Clopidogrel Two Doses Comparative 1-Year Assessment of Safety and Efficacy (COMPASS) Study in Japanese Patients with Ischemic StrokeUchiyama S.a · Tanahashi N.b · Minematsu K.c · on behalf of the COMPASS (SFY6913) Study Group
aDepartment of Neurology, Tokyo Women’s Medical University School of Medicine, Tokyo, bDepartment of Neurology, Saitama Medical University International Medical Center, Saitama, and cDepartment of Cerebrovascular Medicine, National Cerebral and Cardiovascular Center, Osaka, Japan Corresponding Author
Background: Clopidogrel 75 mg once daily is licensed in Japan for the prevention of recurrent ischemic cerebrovascular events in adults as the usual dosage. However, a lower dose (50 mg) is an option in patients at an increased risk of bleeding depending on age, body weight and symptoms. This study compared the safety of both 75- and 50-mg doses of clopidogrel in patients with noncardioembolic ischemic stroke. Methods: This was a double-blind, double-dummy postmarketing clinical trial carried out across 118 Japanese institutions. Patients with an episode of noncardioembolic ischemic stroke at least 8 days prior to randomization, who were aged <75 years or had a body weight >50 kg were randomized to 50 or 75 mg clopidogrel once daily for 52 weeks. The primary endpoint was the incidence of bleeding adverse events. The secondary safety endpoints included the incidence of serious adverse events, serious bleeding adverse events and other prespecified adverse events. The secondary efficacy endpoint was the incidence of vascular events, including ischemic stroke, myocardial infarction, and peripheral artery disease. Results: A total of 1,110 patients were randomized to clopidogrel 50 mg (n = 558) or 75 mg (n = 552). No significant difference between the groups was detected in the incidence of bleeding adverse events, which was 14.0 and 16.5% in the clopidogrel 50- and 75-mg groups, respectively (hazard ratio = 0.831, 95% CI = 0.615–1.124, p = 0.2274). Additionally, there was no statistical difference with respect to any of the secondary safety endpoints. No significant difference between the groups was detected in the incidence of serious adverse events, which was 8.6 and 9.5% in the clopidogrel 50- and 75-mg groups, respectively (hazard ratio = 0.877, 95% CI = 0.597–1.289, p = 0.5035), and there was no significant difference between the groups in the incidence of serious bleeding events, which was 1.7 and 1.5% in the clopidogrel 50- and 75-mg groups, respectively (hazard ratio = 1.240, 95% CI = 0.489–3.142, p = 0.6496). The percentages of intracranial hemorrhage in the 50- and 75-mg groups were 0.18% (1/558) and 0.18% (1/552), respectively. The cumulative incidence of vascular events was somewhat lower in the 75-mg group, but was not statistically different (2.6 vs. 3.8%; p = 0.4118). Conclusions: Clopidogrel 75 mg provides a clinically acceptable safety profile and suggests better clinical benefit as compared to clopidogrel 50 mg for the secondary prevention of ischemic stroke in Japanese patients who are <75 years old with a body weight >50 kg, considering the balance of safety and efficacy on this trial.
Copyright © 2012 S. Karger AG, Basel
Ischemic stroke is well recognized as a leading cause of death and disability worldwide. In Japan, the incidence of ischemic stroke is 260–357 per 100,000 person-years, corresponding to a mortality rate of 45–68 per 100,000 person-years . Moreover, in patients who survive a first ischemic stroke, the risk of recurrence is high; 4.83% for Japanese patients enrolled in the REACH registry  and 24.5% in Japanese patients with mild strokes  within 2 years. Therefore, administration of antiplatelet agents is essential for the secondary prevention of ischemic stroke .
The antiplatelet agent clopidogrel is a thienopyridine agent that inhibits platelet aggregation by selectively and irreversibly inhibiting the binding of adenosine diphosphate to its platelet receptor P2Y12 . In a large, predominantly Caucasian population at risk of ischemic events due to atherosclerotic vascular disease (manifested as ischemic stroke, myocardial infarction, or symptomatic peripheral arterial disease), clopidogrel 75 mg once daily was shown to be more effective than aspirin 325 mg once daily for the prevention of ischemic events . On the basis of these and other results, clopidogrel 75 mg once daily has been licensed for more than 10 years in Western countries for the prevention of atherothrombotic events in patients with ischemic cerebrovascular, coronary, or peripheral arterial disease [7,8].
Studies conducted in Japan among patients with a history of ischemic stroke have demonstrated that clopidogrel 75 mg once daily is better tolerated than ticlopidine 200 mg once daily, and shows noninferior efficacy with regard to the secondary prevention of vascular events [9,10]. However, any bleeding adverse events requiring attention were observed with a higher incidence among patients receiving clopidogrel who were aged ≥75 years (vs. <75 years) or weighed <50 kg (vs. ≥50 kg) . Consequently, in Japan, clopidogrel is licensed for oral use at a dose of 75 mg once daily, but, depending on the patient’s age and body weight, a dose of 50 mg once daily is recommended.
Thus, the present postmarketing study compared the safety and efficacy of clopidogrel 50 and 75 mg for the secondary stroke prevention in Japanese patients, who were aged <75 years and weighed >50 kg.
Materials and Methods
This study was a multicenter, single-country, randomized, double-blind, double-dummy phase IV trial (NCT00386191) designed to compare the safety of 50 and 75 mg of clopidogrel for the secondary stroke prevention in Japanese patients. The primary safety endpoint was the incidence of bleeding adverse events. The secondary safety endpoints included the incidence of serious adverse events (SAEs), serious bleeding adverse events, and prespecified adverse events of interest comprised of leukopenia, neutropenia, thrombocytopenia, and hepatic dysfunction. The incidence of vascular events was used to compare the efficacy of the two clopidogrel doses.
Patients were recruited to the study from 118 Japanese medical institutions. Patients aged 20–74 years with a body weight >50 kg were eligible for the study if they had experienced an episode of noncardioembolic ischemic stroke at least 8 days prior to randomization and their clinical course preceding randomization was well documented, or if they had experienced an ischemic stroke confirmed at the study institution using diagnostic brain imaging (assessed by CT or MRI) performed <12 weeks before treatment initiation.
The main exclusion criteria for the study were: patients with cardiogenic stroke or disease that could precipitate cardiogenic cerebral thromboembolism, such as atrial fibrillation or valvular heart disease (including valve replacement); patients with a transient ischemic attack occurring after the last episode of ischemic stroke; patients with a serious impairment that would hinder the detection of a new ischemic event (e.g. bedridden, needing total assistance, or dementia); patients with bleeding diathesis, coagulopathy, or hemorrhagic disease; patients with a history of intracranial hemorrhage; patients with diabetic retinopathy; patients with hypertension and insufficient control of blood pressure despite antihypertensive therapy, and patients with malignancy (except complete remission from malignancy <5 years prior to randomization), cardiac disease, serious renal disease or abnormal creatinine levels, leukopenia, neutropenia, thrombocytopenia, severe hepatic dysfunction, or a history of drug allergy.
Within 2 weeks of obtaining informed consent, patients were randomly allocated to receive either clopidogrel 50 or 75 mg. Based on an application from the investigator, a centralized patient registration center confirmed the patient eligibility, and notified his/her registration and an allocation number to identify the study drug among the distributed drugs to the investigator and the sponsor by telephone or facsimile. If a patient was ineligible due to noncompliance with the inclusion and exclusion criteria, the centralized patient registration center notified the fact to the investigator.
Double-blinding was maintained in both treatment groups throughout the course of the study as follows: a controller confirmed the indistinguishability of the study drugs and of their package appearances at the time of randomized allocation and between finalization of all case report forms and the key code breaking. The controller established the key codes and emergency key codes to cope with emergencies. The controller also retained the key codes and emergency key codes in sealed and signed envelopes until the time of key code breaking.
Patients assigned to the 50-mg group received two 25-mg clopidogrel tablets and one 75-mg placebo tablet that was indistinguishable from the 75-mg clopidogrel tablet (3 tablets in total). Those patients assigned to the 75-mg group received two 25-mg placebo tablets that were indistinguishable from the 25-mg clopidogrel tablets and one 75-mg clopidogrel tablet (3 tablets in total). The doses for the 50- and 75-mg groups were administered orally once daily after a meal.
Treatment was continued for 52 weeks and assessments were performed at baseline (day 1) and at weeks 4, 8, 12, 20, 28, 36, 44, and 52. These included assessments of compliance, blood pressure, pulse rate, concomitant drug use, adverse events developed after the last observation, and blood sampling for laboratory tests (hematological and biochemical tests). Additionally, diagnostic brain imaging (CT or MRI) and electrocardiography were performed at baseline and week 52 (or at discontinuation of treatment). Patients were instructed to report any safety-related event occurring within 2 weeks after final study drug intake. Follow-up was performed after 52 weeks until discontinued by the investigator based on an appropriate medical assessment.
Bleeding adverse events were defined as all adverse events with any bleeding, including intracranial hemorrhage, retinal bleeding of the ocular fundus, subcutaneous bleeding, nasal bleeding, intraoral bleeding, hypermenorrhea, and gastrointestinal bleeding. SAEs were defined as any adverse events that met the following criteria: all-cause death, any life-threatening events, hospitalization or prolongation of existing hospitalization, clinically persistent or significant disability/incapacity, or medically important events. Leukopenia was defined as a white blood cell count <3,000/mm3, neutropenia as a neutrophil count <1,500/mm3, and thrombocytopenia as a platelet count <100,000/mm3. Hepatic dysfunction was defined as a change in levels of l-aspartate aminotransferase, alanine aminotransferase, γ-glutamyl transpeptidase, alkaline phosphatase, or total bilirubin ≥120% of the upper limit of normal (if levels were normal at baseline) or ≥200% of the baseline value (if levels were abnormal at baseline), or when jaundice was detected. The following frequency distributions of treat-emergent adverse events (TEAEs) were provided by treatment: overview of all TEAEs, serious TEAEs, TEAEs leading to death and TEAEs leading to treatment discontinuation. Vascular events included ischemic stroke, myocardial infarction and other vascular events in heart, extremities, or other organs that did not fit the previous criteria, e.g. transient ischemic attack or angina.
The study was monitored by the Safety/Efficacy Evaluation Committee, whose role was to evaluate safety variables, especially bleeding adverse events, SAEs, and adverse events of interest and efficacy variables (i.e. vascular events).
The study protocol was approved by the Institutional Review Board/Ethics Committee and all patients gave informed consent. The study was conducted in accordance with ethical principles such as the Declaration of Helsinki, together with good clinical practice guidelines, good postmarketing study practice, good vigilance practice, and domestic pharmaceutical laws and regulations.
The target population of this study excluded elderly and low-weight patients with a possible high bleeding risk. From an unpublished study in this population, the incidence of bleeding adverse events for clopidogrel 75 mg over 52 weeks excepting patients aged ≥75 years or with a weight ≤50 kg was calculated as 26.8%. For clopidogrel 50 mg, the incidence of bleeding adverse events was assumed to be 30% lower than with clopidogrel 75 mg, based on the 30% decrease in platelet aggregation inhibitory effect and 30% decrease in incidence of ≥2-fold prolongation of bleeding time in patients receiving the lower dose. Therefore, 541 patients per group (approximately 1,100 in total) were required in order to show a significant difference in safety between clopidogrel 50 and 75 mg, using a log rank test with a 5% two-sided significance level and 80% statistical power. The study was powered for a definitive evaluation of the primary endpoint. The main safety and efficacy analyses were based on the intention-to-treat (ITT) population (all randomized patients) and other safety analyses were based on the all-treated population (all patients who received ≥1 dose of study drug). Incidence curves and corresponding 95% confidence intervals (CIs) were estimated using the Kaplan-Meier method. Finally, for all parameters measured, no imputation of missing data was performed.
The study was conducted between September 4, 2006 and December 12, 2008. A total of 1,110 patients were randomized to either clopidogrel 50 mg (n = 558) or clopidogrel 75 mg (n = 552) and were, therefore, included in the ITT population (fig. 1). Only 2 patients (both randomized to clopidogrel 50 mg) were not exposed to the study drug, yielding an all-treated population of 1,108 patients.
|Fig. 1. Patient flowchart of the study.|
Similar proportions of patients in the clopidogrel 50-mg group (n = 469, 84.1%) and the 75-mg group (n = 457, 82.8%) completed the 52-week treatment course. The most common reasons for treatment discontinuation were adverse events [n = 53 (9.5%) and n = 67 (12.1%) in the clopidogrel 50- and 75-mg groups, respectively], patient’s request [n = 12 (2.2%) and n = 11 (2.0%)], and poor compliance [n = 5 (0.9%) and n = 5 (0.9%)]. Regardless of treatment discontinuation, a high proportion of patients (n = 1,039, 93.6%) were followed up at week 52 [n = 522 (93.5%) of the 50-mg group and n = 517 (93.7%) of the 75-mg group].
The demographics and clinical characteristics of the study participants were well balanced across the two treatment arms (table 1). Most patients were male (80.0%) and the mean age was 62.1 years, with more than a half of patients (55.7%) younger than 65 years of age. In the majority of patients (85.3%), the time from the most recent stroke was later than 30 days from study enrollment, and lacunar stroke (61.9%) was the most common type of recent stroke.
|Table 1. Patient baseline and disease characteristics|
Previous use of antiplatelet drugs was prevalent among the study cohort, with most patients from both treatment groups having previously received aspirin.
Compliance to treatment throughout the study period was high (99% in both groups). All patients exposed to the study drug were therefore compliant, defined as administration of at least 75% of the prescribed dosage regimen.
Cumulative exposure to clopidogrel 50 and 75 mg was 512.3 and 498.9 person-years, respectively, with a mean duration of exposure of 336.5 and 330.1 days. Indeed, most patients in the clopidogrel 50-mg (84.4%) and 75-mg groups (82.6%) received the study drug for ≥1 year.
No significant difference between the treatment groups was detected with respect to the incidence of bleeding adverse events. Analysis of the time to first bleeding adverse event in the ITT population indicated that bleeding events occurred in 78 of 558 patients and 92 of 552 patients in the clopidogrel 50- and 75-mg groups, respectively (table 2). The cumulative incidences (95% CI) of bleeding at week 52 in the clopidogrel 50- and 75-mg groups were 14.0% (11.1–16.9%) and 16.5% (13.4–19.6%), respectively (fig. 2a; hazard ratio = 0.831, 95% CI = 0.615–1.124, p = 0.2274). Although 2 patients in the 50-mg group did not take the study drug, the time to the first bleeding adverse event in the all-treated population (on-treatment analysis) was consistent with that in the ITT population (50 mg: 13.7%, 95% CI = 10.7–16.6%; 75 mg: 16.0%, 95% CI = 12.9–19.2%; hazard ratio = 0.836, 95% CI = 0.612–1.140, p = 0.2558).
|Table 2. Time to first bleeding adverse event in the ITT population|
|Fig. 2. Kaplan-Meier cumulative incidence curves for (a) time to first bleeding adverse event among the ITT population and (b) time to vascular event among the ITT population.|
Looking at details of bleeding adverse events, severe intensity in the 50- and 75-mg group was 0.4% (2/558) and 0.4% (2/552), moderate intensity was 2.0% (11/558) and 1.8% (10/552), and mild intensity was 11.6% (65/558) and 14.5% (80/552), respectively (for online suppl. table 1, see www.karger.com/doi/10.1159/000342655). Furthermore, the percentage of gastrointestinal bleeding in the ITT population was low in both groups (50 mg: 2.2%, 12/558; 75 mg: 2.5%, 14/552) (table 3), and bleeding requiring transfusion had not been observed in this study. Importantly, the percentages of intracranial hemorrhage in the 50- and 75-mg groups were 0.18% (1/558) and 0.18% (1/552), respectively. In the all-treated population, a significant drop in hematocrit (more than 15% decrease) was observed in 3 patients of the clopidogrel 50-mg group and in 1 patient of the clopidogrel 75-mg group.
|Table 3. Number (%) of patients experiencing gastrointestinal bleeding adverse events presented by preferred term in the ITT population|
No statistically significant differences between the treatment groups were detected with respect to any of the secondary safety endpoints (table 4). Hence, patients receiving clopidogrel 75 mg were not at greater risk than those receiving clopidogrel 50 mg of experiencing an SAE, a serious bleeding adverse event, or the adverse event of interest. The cumulative incidences of SAEs at week 52 in the clopidogrel 50- and 75-mg groups were comparable (8.6%, 95% CI = 6.3–11.0% vs. 9.5%, 95% CI = 7.0–11.9%; hazard ratio = 0.877, 95% CI = 0.597–1.289; p = 0.5035), as were the cumulative incidences of serious bleeding events (1.7%, 95% CI = 0.6–2.8% vs. 1.5%, 95% CI = 0.5–2.5%; hazard ratio = 1.240, 95% CI = 0.489–3.142; p = 0.6496). The cumulative incidences (95% CI) of adverse events of interest at week 52 were 22.4% (18.9–25.9%) in the clopidogrel 50 mg and 23.8% (20.2–27.4%) in the clopidogrel 75-mg group (hazard ratio = 0.935, 95% CI = 0.735–1.190; p = 0.5834) and there were no intergroup differences in the individual incidences of these 4 safety variables.
|Table 4. Time to first SAE, serious bleeding event, and other predefined adverse events in the ITT population|
In the all-treated population, most patients reported a TEAE but there were no clinically meaningful differences between the groups regarding the percentages, nature, or severity of events (table 5). Overall, 83.3% (463/556) of patients in the 50-mg group and 86.4% (477/552) of patients in the 75-mg group reported at least 1 TEAE during the course of the study. The most frequently observed TEAEs were increased eosinophil count [clopidogrel 50 mg: 8.5% (47/556); clopidogrel 75 mg: 9.2% (51/552)]; increased γ-glutamyl transpeptidase [clopidogrel 50 mg: 8.6% (48/556); clopidogrel 75 mg: 7.8% (43/552)]; increased alanine aminotransferase [clopidogrel 50 mg: 5.6% (31/556); clopidogrel 75 mg: 6.2% (34/552)], and nasopharyngitis [clopidogrel 50 mg: 20.9% (116/556); clopidogrel 75 mg: 22.3% (123/552)]. Several TEAEs were more frequent in the 75-mg group, including abnormal liver function test [clopidogrel 50 mg: 2.0% (11/556); clopidogrel 75 mg: 4.2% (23/552)]; bronchitis [clopidogrel 50 mg: 1.4% (8/556); clopidogrel 75 mg: 2.9% (16/552)], and pruritus [clopidogrel 50 mg: 1.4% (8/556); clopidogrel 75 mg: 2.9% (16/552)].
|Table 5. Number (%) of TEAEs among the all-treated population|
The number of patients with serious TEAEs was also similar in both groups [clopidogrel 50 mg: 9.9% (55/556); clopidogrel 75 mg: 9.8% (54/552)]. There was no significant difference in the types of serious TEAEs between the two treatment groups; however, two classes of serious TEAEs [unspecified, benign and malignant neoplasms (including cysts and polyps) and cardiac disorders] were observed more frequently (≥2-fold) in the 50-mg group than in the 75-mg group.
Two patients experienced a TEAE which led to death in the 50-mg group, but none were reported for the 75-mg group. However, for both these patients, there was no causal relationship between the events leading to death and the study drug.
Finally, fewer patients discontinued treatment due to TEAEs in the 50-mg group than in the 75-mg group [9.4% (52/556) and 12.0% (66/552), respectively].
Vascular events occurred in 21 of 558 patients and 16 of 552 patients in the clopidogrel 50- and 75-mg groups, respectively (table 6). The cumulative incidence of vascular events at week 52 was slightly lower in the clopidogrel 75-mg group, although no statistically significant difference was detected (clopidogrel 50 mg: 3.8%, 95% CI = 2.2–5.5%; clopidogrel 75 mg: 2.6%, 95% CI = 1.3–4.0%). Furthermore, the log rank test revealed that patients receiving clopidogrel 50 mg were no more likely to have a vascular event than counterparts receiving clopidogrel 75 mg (fig. 2b; hazard ratio = 1.312, 95% CI = 0.685–2.514, p = 0.4118). Ischemic stroke was observed in 10 patients (1.8%) in the clopidogrel 50-mg group and 12 patients (2.2%) in the clopidogrel 75-mg group. Furthermore, 2 patients from the 75-mg group had cardiogenic cerebral thromboembolism due to arterial fibrillation and patent foramen ovale, despite patients with disease that could precipitate cardiogenic cerebral thromboembolism being excluded from the study. Myocardial infarction was observed in 2 patients (0.4%) in the clopidogrel 50-mg group and 1 patient (0.2%) in the clopidogrel 75-mg group. Other vascular events (e.g. peripheral arterial diseases, transient ischemic attack) were observed in 9 patients (1.6%) in the clopidogrel 50-mg group and 5 patients (0.9%) in the clopidogrel 75-mg group. Finally, there was 1 death due to a vascular event (peripheral arterial disease) in the clopidogrel 50-mg group (table 7).
|Table 6. Time to first vascular event|
|Table 7. Number (%) of vascular events|
Clopidogrel 75 mg has been approved since 2006 in Japan for ‘the reduction of recurrence after ischemic cerebrovascular disorder (excluding cardiogenic brain embolism)’. In addition to the recommended dose of 75 mg, a lower dose (50 mg) is also available for use in Japanese patients who are aged ≥75 years or weigh <50 kg and may have an increased bleeding risk at the standard dose. This is based on data that demonstrate a reduction in the incidence of bleeding events with clopidogrel 50 mg (vs. clopidogrel 75 mg), while a significant level of platelet aggregation inhibition is maintained. The aim of this phase IV randomized controlled study was to satisfy the risk-to-benefit ratio of clopidogrel 75 mg versus clopidogrel 50 mg in Japanese patients with noncardioembolic ischemic stroke who were aged <75 years and weighed >50 kg.
The COMPASS study indicates that treatment with clopidogrel 75 mg is not associated with a significant increase in the risk of bleeding adverse events in the ITT population compared with clopidogrel 50 mg. The trend for cumulative incidence of bleeding adverse events in the all-treated population was consistent with that in the ITT population, a fact which suggests robustness of the study results and denies our hypothesis that clopidogrel 75 mg increases the risk of bleeding adverse event by approximately 30%, as compared to 50 mg. Moreover, the rate of gastrointestinal bleeding in the clopidogrel 75-mg group was similar to that in the clopidogrel 50-mg group and no bleeding events requiring transfusion were identified. During the on-treatment period, 3 patients in the clopidogrel 50-mg group and 1 patient in the clopidogrel 75-mg group experienced a significant drop in hematocrit. The drops in the clopidogrel 50-mg group were due to surgery for arteriosclerosis obliterans, ascending colon cancer and gastric ulcer hemorrhage; however, in the clopidogrel 75-mg group, a cause of the drop could not be specified since it recovered to normal without any treatment or study drug discontinuation. Furthermore, patients treated with clopidogrel 75 mg are not at increased risk of experiencing an SAE, a serious bleeding adverse event or adverse events of interest. Taking into consideration the above facts, the safety profile of clopidogrel 75 mg is thought to be comparable to that of clopidogrel 50 mg.
As regards efficacy, the COMPASS study showed a numerically lower cumulative incidence of vascular events in the group treated with clopidogrel 75 mg (2.6%, 95% CI = 1.3–4.0% vs. 3.8%, 95% CI = 2.2–5.5%; p = 0.4118). Although the standard dose (75 mg) is not associated with a significant reduction in the incidence of vascular events in this clinical trial setting, vascular events defined in this study were considered as life-threatening or serious events. Taking into consideration the facts that 2 patients suffering vascular events in the clopidogrel 75-mg group would have primarily been excluded from the study and standard dose has a comparable safety profile to lower dose of clopidogrel, this numerical difference might mean potential benefit of the standard dose of clopidogrel. In other words, the study may present a medically important risk of the low-dose clopidogrel for patients without bleeding risk, even in the absence of a statistically significant difference. It should also be noted that this study was not powered to detect a difference in efficacy.
The incidences of bleeding adverse events and other adverse events in the present study are comparable to previous clinical trials with clopidogrel. In the pivotal phase III studies in Japan, major hemorrhagic adverse drug reactions were rare following treatment with clopidogrel 75 mg for 52 weeks, occurring in <2% of Japanese stroke patients . The same study showed an incidence of hepatic dysfunction of approximately 13.4%, with the incidences of leukopenia, neutropenia and thrombocytopenia <2.5%. The PRoFESS study investigated clopidogrel 75 mg versus aspirin and dipyridamole in stroke patients [7,11]. Here, the incidence of major hemorrhagic events was 3.6% in the clopidogrel arm, while any hemorrhagic events occurred in 4.9% of patients. Adverse events leading to treatment discontinuation occurred at a similar frequency in the PRoFESS study (10.6%) to the present study (12.0%).
The incidence of vascular events in the present study (2.6%) was also similar to that found previously in two separate studies with clopidogrel 75 mg in Japanese patients (3.6 and 4.4%, respectively) [9,10]. Vascular events in both studies were infrequent compared with those reported in non-Japanese populations. For example, the incidence of ischemic stroke, myocardial infarction or vascular death was 5.32% in the clopidogrel 75 mg arm of the CAPRIE study and the incidence of stroke, myocardial infarction or vascular death was 13.1% in the PRoFESS study [6,11]. However, cross-study comparisons are often unreliable due to subtle differences in study protocols and distinct patient populations. Indeed, the recurrence of myocardial infarction has been shown to be lower in patients from Japan compared with the rest of the world, which could have had an impact on these results .
The main strengths of this study lie in its design as a randomized, double-blind, double-dummy trial and the fact that following randomization, the patient characteristics in the two treatment arms were very similar. The study also had a high completion rate, with similar proportions of patients from both groups completing the 52-week course, in addition to a high follow-up rate, with 93.6% of patients completing the study undergoing follow-up assessments.
However, there are also several potential limitations to the current study. First, the incidence of bleeding adverse events in the 75-mg group (16.5%) was lower than that estimated during the sample size collection (26.8%), thereby reducing the power of the study to determine a statistically significant difference between the two treatment groups. Second, the trial had many exclusion criteria because the COMPASS study was a postmarketing clinical trial in which patients with off-label use and/or with impairments that may hinder detection of efficacy and safety variables cannot be enrolled. Actually, in Japan, clopidogrel 50 mg has been recommended for patients with higher bleeding risk such as patients aged ≥75 years and/or with a body weight ≤50 kg. That would have resulted in a study population with a small rate of female gender and atherothrombic stroke. Taking into account that mean body weight (SD) in Japanese people in their 60s was 64.2 (9.4) kg in men and 53.3 (8.4) kg in women when the study was conducted, many of the female patients would have been likely to be excluded from the study . Furthermore, it can be speculated that some of the patients with atherothrombic stroke had been excluded because of being bedridden, needing total assistance, or having dementia, which may impede efficacy and safety evaluation considering its larger ischemic area as compared to lacunar stroke. Third, efficacy had been evaluated during 52 weeks of treatment. Nevertheless, the primary objective of the COMPASS study was safety evaluation in terms of bleeding adverse events. In addition, neither new findings nor increasing of the concerned bleeding frequency beyond week 52 had been predicted.
Furthermore, as described in the section of statistical analyses, 541 patients per group (approximately 1,100 in total) were thought to be adequate to show a significant difference in bleeding adverse events during 52 weeks between the 50- and 75-mg group. Therefore, 52 weeks was thought to be an appropriate duration for the postmarketing commitment study for safety evaluation. Lastly, our disclosed relationship with the sponsor would be a potential risk for the interpretation of the study results. However, major endpoints of the study were adjudicated by a third party not related to both authors and sponsor under blinded condition. Thorough evaluation of clopidogrel 50 mg versus 75 mg over 52 weeks for the management of ischemic stroke in Japanese patients aged <75 years and weighing >50 kg demonstrated that there were no statistically significant differences between the two dosages in any safety analyses, and no clear difference in the incidence of vascular events, although this was numerically higher in the low-dose group. Considering the balance between the seriousness of clinical outcomes associated with vascular events and bleeding severity, the benefit of clopidogrel 75 mg in reducing vascular risk could potentially outweigh the risk engendered by a bleeding adverse event. Namely, clopidogrel 75 mg provides a clinically acceptable safety profile and is expected to be of better clinical benefit as compared to clopidogrel 50 mg for the secondary prevention of ischemic stroke in Japanese patients who are <75 years old with a body weight >50 kg.
The authors wish to thank all the patients and investigators involved in the study. All authors contributed to the design, writing and critical appraisal of the manuscript. This study was sponsored and financed by Sanofi-Aventis K.K., Tokyo, Japan.
S. Uchiyama is an advisory board member of Sanofi-Aventis K.K., Nippon Boehringer Ingelheim Co. Ltd. and Bayer Yakuhin Ltd., and has received honoraria as a speaker or moderator and research funds from Sanofi-Aventis K.K., Nippon Boehringer Ingelheim Co. Ltd., Daiichi Sankyo Co. Ltd., Otsuka Pharmaceutical CO., Ltd., and Bayer Yakuhin Ltd. N. Tanahashi has received honoraria as a speaker or moderator and research funds from Sanofi-Aventis K.K., Mitsubishi Tanabe Pharma and Otsuka Pharmaceutical Co., Ltd. K. Minematsu has received research funding from Sanofi-Aventis K.K., and received lecture fees from Sanofi-Aventis K.K., Otsuka Pharmaceutical Co., Ltd., Bayer Yakuhin, Ltd. and AstraZeneca during the past 3 years.
Shinichiro Uchiyama, MD, PhD, FAHA
Department of Neurology
Tokyo Women’s Medical University School of Medicine
8-1 Kawada-cho, Shinjuku-ku, Tokyo 162-8666 (Japan)
Received: May 9, 2012
Accepted: August 13, 2012
Published online: September 25, 2012
Number of Print Pages : 11
Number of Figures : 2, Number of Tables : 7, Number of References : 13
Additional supplementary material is available online - Number of Parts : 1
Vol. 34, No. 3, Year 2012 (Cover Date: October 2012)
Journal Editor: Hennerici M.G. (Mannheim)
ISSN: 1015-9770 (Print), eISSN: 1421-9786 (Online)
For additional information: http://www.karger.com/CED