Escalation With Overdose Control is a dose escalation design for phase I clinical trials such that the probability of overdose is controlled explicitly.
It was first introduced by Babb et al.(1998) and several modifications have been studied along of the years. This R-package has three available designs: the classical EWOC introduced by Babb et al.(1998), the proportional hazards model in discussed Tighioaurt (2014), and the extended parametrization presented by Tighioaurt et al (2017).
Before installing the R-package EWOC, you may need to install Just Another Gibbs Sampler.
The R-package EWOC can be installed from GitHub with:
# install.packages("devtools")
devtools::install_github("dnzmarcio/ewoc")A new dose using the classical EWOC can be calculated:
library(ewoc)
DLT <- 0
dose <- 20
test <- ewoc_d1classical(DLT ~ dose, type = 'discrete',
theta = 0.33, alpha = 0.25,
min_dose = 0, max_dose = 100,
dose_set = seq(0, 100, 20),
rho_prior = matrix(1, ncol = 2, nrow = 1),
mtd_prior = matrix(1, ncol = 2, nrow = 1),
rounding = "nearest")summary(test)
#> Conditions
#> Minimum Dose Maximum Dose Theta Alpha Number of patients
#> 1 0 100 0.33 0.25 1
#>
#> Next Dose
#> Estimate 95% HPD
#> 1 40 (12.28 ; 99.76)
#>
#> P(DLT| next dose)
#> Estimate 95% HPD
#> 1 0.3 (0.05 ; 0.77)In addition, simulations also can be performed to evaluate a design:
library(ewoc)
DLT <- 0
dose <- 30
step_zero <- ewoc_d1classical(DLT ~ dose, type = 'discrete',
theta = 0.33, alpha = 0.25,
min_dose = 0, max_dose = 100,
dose_set = seq(0, 100, 20),
rho_prior = matrix(1, ncol = 2, nrow = 1),
mtd_prior = matrix(1, ncol = 2, nrow = 1),
rounding = "nearest")
response_sim <- response_d1classical(rho = 0.05, mtd = 20, theta = 0.33,
min_dose = 10, max_dose = 50)
sim <- ewoc_simulation(step_zero = step_zero,
n_sim = 1, sample_size = 30,
alpha_strategy = "conditional",
response_sim = response_sim,
ncores = 1)Babb, J., Rogatko, A., & Zacks, S. (1998). Cancer phase I clinical trials: efficient dose escalation with overdose control. Statistics in medicine, 17(10), 1103-1120.
Tighiouart, M., Liu, Y., & Rogatko, A. (2014). Escalation with overdose control using time to toxicity for cancer phase I clinical trials. PloS one, 9(3), e93070.
Tighiouart, M., Cook-Wiens, G., & Rogatko, A. (2017). A Bayesian Adaptive Design for Cancer Phase I Trials Using a Flexible Range of Doses. Journal of Biopharmaceutical Statistics, (just-accepted).