MultAdj: Simulation-based design of traditional trials with multiple...

View source: R/MultAdj.r

MultAdjR Documentation

Simulation-based design of traditional trials with multiple objectives

Description

This function supports power calculations for a broad class of Phase III clinical trials with multiple objectives, including

  • Trials with a single source of multiplicity, i.e., trials with a single endpoint and several dose-control comparisons, or two-arm trials with several endpoints.

  • Trials with several sources of multiplicity, i.e., trials with several endpoints and several dose-placebo comparisons.

A fixed-sample design is assumed in this setting. Most commonly used multiplicity adjustments are supported for trials with a single source of multiplicity. In addition, global testing procedures can be applied in two-arm trials with several endpoints. Popular gatekeeping procedures are supported in the advanced multiplicity problems with several sources. For examples of the function call, see MultAdjExample1, MultAdjExample2 or MultAdjExample3.

Usage

MultAdj(parameters)

Arguments

parameters

List of the trial design and other parameters. The required elements are defined below:

  • endpoint_type: Character value defining the common type of trial endpoints. Possible values:

    • "Normal": Normally distributed endpoint.

    • "Binary": Binary endpoint.

  • direction: Character value defining the common direction of favorable outcome for all endpoints. Possible values: "Higher" (a higher value of each endpoint indicates a more favorable outcome) and "Lower" (a lower value of each endpoint indicates a more favorable outcome).

  • n_comparisons: Integer value defining the number of dose-control comparisons in the trial. This value must be positive.

  • n_endpoints: Integer value defining the number of endpoints in the trial. This value must be positive. Either n_comparisons or n_endpoints must be greater than 1.

  • sample_size: Integer vector defining the number of enrolled patients in each trial arm (control and experimental treatments). Each element must be positive.

  • control_mean: Numeric vector defining the mean of each endpoint in the control arm. This parameter is required only with normally distributed endpoints (endpoint_type="Normal").

  • control_sd: Numeric vector defining the standard deviation of each endpoint in the control arm. Each element must be positive. This parameter is required only with normally distributed endpoints.

  • treatment_mean: Numeric vector or matrix defining the mean of each endpoint in each experimental treatment arm. In clinical trials with several endpoints and several dose-placebo comparisons, the rows corresponds to the endpoints and the columns corresponds to the treatment-control comparisons. This parameter is required only with normally distributed endpoints.

  • treatment_sd: Numeric vector or matrix defining the standard deviation of each endpoint in each experimental treatment arm. In clinical trials with several endpoints and several dose-placebo comparisons, the rows corresponds to the endpoints and the columns corresponds to the treatment-control comparisons. Each element must be positive. This parameter is required only with normally distributed endpoints.

  • control_rate: Numeric vector defining the proportion or response rate for each endpoint in the control arm. Each element must be between 0 and 1. This parameter is required only with binary endpoints (endpoint_type= "Binary").

  • treatment_rate: Numeric vector or matrix defining the proportion or response rate for each endpoint in each experimental treatment arm. In clinical trials with several endpoints and several dose-placebo comparisons, the rows corresponds to the endpoints and the columns corresponds to the treatment-control comparisons. Each element must be between 0 and 1. This parameter is required only with binary endpoints.

  • endpoint_correlation: Numeric matrix defining the pairwise correlations among the endpoint-specific test statistics. Each element must be between 0 and 1 and the matrix must be positive definite. This parameter is required only in trials with multiple endpoints.

  • mult_test: Character value defining the multiple testing procedure, global testing procedure or gatekeeping procedure. Possible values:

    • "Bonferroni": Bonferroni multiple testing procedure.

    • "Holm": Holm multiple testing procedure in trials with a single source of multiplicity or Holm-based gatekeeping procedure in trials with several sources of multiplicity.

    • "Fixed-sequence": Fixed-sequence multiple testing procedure.

    • "Chain": Chain multiple testing procedure.

    • "Hochberg": Hochberg multiple testing procedure in trials with a single source of multiplicity or Hochberg-based gatekeeping procedure in trials with several sources of multiplicity.

    • "Hommel": Hommel multiple testing procedure in trials with a single source of multiplicity or Hommel-based gatekeeping procedure in trials with several sources of multiplicity.

    • "O'Brien": O'Brien global testing procedure.

    Note that the O'Brien procedure can be used only in two-arm trials with several endpoints, similarly gatekeeping procedures can be used only in trials with several endpoints and several dose-placebo comparisons.

  • weights: Numeric vector defining the initial hypothesis weights. Each element must be between 0 and 1. This parameter is required only with multiple testing procedures.

  • transition: Numeric matrix defining the hypothesis transition parameters. Each element must be between 0 and 1 and the sum of elements in each row must be less than or equal to 1. This parameter is required only with the chain multiple testing procedure.

  • sequence: Integer vector defining the hypothesis testing sequence. This parameter is required only with the fixed-sequence multiple testing procedures.

  • mult_method: Character value defining the mixture method for the gatekeeping procedure. Possible values:

    • "Standard": Standard mixture method.

    • "Modified": Modified mixture method.

    • "Enhanced": Enhanced mixture method.

    This parameter is required only with gatekeeping procedures.

  • mult_test_gamma: Numeric vector defining the truncation parameter for each endpoint-specific family of hypotheses. The vector's length must be equal to the number of endpoints. Each element must be between 0 and 1, the last element may be equal to 1 whereas the other elements must be strictly less than 1. This parameter is required only with gatekeeping procedures.

  • dropout_rate: Numeric value defining the patient dropout rate. A uniform patient dropout process is assumed and thus this parameter defines the fraction of patients that will be excluded from the analysis. This value must be between 0 and 1.

  • alpha: Numeric value defining the overall one-sided Type I error rate. The default value is 0.025.

  • random_seed: Integer value defining the random number generator seed. The default value is 49283.

  • nsims: Integer value defining the number of simulation runs.

  • ncores: Integer value defining the number of cores for parallel calculations. The number of cores cannot exceed the maximum available number of cores. The default value is 1.

Value

The function returns an object of class ‘⁠MultAdjResults⁠’. This object is a list with the following components:

parameters

List containing the user-specified parameters.

sim_results

Data frame containing the raw and adjusted p-values generated by the hypothesis tests for each simulation run. The first set of n columns correspond to the raw p-values for the n null hypotheses and the next set of n columns correspond to the adjusted p-values for the n null hypotheses.

sim_summary

List containing the power calculation results for the specified multiple testing procedure, global testing procedure or gatekeeping procedure.

A detailed summary of the simulation results can be created using the GenerateReport function.

See Also

MultAdjApp, MultAdjExample1, MultAdjExample2, MultAdjExample3


MedianaDesigner documentation built on Aug. 28, 2023, 9:06 a.m.