Teun M. Post (1,2), Nelleke Snelder (1), Richard Hooijmaijers (1)
(1) LAP&P Consultants BV, Leiden, The Netherlands; (2) Division of Pharmacology, Leiden Academic Centre for Drug Research, Leiden, The Netherlands
Objectives: The objective was to develop a workflow in which established and experimental treatment effects (structure and data) can be compared and therefore discussed within a team setting to support decision-making.
Methods: A Shiny application [1,2] provides an easy-to-generate and easy-to-use interface to visualize models in R, for instance to present its structure, and to compare model outcomes and observed data.
A previously reported preclinical Cardiovascular Effects Systems Pharmacology framework [4,5] was transformed from NONMEM to R [3] with an in-house developed R package, and implemented in Shiny. The framework contains several library compounds (e.g., propranolol, amlodipine) with their PK and corresponding effects on mean arterial blood pressure (MAP), cardiac output (CO), total peripheral resistance (TPR), heart rate (HR), and stroke volume (SV). The corresponding observed preclinical data was added to the application.
An option was incorporated to interactively visualize the output of new treatments based on its experimental or anticipated PK and PD characteristics, including the possibility to investigate a site of action and a type of exposure-response function. The outcome of the effects could be overlaid and combined, also with available observations.
Results: The user can:
- Select a library model
- Select dose/dosing regimen
- Display the model structure and selected site of action
- Plot its PK and PD
- Overlay with observations or combine with other treatments
- Select experimental PK and PD properties
- Select site of action
- Select type of exposure-response function
- Select dose/dosing regimen
- Select rat strain
- Overlay with observations or combine with other treatments
- Change PK and/or PD characteristics based on discussions or known properties (automatic update to outputs)
- Upload additional data
- Selection of plot types, layout and simulation characteristics
- Save and export results for reporting or re-use
Conclusions: The workflow provided a means to 1) visualize complex systems pharmacology model structures, 2) compare treatment effects and observations, 3) enable team discussions (“what-if” questions), and 4) support decision-making within a broader team. Due to the modular setup of the application, it can be extended to include items like additional treatment effects, data, and even analysis workflows (e.g. optimize selected model structures in NONMEM), and it can easily be converted to use within different projects.
References:
[1] RStudio Team (2015). RStudio: Integrated Development for R. RStudio, Inc., Boston, MA URL http://www.rstudio.com/
[2] http://shiny.rstudio.com/
[3] R Core Team (2015). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/
[4] Snelder N, et al., Mechanism-based PKPD modeling of cardiovascular effects in conscious rats – an application to fingolimod, PAGE 22 (2013) Abstr 2686 [www.page-meeting.org/?abstract=2686]
[5] Snelder N, et al., Drug effects on the CVS in conscious rats: separating cardiac output into heart rate and stroke volume using PKPD modelling. Br J Pharmacol. 2014 Nov;171(22):5076-92
Reference: PAGE 25 (2016) Abstr 5733 [www.page-meeting.org/?abstract=5733]
Poster: Methodology - Other topics