Bioequivalence and Bioavailability Forum

Hi stic-stats,

❝ ❝ […] Since AUC₇₂ is a primary PK metric consider hospitalizing subjects in the future. By opting for ambulatory sampling you saved some rupees but risked to substantially loose power.

❝

❝ May I know what type of power loose.

Fewer eligible subjects will always decrease power. To which extent is unclear (mmw posted more than five years ago) and without the CV and sample size we would need a crystal ball.
Example (2×2 crossover, T/R 0.95, target power 0.80):

library(PowerTOST)
CV  <- seq(20, 40, 5) # intra-subject CV in percent
dor <- seq(0, 15, 5)  # dropout-rate in percent
res <- data.frame(CV = rep(CV, each = length(dr)), n = NA, dor = dor,
                  elig = NA, power = NA)
for (j in 1:nrow(res)) {
  tmp <- sampleN.TOST(CV = res$CV[j]/100, print = FALSE, details = FALSE)
  res$n[j] <- tmp[["Sample size"]]
  if (res$dor[j] == 0) {
    res$elig[j]  <- res$n[j]
    res$power[j] <- tmp[["Achieved power"]]
  } else {
    res$elig[j]  <- floor(tmp[["Sample size"]]*(100-res$dor[j])/100)
    res$power[j] <- suppressMessages(power.TOST(CV = res$CV[j]/100,
                                                 n = res$elig[j]))
  }
}
res$power <- signif(res$power, 5)
names(res)[1:4] <- c("CV (%)", "dosed", "dropouts (%)", "eligible")
print(res, row.names = FALSE)


Gives:

 CV (%) dosed dropouts (%) eligible   power
     20    20            0       20 0.83468
     20    20            5       19 0.81324
     20    20           10       18 0.79124
     20    20           15       17 0.76365
     25    28            0       28 0.80744
     25    28            5       26 0.77606
     25    28           10       25 0.75766
     25    28           15       23 0.71729
     30    40            0       40 0.81585
     30    40            5       38 0.79533
     30    40           10       36 0.77239
     30    40           15       34 0.74666
     35    52            0       52 0.80747
     35    52            5       49 0.78311
     35    52           10       46 0.75583
     35    52           15       44 0.73545
     40    66            0       66 0.80525
     40    66            5       62 0.77978
     40    66           10       59 0.75824
     40    66           15       56 0.73466

❝ Is their any impact on calculation of AUC due to number of missed sample?

Yes. Since generally studies are planned for the PK-metric with the highest variability (likely C_max) the overall impact on power should be limited.

❝ Could these subjects be excluded from statistical analysis? If yes, on what basis we exclude these subjects?

Everything is possible – if stated in the statistical protocol. You could exclude them (only AUC whilst keeping C_max), use an estimate (if λ_z is reliable), use the AUC to the last time point where you have concentrations for both T and R,…
Please search the forum; lots of posts.