In this chapter, we review six case studies of adaptive clinical trials. This chapter is intended to complement the previous chapters. We have organised the case studies by their main adaptive design feature. Each case study describes the background of individual trial and summarises key design details and overall results. Our presentation and discussion do not necessarily represent the official view of the trial investigators. Our discussion is intended to improve the reader’s general literacy and comprehension of published results of adaptive clinical trials.

In this chapter, we review the history of clinical trials. We review the famous 1948 streptomycin trial for tuberculosis to highlight how the discipline of randomised clinical trials was born out of economic hardship. Other historical events of adaptive trial designs and master protocols are discussed in this chapter.

In this chpater, we review the standards and guidelines for adaptive trial designs and master protocols. Methodological rigour and transparent reporting are required in all clinical trials. Adaptive trial designs and master protocols are no exception. The standard reporting guidelines and risk-of-bias assessments for conventional trials can be applied to adaptive trial designs and master protocols. Adaptive trial designs have pre-planned adaptations and analysis plans that are specified in a formal way that outline how the data will be used to guide the design. Planning of adaptive clinical trials, basket trials, umbrella trials, and platform trials will require early engagement with the stakeholders and methodologists to think through potential hurdles and challenges of the statistical design and its implementation.

In this chapter, we review practical considerations for adaptive trial designs and master protocols. Planning adaptive trial designs and master protocols require resources and time. It is best to plan ahead with key stakeholders with statistical, content, and operational expertise to make the trial possible. Customised education and training plans will likely be required for the vendors, investigators, and other personnel involved in the trial. Critical thinking is needed from the personnel involved to create flexible technology systems and procedures required to execute these clinical trials. During the conduct, it is important to document what happened, maintain a proper firewall, and manage external communications effectively. For long-term platform trials, study adjustments may be unavoidable, but it is important that these adjustments are made before the patients are enrolled into the new study arm.

This chapter discusses the property and principles of adaptive trial designs. Adaptive trial designs refer to trial designs that offer pre-planned opportunities to modify the design of an ongoing trial based on accumulating trial data. Decisions for potential adaptations are made during the trial based on interim data, but flexibilities in adaptive trial designs are established and outlined in the study documents before any patient is recruited. For statistical planning, a simulation-guided approach is often used to evaluate the statistical properties of the design. It is generally required to demonstrate control of false positive rates for the regulatory, ethics, and funding bodies. Measures to mitigate and plan for operational bias and complexity in adaptive trial designs are needed.

This chapter introduces clinical research concepts and randomised clinical trials, covering the basics and building blocks that are necessary to understand the topics of adaptive trial designs and master protocols. Clinical trials are a type of prospective experimental studies in which human volunteers receive specific interventions according to the research protocol, then are followed longitudinally over time. Clinical trials are typically conducted in a sequence (from phase I, phase IIA, phase IIB, and phase III) that builds on knowledge accumulated from non-clinical and previous clinical studies. Randomisation is a process of random assignment of clinical trial participants to one or more intervention group(s) or control group under comparison. The use of randomisation provides a sound basis for making statistical causal inference when estimating the comparative treatment effects between groups. Fixed sample trial design refers to a type of designs where the trial data is only analysed once when a priori determined sample size has been reached. Fixed sample trial designs are designed with a fixed maximum sample size, a fixed number of interventions, and a defined end to the trial. This is the most common approach to clinical trial research.

This chapter discusses the common types of adaptive trial designs and their motivations and challenges. The common types covered in this chapter include: sequential designs, sample size re-assessment, adaptive randomisation, minimisation, response adaptive randomisation, enrichment design, and seamless design. Sequential designs that use interim analyses to allow for early stopping are the most common type of adaptive trial design. Sequential designs are often used with other types of adaptive trial designs. Other common types of adaptive trial designs include sample size re-assessment that uses blinded or unblinded data for re-estimation of sample size during the trial; response adaptive randomisation that preferentially increases allocation ratio in favour of treatment arm based on interim trial data; adaptive enrichment design that allows for modification of patient eligibility criteria; and seamless design that combines two phases of clinical trial research into one trial. Adaptive trial designs offer more flexibility in comparison to conventional fixed trial designs, but this comes with complexities that usually require more thorough and thoughtful planning in the design stage. Trade-offs for different design options should be carefully considered.

In this chapter, we review ten common misconceptions of adaptive trial designs and master protocols encountered during our collective experience in teaching and working in the field of clinical trial research.