www.gfbr.global Ethics of alternative clinical trial designs and methods in low- and middle- income country research 28 – 29 November 2017, Bangkok

The GFBR Funded by Bill & Melinda Gates Foundation, UK Medical Research Council, US National Institutes of Health, and Wellcome Seeks to bring voices and perspectives of low- and middle-income countries (LMIC) to the fore in discussions on emerging ethical issues in research Multi-layered approach, involving: researchers, clinicians, IRB/REC members, regional policy-makers, and global thought leaders Strong emphasis on developing networks and collaborations for participants, and learning across borders

The topic Alternative clinical trial designs and methods are increasingly being used in place of the conventional randomised controlled clinical trial in LMICs. These approaches – including adaptive, cluster randomised and stepped wedge designs and controlled human infection models – offer a number of potential advantages, including accelerating vaccine or drug development and making the clinical trial process more socially acceptable. The ethical implications - on risks and potential benefits to participants, consent, scientific rigour, trial efficiency - have not been adequately addressed. Current guidance was largely written without special consideration of new trial designs, leaving researchers, research ethics committees and regulators with little support in how to evaluate, implement and run these often complex trials. This meeting aimed to bring different socio-cultural and disciplinary perspectives together and provide mutual ground for discussion and a shared understanding of the challenges and opportunities presented by these alternative approaches.

Participants - 93 participants - 35 countries - Perspectives from policy-making; public health; clinical care and research; biostatistics; bioethics; regulatory agencies

Case studies Cluster randomised trials (CRT) Stepped wedge Ethical issues of the PolyIran study: A cluster randomized trial nested within Golestan cohort study, Gholamreza Roshandel, Iran Lessons from an adaptive multi-arm multi-stage trial of strategies for improving linkage into HIV care or prevention in Malawi, Augustine Choko, Malawi/UK Stepped wedge ‘Que Vivan Las Madres´: Scaling up an integrated approach to reduce maternal and perinatal mortality in Northern Guatemala, Karla Hemming, UK and Guillermo Ambrosio, Guatemala ATMIYATA: Testing effectiveness of counselling delivered by community volunteers to people with common mental health issues in rural parts of Gujarat, Kaustubh Joag, India

Case studies Adaptive platforms The design and implementation of an adaptive platform trial for the treatment of Ebola in West Africa, Scott Berry, USA Critically ill children and adaptive trials for comparative effectiveness research, Srinivas Murthy, Canada Controlled human infection models (CHIMs) Control of invasive Salmonella in Africa and Asia – Is there a role for establishing controlled human infection models in endemic countries? Meriel Raymond, UK Experiences and perceptions of study participants in a malaria challenge study in Kilifi, Dorcas Kamuya, Kenya The case of Zika virus human challenge studies, Ricardo Palacios Gomez, Brazil

Guidance and regulation Guidance and regulation from different perspectives Overview: Current international ethical guidance and regulations on use of alternative research designs, Rieke van der Graaf, Netherlands Funder: CHIMS: What does it mean to be a responsible research funder? Katherine Littler, UK Regional perspectives: East Africa - Claude Kirimuhuzya, Uganda Caribbean - Derrick Aarons, Trinidad and Tobago Latin America - Carla Saenz, USA Southeast Asia - Cristina Torres, Philippines

Key points There are a variety of reasons to use a non-traditional design including: Scientific (validity and contamination) Practical (recruitment, consent, feasibility in implementation, costs) Ethical (beneficence) Issues and challenges include: Methodological (scientific rigour, applicability of the concept of clinical equipoise) Distinguishing research and implementation study Defining the participants (with implications for consent and post-trial provision obligations) Informed consent (reasons for waiver, complexity of information) Determining appropriate compensation (reimbursement vs incentive) Standard of care Justice (e.g. equity in delayed participation and its relationship with equipoise) Political interference (e.g. choice of first cluster v randomisation in Stepped Wedge CRT)

Key points - context Instead of talking about ‘alternative’ study designs, it’s more helpful to have a contextual understanding and ask oneself: what is the best design for the goals of this study? The context will depend on the: Nature of health or disease condition being studied (public/population health, emergency) Nature of treatment (likely effectiveness, range of therapeutic options, duration in evaluating study endpoints) Resources Preference of stakeholders (e.g. community acceptability) Behaviour of participants (e.g. sharing of medicine and contamination)

We need flexible, adaptive ethics as well as adaptive trials Key points - context Being explicit about goals also helps clarify some of the ethical and regulatory questions about the study: there are grey areas (particularly in the use of stepped wedge studies) between evaluating roll-out of a new intervention, and conducting research into how the intervention works. In such ‘grey areas’, it is helpful to separate out what may be required in terms of regulation (does the law in this jurisdiction require scrutiny by a research ethics committee (REC)?); and what is actually ethically demanded (what kind of oversight would help identify and respond to ethical concerns?). There is a need for some kind of mechanism (not necessarily a REC) that would exercise appropriate ethical oversight of studies that are primarily concerned with evaluation and quality improvement. We need flexible, adaptive ethics as well as adaptive trials

Key points - capacity Researchers, regulators and ethics committees should be supported to better understand the facets of this nimble science. Not only the scientific aspects - risk, bias, power calculations, limitations and strengths - but also to better discern the key and emerging ethical issues in such approaches. This need is particularly pressing as complex mathematical simulation are increasingly being used to assess risk and benefits within a trial design. There was a proposal for research grants to include additional provision for training/support for ethics committees so they are sufficiently knowledgeable and confident to assess novel trial designs appropriately, particularly in low income environments or for international organisations to provide training (e.g. WHO, PAHO). The need for wider public education/engagement was also recognised to enhance greater understanding of what is involved, along with empirical data on whether particular trial designs (e.g. CRTs) are more acceptable in practice to local communities/possible participants.

Key points - guidance There is limited international guidance on these trial designs and scarce guidance specific to LMICs. The Ottawa statement (2012) and CIOMS guidelines 2016 offer guidance on the ethical design and conduct of CRTs but there is a lack of guidance completely for adaptive platforms. Some participants called for international and LMIC specific ethical guidance, the uptake of which would depend on international organisations taking the lead on development and publication. The existence of such guidance should avoid ‘precautionary’ approaches being taken by many countries (i.e. rejecting any kind of novel design).

Conclusions Many of the ethical issues that arise in these study designs are common to other kinds of research in LMICs and in HICs too: e.g. achieving genuinely informed and voluntary consent; reimbursement and compensation that is fair and appropriate but does not constitute undue inducement etc. However, novel designs present issues that may complicate these ‘standard’ concerns include determining who is the study participant in a CRT; determining when a waiver of consent is appropriate and questions of public acceptability and impact on trust, particularly in challenge studies. Enhancing the capacity of researchers, research ethics committees and regulators to implement and review these designs should facilitate a more nuanced understanding. By better understanding the risks, benefits and contextual considerations, the full potential of novel designs to address the health needs in LMICs can be realised.