From: "Gordi, Toufigh" Toufigh.Gordi@cvt.com
Subject: [NMusers] Simulated PD value at time zero 
Date: Tue, February 1, 2005 1:45 pm

Dear all,

I am simulating an indirect response model with drug concentrations
(following a tri-exponential decrease) decreasing the production rate of
the effect. Part of the code is presented here:

$PK
...
BASE=KIN/KOUT
S2=VC
F1=BAV
F5=BASE

$DES
...
CONC=A(2)/VC
HIT=IEFF*CONC/(IC50+CONC)
...
DADT(5)=KIN*(1-HIT)-A(5)*KOUT

What seems strange to me is that the simulated value for compartment 5
(the effect) is equal to 0 and not BASE at time 0 (dose given at time
0.001). I have tried simulation data files with and without a value of 1
as dose for compartment 5 but the results are the same. Sufficient time
after dose, the PD value returns and stays at the BASE value. Also in
the actual NONMEM run, the initial value is equal to BASE.

Any suggestions why the 0 time values for the PD are predicted to 0 and
not BASE?

Toufigh Gordi   
_______________________________________________________

From: "AbuTarif, Malaz" malaz.abutarif@spcorp.com
Subject: RE: [NMusers] Simulated PD value at time zero
Date: Tue, February 1, 2005 3:55 pm

Dear,
I think this is similar to what Max was asking earlier in the week [NMusers] Indirect response PD model. 
Without the drug, baseline PD is basically a steady state situation. NonMem
does not know that. So, the solution to your problem is exactly the solution
to Max's problem, I think. Initialize the effect or PD compartment with a
bolus "PD dose" so it starts from ss conditions level (baseline), or define
the effect compartment as a ss situation.

Malaz

Malaz A. AbuTarif, 
B.Sc. (Pharmacy), Ph.D., G.C.P.M.
Associate Principal Scientist
Schering-Plough Research Institute
2015 Galloping Hill Rd., K15-2-2650
Kenilworth, NJ 07033
Tel: (908) 740-3835, Fax: (908) 740-2916
malaz.abutarif@spcorp.com
_______________________________________________________