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

Description Usage Arguments Value See Also

View source: R/MultAdj.r

Description

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

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

1
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.

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

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 Oct. 11, 2021, 9:10 a.m.