Mario González-Sales (1,2), Olivier Barrière (2), Pierre Olivier Tremblay (2), Fahima Nekka (1), Mario Tanguay (1,2).
(1) Université de Montréal, Montréal, Canada, (2) Inventiv Health Clinical, Montréal, Canada.
Objectives: The pharmacometrician workflow can be summarized in five common steps: 1) build the dataset, 2) explore the data, 3) model the data, 4) validate the model and 5) communicate the findings. The automatization of these steps saves time, money and avoids errors. The objective of this work was to create a flexible and user-friendly tool to help the pharmacometrician in his daily tasks.
Methods: The inVentR package consists of a series of functions developed in R [1], for data assembling, input and output visualization, and model development. An important focus has been put on user friendliness, documentation, training examples, and extensibility.
Results: A function to build a dataset in the specific format required for the NONMEM® analysis from basic tabulated files is provided. This function can handle PK and PD data, multiple analytes and responses, and also different types of administration routes including: bolus, infusion, and first and/or zero order absorption processes. An exhaustive list of functions is made available for plotting purposes. These functions use a friendly syntax with a high level of abstraction that allows vectorized input, such as multiple thetas or covariates, and return ggplot [2, 3] objects. This way, we take advantage of the modularity of ggplot to allow different layers to be combined or the plot to be faceted later. Moreover, there are specific functions dedicated to increase the efficacy and reproducibility during the modeling process. These functions can, among others, provide reasonable initial estimates, summarize model parameters, duplicate a model updating the value of the parameters, and run NONMEM® from R. inVentR package can be used to easily generate professional reports and informative outputs.
Conclusions: The inVentR package is a very flexible, powerful and efficient tool that helps the pharmacometrician through the complex model building process. inVentR will be available as an open source package from the Comprehensive R Archive Network (CRAN).
References:
[1] R, http://cran.r-project.org.
[2] Winston Chang (2012), R Graphics Cookbook Practical Recipes for Visualizing Data, O’Reilly Media.
[3] Ito K, Murphy D, Application of ggplot2 to Pharmacometric Graphics. CPT Pharmacometrics Syst Pharmacol (2013) 2(10): e79.
Reference: PAGE 24 (2015) Abstr 3422 [www.page-meeting.org/?abstract=3422]
Poster: Methodology - Other topics