Statistical modeling to adjust for time trends in adaptive platform trials utilizing non-concurrent controls
This is an accompanying repository for the paper “Statistical modeling to adjust for time trends in adaptive platform trials utilizing non-concurrent controls’’ by Pavla Krotka, Martin Posch, Mohamed Gewily, Günter Höglinger, and Marta Bofill Roig. It contains the code to reproduce all simulations and figures, as well as the case study presented in the paper.
The repository is structured as follows:
-
Folder simulations:
- NCC_FreqModels_simscript.R: This script contains the code to reproduce all simulations included in the paper. The results for the individual scenarios are then saved in the subfolder results.
- NCC_FreqModels_figures.Rmd: This file contains the code to create all figures presented in the paper (Section 4) and the supplementary material (Section C). The figures are saved in the subfolder figures in a .pdf, .png and .tiff formats.
-
Folder case_studies:
-
Subfolder PSP:
- NCC_FreqModels_case_study_PSP.Rmd: This file includes the code used for illustrating the use of the methods from the paper on the data from the ABBV-8E12 trial, as shown in Section 5. It also contains the code to reproduce Figure 13, which is saved in the subfolder figures. For data privacy reasons the dataset used for the illustration could not be uploaded.
- NCC_FreqModels_case_study_PSP_synthetic.Rmd: This file includes the code used for illustrating the use of the methods from the paper on synthetic data that mimic the data from the ABBV-8E12 trial.
-
Subfolder FLAIR:
- NCC_FreqModels_case_study_PSP.Rmd: This file includes the code used for illustrating the use of the methods from the paper on simulated data based on the FLAIR trial, as shown in the supplementary material (Section D). It also contains the code to reproduce Figure S12, which is saved in the subfolder figures.
-
The version of the NCC R-package
that was used for the simulation study is included in this repository.
This version is labeled as Release 1.4 on GitHub and can be installed by running the following code:
# install.packages("devtools")
devtools::install_github("pavlakrotka/NCC@v1.4", force = TRUE, build_vignettes = TRUE)Please note that prior to installing the NCC package, the
JAGS (>=3.4.1) library needs to be
installed on your computer, as it is an external dependency of the
package. For more details, see
https://pavlakrotka.github.io/NCC/articles/installation.html.
The required working directory for each code file is the folder where
this file is located. E.g., for the files NCC_FreqModels_simscript.R
and NCC_FreqModels_figures.Rmd, the required working directory is the
folder simulations.
See below the R version, operating system, and the versions of all R packages used for running the code for this manuscript (simulations, figures, as well as case studies).
> sessionInfo()
R version 4.4.2 (2024-10-31)
Platform: x86_64-pc-linux-gnu
Running under: Debian GNU/Linux 12 (bookworm)
Matrix products: default
BLAS: /usr/lib/x86_64-linux-gnu/blas/libblas.so.3.11.0
LAPACK: /usr/lib/x86_64-linux-gnu/lapack/liblapack.so.3.11.0
locale:
[1] LC_CTYPE=en_US.UTF-8 LC_NUMERIC=C LC_TIME=en_US.UTF-8 LC_COLLATE=en_US.UTF-8 LC_MONETARY=en_US.UTF-8
[6] LC_MESSAGES=en_US.UTF-8 LC_PAPER=en_US.UTF-8 LC_NAME=C LC_ADDRESS=C LC_TELEPHONE=C
[11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C
time zone: Europe/Vienna
tzcode source: system (glibc)
attached base packages:
[1] splines stats graphics grDevices utils datasets methods base
other attached packages:
[1] viridis_0.6.5 viridisLite_0.4.2 spaMM_4.5.0 lmerTest_3.1-3 lme4_1.1-36 Matrix_1.7-0 scales_1.3.0
[8] ggpubr_0.6.0 latex2exp_0.9.6 kableExtra_1.4.0 lubridate_1.9.3 forcats_1.0.0 stringr_1.5.1 dplyr_1.1.4
[15] purrr_1.0.2 readr_2.1.5 tidyr_1.3.1 tibble_3.2.1 ggplot2_3.5.1 tidyverse_2.0.0 NCC_1.0
loaded via a namespace (and not attached):
[1] tidyselect_1.2.1 rjags_4-16 loo_2.8.0 fastmap_1.2.0 digest_0.6.37 BayesPPD_1.1.3 timechange_0.3.0
[8] lifecycle_1.0.4 StanHeaders_2.32.10 magrittr_2.0.3 compiler_4.4.2 rlang_1.1.5 tools_4.4.2 ggsignif_0.6.4
[15] knitr_1.46 pkgbuild_1.4.6 curl_5.2.1 xml2_1.3.6 abind_1.4-8 registry_0.5-1 withr_3.0.2
[22] numDeriv_2016.8-1.1 grid_4.4.2 stats4_4.4.2 colorspace_2.1-1 future_1.34.0 inline_0.3.21 globals_0.16.3
[29] iterators_1.0.14 MASS_7.3-61 cli_3.6.4 mvtnorm_1.3-3 rmarkdown_2.26 crayon_1.5.3 reformulas_0.4.0
[36] generics_0.1.3 RcppParallel_5.1.10 rstudioapi_0.16.0 future.apply_1.11.3 tzdb_0.4.0 minqa_1.2.8 pbapply_1.7-2
[43] proxy_0.4-27 rstan_2.32.7 assertthat_0.2.1 parallel_4.4.2 matrixStats_1.5.0 vctrs_0.6.5 boot_1.3-31
[50] carData_3.0-5 jsonlite_1.8.8 slam_0.1-55 car_3.1-2 hms_1.1.3 rstatix_0.7.2 Formula_1.2-5
[57] listenv_0.9.1 systemfonts_1.0.6 foreach_1.5.2 glue_1.8.0 ROI_1.0-1 parallelly_1.42.0 nloptr_2.1.1
[64] codetools_0.2-20 stringi_1.8.4 gtable_0.3.6 QuickJSR_1.6.0 munsell_0.5.1 doFuture_1.0.1 pillar_1.10.1
[71] htmltools_0.5.8.1 R6_2.6.1 Rdpack_2.6.2 evaluate_0.24.0 lattice_0.22-6 rbibutils_2.3 backports_1.5.0
[78] RBesT_1.8-1 broom_1.0.5 rstantools_2.4.0 Rcpp_1.0.14 svglite_2.1.3 coda_0.19-4.1 gridExtra_2.3
[85] nlme_3.1-166 checkmate_2.3.2 xfun_0.46 mgcv_1.9-1 pkgconfig_2.0.3 Funding
This research was funded in whole or in part by the Austrian Science Fund (FWF) [ESP 442 ESPRIT-Programm].
P.K., M.P., M.G., and G.H. received funding from the European Joint Programme on Rare Diseases (Improve-PSP).