Table 1 Characteristics of included studies
From: Stakeholder perspectives on adaptive clinical trials: a scoping review
Lead author and date | Population | Study design | Main findings |
|---|---|---|---|
Benda et al. 2010 [7] | International representatives from industry, academia, and regulatory agencies | Statistical considerations and issues session conference proceedings | -Adaptive designs (ADs) should ensure the integrity of a trial and the validity of its conclusions -Adaptivity is a fundamentally important concept that can be applied to many different stages of drug discovery and development |
Chappell et al. 2017 [8] | Researchers and representatives from industry | Conference proceedings on statistical issues in clinical trials | -Grants for developing a response-adaptive randomization design are crucial -There are ethical considerations that arise with respect to response-adaptive randomization |
Coffey et al. 2012 [9] | Representatives from the National Institutes of Health (NIH), the Food and Drug Administration (FDA), the European Medicines Agency (EMA), the pharmaceutical industry, non-profit foundations, the patient advocacy community, and academia | Recommendations from a workshop | 6 recommendations: -Need for a better-defined taxonomy of the framework for ADs -Need to better quantify the statistical risks (e.g., statistical bias, potential increase in type I error rates, and risk for covariate imbalance) -Better understanding of the concept of “adaptive by design” -NIH should offer more recognition and funding for planning clinical trials that might benefit from ADs -Use of ADs may require a different way of thinking about the structure and conduct of data safety monitoring branches -Need for education of the clinical trials community regarding the use of ADs |
DellaCioppa 2013 [10] | Senior executives from more than 20 companies, including 11 of the top 15 pharmaceutical companies | Executive roundtable on AD trials | -AD needs to be even more widely adopted across industry -Simple ADs should be considered part of Good Clinical Practice |
Dibao-Dina et al. 2018 [11] | International Review Boards members | Web-based survey via Sphinx Survey, clinical vignettes | -The most important ethical justifications for unequal randomization are gaining experience in treatment and reducing drop-outs -Different definitions of equipoise exist which may have caused the discrepancy when determining ethics of adaptive clinical trials (ACTs) |
Dimairo et al. 2015a [12] | Key stakeholders in clinical trials research (Clinical Trials Unit [CTU] directors, funding board and panel members, statisticians, regulators, chief investigators, data monitoring committee members, and health economists) | Semi-structured, in-depth interviews of key stakeholders | 5 major barriers identified: -Lack of practical knowledge and applied training with insufficient case studies -Time constraints for planning -Lack of awareness of AD opportunities or acceptable scope -Statistical and operational complexities -Conservatism regarding use of ADs Recommendations: -Accessible publications of successful and unsuccessful trials -Toolkit for ACT design |
Dimairo et al. 2015b [13] | Key research stakeholders—predominantly in the UK, CTUs, public funders, and private sector organizations | Cross-sectional, online parallel surveys | -Top-ranked barriers toward using ACTs: lack of bridge funding to support design work of complex AD, lack of practical implementation knowledge and hands-on experience, researcher providing inadequate rationale for use of ACTs -Need exists for a guidance document or troubleshooting kit regarding ACT design and implementation |
Elsäßer et al. 2014 [14] | EMA | Text search of scientific advice letters regarding ADs | Questions that are generally addressed by assessors when evaluating adaptive clinical trial proposals: -Is there a good rationale? Have alternative, more standard trial designs been considered? -Does the proposal fit well in the context of the development program? -Can the proposal be implemented without important damage to trial integrity? -Is the type I error rate controlled? -Has the potential bias of treatment effect estimates been evaluated? -Is the proposal practical and feasible? |
Food and Drug Administration 2018 [15] | Food and Drug Administration (FDA or Agency) on ADs | The draft guidance to represent the current thinking of the FDA | Advantages of ADs include: -Statistical efficiency -Ethical considerations -Advantages in generalizability and improved understanding of drug effect -Acceptability to stakeholders Barriers include: -Complex analytical methods required -Gains in efficiency in some respects may be offset by losses in other respects -Efficiency gains may be limited in certain clinical settings -An adaptive change to a trial design may lead to results after the adaptation that are not similar to those before the adaptation |
Guetterman et al. 2015 [16] | Statisticians, clinician researchers, and representatives from the FDA and NIH who took part in the ADAPT-IT program | Qualitative study via telephone using semi-structured interview protocol | -Participants thought ADs could increase efficiency of trials and decrease costs -Indicates need to educate the broader research community and more training resources to continue learning |
Hartford et al. 2018 [17] | Pharmaceutical companies, academic institutions, and contract research organizations | Survey | -Stopping early for safety and changing the endpoint of the analyses were rarely mentioned in literature prior to 2012 but are now appearing more frequently -The barriers of change management and negative experiences by some institutions with ADs remain a source of concern -Consistent training would be helpful to choose the right adaptation(s) -Perceived barrier of regulatory acceptance also remains a concern, which could be alleviated by an update of the FDA draft guidance to industry on ADs |
Jaki 2013 [18] | UK CTUs | Survey | Barriers to using ADs: -Lack of expertise -Lack of time -Lack of software -Lack of funding structure -Investigator insistence on traditional designs |
Legocki et al. 2015 [19] | Clinical trial experts: academic biostatisticians, consultant biostatisticians, academic clinicians, and other stakeholders including patient advocacy, NIH, and FDA representatives | Self-administered ACT beliefs survey, open-ended questions, mini-focus groups | -Benefits of ACTs include higher probability of receiving an effective intervention, optimizing resource utilization, and accelerating treatment discovery -Disadvantages of ACTs include possible bias and lack of informed consent |
Lin et al. 2016 [20] | Center for Biologics Evaluation and Research | Retrospective survey within Center for Biologics Evaluation and Research | The following questions should be asked before utilizing ADs: -Why use ADs? -What features does proposed design have and are they clear in protocol? -Is type I error controlled? -Are the simulation studies adequate? -Is operational bias minimized? -Are success criteria and stopping rules specified? |
Meurer et al. 2016 [21] | Clinicians, preclinical scientists, and biostatisticians who were planning clinical trials and part of the ADAPT-IT program | Visual analog scale (VAS) items through paper survey/web-based survey, mini-focus groups | -There is a need for greater community education regarding ACTs as many think clinicians don’t understand it -ACTs can help reduce sample size needed for studies but may limit assessment of secondary outcomes |
Morgan et al. 2014 [22] | 92 organizations worldwide inclusive of industry (pharma, biotech, contract research organizations [CROs]) and academia | Web-based survey | Barriers to ADs include: -Lack of education about design leads to uncertainties about methodology -Time management and planning for simulations -Risk of regulatory acceptance -Academia has a lack of resources for modeling |
Quinlan et al. 2010 [23] | 13 large- and medium-sized pharmaceutical companies and 3 statistical consultancies | Survey | 5 main barriers to ADs: -Requirement of additional planning time -Willingness of the project team to engage in the additional activities and simulations -Availability of statistical and clinical expertise and software tools -Impact of adaptive approaches to functional lines supporting clinical development -Insufficient top-down financial and motivational support from the R&D organizations |
Wang 2010 [24] | A panel with international representatives from industry, academia, and regulatory agencies | Conference proceedings titled “On “Perspectives on the Use of Adaptive Designs in Clinical Trials” | Principal statistical issues of an AD include: -The probability of correct selection in lieu of type I error -Type II error in exploratory adaptive trials -Committing a type I error in confirmatory adaptive trials -Bias in treatment effect estimates and its related issues with the interpretability of a positive study result due to adaptation |