BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Causal inference and stratified medicine: an illustration of trial
  designs incorporating biomarker information for mechanisms evaluation - R
 ichard Emsley\, University of Manchester
DTSTART:20130122T143000Z
DTEND:20130122T153000Z
UID:TALK42668@talks.cam.ac.uk
CONTACT:Dr Jack Bowden
DESCRIPTION:The development of personalised (stratified) medicine is intri
 nsically dependent on an understanding of treatment-effect mechanisms (eff
 ects on therapeutic targets that mediate the effect of the treatment on cl
 inical outcomes). There is a need for novel clinical trial designs for the
  joint evaluation of treatment efficacy\, the utility of predictive marker
 s as indicators of treatment efficacy\, and the meditational mechanisms pr
 oposed as the explanation of these effects. We review the problem of confo
 unding (common causes) for the drawing of valid inferences concerning trea
 tment-effect mechanisms\, even when the data has been generated using a ra
 ndomised controlled trial. We illustrate the potential of the predictive b
 iomarker-stratified design\, together with baseline measurement of all kno
 wn prognostic markers\, to enable us to evaluate both the utility of the p
 redictive biomarker in such a stratification and\, perhaps more importantl
 y\, to estimate how much of the treatment's effect is actually explained b
 y changes in the putative mediator. We call this a biomarker stratified ef
 ficacy and mechanisms evaluation (BS-EME) trial. The analysis strategy inv
 olves the use of instrumental variable regression\, using the treatment by
  predictive biomarker interaction as an instrumental variable - a refined\
 , subtle and potentially more powerful use of Mendelian randomisation. 
LOCATION:Large  Seminar Room\, 1st Floor\, Institute of Public Health\, Un
 iversity Forvie Site\, Robinson Way\, Cambridge
END:VEVENT
END:VCALENDAR