getDesignFixed() can be used to create
a fixed (single-stage) trial design. This convenience wrapper constructs
an object of class TrialDesignFixed with
kMax = 1.getSimulationMultiArmSurvival() and
getSimulationEnrichmentSurvival() now support patient-wise
survival simulations in addition to the existing test-statistic-based
simulation approach. The new argument simulationType can be
used to choose between the new C++ implementation
("patientWise"), the previous test-statistic-based approach
("testStatisticBased"), and a basic R reference
implementation ("patientWiseBasic") intended for validation
and testing. The default "auto" setting selects the
appropriate simulation approach based on the specified arguments.
Patient-wise simulation enables scenarios based on individual patient
accrual, event, censoring, dropout, and maximum-subject constraints
while preserving the previous test-statistic-based workflow for existing
use cases. Survival planning reports derived
expectedDropoutProportion1 and
expectedDropoutProportion2 at each interim and final
planned analysis time, accounting for competing events and deterministic
planned accrual. Patient-wise two-arm, multi-arm, and enrichment
simulations report observedDropoutProportion1 and
observedDropoutProportion2 by stage, averaging
replication-specific proportions among replications contributing at that
stage. Stopped trials and discontinued groups do not contribute to later
stages. Outputs retain stage and scenario dimensions, with an additional
active-arm or population dimension where applicable. When both input
dropout rates are zero, the proportions remain in the result object but
are omitted from print and summary output. The accompanying stage-wise
dropoutObservationTime1 and
dropoutObservationTime2 remain available in result objects
and are no longer displayed. These outputs differ from input dropout
probabilities and are unavailable for test-statistic-based
simulations.getPiecewiseSurvivalTimeMultiArm()
and getPiecewiseSurvivalTimeEnrichment() define piecewise
exponential survival distributions for patient-wise multi-arm and
enrichment simulations. They support interval-specific hazard ratios and
multiple named simulation situations, with a common control-hazard
profile across treatment arms for multi-arm designs and
subgroup-specific control hazards for enrichment designs. Control and
treatment hazards can alternatively be specified through exponential
medians for each interval using medianControl (multi-arm),
medianControls (enrichment), and
medianTreatments; these interval parameters are not the
medians of the complete piecewise survival distributions. The new helper
getPiecewiseHazardRatioArray() combines group-by-interval
hazard-ratio matrices into an array of simulation situations. The
resulting survival specifications support console and Markdown output,
data-frame and matrix conversion, and R-code reconstruction with
getObjectRCode().getDataset() now supports datasets of general,
endpoint-independent estimates through the new
DatasetGeneral class. Stage-wise estimates and standard
errors can be provided using est and se;
degrees of freedom (df) are optional and default internally
to Inf, corresponding to a normal approximation.criticalValues and
futilityBounds for designs with
directionUpper = FALSE. If the direction of the alternative
is already specified when creating a design, e.g., with
getDesignGroupSequential(directionUpper = FALSE), the
boundaries are now shown in the correct direction, typically as negative
values. This also applies to futility bounds generated for beta-spending
designs.getFisherInformation() can be used to
calculate cumulative or stage-wise Fisher information at selected
planned analyses for a design plan or simulation results object for
means, rates, survival, or count data endpoints. Its typed output can be
passed directly to getFutilityBounds(), which validates
that cumulative or stage-wise information is used as required by the
requested conversion. Complete multi-stage and multi-situation output
can be piped into getFutilityBounds(); the calculation is
then carried out separately for every interim stage and planning
situation. The print methods identify the target scale, information
type, requested analysis stages and, where applicable, the planning
situations represented by the values. Trial design plans and simulation
results can also be piped directly into
getFutilityBounds(); the required Fisher information and
its cumulative or stage-wise type are then determined internally from
the requested target scale.directionUpper has changed from TRUE to
FALSE. Accordingly, the default event rates have been
adjusted to provide meaningful examples for an alternative in the lower
direction: pi2 (and piControl) now defaults to
0.5, while the sample size default for pi1 is
now c(0.1, 0.2, 0.3); the power and simulation default for
pi1 remains seq(0.2, 0.5, 0.1). One-sided
power calculations and simulations issue a warning if
directionUpper is not specified explicitly, making users
aware of the changed default. No warning is issued for sample size
calculations because the directionUpper argument has only
been added to these functions in this version. For sample size
calculations, the direction is always derived from the specified
alternative; the argument is provided solely to validate that this
derived direction is consistent with a directionUpper value
that may already have been set in the design.getComparison() and the R6
TrialComparison class for comparing designs, design
characteristics, sample size and power calculations, and simulations.
Named or unnamed objects, nested lists, and existing comparisons can be
combined, extended with add(), and edited with
delete(); getObjects(), fetch(),
$, and [[ provide access to the contained
objects. Duplicate names are rejected. Compact print and summary output
presents shared and differing parameter profiles using descriptive
object labels, context-specific captions, and titles, with display
limits for large comparisons. Full data-frame conversion, Markdown
output, and R-code reconstruction are supported, including stage
coordinates for fixed designs and robust numeric type inference. The
plot() of a TrialComparison result supports
multiple category columns and y parameters, with automatic or explicit
left/right axis assignment through yAxisSide.
getComparisonPlotParameters() previews selections, axis
transformations, and diagnostic messages. Comparison axes use
context-specific parameter captions for both single and multiple y
parameters. Comparison legends use object-aware titles and adaptive
category/parameter layouts, with measured wrapping, independent
positions, and font reductions limited to ten percent. Automatic legend
placement (legendPosition = "auto") is the default across
all rpact plots with legends, while manual positions and legacy
NA behavior remain available. Comparison operations and
package testing provide structured errors through the central error
helpers, preserving worker failure details. compare()
starts a trial comparison from supported planning objects and extends an
existing TrialComparison by reference, supporting pipelines
such as design |> compare(plan1) |> compare(plan2).
Trial comparison data frames support digits,
ceilingEnabled, and removeVariableParameters
for numeric formatting and shared profiles. The output
modes "auto", "html", and "pdf"
retain data frames and render scrollable HTML or split LaTeX longtables
automatically through knitr, without embedded report helpers.%vs% operator combines trial planning objects,
comparisons, and their summaries through getComparison()
and returns a comparison summary. It supports chaining and plotting with
explicit axis selections.textMode, width, and showAll
controls. All display limits and these controls can also be set through
RPACT_COMPARISON_SUMMARY_* environment variables or
rpact.comparison.summary.* options. Omitted content has
actionable display hints; Markdown tables retain full text.testPackage() now supports parallel test execution on
Windows, Linux, and macOS while retaining its HTML and PDF reports. The
new workers argument defaults to automatic selection
(available CPUs minus two, between one and 16); use
workers = 1 for serial execution. Local test scheduling
priorities are respected, and the result includes the worker count and
separate timings for unit tests, R basic package tests, and HTML/PDF
rendering. On Windows, testPackage() uses parallel R
processes with file-backed standard streams and result files to avoid
the pipe-related access violation in R bug 19104.testPackage() now runs R basic package tests in a
separate R process before starting parallel workers and isolates
configuration writes per test process. Worker crashes preserve partial
results and are explicitly reported as incomplete execution, with the
expected test count from META, the full diagnostic log
path, and a hint to retry with workers = 1. Partial reports
cannot claim successful installation qualification.print() of rpact result objects improved: Parameter
listings now align their colon column using applicable parameter
captions without a fixed four-character padding allowance. Technical
summaries account separately for parameter type prefixes. Caption
padding remains capped at 40 characters, and longer captions are
retained in full.NA with a
warning) for sequential designs with user-defined beta spending
(typeBetaSpending = "bsUser"), for both binding and
nonbinding futility. This avoids unreliable inversion results and C++
bounds warnings. The restriction also applies to multi-arm and
enrichment analyses; other analysis results remain available.kappa != 1 in survival planning and simulation, including
multi-arm and enrichment simulations. Lambda remains the inverse Weibull
scale; the hazard ratio is (lambda1 / lambda2)^kappa.
Exponential calculations (kappa = 1) are unchanged.typeOfSelection for
multi-arm and enrichment simulations, explaining all five selection
rules, their interaction with adaptations and thresholds, and how to
drop one worst arm at each interim using a custom selection function
(issue #159).getObjectRCode() now removes duplicate precondition
blocks while preserving complete user-defined function bodies and pipe
chains.getSimulationSurvival(), the early stopping
probability (“Early stop”) is now displayed stage-wise instead of
overallgetSimulationSurvival(),
numberOfSubjects1 and numberOfSubjects2 are no
longer shown in the print output (though still available in the result
object), and singleEventsPerStage is now displayed.getSimulationMultiArmMeans(),
getSimulationMultiArmRates(), and
getSimulationMultiArmSurvival(), the field
numberOfActiveArms has been replaced by
numberOfSelectedArms.stopIllegalArgument,
stopMissingArgument) for clearer and more consistent error
messages.getSampleSizeCounts(): field ‘expectedStudyDurationH1’
in result object was replaced by ‘studyDuration’, i.e.,
‘expectedStudyDurationH1’ is deprecatedarmNumber column was
renamed to activeArmdirection field in getStageResults() result
objects was renamed to directionUpper.Trial comparison data frames support digits,
ceilingEnabled, and removeVariableParameters
for numeric formatting and shared profiles. The output
modes "auto", "html", and "pdf"
retain data frames and render scrollable HTML or split LaTeX longtables
automatically through knitr, without embedded report helpers.
Extended testPackage() with additional options for
report customization and metadata handling. A new argument
reportFileBaseName allows explicit control over the base
name of all generated report files.
testPackage() now supports structured metadata via
the new arguments metaData and metaDataFile.
Users can either pass a named list of key–value pairs directly or
provide an external text file in KEY=VALUE format. This
metadata is included in the generated test reports and supports
reproducible, auditable installation qualification and validation
workflows.
Added two internal utility functions,
writeKeyValueFile() and readKeyValueFile(),
for writing and reading human-editable key–value metadata files
(KEY=VALUE, INI/.env-like format). These functions are
implemented using base R only, support UTF-8 encoding, comments, and
robust parsing, and are intended for use in regulated
environments.
The new key–value file utilities are used internally by
testPackage() to support structured metadata handling via
the metaData and metaDataFile arguments,
enabling reproducible and auditable inclusion of environment- and
system-specific information in installation qualification and validation
reports.
identical() for user-defined parameter
checks.getHazardRatioByPi() to
return the hazard ratio directly derived from the corresponding rate
parameters, i.e., the hazard ratio is defined as lambda1 /
lambda2; the Weibull shape parameter kappa
controls time-dependence only.finalPValues for two-sided
group-sequential designs in specific edge cases (see issue #114).Trial comparison data frames support digits,
ceilingEnabled, and removeVariableParameters
for numeric formatting and shared profiles. The output
modes "auto", "html", and "pdf"
retain data frames and render scrollable HTML or split LaTeX longtables
automatically through knitr, without embedded report helpers.
The new function getFutilityBounds() converts
futility bounds between different scales such as z-value, p-value,
conditional power, predictive power, reverse conditional power, and
effect estimate.
The new argument futilityBoundsScale in
getDesignInverseNormal() and
getDesignGroupSequential() allows specifying futility
bounds directly on alternative scales (e.g. p-value), which are
internally converted as needed.
The new argument alpha0Scale in
getDesignFisher() allows specifying alpha0 bounds directly
on alternative scales, which are internally converted as
needed.
maxNumberOfRawDatasetsPerStage in
survival clarifiedgetTestLabel() introducedTrial comparison data frames support digits,
ceilingEnabled, and removeVariableParameters
for numeric formatting and shared profiles. The output
modes "auto", "html", and "pdf"
retain data frames and render scrollable HTML or split LaTeX longtables
automatically through knitr, without embedded report helpers.
efficacyStops and futilityStops
parameter added (issue #88)
parameter stdErrorEstimate ("pooled" or
"unpooled") added for calculation of final confidence
intervals in two-sample situation for rates
testPackage() returns an
InstallationQualificationResult object
getSimulationCounts() (issue #84)Trial comparison data frames support digits,
ceilingEnabled, and removeVariableParameters
for numeric formatting and shared profiles. The output
modes "auto", "html", and "pdf"
retain data frames and render scrollable HTML or split LaTeX longtables
automatically through knitr, without embedded report helpers.
For the functions getSimulationMultiArmMeans(),
getSimulationMultiArmRates(), and
getSimulationMultiArmSurvival() it is now possible to
specify a parameter doseLevels to define the dose levels
for a linear or sigmoidEmax dose-response
relationship (see feature request #63)
Added support for unequal variances between two groups in
getSampleSizeMeans(), getPowerMeans(), and
getSimulationMeans() functions, see enhancement #70
testPackage() produces a comprehensive installation
qualification report in html and pdf format (see new vignette Installation
Qualification of rpact)
setupPackageTests() sets up the package tests by
downloading the test files and copying them to the rpact installation
directory
saveOptions() saves the current rpact
options to a configuration file
resetOptions() resets the rpact options
to their default values
Argument conservative added to
getSampleSizeRates() function, see enhancement #39
Enable futility boundaries in Boundaries p Values Scale
plot plot (type = 3) using
options("rpact.plot.show.futility.on.pvalue.scale" = TRUE)
or argument showFutilityBounds = TRUE, see enhancement #79
Enable beta-spending in Error Spending plot (type = 4)
using options("rpact.plot.show.beta.spent" = TRUE) or
argument showBetaSpent = TRUE, see enhancement #80.
Furthermore, options("rpact.plot.show.alpha.spent" = FALSE)
or argument showAlphaSpent = FALSE can be used to show only
beta-spending in the plot
getSampleSizeMeans(), getPowerMeans(),
getSimulationMeans(): For two-armed trials, it is allowed
to specify the standard deviations (stDev) separately,
i.e., as vector with two elements.
getDesignConditionalDunnett()) in analysis tool is
fixed.maxInformation improved (see enhancement #65)getObjectRCode() improved
(see #81)testPackage(): additional warning details will be added
to the test report if warnings exist* Issue #61 fixedgetPiecewiseSurvivalTime()
improvedinformationRates, userAlphaSpending, and
userBetaSpendingTrial comparison data frames support digits,
ceilingEnabled, and removeVariableParameters
for numeric formatting and shared profiles. The output
modes "auto", "html", and "pdf"
retain data frames and render scrollable HTML or split LaTeX longtables
automatically through knitr, without embedded report helpers.
The new function getSimulationCounts() can be used
to perform power simulations for clinical trials with negative binomial
distributed count data. The function returns the simulated power,
stopping probabilities, conditional power, and expected sample size for
testing mean rates for negative binomial distributed event numbers in
the two treatment groups testing situation
The functions getDesignGroupSequential(),
getDesignInverseNormal(), and
getDesignFisher() now support the argument
directionUpper to specify the direction of the alternative
for one-sided testing early at the design phase, see enhancement #26
getSampleSizeCounts() and
getPowerCounts() output boundary values also on the
treatment effect scale, see enhancement #40
The fetch() and obtain() functions can
be used to extract multiple parameters from an rpact result object and
support various output formats
kable() for rpact result objects marked
as deprecated, as the formatting and display will be handled
automatically by rpactggplot2 changed
from 2.2.0 to 3.2.0directionUpper = FALSE has no influence
in simulation for testing rates in one-sample situationTrial comparison data frames support digits,
ceilingEnabled, and removeVariableParameters
for numeric formatting and shared profiles. The output
modes "auto", "html", and "pdf"
retain data frames and render scrollable HTML or split LaTeX longtables
automatically through knitr, without embedded report helpers.
All reference classes in the package have been replaced by R6 classes. This change brings significant advantages, including improved performance, more flexible and cleaner object-oriented programming, and enhanced encapsulation of methods and properties. The transition to R6 classes allows for more efficient memory management and faster execution, making the package more robust and scalable. Additionally, R6 classes provide a more intuitive and user-friendly interface for developers, facilitating the creation and maintenance of complex data structures and workflows
Extension of the function getPerformanceScore() for
sample size recalculation rules to the setting of binary endpoints
according to Bokelmann et
al. (2024)
The getSimulationMultiArmMeans(),
getSimulationMultiArmRates(), and
getSimulationMultiArmSurvival() functions now support an
enhanced selectArmsFunction argument. Previously, only
effectVector and stage were allowed as
arguments. Now, users can optionally utilize additional arguments for
more powerful custom function implementations, including
conditionalPower, conditionalCriticalValue,
plannedSubjects/plannedEvents,
allocationRatioPlanned, selectedArms,
thetaH1 (for means and survival), stDevH1 (for
means), overallEffects, and for rates additionally:
piTreatmentsH1, piControlH1,
overallRates, and overallRatesControl
Same as above for getSimulationEnrichmentMeans(),
getSimulationEnrichmentRates(), and
getSimulationEnrichmentSurvival(). Specifically, support
for population selection with selectPopulationsFunction
argument based on predictive/posterior probabilities added (see #32)
The fetch() and obtain() functions can
be used to extract a single parameter from an rpact result object, which
is useful for writing pipe-operator linked commands
.parameterNames and
.parameterFormatFunctions were removed from all rpact
result objects in favor of a more efficient solutiongetSampleSizeSurvival() /
getPowerSurvival():
eventsPerStage replaced by
cumulativeEventsPerStagesingleEventsPerStage addedgetSimulationSurvival():
eventsPerStage replaced by
singleEventsPerStageoverallEventsPerStage replaced by
cumulativeEventsPerStagegetSimulationMultiArmSurvival():
eventsPerStage replaced by
cumulativeEventsPerStagesingleNumberOfEventsPerStage replaced by
singleEventsPerArmAndStagesingleEventsPerStage addedgetSimulationEnrichmentSurvival():
singleNumberOfEventsPerStage replaced by
singleEventsPerSubsetAndStagecovr and uploads the results to codecov.ioTrial comparison data frames support digits,
ceilingEnabled, and removeVariableParameters
for numeric formatting and shared profiles. The output
modes "auto", "html", and "pdf"
retain data frames and render scrollable HTML or split LaTeX longtables
automatically through knitr, without embedded report helpers.
The new functions getSampleSizeCounts() and
getPowerCounts() can be used to perform sample size
calculations and the assessment of test characteristics for clinical
trials with negative binomial distributed count data. This is possible
for fixed sample size and group sequential designs. For the latter, the
methodology described in Muetze et al. (2019) is implemented. These
functions can also be used to perform blinded sample size reassessments
according to Friede and Schmidli (2010).
mvnprd, mvstud,
as251Normal, and as251StudentTmnormt dependency has been removedtheta can be used for plotting of sample size
and power resultsTrial comparison data frames support digits,
ceilingEnabled, and removeVariableParameters
for numeric formatting and shared profiles. The output
modes "auto", "html", and "pdf"
retain data frames and render scrollable HTML or split LaTeX longtables
automatically through knitr, without embedded report helpers.
The new function getPerformanceScore() calculates
the conditional performance score, its sub-scores and components
according to Herrmann et
al. (2020) for a given simulation result from a two-stage
design
allocationRatioPlanned for simulating multi-arm and
enrichment designs can be a vector of length kMax, the number of
stages
getObjectRCode() (short: rcmd()): with
the new arguments pipeOperator and output many
new output variants can be specified, e.g., the native R pipe operator
or the magrittr pipe operator can be used
Generic function knitr::knit_print for all result
objects implemented and automatic code chunk option
results = 'asis' activated
df <= 500
because of erroneous results in mnormt package otherwise.
For df > 500, multivariate normal distribution is
usedomega to chi in class
TrialDesignPlanSurvivalsapply removed from C++ code to
stop deprecated warnings on r-devel-linux-x86_64-fedora-clangallocationRatioPlanned for simulating means and rates
for a two treatment groups design can be a vector of length kMax, the
number of stagescalcSubjectsFunction can be used in C++ version for
simulating means and ratescalcEventsFunction added in
getSimulationSurvival()getPerformanceScore() added: calculates the performance
score for simulation means results (1 and 2 groups; 2 stages)getDataset() to enable
pipe syntax for analysis, e.g.,
getDesignGroupSequential() |> getDataset(dataMeans) |> getAnalysisResults()SystemRequirements: C++11 added to DESCRIPTION to
enable C++ 11 compilation on R 3.xbetaAdjustment can also be used in
getDesignInverseNormal()subsets removed from result of
getWideFormat() for non-enrichment datasetspopulations in
getSimulationEnrichmentMeans(),
getSimulationEnrichmentRates(), and
getSimulationEnrichmentSurvival() has been removed since it
is always derived from effectListgetSimulationEnrichmentRates() for
calculated non-integer number of subjectsgetRawData(): the resulting data.frame now
contains the correct stopStage and
lastObservationTime (formerly
observationTime)deltaWT is provided with three decimal points for
typeOfDesign = “WToptimum”as.data.frame functions improvedTrial comparison data frames support digits,
ceilingEnabled, and removeVariableParameters
for numeric formatting and shared profiles. The output
modes "auto", "html", and "pdf"
retain data frames and render scrollable HTML or split LaTeX longtables
automatically through knitr, without embedded report helpers.
Two-sided beta-spending approach with binding and non-binding futility bounds
Delayed response utility added in design specification
getSimulationMultiArmSurvival(): single stage treatment
arm specific event numbers account for selection proceduregetSimulationEnrichmentRates() and
getSimulationEnrichmentSurvival()getDesignCharacteristics()getSimulationSurvival(): the result object now contains
the new parameter overallEventsPerStage, which contains the
values previously given in eventsPerStage (it was
“cumulative” by mistake); eventsPerStage contains now the
non-cumulative values as expectedstats::qnorm(1e-323) to
stats::qnorm(1e-100)getAnalysisResults(): issue with zero values in the
argument ‘userAlphaSpending’ fixedTrial comparison data frames support digits,
ceilingEnabled, and removeVariableParameters
for numeric formatting and shared profiles. The output
modes "auto", "html", and "pdf"
retain data frames and render scrollable HTML or split LaTeX longtables
automatically through knitr, without embedded report helpers.
Simulation tools for enrichment design testing means, rates, and
hazard ratios: function getSimulationEnrichmentMeans(),
getSimulationEnrichmentRates(),
getSimulationEnrichmentSurvival() available for simulation
of enrichment designs; note that this is a novel implementation, hence
experimental
getDesignGroupSequential() /
getDesignInverseNormal(): new typeOfDesign =
“noEarlyEfficacy” added
getSimulationSurvival(): bug fixed for
accruallIntensity = 0 at some accrual intervalsgetSimulationMultiArmMeans(),
getSimulationMultiArmRates(), and
getSimulationMultiArmSurvival()testPackage(): a problem with downloading full
set of unit tests under Debian/Linux has been fixedkable() improved: optional
knitr::kable arguments enabled, e.g., formatqnorm() calculations
improvedas.data.frame()Trial comparison data frames support digits,
ceilingEnabled, and removeVariableParameters
for numeric formatting and shared profiles. The output
modes "auto", "html", and "pdf"
retain data frames and render scrollable HTML or split LaTeX longtables
automatically through knitr, without embedded report helpers.
Analysis tools for enrichment design testing means, rates, and
hazard ratios: function getAnalysisResults() generalized
for enrichment designs; function getDataset() generalized
for entering stratified data; manual extended for enrichment
designs
Automatic boundary recalculations during the trial for analysis
with alpha spending approach, including under- and over-running: setup
via the optional parameters ‘maxInformation’ and ‘informationEpsilon’ in
function getAnalysisResults()
The new function getObjectRCode() (short:
rcmd()) returns the original R command which produced any
rpact result object, including all dependencies
getWideFormat() and getLongFormat()
return a dataset object in wide format (unstacked) or long format
(narrow, stacked)
Generic function kable() returns the output of an
rpact result object formatted in Markdown.
Generic function t() returns the transpose of an
rpact result object
getDesignFisher() fixed:
getDesignFisher(method = "noInteraction", kMax = 3) and
getDesignFisher(method = "noInteraction") produced
different resultstestPackage(): the default call is now running only a
small subset of all available unit tests; with the new argument
‘connection’ the owners of the rpact validation documentation can enter
a ‘token’ and a ‘secret’ to get full access to all unit testsgetSampleSizeSurvival(),
getSimulationSurvival(),
getNumberOfSubjects(), and
getEventProbabilities()getParameterCaption() and
getParameterName() implementedas.matrix() improved for several
result objectsgetAvailablePlotTypes() for sample size and
power results fixedgetDesignFisher(kMax = 1) in
getSimulationMultiArm...() fixedgetSimulationMultiArmSurvival(): correlation of
log-rank statistics revised and improvedgetSimulationMultiArmMeans(): name of the first
effectMeasure option “effectDifference” changed to “effectEstimate”getSimulation[MultiArm][Means/Rates/Survival]():
argument ‘showStatistics’ now works correctly and is consistently FALSE
by default for multi-arm and non-multi-armgetSimulation[MultiArm]Survival(): generic function
summary() improvedgetAnalysisResults(): generic function
summary() improvedgetAccrualTime(): improved and new argument
‘accrualIntensityType’ addedgetSampleSizeSurvival(): field ‘studyDurationH1’ in
result object was replaced by ‘studyDuration’, i.e., ‘studyDurationH1’
is deprecated and will be removed in future versionsgetSimulationMultiArmSurvival(): plannedEvents
redefined as overall events over treatment armsgetStageResults(): element overallPooledStDevs added;
print output improvedgetSampleSizeSurvival() with user
defined lambdas with different lengths: issue fixedsummary() improved for several result
objectstestPackage() improvedgetSimulationMultiArm[Means/Rates/Survival](): stage
index corrected for user defined calcSubjectsFunction or
calcEventsFunctiongetSimulationMultiArmRates(): adjustment for identical
simulated rates to account for tiesgetSimulationMultiArmSurvival(): corrected correlation
of test statisticsTrial comparison data frames support digits,
ceilingEnabled, and removeVariableParameters
for numeric formatting and shared profiles. The output
modes "auto", "html", and "pdf"
retain data frames and render scrollable HTML or split LaTeX longtables
automatically through knitr, without embedded report helpers.
Simulation tools for multi-arm design testing means, rates, and hazard ratios
Analysis tools for multi-arm design testing means, rates, and hazard ratios
getSimulationRates(): exact versions for testing a
rate (one-sample case) and equality of rates (two-sample case)
getDataset: multi-arm datasets for means, rates, and survival data
Analysis of fixed designs
Summary for analysis and simulation result objects newly implemented
Summary for most rpact result objects substantially improved and enhanced
getEventProbabilities(): plot of result
object
getNumberOfSubjects(): plot of result
object
Visual comparison of two designs:
plot(design1, design2)
Functions setOutputFormat and getOutputFormat implemented: definition of user defined output formats
getSimulationMeans(): thetaH1 and stDevH1 can be
specified for assessment of sample size recalculation (replaces
thetaStandardized)
getSimulationSurvival(): separate p-values added to
the aggregated simulation data for Fisher designs
getSimulationMeans(),
getSimulationRates(): Cumulated number of subjects
integrated in getData object
getSimulation[MultiArm][Means/Rates/Survival](): new
logical argument ‘showStatistics’ added
Example datasets (csv files) added to the package
plot type “all”: plot all available plots of an object in one
step using plot(x, type = "all")
plot type improved: ‘type’ now can be a vector, e.g.,
plot(x, type = c(1, 3))
plot(x, grid = 1): new plot argument ‘grid’ enables
the plotting of 2 or more plots in one graphic
getAnalysisResults(): list output implemented analogous
to the output of all other rpact objectsgetAnalysisResults(): the following stage result
arguments were removed from result object because they were redundant:
effectSizes, testStatistics, and pValues. Please use the ‘.stageResults’
object to access them, e.g., results$.stageResults$effectSizesgetAnalysisResults(): the following design arguments
were removed from result object because they were redundant: stages,
informationRates, criticalValues, futilityBounds, alphaSpent, and
stageLevels. Please use the ‘.design’ object to access them, e.g.,
results$.design$informationRatesplot(x, showSource = TRUE) improved for all rpact
result objects xgetSimulationRates(): issue for futility stopping for
Fisher’s combination test fixedgetSimulationSurvival(): issue for expected number of
events fixedgetSimulationSurvival(): if eventsNotAchieved > 0,
rejection/futility rate and analysis time is estimated for valid
simulation runsgetSimulationSurvival(): output improved for
lambda1/median1/hazardRatio with length > 1getSampleSizeSurvival(): calculation of the maximum
number of subjects given the provided argument ‘followUpTime’
improvedgetPiecewiseSurvivalTime(): delayed response via
list-based piecewiseSurvivalTime definition enabledgetAccrualTime() /
getSimulationSurvival(): issue with the calculation of
absolute accrual intensity by given relative accrual intensity
fixedgetRawData(): issue for multiple pi1 solvedgetAnalysisSurvival(): calculation of stage wise
results not more in getStageResultsgetStageResults(): the calculation of ‘effectSizes’ for
survival data and thetaH0 != 1 was correctedgetDataset() of survival data: issue with the internal
storage of log ranks fixedgetSampleSizeSurvival() with piecewise survival time:
issue with calculation of ‘maxNumberOfSubjects’ for given ‘followUpTime’
fixedsummary() improvedgetSampleSizeSurvival() with given
maxNumberOfSubjects improvedget[SampleSize/Power]Survival() for Kappa !=
1 improvedsummary() for getDesign[...]()
fixedsummary() fixed for
getSampleSize[...]() and getPower[...]()Trial comparison data frames support digits,
ceilingEnabled, and removeVariableParameters
for numeric formatting and shared profiles. The output
modes "auto", "html", and "pdf"
retain data frames and render scrollable HTML or split LaTeX longtables
automatically through knitr, without embedded report helpers.
Plot function for Fisher design implemented
Generic function summary() implemented for
getDesign[...](), getSampleSize[...](),
getPower[...](), and getSimulation[...]()
results: a simple boundary summary will be displayed
getDesign[...](), getSampleSize[...](),
getPower[...](), and getSimulation[...]()
resultsgetStageResults() improvedgetAccrualTime() improvedgetSampleSizeSurvival() improved: numeric
search for accrualTime if followUpTime is givenTrial comparison data frames support digits,
ceilingEnabled, and removeVariableParameters
for numeric formatting and shared profiles. The output
modes "auto", "html", and "pdf"
retain data frames and render scrollable HTML or split LaTeX longtables
automatically through knitr, without embedded report helpers.
Power calculation at given or adapted sample size for means, rates and survival data
Sample size and power calculation for survival trials with piecewise accrual time and intensity
Sample size and power calculation for survival trials with exponential survival time, piecewise exponential survival time and survival times that follow a Weibull distribution
Simulation tool for survival trials; our simulator is very fast because it was implemented with C++. Adaptive event number recalculations based on conditional power can be assessed
Simulation tool for designs with continuous and binary endpoints. Adaptive sample size recalculations based on conditional power can be assessed
Comprehensive and unified tool for performing sample size calculation for fixed sample size design
Enhanced plot functionalities
set.seed() calls ‘kind’ and ‘normal.kind’ were
specified as follows: kind = “Mersenne-Twister”, normal.kind =
“Inversion”readDatasets() fixed: variable names ‘group’ and
‘groups’ are now acceptedgetSampleSizeSurvival(): argument ‘maxNumberOfPatients’
was renamed in ‘maxNumberOfSubjects’