1. What are adaptive clinical trials? According to the FDA guideline, adaptive clinical study is a study that allows for opportunity of change and modification in the hypothesis and study design using data obtained from clinical trial participants.
The collection and analysis of data are completed at specific time points throughout the study. The subjects and investigators can either be blinded or unblinded and the analysis of data can be performed with or without testing the statistical hypothesis.
There is an important difference between changes using unblinded results via an interim analysis of the controlled study and changes established on blinded results through a non-comparative interim analysis. Interim analysis is the testing of the data in a trial while the study is still occurring.
The data collected during this examination can include PD data, biomarkers, and efficacy and safety outcome data. There should be frequent revisions of the study protocol in case of an unblinded analysis to avoid any misinterpretation of results.
On the other hand revision of study protocol based on a blinded iterim analysis are not required frequently as they do not present a statistical bias. Some changes such as sample size do not have to be planned for, prior to the study as those are based on the collection of information throughout the trial.
Concerns are raised in a study if the revisions were not planned before the study or suggested after the unblinded analysis. (FDA Guideline) 2) Types of adaptive clinical trial: ? Adaptive randomization trials: The main goal from these trials is to maximize the chance of success. Researchers are allowed to modify randomization taking into consideration the inequalities in probability of treatment.
The use of this model is not allowed in longer trials as its results depend on the response of patients while the study is still ongoing, which might lead to a further delay if the trial has an initial long duration (Mahajan & Gupta, 2010). ? Sequential group trial: In this adaptive design, there is a possibility of adding modifications or stopping a study if there are problems with safety or efficacy based on results of interim analysis. Group sequential trials are commonly used in the oncology field. One of the examples is 3+3 phase1 design to identify the maximum tolerated dose.
In this model 3 patients are given a certain dose and if they do not experience any adverse effects, 3 additional patients are given an increased dose (Mahajan & Gupta, 2010). ? Change in sample size: Based on the data and results obtained during the study, we can either change or re-estimate the sample size. This can be done when the study is blinded or unblinded but it is preferable that the study does not begin with a smaller sample size, only to conduct a re-estimation later because the clinical differences and results might not be clear and significantly different due to the small number of patients.
Modification in study design also includes but is not limited to: ?Inclusion and exclusion criteria ?Change of treatment for different study groups such as dosage, time of administration and duration ?The use of concomitant medications ?Modifying the initial scheduled time of subject observation, evaluation and duration of the study in order to collect data ?Primary and secondary endpoints ?Statistical methodologies to evaluate endpoints 3) Why are adaptive trials used? Adaptive clinical trials are used because they are a flexible and an efficient way to identify benefits of a drug under investigation.
Using adaptive clinical trials we can also identify any signals, patterns, or trends that a drug causes after observing the study subjects. It is a flexible way because of their ability to adapt to several changes in the trial condition and procedure.
The efficiency can be measured in terms of the time that can be saved by conducting adaptive trial compared to the usual development time. Sponsors lean towards adaptive clinical trials because they increase the chances of success. They may also be ethical in terms of safety and efficacy of the tested therapy.
Adaptive design trials are different from standard trials because we can make modifications based on the data collected from the study while it is still going. In phase III trials, adaptive design is commonly used because it allows for stopping the study in case of proved efficacy, inefficacy or if there were issues with safety.
From a business perspective adaptive trials are also used to get as much information as possible per dollar invested through making the right decisions; while developing a certain product.
These right decisions can definitely help in accelerating the process and saving more money for sponsors. 4) What are the pros and cons of using adaptive designs? The benefits of using adaptive clinical trials is that we can decrease the time of patient exposure to investigational drug and expand the chances to determine the best dose that can be given to the patients.
Through the modification of study design we can reduce the sample size and save time on duration of development. Adaptive methods can also allow for mergers and reconstruction of several studies instead of conducting them as separate ones, which may in turn prove to be efficient in terms of the time taken, cost and the resources. Another great benefit of adaptive trial is the use of accumulating data as the research advances.
The use of this data allows for modification during the study that can lead to faster results (Judith A Quinlan & Michael Krams, 2006). The challenges in conducting an adaptive clinical trial is that although changes can be beneficial; they can sometimes show misleading results. It is common to see three to five changes in the protocol but usually the main impact of these amendments is the population studied in the trial.
Making these amendments can cause the population to change during the study and sometimes it may results in a totally new target population by the end of the study. A good example would be in the case of making major changes to the inclusion and exclusion criteria. This can result in a different population than the fixed one that was planned in the beginning of the trial (Chow & Chang, 2008). Modifications in the dosage and the study endpoints can lead to bias in the collection of data during the trial.
To account for the numerous variations, researchers have to conduct statistical test to keep the type I error and modify the sample size formula to reach the power anticipated (Chow & Chang, 2008). The statistical analysis required for adaptive designs are usually much more complicated than the ones needed for conventional studies.
A Data Monitoring Committee (DMC) is the body that supervises the integrity of the study’s results and the review of the data is done in a confidential manner (PhRMA Working Group, 2006). The challenge in this is that sometimes the adaptations in the design are beyond the scope of the DMC and members with specific expertise should be brought to help in making the decision. 5) How do regulatory agencies view adaptive clinical trials?
FDA encourages the use of adaptive clinical trials and Bayesian statistical approach in research and drug development. The EMA is also very encouraging of the adaptive design. The European agency issued a document in 2006 discussing adaptive clinical trials and their use in novel drug development. Whereas the other regulatory agencies are not very supportive of adaptive designs (Mahajan & Gupta, 2010). There are regulatory differences regarding the investigative and confirmatory phases of the adaptive clinical trials, as they are not very clear.
Modifications are supported in phase I and IIa but criticized in phase III. There are also controversies in changing the hypothesis, statistical tests, or endpoints during the studies. One of the main aims of modifying the protocols is to provide the investigators with flexibility to be able to find the best therapeutic treatment and benefits of a certain investigational drug but at the same time not jeopardizing the integrity of the process. The trials bridge the gap between agencies and sponsors because both have the same goal, which is to reduce the rate of failure.
6) What is an example of an adaptive design that has been used to support approval of a new drug? An example of an adaptive design that was used to support an approval of a new drug is performing an investigation of serial studies to predict the therapeutic response with imaging and molecular analysis 2 (I-SPY 2) in patients with breast cancer(Barker et al. , 2009). I-SPY 2 is a new model to address biomarker barrier.
It was done as neoadjuvant study in women with a higher risk, as their cancer was in an advanced yet curable stage. I-SPY 2 compared the combination of several drugs with chemotherapy to the use of therapy by itself.
The main purpose of this adaptive design was to identify a new treatment procedure based on the molecular characteristic and in this case it was the biomarker signature. The response to the treatment and endpoint was a pathology response measurement. I-SPY 2 was also analyzing, testing, validating, and quantifying biomarkers while new products were being tested (Barker et al. , 2009).
The methods that demonstrate higher Bayesian probability proving more efficacy than the standard therapy proceed with the corresponding biomarker. Otherwise, methods showing low Bayesian probability are dropped along with their corresponding biomarker signatures.
This adaptive design is a fast way to assess novel drugs and drug combinations to identify the most effective ones leading to approving the drug that mostly benefit women suffering from breast cancer. REFERENCES ?Barker, A. , Sigman, C. , Kelloff, G. , Hylton, N. , Berry, D. , & Esserman, L. (2009). I-SPY2:
An adaptive breast cancer trial design in the setting of neoadjuvant chemotherapy. Clinical Pharmacology & Therapeutics, 86(1), 97-100. doi:10. 1038/clpt. 2009. 68 ?Chow, S. , & Chang, M. (2008). Adaptive design methods in clinical trials – a review. Orphanet Journal of Rare Diseases, 3(1), 11-11.
doi:10. 1186/1750-1172-3-11 ?Judith A Quinlan, & Michael Krams. (2006). Implementing adaptive designs: Logistical and operational considerations. Drug Information Journal, 40(4), 437. ?Mahajan, R. , & Gupta, K. (2010). Adaptive design clinical trials: Methodology, challenges and prospect. Indian Journal of Pharmacology, 42(4), 201-207. doi:10. 4103/0253-7613. 68417 ?PhRMA Working Group.
(2006). Adaptive designs in clinical drug development- an executive summary of the PhRMA working group. Journal of Biopharmaceutical Statistics, 16(3), 275-283. doi:10. 1080/10543400600614742.