rankTrial {seqDesign}

Ranking and Selection, and Head-to-Head Comparison of Individual and Pooled Treatment Arms

Description

rankTrial assesses the probability of correctly selecting the winning most efficacious (individual and/or pooled) treatment arm, and assesses power to detect relative treatment efficacy in head-to-head comparisons of (individual and/or pooled) treatment arms.

Usage

rankTrial(censFile, idxHighestVE, headHead = NULL, poolHead = NULL,
  lowerVE, stage1, stage2, alpha, saveDir = NULL, verbose = TRUE)

Arguments

censFile

if saveDir = NULL, a list returned by censTrial; otherwise a name (character string) of an .RData file created by censTrial

idxHighestVE

an integer value identifying the treatment (vaccine) arm with the true highest VE(0–stage2)

headHead

a matrix (ncol = 2) of treatment arm indices for head-to-head comparisons, where the treatment with higher efficacy is listed first in each row

poolHead

a matrix (ncol equals 3 or 4) of treatment arm indices for pooled-arm comparisons, where the pooled treatment with higher efficacy pooled over the first two columns is compared with the (pooled) treatment defined by columns 3 and onward. Ranking and selection of pooled arms is performed separately for each row of poolHead.

lowerVE

a numeric value defining a ‘winning’ treatment arm as one with sufficient evidence for rejecting the null hypothesis H0: VE(0–stage1) \le lowerVE x 100% (typically set equal to 0)

stage1

the final week of stage 1 in a two-stage trial

stage2

the final week of stage 2 in a two-stage trial, i.e., the maximum follow-up time

alpha

one minus the nominal confidence level of the two-sided confidence interval used for testing a null hypothesis H0: VE(0–stage1) \le b x 100% against an alternative hypothesis H1: VE(0–stage1) > b x 100%

saveDir

a character string specifying a path for censFile. If supplied, the output is also saved as an .RData file in this directory; otherwise the output is returned as a list.

verbose

a logical value indicating whether information on the output directory and file name should be printed out (default is TRUE)

Details

All time variables use week as the unit of time. Month is defined as 52/12 weeks.

The probability of correct treatment selection is defined as the probability that the treatment arm with the highest estimated VE(0–stage2) is the one with the true highest VE(0–stage2) and, for this treatment arm, the null hypothesis H0: VE(0–stage1) \le lowerVE x 100% is rejected. If poolHead is specified, the probability of correct pooled treatment selection is assessed for each set of two pooled treatment arms.

VE(0–t) is estimated as one minus the ratio of Nelson-Aalen-based cumulative incidence functions. The null hypothesis H0: VE(0–t) \le b x 100% is rejected if the lower bound of the two-sided (1-alpha) x 100% confidence interval for VE(0–t) lies above b.

For head-to-head individual and pooled treatment comparisons, powers to reject the null hypotheses that relative VE(0–stage1) \le 0% and relative VE(0–stage2) \le 0% are assessed using the aforementioned testing rule.

Value

If saveDir is specified, the output list (named out) is saved as an .RData file in saveDir (the path to saveDir is printed); otherwise it is returned. The output object is a list with the following components:

• rankSelectPw: the probability of correct selection of the winning most efficacious individual treatment

• headHeadPw: if headHead is specified, a matrix of powers to detect relative VE(0–stage1) (column 1) and relative VE(0–stage2) (column 2) in head-to-head comparisons of individual treatment arms

• poolRankSelectPw: if poolHead is specified, a numeric vector of the probabilities of correct selection of the winning most efficacious pooled treatment for each set of pooled treatments

• poolHeadPw: if poolHead is specified, a matrix of powers to detect relative VE(0–stage1) (column 1) and relative VE(0–stage2) (column 2) in head-to-head comparisons of pooled treatment arms

See Also

simTrial, monitorTrial, and censTrial

Examples

simData <- simTrial(N=c(1000, rep(700, 2)), aveVE=seq(0, 0.4, by=0.2), 
                    VEmodel="half", vePeriods=c(1, 27, 79), enrollPeriod=78, 
                    enrollPartial=13, enrollPartialRelRate=0.5, dropoutRate=0.05, 
                    infecRate=0.04, fuTime=156, 
                    visitSchedule=c(0, (13/3)*(1:4), seq(13*6/3, 156, by=13*2/3)),
                    missVaccProb=c(0,0.05,0.1,0.15), VEcutoffWeek=26, nTrials=5, 
                    stage1=78, randomSeed=300)

monitorData <- monitorTrial(dataFile=simData, stage1=78, stage2=156, 
                            harmMonitorRange=c(10,100), alphaPerTest=NULL, 
                            nonEffStartMethod="FKG", nonEffInterval=20, 
                            lowerVEnoneff=0, upperVEnoneff=0.4, 
                            highVE=0.7, stage1VE=0, lowerVEuncPower=0, 
                            alphaNoneff=0.05, alphaHigh=0.05, alphaStage1=0.05, 
                            alphaUncPower=0.05, estimand="cuminc", lagTime=26)

censData <- censTrial(dataFile=simData, monitorFile=monitorData, stage1=78, stage2=156)
                       
rankData <- rankTrial(censFile=censData, idxHighestVE=2, 
                      headHead=matrix(2:1, nrow=1, ncol=2), lowerVE=0, stage1=78, 
                      stage2=156, alpha=0.05)

### alternatively, to save the .RData output file (no '<-' needed):
###
### simTrial(N=c(1400, rep(1000, 2)), aveVE=seq(0, 0.4, by=0.2), VEmodel="half", 
###          vePeriods=c(1, 27, 79), enrollPeriod=78, enrollPartial=13, 
###          enrollPartialRelRate=0.5, dropoutRate=0.05, infecRate=0.04, fuTime=156, 
###          visitSchedule=c(0, (13/3)*(1:4), seq(13*6/3, 156, by=13*2/3)), 
###          missVaccProb=c(0,0.05,0.1,0.15), VEcutoffWeek=26, nTrials=30, 
###          stage1=78, saveDir="./", randomSeed=300)
###
### monitorTrial(dataFile=
###          "simTrial_nPlac=1400_nVacc=1000_1000_aveVE=0.2_0.4_infRate=0.04.RData", 
###          stage1=78, stage2=156, harmMonitorRange=c(10,100), alphaPerTest=NULL, 
###          nonEffStartMethod="FKG", nonEffInterval=20, 
###          lowerVEnoneff=0, upperVEnoneff=0.4, highVE=0.7, stage1VE=0, 
###          lowerVEuncPower=0, alphaNoneff=0.05, alphaHigh=0.05, alphaStage1=0.05, 
###          alphaUncPower=0.05, estimand="cuminc", lagTime=26, saveDir="./")
###
### censTrial(dataFile=
###  "simTrial_nPlac=1400_nVacc=1000_1000_aveVE=0.2_0.4_infRate=0.04.RData",
###  monitorFile=
###  "monitorTrial_nPlac=1400_nVacc=1000_1000_aveVE=0.2_0.4_infRate=0.04_cuminc.RData",
###  stage1=78, stage2=156, saveDir="./")
###
### rankTrial(censFile=
###  "trialDataCens_nPlac=1400_nVacc=1000_1000_aveVE=0.2_0.4_infRate=0.04_cuminc.RData",
###  idxHighestVE=2, headHead=matrix(2:1, nrow=1, ncol=2), lowerVE=0, stage1=78, 
###  stage2=156, alpha=0.05, saveDir="./")