Back to the Top
The following message was posted to: PharmPK
I have a question on missing samples in bioequivalence studies. We were
conducting a bioequivalence study on a drug with a long half-life
hrs). in the second period the samples at 72, 96, 120, and144 hours
collected due to snowing where the roads were closed and no volunteers
showed up at the study center. However the data from period 1 showed
the terminal phase in the conc-time profile started at 24 hours which
enabled us to use at least 2 data points before the missing samples and
after them with values for R^2 exceeding 0.95 in the 24 volunteers.
My questions are
1) Is there a guidance for handling such case
2) Is the value of R^2 and similar values for the half-lives in the two
periods enough to justify the missing of 4 samples in all volunteers
reference and 12 on test).
3) How to approach this problem when dealing with regulators.
Back to the Top
To my knowledge there is no guidance on this issue.
My advice is as follows:
I understood there are two more sampling times after the 144 hours and
these data points were used together with two data points before the
gap (probably 24 and 48 hours) to estimate the apparent elimination
rate constant in the second period. If this is true than the issue of
missing data points may be defended. The similarity between the
half-lives estimated within each subject in the two periods may be used
as an argument for the acceptability of the AUCinf data.
Currently, truncated AUC is an acceptable metrics. Therefore, an
analysis of AUC0-48 should be performed and the contrast of this
parameter between periods should be used to prove bioequivalence in
addition to the existent AUCt and AUCinf.
There is no guarantee that the reviewer will accept these results but
if this is only one study out of more within the submission you have a
PharmPK Discussion List Archive Index page
Copyright 1995-2010 David W. A. Bourne (firstname.lastname@example.org)