A PRIM approach to predictive-signature development for patient stratification

Stat Med. 2015 Jan 30;34(2):317-42. doi: 10.1002/sim.6343. Epub 2014 Oct 27.

Abstract

Patients often respond differently to a treatment because of individual heterogeneity. Failures of clinical trials can be substantially reduced if, prior to an investigational treatment, patients are stratified into responders and nonresponders based on biological or demographic characteristics. These characteristics are captured by a predictive signature. In this paper, we propose a procedure to search for predictive signatures based on the approach of patient rule induction method. Specifically, we discuss selection of a proper objective function for the search, present its algorithm, and describe a resampling scheme that can enhance search performance. Through simulations, we characterize conditions under which the procedure works well. To demonstrate practical uses of the procedure, we apply it to two real-world data sets. We also compare the results with those obtained from a recent regression-based approach, Adaptive Index Models, and discuss their respective advantages. In this study, we focus on oncology applications with survival responses.

Keywords: biomarker; patient rule induction method (PRIM); patient stratification; predictive signature; subgroup analysis.

Publication types

  • Research Support, N.I.H., Extramural

MeSH terms

  • Antineoplastic Agents / therapeutic use
  • Biomarkers / analysis
  • Breast Neoplasms / drug therapy
  • Breast Neoplasms / genetics*
  • Clinical Trials as Topic / methods
  • Clinical Trials as Topic / statistics & numerical data*
  • Computer Simulation
  • Disease-Free Survival
  • Drug Therapy, Combination
  • Estrogen Antagonists / therapeutic use
  • Female
  • Gene Expression
  • Humans
  • Lymphoma, Large B-Cell, Diffuse / drug therapy
  • Lymphoma, Large B-Cell, Diffuse / genetics*
  • Microarray Analysis / methods
  • Neoplasms / drug therapy
  • Neoplasms / genetics*
  • Outcome Assessment, Health Care / methods
  • Outcome Assessment, Health Care / statistics & numerical data
  • Patient Selection*
  • Pharmacogenetics / methods*
  • Predictive Value of Tests*
  • Receptors, Estrogen / drug effects
  • Receptors, Estrogen / genetics
  • Research Design
  • Retrospective Studies
  • Tamoxifen / therapeutic use

Substances

  • Antineoplastic Agents
  • Biomarkers
  • Estrogen Antagonists
  • Receptors, Estrogen
  • Tamoxifen