What are Adaptive Trials in Clinical Research?

Introduction

Clinical trials are the cornerstone of medical advancements, providing the evidence needed to bring new and best treatments to patients. Traditionally, these trials have followed a rigid, sequential process, often taking years to complete and requiring significant resources. As medical research and technology evolve, so too must the methods we use to evaluate treatments.

Enter adaptive trials—a revolutionary approach that is transforming the landscape of clinical research. But what exactly are adaptive trials, and why are they considered smarter and faster than traditional trials? This blog will explore these questions, shedding light on the benefits and key features of adaptive trials and their potential to revolutionise clinical research.

Traditional vs. Adaptive Trials

To understand the significance of adaptive trials, it's essential first to grasp the limitations of traditional clinical trials. In a conventional trial, the design is fixed from the outset.

Researchers determine the sample size, patient population, dosing schedules, and endpoints before the trial begins, and these elements remain unchanged throughout the study. This rigidity can be a double-edged sword: while it ensures validity if the trial is positive, it also means that if the initial assumptions are incorrect, the entire trial could fail to identify a treatment that works, leading to wasted time and resources.

Nothing is more frustrating than abandoning a promising treatment and not knowing whether it failed because it's not actually effective or because the trial design was flawed.

Adaptive trials, offer a flexible and dynamic alternative. These trials are designed to evolve in response to adaptive analyses that occur as the trial progresses.  As participants are recruited the exact design of the study can change according to design rules that are specified before the trial commences.  

This allows modifications to aspects of the design, such as the number of participants, dosage levels, or even the patient population, in order to optimise the value of information generated by the trial. There are three major advantages: on average, results will be obtained sooner, the cost of the trial will be lower, and the risk of failing to identify an effective treatment will be reduced.

Key Features of Adaptive Trials

Adaptive trials are characterised by several key features that set them apart from traditional trials:

1. Pre-specified but Flexible Design: Adaptive trials are inherently flexible. The trial protocol is designed with predefined rules that allow for modifications based on results at each adaptive analysis. This flexibility ensures that the trial remains relevant and efficient, across a range of plausible assumptions about size of treatment effect and which patients will benefit from the treatment.
2. Adaptive Analyses: One of the most critical components of an adaptive trial is the adaptive analyses. These analyses are conducted as data accrues throughout the trial. Based on the results and pre-specified rules, the trial's course can be modified, such as adjusting the dosage, expanding or narrowing the patient population, or even stopping the trial if a pre-specified effect (such as superiority of an intervention) is shown in the accrued data.
3. Real-Time Data Utilisation: Because of the combination of adaptive analyses and pre-specified design rules, data is used when its most valuable, i.e., as soon as its available.  In a fixed trial design data is only used once, at the end of the trial, and if the design assumptions are wrong the only option is to start again and design a new trial.
4. Multiple Objectives: Adaptive trials can address more objectives within a single study than can occur with a fixed design. For example, a trial might simultaneously evaluate different dosages or patient population, allowing for a more comprehensive understanding of the intervention's value.
5. Never slower than a traditional design: If an adaptive design specifies a maximum sample size that is the same as the sample size specified for a traditional trial, the adaptive design will never take longer than a traditional design, and will often be faster.

Examples of Success

Adaptive trials have already proven their value in several high-profile studies. One notable example is the I-SPY 2 trial (refer to trial registration and protocol paper in study1; study2), an adaptive platform trial designed to evaluate treatments for breast cancer. By using an adaptive design, the I-SPY 2 trial has been able to efficiently test multiple therapies simultaneously (using a common control arm) and match treatments to different categories of breast cancer, rapidly identifying the most promising treatments and accelerating their development.

Another example is the REMAP-CAP trial (see study1; study2), which evaluated multiple re-purposed drugs to treat severe COVID-19.  Taking only 8 months the trial was the first to identify tocilizumab as a highly effective treatment for patients who were critically ill from COVID. It was the fastest among multiple trials because it used a design without a fixed sample size, conducting adaptive analyses, until it could be concluded that the treatment worked.

Conclusion

Adaptive trials are most useful at phase 2, when little is known about how a new treatment works and its effectiveness. It’s counterintuitive that, at this time, a traditional trial design requires all design parameters to be fixed. Adaptive designs offer a smarter, faster,and more efficient alternative to traditional trials.

By allowing for real-time adjustments based on accruing data, adaptive trials can optimise resource use, reduce trial duration, and improve patient outcomes. As the healthcare industry continues to embrace innovation, adaptive trials are poised to become an integral part of clinical development, particularly at phase 2, reducing the time from the start of phase 2 to registration by many months or years.