IDDI's Think Tank is a group of experts from various professional and academic backgrounds. The aim of the Think Tank is to drive innovative strategic thinking and to envision new paradigms for drug development.
The Group also helps solve problems encountered by clinical trial sponsors based on the considerable experience of its members.
Current reflection themes
Phase I trials in oncology
The methodology of phase I clinical trials in oncology is currently evolving in two major ways: first, by using optimized statistical methods to determine the maximum tolerated dose and second, by developing new approaches to the identification of the optimal biological dose. These ideas are further discussed in a recently published book (Eisenhauer E, Twelves C, Buyse M. Phase I Clinical Trials in Cancer. Oxford University Press (329 p.), Oxford, 2006).
Adaptive trial designs
Recent developments in biostatistics and bioinformatics have reshaped the landscape of clinical research. With the number of promising new molecules available for clinical testing, trials need to detect a drug's benefit and harm as fast as possible. An unprecedented number of techniques are now available to do so: group sequential designs allow for early termination in case of extreme efficacy or futility, while adaptive designs use information from interim analyses to adapt the trial characteristics in a statistically controlled manner.
Techniques such as sample size adjustment, stochastic curtailment based on conditional power, "drop the loser", "play the winner", adaptive randomization, Bayesian adaptive designs, and seamless transition designs are all available even though their use raises formidable challenges in practice.
The role of Independent Data Monitoring Committees, in particular, may need to evolve from a simple monitoring of safety to a broader role in helping to bring effective drugs to the market. Many of these ideas are further discussed in a chapter to appear in a forthcoming book (Herson J, Wittes J, Buyse M. Practical Issues in Interim Analysis of Clinical Trials. In: Design of Clinical Studies (Harrington D, Editor), Springer [in preparation]).
Biomarkers
Advances in molecular biology, high throughput technologies and imaging techniques, provide investigators with an ever growing number of biomarkers which are already used for a variety of purposes: to take early go/no go decisions in drug development, to stratify patients, to target "responding" subsets, to adjust treatments for individual patients, or to replace clinical endpoints for decision making. Here too challenging statistical issues are involved in identifying and validating biomarkers that have prognostic and / or predictive ability, or those that can potentially be used as surrogate endpoints in clinical trials. These ideas are taken up in a book chapter to appear soon (Buyse M, Michiels S. Biomarkers and Surrogate Endpoints in Clinical Trials. In: Fundamentals of Oncology Clinical Trials (Kelly K, Halabi S, Editors). Demos Medical Publishing [in preparation]).
Endpoints
The need for speed in clinical development leads to the search of endpoints that are observed as early as possible in the development of new drugs, yet reliably predict the hard clinical endpoints which should ultimately be affected by these new drugs. Considerable work has been devoted to the evaluation of surrogate endpoints, and has resulted in several key papers describing statistical approaches to the validation of surrogate endpoints. Much of this work has also been published as a book (Burzykowski T, Molenberghs G, Buyse M (Editors). The Evaluation of Surrogate Endpoints. Springer (408 p.), New York, 2005).
Complex randomization
The most crucial design component of a randomized trial is the method employed to randomize patients to the various treatments under investigation. The need to use adaptive randomization in some trials, as well as the complexities of drug supply management, make randomization an increasingly complex endeavour to be managed by a group having appropriate expertise in biostatistics as well as trial logistics. Some of the recent developments for complex randomizations will be discussed in a paper currently in development (Buyse et al. Covariate adaptive randomization in clinical trials [in preparation]).