| Title: | Generalized Two-Stage Designs for Phase II Single-Arm Studies |
|---|---|
| Description: | One can find single-stage and two-stage designs for a phase II single-arm study with either efficacy or safety/toxicity endpoints as described in Kim and Wong (2019) <doi:10.29220/CSAM.2019.26.2.163>. |
| Authors: | Seongho Kim |
| Maintainer: | Seongho Kim <[email protected]> |
| License: | GPL (>= 2) |
| Version: | 1.0 |
| Built: | 2024-12-01 08:00:06 UTC |
| Source: | CRAN |
The R package gen2stage can generate single-stage and two-stage designs
for phase II single-arm efficacy or safety studies.
| Package: | gen2stage |
| Type: | Package |
| Version: | 1.0 |
| Date: | 2017-10-05 |
| License: | GPL-2 |
Seongho Kim <[email protected]>
Kim S and Wong WK. Phase II Two-Stage Single-Arm Clinical Trials for Testing Toxicity Levels. Commun Stat Appl Methods. 2019 Mar;26(2):163-173. https://www.ncbi.nlm.nih.gov/pubmed/31106162.
# Single-stage safety design with pu (p0) = 0.33 vs. pa (p1) = 0.20 gen2single(0.33, 0.20, 0.05, 0.20) # Single-stage efficacy design with pu (p0) = 0.67 vs. pa (p1) = 0.80 gen2single(0.67, 0.80, 0.05, 0.20) # save and print safety1 = gen2single(0.33, 0.20, 0.05, 0.20) print(safety1) # Two-stage safety design with pu (p0) = 0.33 vs. pa (p1) = 0.20 gen2simon(0.33, 0.20, 0.05, 0.20) gen2simon(0.33, 0.20, 0.05, 0.10, nmax=150) # Two-stage efficacy design with pu (p0) = 0.67 vs. pa (p1) = 0.80 gen2simon(0.67, 0.80, 0.05, 0.20) gen2simon(0.67, 0.80, 0.05, 0.10, nmax=150) # save, print and plot safety2 = gen2simon(0.33, 0.20, 0.05, 0.20) print(safety2) plot(safety2)# Single-stage safety design with pu (p0) = 0.33 vs. pa (p1) = 0.20 gen2single(0.33, 0.20, 0.05, 0.20) # Single-stage efficacy design with pu (p0) = 0.67 vs. pa (p1) = 0.80 gen2single(0.67, 0.80, 0.05, 0.20) # save and print safety1 = gen2single(0.33, 0.20, 0.05, 0.20) print(safety1) # Two-stage safety design with pu (p0) = 0.33 vs. pa (p1) = 0.20 gen2simon(0.33, 0.20, 0.05, 0.20) gen2simon(0.33, 0.20, 0.05, 0.10, nmax=150) # Two-stage efficacy design with pu (p0) = 0.67 vs. pa (p1) = 0.80 gen2simon(0.67, 0.80, 0.05, 0.20) gen2simon(0.67, 0.80, 0.05, 0.10, nmax=150) # save, print and plot safety2 = gen2simon(0.33, 0.20, 0.05, 0.20) print(safety2) plot(safety2)
Calculates generalized optimal and minimax 2-stage phase II designs based on the R function ph2simon.
gen2simon(pu, pa, ep1, ep2, nmax=100) ## S3 method for class 'gen2simon' print(x, ...) ## S3 method for class 'gen2simon' plot(x, ...)gen2simon(pu, pa, ep1, ep2, nmax=100) ## S3 method for class 'gen2simon' print(x, ...) ## S3 method for class 'gen2simon' plot(x, ...)
pu |
unacceptable response/toxicity rate |
pa |
response/toxicity rate that is desirable |
ep1 |
threshold for the probability of declaring drug/treatment desirable under p0 |
ep2 |
threshold for the probability of rejecting the drug/treatment under p1 |
nmax |
maximum total sample size (default 100; can be at most 500) |
x |
object returned by gen2simon |
... |
arguments to be passed onto plot and print commands called within |
gen2simon returns a list with pu, pa, alpha, beta and nmax as above and:
out |
matrix of best 2 stage designs for each value of total sample size n. The 6 columns are: r1, n1, r, n, EN(p0), PET(p0), alpha, beta |
The "print" method formats and returns the minimax and optimal designs. The "plot" plots the expected sample size agains the maximum sample size as in Jung et al., 2001
Kim S and Wong WK. Phase II Two-Stage Single-Arm Clinical Trials for Testing Toxicity Levels. Commun Stat Appl Methods. 2019 Mar;26(2):163-173. https://www.ncbi.nlm.nih.gov/pubmed/31106162.
Jung SH, Carey M and Kim KM. (2001). Graphical Search for Two-Stage Designs for Phase II Clinical Trials. Controlled Clinical Trials 22, 367-372.
Simon R. (1989). Optimal Two-Stage Designs for Phase II Clinical Trials. Controlled Clinical Trials 10, 1-10.
# Two-stage safety design with pu (p0) = 0.33 vs. pa (p1) = 0.20 gen2simon(0.33, 0.20, 0.05, 0.20) gen2simon(0.33, 0.20, 0.05, 0.10, nmax=150) # Two-stage efficacy design with pu (p0) = 0.67 vs. pa (p1) = 0.80 gen2simon(0.67, 0.80, 0.05, 0.20) gen2simon(0.67, 0.80, 0.05, 0.10, nmax=150) # save, print and plot safety2 = gen2simon(0.33, 0.20, 0.05, 0.20) print(safety2) plot(safety2)# Two-stage safety design with pu (p0) = 0.33 vs. pa (p1) = 0.20 gen2simon(0.33, 0.20, 0.05, 0.20) gen2simon(0.33, 0.20, 0.05, 0.10, nmax=150) # Two-stage efficacy design with pu (p0) = 0.67 vs. pa (p1) = 0.80 gen2simon(0.67, 0.80, 0.05, 0.20) gen2simon(0.67, 0.80, 0.05, 0.10, nmax=150) # save, print and plot safety2 = gen2simon(0.33, 0.20, 0.05, 0.20) print(safety2) plot(safety2)
Calculates the generalized exact one stage phase II design based on the R function ph2single.
gen2single(pu,pa,ep1,ep2,nsoln=5) ## S3 method for class 'gen2single' print(x, ...)gen2single(pu,pa,ep1,ep2,nsoln=5) ## S3 method for class 'gen2single' print(x, ...)
pu |
unacceptable response/toxicity rate |
pa |
response/toxicity rate that is desirable |
ep1 |
threshold for the probability of declaring drug/treatment desirable under p0 |
ep2 |
threshold for the probability of rejecting the drug/treatment under p1 |
nsoln |
number of designs with given alpha and beta |
x |
object returned by gen2single |
... |
arguments to be passed onto print command called within |
gen2single returns the optimal design with pu, pa, alpha, and beta as above and:
out |
matrix of the single-stage designs up to nsoln. The 4 columns are: r, n, alpha (type I error), beta (type II erro) |
The "print" method formats and returns the optimal design.
Kim S and Wong WK. Phase II Two-Stage Single-Arm Clinical Trials for Testing Toxicity Levels. Commun Stat Appl Methods. 2019 Mar;26(2):163-173. https://www.ncbi.nlm.nih.gov/pubmed/31106162.
# Single-stage safety design with pu (p0) = 0.33 vs. pa (p1) = 0.20 gen2single(0.33, 0.20, 0.05, 0.20) # Single-stage efficacy design with pu (p0) = 0.67 vs. pa (p1) = 0.80 gen2single(0.67, 0.80, 0.05, 0.20) # save and print safety1 = gen2single(0.33, 0.20, 0.05, 0.20) print(safety1)# Single-stage safety design with pu (p0) = 0.33 vs. pa (p1) = 0.20 gen2single(0.33, 0.20, 0.05, 0.20) # Single-stage efficacy design with pu (p0) = 0.67 vs. pa (p1) = 0.80 gen2single(0.67, 0.80, 0.05, 0.20) # save and print safety1 = gen2single(0.33, 0.20, 0.05, 0.20) print(safety1)
Calculates the operating characteristics of a two-stage boundary based on the R function oc.twostage.bdry.
oc.gentwostage.bdry(pu, pa, r1, n1, r, n)oc.gentwostage.bdry(pu, pa, r1, n1, r, n)
pu |
unacceptable response rate |
pa |
response rate that is desirable |
r1 |
first stage threshold to declare treatment undesirable |
n1 |
first stage sample size |
r |
overall threshold to declare treatment undesirable |
n |
total sample size |
oc.gentwostage.bdry returns the type I and II error rates as well as the probability of early temination and expected sample size under pu for a specific boundary.
Kim S and Wong WK. Phase II Two-Stage Single-Arm Clinical Trials for Testing Toxicity Levels. Commun Stat Appl Methods. 2019 Mar;26(2):163-173. https://www.ncbi.nlm.nih.gov/pubmed/31106162.
# Optimal two-stage safety design with pu (p0) = 0.33 vs. pa (p1) = 0.20 oc.gentwostage.bdry(0.33, 0.20, 8, 26, 22, 85) # Optimal two-stage efficacy design with pu (p0) = 0.67 vs. pa (p1) = 0.80 oc.gentwostage.bdry(0.67, 0.80, 18, 26, 63, 85)# Optimal two-stage safety design with pu (p0) = 0.33 vs. pa (p1) = 0.20 oc.gentwostage.bdry(0.33, 0.20, 8, 26, 22, 85) # Optimal two-stage efficacy design with pu (p0) = 0.67 vs. pa (p1) = 0.80 oc.gentwostage.bdry(0.67, 0.80, 18, 26, 63, 85)