Hi allBack to the Top
I have a couple questions regarding non-compartmental modeling. We would
like to determine the Ka of lopinavir from steady-state PK data (PK
profiles). LPV is given orally (assumes 1st order absorption and 1st order
elimination), and all patients are at steady state. We modeled our data
using WinNonlin 4.1, non-compartmental analyses. Here are my questions:
1. Is it true that AUMC(0-infinity) = AUMC (tau, steady-state)? Just like
AUC (0-infinity) = AUC (tau)?
2. from my previous studies, MRT (oral) = MRT (iv) + 1/Ka
and also MRT (oral) = 1/K + 1/Ka
where K is the elimination rate constant and Ka the absorption rate
constant. Is it valid for me to calculate Ka from my non-compartmental
results using K=lambda z, and MRT(oral) = AUMC(tau)/AUC(tau)?
I would appreciate any advice in this matter.
Thank you!
Lillian Ting
University of British Columbia
Dear Lillian,Back to the Top
With respect tu your question:
> 1. Is it true that AUMC(0-infinity) = AUMC (tau, steady-state)? Just
like
> AUC (0-infinity) = AUC (tau)?
No, this is not true, as can be concluded from a simple numerical example. I
can send you a spreadsheet demonstrating this. If k is large, both values
are close to each other, since both profiles are close to each other. In
'normal' cases, AUC(0-infinity) is larger than AUC(tau); e.g. for k = 0.05
/h, ka = 2 /h, and tau = 24 h, the difference is about a factor 2.
Consequently, the answer to your second question is also negative.
Best regards,
Hans Proost
Johannes H. Proost
Dept. of Pharmacokinetics and Drug Delivery
University Centre for Pharmacy
Antonius Deusinglaan 1
9713 AV Groningen, The Netherlands
tel. 31-50 363 3292
fax 31-50 363 3247
Email: j.h.proost.aaa.farm.rug.nl
Hans,Back to the Top
Have to disagree with you about being able to evaluate Ka from steady-state
data using non-compartment analysis. I do agree that AUMC(0-inf) single
dose is not equal to AUMC(0-tau) at steady-state, but it is a simple
spreadsheet exercise to conduct reverse superposition on the steady-state
data to acquire single dose concentrations. From the single dose data AUC
and AUMC will give correct MRT and thus ka. There are several
assumptions/conditions however that need to be met: 1. The terminal
disposition can be determined from the steady-state data (study design may
require wash out data) 2. Absorption is first order (true for single dose
as well but if one only wants MAT the order of absorption in not
important). The reverse superposition doesn't even require data to be at
steady state as long as you have characterized the dosing interval and
terminal disposition appropriately with the sampling scheme. Again you may
or may not need wash out to get K. I think there was a paper on
this topic in J. Pharm. Sci. a number of years ago.
Art Straughn, Pharm.D.
Professor and Director Drug Research Laboratory
Department of Pharmaceutical Sciences
University of Tennessee
Memphis, TN
astraughn.aaa.utmem.edu
Hi HansBack to the Top
Thanks for the input. Could you send me the spreadsheet you mentioned so I
can take a look?
Unfortunately we only have steady-state data. Maybe there is some other
way to calculate Ka?
Thanks!
Lillian
Univeristy of British Columbia
Dear Arthur,Back to the Top
You wrote:
>"Have to disagree with you about being able to evaluate Ka from
>steady-state data using non-compartment analysis."
I did not say whether or not Ka can be calculated. I said that AUMC after
single dose and at steady state are different. This implies that the
suggested method for calculating Ka is not correct.
>"... it is a simple spreadsheet exercise to conduct reverse superposition
>on the steady-state data to acquire single dose concentrations."
Thank you for pointing to this method. I did not know it, and I will have a
look at the reference in J. Pharm. Sci. But I have my doubts on such
methods with respect to their accuracy and precision, apart from the
numerous assumptions that cannot be checked properly.
Personally I am not in favor of noncompartmental pharmacokinetics, mainly
because it does not allow a prediction of the time course of the
concentration. The latter is essential for relating dose to therapeutic
activity of a drug, a key issue in pharmacokinetics. In my view, the
advantage of not using a (compartmental) model does not counterbalance this
disadvantage. What can we learn from noncompartmental pharmacokinetics? I
posed this question in the PharmPK group some time ago, but did not learn much.
Best regards,
Hans Proost
Johannes H. Proost
Dept. of Pharmacokinetics and Drug Delivery
University Centre for Pharmacy
Antonius Deusinglaan 1
9713 AV Groningen, The Netherlands
tel. 31-50 363 3292
fax 31-50 363 3247
Email: j.h.proost.at.farm.rug.nl
Hans,Back to the Top
Sorry, I just meant to disagree with the implications (not the statement),
that since AUMC(0-inf) single dose is not the same as AUMC(0-Tau), then Ka
could not be determined using non-compartment analysis with ss data. I am
not a fan of "connect-the-dots" for AUMC determinations anyway because of
the extensive extrapolation (usually 20% or more) when compared to
extrapolation of AUC (5 to 10%). As you point out, questions about
precision and accuracy are brought into the picture. This though, should not
distract from the value these kinetic parameters have had related to
absorption and disposition modeling.
As far as which is better: "connect-the-dots" (non-compartment) vs.
"minimizing SSE"(compartment), it depends on what "truth" you want to glean
from the data. Generally, regulatory truth requires the use of
"connect-the-dots", while intellectual truth mandates the use of "Minimize
SSE". The real truth is probably somewhere in between. Please note: I am
not implying regulators only accept "connect-the-dots" analysis as fondly
referred to by Nick H. as Mickey Mouse Kinetics.
Art Straughn, Pharm.D.
Professor and Director
Drug Research Laboratory
University of Tennessee
874 Union Ave
Suite 5P
Memphis, TN 38163
E-mail: ASTRAUGHN.at.UTMEM.EDU
Phone: (901) 448-6033
Fax: (901) 448-6940
Hi ArthurBack to the Top
Thanks for your suggestion. Could you send me that spreadheet that does
reverse superposition on steady-state data please? I didn't know about
this method and would like to give it a try.
I only have steady-state data, and no wash-out period in this
study. Hopefully we can still utilize such data.
Thanks a lot!
Cheers,
Lillian
University of British Columbia
PharmPK Discussion List Archive Index page
Copyright 1995-2010 David W. A. Bourne (david@boomer.org)