Sample Size Calculator for Survival Endpoint Clinical Trials
An interactive 'shiny' application for sample size and power
calculation under general conditions for clinical trials with survival
endpoints. Implements the weighted log-rank method of Lu (2021)
via the 'lrstat' package, supporting
non-proportional hazards, delayed treatment effects, unequal allocation
and dropout, as well as the classic method of Freedman (1982)
via the 'powerSurvEpi' package. Results are
presented interactively with survival curves and event-prediction
timelines.
survSampleSize
survSampleSize provides an interactive Shiny application for sample size
and power calculation in clinical trials with a survival (time-to-event)
endpoint, under general design conditions.
Two complementary methods are implemented:
- Lu (2021) weighted log-rank method (via the
lrstat package), supporting
non-proportional hazards, delayed treatment effects (DTE), unequal
allocation, dropout, non-inferiority testing, and Fleming-Harrington
weighted log-rank statistics.
- Freedman (1982) classic method (via the
powerSurvEpi package)
for the proportional-hazards setting.
The app also displays theoretical survival curves and a calendar-time
event-prediction timeline, and offers a side-by-side comparison of the two
methods.
Installation
Install from CRAN:
install.packages("survSampleSize")
Or install the development version:
# install.packages("remotes")
remotes::install_github("wettlinmalfa629-hue/survSampleSize")
The interactive app relies on several packages declared in Suggests.
Install them with:
install.packages(c(
"lrstat", "powerSurvEpi", "DT", "ggplot2", "bslib", "plotly"
))
Usage
Launch the application with:
library(survSampleSize)
run_app()
This opens the Shiny app in your default browser. From there you can:
- Choose a calculation method (Lu 2021 or Freedman 1982) and direction
(solve for sample size N given power, or solve for power given N).
- Set the statistical design parameters (alpha, power, test type,
allocation ratio, non-inferiority margin).
- Set the time parameters (accrual duration, follow-up time).
- Set the survival and effect-size parameters (control median survival,
target hazard ratio, delayed-effect time, dropout rate, accrual rate).
- Click Calculate to view the results, survival curves, event-prediction
timeline, and a method comparison.
References
- Freedman, L. S. (1982). Tables of the number of patients required in
clinical trials using the log-rank test. Statistics in Medicine,
1(2), 121-129. doi:10.1002/sim.4780010204
- Lu, K. (2021). Sample size calculation for logrank test and prediction of
number of events over time. Pharmaceutical Statistics, 20(2), 229-244.
doi:10.1002/pst.2069