Back to the Top
Hi everybody,
I have a problem with dealing the concentration at 0 hour. After the
assay of plasma, we found that one C-T curve of the second period
( first period 20mg / second 10mg) had a concentration above LLOQ at
0 hour. But both the DB and BK of that curve had no concentration.
The concentration is 21.8ng/ml and Cmax of that person in the same
period is 1360ng/ml. No re-tests have been done for that person. I am
not sure how to deal with this curve when calculating PK parameters.
Should I drop out this curve or replace the concentration with 0? Is
there any regulation, guidance or SOP about this that I can follow?
Thanks a lot!
Maurice
Back to the Top
Maurice-
Regarding how to handle a measurable but low value (compared to
Cmax) in a predose sample after a previous dosing to the individual:
If one concentration vs time profile in the second period of dosing
had a predose value that was above the LLOQ, there are several
possible explanations. I am assuming that the same person received
the 20-mg dose and subsequently received the 10-mg dose. Some
possible explanations are:
Inadequate time to clear all of the first dose.
Contamination of the sample during the collection period (permanent
contamination).
Contamination during the analytical process (contamination of the
extract but not the remainder of the sample).
An interfering chromatographic peak that produces a response read as
analyte.
The value is near the LLOQ and is noise due to statistical
fluctuations of background peaks.
A metabolite is present that can revert to parent.
If inspection or analysis of other C-T profiles shows that inadequate
washout is possible, then the value is likely to be accurate and not
an artifact of contamination. Keep the value and calculate PK
parameters according to established techniques.
If inadequate washout is not readily apparent, then you can choose to
reanalyze the sample (or an aliquot of it) if sufficient sample
remains. For reanalysis, you ideally should have an SOP in place
that triggers reanalysis of a predose or control sample that is found
to give a measurable amount of drug. If not, you will need to
document the reason for reanalysis very clearly.
If the value is not near the LLOQ and the reanalyzed value is
substantially lower (e.g., undetectable or below the quantification
limit), then you will also need an SOP or rule in place for selecting
a value to report. Often in such cases the sample is reanalyzed in
duplicate which provides more evidence that the reanalysis value is
correct.
If reanalysis confirms the initial value, then either the value is
real or the sample is permanently contaminated..
You may be able to justify elimination of the value if the previous
sample from the same individual during the first period was
undetectable or substantially less than the value in question,
especially if it was undetectable for two or more timepoints.
Finally, if you find no compelling reason to "censor" that value,
then you will need to use it, rather than eliminate it just because
you don't like it. Since the questionable value (21.8) is only 1-2%
of the Cmax (1360), it will probably not affect most calculated
parameters in any meaningful way. All things considered, this is
often the best approach.
Tom
Thomas L. Tarnowski, Ph.D.
Elan Pharmaceuticals
800 Gateway Boulevard
South San Francisco, CA 94080
ttarnowski1.aaa.aol.com
Want to post a follow-up message on this topic?
If this link does not work with your browser send a follow-up message to PharmPK@boomer.org with "Concentration at 0 hour" as the subject | Support PharmPK by using the |
Copyright 1995-2011 David W. A. Bourne (david@boomer.org)