In the recent times, the usage of adaptive trial design in clinical trials based has become much popular because of its flexibility as well as efficiency. Adaptive designs in clinical trials are carrying the potential to accelerate the studies when at the same time cutting costs to a larger extent. (1), (2) So, what is an adaptive clinical trial? Where this design can be implemented? And, when it can be implemented? The following article will answer all these questions for you ...
Defining an Adaptive Clinical Trial:
In simple terms, an adaptive clinical trial can be defined as a clinical trial that is designed initially for evaluating the trial participants' reactions to an investigational drug and which can be modified at the later stages in accordance with the findings. As such, in an adaptive clinical trial, the following parameters can be altered during the course of the trial in relation with the findings:
- Study eligibility criteria
- Total sample size of the study
- Treatment regimens of various study groups
- Cocktail mix
- Randomization procedure
- Concomitant treatments that are being used
- Planned schedules of patient evaluation for data collection
- Primary endpoints
- Selection and order of secondary endpoints
- Analytical methods that are being used to evaluate the endpoints
- Treatment duration
- Diagnostic procedures
- Criteria for evaluation
- Assessment of clinical responses
And, the undertaken adaptations will be continued throughout the clinical trial. (1), (2), (3), (4)
What could be the Aim of Adaptive Clinical Trials?
The primary aim of any adaptive clinical trial will be to bring technological advances to the patients in a much effective way. Another aim of the adaptive clinical trials is to identify the drugs that have a therapeutic effect more quickly. And yes, one another purpose is to make the clinical trials more flexible. (1), (2), (3)
Type of Adaptive Trial Designs:
Based on the type of adaptation employed, the adaptive trials can be classified as follows:
- Adaptive Randomization Design: This is a study design which would allow for a modification of randomization schedules depending upon the varied probabilities of treatment allotment. The primary objective of this adaptive trial design is to increase the probability of success of the particular clinical trial. The adaptive randomization procedures include the following:
- Treatment adaptive randomization
- Covariate adaptive randomization
- Response adaptive randomization
- Group Sequential Design: This is a study design that would allow for premature termination of the trial due to the concerns of safety/efficacy or both. In this trial design, there will be options of additional adaptations depending upon the results that are obtained from interim analysis.
- Sample Size Re-estimation Design: This is a type of adaptive study design which would allow for adjustments in sample size or re-estimation depending upon the data that are obtained from the interim analysis. As such, the adaptations can be done in a blinding/non-blinding fashion depending upon the following criteria:
- Treatment effect size
- Conditional power
- Reproducibility probability
- Drop-the-losers Design: This is a study design which would allow for dropping off the inferior treatment arms. This design would allow for the addition of extra treatment arms. This design will be particularly useful in phase 2 clinical studies where the uncertainties in dosage levels arise. In this study design, the selection criteria along with the decision rules play a major role.
- Adaptive Dose Finding Design: This is a type of adaptive study design which is often employed in early phase of clinical development for the purpose of identifying the minimum effective dose and the maximum tolerable dose that are being used to determine the dosage levels for the next phase of clinical trials.
- Adaptive Treatment-switching Design: This is a study design which would allow the trial investigator to shift a patient's treatment from the initial assignment to another treatment when an evidence of lack of efficacy/safety of the initial treatment has been observed.
- Adaptive Hypotheses Design: This adaptive trial design would allow for modifications in hypotheses in accordance with the interim results. Most often, this type of adaptive trial design will be considered prior to database lock or data unblinding.
- Biomarker Adaptive Design: This type of adaptive study design would allow for the adaptations in a clinical trial depending upon the response of biomarkers like the genomic markers. Any biomarker adaptive study design would involve the following:
- Biomarker qualification and standards
- Optimal screening design
- Model selection as well as validation
A biomarker adaptive clinical trial design can be used for the following purposes:
- Selecting the correct patient population
- Identifying the natural course of the particular disease
- Early detective of the disease
- Helping to develop personalized medicines
- Adaptive seamless Phase II/III Design: This is an adaptive trial design which would use the data from the patients who are enrolled before/after the adaptations in the final analysis. Basically, this is a two-stage trial design that consists of: learning and confirmatory stages.
- Multiple Adaptive Design: This is a combination of any of the above adaptive clinical trial designs. The most common combinations are:
- Adaptive group sequential design + Drop-the-losers design + Adaptive seamless trial design
- Adaptive dose escalation design + Adaptive randomization (1), (5)
Questions to be answered while Implementing Adaptive Design:
The following questions should be answered while implementing adaptive trial design:
- Will the trial design need any extraordinary effort for implementation?
- Are the levels of difficulty as well as the cost justifiable?
- Will the adaptive design implementation delay patient recruitment?
- Will the delayed response diminish the benefits of the adaptive design?
- How often the unblinded analyses are practical?
- To whom should the data be unblinded?
- How the impacts of a Data Monitoring Committee's decisions should be considered?
- Will the unblinding cause biases in assessing treatments?
- Will the adaptive design destroy the randomness?
- Will the protocol deviation and violations invalidate the adaptive design?
- How will the unexpected DMC action affect the power as well as the validity of the adaptive design? (1)