Back to the Top
The following message was posted to: PharmPK
Dear All,
I currently am struggeling with the pharmacokinetic analysis of single
or multiple bolus plus continous infusion data without elimination phase
samples in pediatric patients of a highly lipophilic drug, that is
hepatically eliminiated by different CYPs and is known to exhibit high
interindividual variability. Unfortunately there are only 16 Patients
available, with 10-25 samples taken over 1-7 days. In addition i have
countious measurements of heart rate data and other variables. I am
wondering how to approach the pharmacokinetic data, as lacking
elimination samples seems to preclude quite some options. Is a NCA
approach with calculation of clearance valid in this data set? Which
parameters of the analysis could i try to correlate to the heart rate
data? Which options to I overlook...?
Thank you all for your help in advance,
sincerely yours,
Sergej Ramusovic
PhD Student
[Do you have data during the initial upswing? Any variation from steady state.
Circadian effects might be a nuisance but you might be able to model your
data, starting with something small (one compartment for example) - db]
Back to the Top
The following message was posted to: PharmPK
Sergej,
There are lots of things you could do to understand this data set.
"Unfortunately there are only 16 Patients
available, with 10-25 samples taken over 1-7 days. "
First of all forget about NCA.
Then take a course in population PKPD analysis.
Finally you might wonder how it can be considered ethical to take 10 to 25
samples from children without any idea of how to analyse the data...
If you would like to write to me directly to discuss the details I'd be happy to
try and help you.
Nick
Back to the Top
The following message was posted to: PharmPK
Dear Sergei:
I agree with Nick. It would be good if you knew population modeling.
There are basically two approaches currently - parametric, which is what
NONMEM does, for example. There are other parametric pop methods which have
exact likelihood computations rather than the approximate ones in NONMEM.
Parametric approaches are also limited by the fact that they must make
assumptions about the expected shape of the model parameter distributions
such as normal, lognormal, multimodal, etc. If the assumption is not true,
the results obtained will not be correct. These are important limitations of
the parametric approach to population modeling. The third important
limitation of parametric approaches is that dosage regimens developed using
them can never evaluate the expected precision with which a dosage regimen
will hit a desired target goal. This has been well shown by the separation
principle. It states that whenever you seek to control a system by first,
getting point estimates of the model parameters, and second, using these
point estimates to control the system, the control (dosage regimen in this
case) is done suboptimally, as there is no performance criterion to
optimize.
The other approach is nonparametric (NP). In this case one does not
need to make any assumptions at all about the shape of the parameter
distributions. The entire distributions are estimated, not just means, SD's,
medians, covariances, etc. This assures that results with maximum likelihood
are obtained. In addition, pop models made with NP methods can be directly
linked with "multiple model (MM)" dosage design. Regimens developed here are
maximally precise as they use the entire model parameter distributions to
find the specific dosage regimen that hits the desired target with minimum
expected weighted squared error. This avoids the problems of "separation
principle" control described above. Of the currently available maximum
likelihood software for pop modeling, this meets the needs the best. Using
NP models and MM dosage design, you will see that in addition to the usual
covariates, the dosage regimen itself becomes a most important tool to
reduce variability in patient response about a chosen target goal. If you
use parametric models, you will never see this and can never be aware of it.
You might look at
1. Jelliffe R, Schumitzky A, Bayard D, Leary R, Botnen A, Van Guilder
M, Bustad A, and Neely M: Human Genetic variation, Population
Pharmacokinetic - Dynamic Models, Bayesian feedback control, and Maximally
precise Individualized drug dosage regimens. Current Pharmacogenomics and
Personalized Medicine, 7: 249-262, 2009.
2. Bustad A, Terziivanov D, Leary R, Port R, Schumitzky A, and Jelliffe
R: Parametric and Nonparametric Population Methods: Their Comparative
Performance in Analysing a Clinical Data Set and Two Monte Carlo Simulation
Studies. Clin. Pharmacokinet., 45: 365-383, 2006.
3. Jelliffe R: Estimation of Creatinine Clearance in Patients with
Unstable Renal Function, without a Urine Specimen. Am. J. Nephrology, 22:
320-324, 2002.
4. Leary, R., Jelliffe R., Schumitzky, A., and Van Guilder, M An
adaptive grid non-parametric approach to pharmacokinetic and dynamic(PK/PD)
population models, 14-th IEEE Symposium on Computer Based Medical Systems,
389-394, 2001.
5. Jelliffe R: Goal-Oriented, Model-Based Drug Regimens: Setting
Individualized Goals for each Patient. Therap. Drug Monit. 22: 325-329,
2000.
6. Jelliffe R, Bayard D, Milman M, Van Guilder M, and Schumitzky A:
Achieving Target Goals most Precisely using Nonparametric Compartmental
Models and "Multiple Model" Design of Dosage Regimens. Therap. Drug Monit.
22: 346-353, 2000.
7. Jelliffe R, Schumitzky A, Bayard D, Milman M, Van Guilder M, Wang X,
Jiang F, Barbaut X, and Maire P: Model-Based, Goal-Oriented, Individualized
Drug Therapy: Linkage of Population Modeling, New "Multiple Model" Dosage
Design, Bayesian Feedback, and Individualized Target Goals. Clin.
Pharmacokinet. 34: 57-77, 1998.
We welcome visitors at any time, and emails and phone calls. We will
be at the PAGE meeting this June 5-8 in Venice, and we look forward to
seeing you and anyone else who is interested in optimal pop modeling and
optimally precise drug therapy. Also, our web site is below. Come and
visit!
Very best regards,
Roger W. Jelliffe, M.D., F.C.P.
Professor of Medicine,
Co-Director, Laboratory of Applied Pharmacokinetics
www.lapk.org
USC Keck School of Medicine
2250 Alcazar St, Room 134-B
Los Angeles CA 90033
Back to the Top
The following message was posted to: PharmPK
Dear All,
I currently am struggeling with the pharmacokinetic analysis of single
or multiple bolus plus continous infusion data without elimination phase
samples in pediatric patients of a highly lipophilic drug, that is
hepatically eliminiated by different CYPs and is known to exhibit high
interindividual variability. Unfortunately there are only 16 Patients
available, with 10-25 samples taken over 1-7 days. In addition i have
countious measurements of heart rate data and other variables. I am
wondering how to approach the pharmacokinetic data, as lacking
elimination samples seems to preclude quite some options. Is a NCA
approach with calculation of clearance valid in this data set? Which
parameters of the analysis could i try to correlate to the heart rate
data? Which options to I overlook...?
Thank you all for your help in advance,
sincerely yours,
Sergej Ramusovic
PhD Student
[Do you have data during the initial upswing? Any variation from steady state.
Circadian effects might be a nuisance but you might be able to model your
data, starting with something small (one compartment for example) - db]
Back to the Top
Ramusovic,
Can you post a clear study design to share some thoughts!
Best,
Rao
Back to the Top
The following message was posted to: PharmPK
Dear Sergej,
Is the molecule you are working with a well known drug or NCE? If you have more
complete adult data and information on in vitro metabolism, our PBPK software "GastroPlus"
or "SimCYP" or "PKSim" would be appropriate for building a model for the adult that could
easily be converted to predict the pediatric subjects. Even though you probably can't fit
the data using classical PK approaches, if you build a PBPK/PD model using adult IV data
you'll be in the best position to understand the pediatric data.
Even if this is an NCE and you only have pediatric data (not likely!) you could calculate
the physicochemical properties and estimated Vss using ADMET Predictor and apply in vitro
- in vivo extrapolation (IVIVE) to build the pediatric model.
Mike
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
* Michael B. Bolger, Ph.D.
* Chief Scientist
* Phone: 707-303-6067 or (661) 723-7723 x 301
* FAX: (661) 723-5524
* Simulations Plus, Inc.
* 42505 10th Street West
* Lancaster, CA 93534
* U.S.A.
* http://www.simulations-plus.com (NASDAQ: SLP)
* bolger.aaa.simulations-plus.com
|
Want to post a follow-up message on this topic?
If this link does not work with your browser send a follow-up message to PharmPK@boomer.org with "Analysis of bolus plus continuous infusion data without elimination samples" as the subject | Support PharmPK by using the |
Copyright 1995-2011 David W. A. Bourne (david@boomer.org)