7 replies to this topic

[shikha rawat]

dear all forum members,

greetings!

 

Is it necessary to have 3 formulations (slow,inter & fast) for developing IVIVC. i have the in vivo as well as in vitro data of a test & reference. how to validate the correlation model formed by using above two formulations (taking test as internal val ) as we don't have any formulation for external validation. and what conclusions can we draw from the levy plot attached below.

 

thanks in advance

stats06

[Simon Davis]

Hi StatsO6, I don't see an attached image of the plot currently - if you have problems attaching you can always email me and I'll add it for you.

 

I think the absolute minimum is two but the more you have the better, will be able to comment more when we see your data,

 

  Simon

 

PS If you are attending APS in Nottingham, consider enrolling in Pharsight IVIVC Toolkit for Phoenix WinNonlin - Sept. 11, 2012

 

The course price is $636 if booked by 11 August 2012. After this date the course price is $795, Student and group discounts also available.

 

http://www.pharsight.com/training/enroll_new.php?course_id=674

[shikha rawat]

hi Simon,

 

thanks for your reply . I am attaching the image for your reviews.

i have another query , suppose i have in vivo data from two studies of (T vs R) with slight change in the polymer ratio in the test product of the 2nd biostudy. now can i consider both the study data simultaneously in IVIVC fomulation ? if yes then which values of the reference (1st or 2nd ) can be used in the reference tab .

 

thanks

stats06 [file name=levy_plot.bmp size=1002118]http://pharsight.com/extranet/media/kunena/attachments/legacy/files/levy_plot.bmp[/file]

Attached Files

•  levy_plot.bmp   978.63KB   454 downloads

[Simon Davis]

Stats06,
Your plot of FDiss vs Fabs shoes that currently your experiments are not especially well matched and you will have to account for that in your correlation model.

Also it only shows Test, you should have reference here too to see if it they are similar shapes.

Regarding your other query, it think the sentence should read;

now can i consider both the study data simultaneously in IVIVC formulation ? if yes then which values of the TEST (1st or 2nd ) can be used in the TARGET field tab . (Target could be a copy of the REFERENCE set if for instance you were trying to match the innovator's formulation

there is only one REFERENCE that you can compare multiple test formulations against at the same time, i assume you have in vitro date for this too, try plotting all on one plot Ref, Test1 and test2 and posting that.

Simon.

[shikha rawat]

hi simon,

 

thanks for your reply..Sorry, was in a hurry yesterday so made a lot of mistakes while posting

 

 What i meant was that the same RLD was evaluated twice in the first as well second study, so which data should i consider here? IF the innovator (RLD) data is used in the target field i am not able to ake take it as the reference in the invivo field. (I am doing all this using IVIVC wizard.)

 

The second way in which i tried to built a correlation model was

 

1.) Deconvolving the RLD in vivo data (using wagner nelson approach as the dosing route was oral & after visually examining the plots i think a one compartmental model will fit my data) and plotting it against its vitro data to have an idea about how my disso data is behaving.

 

2.)using the above formed model to deconvolve the test in vivo data.

 

3.) plot both the RLD as well as test in a single plot and built a correlation model.

 

is the above approach corrrect or am i missing something?

 

thanks & regards

stats 06

[Simon Davis]

Sorry I don't really follow what you have got - can you post the project please?

 

(my guess is your 'vivo' data for reference came from a replicate BE study e.g RTR - probably you would include it all i.e. data form both periods for each subject but let's see if someone with more experience of using this data has some suggestions.)

 

 Simon.

[shikha rawat]

hi simon ,

 

thanks for your reply and sorry if i was'nt clear in my last post as well ...

 

You have guessed it right ... Actually we conducted two studies , firstly a partial replicate (RTR) and then a fully replicate (TRTR,RTRT) .

 

(if its still not clear i'll email my project to you)

 

Now i am confused how to make full use of this data and build a correlation model. I have the in vitro (i.e dissolution data ) of these two studies as well.

(I tried to build the correlation model using the first two periods data from the second study , the Levy plot of the same was attached in my earlier post)

 

@all forum members , has anybody ever done such a study?

 

thanks & regards

stats06

[Simon Davis]

Stats06 - i am going to be out of the office for a few days but if you can upload your project here perhaps someone else can take a look at your data, and make some suggestions.

 

  Simon.