Bioequivalence and Bioavailability Forum • IVIVC output

moulirc
☆

Australia,
2014-02-07 14:49
(4514 d 07:44 ago)

Posting: # 12364
Views: 6,056

IVIVC output [Dissolution / BCS / IVIVC]

Hi, i am trying to run IVIVC for a modified formulation. this is my final pk parameters, can someone please let me know how i can interpret these results.

Formulation  Parameter  Predicted   Target     %PE         Ratio

L1           AUClast    15.867095   19.928375  -20.379386  0.79620614

L1           Cmax        6.7290543   8.72      -22.831946  0.77168054

L2           AUClast    10.00431    19.928375  -49.798668  0.50201332

L2           Cmax        4.46        8.72      -48.853211  0.51146789

L3           AUClast     3.7605866  19.928375  -81.129487  0.18870513

L3           Cmax        1.59        8.72      -81.766055  0.18233945

Thanks
Kind Regards
Mouli

Edit: Category changed and table BBCoded. [Helmut]

JMCardot
☆

France,
2014-02-08 23:12
(4512 d 23:21 ago)

@ moulirc
Posting: # 12372
Views: 5,030

IVIVC output

Post reply

Hi Mouli,

Based on a first look I do not understand what is the "target" in your table, why you have the same target for 3 test formulations, etc… :confused:

Either it is a mistake and you have 3 different observed (corresponding to L1 and L3) or
L1 to L3 are 3 possible dissolution methods used to predict in vivo or
you simulated various options of new formulations vs only one theorical target (!) based on an existing IVIVC that you have establish before but not share with us. In this last case results are possible.

Case (I) In the first case you must have the observed results in one column and the predicted in the another and the %PE in a third one, etc… If it is for a validation of an IVIVC with 3 formulations (internal for example) according to guideline you have limits to accept your IVIVC, for FDA and internal predictability it is «Average absolute percent prediction error (% PE) of 10% or less for Cmax and AUC establishes the predictability of the IVIVC. In addition, the % PE for each formulation should not exceed 15%.❝ In your case that would correspond to the fact that you have not validated your IVIVC.

Case (III) Now let have a look on your data assuming it is 3 different new test formulations vs a target profile. First the error is similar for Cmax and AUC leading to a loss in extent (lower curve but correct shape). Second you have (may be by chance) a nice factor 2 between each formulation. The questions are multiple to explain it. Find some idea below:

Incomplete dissolution
Possible mistake in the initial absorption evaluation of initial IVIVC (flip-flop, etc…)
Time scaling problem: a faster absorption than dissolution (bad disso method)
if L1 is the fastest and L3 the slowest release => very long release vs GI transit time (F problem)
Incorrect parameters for the deconvolution and or convolution
etc…

May I ask you to explain better your problem as here it is not clear (at least for me) what you have done and what you expect? Which type of IVIVC (based on what), what is L1 to L3 etc…

Best regards,

JM

moulirc
☆

Australia,
2014-02-09 02:27
(4512 d 20:06 ago)

@ JMCardot
Posting: # 12373
Views: 4,902

IVIVC output

Post reply

HI JM,

sorry for not being specific, and thank yo for your post.

This is the background.

L1, L2, L3 refers to three formulation i am trying to develop (extended release), different concentration of the matrix.
L1 being the lowest concentration of the matrix, and in the dissolution study more drug was released within the first 30 minutes
L2 - medium concentration of the matrix
L3 - highest concentration of the matrix (less drug was released from the dissolution study)

Reference formulation: this data was extracted from a journal article (immediate release formulation).

its case 2, three different formulation trying to predict invivo pk profile.
We are expecting that L2 and L3 we will be having bio-availability issue and we are trying to prove this.

Please let me know if you need more information

Thank you

Mouli

JMCardot
☆

France,
2014-02-09 19:29
(4512 d 03:04 ago)

@ moulirc
Posting: # 12374
Views: 4,932

IVIVC output

Post reply

Hi Mouli,

In this case if I understood you have a “theoretical IVIVC” and you used your dissolution as 1:1 input function in the convolution vs the published IR.

In this case the results could be easily understood

– Due to dissolution

if the dissolution of your 3 SR formulations are slow and do not reach 100% in your dissolution test.
if you stop the dissolution too early
if you stop to use the in vitro data at a certain time (why ?)

– Due to your “IVIVC” The relation is not a 1:1 as you assumed

To establish a real IVIVC you must have in vivo data with your formulations at a certain moment of the development.

Best regards,
JM