- On 25 Mar 2005 at 14:07:46, "Tremblay Pierre-Olivier" (potremblay.-a-.anapharm.com) sent the message

Back to the Top

The following message was posted to: PharmPK

Dear PharmPK members, I know I'm bringing up old stuff here and not of

very high scientific interest but if someone could give me a hand or at

least give me some hints, it would be greatly appreciated.

Here's the question: How in Winnonlin can you specify multiple dosing

in a user model based on a system of differential equations (NOT the

analytical solution).

I know I could do this quickly using the integrated algebraic functions

but the model I'm working on right now is reasonably complex. I did

symbolic integration in a math software and the resulting expressions

take several pages. The expressions were slimer using numerical

integration of course but then I figured I'd lose some control over

parameter modification for simulation purposes (right ?).

So in the end, I thought that getting the trick for performing

multiple-dosing in WNL using differentials would be simpler than to

integrate every time I change the value of a micro-constant.

Many thanks in advance.

Pierre-Olivier :o)

P.S. I did browse through the PharmPK archive and got a few hits but I

had trouble interpreting the the answers. - On 25 Mar 2005 at 13:37:10, "Walt Woltosz" (walt.at.simulations-plus.com) sent the message

Back to the Top

The following message was posted to: PharmPK

Dear Pierre-Oliver,

You might want to consider using GastroPlus for your simulations.

The type of analysis you describe is quite straightforward with it. You

can even easily simulate, and fit models to, mixed multiple doses if you

wish (e.g., mixtures of iv bolus, iv infusion, immediate release, and

controlled release doses given at various times). Any combination of

pharmacokinetic and pharmacodynamic model parameters can be fitted

across single or multiple observation sets. Saturable metabolism and

transport can also be include if needed. Simple 1-, 2-, or 3-compartment

PK can be used, or complete PBPK models if you prefer.

An interactive web demonstration and a short trial period can be

arranged if you are interested.

Best regards,

Walt Woltosz

Chairman & CEO

Simulations Plus, Inc. (AMEX: SLP)

1220 W. Avenue J

Lancaster, CA 93534-2902

U.S.A.

http://www.simulations-plus.com

Phone: (661) 723-7723

FAX: (661) 723-5524

E-mail: walt.at.simulations-plus.com - On 29 Mar 2005 at 11:16:00, "Mark Lovern" (mlovern.-at-.Pharsight.com) sent the message

Back to the Top

Dear Pierre,

Code for implementing multiple dosing for a differential equation model

in WinNonlin will vary depending upon the type of input (ie 1st-order

absorption, IV infusion etc.) Also, even for a particular model, there

are a variety of ways to implement the dosing code. For example,

perhaps, the most conceptually straightforward approach for a 1st-order

oral absorption model is to write code that simply 'dumps' drug into

the gut compartment at each dosing time. However, this approach does

have the limitation that the exact times for each dose must be embedded

in your data file. An alternative approach for the same model would be

to model the amount of drug in the gut (Ag) as a superposition of

exponential terms, then incorporate the term for Ag into the

differential equation for the central compartment. For your reference,

please find below annotated code for implementing each approach for a

1-compartment model.

EXAMPLE I: CODE FOR A FULL DIFFERENTIAL EQUATION MODEL

Model 1

remark ******************************************************

remark Developer:

remark Model Date: 29-Mar-2005

remark Model Version: 1.0

remark ******************************************************

remark

remark - define model-specific commands

remark 1-Compartment Model with 1st-order extravascular absorption

remark no. parameter constant

remark --- --------- ------------

remark 1 Ka # of Doses

remark 2 CL Dose #1

remark 3 V1 Time of Dose #1, etc.

remark **Warning!**Warning!**Warning!**Warning!**Warning!**Warning!

remark ** In order for this model to work, you must have time **

remark ** points in your data set that correspond to each dose **

remark ** administration time and each dump time from the bile. **

remark **Warning!**Warning!**Warning!**Warning!**Warning!**Warning!

remark - define differential equations starting values

COMMANDS

NFUNCTIONS 1

NDERIVATIVES 2

NPARAMETERS 3

PNAMES 'Ka', 'CL', 'V1'

END

remark - define temporary variables

TEMPORARY

T=X

NDOSE = CON(1)

DOSE1 = CON(2)

END

START

Z(1) = DOSE1

Z(2) = 0

END

remark - define differential equations

DIFFERENTIAL

remark The next block of code handles the multiple dosing.

J=3

remark Count up the number of doses administered by time = T.

remark NDOSE is the total number of Doses given to each subject.

DO I = 1 to NDOSE-1

remark The next conditional compares the current value of the

remark integration time index with the administration time

remark of the ith dose.

remark Note: The 3rd cell in the dosing grid contains the admin

remark time for the 1st dose. After that, dose times will be

remark in every second cell. Thus, CON(J) will

remark contain the dose time for the Jth dose.

remark If the administration time for the Jth dose is < the

remark current value of the integration time index then

remark J (# administered doses) is incremented and

remark the DO Loop is iterated another time.

remark Otherwise control "skips" to the block of

remark code labelled "RED:"

IF T > CON(J) THEN

J = J+2

ELSE

GOTO RED

ENDIF

NEXT

remark The next block of code checks to see if the (very)last dose has

remark already been administered. If so, then the time and amount of

remark the last dose are saved into the variables LASTDOSE and

remark TLSTDOSE.

IF INT(((J-1)/2)+.001) = NDOSE AND T > CON(J) THEN

LASTDOSE = CON(J-1)

TLSTDOSE = CON(J)

GOTO GREEN

ENDIF

RED:

remark The following conditional ensures that the dose at Time

remark zero is not counted twice.

IF J > 3 THEN

remark The next conditional checks to see if the integration

remark time index is exactly equal to the Jth dosing time. If

remark so, then the dose is "dumped" into the absorption

remark compartment. LASTDOSE is the amount of the last administered

remark dose, TLSTDOSE is the administration time of the last dose.

IF T = CON(J) THEN

LASTDOSE = CON(J-1)

TLSTDOSE = CON(J)

Z(1) = Z(1)+LASTDOSE

ELSE

LASTDOSE = CON(J-3)

TLSTDOSE = CON(J-2)

ENDIF

ELSE

LASTDOSE = DOSE1

TLSTDOSE = 0

ENDIF

GREEN:

DZ(1) = -Z(1)*Ka

DZ(2) = (Z(1)*Ka - Z(2)*CL)/V1

END

remark - define algebraic functions

FUNCTION 1

F= Z(2)

END

remark - define any secondary parameters

remark - end of model

EOM

EXAMPLE 2: CODE FOR A MIXED-SPECIFICATION MODEL

Model 1

remark one compartment model - first order input and output

rema

rema no. parameter constant secondary parm.

rema --- --------- -------- ---------------

rema 1 v_f # doses k10

rema 2 ka dose 1 auc

rema 3 k10 time, etc ka half life

rema 4 k10 half life

rema 5 tmax

rema 6 cmax

rema*************************************************************

rema i--------------------i

rema i i

rema ka --> i compartment 1 i ---> k10

rema i i

rema i--------------------i

rema*************************************************************

comm

nder 1

nparm 3

nsec 6

pnames 'v_f', 'ka', 'CL_f'

snames 'k10','auc', 'ka_hl', 'k10_hl', 'tmax', 'cmax'

end

START

Z(1) = 0

END

DIFF

j=1

ndose=con(1)

rema Count up the number of doses administered up to time x

do i = 1 to ndose

j=j+2

if x <= con(j) then goto red

endif

next

red:

ndose = i-1

rema Perform superposition

remark Agut is the total amount of drug in the gut

Agut=0

j=1

do i = 1 to ndose

j=j+2

t=x - con(j)

d=con(j-1)

remark A_ds_i is the amount of drug in the gut associated

remark with the ith dose

A_ds_i=d*dexp(-ka*t)

Agut=Agut+ A_ds_i

next

DZ(1) = (ka*Agut-CL_F*Z(1))/V_F

END

func1

f=Z(1)

end

seco

k10=CL_F/V_F

d=con(2)

auc=d/v_F/k10

ka_hl=-dlog(.5)/ka

k10_hl=-dlog(.5)/k10

tmax=(dlog(ka/k10)/(ka-k10))

cmax=(d/v_f)*exp(-k10*tmax)

end

EOM

Of course, code for implementing infusion dosing or zero-order

absorption processes will differ slightly from the attached examples.

So, if you are faced with either of these cases, and are having trouble

please let me know.

I hope this information proves helpful. If there is any way that I may

be of further assistance, please let me know.

With Best Regards,

Mark

Mark R. Lovern, Ph. D.

Director, Marketing Services

Pharsight Corporation

Phone: (919) 852-4607

Mobile (919) 622-2296

FAX: (919) 859-6871

5520 Dillard Drive, Suite 210

Cary, NC 27511 - On 1 Apr 2005 at 08:48:53, "Walt Woltosz" (walt.-a-.simulations-plus.com) sent the message

Back to the Top

The following message was posted to: PharmPK

[David - welcome back. Up for a little humor today?]

By comparison, you could use GastroPlus and do the following:

Enter the properties of your compound:

MWt

Solubility at a reference pH

pKa(s)

logP or logD at a reference pH

effective permeability (usually estimated from in vitro, or can

be fitted)

Enter the PK parameters (CL, Vd, multi-compartment distribution rate

constants if you have them)

Create a simple text file for multiple dosing that gives the time,

amount, and dosage form (tablet, iv bolus, iv infusion, tablet, capsule,

suspension, etc.) for each dose. If you like, you can also change

physiological models at any time by putting the name of the ACAT model

at the end of each line (e.g., fasted, fed) if you want to simulate such

changes. A sample file might look like this:

'Dosage Form Amount Start Time

End Time Physiology

IV: Bolus 10 0

Human Fasted

IV: Infusion 20 2

8 Human Fasted

IR: Tablet 20 12

Human Fed

IR: Tablet 0 14

Human Fasted

And so on. The last line has no dose, but is used to switch back to

fasted state.

**Warning!**Warning!**Warning!**Warning!**Warning!**Warning!

The next step is essential.

**Warning!**Warning!**Warning!**Warning!**Warning!**Warning!

Click Start to run the simulation.

: )

Walt Woltosz

Chairman & CEO

Simulations Plus, Inc. (AMEX: SLP)

1220 W. Avenue J

Lancaster, CA 93534-2902

U.S.A.

http://www.simulations-plus.com

Phone: (661) 723-7723

FAX: (661) 723-5524

E-mail: walt.-at-.simulations-plus.com - On 2 Apr 2005 at 17:07:20, "Yung-jin Lee" (yjlee168.at.kmu.edu.tw) sent the message

Back to the Top

The following message was posted to: PharmPK

Hi,

I think that Boomer (http://www.boomer.org/) is a very good choice

to pick up, espcially if you like to use differential eqs. to

define your models with multiple dosing. It's not very difficult

to learn how to use Boomer. And it is free.

Best regards,

Yung-jin Lee, Ph.D.

Associate Professor & Deputy Head of

Graduate Institute of Clinical Pharmacy,

College of Pharmacy,

Kaohsiung Medical University,

Kaohsiung, TAIWAN 807

Phone: (07)312-1101 ext. 2262

Fax: (07)313-6316

http://clinpharm.kmu.edu.tw/

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 "Multiple dosing in WNL" as the subject

PharmPK Discussion List Archive Index page

Copyright 1995-2010 David W. A. Bourne (david@boomer.org)