-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathREADME.Rmd
69 lines (49 loc) · 3.4 KB
/
README.Rmd
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
---
output: github_document
bibliography: "inst/references.bib"
csl: "inst/asa.csl"
---
<!-- README.md is generated from README.Rmd. Please edit that file -->
```{r, include = FALSE}
knitr::opts_chunk$set(
collapse = TRUE,
comment = "#",
fig.path = "man/figures/README-",
out.width = "100%"
)
```
# BPrinStratTTE <a href="https://boehringer-ingelheim.github.io/BPrinStratTTE/"><img src="man/figures/logo.svg" align="right" height="139" alt="BPrinStratTTE website" /></a>
<!-- badges: start -->
[](https://lifecycle.r-lib.org/articles/stages.html#experimental)
[](https://CRAN.R-project.org/package=BPrinStratTTE)
<!-- badges: end -->
Bayesian models to estimate causal effects of biological treatments on time-to-event endpoints in clinical trials with principal strata defined by the occurrence of antidrug antibodies.
## Scope
- The package contains functions to fit Bayesian principal stratification models and to perform clinical trial simulations to determine operating characteristics for given scenarios.
- Two-arm clinical trials of biological therapies are considered
- with an intercurrent event (determining the principal stratum of interest) that can only occur in the treated arm (such as the development of antidrug antibodies), and
- with a time-to-event endpoint that is assumed to follow an exponential distribution.
- Effect estimators are hazard ratios and restricted mean survival times.
- Potential predictors of the intercurrent event can be taken into account.
- The models are fitted by Monte Carlo Markov Chain (MCMC) sampling, they are coded in [Stan](https://mc-stan.org/) and precompiled.
- More flexible time-to-event distributions (piecewise-exponential and Weibull) will be considered in future versions of the package.
## Principal stratification methodology
- Principal stratification is an approach to estimate causal effects in partitions of subjects determined by post-treatment events. It was introduced in the biostatistical literature by @Frangakis2002.
- The ICH E9 (R1) addendum on estimands and sensitivity analysis in clinical trials proposed principal stratification as one approach to deal with intercurrent events in clinical trials (@ICHE9R1Guideline).
- Principal stratum membership is typically not known with certainty. A Bayesian approach may be particularly suited to deal with this type of uncertainty. Following a proposal by @Imbens1997, principal stratum membership can be treated as a latent mixture variable.
- Motivated by scientific questions arising in clinical trials of biological therapies, in this package the approach by @Imbens1997 is adapted to a specific clinical trial setting with a time-to-event endpoint and the intercurrent event only occurring in the treated group.
- For recent reviews of applications to clinical trials see @Lipkovich2022 and @Bornkamp2021.
<!-- <font size="3"> -->
References: <br>
<div id="refs"></div>
<!-- </font> -->
## Installation
The current stable version of the package can be installed from CRAN with:
```{r cran-installation, eval=FALSE}
install.packages("BPrinStratTTE")
```
The development version of the package can be installed from GitHub with:
```{r gh-installation, eval=FALSE}
if (!require("remotes")) {install.packages("remotes")}
remotes::install_github("Boehringer-Ingelheim/BPrinStratTTE")
```