How can be explained this result?

Phoenix Certara Forums Phoenix WNL basics
1

Dear all,
I have some issues with explanation of this result. 90% confidence interval is acceptable but ANOVA results are not.

Can be the reason in this form of PK graph?
All images are in attached files.

Best regards,
Veronica.

1.png
1.png1200×149 85.3 KB
2.png
2.png685×466 70.1 KB
3.png
3.png848×400 29.4 KB

2

Veronika, My reading of the guidelines e.g. in the EMA BE document is;

Statistical analysis
The assessment of bioequivalence is based upon 90% confidence intervals for the ratio of the population geometric means (test/reference) for the parameters under consideration

However your test product does seem to have markedly different profile (is this mean data?) that would concern me regarding consisitency of the formulation’s fragmentation/dissolution.

To answer more fully I think you would have to provide the project and it may be that you need to look at recruiting a consultant to advise you.

WHich parts in the second table are you picking out that concern you? you may be better asking your question at the BEBAC forum…http://forum.bebac.at

3

Hi Veronika,

yes, your results are unusual

From the graph I can see that the mean PK profiles (are they?) are comparable but not close enough.

Due to narrow CI and PE away from 100% there’s a significant formulation effect.

Subject effect - not a big deal

Period effect (equal carry over): the subjects (or IMPs or blood samples) were inflated by the same factor during one period. Which one? Who knows?

May be you will see that if you compare mean graphs like (R period 1 vs R period2) and (T period 1 vs T period2)

Mittyright

4

Mitty, Veronika inadvertently used Report rather than reply to inform me…

Thank you for your reply! It is not a mean profile, it is profile of one of the volunteer.

Simon

5

Mittyright , thank you!

6

Dear Veronika,

assuming balanced sequences I calculated a CVintra of 9.04% for AUClast and 16.3% for Cmax. Given such a low variability the study was “overpowered” for those PK metrics and a significant treatment effect (note: the CI does not include 100%) as expected.

Did you administer two IR-doses four hours apart or was this one dose of a biphasic MR or pulsatile release product? In the latter case the EMA requires not only the PK metrics you already assessed but also the two partial AUCs and the first/second Cmax (see section 6.8.1. of the MR-guideline). If most of the subjects show profiles similar to the one you posted, I think that you may fail to demonstrate BE for pAUC4–t and the second Cmax.