1 reply to this topic

[mjons]

Hi,

I have a somewhat complex setup for powder dissolution testing that I would like to model in Phoenix. The issue is that I would like to be able to fit parameters on multiple levels in the model:

-Some parameters I would like to fit per individual test

-Some parameters I would like to fit per formulation (but keep constant between replicates of the same formulation)

-Some parameters I would like to fit as a single parameter for all tests (across formulations and replicates)

 

Could this be achieved in WinNonlin?

[Simon Davis]

In short yes this should be possible since you can write pretty much anything you would like and freeze certain parameters.  If you haven't written much code before then I would look at https://www.certara....ing/pml-school/

 

 

and there is a searchable library here; https://certara.gith...ml_library.html

here's one example. Description: PML model name Time Dependent Dissolution

Reference: Gabrielsson J, Weiner D. Pharmacokinetic and Pharmacodynamic Data Analysis, Concepts and Applications. 5th ed. Sweden: Swedish Pharmaceutical Press; 2016.

    #This is an example of how to implement a conditional
#model based on time.  The example is a dissolution
#model where at time<tx, Fdiss=kd*(Time^n) and where
#time>=tx, Fdiss=A^(Time).  tx is estimated.
#Note that the C++/C syntax for a switch statement is used
#(cond) ? actionA:actionB.

test(){

covariate(Time)
stparm(Fdiss = Time<tx? kd*(Time^n): A^(Time))
error(FdissEps = 1)
observe(FdissObs(Time) = Fdiss + FdissEps)
fixef(tx = c(,1,))
fixef(kd = c(, 1, ))
fixef(n = c(, 1, ))
fixef(A = c(, 1, ))
}