6 replies to this topic

[hppatel19886]

Hi,

 

I would like to calculate PK parameters in Phoenix for the IV bolus plus IV infusion. Could anyone tell me how to do?

 

The scenario is here.

 

1 gm IV bolus at 0 min and 333 mg/hr infusion started after 5 min and continued for 5 hrs. I have data points from 5 min to 5 hrs. Please tell me how can I calculate the NCA PK parameter?

I am unable to add both dosings in NCA. It would be great if you can show how to add dosing and how to prepare a data file for NCA. I am new to Phoenix.

 

 

[Simon Davis]

NCA theory and models are defined for a single dose event, if you give more than one dose even it will return results from only the last dose for each profile as defined by the sort keys .  If you want to analyse this as an NCA then I think you will have to make a compromise of defining this as an infusion only. However I would recommend that you avoid this design in future and understand where the differences may come from.

 

In the attached project I have performed a number of simulations with clearance of 6L/h and 60L/h  (typical and high values for a drug)  and NCA's on a very rich profile from the simualtion and one with a more realistic sample schedule.

Subject 1 is set up to be simualtes with the true dosing info and Subject 2 as if the same total amount of drug was all given as a constant infusion 0 to 5 hours.  You can see in the table below how some common parameters might vary in their estimation.

 

I hope this helps, SImon.

 

PS   I'm not sure where you're based but we hold face to face training courses all around the world, the next one in India is

 

Bengaluru, India – Introduction to Phoenix WinNonlin with IVIVC Toolkit, April 2-4, 2019 – 2 Registrations (Chandramouli Radhakrishnan

 

and the whole 2019 calendar  is here

https://s3.amazonaws.com/certara-training/LMS/website/files/phoenix/schedule/2019+Training+Calendar+by+Location.pdf

 

Alternatively you can take on demand Training over the web where you will have the same content but you can watch the videos at your own pace, please explore these options by registering here: http://www.certarauniversity.com/lms/

 

Attached Thumbnails

• 
• 
• 

Attached Files

•  IVbolus_Inf.phxproj   1.58MB   557 downloads

[neotrident2018]

Dear Simon,
It is a good topic. Recently, I met the same problem. I recommendate the infusion type to be selected in NCA and the dose option set the toatal dose. but there woule be a problem that some paramenters will not same as real toatal dose infusion. The type of data is acquired from the clinical Phase I. when the NCA method is recommendated by the guidline, you think if we can use the compartmental modeling to calculate the PK parameters for this type of data?

I will Highly appreciate if you can clear it.
Fu Yangyang

Edited by Simon Davis, 23 January 2019 - 08:16 AM.
removing re-quoting of original post

[Simon Davis]

Correct, for an NCA, I think you have to approximate it as one infusion. My belief is that you will still need to use NCA as your primary analysis, but I think you should be very clear yourself, and to those whom you present your results, of the limitations and assumptions of this approach.

It maybe useful to present a secondary analysis using compartmental fitting, however I think it would be better for these conversations to be had before the study is run - and perhaps avoid these designs where possible.

Simon.

Edited by Simon Davis, 23 January 2019 - 10:51 AM.

[neotrident2018]

Dear Simon,
Thanks so much for your efforts. I think sometimes the infusion is OK. However, you can see the attached plot. The dose regimen is 1500ug via bolus in 1 minute and then dosing 4500ug via infusion in 8 hours. The first blood sampling time is 5 minutes aftter dosing. I think the bolus may be more suitable for this type of data. Because IV Infusion type estimate C0 as value of zero, while the IV Bolus type could calculate C0 via extrapolation. Do you think it is reasonable for this thought?

Attached Thumbnails

• 

Edited by Simon Davis, 23 January 2019 - 10:13 AM.
removed the quoting of previous post

[Simon Davis]

C0 is indeed the parameter that is most impacted in these examples. I htink you have two options, and perhaps you cna do both of these.

 

a) just report the observed first conc as C0, in your case it looks like that's not a bad estimate and you can expalin why.

 

perform two NCAs, the main analysis based on the infusion dose and all data, and a second that focus on just the IV bolus segment to 7 or 8 hours to report the extrapolated C0.  Remember if you have an observation at time 0 that is what will be reported regardless, so be careful about having a BLQ set to 0 at dose time., better to leave as blank or non-numeric e.g BLQ or <10

[neotrident2018]

Dear Simon,
Thank you so much for your warm heart.

Fu Yangyang

Edited by Simon Davis, 23 January 2019 - 01:39 PM.
removed the quoting of previous post