Objectives: Develop a web app to easily setup PK/PD simulations.
Enable online model code editing and simulation datasets generation.
Include a library with 100+ models (PK, PD, TMDD, etc.) for immediate simulation without pharmacometrics programming knowledge.
Automatically generate downloadable R code for documentation and offline execution to maintain full reproducibility.
Methods: The e-Campsis simulator is built on the open-source packages campsis [1] and campsismod [2], using the S4 object system in R to structure pharmacometric models and simulation datasets. Campsis provides an abstraction layer delegating simulation work to mrgsolve [3] or rxode2 [4].
The e-Campsis graphical user interface comprises the following key sections:
Model: Users select a model from the library or upload a Campsis model, which is displayed with syntax highlighting in an editor window. Parameters appear in dynamic JavaScript tables for modification.
Trial
Design: Users define the study design (amounts, times, intervals) and observation times, specify covariates, and enter dose adaptation formulas if needed.
Simulation: e-Campsis runs the simulation and plots results for any desired outputs in the model code. The tab has three graphical components: parameter values and scenarios, settings configuration (variability components, seed number, simulation engine), and plot customization using ggplot2 or plotly.
Post-processing: (pro version only) Simulation results can be further analyzed per individual by e.g. deriving non-compartmental PK metrics, such as AUC, Cmax or Tmax, as well as the percentage above/below a threshold level. User-defined functions as R-code enable full customization. Summary statistics (geometric mean, CV%, etc.) are derived from individual metrics and displayed in an HTML table using gtsummary [5].
Download: This tab allows the user to preview the generated R code and to download the entire project. Using this download, the user is able to reproduce the simulation results offline in R using Campsis.
Results: e-Campsis was released to the pharmacometrics community in June 2023 and is freely accessible at www.e-campsis.com. An advanced version with additional features like saving/loading the state of the application (input fields, etc.), automatic import and translation of NONMEM code, and replicate simulations was first released in December 2023 and is under active development.
Conclusion: The web-based application e-Campsis was developed to satisfy different needs: It allows the user to explore models, simulate different designs and analyze the output. The automatic generation of human readable R-code ensures reproducibility and consistency within an organization. e-Campsis can greatly support the communication of modelling and simulation results with clinical teams by interactive simulations in meetings.
