Can Phoenix include Cmax or only timepoints after Cmax for predicted Y vs X?
When using the best-fit method, Phoenix will not include Cmax for non-bolus models. But a user-specified range can be used to include Cmax.
Hi, this is not a good idea since absorption is not completed at t[sub]max[/sub]. Even the first time point after t[sub]max[/sub] might be to early. For a one-compartment model the inflection point of the profile (at 2× t[sub]max[/sub]) is a good starting point. See: Scheerans C, Derendorf H and C Kloft Proposal for a Standardised Identification of the Mono-Exponential Terminal Phase for Orally Administered Drugs Biopharm Drug Dispos 29, 145–57 (2008)
Thanks Helmut, very helpful! For first ‘screening’, I usually use the Phoenix best-fit method (and I wondered whether Phoenix ever includes Cmax, which now is clear to me from lhughes’ comment), and then look for outliers and differences when running it with user-specified ranges, sometimes with different starting points.
Hi, this was a major change in Phoenix compared to “classical” WinNonlin where t[sub]max[/sub] might have been included. Personally I’m not happy with the max R ²[sub]adj[/sub]-method anyway. Fine as a first screening, but visual inspection of fits is mandatory anyway. Max. R ²[sub]ad[/sub]j regularly fails in fitting multiple peaks (pulsatile formulations) and controlled release formulations (“flat” peaks). I hope that Pharsight considers implementing alternative methods (e.g., TTT, max AIC) as an option in a future release. 
Helmut, Can you please elaborate on “How Adj R2 approach fails in multiple peaks and flat peaks cases and in what ways is TTT method better in such cases” Regards Kaushik
Kaushik, there is quite some discussion on this method Bioequivalence and Bioavailability Forum • bear v1.0.0 - a data analytical tool for ABE in R personally I’m not sure it is ‘worth’ the effort of applying as it is primarily intended for 1 compartment curves whereas it is in profiles with a greater number of exponentials that cause more issues in my experience when trying to apply an NCA anaylsis. Simon.