(461e) Kpred and Vdss: Free Tools On the Pharmahub for Exploring Drug Tissue Partitioning and Distribution

Translating cutting edge simulation models to the next generation of pharmaceutical professionals and students is more than a problem of information management. Papers published with complex mathematical models often include summarizing remarks and discuss general trends, but the real utility of such models comes from being able to challenge them with so called “what-if” scenarios. In the case of many simple linear models, these “what if” scenarios can be done via inspection, but not always. Many nonlinear models common to pharmacokinetics do not have obvious trends, and the impact of a trend may change in different regions of parameter space. For students to explore these types of relationships, software implementation of the models with easy to use interfaces are required.

In this work, we will briefly review the PharmaHub (pharmahub.org) as an online community for exploring all aspects of pharmaceutical research from fundamental pharmacokinetics to manufacturing. We then focus on two modules in the phubpk suite of tools for pharmacokinetic modeling. At the heart of both tools is a 13-20 compartment whole body physiologically based pharmacokinetic model where each compartment corresponds to a different physiological tissue.

Kpred is a tool to explore a model for predicting the blood tissue partition coefficients for a given drug molecule (Rodgers  and Rowland. Pharm Res. 24, 5 2007). In this highly nonlinear model, physicochemical descriptors of the drug molecule are coupled with information about how the drug distributes in the blood and plasma to estimate the partition coefficient.  Users of the tool are presented with a simple interface to select drugs with known input parameters or the user can enter custom values to consider many different cases or scenarios. The Kpred tool also allows users to do Monte Carlo sampling to understand the sensitivity of the input parameters on the partition coefficients.

To gain more physical insight into the effect of different partition coefficient values mean for drug distribution, Vdss is a complimentary tool that allows the user to couple the partition coefficients with physiological tissue volumes to compute the steady state volume of distribution. A user is allowed to select a standard 70kg subject with nominal tissue volumes, or they may define a custom individual with nonstandard tissue distribution, i.e. an obese individual. The user can also use Monte Carlo sampling to understand the effect of the uncertainty in the partition coefficients and/or tissue volumes on the predicted steady state volume of distribution.

These tools serve as an example of translating complex models to audiences who would otherwise not have the resources to explore them.