Skip to contents

Computing Toxicity Probabilities for a Given Dose, Model and Samples

Source: R/Model-methods.R
prob.Rd

[Stable]

A function that computes the probability of the occurrence of a DLE at a specified dose level, based on the model parameters (samples).

Usage 

prob(dose, model, samples, ...)

# S4 method for class 'numeric,LogisticNormal,Samples'
prob(dose, model, samples, ...)

# S4 method for class 'numeric,LogisticLogNormal,Samples'
prob(dose, model, samples, ...)

# S4 method for class 'numeric,LogisticLogNormalSub,Samples'
prob(dose, model, samples, ...)

# S4 method for class 'numeric,ProbitLogNormal,Samples'
prob(dose, model, samples, ...)

# S4 method for class 'numeric,ProbitLogNormalRel,Samples'
prob(dose, model, samples, ...)

# S4 method for class 'numeric,LogisticLogNormalGrouped,Samples'
prob(dose, model, samples, group, ...)

# S4 method for class 'numeric,TwoDrugsCombo,Samples'
prob(dose, model, samples, ...)

# S4 method for class 'matrix,TwoDrugsCombo,Samples'
prob(dose, model, samples, ...)

# S4 method for class 'numeric,LogisticKadane,Samples'
prob(dose, model, samples, ...)

# S4 method for class 'numeric,LogisticKadaneBetaGamma,Samples'
prob(dose, model, samples, ...)

# S4 method for class 'numeric,LogisticNormalMixture,Samples'
prob(dose, model, samples, ...)

# S4 method for class 'numeric,LogisticNormalFixedMixture,Samples'
prob(dose, model, samples, ...)

# S4 method for class 'numeric,LogisticLogNormalMixture,Samples'
prob(dose, model, samples, ...)

# S4 method for class 'numeric,DualEndpoint,Samples'
prob(dose, model, samples, ...)

# S4 method for class 'numeric,LogisticIndepBeta,Samples'
prob(dose, model, samples, ...)

# S4 method for class 'numeric,LogisticIndepBeta,missing'
prob(dose, model, samples, ...)

# S4 method for class 'numeric,OneParLogNormalPrior,Samples'
prob(dose, model, samples, ...)

# S4 method for class 'numeric,OneParExpPrior,Samples'
prob(dose, model, samples, ...)

# S4 method for class 'numeric,LogisticLogNormalOrdinal,Samples'
prob(dose, model, samples, grade, cumulative = TRUE, ...)

Arguments

dose

(number or numeric)
the dose which is targeted. The following recycling rule applies when samples is not missing: vectors of size 1 will be recycled to the size of the sample (i.e. size(samples)). Otherwise, dose must have the same size as the sample.

model

(GeneralModel or ModelTox)
the model for single agent dose escalation or pseudo DLE (dose-limiting events)/toxicity model.

samples

(Samples)
the samples of model's parameters that will be used to compute toxicity probabilities. Can also be missing for some models.

...

model specific parameters when samples are not used.

group

(character or factor)
for LogisticLogNormalGrouped, indicating whether to calculate the probability for the mono or for the combo arm.

grade

(integer or integer_vector)
The toxicity grade for which probabilities are required

cumulative

(flag)
Should the returned probability be cumulative (the default) or grade-specific?

Value

A proportion or numeric vector with the toxicity probabilities, or a numeric matrix for methods that evaluate multiple dose combinations at once. If non-scalar samples were used, then every element in the returned vector corresponds to one element of a sample. Hence, in this case, the output vector is of the same length as the sample vector. If scalar samples were used or no samples were used, e.g. for pseudo DLE/toxicity model, then the output is of the same length as the length of the dose. For matrix-valued dose inputs, the returned matrix contains one column per dose combination. In the case of LogisticLogNormalOrdinal, the probabilities relate to toxicities of grade given by grade.

Details

The prob() function computes the probability of toxicity for given doses, using samples of the model parameter(s). If you work with multivariate model parameters, then assume that your model specific prob() method receives a samples matrix where the rows correspond to the sampling index, i.e. the layout is then nSamples x dimParameter.

Functions

  • prob(dose = numeric, model = LogisticNormal, samples = Samples): Calculate toxicity probabilities for a LogisticNormal model.

  • prob(dose = numeric, model = LogisticLogNormal, samples = Samples): Calculate toxicity probabilities for a LogisticLogNormal model.

  • prob(dose = numeric, model = LogisticLogNormalSub, samples = Samples): Calculate toxicity probabilities for a LogisticLogNormalSub model.

  • prob(dose = numeric, model = ProbitLogNormal, samples = Samples): Calculate toxicity probabilities for a ProbitLogNormal model.

  • prob(dose = numeric, model = ProbitLogNormalRel, samples = Samples): Calculate toxicity probabilities for a ProbitLogNormalRel model.

  • prob(dose = numeric, model = LogisticLogNormalGrouped, samples = Samples): method for LogisticLogNormalGrouped which needs group argument in addition.

  • prob(dose = numeric, model = TwoDrugsCombo, samples = Samples): method for TwoDrugsCombo for a single dose combination provided as a named numeric vector.

  • prob(dose = matrix, model = TwoDrugsCombo, samples = Samples): method for TwoDrugsCombo for one or more dose combinations provided in the rows of a numeric matrix.

  • prob(dose = numeric, model = LogisticKadane, samples = Samples): Calculate toxicity probabilities for a LogisticKadane model.

  • prob(dose = numeric, model = LogisticKadaneBetaGamma, samples = Samples): Calculate toxicity probabilities for a LogisticKadaneBetaGamma model.

  • prob(dose = numeric, model = LogisticNormalMixture, samples = Samples): Calculate toxicity probabilities for a LogisticNormalMixture model.

  • prob(dose = numeric, model = LogisticNormalFixedMixture, samples = Samples): Calculate toxicity probabilities for a LogisticNormalFixedMixture model.

  • prob(dose = numeric, model = LogisticLogNormalMixture, samples = Samples): Calculate toxicity probabilities for a LogisticLogNormalMixture model.

  • prob(dose = numeric, model = DualEndpoint, samples = Samples): Calculate toxicity probabilities for a DualEndpoint model.

  • prob(dose = numeric, model = LogisticIndepBeta, samples = Samples): compute toxicity probabilities of the occurrence of a DLE at a specified dose level, based on the samples of LogisticIndepBeta model parameters.

  • prob(dose = numeric, model = LogisticIndepBeta, samples = missing): compute toxicity probabilities of the occurrence of a DLE at a specified dose level, based on the LogisticIndepBeta model parameters. All model parameters (except dose) should be present in the model object.

  • prob(dose = numeric, model = OneParLogNormalPrior, samples = Samples): Calculate toxicity probabilities for a OneParLogNormalPrior model.

  • prob(dose = numeric, model = OneParExpPrior, samples = Samples): Calculate toxicity probabilities for a OneParExpPrior model.

  • prob(dose = numeric, model = LogisticLogNormalOrdinal, samples = Samples): Calculate grade-specific toxicity probabilities for a LogisticLogNormalOrdinal model.

Note

The prob() and dose() functions are the inverse of each other, for all dose() methods for which its first argument, i.e. a given independent variable that dose depends on, represents toxicity probability.

See also

probFunction(), dose(), efficacy().

Examples 

# Create some data.
my_data <- Data(
  x = c(0.1, 0.5, 1.5, 3, 6, 10, 10, 10),
  y = c(0, 0, 0, 0, 0, 0, 1, 0),
  cohort = c(0, 1, 2, 3, 4, 5, 5, 5),
  doseGrid = c(0.1, 0.5, 1.5, 3, 6, seq(from = 10, to = 80, by = 2))
)
#> Used default patient IDs!

# Initialize a model, e.g. 'LogisticLogNormal'.
my_model <- LogisticLogNormal(
  mean = c(-0.85, 1),
  cov = matrix(c(1, -0.5, -0.5, 1), nrow = 2),
  ref_dose = 56
)

# Get samples from posterior.
my_options <- McmcOptions(burnin = 100, step = 2, samples = 20)
my_samples <- mcmc(data = my_data, model = my_model, options = my_options)

# Posterior for Prob(DLT | dose = 50).
prob(dose = 50, model = my_model, samples = my_samples)
#>  [1] 0.4665401 0.1890683 0.3162080 0.5727320 0.5727320 0.3431303 0.3431303
#>  [8] 0.3431303 0.3711676 0.3711676 0.3711676 0.3251602 0.3251602 0.5028755
#> [15] 0.5028755 0.5835603 0.5835603 0.1981376 0.2479874 0.2574285

# Create data from the 'DataDual' class.
data_dual <- DataDual(
  x = c(25, 50, 25, 50, 75, 300, 250, 150),
  y = c(0, 0, 0, 0, 0, 1, 1, 0),
  w = c(0.31, 0.42, 0.59, 0.45, 0.6, 0.7, 0.6, 0.52),
  doseGrid = seq(from = 25, to = 300, by = 25)
)
#> Used default patient IDs!
#> Used best guess cohort indices!

# Initialize a toxicity model using 'LogisticIndepBeta' model.
dlt_model <- LogisticIndepBeta(
  binDLE = c(1.05, 1.8),
  DLEweights = c(3, 3),
  DLEdose = c(25, 300),
  data = data_dual
)

# Get samples from posterior.
dlt_sample <- mcmc(data = data_dual, model = dlt_model, options = my_options)

# Posterior for Prob(DLT | dose = 100).
prob(dose = 100, model = dlt_model, samples = dlt_sample)
#>  [1] 0.3331095 0.3331095 0.3331095 0.3331095 0.3780942 0.3780942 0.4100337
#>  [8] 0.4100337 0.4100337 0.4100337 0.4070378 0.4070378 0.4070378 0.4070378
#> [15] 0.4070378 0.4070378 0.4070378 0.5868325 0.6399562 0.1724915
prob(dose = c(50, 150), model = dlt_model)
#> [1] 0.1981823 0.4601234