Transcript G2_Panel Questions_Bayesian

delete these guides from slide master before printing or giving to the client
Bayesian Trial Design – Case studies
Panelists
 Andrew Mugglin, Medtronic
 Don Berry, MD Anderson
 Michael Krams, Pfizer
 Stacy Lindborg, Eli Lilly
 Suman Bhattacharya, Biogen Idec
 Telba Irony, FDA-CDRH
2005 FDA/Industry Workshop
delete these guides from slide master before printing or giving to the client
1
delete these guides from slide master before printing or giving to the client
Questions to the Panel
Introduction
How have you been involved with
Bayesian Clinical Trials?
2005 FDA/Industry Workshop
delete these guides from slide master before printing or giving to the client
2
delete these guides from slide master before printing or giving to the client
1) In what kinds of clinical trial is it
advantageous to be Bayesian?
How does one make the determination
that using a Bayesian design is
worthwhile?
2005 FDA/Industry Workshop
delete these guides from slide master before printing or giving to the client
3
delete these guides from slide master before printing or giving to the client
2) What challenges should one expect
when implementing a Bayesian trial?
(Including non-statistical and practical
challenges)
2005 FDA/Industry Workshop
delete these guides from slide master before printing or giving to the client
4
delete these guides from slide master before printing or giving to the client
3) Are there any therapeutic areas in drug development
where the Bayesian approach may be more
advantageous and ethical than the traditional approach?
3a) To what stage of drug development are Bayesian
methods most suited?
3b) Are there examples of clinical trials where
Bayesian methods may not be appropriate?
3c) What sort of Bayesian design and analysis is
most likely to win regulatory acceptance in the drug
world?
2005 FDA/Industry Workshop
delete these guides from slide master before printing or giving to the client
5
delete these guides from slide master before printing or giving to the client
4) Should Type-1 error rates and
frequentist operating characteristics
be the standard by which Bayesian
designs are judged?
2005 FDA/Industry Workshop
delete these guides from slide master before printing or giving to the client
6
delete these guides from slide master before printing or giving to the client
Thank you
Luncheon Round Tables
2005 FDA/Industry Workshop
delete these guides from slide master before printing or giving to the client
7