Comparison of Prasugrel and Clopidogrel in Thrombotic Stroke Patients with Risk Factors for Ischemic Stroke Recurrence: An Integrated Analysis of PRASTRO-I, PRASTRO-II, and PRASTRO-III

Commonly recommended antiplatelet agents for secondary prevention of non-cardioembolic stroke are clopidogrel and aspirin; however, a high risk of stroke recurrence (between 3% and 10% at 1 year after the index event) is reported with these treatments [1]. Genetic polymorphisms in cytochrome P450 (CYP) 2C19 have been identified as the major cause of a poor response to clopidogrel [2], raising concerns about clopidogrel monotherapy in Asian people, who are more likely than Caucasian people to be CYP2C19 poor metabolizers [3].

Prasugrel, a P2Y12 receptor antagonist, inhibits platelet aggregation more rapidly and consistently, and to a greater extent, than clopidogrel, in healthy subjects and patients with high on-treatment platelet reactivity after percutaneous coronary intervention independently of CYP2C19 genetic polymorphisms [4‒7]. The rapid and stable efficacy of prasugrel for patients with acute coronary syndrome was established in TRITON-TIMI 38 [8], and the safety and efficacy of low-dose prasugrel (3.75 mg daily) for these patients were verified in the PRASFIT-ACS trial in Japan [9]. A previous study in patients with non-cardioembolic stroke treated with clopidogrel reported that CYP2C19 polymorphisms did not affect the antiplatelet effects of prasugrel or the plasma concentrations of its active metabolite, which differed from observations with clopidogrel [10]. Thus, prasugrel is expected potentially be an alternative to clopidogrel in the treatment of ischemic cerebrovascular disease.

Three phase 3 trials of prasugrel have been conducted in Japanese patients with ischemic cerebrovascular disease: PRASTRO-I (JapicCTI-111582) [11], PRASTRO-II (JapicCTI-121901) [12], and PRASTRO-III (JapicCTI-184141) [13]. All three trials had the same primary efficacy endpoint, a composite incidence of stroke, myocardial infarction (MI), and death from other vascular causes [11‒13]. Although PRASTRO-I did not prove the non-inferiority of prasugrel to clopidogrel [11], subgroup analysis revealed a 20% risk reduction in patients with large-artery atherosclerosis or small-artery occlusion treated with prasugrel versus clopidogrel [14]. PRASTRO-II was conducted in Japanese patients who were elderly and/or had low bodyweight, as these patient populations were not included in PRASTRO-I. PRASTRO-II showed that the efficacy and safety of prasugrel and clopidogrel were similar [12]. PRASTRO-III, which included patients ≥50 years of age who had thrombotic stroke (large-artery atherosclerosis and small-artery occlusion) and were at risk of ischemic stroke (hypertension, dyslipidemia, diabetes mellitus, chronic kidney disease, or ischemic stroke history), showed the possible efficacy of prasugrel; however, there was no significant difference between treatment groups [13]. Based on these findings, prasugrel was approved in Japan for the indication of recurrence prevention after ischemic cerebrovascular disease (associated with large-artery atherosclerosis or small-artery occlusion) in patients with high stroke risk [15]. However, the accuracy of estimates was low given the small number of patients who completed PRASTRO-III (N = 230; prasugrel, N = 118; clopidogrel, N = 112) [13]. Thus, to confirm the generalizability of the PRASTRO-III findings and to supplement the small sample size, we conducted an integrated analysis of data from PRASTRO-III with data from patients in PRASTRO-I and PRASTRO-II who comprised the same patient population as that in PRASTRO-III.

In this integrated analysis of PRASTRO-I/II/III [11‒13], patients were treated with either prasugrel (3.75 mg) or clopidogrel (50 mg [PRASTRO-II and PRASTRO-III] or 75 mg [PRASTRO-I and PRASTRO-III]) orally once daily. Patients were followed up for 96–104 weeks (PRASTRO-I), 48 weeks (PRASTRO-II), or 24–48 weeks (PRASTRO-III).

Patients were selected from PRASTRO-I and II according to the selection criteria (risk factors for recurrent stroke) of PRASTRO-III for an integrated analysis. The full analysis set included patients with ischemic stroke (classified as large-artery atherosclerosis or small-artery occlusion according to the Trial of Org 10,172 in Acute Stroke Treatment (TOAST) classification [16]) and the presence of one or more of the following: hypertension, dyslipidemia, diabetes mellitus, chronic kidney disease, and history of ischemic stroke. The eligibility criteria for each individual study have been described previously [11‒13].

The primary efficacy endpoint was the composite incidence of ischemic stroke (fatal and nonfatal), MI (fatal and nonfatal), and death from other vascular causes in the intention-to-treat population. Bleeding events (life-threatening bleeding, major bleeding, and clinically relevant bleeding) were evaluated as the primary safety endpoint [13]. Adverse events were classified using the Medical Dictionary for Regulatory Activities/Japanese version 23.0.

The Kaplan-Meier method was used to calculate cumulative incidences and 95% confidence intervals (CIs) for the study outcomes. Hazard ratios (HRs) and 95% CIs were calculated using the Cox regression model. Sensitivity analysis was performed by weighting (age and bodyweight) on the background distribution of eligible patients. These factors were selected based on those used in previous studies [17, 18]. Statistical analyses were performed using SAS version 9.2 or later (SAS Institute Inc., Cary, NC, USA).

In total, 2,688 patients (PRASTRO-I, n = 2,184; PRA­STRO-II, n = 274; PRASTRO-III, n = 230) met the study criteria; both the safety analysis set and full analysis set included 2,688 patients (prasugrel, n = 1,337; clopidogrel, n = 1,351) (online suppl. Fig. 1; for all online suppl. material, see www.karger.com/doi/10.1159/0005282009149).

Patients’ baseline characteristics were similar between the groups (Table 1). At enrollment, stroke classification was large-artery atherosclerosis in 49.3% of patients and small-artery occlusion in 50.7% of patients. Reduced CYP2C19 function (i.e., CYP2C19 poor metabolizers) was reported in 17.4% of patients.

The composite incidence of the primary efficacy endpoint was numerically lower with prasugrel versus clopidogrel (3.4% vs. 4.3%, HR: 0.771, 95% CI: 0.522–1.138) (Table 2). The composite incidence was similar in both groups early in treatment; at around 240 days of treatment, a decreasing trend was observed with prasugrel versus clopidogrel treatment that continued to the end of the study (Fig. 1a). The incidences of each component of the primary efficacy endpoint were 3.1% versus 4.1% for ischemic stroke (HR: 0.741, 95% CI: 0.494–1.110) and 0.3% versus 0.2% for MI (HR: 1.323, 95% CI: 0.296–5.914); no events of death from other vascular causes were observed (Table 2). For most background factors, the incidence of the primary efficacy endpoint tended to be lower with prasugrel than clopidogrel (Fig. 2a), especially for patients who had creatinine clearance 30–50 (HR: 0.379, 95% CI: 0.094–1.522), were CYP2C19 poor metabolizers (HR 0.396, 95% CI: 0.153–1.020), and had a <4-week duration from onset (HR: 0.491, 95% CI: 0.239–1.007). In the sensitivity analysis, which was weighted by age and bodyweight, the incidence of the primary efficacy endpoint was 2.5% (33.5/1,342.4) in the prasugrel group and 3.8% (51.1/1,345.6) in the clopidogrel group (HR: 0.638, 95% CI: 0.413–0.987).

The composite incidence of the primary safety endpoint for prasugrel versus clopidogrel was 6.0% versus 5.5% (HR: 1.074, 95% CI: 0.783–1.473) (Table 2; Fig. 1b). The incidence of each bleeding event for prasugrel versus clopidogrel was 1.2% (n = 16) versus 1.1% (n = 15) for life-threatening bleeding, 0.1% (n = 2) versus 0.1% (n = 1) for major bleeding, and 4.8% (n = 64) versus 4.3% (n = 58) for clinically relevant bleeding (Table 2). The incidence of the primary safety endpoint by patient background is shown in Figure 2b. A full list of adverse events is provided in online supplementary Table 1. Sensitivity analysis weighted by age and bodyweight revealed an incidence of 5.6% (75.5/1,342.4) in the prasugrel group and 5.1% (68.2/1,345.6) in the clopidogrel group for the primary safety endpoint (HR: 1.080, 95% CI: 0.778–1.498).

In this integrated analysis of the PRASTRO-I/II/III studies, a lower overall incidence of efficacy events was observed for prasugrel than clopidogrel in patients with ischemic stroke who were at a high risk of stroke recurrence. Although PRASTRO-I included patients with ischemic stroke of undetermined etiology (TOAST classification) [16], subgroup analysis found potential differences in responses among stroke subtypes. Patients with ischemic stroke of unknown cause responded less well to prasugrel than clopidogrel (primary event incidence: 4.6% vs. 3.0%, respectively), indicating that these patients may be poor candidates for more potent antiplatelet therapy [14]. Additionally, older (≥75 years) and underweight (≤50 kg) patients were excluded [11]. PRASTRO-II was limited whereby it only included older (≥75 years) and underweight (≤50 kg) patients. Moreover, the sample size was small (N = 654) [12]. PRASTRO-III limited enrollment to patients with stroke subtypes most likely to benefit from antiplatelet treatment (i.e., large-artery atherosclerosis and small-artery occlusion) who had a high stroke risk; however, the sample size was small (prasugrel, n = 118; clopidogrel, n = 112) [13].

The present analysis selected patients from PRASTRO-I and PRASTRO-II who were targeted in the PRASTRO-III trial, merging patients from all three trials to evaluate the generalizability of the PRASTRO-III findings and to supplement the small population. For the primary efficacy endpoint composite incidence, the between-group difference was not significant (HR: 0.771, 95% CI: 0.522–1.138); the HR was <1, which was a more pronounced result than observed in PRASTRO-III. This analysis demonstrates that the strong antiplatelet effects of prasugrel can reduce stroke recurrence in patients with large-artery atherosclerosis or small-artery occlusion who are at a high risk of ischemic stroke recurrence. Because age and bodyweight differed between these studies, a weighted sensitivity analysis was conducted based on the background factor distribution in a real-world clinical setting, but the trend for efficacy and safety remained the same.

Previous trials have reported that aspirin reduces the risk of recurrent ischemic events by about 20% versus placebo or no treatment [19, 20]; clopidogrel has been reported to reduce the risk of ischemic events by 7.3% compared with aspirin in a study conducted outside Japan [21]. Based on this evidence, aspirin and clopidogrel are recommended as first-line therapies for the prevention of recurrent ischemic stroke [22, 23]. In this study, limited to patients with ischemic stroke classified as large-artery atherosclerosis or small-artery occlusion, >20% risk reduction was achieved with prasugrel compared with clopidogrel. Prasugrel may contribute to secondary prevention in stroke patients at high risk of recurrence.

In PRASTRO-III, bleeding events were reported for six patients (5.0%) in the prasugrel group and four (3.5%) in the clopidogrel group; however, these findings could not be fully evaluated because of the small number of events [13]. In this integrated study, bleeding events occurred in 80 patients (6.0%) in the prasugrel group and 74 (5.5%) in the clopidogrel group, showing a nonsignificant but numerically higher incidence in the prasugrel group, which is similar to the findings of PRASTRO-III. Overall, there were no major safety issues observed for prasugrel.

The PENDULUM registry reported that ischemic stroke, 12 months after percutaneous coronary intervention, is more common in patients with higher platelet aggregation [24]. Previous studies have shown that drugs with strong antiplatelet action (e.g., prasugrel, ticagrelor) are effective at reducing cardiovascular events in patients with atherosclerotic stroke [8, 25]. In these studies, there was no increase in bleeding events with prasugrel versus clopidogrel, possibly because of the low dose of prasugrel used (3.75 mg/day). These results are consistent with those of PRASFIT-ACS, which used a 3.75 mg/day prasugrel maintenance dose and reported a similar bleeding incidence between prasugrel and clopidogrel [9]. The subgroup analysis favored prasugrel over clopidogrel overall including all stroke risk factors.

Notably, when considering CYP2C19 gene polymorphisms, the HR (prasugrel vs. clopidogrel) of the primary efficacy endpoint was 0.396 (95% CI: 0.153–1.020) in poor metabolizers and 1.517 (95% CI: 0.884–2.602) in extensive metabolizers. The CHANCE-2 trial demonstrated a modestly reduced stroke risk at 90 days after minor ischemic stroke or transient ischemic attack with ticagrelor compared with clopidogrel among CYP2C19 loss-of-function carriers (HR 0.77, 95% CI: 0.64–0.94) [26]. However, bleeding events were more frequent with ticagrelor (5.3%) than clopidogrel (2.5%). Although prasugrel may be a treatment option for patients who are CYP2C19 poor metabolizers, measuring P2Y12 reaction units and gene polymorphisms is difficult in routine clinical practice. Given that the reduction in P2Y12 reaction unit by prasugrel is effective in preventing stroke recurrence, and that, unlike clopidogrel, it is effective in CYP2C19 poor metabolizers, prasugrel may be an option for patients with large-artery atherosclerosis or small-artery occlusion regardless of CYP2C19 status.

This study has several limitations. First, this was an integrated analysis and the conditions of each clinical trial were not identical. Second, our analysis was not able to provide acute event data to compare prasugrel with clopidogrel. Third, patient background characteristics (e.g., age and bodyweight) differed from those in the real-world clinical setting, which likely includes older patients and more pronounced risk differences, and therefore, potentially limiting the generalizability of these findings. Finally, our analysis only included Japanese patients; therefore, the study findings are not necessarily generalizable to non-Japanese populations.

This integrated analysis reinforced the findings of PRASTRO-III and showed that prasugrel is a promising medication for a numerical reduction in the composite incidence of ischemic stroke, MI, and death from other vascular causes in patients with ischemic stroke who are at a high risk of stroke recurrence. No major safety issues were reported for prasugrel.

The authors thank Sarah Bubeck, PhD, and Emily Woodhouse, PhD, of Edanz (www.edanz.com) for providing medical writing support, which was supported by Daiichi Sankyo Co., Ltd. (Tokyo, Japan), in accordance with Good Publication Practice (GPP3) guidelines (http://www.ismpp.org/gpp3).

The PRASTRO-I, PRASTRO-II, and PRASTRO-III studies each fully adhered to the ethical principles of the Declaration of Helsinki as well as Good Clinical Practice guidelines. Each of the study protocols was approved by the Ethics Committee of each participating institution. All study participants provided written informed consent.

T.Kitazono received payment from Daiichi Sankyo Co., Ltd. to support the present study (payment was made directly to T.Kitazono); grants or contracts from Ono Pharmaceutical Co., Ltd., MSD K.K., Kissei Pharmaceutical Co., Ltd., Takeda Pharmaceutical Co., Ltd., Chugai Pharmaceutical Co., Ltd., Astellas Pharma Inc., Nippon Boehringer Ingelheim Co., Ltd., Bristol-Myers Squibb K.K., EA Pharma Co., Ltd., Shionogi Inc., Mitsubishi Tanabe Pharma Corporation, Torii Pharmaceutical Co., Ltd., Otsuka Pharmaceutical Co., Ltd., and Asahi Kasei Medical Co., Ltd. (payments were made to the Department of Medicine and Clinical Science, Graduate School of Medical Sciences, Kyushu University); and honoraria from Daiichi Sankyo Co., Ltd., and Bayer Yakuhin, Ltd. (payments were made directly to T.Kitazono). M.K. received honoraria from Daiichi Sankyo Co., Ltd. and Bristol-Myers Squibb K.K. and payment for participation on a Data Safety Monitoring Board or Advisory Board (all payments were made directly to M.K.). Y.M. received consulting fees from E.P. Medical Inc. (payments were made directly to Y.M.); lecture fees from Medtronic Japan Co., Ltd., Stryker Japan K.K., Terumo Corporation, Johnson & Johnson K.K., Kaneka Corporation, JIMRO Co., Ltd., AstraZeneca K.K., Daiichi Sakyo Co., Ltd., Pfizer Japan Inc., Otsuka Pharmaceutical Co., Ltd., and Bristol-Myers Squibb K.K. (payments were made directly to Y.M.); and owns shares in E.P. Medical Inc. K.K. has received grants or contracts from Daiichi Sankyo Co., Ltd. (payments were made to the Department of Neurology Science, Graduate School of Medicine, Nippon Medical School); consulting fees from Daiichi Sankyo Co., Ltd. (payments were made directly to K.K.); honoraria from Daiichi Sankyo Co., Ltd. (payments were made directly to K.K.); and support for attending meetings/travel from Daiichi Sankyo Co., Ltd. (payments were made directly to K.K.). T.S., A.T., and T.Kuroda are employees of Daiichi Sankyo Co., Ltd.

The study was designed and supported by Daiichi Sankyo Co., Ltd. (Tokyo, Japan).

Takanari Kitazono, Masahiro Kamouchi, Yuji Matsumaru, Toshiaki Shirai, Atsushi Takita, Takeshi Kuroda, and Kazumi Kimura contributed to the conception and design of the study, interpretation of the data, and drafting of the manuscript, and critically revised it for intellectual content. All authors have read and approved the final version of the manuscript and agree to be held responsible for the integrity of the work.

Deidentified individual participant data and applicable supporting clinical study documents are available upon reasonable request at https://vivli.org/. In cases where clinical study data and supporting documents are provided pursuant to company policies and procedures, Daiichi Sankyo Co., Ltd. will continue to protect the privacy of clinical study participants. Details on data sharing criteria and the procedure for requesting access are available at https://vivli.org/ourmember/daiichi-sankyo/.