CombPlateau_sim: Combination design Simulator for the combination of two agents where toxicity is increasing with the dose of both agent and efficacy is increasing and possibly plateaus with the dose of one agent

Description

CombPlateau_sim is used to generate simulation replicates of phase I/II clinical trial for combination studies where the toxicity is assumed to increase with the dose of both agents, and the efficacy is asummed to increase with one agent and increase and possibily plateaus with the second agent. This phase I/II adaptive design is performed using the design proposed by Riviere et al. entitled "A Bayesian dose finding design for clinical trials combining a cytotoxic agent with a molecularly targeted agent".

Usage

CombPlateau_sim(ndose_a1, ndose_a2, p_tox, p_eff, tox_max, eff_min, 
prior_tox_a1, prior_tox_a2, prior_eff_a1, prior_eff_a2, n, cohort_start=3, 
cohort=3, time_full, poisson_rate, cycle=0, nsim, c_tox=0.85, c_eff=0.10, 
seed = 2174892, threads=0)

Arguments

ndose_a1

Number of dose levels for agent 1.

ndose_a2

Number of dose levels for agent 2.

p_tox

A matrix of the true toxicity probabilities associated with the combinations. True toxicity probabilities should be entered with agent 1 in row and agent 2 in column, with increasing toxicity probabilities with both row and column numbers (see examples).

p_eff

A matrix of the true efficacy probabilities associated with the combinations. True efficacy probabilities should be entered with agent 1 in row and agent 2 in column, with increasing (or plateau) efficacy probabilities with both row and column numbers (se

tox_max

Maximum acceptable toxicity probability.

eff_min

Minimum efficacy probability desired.

prior_tox_a1

A vector of initial guesses of toxicity probabilities associated with the doses of agent 1. Must be of length ndose_a1.

prior_tox_a2

A vector of initial guesses of toxicity probabilities associated with the doses of agent 2. Must be of length ndose_a2.

prior_eff_a1

A vector of initial guesses of efficacy probabilities associated with the doses of agent 1. Must be of length ndose_a1.

prior_eff_a2

A vector of initial guesses of efficacy probabilities associated with the doses of agent 2. Must be of length ndose_a2.

Total number of patients to include in the trial.

cohort_start

Cohort size for the start-up phase. Default is set at 3 (recommended).

cohort

Cohort size for the model-based phase. Default is set at 3.

time_full

Full follow-up time window for efficacy evaluation.

poisson_rate

A value indicating the rate for the Poisson process used to simulate patient arrival, i.e. expected number of arrivals per observation window.

cycle

Minimum waiting time between two dose cohorts (usually a toxicity cycle). Default value is set at 0.

nsim

Number of simulations.

c_tox

Tocixity threshold for decision rules. The default value is set at 0.85.

c_eff

Efficacy threshold for decision rules. The default value is set at 0.10.

seed

Seed of the random number generator. Default value is set at 2174892.

threads

Number of threads to use to do the computations. If 0, it uses as many threads as available processors.

Value

An object of class "CombPlateau_sim" is returned, consisting of the operating characteristics of the design specified. Objects generated by CombPlateau_sim contain at least the following components:
p_toxTrue toxicities.
p_effTrue efficacies.
rec_dosePercentage of Selection.
n_pat_doseNumber of patients at each combination.
n_tox_doseNumber of toxicities at each combination.
n_eff_doseNumber of toxicities at each combination.
inconcPercentage of inclusive trials.
nsimNumber of simulations.
cohort_startCohort size for the start-up phase.
cohortCohort size for the model-based phase.
nTotal number of patients planned in the trial.
pat_totTotal patients accrued.
tox_maxToxicity upper bound.
eff_minEfficacy lower bound.
prior_tox_a1Prior toxicity probabilities for agent 1.
prior_tox_a2Prior toxicity probabilities for agent 2.
prior_eff_a1Prior efficacy probabilities for agent 1.
prior_eff_a2Prior efficacy probabilities for agent 2.
c_toxTocixity threshold.
c_effEfficacy threshold.
time_fullFull follow-up time for efficacy is also reminded.
poisson_rateRate for Poisson process is also reminded.
durationTrial mean duration.

References

Riviere, M-K., Yuan, Y., Dubois, F., and Zohar, S. (2015). A Bayesian dose finding design for clinical trials combining a cytotoxic agent with a molecularly targeted agent. Journal of the Royal Statistical Society - Series C.

Examples

Run this code

p_tox_sc1 = t(matrix(c(0.10,0.15,0.30,0.45,
                     0.15,0.30,0.45,0.50,
                     0.30,0.45,0.55,0.65),nrow=4,ncol=3))
p_eff_sc1 = t(matrix(c(0.25,0.25,0.26,0.27,
                     0.40,0.41,0.41,0.42,
                     0.55,0.55,0.56,0.56),nrow=4,ncol=3))
p_tox_sc4 = t(matrix(c(0.01,0.04,0.08,0.10,
                     0.03,0.05,0.10,0.15,
                     0.07,0.10,0.15,0.30),nrow=4,ncol=3))
p_eff_sc4 = t(matrix(c(0.05,0.20,0.30,0.32,
                     0.10,0.30,0.45,0.46,
                     0.20,0.40,0.60,0.61),nrow=4,ncol=3))                     
prior_tox_a1 = c(0.2, 0.3, 0.4)
prior_eff_a1 = c(0.3, 0.4, 0.5)
prior_tox_a2 = c(0.12, 0.2, 0.3, 0.4)
prior_eff_a2 = c(0.3, 0.4, 0.5, 0.59)

# UNCOMMENT THOSE EXAMPLES
#sim1 = CombPlateau_sim(ndose_a1=3, ndose_a2=4, p_tox=p_tox_sc1, 
#p_eff=p_eff_sc1, tox_max=0.30, eff_min=0.20, prior_tox_a1=prior_tox_a1, 
#prior_tox_a2=prior_tox_a2, prior_eff_a1=prior_eff_a1, 
#prior_eff_a2=prior_eff_a2, n=75, cohort_start=3, cohort=3, time_full=7, 
#poisson_rate=0.28, cycle=0, nsim=2000, c_tox=0.85, c_eff=0.10, seed = 2174892, 
#threads=0)
                        
#sim1
                        
#sim2 = CombPlateau_sim(ndose_a1=3, ndose_a2=4, p_tox=p_tox_sc4, 
#p_eff=p_eff_sc4, tox_max=0.30, eff_min=0.20, prior_tox_a1=prior_tox_a1, 
#prior_tox_a2=prior_tox_a2, prior_eff_a1=prior_eff_a1, 
#prior_eff_a2=prior_eff_a2, n=75, cohort=3, time_full=7, poisson_rate=0.28, 
#nsim=1000)

#sim2

# Dummy example, running quickly
useless = CombPlateau_sim(ndose_a1=2, ndose_a2=2, 
p_tox=matrix(c(0.05,0.10,0.15,0.25),nrow=2), 
p_eff=matrix(c(0.10,0.35,0.30,0.65),nrow=2), tox_max=0.35, eff_min=0.20, 
prior_tox_a1=c(0.1,0.3), prior_tox_a2=c(0.1,0.3), prior_eff_a1=c(0.2,0.4),
prior_eff_a2=c(0.2,0.4),
n=15, cohort=3, time_full=7, poisson_rate=1, nsim=1)