## ABEL vs ABE, randomization [🇷 for BE/BA]

Hi Bebac user,

❝ For a drug that has CV = 25%, we need to design a full replicated study

CV 25% of what? Unusual that Cmax and AUC have the same one. If your authority accepts reference-scaling for Cmax, generally AUC drives the sample size because you have to use conventional ABE for it.

❝ To estimate the sample size, which one is the right to use in R

❝ SampleN.tost OR sampleN.scabel??

None would work. is case-sensitive. You are interested in sampleN.TOST() and sampleN.scABEL().

If really both metrics have 25% (please check), chances are pretty low that ABEL can be applied (only if CVwR > 30%):

library(PowerTOST) sampleN.TOST(CV = 0.25, design = "2x2x4") +++++++++++ Equivalence test - TOST +++++++++++             Sample size estimation ----------------------------------------------- Study design: 2x2x4 (4 period full replicate) log-transformed data (multiplicative model) alpha = 0.05, target power = 0.8 BE margins = 0.8 ... 1.25 True ratio = 0.95,  CV = 0.25 Sample size (total)  n     power 14   0.813985 sampleN.scABEL(CV = 0.25, theta0 = 0.95, design = "2x2x4") +++++++++++ scaled (widened) ABEL +++++++++++             Sample size estimation    (simulation based on ANOVA evaluation) --------------------------------------------- Study design: 2x2x4 (4 period full replicate) log-transformed data (multiplicative model) 1e+05 studies for each step simulated. alpha  = 0.05, target power = 0.8 CVw(T) = 0.25; CVw(R) = 0.25 True ratio = 0.95 ABE limits / PE constraint = 0.8 ... 1.25 EMA regulatory settings - CVswitch            = 0.3 - cap on scABEL if CVw(R) > 0.5 - regulatory constant = 0.76 - pe constraint applied Sample size search  n     power 12   0.7659 14   0.8266

Therefore, you would need the same sample size, though you would achieve a slightly higher power with ABEL.

❝ I am using a RandomizBE to make randomization but I want to divide them into period 1, period 2 and so on. How?

Extract the treatment from the sequence and create additional columns:

library(randomizeBE) n       <- 14 seqs    <- c("TRTR", "RTRT")   # any crossover or replicate design pers    <- nchar(seqs)      # number of periods periods <- paste0("p", 1:pers) # column names random  <- RL4(nsubj = n, seqs = seqs, randctrl = TRUE) random$rl[ , periods] <- "" # empty period columns for (i in 1:pers) { # extract treatment / period random$rl[i + 3] <- substring(random$rl$sequence, i, i) } print(random) Randomization table          created: 2022-06-04 15:32:05 (seed: 1892562 blocksize: 4 4 4 2 )  subject seqno sequence p1 p2 p3 p4        1     2     RTRT  R  T  R  T        2     1     TRTR  T  R  T  R        3     2     RTRT  R  T  R  T        4     1     TRTR  T  R  T  R        5     2     RTRT  R  T  R  T        6     1     TRTR  T  R  T  R        7     2     RTRT  R  T  R  T        8     1     TRTR  T  R  T  R        9     2     RTRT  R  T  R  T       10     2     RTRT  R  T  R  T       11     1     TRTR  T  R  T  R       12     1     TRTR  T  R  T  R       13     1     TRTR  T  R  T  R       14     2     RTRT  R  T  R  T

Dif-tor heh smusma 🖖🏼 Довге життя Україна! Helmut Schütz The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes  Ing. Helmut Schütz 