Jump to content


Photo

Transit compartment PK for absorption delay


  • Please log in to reply
7 replies to this topic

#1 jgjguo@gmail.com

jgjguo@gmail.com

    Newbie

  • Members
  • Pip
  • 7 posts

Posted 23 October 2020 - 05:12 AM

Hi,

 

I am trying to build a transit compartment  model to fit the plasma PK with two peaks especially at  high dose groups. It is reported that  transit compartment is a good choose. I add a transit statement. But unfortunately couldn’t fit data. No effect was observed when I tried to adjust mtt, ktr and ntr. Sounds like there is a glitch.

 

It will be appreciated  if you can help in  fitting the PK data to transit compartment model. I am attaching project file with existing mode for only high dose PK.

 

Thanks

 

Jian

 

Attached Files



#2 Simon Davis

Simon Davis

    Advanced Member

  • Administrators
  • 1,122 posts

Posted 23 October 2020 - 08:20 AM

i think it will be more informative if you fit all the data you have i.e. with the different dose levels.  I will try to take a look at your project at the week end.  Simon.



#3 cradhakr

cradhakr

    Newbie

  • Members
  • Pip
  • 9 posts

Posted 23 October 2020 - 01:40 PM

Had a quick look at your project file and looks like this is not a Transit model. I have attached a link to a YouTube/PML school on enterohepatic circulation and hope this should be useful for you. From the forum, you could download the PML code for your reference.

 

https://support.certara.com/forums/topic/990-lesson-7-enterohepatic-recirculation/

 

 

Thanks and regards

Chandramouli 



#4 f_yc

f_yc

    Newbie

  • Members
  • Pip
  • 4 posts

Posted 23 October 2020 - 02:19 PM

Hi Jian,

Obviously wrong, you need to give At1 an initial value, which needs to be done through the dosepoint() statement.
like:
dosepoint(At1)

But in order to distinguish it from the original Aa and share the same total dose, you'd better add another bioavailability parameter F.
So:
dosepoint(Aa, bioavail = (f))
dosepoint(At1, bioavail = (1-f))
fixef(f = c(, 0.5, ))

Attached Files


Edited by f_yc, 23 October 2020 - 02:21 PM.


#5 jgjguo@gmail.com

jgjguo@gmail.com

    Newbie

  • Members
  • Pip
  • 7 posts

Posted 23 October 2020 - 04:17 PM

Thanks Chanramouli for your feedback and forwarding PML code,  Agree, enterohepatic re-circulation is an option for modeling this kind of double peak PK. I did create a  model. It fit the data. 

 

Thanks

 

Jian


Edited by jgjguo@gmail.com, 25 October 2020 - 04:14 PM.


#6 jgjguo@gmail.com

jgjguo@gmail.com

    Newbie

  • Members
  • Pip
  • 7 posts

Posted 23 October 2020 - 04:48 PM

Awesome!

 

Thank you so much for your help! Having two parts of dosepoint statement with bio-availability parameter makes much sense and is simple for transit compartment. Thanks for the coding as well. 

 

Best,

 

Jian 


Edited by jgjguo@gmail.com, 25 October 2020 - 04:14 PM.


#7 jgjguo@gmail.com

jgjguo@gmail.com

    Newbie

  • Members
  • Pip
  • 7 posts

Posted 25 October 2020 - 04:27 PM

Hi,

 

I  tried to use this transit model to fit three dose PK. Sounds like model fit  100 mg and 200 mpk dose well, but not on 30 mpk. I did minor modification by adding saturable absorption and elimination based on attached paper.  Author emphasized that this model can describe PK with late absorption peak at HIGHER doses (that is exactly my case here). But it is no significant improvement on my case. Model  showed second peak on low dose 30 mpk.  Paper and my project are attached. Do you have any suggestions?

 

Many thanks

 

Jian

Attached Files


Edited by jgjguo@gmail.com, 25 October 2020 - 05:48 PM.


#8 bwendt@certara.com

bwendt@certara.com

    Advanced Member

  • Administrators
  • 190 posts

Posted 16 November 2020 - 02:26 PM

Hi Jian,

 

a major problem with the models in your project file is that they have too many parameters, only a small fraction of the parameters seems to be identifiable. That is why I am thinking that the idea of a transit model does not fit here.

There is one additional alternative to the transit model that you are in favor of:

 

In one of our PML School we presented a scenario with sublingual dosing. At first sight, this might not be related to your model, however, it describes a similar scenario: 

 

you have one immediate fast absorption

followed by slow absorption delayed by some time

 

Please see the link to our materials here:

 

https://support.cert...-model-builder/

 

I have adapted this model to your data, please see the additional object in the workflow called PK43:

 

Attached File  mouse PK_43.phxproj   6.01MB   3 downloads

 

Let me know if there are questions or comments.

 

Bernd

 






0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users