Short Courses
|
1st Course (a.m.) |
2nd Course (p.m.) |
room 1 |
Analysis of multiple endpoints in clinical trials (1 Day Course)(*) |
room 2 |
Multiple comparison procedures
(including dose-response analysis) |
Adaptive designs and group sequential designs |
(*)If you participate in 1 Day Course, please register to 1st and 2nd courses.
Analysis of multiple endpoints in clinical trials
Ajit Tamhane (Northwestern University) and
Alex Dmitrienko (Eli Lilly and Company)
This full-day course will focus on issues arising in
clinical trials with multiple endpoints, including analysis of
multiple co-primary endpoints (procedures aimed at an overall analysis
followed by an analysis of individual endpoints) and follow-up
analyses of secondary endpoints using gatekeeping procedures. It
provides a detailed overview of well-established methods as well as
novel statistical approaches developed over the past five years. The
course offers a well-balanced mix of theory and applications,
including software implementation of the described statistical methods
in SAS. Examples from clinical trials will be used throughout the
discussion to illustrate the statistical approaches discussed in the
course.
Principles and Techniques of Multiple Comparisons with Applications to Dose-Response and Microarray Studies
This course will cover concepts of error rates and techniques for controlling them. Not merely a cookbook course, this course is intended to help one choose from a fixed menu of dishes, and to let one potentially create new dishes from fresh ingredients.
For error rate control, we will differentiate between controlling the number or the proportion of incorrect rejections. The advantage and disadvantage of controlling their exceedance probabilities or expectations will be discussed. Confusing between conditional and unconditional reporting of error rates will be clarified. The error rates considered include familywise error rate (FWER), generalized familywise error rate (gFWER), false discover rate (FDR), and Fdr (a version of “conditional” FDR).
The closed testing principle and the partition testing principle will be described in some detail, and shown to generate familiar multiple tests such as Holm’s and Hochberg’s step-wise tests. Special emphasis will be given to conditions needed for closed or partition tests to have valid step-wise shortcuts.
Real applications will be used to illustrate appropriate choice of meaningful error rate, as well as construction of multiple tests to control it. Dose-response studies will be used to show how null hypotheses can be formulated to reflect different kinds of decision paths (e.g., serial or parallel), which in turn guide multiplicity adjustment calculations. Microarray studies will be used to illustrate special considerations in large scale multiple testing (including studies of gene expression levels, SNPs, and other biomarkers). Choice of error rate to control, strategies for taking dependence into account, and the advantage and disadvantage of modeling versus resampling-based analysis, will be discussed.
Adaptive Designs for Clinical Trials
Werner Brannath, Medical University of Vienna
Frank Bretz, Novartis Pharma AG
Martin Posch, Medical University of Vienna
The aim of this short course is to give an introduction to the principles and methodologies of adaptive designs for clinical trials. Adaptive designs allow for mid-course design adaptations that are based on un-blinded interim data as well as on information from outside the trial without compromising the overall type I error rate. Examples of design adaptations are the adjustment of sample sizes or the number and timing of interim analyses. These design parameters may be adapted depending on interim estimates of the variance, the treatment effect and safety parameters. Adaptive designs are particularly useful in trials with several experimental treatments/doses. In this case one can select promising treatments/doses at interim analyses. This allows to combine different phases (e.g. II and III) of a drug development program into a single trial and may lead to substantial savings in cost and time. In order to control the type I error rate one need not prefix or fully anticipate the adaptation rules in advance. This allows to react to unforeseen events and to implement complex decisions that are based on convoluted internal and external information. However especially in regulatory settings, careful planning of adaptive designs is required.
The course will provide an overview of methods from the published literature of the last two decades including the most recent developments. Special emphasis is put on sample size adjustment, estimation, and multiple hypotheses testing with adaptive designs. Additionally, regulatory issues will be discussed. All methods will be illustrated by examples or case studies.
Outline:
- Introduction to adaptive designs
- Basic principles of adaptive designs with applications to sample size adjustment and parameter estimation.
- Adaptive multiple testing procedures with applications to treatment selection
- Case studies
Bio sketches:
Werner Brannath is Associate Professor in Biostatistics at the Medical University of Vienna. He received his Ph.D. in mathematics in 1997. He has written several papers on sequential and adaptive designs designs, and on multiple testing. He is consultant for major pharmaceutical companies, has been invited speaker at numerous conferences like e.g. the annual ICB meeting 2008 and the Euro DIA meeting 2007, and he has co-authored a large number of medical papers. Additionally, he serves as Associated Editor of the Biometrical Journal.
Frank Bretz, Ph.D., is Biometrical Fellow, Novartis Pharma AG. He has been working on multiplicity issues arising in clinical trials, dose finding trials and adaptive designs since 1997 and received his PhD on this topic in 1999 at the University of Hannover. Since 2007 he is Adjunct Professor at the Medical University of Hannover. Additionally, he serves as
Associate Editor of Biometrics, Biometrical Journal and the Journal of Statistical Computation and Simulation.
Martin Posch is Associate Professor of Medical Statistics at the Medical University of Vienna which he joined in 1996. He received his Ph.D. in Mathematics and contributed to the theory of adaptive designs and multiple testing. He participated in the planning and implementation of clinical trials with adaptive interim analysis. Additionally, he serves as Associate Editor of Biometrics and Biometrical Journal.
|
|
|
Home
Invitation to MCP 2009 Tokyo
Important dates
Keynote talk, invited speakers and topics
Short Courses
Program
Registration
Registration fees
Special Issue Deadline: August 31th
Presentation File (new)
Organizing Committee
Sponsors
Conference Venue
Society for the Support of the International MCP conference
Downloads
MCP-2009 Flyer
Archive
MCP-2007 - Vienna
(Special issue of the MCP 2007)
MCP-2005 - Shanghai
MCP-2002 - Bethesda
MCP-2000 - Berlin
MCP-1996 - Tel Aviv
|