Back to the Top
Dear PK group,
I would like to ask- does anyone know if I can use Winnonlin program for
calculation of absorption by the convolution method?
Liat Lomnitski, Phd
Back to the Top
Dear Dr. Lomnitski,
I cannot answer regarding WinNonlin, but I can offer to provide information
regarding GastroPlus(TM) - the state-of-the-art absorption simulation
program now in use in the U.S., Europe, and Japan, and by most of the major
pharmaceutical companies in the world. GastroPlus(TM) is produced by our
company, Simulations Plus, Inc., in Lancaster, California.
GastroPlus(TM) is an extension of the work originally done by Gordon Amidon
and his students and colleagues. Written from scratch, it incorporates a
compartmental absorption and transit model based on a series of 41
differential equations which are integrated by a 4th-5th order variable
step Runge-Kutta integration scheme. There are many, many features to the
program, and we offer an evaluation CD-ROM that lets you experience the
full power of GastroPlus(TM) for a limited time.
Please feel free to contact us if you are interested.
Walt Woltosz Phone: (661) 723-7723
Chairman & CEO FAX: (661) 723-5524
Simulations Plus, Inc. (NASDAQ:SIMU)
1220 West Avenue J
Lancaster, CA 93534-2902
Back to the Top
The following summary on convolution and deconvolution methods and
references was kindly provided by Bill Gillespie. Updates on our product
releases can always be obtained at our web site www.pharsight.com.
A future release of WinNonlin will implement a deconvolution method. It is a
constrained deconvolution method that produces a smooth estimate of the
input rate which is constrained to be non-negative. The degree of smoothness
may specified by the user via adjustment of a single tuning parameter, or
that parameter may be estimated automatically via cross-validation.
The estimated release date for the WinNonlin version containing the
deconvolution components is December, 1999. Otherwise I am not aware of any
good commercial software products for convolution and deconvolution.
Kinetica implements a relatively crude point-area deconvolution method that
I cannot recommend due to its poor performance with data containing any
realistic amount of variability. If your needs are more immediate, the best
noncommercial endeavors I am familiar with are programs developed by Roman
Hovorka at City University, London
(http://www.city.ac.uk/~er583/mmg/software.html) and Davide Verotta at UCSF.
Some references on their work follows:
R Hovorka, MJ Chappell, KR Godfrey, FN Madden, MK Rouse, PA Soons. CODE: A
deconvolution program implementing a regularisation method of deconvolution
constrained to non-negative values. Description and pilot evaluation.
Biopharm Drug Dispos 19:39-53 (1998).
FN Madden, KR Godfrey, MJ Chappell, R Hovorka, RA Bates. A comparison of six
deconvolution techniques. J Pharmacokin Biopharm 24:283-299 (1996). D
Verotta. Two constrained deconvolution methods using spline functions. J
Pharmacokin Biopharm 21:609-636 (1993).
D Verotta. Estimation and model selection in constrained deconvolution.
Annals Biomed Eng 21:605-620 (1993).
K E Fattinger, D Verotta. A non-parametric subject-specific population
method for deconvolution: I. Description, internal validation and real data
examples. J Pharmacokin Biopharm 23:581-609 (1996).
D Verotta. An inequality-constrained least-squares deconvolution method. J
Pharmacokin Biopharm 17:269-289 (1989)
Jeff Wald, Ph.D. | Manager, Scientific Support Services
Pharsight Corporation | www.Pharsight.Com
Direct: Jeff.Wald.-a-.Pharsight.com | Technical inquiries:
Back to the Top
In determination of the rate of the drug bioavailability, the
deconvolution methods can be used if intra- and
extravascular drug inputs have equal forms. For example, if a drug
is administered inravenously and orally in the single doses D_iv and
D_po, respectively, where
D_iv = k D_po,
and k is a constant k>0.
In our study (1), we presented the method and software (a commercial
software product) for determination of the rate of the drug
bioavailability which can be used for equal and/or unequal
forms of intra- and extravascular drug inputs. For example, if
a drug is administered inravenously in a short or long time
infusion and orally in a single dose. This method is based on the
deterministic circulatory model introduced in the same study and
allows to determine the model of the rate of the drug bioavailability
in a numerical and/or analytical form. No smoothing methods and no
prior hypothesis about the analytical form of the bioavailability
rate are necessary. In our next study (2), we introduced the
criterion for testing equivalence of two weighting functions. This
criterion can be used e.g. to test similarity of the two rates of
the drug bioavailability from the two different drug formulations.
The criterion has two limit values: 0 % and 100 %. The closer the
value of the criterion to 100 %, the higher the probability that the
two rates of the drug bioavailability are similar.
1. Dedik, L., Durisova, M. Computer Methods and Programs in
Biomedicine, 51, 1996, 183-192.
2. Durisova, M., Dedik, L., Phramaceutical Research, 14, 1997,
PharmPK Discussion List Archive Index page
Copyright 1995-2010 David W. A. Bourne (firstname.lastname@example.org)