The safety of early phase clinical trials: Bridging the gap between governance and practice

Clinical trials are governed by an extensive set of laws, regulations, guidance and ethical frameworks aimed at safeguarding human health and managing the risks of medical research. These documents cover every phase of clinical research and together provide an extremely comprehensive set of statements about the requirements for the safe non-clinical and clinical development of new medicines. However, as every safety-critical industry has learned, developing and disseminating regulations and guidelines, however comprehensive, does not necessarily translate into safe practice on the ground.

Current frameworks and regulations provide an excellent foundation for the safe conduct of clinical trials. The Helsinki Declaration¹ is widely accepted as an underlying ethical framework. The International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use (ICH) brings together global regulatory authorities and pharmaceutical industry to discuss scientific and technical aspects of pharmaceuticals and develop ICH guidelines that are grouped under four topics: quality, safety, efficacy and multidisciplinary.²

In Europe, the European Medicines Agency provides regional comprehensive guidance on the implementation of European clinical trials regulation relating to safe development of medicines and the conduct of trials.^{3, 4} Locally, in the United Kingdom, the Medicines and Healthcare products Regulatory Agency (MHRA) offer guidance on the conduct of trials in practice and have powers of regulation and inspection of clinical trials. MHRA guidelines and requirements are comprehensive, and the Phase 1 accreditation scheme provides a valuable best practice forum for accredited early phase clinical trials units and their regulator.^{5, 6}

Guidance and regulations are however heavily weighted towards the manufacture of drugs, non-clinical research and later phase trials. There is limited consideration of early phase clinical research, which is detrimental to drug development as early phase research is key to efficient and fast development of new treatments. Early phase trials are of particular relevance as they are now highly varied due to the complexity of the medicinal products being tested, complexity and comorbidities of patients in the trial and the consequent variability in trial design and monitoring and compliance regimes. This variability means that it is all but impossible for regulators to develop and maintain guidelines that can encompass all potential innovative trial formats. In addition, any guidelines that are developed inevitably lag behind the evolution of trial design and practice.

The most important challenge facing regulators and those developing guidelines however is that, for a variety of reasons, rules are never sufficient to ensure safety in practice. Clinical trial guidelines are generally written by people who understand the principles but have no current practical expertise designing or performing clinical trials. Guidance on risk management and quality control set out the broad features of risk management systems. Implementation within this framework varies significantly, depending on local regulation and site facilities and expertise. Compliance monitoring is often conducted using rigid historical interpretation of guidelines, whereas in practice investigators often need to adapt to the local environment and to the evolving understanding gained from the trial itself.⁷

Reviews of the conduct of clinical trials have identified multiple safety issues.^{8, 9} These include inadequate procedures and documentation of dose escalation, poor and narrowly focussed training for emergencies, inadequate procedures for contacting doctors out of hours, incorrect dosing of participants, due to a lack of adequate procedures and resources, and lack of robust procedures for the scheduling to ensure all protocol assessments were performed and that staff allocated were suitably trained and capable.

The difficulty of translating comprehensive regulation and guidelines into practice is exemplified in a recent study examining work practices in a clinical research facility during a SARS CoV-2 controlled human infection model.¹⁰ Whilst controlled human infection model studies represent a specific and unusual subtype of experimental studies, the underlying principles of study set-up and conduct are transferrable to other experimental and early phase clinical trials. Human factors researchers, skilled in the analysis of work systems, reviewed documents, conducted interviews, observed the work environment, walked through the trial processes and made detailed, structured observations of three safety critical tasks. They identified a number of risks to patients and staff, including points where guidance was imperfectly implemented, areas where voluminous guidance was simply impossible to follow and a variety of challenges in the design of the facility that potentially compromised safety. The study is of particular interest because the authors stress that those running the trial were highly professional, collaborative, were careful with procedures and had an excellent safety culture. The problems observed were inherent in the processes and systems being used. For instance, PPE requirements were different in the trial setting and in the linked hospital; instructions to not touch surfaces during transfer were impossible to follow because of the design of the facility; written procedures were long and cumbersome, not at all suitable as practical work instructions; participants in the trial could not easily communicate with those running the trial. Researchers and trial leaders jointly developed 14 recommendations for enhancing safety in future trials, all concerned with improving the work practices and trial environment.⁸

This study illustrates how ‘work as imagined’ can differ from ‘work as done’.¹¹ The trial was governed by thoughtful and well-intentioned standards guidelines. Those leading the trial followed these guidelines to the best of their ability. In spite of this, a number of risks to patients and staff were identified. The principal lesson is that standards and guidelines, however careful and comprehensive, are necessary but not sufficient to maintain trial safety even when all those involved are acting in a highly professional manner.^{10, 12} These problems are likely to be compounded by the increasing heterogeneity of clinical trials in which variation from standard and historical practice becomes increasingly common, meaning that any local or global regulation and guidelines will inevitably lag behind innovations in practice.

We need therefore to pay much more attention to how safety is achieved in practice, across multiple trial types and settings. Standardization and compliance should be combined with agility to adapt to change and innovation. The overarching principle should be simplification, rather than adding more documents and checklists, to achieve lean processes that can be easily understood and followed. We suggest the following:

• Investigators and clinical researchers should be well represented in the development of relevant regulation and their supplementary guidelines. Engaging experienced investigators/researchers and respecting their expertise can help making regulation and guidance more relevant and also provides a chance to nurture the next generation of professional, experienced, clinical trials investigators and researchers.
• Regulators need to recognize that any guidance will need to be adapted by clinical trial professionals into practical local procedures that meet quality standards. Regulation and regulatory guidance themselves should also be modular so that they can be updated more frequently to counteract the current lag to clinical trial practice.
• Many trials now have unique features making it difficult to define standard procedures across multiple trials and sometimes even across multiple centres of one trial. Standard operating procedures are still necessary but need to be developed for each particular trial and potentially each site. Standardization is achieved at any point in time via current modular work instructions and process maps. Continual improvement formalizes the necessary incremental changes through documented customization and optimization.
• Compliance monitoring should be intelligent and focus on overarching concepts and frameworks, whilst allowing flexibility of local practical implementation.⁷
• Clinical trial specialists need a greater understanding of human factors and ergonomics approaches that are used in other industries to understand work systems and develop effective safety management systems. The approaches used by Higham and colleagues in their study or an early phase trial provide a valuable starting point (10).
• Continual and adaptive training and upskilling of investigators and researchers, combined with collaboration with experts in other specialties, are fundamental in delivering early phase trials.
• In line with modular and adaptable work instructions and process maps, learning and development programmes should be made available that are on-demand and have flexible curricula tailored to the scope of work of individual early phase investigators and researchers.
• Providing investigators, researchers and sites with certifications, such as ‘trusted’ badges could encourage participation in such learning programmes, which would be further enhanced by offering other appropriate incentives, such as faster regulatory authorization timelines for trusted investigators and sites, for example, via notification schemes.

Clinical trials are the bedrock of medical research and are trusted because of their rigorous design and because they conform to strict guidelines in practice. We are emphatically not suggesting any dilution of standards or that those designing or conducting trials have carte blanche to conduct trials in whatever way they think best. Early phase trials are however becoming increasingly heterogeneous, and this requires a different approach to standards and guidelines akin to the ‘mass customisation’ seen in industry.¹³ Mass customization allows products to be made to individual customer requirements but is only achievable because each of the many components is built in a standard fashion with rigorous quality control.

We therefore need a toolbox of standards, guidelines and local work instructions, which can be customized to the requirements of any particular trial and patient group. Each trial will still have clear and defined standards and safety procedures, but the precise nature of these standards and procedures may vary between trials. To achieve this objective, we will need to train a new generation of clinical trial researchers. They will need the same basic knowledge of pharmacology, trial methodology, national and local regulations and guidelines. They will also however need to have a solid understanding of human factors, ergonomics and the essentials of safety science so that they can both specify the standards needed in each trial and have the knowledge and skills to manage risk and protect patients in the rapidly evolving landscape of early phase trials.