JSM 2015 Preliminary Program

Online Program Home
My Program

Abstract Details

Activity Number: 700
Type: Contributed
Date/Time: Thursday, August 13, 2015 : 10:30 AM to 12:20 PM
Sponsor: Biopharmaceutical Section
Abstract #316223 View Presentation
Title: Patient-Specific Meta-Analysis with Application to a Genomic Prostate Cancer Diagnostic
Author(s): Michael Crager* and Nan Zhang and Tara Maddala
Companies: Genomic Health, Inc. and Genomic Health, Inc. and Genomic Health, Inc.
Keywords: meta-analysis ; prediction ; cancer diagnostics
Abstract:

Patient-specific meta-analysis (PSMA, Crager and Tang, J. Appl. Stat. 2014) assesses prognosis or prediction of a clinical event for an individual patient using multiple studies. The method makes predictions for a patient by combining prediction-related estimates for the patient from each study, with weighting based on estimate precision. Fixed effects PSMA combines estimates from studies using inverse variance weighting. Random effects PSMA treats between-study differences as random. Both methods produce point estimates and confidence intervals. Simulation studies suggest the random effects model gives valid confidence intervals with 6 or more studies. The method can be used with proportional hazards regression, logistic regression, and linear models of continuous numeric outcome. An example application combines information from 2 validation studies of a prostate cancer diagnostic for assessing the probability of adverse pathology using a 17-gene genomic assay of biopsy tissue. PSMA estimates are more precise and reflect the body of evidence collected across these studies, resulting in better discrimination between aggressive and indolent disease.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2015 program





For program information, contact the JSM Registration Department or phone (888) 231-3473.

For Professional Development information, contact the Education Department.

The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.

2015 JSM Online Program Home