Study Applies Standardized Bleeding Definition to Disparate Antiplatelet Trials
Differences in major bleeding rates among trials comparing P2Y12 inhibitors in patients with acute coronary syndromes (ACS) are minimized when standardized bleeding definitions and event reporting replace individual trial practices. The finding, which appeared online May 30, 2011, ahead of print in the European Heart Journal, suggests that consistency in the reporting of major bleeding enables more meaningful comparison of trial data and accurate assessment of the safety of new antiplatelet agents or regimens.
Investigators led by John W. Eikelboom, MD, of McMaster University (Hamilton, Canada), searched databases from 2001 to 2010 for randomized phase 3 trials that compared antiplatelet regimens and reported major bleeding outcomes. Six studies involving 119,020 patients met the criteria. Three were placebo-controlled studies of standard clopidogrel:
- CURE (Clopidogrel in Unstable angina to prevent Recurrent ischemic Events)
- CLARITY-TIMI 28 (Clopidogrel as Adjunctive Reperfusion Therapy-Thrombolysis in Myocardial Infarction 28)
- COMMIT (ClOpidogrel Metoprolol in Myocardial Infarction Trial)
And 3 tested standard clopidogrel against prasugrel, ticagrelor, or high-dose clopidogrel:
- TRITON-TIMI 38 (TRial to Assess Improvement in Therapeutic Outcomes by Optimizing Platelet InhibitioN with Prasugrel-Thrombolysis In Myocardial Infarction 38)
- PLATO (PLATelet inhibition and patient Outcomes)
- CURRENT-OASIS 7 (Clopidogrel optimal loading dose Usage to Reduce Recurrent EveNTs)
Different Definitions, Different Results
Each of the 6 trials used a different definition for the primary major bleeding endpoint. CURE, COMMIT, TRITON-TIMI 38, and PLATO further characterized major bleeding as life-threatening or non-life-threatening. Five trials reported TIMI major bleeding, which served as the standard for comparison in this analysis. In CURE, CLARITY-TIMI 28, and TRITON-TIMI 38 the events were adjudicated, while in PLATO and CURRENT-OASIS 7 they were not. In addition, the timing of reporting of major bleeding ranged from 15 days in COMMIT to 15 months in TRITON-TIMI 38.
Using the primary trial definition, rates of major bleeding among patients receiving standard-dose clopidogrel ranged from 0.6% in COMMIT to 11.2% in PLATO. Estimated risk ratios did not vary among the 3 placebo-controlled trials (P for heterogeneity = 0.25). On the other hand, risk ratios in the 3 active-controlled trials showed significant heterogeneity (P = 0.05). In PLATO, ticagrelor did not increase major bleeding compared with clopidogrel (11.6% vs. 11.2%), whereas in TRITON-TIMI 38 prasugrel added risk (2.4% vs. 1.8%), as did double-dose clopidogrel in CURRENT-OASIS 7 (2.5% vs. 2.0%).
By contrast, major bleeding rates were substantially lower in CURE, PLATO, and CURRENT-OASIS 7 when TIMI criteria rather than the primary trial definitions were applied. PLATO was an outlier; its high TIMI major bleeding rate (7.7%) was attributed to a higher proportion of CABG-related bleeding events.
Risk ratios for TIMI major bleeding were similar in the 2 placebo-controlled trials that reported this endpoint (RR 0.94; 95% CI 0.68-1.30 for CURE and RR 1.09; 95% CI 0.67-1.78 for CLARITY-TIMI 28; P for heterogeneity = 0.62). However, trials with active comparators did show heterogeneity (P = 0.01). The same pattern of increased bleeding with prasugrel and double-dose clopidogrel but not ticagrelor was observed with the TIMI definition as with the trial definitions.
Factoring Out Procedural Bleeding
In the 4 trials that separately reported non-CABG TIMI major bleeding, these rates were substantially lower than those for overall TIMI major bleeding, ranging from 0.6% in CURRENT-OASIS 7 to 2.2% in PLATO for patients who received standard-dose clopidogrel. Non-CABG-related TIMI major bleeding was significantly increased by ticagrelor in PLATO and by double-dose clopidogrel in CURRENT-OASIS 7, with hazard ratio estimates ranging from 1.25 to 1.40.
Standardized Reporting Can Change Risk Assessments
The authors write, “Application of a standardized approach to the reporting of major bleeding by using the same definition of bleeding at a consistent time point virtually eliminated [the] differences [in reported rates of major bleeding with standard-dose dual antiplatelet therapy] and also altered the interpretation of the relative safety of new P2Y12 inhibitors compared with standard-dose clopidogrel.
“For example, the primary trial reports prasugrel, but not ticagrelor, increased major bleeding compared with standard-dose clopidogrel,” they continue, “whereas our re-assessment using a standardized bleeding definition and excluding procedure-related events indicates that ticagrelor, but not prasugrel, increases non-CABG TIMI major bleeding.”
Most of the variability in outcomes can be explained by 3 factors, the investigators say:
- The definition of major bleeding
- The timing of reporting of the primary bleeding outcome
- Rates of CABG
Other, smaller potential contributors include differences in patient characteristics (eg, age, sex, or renal function), clinical presentation (eg, NSTE ACS vs. STEMI), concomitant therapies (eg, IV glycoprotein IIb/IIIa inhibitors or thrombolytics), and bleeding treatments (eg, rates of red cell transfusion).
In the end, accurate assessment of the safety of alternative antiplatelet regimens, including new, more potent P2Y12 inhibitors, depends on the ability to analyze comparable trial data based on standardized reporting and bleeding definitions, the authors comment.
TIMI Major Too Restrictive?
“This kind of cross-trial analysis is very important because it shows what a difference [study-specific outcome] definitions can make,” said Sorin J. Brener, MD, of Weill Cornell Medical College (New York, NY).
Using a standardized TIMI major definition “evens out the playing field,” he told TCTMD in a telephone interview. But trialists should aim for a clinically relevant definition of bleeding, he added, one that is neither too broad nor too restrictive. “‘TIMI major’ is one of the most exclusive,” he noted. “When it is used as a standardized definition, bleeding rates tend to be more uniform and the differences between treatment arms much smaller.”
On the downside, this restrictive definition could lead to questioning the usefulness of a treatment whose only benefit is its impact on bleeding, he observed. “The trials of bivalirudin are a good example,” Dr. Brener said. “The reason they could show a reduction in major bleeding is because the events were inflated by a very inclusive definition. More inclusive definitions have a better ability to distill differences between treatment arms.”
Onward to BARC
On the other hand, too liberal a definition of major bleeding “might pick up every little oozing,” which would likewise fail the test of clinical relevance. “I think the BARC definitions are a very good compromise,” Dr. Brener said, referring to the recently published standardized bleeding definitions from the Bleeding Academic Research Consortium (Mehran R, et al. Circulation. 2011;123:2736-2747).
In fact, the timing of this study could not be better, commented Deepak L. Bhatt, MD, MPH, of Brigham and Women’s Hospital (Boston, MA), in a telephone interview, because its message—that “bleeding definitions are all over the place and make comparing results from one trial to another almost impossible”—serves as perfect lead-in to the BARC debut.
These consensus definitions, which cover 5 degrees of severity, provide both sensitivity and specificity and will enable a more refined estimate of differences in bleeding caused by different treatments, he told TCTMD. Although TIMI and other criteria may still be reported in trials that are already under way, BARC definitions will increasingly become the standard and will “facilitate across-trial and across-registry comparisons,” Dr. Bhatt asserted.
He downplayed the apparent reversal of fortune for prasugrel in terms of bleeding impact when the standardized TIMI major definition was applied, noting that both the TRITON-TIMI 38 results and biology support the notion that there is more bleeding with prasugrel than clopidogrel.
And ticagrelor, as a potent antiplatelet agent, would also be expected to produce a modest increase in bleeding, Dr. Bhatt said, adding, “What this study shows is that those differences are not enormous. In absolute terms, the risk of major bleeding is pretty low with either agent.”
For practitioners, “the real take-away message is that just looking at the last line of study abstracts without understanding [how results were derived] is not good enough—it’s an easy way to get burned,” Dr. Brener cautioned.
The mean age ranged from 57.5 to 64.2 years, and the majority of patients in each trial were male. CLARITY-TIMI 28 included only patients with STEMI, COMMIT enrolled patients with suspected STEMI, and CURE included only patients with NSTE ACS. In the remaining trials, about two-thirds of patients had NSTE ACS and one-third STEMI.
Quinlan DJ, Eikelboom JW, Goodman SG, et al. Implications of variability in definition and reporting of major bleeding in randomized trials of oral P2Y12 inhibitors for acute coronary syndromes. Eur Heart J. 2011;Epub ahead of print.
Study Applies Standardized Bleeding Definition to Disparate Antiplatelet Trials
- The study was supported by an unrestricted educational grant from AstraZeneca.
- Dr. Eikelboom reports serving on the advisory board or receiving honoraria from AstraZeneca, Bayer, Bristol-Myers Squibb, Daiichi Sankyo, Eisai, GlaxoSmithKline, Johnson and Johnson, Merck, Novartis, Portola, and Sanofi-Aventis.
- Dr. Brener reports serving as a consultant for and receiving speaker’s fees from Eli Lilly.
- Dr. Bhatt reports receiving research grants from AstraZeneca, Bristol-Myers Squibb, Eisai, Sanofi Aventis, and The Medicines Company.