Individualized Prediction in Prostate Cancer Using Joint Longitudinal and Survival Models

Thursday, 23 February 2012
Murdoch Childrens Research Institute (MCRI), Royal Childrens Hospital
Flemington Road

Patients with localized prostate cancer are frequently treated with radiation therapy. Following treatment, prostate-specific antigen (PSA) measurements are typically obtained at regular intervals for the purpose of monitoring and obtaining an early indication of disease recurrence.

In this talk I will present a statistical model that describes serial PSA measurements and clinical recurrence. This joint longitudinal survival model will be fit to data on over 3000 patients using Markov Chain Monte Carlo (MCMC) methods. The model can be used to provide predictions of the future probability of disease recurrence based on a patient's series of PSA values. For a patient who shows a pattern of increasing PSA values but no clinical symptoms a possible intervention is hormone therapy.

This statistical model provides useful quantitative information that can aid in the clinical decision making about whether to initiate hormone therapy.

A website calculator has been developed so that patients and their physicians can make individualized predictions.


Prof. Jeremy Taylor

Professor Jeremy Taylor has broad interests and expertise in developing statistical methodology and applying it in biomedical research, particularly in cancer research.

He has been the PI or co-investigator on numerous NIH grants over the last twenty years. His own research undertaken at UCLA and the University of Michigan is in longitudinal modelling, power transformations, survival analysis particularly cure models, and in modelling radiation oncology related data.

Recent interests, specifically related to cancer, are in statistical methods for genomic data, statistical methods evaluating cancer biomarkers, Phase I trial design, and model validation. Prof Taylor is also the director of the Biostatistics Unit at the Cancer Center, whose function is to support and collaborate with cancer researchers at the University of Michigan in basic, clinical and prevention research.

PDF icon Jeremy_Taylor_seminar_Feb2012.pdf222.1 KB