BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:A population-finding design with non-parametric Bayesian response 
 model - Mueller\, P (University of Texas at Austin)
DTSTART:20150706T151500Z
DTEND:20150706T160000Z
UID:TALK60044@talks.cam.ac.uk
CONTACT:42080
DESCRIPTION:Targeted therapies on the basis of genomic aberrations analysi
 s of the tumor have become a mainstream direction of cancer prognosis and 
 treatment.  Studies that match patients to targeted therapies for their pa
 rticular genomic aberrations\, across different cancer types\, are known a
 s basket trials.  For such trials it is important to find and identify the
  subgroup of patients who can most benefit from an aberration-specific tar
 geted therapy\, possibly across multiple cancer types.   \n\nWe propose an
  adaptive Bayesian clinical trial design for such subgroup identification 
 and adaptive patient allocation. We start with a decision theoretic approa
 ch\, then construct a utility function and a flexible non-parametric Bayes
 ian response model. The main features\nof the proposed design and populati
 on finding methods are that we allow for variable sets of covariates to be
  recorded by different patients and\, at least in principle\, high order i
 nteractions of covariates. The separation of the decision problem and the 
 probability model allows for the use of highly flexible response models. A
 nother important feature is the adaptive allocation of patients to an opti
 mal treatment arm based on posterior predictive probabilities.  The propos
 ed approach is demonstrated via extensive simulation studies.\n
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR