Article Text

Download PDFPDF

Streamlining cardiovascular clinical trials to improve efficiency and generalisability
  1. Faiez Zannad1,
  2. Marc A Pfeffer2,
  3. Deepak L Bhatt3,
  4. Denise E Bonds4,
  5. Jeffrey S Borer5,6,
  6. Gonzalo Calvo-Rojas7,
  7. Louis Fiore8,
  8. Lars H Lund9,
  9. David Madigan10,
  10. Aldo Pietro Maggioni11,
  11. Catherine M Meyers12,
  12. Yves Rosenberg4,
  13. Tabassome Simon13,14,
  14. Wendy Gattis Stough15,
  15. Andrew Zalewski16,
  16. Nevine Zariffa17,
  17. Robert Temple18
  1. 1 Clinical Investigation Center, Centre Hospitalier Universitaire de Nancy, Nancy, France
  2. 2 Cardiovascular Division, Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts, USA
  3. 3 Brigham and Women's Heart and Vascular Center, Harvard Medical School, Boston, Massachusetts, USA
  4. 4 National Heart, Lung, and Blood Institute, National Institutes of Health, Bethesda, Maryland, USA
  5. 5 The Howard Gilman Institute, New York, New York, USA
  6. 6 State University of New York Downstate Medical Center, Brooklyn, New York, USA
  7. 7 Department of Clinical Pharmacology, Hospital Clínic, University of Barcelona, Barcelona, Spain
  8. 8 Department of Veterans Affairs, Cooperative Studies Program, Boston, Massachusetts, USA
  9. 9 Department of Medicine, Unit of Cardiology, Karolinska Institute, Stockholm, Sweden
  10. 10 Department of Statistics, Columbia University, New York, New York, USA
  11. 11 ANMCO Research Center, Florence, Italy
  12. 12 National Center for Complementary and Integrative Health, National Institutes of Health, Bethesda, Maryland, USA
  13. 13 Assistance Publique-Hôpitaux de Paris, Saint Antoine Hospital, Paris, France
  14. 14 Université Pierre et Marie Curie, Paris, France
  15. 15 Campbell University College of Pharmacy and Health Sciences, Research Triangle Park, North Carolina, USA
  16. 16 Glaxo Smith Kline, King of Prussia, Pennsylvania, USA
  17. 17 AstraZeneca, Cambridge, UK
  18. 18 Center for Drug Evaluation and Research, United States Food and Drug Administration, Silver Spring, Maryland, USA
  1. Correspondence to Professor Faiez Zannad, Clinical Investigation Center, Centre Hospitalier Universitaire de Nancy, Université de Lorraine and INI-CRCT (F-CRIN); Institut Lorrain du Coeur et des Vaisseaux, Nancy 54500, France; f.zannad{at}chru-nancy.fr

Abstract

Controlled trials provide the most valid determination of the efficacy and safety of an intervention, but large cardiovascular clinical trials have become extremely costly and complex, making it difficult to study many important clinical questions. A critical question, and the main objective of this review, is how trials might be simplified while maintaining randomisation to preserve scientific integrity and unbiased efficacy assessments. Experience with alternative approaches is accumulating, specifically with registry-based randomised controlled trials that make use of data already collected. This approach addresses bias concerns while still capitalising on the benefits and efficiencies of a registry. Several completed or ongoing trials illustrate the feasibility of using registry-based controlled trials to answer important questions relevant to daily clinical practice. Randomised trials within healthcare organisation databases may also represent streamlined solutions for some types of investigations, although data quality (endpoint assessment) is likely to be a greater concern in those settings. These approaches are not without challenges, and issues pertaining to informed consent, blinding, data quality and regulatory standards remain to be fully explored. Collaboration among stakeholders is necessary to achieve standards for data management and analysis, to validate large data sources for use in randomised trials, and to re-evaluate ethical standards to encourage research while also ensuring that patients are protected. The rapidly evolving efforts to streamline cardiovascular clinical trials have the potential to lead to major advances in promoting better care and outcomes for patients with cardiovascular disease.

  • clinical trials as topic
  • cardiovascular disease
  • pragmatic clinical trials as topic

Statistics from Altmetric.com

Introduction

Cardiovascular randomised controlled trials are essential to ensure that therapeutic interventions are effective and safe for their intended populations. Impediments in the current research enterprise have been noted, including the need to recruit large numbers of patients often over several years, costs associated with extensive and sometimes unnecessary data collection, complex designs, and inefficient research infrastructure.1–3 Important clinical questions may remain unanswered because of the increasing complexity and costs of clinical research.4

Randomised controlled trials remain the most trusted approach to determining the efficacy and safety of an intervention; thus, streamlining clinical trials to minimise impediments is a high priority. Several alternative methods have been proposed that allow randomisation, and possibly blinding, while streamlining trial operations, such as using registries or data already collected for routine clinical care.5 Some trials have implemented these designs,6–10 but more experience is needed prior to their broad adoption. Some stakeholders may be reluctant to promote experience with these methods because of unfamiliarity with their use or concerns that the results will not be accepted.1 11 Therefore, a consensus must be developed about the best way to evaluate their appropriateness and utility.

These concepts were discussed among a group of cardiovascular clinical trialists, clinicians, biostatisticians, European and US regulators, and government and industry sponsors during the annual Global Cardiovascular Clinical Trialists Forum held in Washington, DC (http://www.globalcvctforum.com). This article summarises important conclusions, remaining uncertainties and priorities for future research with these new approaches.

Value of streamlined clinical trials

The infrastructure to support cardiovascular clinical research was designed initially to perform specific functions needed for regulatory compliance, data collection, data management systems and site-based monitoring. Other functions have been added over time to anticipate possible information that might be requested by regulators. These activities are often attributed to regulations (eg, some types of safety reporting), but on close inspection may reflect overinterpretation of regulations.12 Moreover, additional data are often collected to support secondary and exploratory analyses and publications. An estimated 18% (US$1.1 million) of a study’s budget is spent on supplementary or exploratory endpoints, and US$4–6 billion is spent annually on procedures that generate non-critical data in phase 2 and 3 clinical trials (not limited to cardiovascular trials) (Tufts Center for the Study of Drug Development. Extraneous data collected in clinical trials cost drug developers $4 billion to $6 billion annually, according to Tufts Center for the Study of Drug Development (Press release 6 Nov 2012)). The operations to support collection of these data have increased and are costly, and they often duplicate data collected during routine clinical care. In some cases, this increased trial complexity may contribute to disturbingly low enrolment and retention rates because of frequent visits and procedures, which can be burdensome to patients13; other time-intensive administrative hurdles, for example legal contracts, may also discourage individual researchers or their institutions from participating in clinical trials. A further problem is that the incremental treatment effects for most new cardiovascular therapies are modest in magnitude because of the major advances already realised for most cardiovascular diseases, so that thousands to tens of thousands of patients are required to achieve adequate power to demonstrate the smaller increment.11 Streamlining clinical trials by using existing infrastructure, personnel and data collected as part of routine clinical care has the potential to enhance efficiency, achieve faster recruitment, lower costs and restore the feasibility of some types of cardiovascular clinical trials (table 1). Opportunities may be increased for development of new therapies and for studying critical questions that arise post licensing, including comparative effectiveness and safety studies in particular populations, and studies in patients who have failed prior therapy.

Table 1

Strengths and limitations of registry-based randomised controlled trials

Streamlined research methodologies suitable for cardiovascular clinical trials

Examples of registry-based randomised controlled trials

Observational registries have often been used as data sources for clinical research, but comparisons of different treatments within registries can be biased because many factors influence the choice to use an intervention for a specific patient. It is not possible to overcome this potential for bias even with sophisticated statistical methods.14 Randomisation within a registry system addresses the problem of patient allocation bias while still capitalising on the benefits and efficiencies of the registry (eg, using data already being collected or an infrastructure already in place to collect data in the more generalisable setting of clinical care)11 (figure 1). The Thrombus Aspiration in ST-Elevation Myocardial Infarction in Scandinavia (TASTE) was a landmark trial comparing the use of thrombus aspiration with no aspiration prior to percutaneous coronary intervention (PCI) that demonstrated the feasibility of the registry-based randomised controlled trial.6 TASTE capitalised on the existing infrastructure of the nationwide Swedish Coronary Angiography and Angioplasty Registry and the Swedish Web System for Enhancement and Development of Evidence-based Care in Heart Disease Evaluated According to Recommended Therapies (SWEDEHEART). These registries are both managed by an academic research organisation, and facilitate online collection of baseline and procedural data and data monitoring among all centres performing PCI in Sweden. The registry is also linked to the Swedish National Population Registry, from which the primary endpoint (all-cause mortality at 30 days) was ascertained.6 An online randomisation module was the only addition to the existing registry platform specifically for the TASTE trial; baseline and procedural data were entered online directly into the registry per the standard procedure for patients undergoing PCI in Sweden. The study was not blinded, but the mortality endpoint is less susceptible to confounding and bias than non-mortality endpoints. Data monitoring and adjudication were performed as part of routine registry validation, and no monitoring specific for the trial was performed. Patients provided initial oral consent, followed by written informed consent within 24 hours. The study was conducted at a remarkably low incremental cost.11 15 16 The high proportion (61%) of eligible patients that were enrolled was another strength of this study.6 TASTE answered an important clinical question, finding no significant benefit of aspiration thrombectomy in patients with ST-elevation myocardial infarction (MI) undergoing PCI. The finding from TASTE was confirmed in a traditional randomised controlled trial, the Trial of Routine Aspiration Thrombectomy with PCI versus PCI Alone in Patients with STEMI (TOTAL), which showed no benefit of aspiration thrombectomy on the primary composite endpoint of cardiovascular death, recurrent MI, cardiogenic shock or New York Heart Association class IV heart failure within 180 days and an increased risk of stroke within 30 days.17

Figure 1

Schematic of registry-based randomised clinical trial concept.

Other examples of registry-based randomised controlled trials illustrate the diverse means used to gather information on outcomes, baseline characteristics and other critical data. In the Study of Access Site for Enhancement of Percutaneous Coronary Intervention for Women, women undergoing urgent or elective PCI or diagnostic cardiac catheterisation with the possibility of PCI were randomised to radial or femoral arterial access using an online randomisation module within the existing CathPCI Registry database. Demographics, medical history, concomitant medications, procedural information and index hospital outcomes from consenting patients were extracted from the existing National Cardiovascular Data Registry (NCDR). Additional study-specific data were collected to supplement NCDR data fields.8 9 The site network participating in the NCDR′s CathPCI Registry was the source of participating investigators in this study. Investigators reported a reduction in site coordinator workload.8

The VALIDATE-SWEDEHEART trial will randomise patients within the SWEDEHEART registry to receive heparin only or bivalirudin (with or without low-dose heparin according to local practice prior to randomisation) to determine if bivalirudin is superior to heparin alone in patients with MI who are undergoing PCI and pretreated with ticagrelor or prasugrel.16 18 The SWEDEHEART registry will be used to perform randomisation and collect baseline data. The primary endpoint of this study is the composite of death, MI and major bleeding. Clinical endpoint data will be collected from national registries, public databases and telephone follow-up; endpoints will be adjudicated by a blinded central endpoint adjudication committee.16 18 Preliminary data suggest that patients representative of a ‘real-life’ population are enrolled.18

Broader application of registry-based randomised trials

These examples illustrate the feasibility of registry-based randomised controlled trials to answer questions about effectiveness relevant to daily clinical practice that were unlikely to be supported using traditional research sponsors and methods (eg, drugs were generic, or devices or procedures were available and inexpensive). Translating the experience of TASTE and similar trials, all of which involved existing treatments, to industry-sponsored studies intended to support new drug approvals or new uses is also of interest.19 To our knowledge, large, phase 3, industry-sponsored, pivotal trials have not yet implemented registry-based randomised controlled trial designs, but stakeholders should consider whether this design is feasible and can be validated for such an application, for postapproval studies or for head-to-head comparisons of existing treatments. VALIDATE-SWEDEHEART will contribute important knowledge in this regard. However, there are some applications for which registry-based randomised trial designs would not be suitable, for example where internal validity (ie, ability to attribute observed effects to a treatment or study arm rather than to confounding) is a high priority or where substantial amounts of data are needed that are not collected in the setting of routine clinical care (table 2).

Table 2

Applications of randomised registry trials

An important question is how the structure could also facilitate blinded trials. The examples given are all open-label studies. While the open-label design can be viewed as a limitation, an all-cause mortality endpoint or use of blinded endpoint committee adjudication minimises potential bias in endpoint assessment for these unblinded trials. An open-label, registry-based randomised controlled trial design would not appear appropriate, however, for studies where knowledge of assignment could undermine the integrity of the study, or in which subjective primary endpoints were used, as these would almost invariably involve judgement that could be affected by knowledge of treatment (table 2). There is, however, no obvious impediment to incorporation of placebo controls and specified investigator assessments within the registry framework, although this would clearly add to costs. This design may be problematic for trials of new experimental drugs or devices that require extensive safety reporting, especially for large global trials where data standard requirements vary across countries. Trials requiring extensive data collection beyond that typically collected during routine patient care (and therefore not likely to be part of electronic health records) will reduce the efficiency of the registry-based randomised design. Moreover, the registry-based trials discussed above were conducted in a society with highly reliable, centralised event-reporting and data availability for inpatients and outpatients, within a single payer healthcare system. The feasibility of translating this experience to non-Scandinavian countries is uncertain at present. Finally, registry-based randomised controlled trials remain subject to some of the same design challenges as traditional randomised controlled trials (eg, confounding effects of prerandomisation treatments, difficulty interpreting endpoints that combine efficacy and safety (death, MI, bleeding)).

Pragmatic trials in integrated clinical research networks

In current healthcare systems, research is not necessarily seen as an integral part of the system. Embedding research into health service systems could deliver a ‘learning health service’. The concept is possible because most health services already collect the sort of data that one would wish to collect in clinical trials. Evolution towards an environment where participation in clinical research within a learning health system is the norm rather than the exception (with opt-out provisions as appropriate) is critically needed.

Randomised trials within healthcare organisations or patient networks seem easiest to consider for studies testing two (or more) accepted treatment options (eg, to determine the best treatment option among different drugs within the same class). Comparison of more than two treatments is rarely conducted by the drug development industry but is of interest (eg, anticoagulants and antiplatelet therapies). Comparing standard of care with standard of care plus new drug would also be of interest, either to assess effectiveness or to determine the needed duration of added therapy. These trials need to be blinded, but this seems feasible.

The Health Care Systems Research Collaboratory is an initiative of the US National Institutes of Health (NIH) that supports the partnership between researchers and healthcare organisations to answer clinical questions in the setting of routine care. These point-of-care clinical trials are conducted within the existing clinical infrastructure using clinical staff and healthcare system leadership and can use individual or cluster randomisation. Electronic health records are used as the source for most data.20 This approach eliminates the need for a duplicate infrastructure to support research, which can reduce costs,21 although an initial investment to establish and maintain the infrastructure remains necessary.

The Patient-Centered Outcomes Research Institute (PCORI) PCORnet uses integrated health and patient networks to conduct observational and interventional comparative effectiveness research.22 23 The ADAPTABLE (Aspirin Dosing: A Patient-centric Trial Assessing Benefits and Long-Term Effectiveness) trial is a randomised, pragmatic trial comparing the effect of aspirin 325 mg/day or aspirin 81 mg/day on the primary composite outcome of death, hospitalisation for non-fatal MI or stroke embedded within usual care that is conducted within this framework.10 It has broad eligibility criteria with few exclusions and achieves data collection using existing electronic health records, health system encounters, claims and patient-reported data.10 The primary composite efficacy endpoint is death, hospitalisation for non-fatal MI or stroke, and the primary safety endpoint is major bleeding complications. These endpoint data will be systematically collected from health system encounters mapped to a common data model, claims data and patient reporting.10 Non-cardiovascular point-of-care trials have been successfully conducted.24

Challenges to implementing streamlined clinical trials

Ethical considerations and informed consent

The regulations that guide the conduct of human subjects research originated largely in response to several events that harmed or exploited research subjects.25 While these regulations importantly advanced protections for patients participating in clinical trials, their application to research that compares accepted standards of care (eg, comparative effectiveness of existing therapies, quality improvement, point-of-care cluster randomisation) needs further consideration. Obtaining proper informed consent remains an important need, but the means of consenting patients in such cases might be modified, particularly for cluster randomisation.

Informed consent concerns have in some cases unnecessarily delayed streamlined, pragmatic trials that did not appear to pose important risks to subjects.26 The need to revise existing regulations to address the unique needs of streamlined or pragmatic trials (particularly randomised comparative effectiveness studies) has been recognised. The UK has implemented processes to achieve ethics reviews and oversight that is proportional to the risk of the individual study.27 The European Union Regulation (No 536/2014) includes a provision for simplified informed consent procedures in the setting of cluster randomisation.28 The US Department of Health and Human Services has proposed revisions to the Federal Policy for the Protection of Human Subjects.29 The National Heart Lung and Blood Institute has called for investigators to develop and test new methods of obtaining informed consent (HL-12–019).15

Although there is a view that the usual requirements for study conduct may be unnecessarily burdensome for some of the studies described above, there is no clear consensus on change that would relieve this burden while still protecting patients’ rights to make decisions about their care and the privacy of their medical information, including future reuse of their data for further advancement of science outside the initial research hypothesis.30–41 It seems obvious that informed consent for comparisons of standard treatments could be shorter than what is needed for a treatment with uncertain risks. Centralised institutional review board, an underutilised approach, would also be appropriate for many trials.

Data collection and analysis considerations

Prospective, randomised trials that use existing large databases or electronic health records to identify and enrol patients and to assess (or partially assess) outcomes are entering new territory regarding data quality, management and analysis. Traditional cardiovascular clinical trials have study-specific case report forms and databases that trained study personnel use to capture data deemed relevant for the specific study. In contrast, electronic health records and often large registry data are collected and entered by the personnel involved in the routine clinical care of the patient, although the personnel and organisational resources vary across regions of the world. In some cases, the data may be coded by non-clinical personnel based on medical record documentation, and data that are coded for the purpose of billing may introduce certain biases.42 Thus, data quality concerns may arise with regard to data accuracy or completeness. ‘Noisy’ data will, in general, create a ‘bias toward the null’, making it more difficult to detect treatment differences. Timeliness of data may also be an issue if data are not entered at the point of care but require subsequent coding.42 Creating linkages between distinct electronic systems with different data structures and data definitions, or mapping electronic health record fields to electronic case report form fields, can be technically problematic, especially in multicentre, international trials. Professional societies have advocated for the conduct of validation studies that use electronic data sources (eg, registries or electronic health records) in parallel with traditional electronic case report forms for data collection to determine whether the results are influenced by the data collection method.42 43 Multistakeholder collaboration to develop standards for data sharing, communication between electronic health records and electronic case report forms, harmonisation efforts across systems, common data standards and variables, and linkages between electronic health records and national death registries have been proposed as potential solutions to the challenges encountered when using electronic health record data for clinical research.42

Data security and privacy are major concerns for observational studies conducted using large databases and for pragmatic, randomised trials. Programmes such as the NIH Health Care Systems Research Collaboratory, PCORI and Mini-Sentinel have carefully developed procedures to ensure data privacy by using distributed analyses where the data remain under the participating institution’s control and only aggregated results are transmitted.10 22 23 44

Regulatory considerations

The concept of large, simple trials is not new from a regulatory perspective.45 46 Broad entry criteria enhance the generalisability of trial results and increase the likelihood that an effective therapy will be rapidly adopted into clinical practice. The European Medicines Agency and the US Food and Drug Administration are supportive of reducing the data collection and monitoring burden to focus on information that contributes to achieving meaningful knowledge about the therapy; several guidance documents and regulations have urged this.28 47 48 Other guidelines (eg, International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use – Good Clinical Practice), or their overinterpretation, may lead to complexities that inhibit adoption of innovative approaches to pragmatic trials, for example the perceived need to collect information on non-serious adverse events, especially for late-stage trials.12 The involvement of all stakeholders is needed to ensure that global clinical trial regulations reflect activities that ensure high scientific integrity and produce reliable, clinically relevant results that are of importance to patients.

While regulators support large simple trials, the acceptance of novel designs (eg, registry-based randomised controlled trials or point-of-care randomised trials) in particular cases is less certain, especially for pivotal trials supporting a new drug application (table 2). The analytic and technical issues previously discussed are major determinants of whether these trial designs will be acceptable to provide the level of unbiased evidence required for approval.42 A particular problem would be the conduct of non-inferiority studies using such approaches, as the ability of the control agent to have its expected effect in novel circumstances could be open to doubt,49 and, as noted, ‘noisy’ data introduce a ‘bias toward the null’, that is, a finding of non-inferiority. Registry-based randomised controlled trials or randomised point-of-care designs are especially attractive for many postmarketing studies of both drugs and devices, which should make postmarketing evaluations even more informative, given that many of them are currently only observational. This action would enhance the quality of postmarketing studies through enabling randomisation at a reasonable cost, while simultaneously allowing sponsors and regulators to gain experience with these methods of randomisation, assess data quality, perform validation studies and determine their suitability for pivotal trials.

Conclusions

The many ongoing initiatives to streamline cardiovascular clinical trials and conduct pragmatic research are rapidly advancing the field. Registry-based randomised or pragmatic trial designs may make economically viable comparative effectiveness studies a reality, although experience with comparative drug trials is extremely limited. An important challenge is applying these concepts to trials of drugs and devices under development. Collaboration among stakeholders is necessary to achieve standards for data management and analysis, to validate large data sources for use in randomised trials, and to re-evaluate ethical standards to encourage research and ensure patients are protected.

Acknowledgments

The authors acknowledge the contributions of Robert Golub, MD, John Jarcho, MD, Stuart Spencer, MD, and Robert Califf, MD, for their contributions on the faculty panel during the CVCT meeting from which this paper is based. The contents do not represent the official views of the National Institutes of Health, the U.S. Department of Veterans Affairs, or the U.S. Government. This article reflects the views of the authors and should not be construed to represent FDA’s views or policies.

References

Footnotes

  • Contributors FZ: Developed workshop agenda and participated in workshop discussion of this topic; drafted the paper; revised the paper for important intellectual content; approved the final submitted version; agrees to be accountable for all aspects of the work. MAP, DLB, DEB, JSB, GC, LF, LHL, DM, APM, CMM, YR, TS, AZ, NZ, RT: Participated as a participant and discussant during the workshop from which this paper was developed; drafted the paper; revised the paper for important intellectual content; approved the final submitted version; agrees to be accountable for all aspects of the work. WGS: Drafted the paper; revised the paper for important intellectual content; approved the final submitted version; agrees to be accountable for all aspects of the work.

  • Funding This work was generated from discussions during the 11th Global Cardiovascular Clinical Trialists (CVCT) Forum held in Washington, DC in December 2014. CVCT was organised by the Clinical Investigation Center (CIC) Inserm, CHU, and the University of Lorraine, France, and INI-CRCT (F-CRIN), Nancy, France, and funded by an unrestricted educational grant from Association de Recherche et d'Information en Cardiologie (ARISC), a non-profit educational organisation in Nancy, France. ARISC had no involvement in preparation, review or approval of the manuscript for publication.

  • Competing interests DLB: 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: Cleveland Clinic, 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, ACC.org), 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 (Chair), VA CART Research and Publications Committee (Chair); Research Funding: Amarin, Amgen, AstraZeneca, Bristol-Myers Squibb, Eisai, Ethicon, Forest Laboratories, Ischemix, Lilly, Medtronic, Pfizer, Roche, Sanofi Aventis, The Medicines Company; Royalties: Elsevier (Editor, Cardiovascular Intervention: A Companion to Braunwald's Heart Disease); Site Co-Investigator: Biotronik, Boston Scientific, St Jude Medical; Trustee: American College of Cardiology; Unfunded Research: FlowCo, PLx Pharma, Takeda. DB: This work was done while Dr. Bonds was an employee of the National Heart, Lung and Blood Institute. JSB: Personal fees from Cardiorentis (DSMB), AstraZeneca (event adjudication committee), Novartis (DSMB), ARMGO (advisory board), Celladon (DSMB), BioTRONIK (Trial Executive Committee), Boerhinger-Ingelheim (consultant), Abbott Laboratories (consultant), Sarepta (consultant), Amgen (consultant), Servier (Trial Executive Committee, consultant), GSK (DSMB), Pfizer (DSMB). LHL: Consultant to Bayer Pharma, Novartis, AstraZeneca, Vifor Pharma; Research grants from AstraZeneca, Boston Scientific; Honoraria for lectures or speaker's bureaus from Novartis, St Jude. DM: Within the past 4 years, DM has been a paid consultant to Endo, Jarvik Heart, Lilly, Merck and Pfizer. In addition he has testified on behalf of plaintiffs in litigation related to pharmaceutical products and medical devices. APM: Research grants from Novartis, Cardiorentis, Bayer; Honoraria for lectures from Servier. TS: Research grants from AstraZeneca, Daiichi-Sankyo, GSK, MSD, Novartis; Personal fees (board and consultancy) from AstraZeneca, Astellas, Bayer, BMS, Lilly, MSD, Novartis, Sanofi. WGS: Consultant to Overcome (travel expense reimbursement to attend CVCT and professional time related to preparation of this paper), European Society of Cardiology, Heart Failure Association of the European Society of Cardiology, Heart Failure Society of American, Relypsa, Stealth Peptides, Celyad and Respicardia. AZ: Employee of GSK. NZ: Employee of AstraZeneca.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.