Joint detection of important biomarkers and optimal dose-response model using penalties

Stat Med. 2007 Nov 30;26(27):4876-88. doi: 10.1002/sim.2960.

Abstract

We propose a method to jointly detect influential biomarkers and estimate how they change with dose. The assessment is made in dose-ranging trials collecting multiple outcomes for efficacy, safety, pharmacokinetics or pharmacodynamics. We regress all these outcomes versus a non-parametric transformation of the dose. The regression coefficients and the parameters from the dose-response model are simultaneously estimated using a penalized alternating least-squares method. We illustrate the technique with a phase I clinical trial and a metabonomic experiment in rats.

MeSH terms

  • Animals
  • Biomarkers / analysis*
  • Chemical and Drug Induced Liver Injury
  • Clinical Trials as Topic / methods
  • Computer Simulation
  • Diethylhexyl Phthalate / pharmacokinetics
  • Diethylhexyl Phthalate / toxicity
  • Diethylhexyl Phthalate / urine
  • Dose-Response Relationship, Drug*
  • Humans
  • Least-Squares Analysis*
  • Liver Diseases / metabolism
  • Models, Biological*
  • Rats

Substances

  • Biomarkers
  • Diethylhexyl Phthalate