| Title: | Partially Optimized Bayesian Phase II Efficacy and Futility Design |
| Version: | 0.1.0 |
| Description: | Implements a partially optimized Bayesian single-arm phase II design for a binary efficacy endpoint with efficacy and futility stopping. The design searches candidate response-count boundaries and, subject to prespecified error constraints, selects the design that minimizes the grey-zone at the final look. The package creates decision tables and evaluates operating characteristics by exact trial-path enumeration or Monte Carlo simulation. |
| License: | MIT + file LICENSE |
| Encoding: | UTF-8 |
| Suggests: | knitr, testthat (≥ 3.0.0) |
| VignetteBuilder: | knitr |
| Config/testthat/edition: | 3 |
| Config/roxygen2/version: | 8.1.0 |
| NeedsCompilation: | no |
| Packaged: | 2026-10-30 06:23:31 UTC; feinan.lu |
| Author: | Feinan Lu [aut, cre] |
| Maintainer: | Feinan Lu <feinan.lu@hotmail.com> |
| Repository: | CRAN |
| Date/Publication: | 2026-11-17 12:10:02 UTC |
Partially Optimized Bayesian Phase II Efficacy and Futility Design
Description
Tools for partially optimized Bayesian single-arm phase II designs with a binary endpoint and efficacy and futility stopping. Subject to prespecified error constraints, the design selects response-count boundaries that minimize the final grey-zone probability. The package provides decision tables and operating-characteristic calculations.
Author(s)
Maintainer: Feinan Lu feinan.lu@hotmail.com
Authors:
Feinan Lu feinan.lu@hotmail.com
References
Zhou, H., Lee, J. J., and Yuan, Y. (2017). BOP2: Bayesian optimal design for phase II clinical trials with simple and complex endpoints. Statistics in Medicine, 36(21), 3302-3314. doi:10.1002/sim.7338
Xu, X., Hashimoto, A., Yimer, B. B., and Takeda, K. (2025). BOP2-FE: Bayesian optimal phase II design with futility and efficacy-stopping boundaries. Journal of Biopharmaceutical Statistics. doi:10.1080/10543406.2025.2558142
Create a response-count decision table
Description
Returns posterior evidence, probability boundaries, and the futility, superiority, continue, or final grey-zone action for every possible response count at each planned analysis.
Usage
decision_table(design)
Arguments
design |
A |
Value
A data frame with one row per possible cumulative response count at each planned look.
Examples
design <- select_design(
n_max = 20, looks = c(10, 20), p0 = 0.20, p1 = 0.40,
alpha = 0.10, beta = 0.20,
lambda1_grid = 0.50, lambda2_grid = 0.20, gamma_grid = 0, eta_grid = 0
)
decision_table(design)
Plot decision boundaries
Description
Displays every response-count action at each planned analysis as a filled grid. Interim continuation and final grey-zone outcomes are combined as the grey-zone display category.
Usage
plot_decision_boundaries(design, ...)
Arguments
design |
A |
... |
Additional arguments passed to |
Value
Invisibly returns the response-count decision table used for the figure.
Examples
design <- select_design(
n_max = 20, looks = c(10, 20), p0 = 0.20, p1 = 0.40,
alpha = 0.10, beta = 0.20,
lambda1_grid = 0.50, lambda2_grid = 0.20, gamma_grid = 0, eta_grid = 0
)
plot_decision_boundaries(design)
Plot simulated operating characteristics
Description
Plot simulated operating characteristics
Usage
plot_operating_characteristics(
simulation,
metric = "superiority_probability",
...
)
Arguments
simulation |
A |
metric |
Operating characteristic to display. |
... |
Additional arguments passed to |
Value
Invisibly returns the simulation results used for the figure.
Examples
design <- select_design(
n_max = 20, looks = c(10, 20), p0 = 0.20, p1 = 0.40,
alpha = 0.10, beta = 0.20,
lambda1_grid = 0.50, lambda2_grid = 0.20, gamma_grid = 0, eta_grid = 0
)
simulation <- simulate_design(
design, true_rates = c(0.20, 0.30, 0.40), n_sim = 100, seed = 2026
)
plot_operating_characteristics(
simulation,
metric = "superiority_probability"
)
Select a partially optimized Bayesian phase II design
Description
Searches a grid of boundary parameters and returns the feasible candidate that minimizes the final grey-zone probability at the desirable response rate subject to the prespecified error constraints.
Usage
select_design(
n_max,
looks,
p0,
p1,
alpha,
beta,
prior = c(1, 1),
lambda1_grid = seq(0.05, 0.5, by = 0.05),
lambda2_grid = seq(0.05, 0.45, by = 0.05),
gamma_grid = 0:3,
eta_grid = 0:3
)
Arguments
n_max |
Maximum evaluable sample size. |
looks |
Increasing planned analysis sample sizes ending at |
p0 |
Unacceptable response-rate benchmark. |
p1 |
Clinically desirable response-rate target, strictly greater than
|
alpha |
Maximum type I error, defined as superiority probability at
|
beta |
Maximum type II error, defined as futility probability at |
prior |
Two positive beta-prior shape parameters. |
lambda1_grid |
Candidate values for the futility boundary level. |
lambda2_grid |
Candidate values for the superiority boundary level. |
gamma_grid |
Candidate values for the futility boundary shape. |
eta_grid |
Candidate values for the superiority boundary shape. |
Value
An object of class pob2ef_design containing the selected
parameters, operating characteristics, feasible candidates, and all
evaluated candidates.
Examples
design <- select_design(
n_max = 20,
looks = c(10, 20),
p0 = 0.20,
p1 = 0.40,
alpha = 0.10,
beta = 0.20,
lambda1_grid = 0.50,
lambda2_grid = 0.20,
gamma_grid = 0,
eta_grid = 0
)
design$selected
Simulate a selected design
Description
Estimates operating characteristics by Monte Carlo simulation for one or more true response rates.
Usage
simulate_design(design, true_rates, n_sim = 10000L, seed = NULL)
Arguments
design |
A |
true_rates |
True response rates to simulate. |
n_sim |
Number of simulated trials per true response rate. |
seed |
Optional integer random-number seed. |
Value
An object of class pob2ef_simulation containing scenario-level
operating-characteristic estimates and Monte Carlo standard errors.
Examples
design <- select_design(
n_max = 20, looks = c(10, 20), p0 = 0.20, p1 = 0.40,
alpha = 0.10, beta = 0.20,
lambda1_grid = 0.50, lambda2_grid = 0.20, gamma_grid = 0, eta_grid = 0
)
simulation <- simulate_design(
design, true_rates = c(0.20, 0.30, 0.40), n_sim = 100, seed = 2026
)
simulation$results