6 replies to this topic

[RaphBil]

Hello,

 

I'm trying to model data from an extravascular sustained-release formulation which exhibits nonlinear release rate. Is there any option in WinNonlin to model infusion that are not of zero order?

Thanks!

 

Mendy

[Simon Davis]

Mendy, Using the graphical model builder you can set up to have several dose points to the absorption compartment.  And modelling of each one can be by default first order or zero order; the latter having several options as indicated.

Perhaps by using a combination of these  options you can model your profile, please let us know how you get on - if you want some more help I think we'll need some more detail e.g. can you give us an example of the dataset that you are trying to fit? 

  Simon.

Attached Thumbnails

• 

[RaphBil]

Mendy, Using the graphical model builder you can set up to have several dose points to the absorption compartment.  And modelling of each one can be by default first order or zero order; the latter having several options as indicated.

Perhaps by using a combination of these  options you can model your profile, please let us know how you get on - if you want some more help I think we'll need some more detail e.g. can you give us an example of the dataset that you are trying to fit? 

  Simon.

Hello Simon,

 

Thanks a lot for your reply.

I have actually a single administration in an extravascular compartment of a depot (sustained-release formulation) containing fixed amount of drug.

This depot is releasing the drug for a long period of time (1 month) but not at constant rate.

I have concentrations both in the extravascular compartment and in plasma over the 1M period.

Is that enough info to help you better understand the problem?

 

Mendy

[Simon Davis]

It might be, I get that you say it is not releasing at a constant rate, but is it releasing in some 'predictable' way, I mean when you created the forumulation matrix did you have some target rate of release or is it changing in a 'random'/unexpected way.

I'm intrigued as to what form of matrix you're sampling in the extravascular compartment - or are you sampling from within the actual Depot ?implant?

 

 Simon.

PS to attach items e.g. CSV, XLS, PHXPROJ etc. then you will probably have to click the button 'more reply options'

 

[RaphBil]

Actually I have only in vitro data, but no idea of what is the release rate in vivo.

The matrix is a body fluid which is not blood, and the drug is diffusing from the depot in that fluid and can also leak to the systemic circulation.

I have concentration in the fluid and in plasma over a month.

Unfortunately I cannot disclose more info on the forum.

 

Thanks for your help!

Mendy

[Simon Davis]

OK in that case I'd be tempted to built up something like this, you can alter the Ka rates as you see fit. but basically I've given you a depot compartment to which you dose.

 

From there KaDiff is rate to ECM, (you have ECMObs for these) and then Ka to go to Plasma (CObs)

 

plus there is KaLeak which is straight to Plasma, (for which  you have CObs)

 

You may of course have identifiability issues, perhaps KaDiff could be fixed to the in vitro rate you have.

 

 Simon.

 

PS Remember you may also contact Certara directly for consulting in complete confidence if you have some budget to address this problem; some our customers combine a training course with a workshop to show how a model such as this is developed and tested/validated.

Attached Thumbnails

• 

Attached Files

•  Depot_ECM.wnlt   42.72KB   356 downloads

[RaphBil]

OK in that case I'd be tempted to built up something like this, you can alter the Ka rates as you see fit. but basically I've given you a depot compartment to which you dose.

 

From there KaDiff is rate to ECM, (you have ECMObs for these) and then Ka to go to Plasma (CObs)

 

plus there is KaLeak which is straight to Plasma, (for which  you have CObs)

 

You may of course have identifiability issues, perhaps KaDiff could be fixed to the in vitro rate you have.

 

 Simon.

 

PS Remember you may also contact Certara directly for consulting in complete confidence if you have some budget to address this problem; some our customers combine a training course with a workshop to show how a model such as this is developed and tested/validated.

Thanks a lot for your help.

I'll consider the option you just mentioned.

Best,

Mendy