BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Individual Prediction in Prostate Cancer Studies Using a Joint Lon
 gitudinal-Survival Model - Jeremy Taylor\, University of Michigan
DTSTART:20090427T133000Z
DTEND:20090427T143000Z
UID:TALK16129@talks.cam.ac.uk
CONTACT:Michael Sweeting
DESCRIPTION:For monitoring patients treated for prostate cancer\, Prostate
  Specific Antigen (PSA) is measured periodically after they receive treatm
 ent. Increases in PSA are suggestive of recurrence of the cancer and are u
 sed in making decisions about possible new treatments. The data from studi
 es of such patients typically consist of longitudinal PSA measurements\, c
 ensored event times and\nbaseline covariates. Methods for the combined ana
 lysis of both longitudinal and survival data have been developed in recent
  years\, with the main emphasis being on modeling and estimation. We analy
 ze data from a prostate cancer study in which the patients are treated wit
 h radiation therapy using a joint model. Here we focus on utilizing the mo
 del to make individualized prediction of disease progression for censored 
 and alive patients\, based on all their available pre-treatment and follow
 -up data.\n\nIn this model the longitudinal PSA data follows a non-linear 
 hierarchical mixed model. The clinical recurrences are modeled using a tim
 e-dependent\nproportional hazards model where the time dependent covariate
 s include both the current value\nand the slope of post-treatment PSA prof
 ile. Estimates of the parameters in the model are obtained by the Markov c
 hain Monte Carlo (MCMC) technique.  The model is used to give individual p
 redictions of both future PSA values and the predicted probability of recu
 rrence up to four years in the future. An efficient algorithm is developed
  to give individual predictions for subjects who were not part of the orig
 inal data from which the model was developed. Thus the model can be used b
 y others remotely through a website portal\, to give individual prediction
 s that can be updated as more follow-up data is obtained. In this talk I w
 ill discuss the data\, the models\, the estimation methods\, the statistic
 al issues and the website\, psacalc.sph.umich.edu .\n\nThis is joint work 
 with Menggang Yu\, Donna Ankerst\, Cecile Proust-Lima\, Ning Liu\, Yongseo
 k Park and Howard Sandler.
LOCATION:Large Seminar Room\, 1st Floor\, Institute of Public Health\, Uni
 versity Forvie Site\, Robinson Way\, Cambridge
END:VEVENT
END:VCALENDAR