Seminar über Neuere Methoden der Biometrie: Vortragsthema: Adaptive designs in rare disease trials incorporating historical data
Mi.: 21.02.2018, 17:00 Uhr - 18:30 Uhr
Ort: Institut für Biometrie und Klinische Forschung, Schmeddingstraße 56, 3. OG, Besprechungsraum 1
Dozent: Dr. Katharina Hees (Lehrstuhl für Statistik in den Biowissenschaften, TU Dortmund)
Thema: Adaptive designs in rare disease trials incorporating historical data
Diese Fortbildungsmaßnahme ist im Rahmen der Zertifizierung der ärztlichen Fortbildung der Ärztekammer Westfalen-Lippe mit 2 Punkten (Kategorie: A) anrechenbar
Abstract:
Various Bayesian methods for the incorporation of historical information in present trials have been proposed in the literature. In case that the current data match sufficiently well with the historical data, these approaches lead to increased power and thus reduce the recruitment burden. Regarding the fact that recruiting sufficient patients within an acceptable time horizon is an issue for most clinical trials, especially in the field of rare diseases, this seems to be an attractive option. However, if there is a conflict between the prior and the data of the actual trial, the inclusion of historical data can even lead to decreased power or an inflated type I error. Therefore, so-called data-adaptive prior distributions are well-suited. In case of a prior-data conflict, they down-weight the extent to which the historical data is incorporated. When planning the sample size for trials incorporating historical data, not only the type I error rate, the power, and the treatment group difference but additionally the variance and the weight of the historical data have to be specified. Usually, there is some uncertainty in the planning phase about the value of these parameters. Adaptive designs gave the opportunity to recalculate the sample size based on estimated nuisance parameters mid-course. We will have a look at operating characteristics of these methods in terms of actual type I error rate, power, and expected sample size. Furthermore, I will discuss the pros and contras from a Bayesian as well from a Frequentist view point. Application is illustrated with a clinical trial example in patients with systemic sclerosis, a rare connective tissue disorder.
Ort: Institut für Biometrie und Klinische Forschung, Schmeddingstraße 56, 3. OG, Besprechungsraum 1
Dozent: Dr. Katharina Hees (Lehrstuhl für Statistik in den Biowissenschaften, TU Dortmund)
Thema: Adaptive designs in rare disease trials incorporating historical data
Diese Fortbildungsmaßnahme ist im Rahmen der Zertifizierung der ärztlichen Fortbildung der Ärztekammer Westfalen-Lippe mit 2 Punkten (Kategorie: A) anrechenbar
Abstract:
Various Bayesian methods for the incorporation of historical information in present trials have been proposed in the literature. In case that the current data match sufficiently well with the historical data, these approaches lead to increased power and thus reduce the recruitment burden. Regarding the fact that recruiting sufficient patients within an acceptable time horizon is an issue for most clinical trials, especially in the field of rare diseases, this seems to be an attractive option. However, if there is a conflict between the prior and the data of the actual trial, the inclusion of historical data can even lead to decreased power or an inflated type I error. Therefore, so-called data-adaptive prior distributions are well-suited. In case of a prior-data conflict, they down-weight the extent to which the historical data is incorporated. When planning the sample size for trials incorporating historical data, not only the type I error rate, the power, and the treatment group difference but additionally the variance and the weight of the historical data have to be specified. Usually, there is some uncertainty in the planning phase about the value of these parameters. Adaptive designs gave the opportunity to recalculate the sample size based on estimated nuisance parameters mid-course. We will have a look at operating characteristics of these methods in terms of actual type I error rate, power, and expected sample size. Furthermore, I will discuss the pros and contras from a Bayesian as well from a Frequentist view point. Application is illustrated with a clinical trial example in patients with systemic sclerosis, a rare connective tissue disorder.
Termin in lokalen Kalender importieren