4 replies to this topic

[Omamah]

dear Certara team,

I am working on some data for oral drug that is taken twice daily. The blood concentration was measured after at least 3 weeks of taking the drug and thus steady states is highly expected. 

when I tried to do PK modeling, I found that DV vs IPRED and the DV vs PRED graphs didnt explain the pre-dose concentration. I set the predose as time 0. I checked the concentration at time zero and there were not much BQL. not sure why and how to solve this ?

 

any suggestion will be highly appreciated 

thank you very much

best regards

Omamh

[Simon Davis]

Omamah, without seeing your model set upit is hard to comment.

 

Can you attach your project file?

 

 

Are you performing NCA, WNL  classic or a Phoenix model?

 

Did you inform the model of the preceding 3 weeks of  dosing? the Phoenix model has the option to flag as being at SS, as does NCA, whereas the WNL clasic library models will require you to explicitly  state the doses leading up to your study day.

 

Simon.

[Omamah]

Hi Simon,

thank you very much for the response. I wish I can post the project but it is collaborative work and I will need consent for that. if there is a way I can post my work to better help the understanding kindly suggest.

I did phoenix model, one compartmental (and two compartmental). I tried the additive and the multiplicative error model. I did not perform the PK modeling for the preceding 3 weeks because I have the data of these particular visits (predose, 1, 4, 6 and 8 hours post-dose. I tried to model with and without the steady state. but i still see the same pattern for the predose concentration. but I am not exactly sure what are the numbers needed to fill in the boxes for the steady state option but this is what i did : in the box of the dosepoint I put Aa (i saw an example for that in the help section of the software), in the box of the amount i entered the dose (e.g; 600 mg , eventhough I have some patients who have different dose but I used the majority), and in the box of delta time I put 12 (every 12 hours). 

in the model I selected closed form but I also tried without it and still see the same pattern.

 

best regards

Omamah

[Omamah]

so it seems that there was something wrong when I was trying to flag the steady state. now I got that fixed but I when I run the model I get many times messages that the modelling fail because certain ids couldn't reach the steady state. As I try to fix those through the initial estimate graph adjustment , another message about a different id comes up and so on. I spent alot of time on that and I am wondering if there is away that allow Phoenix to fit the initial estimates for all the ids or do i need to exclude those ids?

 

It might be noteworthy to mention that I changed the time setting, so I calculated the total time from the first time they started taking the drug (about 3 weeks) and until the time the took the dose after which the PK analysis was done 

 

thank you very much

Omamah

Edited by Omamah, 23 August 2018 - 07:25 PM.

[Simon Davis]

I don't think I can help further without seeing the project....if this is confidential data, you can also send to our support team: support@certara.com, where the data will be handled in strict confidence.

Simon.