Good Practices for Adaptive Clinical Trials in Pharmaceutical Product Development
- PDF / 20,298,156 Bytes
- 18 Pages / 612 x 792 pts (letter) Page_size
- 50 Downloads / 192 Views
Brenda Goydos Eli Lilly Keaven M. Anderson Merck
Good Practices for Adaptive Clinical Trials in Pharmaceutical Product Development
Donald Berry MD Anderson Cancer Center
Nancy Burnhan GSK
Christy Ckuong-Stein Pfizer Jennifer Dudinok Roche Parvin Fordipour Wyeth
This article is a summary ofgood adaptivepractices for the planning and implementation of adaptive designs compiled from expm*ences gained in the pharmaceutical industry. The target audience is anyone involved in the planning and execution of clinical trials. The first step prior to planning an adaptive design is to assess the appropriateness of its use. Hence, stmtegic
points to consider when assessing if an adaptive design is the right choice for a trial are discussed. In addition, stmtegic points for consideration at the design and implementation stage are included from operational, regulatorv, clinical, and statistical perspectives. Good pmctices for trial simulation, trial documentation, and data monitoring committees are provided.
Pod Gollo Novartis Sam Givens Roche
Roger Lewis Harbor-UCL4 Medical Center Jeff Maca Novartis Jose Pinkeiro Novartis
Yili Pritckett Abbott Micborl Krans Wyeth Key Words Adaptive design; Best practices; Flexible design; Implementation; Planning
Correspondence Address Brenda Gaydos. Eli Lilly and Co., Lily Corporate Center. Indianapolis. IN 46285 (email: blgdLilly.com).
INTRODUCTION Gallo et al. (1) define an adaptive design as one that uses accumulating data from the ongoing trial to modify certain aspects of the study without undermining the validity and integrity of the trial. The flexibility of adaptive designs has the potential to translate into more ethical treatment of patients within a trial (eg, response-adaptive), increased likelihood of taking the right doses into phase 3 (eg, model-based adaptive dose finding), and also faster product registration (eg, seamless phase 2/3). Furthermore, the ability to check design assumptions in a confirmatory trial and modify the trial appropriately could lead to a higher phase 3 success rate. Because of these potential advantages, interest in adaptive designs has risen to an unprecedented high level in recent years, as evidenced by the numerous workshops and literature publications dedicated to this topic, and by the number of adaptive designs being proposed to regulators. Researchers emphasize that adaptive designs are not a substitute for poor planning. By all accounts, adaptive designs require more upfront planning in terms of statistical and operational considerations. From the statistical perspective, there needs to be a clear understanding of the operating characteristics of the proposed adaptations and decision rules. For confirmatory trials, the key statistical issue in most contexts is preservation of the type I error rate and the ap-
propriate estimation of treatment effect at the end of the trial. Statistical methods for the design and analysis of adaptive designs are often technically and computationally more complex than those associated with conventional d
Data Loading...
