Back to the Top
Dear members,
I have a question regarding "the drug is absorbed into blood,
absorption into systemic circulation" What blood is it? i.e. vain blood
or artery blood is drug absorbed into and why?
Thanks for your instruction.
Tian
Back to the Top
Dear Walt,
Thank you for your extensive reply. A few comments:
> But in the study we are working on now,
> some of the first nonzero data points for the oral doses are many
hours
> after any iv sample would be above LOD.
The characteristics of the drug in the study you are working on now seem
rather unusual (but not unlikely). If clearance from an iv dose is
indeed so
rapid, it may be possible that EHC does not result in a measurable
second
peak, at least not using conventional sampling schemes. In this case
'apparent bioavailability' as obtained from AUC ratio may exceed 1.
Indeed,
it may be useful to talk about F>1 'in the sense that it would tell us
something useful about the drug'. As long as we do not overinterpret
such a
result, and do not forget the real story behind the values, there is no
real
problem.
> I'm interested in your statement that AUC is not affected by EHC
(would
> this also include EEC?). I'm guessing that you are coming from a
> perspective of fitting data.
No, I did not come from a perspective of fitting data. Talking about
AUC I
mean the 'true' AUC. Of course, in practice AUC must be estimated by
some
method, and this will always introduce errors. But if the sampling
scheme is
appropriate ('good design') and the method for estimating AUC is
appropriate
(e.g. trapezoidal rule, or fitting if a 'good model' is chosen), these
errors should be moderate, and bias should be small.
> Now the drug that we
> collected before has an opportunity to get back into the system. Will
> AUC change? Yes, it will increase significantly. So does EHC change
AUC?
> I think it does if the question is interpreted as meaning if all else
is
> the same, does a molecule that undergoes EHC have a different AUC than
> its near twin that does not.
This is correct. However, as I stated earlier, one should compare the
AUC in
case of EHC with the situation without EHC (similar clearance, different
'apparent' volume of distribution; enterohepatic recirculation does not
affect AUC), and not with the situation where the drug disappears in the
same way as during EHC but without returning; in the latter case the
clearance would be much higher, of course.
In your example, the two drugs differ with respect to EHC, but also with
respect to clearance. The drug undergoing EHC is not yet eliminated by
clearance, but it is undergoing a complicated distribution process. Of
course one is free to make other definitions, and to consider any drug
excreted into the bile as eliminated. Then indeed EHC affects AUC, but
the
relationship between clearance and AUC (F x Dose = CL x AUC) is lost,
unless
one accepts that F can exceed 100%, also for intravenous doses. But, to
be
short, I do not like this kind of Mickey Mouse play, and prefer to keep
to
convention whenever possible.
> Again I ask, what is the reason we talk about fraction absorbed?
Please note that in my interpretation 'fraction absorbed' refers to the
fraction of the dose reaching the systemic circulation, thus identical
to
'biovailability'. This fraction can never exceed 100%. Only if it is
calculated using inappropriate data (as in your example; again I
understand
that it is not feasible to get the appropriate data, but that is not the
discussion here), an 'apparent' value >100% can be obtained. I am not
interested in counting molecules passing, but on plasma concentration.
And
for the latter we need 'bioavailability' and not counting molecules.
> If we go to a reference
> some years after an experiment and see Fa = 90% or Fa=108%, which
would
> be more enlightening?
I don't understand your point.
> Absent absorption,
> the nonlinear clearance based on Cp would have a certain value. When
the
> new drug is added by absorption, the clearance increases (if not
> flow-limited). We propose that this difference in clearance is
> effectively the first pass extraction of the newly absorbed drug, and
> that this can be integrated over time to calculate a net first pass
> extraction, and with that known, the bioavailability.
I do not fully understand the procedure:
1) If the concentration increases, clearance decreases (not: increases)
due
to saturation of drug transporters or metabolizing enzymes.
2) From the text I understand that 'clearance' here refers to 'rate of
hepatic drug elimination'. If so, how do you measure this?
3) How can you estimate 'clearance' (by definition #1 or #2) in case of
'no
absorption'?
A further explanation using a numeric example would be helpful. For the
moment, my eyebrows keep raising.
Best regards,
Hans
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.-a-.farm.rug.nl
Back to the Top
Dear Tian,
Your question:
> I have a question regarding "the drug is absorbed into blood,
> absorption into systemic circulation" What blood is it? i.e. vain
blood
> or artery blood is drug absorbed into and why?
Drug absorbed from the gastrointestinal tract is passed mainly via the
portal vein to the liver, and then enters the 'systemic circulation'.
How
this 'systemic circulation' is defined is a good question. I would say
that
this consists of all blood in the body, but excluding the drug
molecules in
the blood in the gut, portal vein, and liver that have not yet passed
the
liver once (i.e. that entered the blood stream during the passage of the
gut). Of course this definition is not perfect, and I would appreciate
if
someone could provide a better one.
So, systemic circulation consists of both venous and arterial blood.
This
does not imply, however, that the drug concentration is the same in
venous
and arterial blood, or plasma. In fact, blood (or plasma) concentration
may
be different in the whole vascular bed, due to (1) incomple mixing after
administration, and (2) distribution and elimination in particular
organs
and tissues. In particular shortly after administration and during
absorption, blood concentrations in different parts may be different.
During
the elimination phase these differences are usually relatively small.
So,
drug concentrations in blood are both site- and time-dependent.
Since drug concentrations are usually measured in a peripheral vein,
e.g. in
the arm, this site is the point of reference. And in most cases we
simply
assume that this is 'the blood (or plasma) concentration'. This is never
completely true, but in many cases sufficiently close to the real
situation.
In case of studies within a very short time period, e.g. as for
anaesthetic
drugs, arterial concentrations are often measured, in particular for
PK-PD
studies, if an arterial line is needed. But in almost any other case, a
peripheral vein is used for sampling.
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
Back to the Top
Dear Hans,
Thank you again for your comments:
Hans: "If clearance from an iv dose is indeed so rapid, it may be
possible that EHC does not result in a measurable Second peak, at least
not using conventional sampling schemes. In this case 'apparent
bioavailability' as obtained from AUC ratio may exceed 1. Indeed, it may
be useful to talk about F>1 'in the sense that it would tell us
something useful about the drug'. As long as we do not overinterpret
such a result, and do not forget the real story behind the values, there
is no real problem."
I agree - but it's not always that the iv clearance is so rapid. It may
simply be that the oral absorption is so spread out in time compared to
the iv data. With slow absorption due to solubility/dissolution/release
limitations, this is not at all uncommon. Also, EHC does not always
result in an observed second peak (sampling times?) and EEC never does.
One problem is that we often don't suspect either until after the data
are taken, and you know how expensive it is to rerun trials, so we have
to work with the data we have.
Hans: "Talking about AUC I mean the 'true' AUC. Of course, in practice
AUC must be estimated by some method, and this will always introduce
errors. But if the sampling scheme is appropriate ('good design') and
the method for estimating AUC is appropriate (e.g. trapezoidal rule, or
fitting if a 'good model' is chosen), these errors should be moderate,
and bias should be small."
Then what do you do when you use the trapezoidal rule or a good model
and the AUC ratio to an iv dose results in a calculated F>1? The AUC is
the AUC, so if the data, taken as best the real world allows, results in
a higher AUC/dose for oral than for iv, how should we proceed? The good
design argument is fine for academia, but with controlled release doses
and others where solubility and/or dissolution are rate-limiting, we
simply cannot get data that is ideal. It's not bad design - it's the
constraints put upon us by Mother Nature and economics.
Hans: ". . . one should compare the AUC in case of EHC with the
situation without EHC (similar clearance, different 'apparent' volume of
distribution; enterohepatic recirculation does not affect AUC), and not
with the situation where the drug disappears in the same way as during
EHC but without returning; in the latter case the clearance would be
much higher, of course. In your example, the two drugs differ with
respect to EHC, but also with respect to clearance. The drug undergoing
EHC is not yet eliminated by clearance, but it is undergoing a
complicated distribution process."
I guess this is another matter of definitions - is it cleared and then
reabsorbed or is it still in the system? I have heard it said many times
that the gastrointestinal tract is effectively outside the body. If so,
then drug secreted in bile, or exsorbed from the enterocytes, could be
considered as having left the body. If it reenters, then it is absorbed
again.
I have no strong preference either way, and I am not out to change the
industry, but I do think there is a value in treating the system in a
way that elucidates what is going on at a mechanistic level. The reason
I asked why we talk about F and Fa was to get to the basics - what are
we trying to communicate with these terms? Does limiting them to <=1
serve that purpose better, or does allowing them to be >1 serve that
purpose better?
Hans: "Of course one is free to make other definitions, and to consider
any drug excreted into the bile as eliminated. Then indeed EHC affects
AUC, but the relationship between clearance and AUC (F x Dose = CL x
AUC) is lost, unless one accepts that F can exceed 100%, also for
intravenous doses. But, to be short, I do not like this kind of Mickey
Mouse play, and prefer to keep to convention whenever possible."
Mickey Mouse? Hey, Mickey's a neighbor of mine here in southern
California, so be nice!
Seriously, I don't see anything Mickey Mouse about it. It seems to me
that it's simply a matter of definitions.
As I asked in an earlier post, suppose with EHC the trapezoidal rule or
good model calculates F=0.99. Nobody gets excited, and this is what
would be published (even though the "real" AUC might only produce
F=0.8). But if the result produces F>1, then people get excited. It
seems to me that there is no harm in F>1, or Fa>1, as long as we
understand how the calculations were made and how to use the values
going forward. In fact, it seems that reporting them as >1 provides more
information than masking the effect with some kind of additional
peripheral compartment for oral doses but not for iv doses, or to say
that the clearance is different between iv and po doses.
Hans: "Please note that in my interpretation 'fraction absorbed' refers
to the fraction of the dose reaching the systemic circulation, thus
identical to 'biovailability'."
I'm sorry, but this is not the accepted definition of absorption by the
FDA, nor of any pharmaceutical company with which we've worked. In
general, F<=FA, not F=Fa. For oral doses, absorption is defined (by the
FDA, Gordon Amidon, and most of the industry) as drug entering the
apical membrane of the enterocytes. Not necessarily even reaching the
portal vein, much less the systemic circulation -- certainly not
bioavailability. In older references (e.g., Wagner-Nelson), the term
"absorption" was used to mean bioavailability, but that is no longer the
accepted standard.
Hans: "This fraction can never exceed 100%. Only if it is calculated
using inappropriate data (as in your example; again I understand that it
is not feasible to get the appropriate data, but that is not the
discussion here), an 'apparent' value >100% can be obtained. I am not
interested in counting molecules passing, but on plasma concentration.
And for the latter we need 'bioavailability' and not counting
molecules."
Actually, the inability to get appropriate data was precisely my
original point - because the real world does not allow obtaining perfect
data, we have to have methods to deal with the data we have. If you're
interested in plasma concentration, then you're interested in molecules
- I don't see how you can separate them.
Hans: "> If we go to a reference some years after an experiment and see
Fa = 90% or Fa=108%, which would be more enlightening?< I don't
understand your point."
My point is that there is information in Fa=108% (from my example a few
posts back) that is not provided if we adjust the AUC and report it as
90% (for the same data). Suppose you pick up a 10-year-old reference on
a drug, and it says Fa=90%, and F=108%. That would tell you something is
going on that is a bit unusual. So reporting F and/or Fa > 1 could
provide useful information.
"Absent absorption, the nonlinear clearance based on Cp would have a
certain value. When
the new drug is added by absorption, the clearance increases (if not
flow-limited). We propose that this difference in clearance is
effectively the first pass extraction of the newly absorbed drug, and
that this can be integrated over time to calculate a net first pass
extraction, and with that known, the bioavailability."
(I'm not sure if Walt meant for this paragraph to be quoted - db)
Hans: "I do not fully understand the procedure: 1) If the concentration
increases, clearance decreases (not: increases) due to saturation of
drug transporters or metabolizing enzymes."
I probably used the word clearance incorrectly - I think I should have
said rate of drug elimination.
At some point in time, if the both the hepatic artery (systemic plasma)
and portal vein plasma have concentrations of 10 mcg/ml, then when they
mix, the concentration is still 10 mg/L. If the clearance is 1 L/h, then
drug would be removed from the system at the rate of 10mg/h.
Now, if the portal vein plasma concentration is increased to 20 mg/L by
absorption, while the hepatic artery plasma remains at 10 mg/L, then the
concentration of the mixture becomes approximately 0.75* 20 + 0.25*10 17.5 mg/L. If the clearance is 1 L/h, then drug is removed at the rate
of 17.5 mg/h.
The difference of 7.5 mg/h is a result of the first pass of the
additional 10 mg/L of portal vein concentration. Suppose the clearance
decreases to 0.9 L/h because of saturation. Then the rate of drug
elimination would be 0.9*17.5 = 15.75 mg/h, and the additional 5.75 mg/h
would be due to first pass extraction. (Clearance would have to become
less than 10/17.5 = 0.57 L/h to result in a lower elimination of drug
from the system.)
That's what I meant when I said clearance was increased - I should have
said elimination was increased. With nonlinear clearance, it is a matter
of doing this calculation at each time step and integrating the
additional elimination over time to get to a net first pass extraction,
which, combined with Fa, allows calculating the bioavailability.
Hans: "2) From the text I understand that 'clearance' here refers to
'rate of hepatic drug elimination'. If so, how do you measure this?"
I stand corrected - elimination is correct. It is fitted/estimated, like
other pharmacokinetic parameters, as described above. (We, of course,
use GastroPlus.)
Hans: "3) How can you estimate 'clearance' (by definition #1 or #2) in
case of 'no absorption'? A further explanation using a numeric example
would be helpful. For the
moment, my eyebrows keep raising."
In the case of no absorption (e.g., iv dose, or absorption is complete),
there is no first pass extraction in liver. The concentration entering
the liver via hepatic artery and portal vein are the same (assuming
complete mixing, etc.), and clearance is calculated using
Michaelis-Menten kinetics for each metabolic pathway, or using a fixed
clearance model if the behavior is linear. There would be no
contribution to the net first pass extraction at that point. First pass
extraction is >0 during oral absorption, and =0 when portal vein and
hepatic artery concentrations are the same. Again, this assumes liver
and gut are the only clearing organs - the contribution of gut
metabolism to first pass extraction would have been calculated after
calculating absorption (drug entering enterocytes) but prior to
calculating the portal vein concentration.
So, if the purpose of Fa is to tell us how much of the drug in a dosage
form was absorbed from the lumen, then if 80 mg of a 100 mg dose was
absorbed, Fa=80%, regardless of how many times it recirculates. If the
purpose of Fa is to say what is the ratio of the number of mg of drug
that entered the enterocytes to the number of mg in the dose, then if 80
mg of a 100 mg dose was absorbed, and 25 mg of that was absorbed again
via recirculation, then Fa = 105%, which tells us something useful
(rather than saying Fa=80% in both cases). By analogy, I would say
similar things about bioavailability.
Walt
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.-at-.simulations-plus.com
Back to the Top
Dear Walt,
The discussion is still vivid and interesting, so I
continue:
>I agree - but it's not always that the iv clearance is so rapid. It may
>simply be that the oral absorption is so spread out in time compared to
>the iv data. With slow absorption due to solubility/dissolution/release
>limitations, this is not at all uncommon.
OK, you have the data, and I accept your view. My
reasoning was as follows: dissolution and release of a
drug from a dosage form should be within a reasonable
time, e.g 12 hours, to avoid irreproducible
bioavailability due to rapid passage and defecation. In
'usual cases', at, say, 4 hours after administration a
substantial amount of drug will be absorbed. If at 4 hours
after iv dosing plasma concentrations is below LOD,
half-life is 'short'.
> Also, EHC does not always
>result in an observed second peak (sampling times?) and
>EEC never does.
Second peak or not, in case of EHC or ECC there is a
'second absorption' of the drug. And this *must* result in
higher plasma concentrations than in the absence of EHC or
ECC.
>It's not bad design - it's the
>constraints put upon us by Mother Nature and economics.
Indeed, my point of view is an academic one. But if a
study design does not give the right answer, the design is
bad. Of course we sometimes must live with it, but it is
still a bad design. IMHO, it is still a matter of
economics. Please don't blame Mother Nature, who allows us
to do all these nice things.
>I guess this is another matter of definitions - is it cleared and then
>reabsorbed or is it still in the system? I have heard it said many
times
>that the gastrointestinal tract is effectively outside the body. If so,
>then drug secreted in bile, or exsorbed from the enterocytes, could be
>considered as having left the body. If it reenters, then it is absorbed
>again.
OK, this is a matter of definitions. The point I wanted to
make is that this definition is not really practical. It
would imply that the absorption should be considered also
for iv dosing. Do you accept that F can exceed 100% for iv
doses?
>But if the result produces F>1, then people get excited. It
>seems to me that there is no harm in F>1, or Fa>1, as long as we
>understand how the calculations were made and how to use the values
>going forward.
I agree; this is exactly what I said in my previous
comment.
>I'm sorry, but this is not the accepted definition of
>absorption by the FDA, nor of any pharmaceutical company with which
we've
>worked. In general, F<=FA, not F=Fa.
OK, you are completely right. Sorry for my inaccuracy.
>If you're interested in plasma concentration, then you're
>interested in molecules - I don't see how you can separate them.
I meant that I am interested primarily in the plasma
concentration profile, i.e. the number of molecules
present in a volume of plasma, because these molecules are
related to the drug effect. This number can be measured in
vivo with a reasonable accuracy. In general, I am less
interested in the number of molecules passing the apical
membrane of the enterocytes. This number cannot be
measured in vivo, and can be obtained only by some
modeling procedure. Of course the process of passing the
enterocytes is extremely important, but is it really
necessary to know whether each molecule passes on average
1.23 times the apical membrane?
>My point is that there is information in Fa=108% (from my example a few
>posts back) that is not provided if we adjust the AUC and report it as
>90% (for the same data).
How do you get the information that Fa is 108% or 90%?
IMHO, Fa cannot be measured, but estimated only using some
modeling procedure. Only F can be calculated (estimated)
using the AUC ratio.
'Adjust the AUC' sounds like a terrible crime! I would
prefer Mickey Mouse over the Great Manipulator.
>We, of course, use GastroPlus.)
So, you are modeling! Yes, now I understand your approach,
at least the idea. But we all know, with Box, that
modeling is always nice, often useful, and sometimes
reasonably close to the real thing. I don't see how
GastroPlus can escape this rule.
>then Fa = 105%, which tells us something useful
>(rather than saying Fa=80% in both cases).
Is this really useful? Fa = 105% could also mean that only
15% of the dose is absorbed (Fa = 15% in my view), and
that the absorbed molecules recirculated on average 6
times, resulting in Fa = 105%. What do we really learn
from your Fa = 105% if we do not know the value of 'Fa in
my view'.
>By analogy, I would say similar things about bioavailability.
Of course you can do, but this inevitably results in the
possibility that F > 100% for an intravenous dose. In
theory there is nothing wrong with this, but 99.9% of the
pharmacokinetic textbooks and papers dealing with
bioavailability need to be rewritten. Perhaps Mickey Mouse
does like this?
By the way, I fully agree with you that F may exceed 100%
if the process of EHC or ECC is really different after
oral dose than after iv dose. This does not need
rewritting the textbooks, and we could use such F values
perfectly, e.g. for dose regimen calculations. As stated
earlier, however, it seems rather unlikely that this could
really be the case, and therefore one should not accept
such a result too easily without a plausible explanation.
Best regards,
Hans
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
Back to the Top
Dear Hans,
Hans: "The discussion is still vivid and interesting, so I Continue"
I agree!
Hans: "OK, you have the data, and I accept your view. My reasoning was
as follows: dissolution and release of a drug from a dosage form should
be within a reasonable time, e.g 12 hours, to avoid irreproducible
bioavailability due to rapid passage and defecation. In 'usual cases',
at, say, 4 hours after administration a substantial amount of drug will
be absorbed. If at 4 hours
after iv dosing plasma concentrations is below LOD, half-life is
'short'."
I agree - in most cases, but we have data showing absorption later than
24, 36, and even 48 hours - it was this data that prompted my original
post on this subject. This is somewhat rare, but it's real, and we have
to deal with it.
Hans:"Second peak or not, in case of EHC or ECC there is a 'second
absorption' of the drug. And this *must* result in higher plasma
concentrations than in the absence of EHC or ECC."
Yes, absolutely. It is these higher concentrations, if/when they occur
only in the oral data, that can result in the F>1 phenomenon.
Hans: ">It's not bad design - it's the
> constraints put upon us by Mother Nature and economics.
Indeed, my point of view is an academic one. But if a study design does
not give the right answer, the design is bad. Of course we sometimes
must live with it, but it is still a bad design. IMHO, it is still a
matter of economics. Please don't blame Mother Nature, who allows us
to do all these nice things."
When I say Mother Nature, I refer to the problems of slowly absorbed
oral doses, whether due to low permeability, low solubility, slow
dissolution, or slow release, and how these affect our ability to get
the same sample times for iv and po doses. If Mother Nature clears the
iv dose in a time that is shorter than the oral (because of prolonged
absorption), then we cannot get an ideal design. To me, the term "bad"
design has implications of human error - someone didn't design the study
properly. But it may be impossible to achieve the ideal design, and if
so, is "bad design" the way to say it (i.e., no one is to blame, it just
is bad from an ideal sense)?
Hans:" >I guess this is another matter of definitions - is it cleared
and then
> reabsorbed or is it still in the system? I have heard it said many
times
> that the gastrointestinal tract is effectively outside the body. If
so,
> then drug secreted in bile, or exsorbed from the enterocytes, could be
> considered as having left the body. If it reenters, then it is
absorbed
> again.
OK, this is a matter of definitions. The point I wanted to make is that
this definition is not really practical. It would imply that the
absorption should be considered also for iv dosing. Do you accept that F
can exceed 100% for iv doses?"
I think the definition is very practical, because it is how we treat
things in GastroPlus, and we've been very successful in understanding
complex drug behaviors as a result. As far as iv doses with F>100%,
well, that's a very good question. If bioavailability is defined as the
percent of the dose that reaches the systemic circulation, and if, in
the case of twin drugs, the behavior of the first (with EEC or EHC)
could be duplicated by administering more of the second (without EEC or
EHC), then perhaps I would accept that. But . . . I have to admit, it
feels a little uncomfortable saying that. : )
I think I would ask - is it useful to talk about F>1 for iv doses? Does
it help us understand our data? If so, then I guess I would be in favor
of it.
Hans: "In general, I am less interested in the number of molecules
passing the apical membrane of the enterocytes. This number cannot be
measured in vivo, and can be obtained only by some modeling procedure.
Of course the process of passing the enterocytes is extremely important,
but is it really necessary to know whether each molecule passes on
average 1.23 times the apical membrane?"
Only if we want to understand mechanistically what is happening to the
drug. I view simulation/modeling as the only tool available for
understanding the many interactions that we often have to deal with, and
I have always strived to make models as physically real as circumstances
would allow. We never have the knowledge base to model everything
correctly, so we make assumptions and we use approximations. A typical
one- two- or three-compartment PK model is not correct, but it is often
very useful. One that adds EHC or that inherently accommodates EEC gives
additional insight. A PBPK model that identifies more accurately where
the drug is going can provide even greater insight. If a drug is
undergoing EHC or EEC, I want to see it and quantify it, if possible -
in a time-dependent model. Why? Not only to have a better understanding
of what happened already, but also to be able to estimate what that same
drug will do if the dose or formulation is changed, or if we want to
develop a controlled release formulation for a drug that is only
available as immediate release. I think mechanistic simulations can
provide better predictions that simplified models when complex phenomena
are involved.
Hans: "How do you get the information that Fa is 108% or 90%? IMHO, Fa
cannot be measured, but estimated only using some modeling procedure.
Only F can be calculated (estimated) using the AUC ratio."
Yes - through some kind of model.
Hans: "'Adjust the AUC' sounds like a terrible crime! I would prefer
Mickey Mouse over the Great Manipulator."
My point exactly! So if you don't adjust the AUC, and it gives you F>1,
then what do you do?
Hans: "So, you are modeling! Yes, now I understand your approach, at
least the idea. But we all know, with Box, that modeling is always nice,
often useful, and sometimes reasonably close to the real thing. I don't
see how GastroPlus can escape this rule."
Of course, it does not. Everything we do, whether fitting simple
noncompartmental parameters, or using a sophisticated simulation, is a
model. I like to say that modeling "helps us to understand what the data
is trying to tell us". So if Fa=108% comes out of a model, and if the
model makes sense within the framework of our definitions, then I think
we should not dismiss it out of prejudice.
Hans: "Is this really useful? Fa = 105% could also mean that only 15% of
the dose is absorbed (Fa = 15% in my view), and that the absorbed
molecules recirculated on average 6 times, resulting in Fa = 105%. What
do we really learn from your Fa = 105% if we do not know the value of
'Fa in my view'."
We would be led to search for why the apparent Fa is so high. (But if it
had been 99%, we probably wouldn't even bother to look, and we would
miss an opportunity for enlightenment.) The correct approach would be to
have a model that predicts both 15% absorption from the original dose,
and the resultant apparent Fa of 105% due to recirculation. That's what
really lets us understand what happened. That is the approach we take
when we say the drug has left the system and reentered, as we do in
GastroPlus. We can see, for example, 10% of a dose excreted as
undissolved particles (so it was never absorbed), with bioavailability
> 100% if we allow sufficient EHC. As long as we are consistent in our
definitions and math, we have a model that provides insight greater than
simply saying the volume of distribution is higher and the clearance is
lower.
Hans: "Of course you can do, but this inevitably results in the
possibility that F > 100% for an intravenous dose. In theory there is
nothing wrong with this, but 99.9% of the pharmacokinetic textbooks and
papers dealing with bioavailability need to be rewritten. Perhaps Mickey
Mouse
does like this?"
That's the part that makes me a bit uncomfortable, as I mentioned above.
But again, if the effect of one of two twin drugs that does not undergo
EHC/EEC can be duplicated by more dose compared to the one that does,
then maybe it's useful. I think 99.9% of the papers and textbooks are
not dealing with the rare situation where this can occur, so maybe only
0.1% would need to be rewritten. Mickey is looking for a job for the
winter months, so maybe he could help. ;o)
Hans - thanks again for your comments on this stimulating thread!
I am still curious about how you think we should handle the situation
where the (real world) data results in calculating F>1 using AUC/Dose
ratios. What should be reported?
Walt
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
Dear Walt,
Thank you again for this nice discussion. It becomes clear now that the
discussion is closely related to a different point of interest rather
than a
different point of view. As I understand you correctly, your point of
interest (as far as related to this discussion) is to get insight in
gastrointestinal absorption and all processes involved, and modeling of
these processes (in GastroPlus). This is indeed a fundamental process in
pharmacokinetics, and the mechanisms should be studied in great detail.
And
from this point of view your part of the discussion is clear and
logical:
'you want to know where all the molecules are, and how they are moving'.
My point of interest (as far as related to this discussion) is a more
global
description of pharmacokinetics, where rate of absorption and
bioavailability are used primarily to describe and model the plasma
concentration profile, among others, for the prediction of dosing
regimens.
In this view the mechanisms are not always relevant (although I fully
agree
that it would be better to take into account the mechanisms, but this
is not
possible on a routine basis; so, 'economics' again). Bioavailability is
here
a key parameter, and F > 1 for an intravenous dose does not work here
('need
to rewrite the textbooks'); even for exceptional drugs I would need a
convincing arguments to accept F > 1 for an oral dose, although this
would
not be really shocking.
> I think I would ask - is it useful to talk about F>1 for iv doses?
Does
> it help us understand our data? If so, then I guess I would be in
favor of
it.
This clarifies the aforementioned different points of interest. It may
indeed be helpful to understand our data, but at the same time, it would
also obscure our basic assumptions in pharmacokinetics, and so we might
easily make the wrong calculations and conclusions.
> I am still curious about how you think we should handle the situation
> where the (real world) data results in calculating F>1 using AUC/Dose
> ratios. What should be reported?
Actual results should be reported, of course. If it is measured, I do
not
have any problem with this result. We need an antenna for 'strange'
results,
but not a filter to leave out any 'unexpected' or 'impossible' result.
To
avoid misunderstanding, however, I would recommend to add a comment to
such
results.
Best regards,
Hans
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.-a-.farm.rug.nl
Back to the Top
Hans,
A very nice summation of this thread. Thank you - I enjoyed it very
much.
Walt
PharmPK Discussion List Archive Index page
Copyright 1995-2010 David W. A. Bourne (david@boomer.org)