From:michael.looby@pharma.novartis.com
Subject:[NMusers] Repeat administration with varying dose interval
Date:Wed, 22 Jan 2003 11:44:04 +0100

Dear NMusers

We have a problem that involves repeated administration of a formulation
that is released by a first order process, but this absorption process is
completely terminated just prior to administration of the next dose.
Furthermore, the inter-dose interval is not constant. How can we code this
within NONMEM?  It is our understanding that this kind of problem would
involve resetting the dosing compartment prior to each subsequent dose in a
mutliple dose setting without resetting the observation compartment.
We would be grateful for your input.

Kind regards
Mick
_________________________________________

From:sophie Callies 
Subject:[NMusers] Re:Repeat administration with varying dose interval
Date:Wed, 22 Jan 2003 14:26:40 +0100 (CET)

Dear Michael

My suggestion to your problem would be to put in the
data set for each patients all the dose record with
their respective time calculated from the first dose
given. This will take into account to problem of the
variability in the dose interval.

Then to had a event identification column EVID (see
nonmem manual) which would be equal to :
4 for when time = 0 (first dose of each patient)
1 for the other dose records with time different of 0
(this reset the dose comparmtent whitout reseting to 0
the amount in the observation compartment)
0 for the observation record

The only issue with my suggestion is the number of
records per patient allowed by NONMEM.

I hope that my suggestion will help

Sophie Callies
_________________________________________

From:"Gastonguay, Marc" 
Subject:[NMusers] FW:Repeat administration with varying dose interval
Date:Wed, 22 Jan 2003 09:13:20 -0500

Mick,
One possible solution is to insert dummy (EVID=2) data records at the time
of termination of the first-order process. On these records, include a
negative dosing compartment number in the CMT data item. Then, insert a
second dummy record (EVID=2) immediately after to turn the compartment on
again with a positive compartment number.

For example, if the dosing compartment is CMT 1 and the absorption process
ends at 8 hours, you could insert the following records:

TIME AMT CMT EVID etc.
8     .   -1  2
8     .    1  2
(continue with data records)

This allows you to turn on/off a single compartment without resetting the
entire system. You'll have to do this for every dose in the data set.

Good luck.

Marc Gastonguay
_________________________________________

From: michael.looby@pharma.novartis.com
Subject: [NMusers] RE:Repeat administration with varying dose interval
Date: Wednesday, January 22, 2003 9:10 AM


Hi Marc

Do you have an idea of how we might apply this for steady-state dosing?

KInd regards
Mick
_________________________________________
From:"Gastonguay, Marc" 
Subject:[NMusers] RE:Repeat administration with varying dose interval
Date:Wed, 22 Jan 2003 09:22:46 -0500

Sorry, Mick, I don't know how to do this with the SS or ADDL short-cuts. I
would just include all dosing records. I'll give this some thought, though.
Maybe someone else has an idea?

Good luck.
Marc
_________________________________________

From:"Sam Liao" 
Subject:RE: [NMusers] Re:Repeat administration with varying dose interval
Date:Wed, 22 Jan 2003 09:37:13 -0500

Dear Michael, Sophie:
It may be easier to specify the dosing event and PKPD observation by the
actual date and clock time in DATE and TIME variables.  NONMEM will
calculate the elapsed time for you.

 Sam Liao, Ph.D.
PharMax Research
20 Second Street,
Jersey City, NJ 07302
phone: 201-7983202
efax: 1-720-2946783
_________________________________________