Overview

The objective of the R Consortium R submission Pilot 1 Project is to test the concept that a R-language based submission package can meet the needs and the expectations of the FDA reviewers, including assessing code review and analyses reproducibility. All submission materials and communications from this pilot are publicly available, with the aim of providing a working example for future R language based FDA submissions. This is a FDA-industry collaboration through the non-profit organization R consortium.

The working group website.

The RConsortium/submissions-pilot1 repo demonstrates an approach to organize internal developed R function and table, listing, figure generation program using an R package.

The RConsortium/submissions-pilot1-to-fda repo demonstrates the eCTD submission package based on the RConsortium/submissions-pilot1 repo.

FDA Response

• Initial submission
  • version: v0.1.0
  • FDA statistical review and evaluation

Running Environment

The project is developed and tested in the environment below:

• OS: Ubuntu 20.04
• R version: R4.1.3
• Snapshot date: 2021-08-31
• Snapshot repository: https://mran.microsoft.com/snapshot/2021-08-31

Folder Structure

The work in this repo is organized as an R package following the concepts discussed in:

• Marwick, B., Boettiger, C., & Mullen, L. (2018). Packaging data analytical work reproducibly using R (and friends). The American Statistician, 72(1), 80-88.
• Wu, P., Palukuru, U. P., Luo, Y., Nepal, S., & Zhang, Y. (2021) Analysis and reporting in regulated clinical trial environment using R. PharmaSUG 2021

More details related to organize a clinical project and submission package can be found in Chapter 10-15 of the R for Clinical Study Reports and Submission book.

R function and Analysis Scripts

In short, the project is organized as an R package.

• pilot1wrappers.Rproj: RStudio project file used to open RStudio project.
• DESCRIPTION: Metadata for a package including authors, license, dependency etc.
• vignettes/: Analysis scripts using Rmarkdown.
• R/: Project specific R functions.
• man/: Manual of project specific R functions.

Datasets

The source dataset is in adam\ folder. The original data is from the PHUSE Github Repository

Startup file

• .Rprofile: Project startup file to setup running environment including R version, repository, folder path etc.
  • We further use inst/startup.R and R/zzz.R to allow the startup file is executed while running. devtools::load_all() and RStudio build panel.

Results

• output\: TLFs output generated by Rmarkdown files in vignettes\ folder.

Reproducibility

The original code is prepared and executed on Ubuntu 20.04.3 LTS. We use renv to ensure reproducibility for R version and R package version.

• renv.lock and renv\ folder: R package management using renv package. (Introduction)

Utilities

• _pkgdown.yml: pkgdown configuration file.