Network pharmacodynamic models for customized cancer therapy.

Abstract

Pharmacokinetics (PKs) and pharmacodynamics (PDs) have always been integral to the design of rational drug dosing regimens. Early on PK-driven approaches came under the auspices of therapeutic drug monitoring that progressed into population-based PK and PK/PD modeling analyses. As the availability of tissue samples for measurement of drug concentrations is limited in patients, the bulk of such model-based methods relied on plasma drug concentrations to both build models and monitor therapy. The continued advances in systems biology and the spawning of systems pharmacology propelled the creation of enhanced PD (ePD) models. One of the main characteristic of ePD models is that they are derived from mechanistically grounded biochemical reaction networks. These models are commonly represented as systems of coupled ordinary differential equations with the ability to tailor each reaction and protein concentration to an individual's genomic/proteomic profile. As patient genomic analyses become more common, many genetic and protein abnormalities can be represented in the ePD models, and thus offer a path toward personalized anticancer therapies. By linking PK models to ePD models, a full spectrum of pharmacological simulation tools is available to design sophisticated multidrug regimens. However, ePD models are not a panacea and face challenges in model identifiability, scaling and parameter estimation. Nonetheless, as new technologies evolve and are coupled with fresh ideas on model implementation, it is likely that ePD and PK/ePD models will be considered a viable enterprise to customize anticancer drug therapy.