Back to the Top
Dear collegues,
after developing a population pharmacokinetic model for premature
infants I wonder how it can be used in clinical practice? It seems to
me impossible to combine Bayesian curve fitting with my defined model
within the software I've used.
Does anybody have experiences or suggestions how to apply Bayesian
curve fitting to population pharmacokinetic models for clinical
practice?
Regards
Susanne Quellmann
[What software are you using? Many pk nonlinear regression programs
include Bayesian estimation - e.g. Adapt II and Boomer etc. - db]
Back to the Top
The following message was posted to: PharmPK
Dear Susanne
Personally I use my own population models of aminoglycosides in
premature patients using PKS sofware with good results.
The use of subpopulations depending of birth weight or gestational age
is a reasonable approach to apply Bayesian curve fitting
for clinical practice in this kind of patients.
Regards.
Dr. Jose M. Lanao
Dpt. Pharmacy and Pharmaceutical Technology
University of Salamanca
Spain
Back to the Top
The following message was posted to: PharmPK
Susanne
> after developing a population pharmacokinetic model for premature
> infants I wonder how it can be used in clinical practice? It seems to
> me impossible to combine Bayesian curve fitting with my defined model
> within the software I've used.
>
> [What software are you using? Many pk nonlinear regression programs
> include Bayesian estimation - e.g. Adapt II and Boomer etc. - db]
This can be a real problem if you are using standard off-the-shelf
Bayesian packages for dose individualisation since either the structural
PK model or the covariate regression models may not be supported by the
package. Sometimes you can trick the package into accepting your model
- but this can be very limited.
Your choices are to either use a more general package for doing Bayesian
MAP estimation (see David's comment above) - or write your own.
Depending on the model etc., it may be possible to write your own in
Excel (which has an inbuilt solver function) without too much
difficulty...
Steve
Stephen Duffull
School of Pharmacy
University of Queensland
Brisbane 4072
Australia
Tel +61 7 3365 8808
Fax +61 7 3365 1688
University Provider Number: 00025B
Email: sduffull.-a-.pharmacy.uq.edu.au
www: http://www.uq.edu.au/pharmacy/sduffull/duffull.htm
PFIM: http://www.uq.edu.au/pharmacy/sduffull/pfim.htm
Back to the Top
The following message was posted to: PharmPK
Dear Dr. Susann Quellmann,
A good software especially designed for clinical use
is USCPACK of Dr. RW. Jelliffe's group from LA. My own
experience is very promising. It is friendly and
goal-oriented.
I recommend it to you.
Back to the Top
The following message was posted to: PharmPK
Dear Susanne, dear all,
I came across the same problem when using WinNonMix for
mixed-effects-modeling (that doesn't provide a tool for Baysian curve
fitting).
The equations are too complex to run them with PKS , but I came across a
user defined model for WinNonLin in "Pharmacokinetic and Pharmacodynamic
Data Analysis" of J. Gabrielsson and D. Weiner.
Has anyone of the group some experience with applying this model?
Regards,
Katja
kgrasmae.-at-.uni-bonn.de
Back to the Top
The following message was posted to: PharmPK
Dear Susanne Quellmann,
Bayesian curve fitting is indeed one of the better solutions for
therapeutic drug monitoring in clinical practice using population data.
We have developed the program MW\Pharm for this purpose. The program is
quite flexible and extensive with respect to number of drugs, PK models
and variance models, and is oriented to clinical practice. MW\Pharm
includes also a tool for developing population models, using an
Iterative Two-Stage Bayesian analysis technique, which performs very
fast and is easy to use; both developing the model and applying the
model to new patients can be done in the same program, thus allowing
easy updating of the models.
The current version of the program runs under DOS, but works perfectly
under Windows 98 (and in general also under Windows XP). We are working
on a Windows version, which is expected to be released this year. For
technical information about the program you can contact me. For
information about prices and how to get the program please contact Eddy
van Essen of Mediware (vanessene.-a-.cs.com).
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
hello susanne
you may have a look at the following papers specifically dealing with
aminoglycosides in premature neobates:
de Hoog M, Mouton JW, Schoemaker RC, Verduin CM and van den Anker JN,
2002. Extended-interval dosing of tobramycin in neonates: implications
for therapeutic drug monitoring. Clin Pharmacol Ther, 71:349-58.
johan mouton
Back to the Top
The following message was posted to: PharmPK
Dear Susanne
You should also have a look at the following reference.
Stickland MD, Kirkpatrick CMJ, Begg EJ, Duffull SB, Oddie SJ, Darlow BA.
An extended interval dosing method for gentamicin in neonates. Journal
of Antimicrobial Chemotherapy 2001;48:887-93.
Cheers
Carl Kirkpatrick
Post-doctoral Research Fellow
School of Pharmacy
University of Queensland
St Lucia 4072
Brisbane
Phone:+617 3365 3195
Fax: +617 3365 1688
URL: http://www.uq.edu.au/pharmacy/ckirkpatrick/kirkpatrick.htm
Bayesian Methods for PKPD analysis (URL):
http://www.uq.edu.au/pharmacy/sduffull/MCMC_eg.htm
Back to the Top
The following message was posted to: PharmPK
Hello!
This seems interesting. Have you published your model yet or could I
have a report ? We are
looking at the issue of infant susceptibility to environmental
pollutants and PK modeling may help.
We are also using Bayesian stats and have developed a software for
hierarchical modeling
(check the MCSim page on http://toxi.ineris.fr , that particular page
is in English, the rest of the site will shortly be translated).
Frederic Y. Bois,
Unite de Toxicologie Experimentale, responsable
INERIS
Parc ALATA, BP 2
60550 Verneuil en Halatte
FRANCE
tel: + 33 (0)3 44 55 65 96
fax: + 33 (0)3 44 55 66 05
email: frederic.bois.-a-.ineris.fr
web: http://www.ineris.fr, http://toxi.ineris.fr
Back to the Top
Dear Susanne:
About Bayesian curve fitting. You start with the population
model. It is the Bayesian prior. You use it to design the dosage
regimen to hit the target goals you have selected for that patient. You
monitor the patient and the serum concentrations, and then use Bayesian
fitting (there are at least 3 types - maximum aposteriori probability -
MAP, multiple model, or sequential interacting multiple model) and make
your individualized model of the behavior of the drug in that patient.
You compare the behavior of the model and the clinical behavior of the
patient, to evaluate the patient's sensitivity to the drug, and to
re-evaluate your target goals. Then you use the individualized model to
develop the next dosage regimen for the patient. You might look at our
USC*PACK clinical software, and at
Jelliffe RW, Schumitzky A, Van Guilder M, Liu M, Hu L, Maire P, Gomis
P, Barbaut X, and Tahani B: Individualizing Drug Dosage Regimens: Roles
of Population Pharmacokinetic and Dynamic Models, Bayesian Fitting, and
Adaptive Control. Therapeutic Drug Monitoring, 15: 380-393, 1993.
There is also more information on our web site, www.lapk.org, under
teaching topics.
Very best regards,
Roger Jelliffe
Roger W. Jelliffe, M.D. Professor of Medicine,
Division of Geriatric Medicine,
Laboratory of Applied Pharmacokinetics,
USC Keck School of Medicine
2250 Alcazar St, Los Angeles CA 90033, USA
Phone (323)442-1300, fax (323)442-1302, email= jelliffe.-at-.usc.edu
Our web site= http://www.lapk.org
PharmPK Discussion List Archive Index page
Copyright 1995-2010 David W. A. Bourne (david@boomer.org)