eselect {eselect}

Endpoint selection and sample size reassessment for composite endpoints based on blinded data

Description

Endpoint selection and sample size reassessment for composite endpoints based on blinded data. The composite endpoint is assumed to be a binary endpoint formed by a combination of two events (E1 and E2). We assume that the endpoint 1 is more relevant for the clinical question than endpoint 2. This function selects between the composite endpoint or the relevant endpoint as the primary endpoint of the study and recalculate the sample size accordingly. The decision criteria to decide between the composite endpoint or the relevant endpoint might be the ratio of the corresponding sample sizes ("SS") or the Asymptotic Relative Efficiency ("ARE"). The algorithm of the function is the following: First, the probabilities of the composite components in the control group and the correlation between them are estimated based on blinded data. Second, using the estimated probabilities and the estimated correlation, the decision criteria is computed and the primary endpoint is selected. Finally, the sample size is recalculated according to the decision.

Usage

eselect(db, p0_e1, OR1, p0_e2, OR2, criteria = "SS", alpha = 0.05, beta = 0.2)

Arguments

db

matrix 2x2 table (pooled sample)

p0_e1

numeric parameter, probability of occurrence E1 in the control group

OR1

numeric parameter, Odds ratio for the endpoint 1

p0_e2

numeric parameter, probability of occurrence E2 in the control group

OR2

numeric parameter, Odds ratio for the endpoint 2

criteria

decision criteria to choose between the composite endpoint or the endpoint 1 as primary endpoint ("SS": Ratio sample sizes, "ARE": Asymptotic Relative Efficiency).

alpha

Type I error.

beta

Type II error.

Value

This function returns the decision (Decision = 1, meaning the chosen endpoint is the composite endpoint; and Decision = 0, meaning the chosen endpoint is the relevant endpoint) and the sample size according to the decision.

References

Bofill Roig, M., Gómez Melis, G., Posch, M., & Koenig, F. (2022). Adaptive clinical trial designs with blinded selection of binary composite endpoints and sample size reassessment. Biostatistics (in press). arXiv e-prints, arXiv-2206 (https://doi.org/10.48550/arXiv.2206.09639). Bofill Roig, M., & Gómez Melis, G. "Selection of composite binary endpoints in clinical trials." Biometrical Journal 60.2 (2018): 246-261.

Examples

# Based on Bofill Roig, M., et al.
# (See supplementary material in https://doi.org/10.48550/arXiv.2206.09639)
p0_e1 = 0.173
p0_e2 = 0.055
p1_e1 = 0.121;
p1_e2 = 0.057;
n1 = 569
n0 = 576
n = n0+n1
p1 = (p0_e1*n0 + p1_e1*n1)/n
p2 = (p0_e2*n0 + p1_e2*n1)/n
p_ce = (0.203*n0 + 0.146*n1)/n
OR1 = 0.7
OR2 = 0.9
x11 = round((p1+p2-p_ce)*n)
x12 = round((p1)*n-x11)
x21 = round((p2)*n- x11)
x22 = round((1-p_ce)*n)
data = matrix(c(x11,x12,x21,x22), nrow = 2 , ncol = 2, byrow = FALSE)
eselect(db=data,p0_e1=0.18,OR1=0.70,p0_e2=0.05,OR2=0.9,criteria="SS",alpha=0.05,beta=0.2)