Simulation-Based Calculation of Basket Trial Operating Characteristics.

basksim

Overview

basksim calculates the operating characteristics of different basket trial designs based on simulation.

Installation

Install the development version with:

# install.packages("devtools")
devtools::install_github("lbau7/basksim")

Usage

With basksim you can calculate the operating characteristics such as rejection probabilities and mean squared error of single-stage basket trials with different designs.

At first, you have to create a design-object using a setup-function. For example to create a design-object for Fujikawa’s design (Fujikawa et al., 2020):

library(basksim)
design <- setup_fujikawa(k = 3, shape1 = 1, shape2 = 1, p0 = 0.2)

k is the number of baskets, shape1 and shape2 are the shape parameters of the Beta-prior of the response probabilities of each baskets and p0 is the response probability that defines the null hypothesis.

Use get_details to estimate several important operating characteristics:

set.seed(123)
get_details(
  design = design,
  n = c(15, 20, 25),
  p1 = c(0.2, 0.5, 0.5),
  lambda = 0.95,
  epsilon = 1.5,
  tau = 0,
  iter = 5000
)

# $Rejection_Probabilities
# [1] 0.4226 0.9824 0.9874
# 
# $FWER
# [1] 0.4226
# 
# $EWP
# [1] 0.999
# 
# $Mean
# [1] 0.2992626 0.4823250 0.4836304
# 
# $MSE
# [1] 0.020532553 0.007330251 0.006862607
# 
# $Lower_CL
# [1] 0.1517281 0.3407342 0.3440962
# 
# $Upper_CL
# [1] 0.4574680 0.6241900 0.6234426
# 
# $ECD
# [1] 2.5472
# 
# $Rejection_Probabilities_SE
# [1] 0.006985832 0.001859583 0.001577418
# 
# $FWER_SE
# [1] 0.006985832
# 
# $EWP_SE
# [1] 0.0004469899
# 
# $ECD_SE
# [1] 0.007147353