Back to the Top
The following message was posted to: PharmPK
Dear Colleagues,
Can anyone think of conditions under which it is more
appropriate/correct/relevant to conduct
non-compartmental analysis rather than compartmental?
Any discussions, examples, references or publications
appreciated.
Rostam
Back to the Top
The following message was posted to: PharmPK
I tried to illustrate the difference between Comp. and Non-Comp. in the
paper "The Rise and Fall of Compartmental Analysis," Pharmacological
Research, Vol 44, pages 337-342 (2001).
Back to the Top
Rostam,
Normally NCA is more appropriate and/or relevant in toxicokinetic
studies where few samples are taken and the only relevant parameters
are Tmax, Cmax and AUC. NCA is also a very useful technique for
bioavailability analysis. In bioavailability calculations, only AUC is
relevant [along with K(el) if AUC(t-inf) is important], and AUC is
easily calculated without model bias using the NCA method.
Nathan S. Teuscher, Ph.D.
Clinical Pharmacokinetics
NPS Pharmaceuticals
Back to the Top
The following message was posted to: PharmPK
Rostam,
The eternal question again. It seems to me that one reason this subject
is often confusing is because the term "compartmental" tends to imply
the use of "compartmental curve fitting" to define a compartmental
model and its associated PK parameters. In fact, the simple use of an
exponential term (extracted from analysis of the plasma data)
constitutes a compartment without any assumption of a model.
It is thus the ambiguity behind the term "compartmental" that gives
rise to the confusion.
If some of our learned contributors could elaborate more, I am sure it
would be beneficial. Perhaps the solution lies in a better description
of the following key terms, as I have attempted:
Compartmental modelling - a software approach based on curve
fitting
Noncompartmental - any approach not based on curve fitting
Curve stripping - method for obtaining exponential terms
Model-independent - results based in part on AUC measurements
NCA - old acronym that implied model-independent approach
The need is to correct the concept that "compartmental" implies
modelling and "noncompartmental" implies "model-independent."
To put this into some perspective, Gibaldi's text is a thorough
compartmental treatment of pharmacokinetics that devotes a chapter to
model-independent methods and an appendix to curve-fitting programs
Back to Rostam's question: here are two cases when noncompartmental
methods (as defined above) are most suitable:
(1) When you don't need to know the compartmental model to get
results.
(2) When you want to do and submit bioequivalence data to the
FDA.
Regards,
David S. Farrier
Summit Research Services
www.SummitPK.com
DFarrier.at.SummitPK.com
David S. Farrier, Ph.D. Phone: 970-249-1389
Summit Research Services Fax:: 970-249-1360
68911 Open Field Dr. Email: DFarrier.at.SummitPK.com
Montrose, CO 81401 Web: http://www.SummitPK.com
Back to the Top
The following message was posted to: PharmPK
Dear David,
My question did not come across clearly so I try to
give an example. You may have a situation that plasma
concentration-time data could be explained by
exponential function (mono or bi) but not all the
assumptions of one or two compartmental models hold.
For example, under situation that there is no rapid
distribution (equilibrium between plasma and tissues
has not been achieved) but plasma profile shows a
mono-phasic decline or there is no or limited
redistribution occurs during the initial decline phase
but plasma profile shows a bi-phasic decline (e.g.,
formulations that actively taken up by macrophages).
So we may have situations that is more appropriate to
use NCA since there is no compartment assumption
associated with it. Does it make sense?
Aldo, would you please comment. Is not this similar
to what you referred in your publication: “fitting the
experimental data with exponential function, by
itself, is not enough to prove the existence of
compartment” or I am missing something?
Rostam
Back to the Top
The following message was posted to: PharmPK
I think there is substantial abuse of the term "non-compartmental
analysis". One can analyze a PK profile by measuring observable
features of the curve...slopes, heights, areas, and moments (SHAM).
There are no compartmental assumptions required to create this sort of
descriptive analysis. However, as soon as you infer some parameter
value with physiological analogy...like a clearance or volume...you
have made some very heavy compartmental assumptions...so why not make
the assumptions more explicit and testable using real model based
approaches?
As for the appropriateness of one approach versus another...you will
find varying philosophies. My philosophy is that as ethical scientists
we must make the most of our data so that we can minimize exposure of
animals, human volunteers, and patients to unproven therapies and allow
patient access to beneficial therapies as soon as possible. If making
educated, scientific, explicit, testable assumptions based on prior
knowledge can facilitate a model based approach that helps to achieve
that goal...by all means use modeling.
Jeff
PharmPK Discussion List Archive Index page
Copyright 1995-2010 David W. A. Bourne (david@boomer.org)