Using a Joint Longitudinal-Survival Model in Prostate Cancer Studies

Jeremy M.G. Taylor
Department of Biostatistics
University of Michigan

Wednesday, October 7, 2009
3:30pm
Weaver-Densford Hall 2-120
Minneapolis Campus

Abstract:

For monitoring patients treated for prostate cancer with radiation therapy, Prostate Specific Antigen (PSA) is measured periodically after they receive treatment. Increases in PSA are suggestive of recurrence of the cancer and are used in making decisions about possible new treatments, such as salvage hormone therapy. The data from studies of such patients typically consist of longitudinal PSA measurements, censored event times and baseline covariates. We develop a joint model to describe these combined longitudinal and survival data. In this model the longitudinal PSA data follows a non-linear hierarchical mixed model. The clinical recurrences are modeled using a time-dependent proportional hazards model where the time dependent covariates include both the current value and the slope of post-treatment PSA profile. Estimates of the parameters in the model are obtained by Markov chain Monte Carlo (MCMC) techniques. One application of this model is to make individualized prediction of disease progression for censored and alive patients, based on all their available pre-treatment and follow-up data. Another application is to estimate the effect of salvage hormone therapy on reducing the risk of clinical recurrence. In this talk I will discuss the data, the models, the estimation methods, the statistical issues and the individual prediction website,
psacalc.sph.umich.edu.

This is joint work with Menggang Yu, Donna Ankerst, Cecile Proust-Lima, Ning Liu, Yongseok Park and Howard Sandler.

A social tea will be held at 3:00 P.M. in A434 Mayo. All are Welcome.
For more details contact 612-624-4655 or see http://www.biostat.umn.edu/seminar_academic.html