An approach to optimal individualized warfarin treatment through clinical trial simulations

Personalized medicine will depend on sophisticated tools, analyses, and molecular level data and clinical information to provide optimized treatment based on each patient's individual characteristics such as health history, current health or disease status, and biochemical and physiological makeup. We discuss an approach to integrate clinical trial simulations with an optimization method to produce predictions of the best individualized treatment. Our objective is to optimize the treatment protocol by minimizing health risk to adverse drug reactions. This approach anticipates the era of genome-based medicine that requires sophisticated engineering, mathematical modeling and simulations to support best practice and clinical use of genetic data.

I. Introduction

Individualized treatment protocols developed from collective physicians' experience or derived from sophisticated clinical trials, aim to improve drug safety and efficacy and minimize patients' risk to serious complications. Genetic discoveries, validated by the statistical analysis of carefully designed clinical studies, have produced additional genotype-dependent protocols and algorithms. Generally, these treatment protocols result in fewer dose-related adverse drug reactions (ADRs) when applied to a particular population. For example, Gedge's warfarin dosing protocol resulted in lower ADRs when tested in an aging population when compared to Cooper's protocol [5].