variable tlast

Phoenix Certara Forums Phoenix WNL basics
1

Hi,

I have a clinical study that allowed final sample to be collected at any time, so after the final in-clinic sample was collected for all subjects at 120h a follow up sample was collected anywhere between D7 to D11. For concentration summaries by timepoint, would you use a predicted concentration for a consistent tlast (e.g. 192h)? For determination of lambda z, could you limit lambda z upper to 120h?

Thanks

2

Are all samples quantifiable at these later timepoints?

I agree that personally I would use some predicted common timepoints, rather than describing it as Tlast as I don’t really think that’s appropriate if they are still quantifiable at these later times.

I certainly would not force Lz calculations to be prematurely end at 120h. I’m not sure how variable your profiles are and if you see with later sampled profiles ‘extra’ phases that means your half life appears longer/different. But that is something you will have to review and include in your discussion,

Simon

3

Hi JCPPM,

I agree with Simon. Perhaps this article is useful.

Did you administer only one formulation or more – aiming at comparative bioavailability or dose proportionality? If the latter, a comparison of AUC0–tlast will be biased. For an unbiased comparison either use AUC0–∞ or – for each subject – AUC to the earliest tlast of all formulations.

  • Fisher D, Kramer W, Burmeister Getz E. Evaluation of a Scenario in Which Estimates of Bioequivalence Are Biased and a Proposed Solution: tlast (Common). J Clin Pharm. 2016; 56(7): 794–800. doi:10.1002/jcph.663. Open Access.
4

Hi Simon and Helmut. Your input is very helpful thank you! I especially appreciate the reference which addresses the question directly. Indeed we had BLQ samples as well as quantifiable samples. We only evaluated one formulation but at multiple dose levels so yes evaluating proportionality. Per your comments, Helmet, and upon reading Fishers et al, it does seem that determining AUC0-t(common, eg. 120h) would mitigate the bias introduced by differences in timeframe. Do I interpret correctly? Thank you!

5

Hi JCPPM,

Yes, you do.

6

Hi Helmut, Thank you for confirming!