Back to the Top
The following message was posted to: PharmPK
Hi All,
Thanks for everyone who provided some input oral CL
(potentially confusing terminology) discussion; I am
switching gear to oral BA now. In some cases, when you
cannot calculate the elimination rate constant from
oral curve due to no appreciable drop in concentration
(flat terminal phase), IV elimination rate constant is
used to estimate AUCinf for oral. In theory, you
expect to have very similar terminal elimination
half-lives for both IV and oral routes but this may
not be true for all compounds. I would like to know if
anyone used this approach to calculate oral BA.
One may say why not using AUClast then the question
is: if the last quantifiable time point for IV and
oral is not the same, is it better to use AUC up to
the point for which we have data for both IV and oral
or just use AUClast regardless of the last
quantifiable time point?
Regards
Rostam
[Have you forgotten 'flip-flop' models? The terminal slope may reflect
absorption, especially with a 'flat terminal phase'. AUC(tlast >
infinity) should use the slowest rate constant which is not necessarily
the elimination rate constant. How about simultaneously modeling of IV
and oral data with linkage between parameters for each roa. That is,
same kel and V - db]
Back to the Top
The following message was posted to: PharmPK
Thanks David. No, I have not forgotten about flip-flop
models but the situation that I am dealing with is
more like a quick/screening PK where sampling was not
taken long enough. Regardless, I am interested to know
in calculating BA, is it better to use AUC up to the
point for which we have data for both IV and oral or
just use AUClast regardless of the last quantifiable
time point? What is your opinion?
Rostam
[I don't have an authoritative opinion but I think the same problem you
have with slow (possibly ka related) terminal slopes is going to cause
you a problem using truncated AUC. I believe (but I haven't read that
literature recently) a requirement for success with truncated AUC is
that absorption is (reasonably) complete by the time the 'last' sample
is collected. If you have slow absorption (flip-flop) then absorption
will still be an important part of the process at the last data point.
If you have IV data (I forget if you do) you could try simultaneous
IV/Oral fitting if assuming an absorption model (possibly a first order
absorption process) is reasonable - db]
Back to the Top
Rostam,
I will like to suggest to you the following approach:
1) if the terminal portion of the iv and oral curves on the semilog
plots are parallel you are entitled to use the elimination rate
constant from the iv profile to approximate the AUCinf for the oral
conc-time profile.
If condition 1) is met, use the AUCt measured up to the last measurable
level. I am under the assumption that the difference between the last
time-point in the iv conc-time profiles is not very different from the
last time-point in the oral conc-time profiles.
2) if the two terminal portions are not parallel, chances are the
absorption may influence the shape of the conc-time profiles after the
oral administration and therefore the kel from the iv dose cannot be
use approximate the AUCinf after the oral dose.
3) you can try to approximate the ratio of AUCinf between the oral and
iv administration with the ratio of AUCt estimated up to the same last
time-point in both profiles.
I will try to explain myself:
F = AUCinf(oral) / AUCinf(iv) ~ AUCt(oral) / AUCt(iv) where
t is the same in the iv and oral conc-time profiles. For simplicity I
assumed the same dose.
This may work if the concentration-time profiles after the oral
administration are not too short, which I believe is your case.
Take the results with a grain of salt anyway!!!
Do not be afraid to include concentration levels that are below LOQ.
Ask the chemist(s) to give you an estimation of the data variability
under the LOQ.
I hope this help.
radu
[Will 3) work if absorption is still taking place at t(last) or are you
assuming the t(last) is late enough that most of the absorption has
occurred - db]
Back to the Top
The following message was posted to: PharmPK
Dear Rostam,
As David has noted, absorption is probably not over at the last time
point when the oral terminal slope is greater than that for iv. Ideally,
your oral data would continue to be taken long enough to capture
clearance after absorption is complete. This may require taking samples
well beyond 24 h for some compounds. But if you already have the data,
this may not be an option.
We prefer fitting iv and oral dosage forms simultaneously when the data
are available, especially if there is any reason to suspect nonlinear
clearance and/or absorption. In the absence of iv data, oral data for
several different dose levels can sometimes provide sufficient data to
deduce bioavailability and pharmacokinetic parameters, including
multi-compartment PK.
For long absorption times, you will have significant colon absorption
(something we find is very often underestimated in the industry) unless
you're using a gastric retention delivery system, and it's likely that
the range of absorption rates in colon are different than the range in
small intestine, so it is important to fit the correct absorption model
as well as the PK parameters. GastroPlus makes this an easy task.
Walt Woltosz
Chairman & CEO
Simulations Plus, Inc. (AMEX: SLP)
1220 W. Avenue J
Lancaster, CA 93534-2902
U.S.A.
http://www.simulations-plus.com
Phone: (661) 723-7723
FAX: (661) 723-5524
E-mail: walt.aaa.simulations-plus.com
Back to the Top
The following message was posted to: PharmPK
Walt and Radu:
Thank you both for your useful comments. Walt
mentioned that absorption is probably not over at the
last time point when the oral terminal slope is
greater than that for iv. For everyone's benefit, we
are talking about declining (negative) slope here (so
bigger numbers are actually smaller). This may be not
be clear to some audience as most PK programs report
absolute values.
Rostam
PharmPK Discussion List Archive Index page
Copyright 1995-2010 David W. A. Bourne (david@boomer.org)