From: "kai wu"
Subject: [NMusers] indirect response model 
Date: Wed, August 18, 2004 9:50 pm

Dear NONMEM users,
  I am trying to use ADVAN6 to fit a indirect response model for a
  endogenous substance suppression by drug. 
The baseline follow a circadian rhythm, this rhythm (Rcort) is controlled by
four parameters which I want to estimate. I do have baseline profile for each
subject. The question is how should I set up the data records and control stream
to incorparate the baseline into my dataset. I  was able to fit the data without
baseline data, but the prediction for time 0 is all 0. Since in my case Rcort is
not a parameter to estimate, the F4 method does not work for me. 

Thanks in advnace!
Kai Wu
Department of Pharmaceutics
University of Florida
Gainesville, Fl 
Office phone #: 352-846-2730

From: "Nick Holford"
Subject: RE: [NMusers] indirect response model 
Date: Wed, August 18, 2004 11:16 pm


Presumably if you could arrange for a constant rate input of something you could
write a model to transform the constant input into a parametric circadian rhythm.

If you set up a large AMT e.g. 1000000 and a nominal RATE of 1 with CMT=4 then you
can have a constant input to play with to make Rcort a rhythmic input for DADT(4).
F4 would be used to set an appropriate scale for the input.

Nick Holford, Dept Pharmacology & Clinical Pharmacology
University of Auckland, 85 Park Rd, Private Bag 92019, Auckland, New Zealand tel:+64(9)373-7599x86730 fax:373-7556

From: "Luann Phillips"
Subject: RE: [NMusers] indirect response model 
Date: Thu, August 19, 2004 10:41 am 


I would try adding your baseline measurements to the dataset starting at
time=0,initialize the amount in compartment 4 to an appropriate value,
and include your equation for Rcort in $PK if it is not dependent on
continuous time (TIME) or in $DES if dependent on continuous time (T).

To initialize the amt in cmt=4, I would add an additional dose record
with cmt=4 at time=0 with amt=the measurement of the endogenous
substance at time=0 (i.e., the first observation from your baseline data
would double as a dose and a measurement).  

If your concerned about measurement error for the time=0 measurement of
the endogenous substance, you can estimate F4=1+eta(n) or F4=
1*(1+eta(n)), etc. where the eta would represent the measurement error
for the substance at time=0.

Note: Because all of the baseline records for the endogenous substance
would come before the first dose of the drug, cp will be zero throughout
the baseline period and eff=1.