Jump to content


PK3 does not match NCA parameters

  • Please log in to reply
1 reply to this topic

#1 boothdm@auburn.edu



  • Members
  • Pip
  • 1 posts

Posted 12 April 2020 - 03:09 AM

I have used the same data for NCA and then, using those parameters to generate initial parameters from Phoenix model,  I have run a PK3 model. Single dose data. The mg/kg dose differed among the subjects (n=7) by two fold so actual Cmax and AUC vary. 
When PK3 is used, Cmax and AUC are about 10 fold less than they should be based on NCA. why the difference? (This is a slow/extended release product, and ka and ke, based on phoenix model, are similar: 0.58 and 0.33 average, respectively). 
My ultimate goal is to simulate multiple dosing at 8 (50 mg/kg), 12 (75 mg/kg) hrs, but when I do so using phoenix model, my predicted concentrations drop dramatically over 24 hrs and then do not increase as might be expected when simulating dosing at intervals equivalent to 2 half-lives. 
Are the two issues related?


Edited by boothdm@auburn.edu, 12 April 2020 - 03:41 AM.

#2 Simon Davis

Simon Davis

    Advanced Member

  • Administrators
  • 1,112 posts

Posted 12 April 2020 - 03:22 AM

The reason would depend on how well PK3 fits to the data, NCA is obviously only reporting observed points, which leads me to think that PK3 is not an optimal model, i would suggest to use the phoenix model that has much more flexibilty for writing your own models to better describe the data.


take a look at this post;

...and see if something like that helps, do you need to consider a tlag?

if you're still stuck please try posting some example project with any confidential info removed e.g compound name.

Edited by Simon Davis, 12 April 2020 - 03:23 AM.

0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users