugrel 10 mg generally caused more bleeding events than the other 2 treatment regimens. Prasugrel 5 mg resulted in fewer bleeding events compared with prasugrel 10 mg and a similar event rate compared with clopidogrel.
This study was limited by being a PK/PD study in a small number of stable patients with CAD treated for a relatively short period of time, and it was not intended to assess clinical outcomes. In addition, the demographic data showed some unavoidable imbalances between the 2 BW cohorts: LBW patients were more likely to be women, to have a previous history of coronary artery bypass graft or percutaneous coronary intervention, to use fewer angiotensin-converting enzyme inhibitor/angiotensin II receptor blocker comedications, and to be smokers. Female sex and smoking have been associated with altered platelet responses previously (9,10) and could have affected platelet response to a minor extent.
In summary, prasugrel 5-mg MD treatment in LBW patients resulted in platelet inhibition that was noninferior compared with that of prasugrel 10-mg MD in HBW patients and achieved active metabolite exposures in a range in which efficacy was maintained and the risk of bleeding was lower than that in the TRITON−TIMI 38 study (5). The PK and PD data were consistent with the lower minor bleeding rates for prasugrel 5 mg (similar to rates for clopidogrel 75 mg) compared with prasugrel 10 mg. These results indicate that the prasugrel 5-mg MD could provide better risk–benefit balance for LBW patients with CAD and support the recommended use of the prasugrel 5-mg MD in LBW patients (3,4). Additional randomized, controlled studies, such as the TRILOGY-ACS trial comparing prasugrel 5-mg MD with clopidogrel 75 mg in LBW patients, are necessary to verify the clinical relevance of these results (8).
The FEATHER study was conducted at the following clinical research institutions: Dr. Angiolillo, University of Florida Health Science Center, University of Florida College of Medicine, Jacksonville, Florida; Dr. Erlinge, Department of Cardiology, Lund University, Skane University Hospital, Lund, Sweden; Dr. Foley, Beaumont Hospital, Clinical Research Centre, Dublin, Ireland; Dr. James, Uppsala Clinical Research Center, Uppsala University Hospital, Uppsala, Sweden; Dr. Ten Berg, Department of Cardiology, St. Antonius, Nieuwegein, the Netherlands. Keri Poi (Eli Lilly and Co.) and Michael Mihm (PharmaNet/i3) were involved in the editing of this manuscript.
For an expanded Methods section, and supplemental tables and references, please see the online version of this article.
The study was funded by Daiichi Sankyo Co., Ltd. and Eli Lilly and Co. Dr. Erlinge has received speaker's fees from Eli Lilly and Co., AstraZeneca, sanofi-aventis, and Accumetrics; and has served on advisory boards for AstraZeneca, Eli Lilly and Co., and Merck. Dr. Angiolillo has reported receiving honoraria for lectures from Bristol-Myers Squibb, sanofi-aventis, Eli Lilly and Co., Daiichi Sankyo, Co., Ltd., and AstraZeneca; consulting fees from Bristol-Myers Squibb, sanofi-aventis, Eli Lilly and Co., Daiichi Sankyo Co., Ltd., The Medicines Co., Portola, Novartis, Medicure, Accumetrics, Arena Pharmaceuticals, Abbott Vascular, and AstraZeneca; and research grants from Bristol-Myers Squibb, sanofi-aventis, GlaxoSmithKline, Otsuka, Eli Lilly and Co., Daiichi Sankyo Co., Ltd., The Medicines Co., Portola, Accumetrics, Schering-Plough, AstraZeneca, and Eisai. Dr. James has received institutional research grants and honoraria from AstraZeneca, Eli Lilly and Co., Merck, and Bristol-Myers Squibb; has served on advisory boards for AstraZeneca, Eli Lilly and Co., and Merck; and has received honoraria from the Medicines Co. Ms. Brown, Ms. Zhou, Drs. Luo, Jakubowski, Small, and Winters, and Mr. Moser are employees of Eli Lilly and Co. Ms. Zhou, and Drs. Jakubowski, and Winters, Ms. Brown, and Mr. Moser, are stockholders in Eli Lilly & Company. Dr. Ten Berg is a consultant to Eli Lilly & Company, Merck, and AstraZeneca. All other authors have reported that they have no relationships to disclose relevant to the contents of this paper.
- Abbreviations and Acronyms
- acute coronary syndrome
- adenosine diphosphate
- area under the concentration curve from time 0 to last measure
- body weight
- coronary artery bypass graft
- coronary artery disease
- higher body weight
- high on-treatment platelet reactivity
- low body weight
- maintenance dose
- maximal platelet aggregation
- vasodilator-associated stimulated phosphoprotein platelet reactivity index
- Received May 10, 2012.
- Revision received July 24, 2012.
- Accepted August 1, 2012.
- American College of Cardiology Foundation
- ↵(2012) Prasugrel (Effient): Package Insert (Ely Lilly and Co. Indianapolis, IN).
- (2012) Prasugrel (Efient) European Union Summary of Product Characteristics (Ely Lilly and Co. Indianapolis, IN).
- Chin C.T.,
- Roe M.T.,
- Fox K.A.,
- et al.
- Bliden K.P.,
- DiChiara J.,
- Lawal L.,
- et al.