Implications of different criteria for percutaneous coronary intervention-related myocardial infarction on study results of three large phase III clinical trials: The CHAMPION experience

Abstract

Aims:

The purpose of this study was to test whether different results between Cangrelor versus standard therapy to acHieve optimal Management of Platelet InhibitiON (CHAMPION) PCI/PLATFORM and PHOENIX trials are due in part to different definitions of percutaneous coronary intervention (PCI)-related myocardial infarction (MI).

Methods and results:

In patients with acute coronary syndrome (ACS), the definition of MI was identical in CHAMPION PCI and PLATFORM and did not require an assessment of baseline cardiac biomarker status, while in PHOENIX specific MI criteria were associated with different patient presentations. The same MI criteria were used in PCI, PLATFORM, and PHOENIX for patients with stable angina. Logistic regression assessed the effect of cangrelor on MI (PCI- and non-PCI related) in the combined PCI/PLATFORM population and in PHOENIX. Consistency of cangrelor’s effect in PCI/PLATFORM and in PHOENIX in patients with stable angina and in those with an ACS (with or without ST elevation) was evaluated. Overall, the incidence of PCI-related MI at 48 h was 6.3% in PCI/PLATFORM and 4.0% in PHOENIX. In patients with ACS, MI incidence was 6.4% in PCI/PLATFORM and 1.7% in PHOENIX, and 6.3% and 5.6%, respectively in stable angina patients. Cangrelor’s effect on PCI-related MI differed between PCI/PLATFORM (odds ratio (OR) 1.03, 95% confidence interval (CI) 0.90–1.17) and PHOENIX (OR 0.80, 95% CI 0.66–0.98) with p_INT=0.04. This difference was mostly evident in patients with ACS (p_INT= 0.06) while the effect was consistent in patients with stable angina (p_INT=0.81). Results were similar when all MIs were analyzed.

Conclusions:

The definition of PCI-related MI has important implications for event rates, treatment effect, and study results. This illustrates the importance of a rigorous assessment of PCI-related MI in clinical trials of patients with an ACS.

Keywords

Percutaneous coronary intervention myocardial infarction acute coronary syndrome randomized controlled trial

Myocardial infarction (MI), the most common endpoint in cardiovascular clinical trials,¹ has been consistently associated with increased mortality in patients with acute coronary syndrome (ACS) undergoing coronary revascularization.^2–4 However, with decreasing time to catheterization in ACS patients undergoing an early invasive strategy,⁵ the ascertainment of MI associated with percutaneous coronary interventions (PCIs) is increasingly challenging.⁶ When detection of early treatment effects is necessary, appropriate criteria to assess PCI-related MI – particularly the accurate identification of early recurrences in the setting of an evolving MI – are thus both difficult and important. To address this need, the Universal Definition of Myocardial Infarction Task Force, both in 2007 and in 2012 emphasized the requirement of cardiac biomarkers (preferably troponin) to be stable or falling before PCI in order to assess PCI-related MI.^7,8 Similarly, in 2013 a consensus document of the Society for Cardiovascular Angiography and Interventions (SCAI), which recommended creatine kinase (CK)-MB instead of troponin for the diagnosis of PCI-related MI, also differentiated PCI-related MI criteria according to baseline status before PCI.⁹

The efficacy of cangrelor, a short-acting intravenous P2Y₁₂ inhibitor, has been assessed in a large phase III program named Cangrelor versus standard therapy to acHieve optimal Management of Platelet InhibitiON (CHAMPION) that included three trials: PCI, PLATFORM, and PHOENIX.^10–12 In CHAMPION PCI and CHAMPION PLATFORM, which started together in 2006, identical criteria were used to adjudicate MI. In patients enrolled with an ACS, the definition of PCI-related MI did not require an explicit assessment of baseline status of cardiac markers before the procedure, including a formal assessment of trend and timing of cardiac markers in relationship to the procedure. Both trials were prematurely stopped for futility. However, in these trials a reduction was observed in cangrelor-treated patients of endpoints not dependent upon biomarkers, including stent thrombosis and Q wave MI.^6,13 These findings generated the hypothesis that cangrelor’s effect might have been confounded by a non-discriminating ascertainment of PCI-related MI, and laid the foundation for the design of CHAMPION PHOENIX.¹⁴

In PHOENIX, an explicit assessment of baseline status for the adjudication of PCI-related MI was implemented based on a combination of serial biomarker samples as well as ischemic symptoms and electrocardiographic (ECG) changes in patients enrolled with a suspected ACS.¹⁴ In this study cangrelor significantly reduced the primary composite endpoint of death, MI, ischemia-driven revascularization, or stent thrombosis at 48 h, with an effect mostly evident on the MI component.

We hypothesized that different results between PCI/PLATFORM and PHOENIX are associated with different criteria for PCI-related MI in patients with ACS. Specifically we hypothesized that there would be a heterogeneous cangrelor effect on MI between PCI/PLATFORM and PHOENIX in patients enrolled with an ACS but a similar cangrelor effect (i.e. no heterogeneity) in patients with stable angina, where the same criteria of PCI-related MI were used.

Methods

Study design and patients

To test our hypothesis, we included patient-level data from three double-blind, double dummy, randomized controlled trials (RCTs) that compared cangrelor with either clopidogrel or placebo in patients undergoing PCI: CHAMPION PCI, CHAMPION PLATFORM, and CHAMPION PHOENIX. In all CHAMPION trials, cangrelor (or matching placebo) was administered with an identical regimen consisting of a bolus (30 μg/kg) followed by an infusion (4 μg/kg per min) to be continued for at least 2 h or until the conclusion of the index PCI, whichever was longer. At the end of the infusion, all patients randomized to cangrelor received 600 mg of clopidogrel. The comparator group differed between the three RCTs. In CHAMPION PCI, the comparator group was 600 mg clopidogrel at the start of PCI; in CHAMPION PLATFORM, clopidogrel 600 mg at the end of PCI; and in CHAMPION PHOENIX clopidogrel 300 mg or 600 mg, as by site standard of care, could be administered either at the start or at the end of PCI. All patients provided written informed consent. In CHAMPION PCI and CHAMPION PHOENIX, patients with stable angina, non-ST-segment elevation ACS (NSTEACS), or ST elevation myocardial infarction (STEMI) could be enrolled, whereas CHAMPION PLATFORM did not include patients with STEMI. Key exclusion criteria were administration of a P2Y₁₂ inhibitor or abciximab at any time in the seven days before randomization (except for CHAMPION PCI, in which patients could be taking clopidogrel before randomization), and receipt of eptifibatide, tirofiban, or fibrinolytic therapy in the 12 h before randomization.

Criteria for MI

All MIs were adjudicated by a clinical event committee (CEC) blinded to randomized treatment allocation before the lock of the database. The criteria for MI in patients with ACS differed between PCI/PLATFORM and PHOENIX. The main differences are summarized in Table 1. An important difference between PCI/PLATFORM and PHOENIX was the definition of PCI-related MI. Specifically, the assessment of baseline status in patients enrolled with an ACS, which was not formally assessed in PCI/PLATFORM, was a key design characteristic in PHOENIX, where specific MI criteria were associated with different baseline categories. Also, CK-MB was assessed by a central lab using a mass assay in PHOENIX while in PCI and PLATFORM CK-MB was determined by sites, using available assays.

Table 1.

Main differences in percutaneous coronary intervention (PCI)-related myocardial infarction (MI) criteria between Cangrelor versus standard therapy to acHieve optimal Management of Platelet InhibitiON (CHAMPION) PCI/PLATFORM and CHAMPION PHOENIX.

	PCI/PLATFORM	PHOENIX	Comments
Stable angina	CK-MB⩾3×ULN post PCI	CK-MB⩾3×ULN post PCI	Same criteria between PCI/PLATFORM and PHOENIX
ACS
NSTEACS	○ 1 baseline sample ○ Biomarker normal at baseline: MI defined as CK-MB⩾3×ULN post PCI○ Biomarker elevated at baseline: elevation in CK-MB⩾3×ULN and 50% increase from baseline sample or ECG changes	○ 2 baseline samples at least 6 h apart required in NSTEACS patients to confirm resolving MI at baseline.○ Baseline normal patients: MI defined as CK-MB⩾3×ULN post PCI○ Baseline abnormal patients were classified into MI increasing or decreasing at baseline:○ Increasing: Re-elevation in CK-MB post PCI (⩾3×ULN and 50% increase from baseline) + additional evidence of ischemia (2 of 2): ECG changes AND angiographic evidence○ Decreasing: Re-elevation in CK-MB post PCI (⩾3×ULN and 50% increase from baseline) + additional evidence of ischemia (1 of 3): ischemic symptoms, ECG changes, or angiographic evidence	PCI/PLATFORM relied on cardiac biomarkers alone, PHOENIX on a rigorous assessment of baseline status, which included biomarkers (preferably troponin) as well as ischemic symptoms and ECG changes
STEMI	Same criteria as NSTE	Event not considered NA	PCI/PLATFORM did not distinguish between STEMI and NSTEACS. PHOENIX excluded PCI related MI in STEMI patients

ACS: acute coronary syndrome; CK: creatine kinase; ECG: electrographic; NSTE: non-ST-segment elevation; NSTEACS: non-ST-segment elevation acute coronary syndrome; STEMI: ST elevation myocardial infarction.

Criteria for MI in CHAMPION PCI and PLATFORM

In CHAMPION PCI and PLATFORM, criteria used to adjudicate MI were identical. The complete definition of PCI-related MI is listed in the Supplementary Material, eAppendix. To summarize, for patients undergoing elective PCI, including patients with stable angina, the adjudication of PCI-related MI required a post-PCI elevation of CK-MB of at least three times the Upper Limit of Normal (ULN).

In patients who qualified with an ACS, the criteria for PCI-related MI required an elevation of CK-MB of at least three times the ULN but, similar to other definitions used prior to the Universal Definition of Myocardial Infarction, stable or falling biomarkers at the time of PCI were not required to assess PCI-related MI. Also, only one biomarker sample was required by the protocol before PCI for patients who qualified for an ACS.

The definition of spontaneous (non-procedural) MI required clinical manifestations of myocardial ischemia based on signs, symptoms and/or electrocardiographic changes plus an elevation of at least 2×ULN of CK-MB, which was favored over troponin.

Criteria for MI in CHAMPION PHOENIX

In CHAMPION PHOENIX, criteria for PCI-related (type 4a) MI were modeled after the second Universal Definition of Myocardial Infarction and expanded to consider additional complications (symptoms, ECG, angiographic), which were later included by the third Universal Definition of Myocardial Infarction and the SCAI definitions.^8,9

The complete definition of PCI-related MI is listed in the Supplementary Material, eAppendix. To assess PCI-related MI each patient was classified by baseline status (Table 1 and Supplementary Material, Table S2), which could be normal or abnormal. As a general rule, the closer the PCI to the index ACS event (or the higher the uncertainty regarding the clinical presentation at baseline, for example no cardiac markers available before PCI), the more stringent were the corresponding baseline status categories and the more additional evidence (symptoms, ECG, and/or angiographic complications) was required to adjudicate a PCI-related MI. Some case examples displaying the trend of cardiac markers in relationship to hospital admission, randomization, and PCI (used to assess baseline status in PHOENIX)⁶ are included in the appendix.

Baseline status was to be determined based on a combination of biomarker samples (preferably troponin), ischemic symptoms and ECG changes. An emphasis was placed on cardiac biomarkers, and the PHOENIX protocol mandated the collection of at least two troponin samples within 72 h before PCI.

The presence of a normal baseline status required three concurrent conditions: (a) normal troponin before PCI (two samples at least 6 h apart required in patients with unstable angina, only one sufficient in patients who qualified for stable angina) (b) no presumed new ECG changes and (c) no symptoms of myocardial ischemia <6 h prior to PCI, with the last two conditions providing supporting evidence for no spontaneous MI at baseline.

If any of these three conditions were not present, baseline was to be considered abnormal. Also, patients with no biomarkers pre-PCI were classified as abnormal. In patients with abnormal baseline, a further assessment was made as to whether biomarkers were increasing or decreasing at the time of PCI. Baseline was considered as “abnormal increasing” not only if troponin was increasing before PCI but also if only one elevated troponin sample was available or if the two troponin values were collected <6 h apart or if the two elevated troponin values were decreasing but less than 20%. If the second of the two troponin values (i.e. the access site sample) was decreasing by 20% or more, the baseline status was considered “abnormal decreasing and remaining abnormal” if the access site sample was elevated, and “abnormal decreasing and return to normal” if the access site sample was below the ULN. Importantly, patients with confirmed ST-elevation MI were not to be evaluated for possible PCI-related MI.

In patients with normal baseline, including patients with stable angina, the adjudication of PCI-related MI required a post-PCI elevation of CK-MB of at least three times the ULN. For patients determined to be “baseline abnormal” (i.e. baseline MI confirmed or cannot be excluded), typically patients with NSTEACS, MI after PCI required complications (ECG, symptoms, or angiographic) in addition to re-elevation of biomarkers after PCI. At least one additional complication was required in patients with an abnormal decreasing baseline status while two complications were needed in patients with baseline categorized as abnormal increasing. Definition of MI unrelated to PCI adhered to the second Universal Definition of Myocardial Infarction (see Supplementary Material, eAppendix).

Statistical analysis

For this analysis, the outcome of interest is MI. We analyzed PCI-related MI events as defined by the protocol in their respective trials (CHAMPION PCI/PLATFORM and PHOENIX) as well as all adjudicated MIs (PCI-related and non PCI-related). For PCI-related MI, we used events categorized by the CEC as PCI-related in CHAMPION PCI and PLATFORM trials and type 4a MI in PHOENIX. For the analysis on all MIs, every adjudicated MI was included. MI endpoints from randomization to 48-hours were considered. Categorical data are presented as proportions and continuous data are presented as medians (25th, 75th percentiles). Cangrelor’s effect on MI (PCI-related and all MIs) was assessed using logistic regression analysis with MI as endpoint and randomized treatment as the only covariate. According to the original primary efficacy analysis in CHAMPION PCI, PLATFORM, and PHOENIX all analyses were performed in the modified intention-to-treat population (mITT), defined as randomized patients who received at least one dose of study drug and underwent the index PCI. Due to the exploratory nature of these analyses, no formal adjustment for multiple comparisons was performed. A p value <0.05 was considered statistically significant.

Results

Patients

In the overall CHAMPION program (PCI, PLATFORM, and PHOENIX) a total of 24,910 patients were included in the mITT population. Of these, 7941 (31.9%) had stable angina as the clinical presentation and 16,969 (68.1%) had an index ACS. Among ACS patients, 14,378 qualified with a NSTEACS (57.7% of the total) and 2591 (10.4%) with STEMI. Proportions of stable angina and ACS patients in PCI/PLATFORM as compared with PHOENIX are presented in Table 2.

Table 2.

Qualifying disease in Cangrelor versus standard therapy to acHieve optimal Management of Platelet InhibitiON (CHAMPION) PCI/PLATFORM and CHAMPION PHOENIX in the modified intention-to-treat population (mITT) population.

	PCI/PLATFORM	PHOENIX	Total
Stable angina n/N (%)	1583/13,968 (11.3)	6358/10,942 (58.1)	7941/24,910 (31.9)
ACS (STE + NSTE) n/N (%)	12,385/13,968 (88.7)	4584/10,942 (41.9)	16,969/24,910 (68.1)
STEMI n/N (%)	899/13,968 (6.4)	1692/10,942 (15.5)	2591/24,910 (10.4)
NSTEACS n/N (%)	11,486/13,968 (82.2)	2892/10,942 (26.4)	14,378/24,910 (57.7)
Total N (%)	13,968 (100.0)	10,942 (100.0)	24,910 (100.0)

ACS: acute coronary syndrome; NSTE: non-ST-segment elevation; NSTEACS: non-ST-segment elevation acute coronary syndrome; STE: ST-segment elevation; STEMI: ST elevation myocardial infarction.

In PHOENIX, there was a preponderance of patients with stable angina (58.1% of the overall population) as opposed to ACS patients (remaining 41.9%), while in CHAMPION PCI/PLATFORM the vast majority of the population was enrolled with an ACS (88.7%).

Incidence of MI in CHAMPION PCI/PLATFORM and PHOENIX according to clinical presentation

Incidence of PCI-related MIs and all MIs in CHAMPION PCI/PLATFORM versus PHOENIX according to clinical presentation – i.e. stable angina and ACS – are shown in Figure 1 and in the Supplementary Material Figure S1.

In CHAMPION PCI/PLATFORM, the incidence of PCI-related MI at 48 h was 6.3% (885 incident MIs in 13,942 patients) and for the overall MI component was 6.6% (924 MIs/13,942 patients). This was similar in patients with stable angina (6.3%; all MIs were PCI-related) and in patients with ACS (PCI-related MIs 6.4%, all MIs 6.7%). The proportion of patients who had at least two troponin samples before PCI in CHAMPION PCI/PLATFORM was 15% (16.9% in patients with ACS) and similar in CHAMPION PCI (13.9%) and CHAMPION PLATFORM (16.9%).

In CHAMPION PHOENIX, the incidence of PCI-related MI was 4.0% (433 MIs in 10,939 patients) and for the overall MI component was 4.2% (462 MIs/10,939 patients). The MI incidence differed substantially according to clinical presentation. Incidence of PCI-related MI was 5.6% in patients with stable angina and 1.7% in patients with ACS. Similarly, incidence of the overall MI component was 5.9% in patients with SA and 1.9% in patients with ACS. The proportion of patients who had at least two troponin samples before PCI was 93% (94% in patients with ACS).

Cangrelor effect in CHAMPION according to clinical presentation

Cangrelor, as compared with control, did not reduce PCI-related MI in PCI/PLATFORM (OR 1.03, 95% CI 0.90–1.18; p=0.66) while a significant 20% OR reduction of PCI-related MI was observed in PHOENIX (OR 0.80, 95% CI 0.66–0.98; p=0.027). This different effect according to trial (PCI/PLATFORM versus PHOENIX) was significant (p_INT 0.04), indicating heterogeneity of cangrelor effect on MI between PCI/PLATFORM and PHOENIX.

As shown in Figure 1, this heterogeneity of cangrelor effect was limited to patients who qualified for an ACS.

In patients with stable angina, the effect of cangrelor was similar in PCI/PLATFORM (OR 0.79, 95% CI 0.53–1.19) as compared with PHOENIX (OR 0.83, 95% CI 0.67–1.03) with no evidence of heterogeneity (p_INT=0.81). In patients with ACS, there was no observed effect of cangrelor on MI in PCI/PLATFORM (OR 1.07, 95% CI 0.92–1.23; p=0.39) while the MI component was numerically reduced in cangrelor-treated patients in PHOENIX (OR 0.67, 95% CI 0.43–1.07; p=0.089), with a p value for interaction that approached statistical significance (p_INT=0.06). Similarly, in the subgroup of patients with NSTEACS, cangrelor, as compared with clopidogrel, had no observed effect on PCI-related MI in PCI/PLATFORM (OR 1.04, 95% CI 0.90–1.21; p=0.565) while was associated with a numerical reduction of PCI-related MI events in PHOENIX (OR 0.68, 95% CI 0.43–1.08; p=0.10) with a p value for interaction of 0.08. In patients with STEMI, among cangrelor-treated patients there was a numerical increase in PCI-related MI events in PCI/PLATFORM (OR 2.65, 95% CI 0.94–7.51; p=0.056). Treatment effect could not be estimated in PHOENIX where PCI-related (type 4a) MI were not considered in patients with STEMI.

Sensitivity analysis analyzing the overall MIs component instead of PCI-related MI yielded consistent results (Supplementary Material, Figure S1).

Discussion

In a large, phase III clinical program of patients undergoing PCI we observed that the adoption of rigorous criteria for the assessment of baseline status and PCI-related MI in patients with ACS, integrating multiple biomarker assessments, ischemic symptoms, and ECG changes, was associated with markedly lower event rates as compared with prior definitions, and also enhanced the ability to detect a treatment effect.

Implications of PCI-related MI criteria on the interpretation of study results: PCI/PLATFORM versus PHOENIX

After CHAMPION PCI and PLATFORM had been stopped for futility, CHAMPION PHOENIX demonstrated that cangrelor reduced the primary composite endpoint of death, MI, ischemia-driven revascularization, or stent thrombosis at 48 h as compared with clopidogrel, with an effect mostly evident on MI.¹⁵

In this analysis, we observed that the difference of the primary study outcome between CHAMPION PCI/PLATFORM and CHAMPION PHOENIX appeared to result from differences in criteria for assessing PCI-related MI in patients with ACS, rather than differences in cangrelor efficacy.

In PHOENIX the effect of cangrelor appeared very consistent between patients with stable angina and with ACS, with a p value for interaction with clinical presentation close to the null hypothesis (p_INT=0.98). On the other hand, this consistency was not observed in PCI/PLATFORM and also, within ACS patients, between PCI/PLATFORM and PHOENIX. It is plausible therefore that the higher proportion of ACS patients enrolled in PCI/PLATFORM as compared with PHOENIX, along with the use of non-discriminating criteria of PCI-related MI, eventually resulted in a neutral cangrelor effect in the former trials.

The MI component typically dominates the primary composite endpoint in cardiovascular clinical trials.¹ The contribution of the MI component to the composite appears to be particularly relevant in interventional RCTs with short-term follow up.¹ In CHAMPION, MI represented ≈ 80% of the primary composite endpoints in PHOENIX and 90–95% of primary endpoint events in PCI/PLATFORM.^10–12

This preponderance of MI events explains the profound impact of different MI definitions on study results and underscores the importance of an appropriate assessment of this endpoint. Recently, the European Society of Cardiology guidelines for the management of acute coronary syndromes in patients presenting without persistent ST-segment elevation included a class IIb, level of evidence A recommendation for cangrelor indicating conflicting evidence between CHAMPION PCI/PLATFORM and PHOENIX.¹⁶ The data we provide may help clarify these discrepancies.

Implications of PCI-related MI criteria on clinical trial design

In PHOENIX, criteria to define PCI-related MI in patients with ACS emphasized a rigorous assessment of baseline status before PCI, including the implementation of methodology to visually assess biomarker trends in relationship with randomization and PCI and, as expected from preliminary work,⁶ were associated with a substantial reduction of MI rates (approximately two-thirds fewer events) and an enhanced treatment effect. The finding of greater treatment effects (but larger confidence intervals due to fewer events) on larger MI has been already documented for cangrelor⁶ as well as other potent antiplatelet agents.^17,18

This observation suggests that the event rate reduction associated with the use of more stringent MI criteria guided by an integrated assessment of baseline status (and the corresponding perceived reduction of statistical efficiency in sample size estimation), appears less important than the ability to discover the presence of a higher and more reliable treatment effect and thus cautions against the use of a PCI-related MI definition with low discriminatory ability in clinical trials.

Notably, event rates associated with a more stringent definition of PCI-related MI in patients with ACS were lower than corresponding event rates in patients with stable angina, a setting where procedural complications are usually less common. This apparent paradox suggests that, when establishing criteria for PCI-related MI, the clinical setting may need to be accounted for. Additional considerations may be the requirements for PCI-related MIs to be prognostically homogeneous across clinical settings (e.g. with similar attributable risk for subsequent death), which may, in turn, require larger elevation of cardiac markers post-PCI in elective patients as compared with patients with ACS.

Finally, larger PCI-related MIs, which could be obtained using a higher threshold for cardiac markers post-PCI and/or additional ECG, clinical, or angiographic supporting criteria, have been previously shown to have an association with mortality similar to that of spontaneous MI,¹⁹ an important design characteristic for appropriate use of composite endpoints in clinical trials. Also, a clear association between PCI-related MI events and mortality in patients with ACS has been observed when a careful assessment of baseline status was implemented,³ further emphasizing the importance of optimal baseline assessment.

Limitations

The presence of the heterogeneous effect of cangrelor between ACS and stable angina patients was not expected in CHAMPION PCI/PLATFORM. Therefore this is a post-hoc, not a pre-specified, analysis. It is possible that other factors, beyond MI definition, may have contributed to different study results in CHAMPION PHOENIX as compared with CHAMPION PCI/PLATFORM including the possibility of using 300 mg of clopidogrel in the comparator arm or the use of other concomitant therapies. Cangrelor efficacy, however, was consistent by clopidogrel loading dose (300 versus 600 mg) in CHAMPION PHOENIX. Also, due to several differences in MI definition between PCI/PLATFORM and PHOENIX it is difficult to clearly establish which specific aspects of the MI definition may have contributed to the final results. To conclusively demonstrate our research question we should probably have applied the PHOENIX MI definition in PCI/PLATFORM and the PCI/PLATFORM MI definition in PHOENIX. However, this is not possible mostly due to the limited number of troponin samples at baseline in PCI/PLATFORM.

Finally, in the CHAMPION program, PCI-related MI was defined using CK-MB, based on a robust association of this cardiac marker with subsequent mortality and more clearly established diagnostic threshold as compared with troponin.^2–4 We do not know, therefore, if, and how, these results would have been different with the use of troponin.

Conclusions

More stringent criteria to diagnose PCI-related MI in patients with ACS, with an emphasis on baseline status before PCI, were associated with a reduced event rate but enhanced ability to detect a treatment effect. These results illustrate the relevant implications that endpoint criteria may have on trial results and emphasize the importance of a careful assessment of PCI-related MI.

Supplemental Material

ACC661692_Supplementary_Material_E-appendix – Supplemental material for Implications of different criteria for percutaneous coronary intervention-related myocardial infarction on study results of three large phase III clinical trials: The CHAMPION experience

Supplemental material, ACC661692_Supplementary_Material_E-appendix for Implications of different criteria for percutaneous coronary intervention-related myocardial infarction on study results of three large phase III clinical trials: The CHAMPION experience by Sergio Leonardi, Renato D Lopes, Ph. Gabriel Steg, Freddy Abnousi, Alberto Menozzi, Jayne Prats, Stacey Mangum, Matthew Wilson, Meredith Todd, Gregg W Stone, C Michael Gibson, Christian W Hamm, Matthew J Price, Harvey D White, Robert A Harrington, Deepak L Bhatt and Kenneth W Mahaffey in European Heart Journal: Acute Cardiovascular Care

Footnotes

Acknowledgements

The authors wish to thank Steven Elkin, The Medicines Company, for his support on data management and analysis.

Conflict of interest

DL Bhatt discloses the following relationships – advisory board: Cardax, Elsevier Practice Update Cardiology, Medscape Cardiology, Regado Biosciences; board of directors: Boston VA Research Institute, Society of Cardiovascular Patient Care; chair: American Heart Association Quality Oversight Committee; Data Monitoring Committees: Duke Clinical Research Institute, Harvard Clinical Research Institute, Mayo Clinic, Population Health Research Institute; honoraria: American College of Cardiology (senior associate editor, Clinical Trials and News, ), Belvoir Publications (editor in chief, Harvard Heart Letter), Duke Clinical Research Institute (clinical trial steering committees), Harvard Clinical Research Institute (clinical trial steering committee), HMP Communications (editor in chief, Journal of Invasive Cardiology), Journal of the American College of Cardiology (guest editor; associate editor), Population Health Research Institute (clinical trial steering committee), Slack Publications (chief medical editor, Cardiology Today’s Intervention), Society of Cardiovascular Patient Care (secretary/treasurer), WebMD (CME steering committees); Other: Clinical Cardiology (deputy editor), NCDR-ACTION Registry Steering Committee (vice-chair), VA CART Research and Publications Committee (chair); research funding: Amarin, AstraZeneca, Bristol-Myers Squibb, Eisai, Ethicon, Forest Laboratories, Ischemix, Medtronic, Pfizer, Roche, Sanofi Aventis, The Medicines Company (co-chair of CHAMPION trials); royalties: Elsevier (editor); site co-investigator: Biotronik, Boston Scientific, St Jude Medical; trustee: American College of Cardiology; unfunded research: FlowCo, PLx Pharma, Takeda.

CM Gibson reports grants and personal fees from Angel Medical Corporation, AstraZeneca, Atrium Medical Systems, Baxter Healthcare, Bayer, Cardiovascular Research Foundation, Consensus Medical Communications, CSL Behring, Cytori Therapeutics, Daiichi Sankyo Company, Eli Lilly and Company, Exeter Group, Genentech, GlaxoSmithKline, Ikaria, Janssen Pharmaceuticals, Johnson and Johnson, Lantheus Medical Imaging, Merck, Ortho-McNeil, Portola Pharmaceuticals, Roche Diagnostics, Sanofi-Aventis, Stealth Peptides, St Jude Medical, The Medicines Company, UpToDate in Cardiovascular Medicine Volcano Corp, and Walk Vascular, during the conduct of the study.

RA Harrington reports grants and personal fees from Astra Zeneca, Bristol-Myers Squibb, Sanofi, The Medicines Company, Lilly, Daiichi Sankyo, GlaxoSmithKline, Johnson and Johnson, Portola, Merck, and Regado, Sanofi, Bristol-Myers Squibb, Merck, Johnson and Johnson, and Gilead, during the conduct of the study.

S Leonardi discloses the following relationships: Advisory board: The Medicines Company, Ely Lilly, Daiichi Sankyo.

KW Mahaffey reports grants and personal fees from Abbott Vascular, Adolor, American College of Cardiology, Amgen, Amylin/BMS, AstraZeneca, Baxter, Bayer, Biotronik, Boehringer Ingelheim, Bristol-Myers Squibb, Cordis, Cubist, Daiichi Sankyo, Dialouges, Duke Center for Educational Excellence, Edwards Lifesciences, Eli Lilly, Elsevier, Forest, Genentech, Gilead Science, GlaxoSmithKline, Guidant, Haemonetics, Ikaria, Janssen, Johns Hopkins University, Johnson and Johnson, Luitpold, Medtronic, Merck, Novartis, Ortho-McNeill, Pfizer, Polymedix, Portola, Pozen, Regado, Regeneron, Roche, Sanofi, Schering-Plough, St Jude Medical, The Medicines Company, South East Area Health Education Center, Springer Publishing, Sun Pharma, and University of British Columbia, during the conduct of the study.

J Prats and M Todd are fulltime employees of the Medicines Company.

PG Steg reports personal fees from Amarin, AstraZeneca, Bayer, Boehringer Ingelheim, Bristol-Myers Squibb, Daiichi Sankyo, GlaxoSmithKline, Lilly, Merck-Sharpe-Dohme, Novartis, Otsuka, Pfizer, Roche, Sanofi, Servier, The Medicines Company, and Vivus during the conduct of the study.

GW Stone reports personal fees from GWS, has been a consultant to Boston Scientific, Eli Lilly, Daiichi Sankyo, and AstraZeneca, during the conduct of the study.

RD Lopes reports personal fees from Consulting fees from Pfizer (significant), Bayer Corporation US (Bayer AG/Bayer in Japan – subsidiaries) Boehringer Ingelheim, Bristol Myers Squibb Glaxo SmithKline (GSK), Merck, Portola, grants from Research Grant from GSK, Bristol Myers Squibb, Merck outside the submitted work.

CW Hamm reports personal fees from CWH has received honoraria from Abbott, AstraZeneca, Bayer, Berlin Chemie, Boehringer Ingelheim, Merck-Sharpe-Dohme, Bristol-Myers Squibb, BRAHMS, Daiichi Sankyo, Essex, GlaxoSmithKline, Medtronic, Lilly, Sanofi-Aventis, Correvio, Pfizer, Roche, The Medicines Company, Boston Scientific, and Gilead., during the conduct of the study.

HD White reports grants from Sanofi-Aventis, grants from Eli Lilly, grants from National Institute of Health, grants from GSK, grants from Merck Sharpe & Dohme, grants and personal fees from AstraZeneca, outside the submitted work.

F Abnousi, S Mangum, M Wilson, and MJ Price have nothing to disclose.

Funding

The CHAMPION PLATFORM, CHAMPION PCI, and CHAMPION PHOENIX trials were funded by The Medicines Company. No funding was received for the preparation of this manuscript.

References

Leonardi

Armstrong

Schulte

. Implementation of standardized assessment and reporting of myocardial infarction in contemporary randomized controlled trials: A systematic review. Eur Heart J 2013; 34: 894–902.

Roe

Mahaffey

Kilaru

. Creatine kinase-MB elevation after percutaneous coronary intervention predicts adverse outcomes in patients with acute coronary syndromes. Eur Heart J 2004; 25: 313–321.

Tricoci

Leonardi

White

. Cardiac troponin after percutaneous coronary intervention and 1-year mortality in non-ST-segment elevation acute coronary syndrome using systematic evaluation of biomarker trends. J Am Coll Cardiol 2013; 62: 242–251.

Bonaca

Wiviott

Braunwald

. American College of Cardiology/American Heart Association/European Society of Cardiology/World Heart Federation universal definition of myocardial infarction classification system and the risk of cardiovascular death: Observations from the TRITON-TIMI 38 trial (Trial to Assess Improvement in Therapeutic Outcomes by Optimizing Platelet Inhibition With Prasugrel-Thrombolysis in Myocardial Infarction 38). Circulation 2012; 125: 577–583.

Capodanno

Angiolillo

. Pretreatment with antiplatelet drugs in invasively managed patients with coronary artery disease in the contemporary era: Review of the evidence and practice guidelines. Circ Cardiovasc Interv 2015; 8: e002301.

Leonardi

Truffa

Neely

. A novel approach to systematically implement the universal definition of myocardial infarction: Insights from the CHAMPION PLATFORM trial. Heart 2013; 99: 1282–1287.

Thygesen

Alpert

White

; Joint ESC/ACCF/AHA/WHF Task Force for the Redefinition of Myocardial Infarction. Universal definition of myocardial infarction. J Am Coll Cardiol 2007; 50: 2173–2195.

Thygesen

Alpert

Jaffe

.;ESC Committee for Practice Guidelines (CPG). Third universal definition of myocardial infarction. Eur Heart J 2012; 33: 2551–2567.

Moussa

Klein

Shah

. Consideration of a new definition of clinically relevant myocardial infarction after coronary revascularization: An expert consensus document from the Society for Cardiovascular Angiography and Interventions (SCAI). J Am Coll Cardiol 2013; 62: 1563–1570.

10.

Harrington

Stone

McNulty

. Platelet inhibition with cangrelor in patients undergoing PCI. N Engl J Med 2009; 361: 2318–2329.

11.

Bhatt

Lincoff

Gibson

.;CHAMPION PLATFORM investigators. Intravenous platelet blockade with cangrelor during PCI. N Engl J Med 2009; 361: 2330–2341.

12.

Bhatt

Stone

Mahaffey

.;CHAMPION PHOENIX investigators. Effect of platelet inhibition with cangrelor during PCI on ischemic events. N Engl J Med 2013; 368: 1303–1313.

13.

White

Chew

Dauerman

. Reduced immediate ischemic events with cangrelor in PCI: A pooled analysis of the CHAMPION trials using the universal definition of myocardial infarction. Am Heart J 2012; 163: 182–190.

14.

Leonardi

Mahaffey

White

. Rationale and design of the Cangrelor versus standard therapy to acHieve optimal Management of Platelet InhibitiON PHOENIX trial. Am Heart J 2012; 163: 768–776.

15.

Steg

Bhatt

Hamm

.;CHAMPION investigators. Effect of cangrelor on periprocedural outcomes in percutaneous coronary interventions: A pooled analysis of patient-level data. Lancet 2013; 382: 1981–1992.

16.

Roffi

Patrono

Collet

. 2015 ESC Guidelines for the management of acute coronary syndromes in patients presenting without persistent ST-segment elevation: Task Force for the Management of Acute Coronary Syndromes in Patients Presenting without Persistent ST-Segment Elevation of the European Society of Cardiology (ESC). Eur Heart J 2016; 37(3): 267–315.

17.

Morrow

Wiviott

White

. Effect of the novel thienopyridine prasugrel compared with clopidogrel on spontaneous and procedural myocardial infarction in the Trial to Assess Improvement in Therapeutic Outcomes by Optimizing Platelet Inhibition with Prasugrel-Thrombolysis in Myocardial Infarction 38: An application of the classification system from the universal definition of myocardial infarction. Circulation 2009; 119: 2758–2764.

18.

Leonardi

Tricoci

White

. Effect of vorapaxar on myocardial infarction in the thrombin receptor antagonist for clinical event reduction in acute coronary syndrome (TRA·CER) trial. Eur Heart J 2013; 34: 1723–1731.

19.

Leonardi

Thomas

Neely

. Comparison of the prognosis of spontaneous and percutaneous coronary intervention-related myocardial infarction. J Am Coll Cardiol 2012; 60: 2296–2304.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

9.86 MB