Bioequivalence and Bioavailability Forum

❝ Can we include outlier test …

❝ … in US, EMA protocols for 2 way crossover and replicate design studies.

• US: Generally no. However, similar to the EMA’s conditions for ABE.
  
  
• EMA: for ABE (any crossover and replicate designs)
  In principle any reason for exclusion is valid provided it is specified in the protocol and the decision to exclude is made before bioanalysis.
  Examples of reasons to exclude the results from a subject in a particular period are events such as vomiting and diarrhoea which could render the plasma concentration-time profile unreliable. In exceptional cases, the use of concomitant medication could be a reason for excluding a subject.
  Exclusion of data cannot be accepted on the basis of statistical analysis […] alone, because it is impossible to distinguish the formulation effects from other effects influencing the pharmacokinetics.
  The exceptions to this are:
  1. A subject with lack of any measurable concentrations or only very low plasma concentrations for reference medicinal product. A subject is considered to have very low plasma concentrations if its AUC is less than 5% of reference medicinal product geometric mean AUC (which should be calculated without inclusion of data from the outlying subject). The exclusion of data due to this reason will only be accepted in exceptional cases and may question the validity of the trial.
     
  2. Subjects with non-zero baseline concentrations > 5% of C_max. Such data should be excluded from bioequivalence calculation.

EMA: for ABEL only
As for ABE. Additionally:
The applicant should justify that the calculated intra-subject variability is a reliable estimate and that it is not the result of outliers.

A statistical test is not acceptable. Box plots of studentized model residuals (in the estimation of \(\small{CV_\textrm{wR}}\), not the T vs R comparison) are preferred. Seemingly a fence of 2 is acceptable. If outliers are detected, \(\small{CV_\textrm{wR}}\) has to be recalculated after exclusion – which will lead to less expansion! However, the 90% CI has to be calculated from the entire data. On a case-by-case basis a study might (‼) be accepted, if both analyses pass ABEL. Personal comment: So what?
For details see the vignette of the package replicateBE and the Q&A-document of the EGA/EMA (London, June 2010).