Model-based Dose Individualization Approaches Using Biomarkers

Objectives: This presentation aims at describing how modeling framework incorporating biomarker data can be a valuable tool for dose individualization in various diseases. Information on specific models of drugs such as sunitinib, warfarin, sitagliptin, etc. with corresponding biomarker data will be discussed in detail. Overview/Description of presentation: Biomarkers are helpful in clinical practice as a diagnostic tool, surrogate endpoint to assess clinical safety and efficacy, and for dose individualization. By incorporating complete time-course of biomarker changes in a model, we can quantitatively characterize the link between exposure, biomarker concentrations, and clinical outcome. An established relationship, therefore, may be used for prediction of changes in biomarker concentration and the resulting clinical outcome under a variety of conditions to evaluate individualized dosing approaches. Several examples are available on how model-based analyses of biomarker data can support the dose individualization approach in various disease states. A model relating exposure of anticancer drug sunitinib, biomarkers (vascular endothelial growth factor (VEGF), soluble vascular endothelial growth factor receptor (sVEGFR)‐2, ‐3, soluble stem cell factor receptor (sKIT)), and tumor growth to overall survival (OS) was developed to be used for dose individualization to maximize OS [1]. A KPD model that describes the relationship between warfarin dose and international normalized ratio (INR) response was developed. The model can be used to manage a priori and a posteriori individualization of warfarin therapy in both adults and children [2]. Prostaglandin E2 (PGE2) levels and thromboxane A2 (TXA2) inhibition were utilized as biomarkers for developing a model to predict drug effects and select efficacious doses in humans [3]. The key steps in the development of a model incorporating biomarkers are: 1) Development of a population model that describes the PKPD relationship of the drug and identify and quantify important predictors for a priori dose individualization 2) Transfer the model to a user-friendly decision support tool for a priori and a posteriori predictions of drug dose and biomarkers response 3) Optimize performance of model using clinical data. Conclusions/Take home message: The models discussed in the presentation serve as examples of how pharmacometrics can be used to assess exposure-biomarker-adverse effects-and clinical outcomes relationship in an integrated manner. These models also provide suitable platforms for dose individualization approaches due to their ability to predict clinical outcomes based on biomarker information