Back to the Top
Hi:
In a rat iv dosed study, we are getting a higher AUC-all than AUC-inf,
using WinNonlin, non-compartmental modelling. Can someone please
explain the possible reasons behind this? The results do show a
distribution (alpha) and ellimination (beta) phase. I have gone
through the graph posted by Dr. Aziz Karim recently showing the
differences between AUC-last, AUC-all and AUC-inf. Based on that
graph, is there any way WinNonlin can figure out whether the last
sampling result in the terminal phase may be below the lqc? If this is
possible, then the program can take upto the last sampling result above
the lqc in the terminal phase and extrapolate from there to derive the
AUC-inf. This AUC-inf would then be lower than the AUC-all.
Thanks
Ananda
Back to the Top
Dear Ananda:
For assessment of single dose PK parameters I would recommend not using
AUC
(all). Just use AUC (lqc) and AUC (inf) as the parameters of total
exposures
using the most sensitive assay available. Also I would not use AUC (inf)
value if there is an unacceptable error in determining lambda_z value
(used
for deriving extra AUC beyond AUC (lqc) or AUC (inf) value is greater
than
20% of the AUC (lqc) value.
The only place where AUC (all) value is useful is in determination of
steady
state AUC within one dose interval especially when a concentration value
before the end of the dose interval (tau) is below lqc.
In the latest version of the WinNonlin software you can choose which PK
parameters you want to appear as an output. For single dose PK
parameter I
simply do not use AUC (all).
With regards and best wishes,
Aziz Karim
Back to the Top
Dear all,
once a year (just my 'educated guess') there is a new discussion about
this
topic.
[That is what the searchable archive is for ;-) See
http://www.boomer.org/pkin/simple.html for the search page - db]
I believe AUC-all is only an 'invention' specific to WinNonlin. I do
not know a
single peer-reviewed article mentioning AUC-all (any hints appreciated).
AUC-last has it's merits in bioequivalence testing, where (assuming
sensitive
analytics and proper sampling time points) extrapolating just adds
variation
whitout gaining additional information (since the main target is a
comparison of
the absorption process). In the EU AUC-last actually is the primary BE
parameter.
[I thought there has been some literature related to AUC truncated,
thus might not AUC-last better refer to a particular time point - db]
If one is interested in doing noncompartmental pharmacokinetics,
AUC-inf is the
adequate parameter (always having one eye on the magnitute of
extrapolated
part...).
Can anybody please give a practical use of AUC-all.
Regards,
Helmut Schütz
Head Biometrics
Biokinet GmbH
Nattergasse 4
A-1170 Vienna/Austria
mailto:helmut.schuetz.-at-.chello.at
Back to the Top
Enclosed some information specifically regarding WinNonlin to build on
the excellent summary previously presented by Dr. Aziz Karim on AUCall
vs AUCinf......
With the introduction of WinNonlin v4.0, a data transformation tool
(e.g. BQL wizard) was added to WinNonlin's functionality. This tool
allows the user to easily automate transformation of the entire dataset
prior to analysis through the use of a wizard-based approach interface.
Prior to version 4.0, the replace and exclude functions are typically
used to substitute zeros or exclude BQL (and other character codes)
prior to analysis. The replace function with a tolerance equal to the
LLOQ can be used to set numeric values below the LLOQ to missing or
zero. WinNonlin v4.0 automates this procedure and allows substitution
of character codes to be conditional upon the sampling time (i.e.
before or after tmax) and sample sequence (i.e. consecutive BQLs). In
addition, an LLOQ can be specified to perform the same substitution
rule to numeric values below the LLOQ. Furthermore, WinNonlin v4.0
allows these substitution rules to be saved, loaded, and applied to new
datasets.
For further technical support regarding WinNonlin or any of our other
products please feel free to contact our support inbox at: www.
pharsight.com/support/index.php
Thanks,
Jeff D. Fischer
Pharsight
PharmPK Discussion List Archive Index page
Copyright 1995-2010 David W. A. Bourne (david@boomer.org)