Back to the Top
[I have received quite a few requests to subscribe and unsubscribe to/from
PharmPK - some of you may have resubscribe thinking that you had been
removed (because of the no activity). In this case you may get an error
message - others wanting to leave the list may still get this and maybe one
or two other messages, I am working through the requests and should have
most of them completed soon. You may receive more than one unsubscribe
message or an error message. Please ignore these for a few days. If they
continue please let me know at david.at.boomer.org - Thanks, db]
Hi
I have a very basic question in Pharmacokinetic/Pharmacodynamic modelling.
I have a concentration profile for a single dose of a drug given by
intravenous bolus route in rats . I also have with me the pharmacodynamic
effect (mean arterial pressure, heart rate) of the same dose of the drug. My
questions are:
1. Is it possible to model with this data? OR
2. Is it essential to give as an intravenous infusion so that i have
the build up of the concentration?. OR
3. Do I need to have different doses and their coreesponding
concentrations and effect.
I would appreciate if someone could provide some references related to
this issue.
Thanks
Cheers
Ramesh
rpvn1.aaa.uic.edu
UIC
*******************************
The Harder You Work... The Luckier You Get...
*******************************
Rameshraja V.N. Palaparthy
1926 W.Harrison Street
#1402 Chicago IL 60612.
Res : (312)-733-7986
Off : (312)-996-3607
EMAIL : rpvn1.aaa.uic.edu
rpalap.-a-.hotmail.com
*******************************
Back to the Top
Dear Ramesh,
It appears you have the data needed to do PK/PD modeling.
Step 1: model conc-time data to describe equation
Step 2: use Hill equation (or other suitable Equation) to do PK/PD modeling
See Gibaldi and Perrier, Pharmacokinetics, Marcel Dekker, 1982 for details
Good luck
Sri
Sri Melethil, Ph.D.
Professor of Pharmaceutics and Medicine
University of Missouri-KC
School of Pharmacy
Room 203-B, 5005 Rockhilll Road
Kansas City, MO 64110
816-235-1794 (fax: 816-235-5190)
Back to the Top
Ramesh,
Please see our three recent publications regarding PK/PD modelling issue. The
drugs we used in our studies are cocaine, benzoylecgonine, norcocaine, and
cocaethylene. We measured plasma concentrations of cocaine and the three
metabolites in both rat plasma and brain ECF for PK. We simultaneously
monitored heart rate, blood pressure, QRS interval, and dopamine in brain ECF
for PD. We then did PK/PD modelling with various methods described in
Garbrielsson and Weiner's book "Pharmacokinetic and Pharmacodynamic Data
Analysis, Concepts and Applications" and in Krzyzanski and Jusko's paper
"Mathematical formalism for the properties of four basic models of indirect
pharmacodynamic responses. J Pharmacokin Biopharm 1997, 25:107-123".
Pan WJ and Hedaya MA: An animal model for simultaneous pharmacokinetic/
pharmacodynamic investigations: Application to cocaine. J Pharmacol Toxicol
Methods 1998, 39:1-8.
Pan, W. J. and Hedaya, M. A.: Cocaine and alcohol interactions in the rat:
effect on cocaine pharmacokinetics and pharmacodynamics. Journal of
Pharmaceutical Sciences. 1999, 88:4, 459-467.
Pan, W. J. and Hedaya, M. A.: Cocaine and alcohol interactions in the rat:
pharmacokinetic and pharmacodynamic contributions of cocaine metabolites.
Journal of Pharmaceutical Sciences. 1999, 88:4, 468-476.
Good luck!
Wei-jian
Wei-Jian Pan, Ph.D.
Dept 4PK, Bldg AP13A-3
Dept of Clinical Pharmacokinetics and Toxicokinetics
Abbott Laboratories
Abbott Park, IL 60064-6104
weijian.pan.aaa.abbott.com
Phone: 847-935-5335
FAX: 847-938-5193
Back to the Top
Hello,
1. Whether you can build a pk/pd model with single dose data or not
depends on the purpose of modelling. If all one wants is a 'feel' for
the data to plan a more elaborate experiment, then probably it is
useful. But, if the model is intended for dose explorations and
extrapolation via simulations to man, then definitely a more robust
model can only developed by careful consideration of the
concentration-effect range.
2. Before performing modelling, we need to understand the mechanism of
action of the drug and the nature of the disease. Only then a
physiologically meaningful model can be derived.
3. Particularly drugs acting on hemodynamic effects have been very well
studied in the past. Two references, while there are many, are given
below:
A. Kleinbloesem, CH. et al. Rate of increase in the plasma
concentration of nifedipine as a major determinant of its hemodynamic
effects in humans. Clin.Pharm.Ther. 1987;41:26-30.
B. Ritchie, RH. et al. Myocardial effect compartment modeling of
metoprolol and sotalol: Importance of mycardial subsite drug
concentration. 1998;87(2), 177-182.
With regards,
Joga
Joga Gobburu
Pharmacometrics,
CDER, FDA
Back to the Top
Hi Ramesh,
With the concentrations of the drug in central compartment and PD data you
have the basis for the PK/PD modelling. You should be in mind: after iv
bolus administration of the drug you have a combination of several transport
processes in parallel (distribution to fat, muscle - renal and or metabolic
elimination - distribution to the target tissue may be with a binding to
specific receptors). Using this vision you have to calculate with an
appropriate PK model to approximate the concentrations profile in the target
tissue. Than start with an E-max model?
Please have a look at this papers:
W. Cawello, Connection of pharmacokinetics and pharmacodynamics - how does
it work?, Int J Clin Pharmacol Ther (1997), 35: 414-417
Willi Cawello and Tammy Antonucci, The correlation between pharmacodynamics
and pharmacokinetics: Basics of Pharmacokinetics-Pharmacodynamics modeling
(1997), 37: 65S-69S
Best Regards,
Willi Cawello, Ph.D.
Humanpharmacology
Schwarz-Pharma company
D40789 Monheim am Rhein, Germany
Email cawello.-at-.schwarzpharma.com
Back to the Top
Hello Ramesh,
One method of pharmacodynamic parameter estimation involves Bayesian
regression. This method involves the minimization of a Bayesian objective
function involving the pharmacodynamic parameters and the physiological
variable being measured.
Obj(Bayes)= Sum [(Pip-Pim)2/(SD Pip)2 + (PysioObs-PysioPred)2/(SD
PhysioObs)2]
Where:
Obj(Bayes)= Bayesian Objective Function
Sum= sum of squared arguments (some of squares)
Pi= Pharmacodynamic parameter i, model estimated
Pip= Pharmacodynamic parameter i, population mean
PhysioObs= observed physiological or pharmacological response
PhisioPred= predicted physiological or pharmacolical response
(Pio-Pip)2= squared difference
(SD Pip)2= standard error of population mean squared= variance
(PhysioObs-PhysioPred)2= sqaured difference in predicted and observed
(SD PhysioObs)2= standard error in measurment squared= variance
A good example of this method of pharmacodynamic parameter estimation
is found in warfarin dosing. Dr.Mungall has his pharmacokinetic/pharmacodynamic
programs available on the internet. He uses the above equation to dose warfarin
in the DrugCalc program, where the physiogical response is prothrombin time or
prothrombin complex activity. The pharmacodynamic parameters are Cmax, Kd
and M
contained in the modified Nagashima equation, or Theophanus-Barile equation.
The concentration of warfarin is predicted using regression equations and is
inserted
into ths pharmacodynamic equation, from which the pharmacodynamic parameters
are estimated
by minimizing the above Bayesian objective function.
A similar approach could be applied to other pharmacodynamic models,
where the
pharmacodynamic equation could be the Hill equation. But, the objective
function would
be similar. A good synopsis of pharmacodynamic models is contained in both
volumes of
Gibaldi and Perrier Pharmacokinetics. The relationship between drug
concentrations and
pharmacodynamic parameters is contained in the particular pharmacodynamic
equation describing
the pharmacological model under study.
Mike Leibold, PharmD, RPH
ML11439.-at-.goodnet.com
Back to the Top
Hi All,
In response to my earlier comments about the InteliSite capsule, I have had
some very differing reports as to transit times and other problems. I am
currently trying to collate data on the subject to try to understand why
such differences exist and I would be grateful for any more info which could
shed light on some of these data.
Thanks for your help!
Neena Washington
Back to the Top
[Two replies - db]
Sender: acy62.aaa.pop.dial.pipex.com
Subject: Re: PharmPK Re: PK-PD modelling - a very basic question
Date: Thu, 24 Jun 99 11:30:18 +0100
x-sender: acy62.at.pop.dial.pipex.com
From: Andrew Sutton
To: "David Bourne"
Mime-Version: 1.0
Dear Neena
There is a very good reference on the variability of gastric, small bowel
and colonic transit times from Degen LP and Phillips SF (1996) at the
Mayo Clinic in: Gut, 1996, 39, 299-305. It shows the significantly
faster transits in men than women using scintigraphy techniques.
Andrew Sutton
Andrew Sutton
ASutton.-at-.gcpl.co.uk
Guildford Clinical Pharmacology
Innovators in Phase 1 model design
and leaders in Phase 2 recruiting.
Telephone +44 (0) 1483 406886
---
Date: Thu, 24 Jun 1999 08:46:49 -0400
From: "Nabil B. Darwazeh"
To: PharmPK.aaa.boomer.org
Subject: PK-PD modelling - a very basic question
Mime-Version: 1.0
Hi Neena
It has been few weeks since we exchanged ideas on this subject. Let us
start by simplifying the problem step by step, we might be able to find
out the source of contradictions between different users of this capsule.
Regards
Nabil
darwazn.at.war.wyeth.com
---
PharmPK Discussion List Archive Index page
Copyright 1995-2010 David W. A. Bourne (david@boomer.org)