- Published 1999 in Pharmaceutical Research

Purpose. 1. To determine properties of the estimated variance component for the subject-by-formulation interaction (σ2 D) in investigations of individual bioequivalence (IBE), and 2. to evaluate the prevalence of interactions in replicate-design studies published by FDA. Methods. Four-period crossover studies evaluating IBE were simulated repeatedly. Generally, the true bioequivalence of the two formulations, including σ2 D= 0, was assumed, σ2 D was then estimated in a linear mixed-effect model by restricted maximum likelihood (REML). The same method was applied for estimating σ2 D for the data sets of FDA. Results. 1.σD estimated by REML was positively biased. The bias and dispersion of the estimated σDincreased approximately linearly with the estimated within-subject standard deviation for the reference formulation (σWR). Only a small proportion of the estimated σD exceeded the estimated σWR. 2. Distributions of the estimated σD were evaluated. At σWR = 0.30, a level of estimated σD= 0.15 was exceeded, by random chance, with a probability of about 25%. 3. Importantly, the behaviour of the σ2 D values estimated from the FDA data sets was similar to that exhibited by the simulated estimates of σ2 D which were generated under the conditions of true bioequivalence. Conclusions. 1. σD estimated by REML is biased; the bias increases proportionately with the estimated σWR. Consequently, exceeding a fixed level of σD (e.g., 0.15) does not indicate substantial interaction. 2. The data sets of FDA are compatible with the hypothesis of σ2 D = 0. Consequently, they do not demonstrate the prevalence of subject-by-formulation interaction. Therefore, it could be sufficient and reasonable to evaluate bioequivalence from 2-period crossover studies.

@article{Endrnyi1999SubjectbyFormulationII,
title={Subject-by-Formulation Interaction in Determinations of Individual Bioequivalence: Bias and Prevalence},
author={L{\'a}szl{\'o} Endr{\'e}nyi and Laszlo Tothfalusi},
journal={Pharmaceutical Research},
year={1999},
volume={16},
pages={186-190}
}