The Power of Adaptive Design
Industry Insight Feb 07, 2014
The U.S. Food and Drug Administration and European Medicines Agency recently extended licenses for Aptiv Solutions ADDPLAN Software to enable improved evaluation of Phase II dose-finding trials.
With the use of adaptive clinical trials set grow, we spoke to Vladimir Dragalin, Ph.D., SVP of Software Development, Consulting at Aptiv Solutions Innovation Center to learn more adaptive design and how Aptiv are facilitating its adoption.
AB: Aptiv Solutions is helping lead the adoption of adaptive design in clinical trials, what is adaptive design and why is it important?
Vladimir Dragalin (VD): Adaptive design is defined as a multistage study design that uses accumulating data to decide how to modify aspects of the study without undermining the validity and integrity of the trial. By validity, we mean the minimization of statistical bias by using the correct statistical methods - and by integrity, we mean the minimization of operational bias through the use of appropriate trial execution technologies and working procedures, including the use of operational firewalls and independent data monitoring committees.
Adaptive clinical trials (ACTs) offer greater flexibility over traditional trials and have the potential to make investigational studies more successful as adaptations are implemented. The chance of a successful trial with a positive result can be increased when an adaptive design is utilized. Furthermore, the probability of obtaining the correct answer to an important research question, such as the correct dose for confirmation in phase III studies, is enhanced. ACTs can result in more ethical treatment of patients from at least two perspectives. First, a larger number of patients can be randomized to more favorable, effective doses with fewer patients exposed to less effective doses. Second, there is the potential to include fewer total patients, thereby reducing their risk of exposure to adverse events.
Sponsors view ACTs as a means to achieve a more rapid product registration or marketing approval. Adaptive designs can shorten the lag time between phases through adaptations implemented at interim points in the trial, eliminate the possibility of an overpowered trial (and thus the time to recruit those last few patients), or more effectively select the optimum dose in Phase II and increase Phase III success rates. Overall, adaptive designs improve development decision-making, which often results in overall time and cost savings.
AB: You recently reported that the FDA purchased licenses for Aptiv Solutions ADDPLAN® Software. What does this software enable and why was it chosen by the FDA?
VD: Adaptive trials currently account for 20% of clinical trials, but that number is going to grow over the next few years. The FDA has even issued a guidance to grant potential priority review status to INDs that utilize adaptive design. The pre-planned adaptions inherent in adaptive designs require advanced statistics and companion software to simulate and implement these adaptations.
ADDPLAN® is commercially validated software for the planning, simulation, and analysis of adaptive design trials covering both exploratory and confirmatory phases of product development.
Along with the EMA and Japan’s PMDA, the FDA is one of three regulatory agencies to license ADDPLAN this year to evaluate trial data from new drug approval applications. ADDPLAN is also available to sponsors and consulting companies.
For example, if a sponsor project statistician used ADDPLAN software to design the study, then an FDA statistician can easily reproduce the design without additional programing and validation. Moreover, the FDA statistician can use ADDPLAN to conduct a sensitivity analysis of the design performance under different deviations from the protocol assumptions.
Similarly, if a sponsor used ADDPLAN to analyze trial data, an FDA statistician can reproduce the results by running the same analysis on their version of ADDPLAN.
Regulatory agencies are responding to an increase in IND submissions utilizing adaptive design. Their interest in and support for adaptive design analysis software, such as ADDPLAN, clearly indicate growing acceptance of adaptive design by sponsors and regulators.
AB: The European Medicines Agency (EMA) published an opinion in October 2013 that the MCP-Mod methodology has the potential to enable more informative Phase II study designs. Do you see this opinion having significant impact on the industry?
VD: Definitely. The MCP-Mod approach is an efficient statistical methodology for model-based design and analysis of Phase II dose finding studies under model uncertainty and the fact that it has been endorsed by EMA could promote better design and analysis of such trials incorporating a wider dose range and an increased number of dose levels. The outcome of designing exploratory phase trials using the MCP-Mod approach is a much better understanding of the true dose-response and, as a consequence, a more informed decision on which dose to advance to subsequent studies.
MCP-Mod has been implemented in ADDPLAN® DF, the dose finding module of ADDPLAN, and it will enable sponsor companies to design more efficient proof of concept and dose-finding trials. The execution of trials designed using ADDPLAN® DF will significantly improve the way in which biopharmaceutical companies identify the optimum dose to take forward to pivotal Phase III trials.
AB: Are the current deficits in the ability of Phase II dose-finding trials acknowledged globally or is this limited to the EMA?
VD: It has been recognized for a long time that the current dose selection methods that use a small number of doses and a narrow dose range, often focused on the upper region of dose-response relationship, and based on multiple pairwise comparisons of the active doses versus placebo are not efficient, and this is one of the major causes of high rate of attrition in confirmatory Phase III trials. For example, the PhRMA Working Group on Adaptive Dose Ranging Studies published several papers1-3 showing in comprehensive simulation studies that these new innovative designs are more efficient than the conventional methods. Of course, a qualification opinion from EMA is very welcome and will definitely encourage sponsors to use these designs more frequently.
AB: ADDPLAN® is clearly enabling improvements in these trials, in your opinion what else can be done?
VD: It is very important to note that executing innovative trials is just as important as designing them, and many sponsor companies will need to understand the requirements for execution specific to such trials. Our company has substantial experience in this field, and can provide adaptive design and execution expertise and a dedicated technology platform - called AptivAdvantageTM - developed specifically for running these sorts of trials.
AptivAdvantageTM integrates electronic data capture (EDC), randomization and drug supply/inventory management into a single execution platform. This platform has been purposefully developed to execute sophisticated adaptive design trials in which a number of pre-planned adaptations may need to be implemented during specific interim analysis stages, while tightly controlling the potential to introduce operational bias.
Aptiv Solutions is well-placed to assist companies in the design and execution of ‘MCP-Mod’ trials.
1. Bornkamp, B., Bretz, F., Dmitrienko, A., Enas, G., Gaydos, B., Hsu, C.H., Konig, F., Krams, M., Liu, Q., Neuenschwander, B., Parke, T. and Pinheiro, J. (2007). Innovative Approaches for Designing and Analyzing Adaptive Dose-Ranging Trials (with discussion), Journal of Biopharmaceutical Statistics, 17:6, 965-995.
2. Pinheiro, J., Sax, F., Antonijevic, Z., Bornkamp, B., Bretz, F., Chuang-Stein, C., Dragalin, V., Fardipour, P., Gallo, P., Gillespie, W., Hsu, C.-H., Miller, F., Padmanabhan, S. K., Patel, N., Perevozskaya, I., Roy, A. Sanil, A., and Smith, J. R. (2010). Adaptive and Model-Based Dose-Ranging Trials: Quantitative Evaluation and Recommendations. White Paper of the PhRMA Working Group on Adaptive Dose-Ranging Studies, Statistics in Biopharmaceutical Research, 2: 435–454.
3. Dragalin, V., Bornkamp, B., Bretz, F., Miller, F., Padmanabhan, S.K., Perevozskaya, I., Pinheiro, J., and Smith, J.R. (2010). A Simulation Study to Compare New Adaptive Dose-Ranging Designs, Statistics in Biopharmaceutical Research, 2: 487–512.