Author + information
- Stefan Koudstaal, MD PhD†,‡,
- Sophie L. Niemansburg, MD§,
- Nabil Dib, MD PhD‖,
- Justin Wallet¶,
- Pieter A. Doevendans, MD PhD†,‡,
- Annelien L. Bredenoord, PhD§ and
- Steven A.J. Chamuleau, MD, PhD†,‡∗ ()
- †Department of Cardiology, Division Heart and Lungs, University Medical Center Utrecht, Utrecht, the Netherlands
- ‡Interuniversity Cardiology Institute of the Netherlands (ICIN), Utrecht, the Netherlands
- §Department of Medical Humanities, Julius Center for Health Sciences and Primary Care, University Medical Center Utrecht, Utrecht, the Netherlands
- ‖Clinical Cardiovascular Cell Therapy, University of California, San Diego, San Diego, California
- ¶Faculty of Medicine, University Medical Center Utrecht, Utrecht, the Netherlands
- ↵∗Department of Cardiology, Division Heart and Lungs, University Medical Center Utrecht, Room E03.511, P.O. Box 85500, 3508 GA Utrecht, the Netherlands
To the Editor:
Cell-based strategies are under intense investigation in the pursuit to develop new effective treatment protocols for ischemic heart disease (IHD). These strategies have been mainly based on the use of tissue-specific autologous stem/progenitor cells such as cells from bone marrow, adipose tissue, or the heart itself (1). Several of these cell types have reached the phase of clinical testing (2,3). Clinical trials, in particular trials investigating cell-based interventions, entail inherent scientific and ethical challenges. Due to the combination of the relative lack of experience with cell-based interventions, the complexity, the variability (especially when autologous cells are used), and the invasive character, traditional ethical issues get a new perspective. One of these issues is the choice for the comparator (4).
Treatment effect is generally assessed in clinical trials by means of superiority of the novel intervention over standard clinical care or placebo. Placebo is used to conceal intervention allocation both for the patient and investigator-physician (5). In pharmaceutical clinical trials, the placebo is often a capsule or tablet indistinguishable from the investigational new drug. However, in clinical trials assessing autologous cell interventions for heart disease, proper blinding can only be achieved when participants allocated to the placebo group undergo the identical harvesting procedure (e.g., bone marrow aspiration in the case of bone marrow mononuclear cells) as well as a sham delivery procedure to the heart undistinguishable from the cell delivery procedure. If the interventional cardiologist is not involved in the follow-up and/or outcome analysis, the sham procedure may not necessitate an actual injection with a placebo solution identical to the cell suspension. Either way, in contrast to placebo tablets, autologous cell interventions expose the control group to risk and harm, raising ethical concerns.
To date, little is known about the extent to which sham interventions in cell-based trials are performed. To address this question, we conducted a systematic review on published reports of randomized clinical trials (RCTs) investigating efficacy of autologous cell interventions in patients with IHD. We assessed how RCTs were designed regarding cell harvesting and/or cardiac sham delivery, whether this depended on the type and stage of cardiac disease, and the adverse events rate in sham procedure patients. A search syntax was developed based on relevant synonyms for domain, which is patients with IHD, and determinant, which is cell intervention delivered to the heart (i.e., intracoronary delivery or intramyocardial injection). The outcome—efficacy of cell intervention—was deliberately withheld from our search syntax to avoid potential reporting and retrieval bias. A systematic literature search was conducted in MEDLINE and EMBASE on the 13th of May 2013. Two reviewers (S.K., J.W.) independently screened title and abstract of studies in accordance of in-/exclusion criteria (see the Online Appendix). The selected articles were crosschecked to identify relevant studies missed by the initial search using ISI Web of Science.
A total of 56 RCTs were identified that were published between 2001 and 2013. In total 3,610 patients were included in these studies, of which 2,189 (61%) received autologous cells, compared to 1,421 (39%) controls. In 75% of studies, bone marrow mononuclear cells were the investigational cell type. Diagnosis was acute myocardial infarction (MI) in 2,463 patients (982 controls [40%]), compared to 1,147 patients (439 controls [38%]) with chronic IHD (i.e., refractory angina or post-MI heart failure).
Combined cell harvesting and sham cardiac delivery, thereby ensuring patient and investigator blinding, was performed in 22 of the 56 studies (39%). Analysis divided for IHD type revealed that 11 of 36 studies (31%) used sham delivery in acute MI and 11 of 20 studies (55%) in chronic IHD (chi-square; p = 0.09). Apparently, use of placebo is generally not preferred in the acute setting of MI in contrast with an elective cell intervention procedure for chronic IHD. Furthermore, in 1 study, participants allocated to the control group underwent cell harvesting but no sham cardiac injections, while the rest of the studies used usual care as a comparator.
Out of 22 studies that used placebo, adverse events in controls were reported in 5 studies (23%), all of which investigating bone marrow cells. A combined endpoint analysis of major cardiovascular events revealed that intracoronary infusion of a placebo solution did not lead to heightened mortality or serious morbidity in these 5 studies. With regard to the cell harvesting procedure no adverse events were reported.
Three major conclusions emanate from our study: 1) 39% of RCTs investigating efficacy of autologous cell intervention used a double-blind trial design based on a sham procedure; 2) trials using sham delivery were less frequently observed in acute MI compared to chronic IHD; and 3) 23% of these trials reported data on (minor) adverse events. This report shows that the choice for the control group for cell-based interventions in cardiology differs among research groups, which could be related to diverging views on scientific necessity and ethical acceptability of sham. The diversity in the choice of the comparator group, shown by this empirical report, supports the need to clarify when and under what conditions sham is scientifically necessary and ethically acceptable, and when another comparator is more appropriate in clinical trials investigating cell-based interventions for cardiology and other medical fields. This will, among others, probably correlate with the study population, and the risk profile of the sham procedure. Hence, this report can function as a starting point to formulate guidelines for researchers that aim to set up a cell-based randomized trial.
Please note: This work was supported by the HGG Group BV to Dr. Koudstaal and the “Wijnand M. Pon Stichting” to Drs. Koudstaal and Chamuleau. Dr. Dib has served as a consultant for Cephalon/TEVA, Cardiovascular Systems, Biologics Delivery Systems/Biosense Webster/Johnson & Johnson, Stemedica, and the University of California. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- American College of Cardiology Foundation
- Dimmeler S.,
- Burchfield J.,
- Zeiher A.M.
- Jeevanantham V.,
- Butler M.,
- Saad A.,
- Abdel-Latif A.,
- Zuba-Surma E.K.,
- Dawn B.