Research ethics for emerging trial designs: does equipoise need to adapt?
BMJ 2018; 360 doi: https://doi.org/10.1136/bmj.k226 (Published 25 January 2018) Cite this as: BMJ 2018;360:k226
All rapid responses
We agree with Hey et al. (1) that emergent clinical trial designs may present new variations on the foundational questions of clinical trial ethics. However, we doubt their assumption that the modern response needs to be based on the concept of equipoise. For, as Hey et al. identify, we still do not have a standardized way of determining when equipoise is fulfilled. As they write, "systematic evidence reviews, formal surveys, and peer critique of a trial's protocol" may all point towards equipoise, but these are not equivalent things.
If equipoise just means uncertainty then yes - on a very superficial level - it should factor into the ethical evaluation of both traditional and novel RCTs. But clearly this is too low a bar to meet: there is always uncertainty, at least with regards to how any one patient will respond to a given treatment (2). And if equipoise specifically means disagreement within a community, then a series of as yet unanswered questions need to be resolved: whose uncertainty, how much, established how, and upon what basis? Without answers to these questions, it becomes too easy to use equipoise to justify almost any design: because of how flexible and multi-varied scientific uncertainty can be, clinical equipoise is unable to objectively discriminate between permissible and impermissible clinical research.
Therefore, we have suggested a series of questions that may more thoroughly ensure that a given trial is ethically sound (3,4).
1. Does the trial have a clear question and a precise hypothesis?
2. Is the trial non-redundant - ie. has the question not already been answered?
3. Is that non-redundancy established through a systematic review of the literature?
4. Are all of the treatment strategies being compared scientifically plausible and reasonable in clinical practice - ie. are patients enrolled in the trial not deprived of accepted treatments?
5. Does the trial have the sufficient methodological rigour (sample size, recruitment strategy, formalized outcome measures) to be likely to produce a meaningful result?
We hope to bring these questions to the broader research community with the intention of ensuring that any RCTs seeking to enroll patients - be they of traditional or emergent design - are scientifically necessary and ethically justified.
(1) Hey SP, Weijer C, Taljaard M, Kesselheim AS. Research ethics for emerging trial designs: does equipoise need to adapt? BMJ 2018;360:k226.
(2) Shamy M, Stahnisch FW, Hill MD. Fallibility: a new perspective on the ethics of clinical trial enrollment. Int J Stroke 2015;10:2-6.
(3) Shamy M, Fedyk M. Why the ethical justification of randomized clinical trials is a scientific question. J Clin Epi 2018; https://doi.org/10.1016/j.clinepi.2017.12.026
(4) De Meulemeester J, Fedyk M, Lurkovic L, et al. Many randomized clinical trials may not be justified: a cross-sectional analysis of the ethics and science of randomized clinical trials. J Clin Epi 2018; https://doi.org/10.1016/j.clin.epi.2017.12.027
Competing interests: No competing interests
Existing guidelines for new design: tools for ethics committee
We read with great interest the paper by Spencer Phillips Hey et coll. focusing on equipoise in emerging trial design. The authors consider that basket and umbrella trials, cluster trials and trials with adaptive design raise ethical challenges; consequently they propose specific guidelines to face these challenges.
Umbrella trial and basket trial main problems are: 1) the accuracy of the diagnostic tests used to compose the groups that are randomized or enrolled; 2) the informed consent that must include explanation about the risk of misclassification for the patient if the assay is not adequately clinically validated.
The problems on informed consent are even more accentuated in adaptative platforms trials because of the dynamic nature of the design; the patients enrolled should be correctly informed about the possible evolutions of the study that change depending on the allocation arm.
The trial clusters, finally, pose problems of competence (since clinicians may not be experts adequate to the challenges posed by these designs) and identification of the recipients of informed consent.
The authors rightly conclude about the need to develop guidelines for the correct evaluation of these emerging study designs and to reconsider the concept of equipoise in the light of such innovations.
We believe that ethical committees need more adequate training and provision of standardized tools for the evaluation of research protocols. We also believe that many of these tools already exist but unfortunately they are not adequately used by ethics committees in the evaluation of research protocols.
The Pediatric Ethics Committee (CEP) of the Tuscany Region, established in 2013, has adopted standardized assessment tools for interventional (SPIRIT 2013 Statement)1, observational (STROBE)2 and diagnostic (QUADAS-2)3 studies.
The diagnostic accuracy problems to be considered for the assessment of the appropriateness of the choice of biomarkers, invoked by Spencer Phillips Hey et al, could be addressed with tools able to evaluate the diagnostic studies supporting the choice of the biomarker. These tools are already available and used since several years by the CEP. On the other hand, guidelines to evaluate the appropriateness of protocols that develop predictive models (TRIPOD)4 are available and must be used by ethics committees in their evaluation process. The same applies for bio-marker evaluation guidelines5-7.
The trial clusters require assessments of appropriateness of the design, partly referred to in the Consort-cluster document (CONSORT Cluster Trials)8. This document identifies the relevant items of a cluster trial, which represent key points of ethical questions to be considered beyond the consensus issues. Adaptive trials design may allows early stopping during the interim analysis (i.e. sequential models 9), thus requiring, as in the case of enrichment procedures, an in-depth analysis of the equipoise that can benefit from existing guidelines.10
Finally, the SPIRIT Statement1 is another available tool, providing indications for the quality assessment of the research protocols (and related ethical implications) that the CEP is using. This document has been recently improved with the inclusion of relevant and patient-centered outcomes.11
Overall our experience testifies the availability of adequate tools that are only waiting to be used by ethics committees and trialists.
References
1. An-Wen Chan et al. SPIRIT 2013 Statement: Defining Standard Protocol Items for Clinical Trials. Ann Intern Med. 2013;158:200-207.
2. Vandenbroucke JP, von Elm E, Altman DG, Gøtzsche PC, Mulrow CD, et al. (2007) Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): Explanation and Elaboration. PLoS Med 4(10):e297. doi:10.1371/journal.pmed.0040297
3. Whiting P, et al. QUADAS-2: A Revised Tool for the Quality Assessment of Diagnostic Accuracy Studies. Ann Intern Med. 2011;155:529-536.
4. Collins GS, Reitsma JB, Altman DG, Moons KG. Transparent reporting of a multivariable prediction model for individual prognosis or diagnosis (TRIPOD): The TRIPOD statement. Ann Intern Med. 2015 Jan 6;162(1):55-63. doi: 10.7326/M14-0697.
5. Concept paper on predictive biomarker-based assay development in the context of drug development and lifecycle. EMA/CHMP/800914/2016
6. Reflection paper on methodological issues associated with pharmacogenomic biomarkers in relation to clinical development and patient selection. EMA/CHMP/446337/2011
7. Concept paper on predictive biomarker-based assay 5 development in the context of drug development and 6 lifecycle. EMA/CHMP/800914/2016
8. Campbell MK, Piaggio G, Elbourne DR, Altman DG; for the CONSORT Group. Consort 2010 statement: extension to cluster randomised trials. BMJ. 2012 Sep 4;345:e5661. PMID: 22951546
9. Pocock SJ. Interim analyses for randomized clinical trials: the group sequential approach. Biometrics. 1982 Mar;38(1):153-62.
10. http://www.fda.gov/Drugs/GuidanceComplianceRegulatoryInformation/Guidanc...
11. Melanie Calvert, PhD1; Derek Kyte, PhD1; Rebecca Mercieca-Bebber, PhD2; et al. Guidelines for Inclusion of Patient-Reported Outcomes in Clinical Trial Protocols. The SPIRIT-PRO Extension. JAMA. 2018;319(5):483-494. doi:10.1001/jama.2017.21903
Competing interests: No competing interests