Bioequivalence and Bioavailability Forum

Hi RK,

❝ […] EMA guideline, it state that ctau is the concentration at the end of the dosing inter­val at steady state. I am not able to find out any parameter in this name in the soft­ware.

AFAIK, correct (not available in “classical” WinNonlin, Phoenix/WinNonlin, Kinetica, EquivTest/PK).

❝ […] ctau=clast?

Not necessarily.* See this thread for some ideas in Phoenix/Win­Non­lin. If you are able to estimate λ_z (≥3 data points) I suggest to use the estimated concentration at t=τ (Ĉ_τ). See also this pre­sen­tation (slide 17 on what I state in my protocols). We face a similar problem in single dose if t_last of the test ≠ t_last of the re­fe­rence. If we compare AUC_t it might mean comparing AUC₁₆ to AUC₂₄. That is another example of “apples-and-oran­ges” sta­tis­tics. However, since AUC is an integrated metric the impact on the BE assessment is minor. C_τ (like C_max) is a single-point metric. The impact on BE might be tre­men­dous. A funny story here (C_last would have failed BE). Two of my studies passed an MRP last year (RMS Germany; CMS Austria, Den­mark, Spain, Sweden, The Netherlands) without any concerns.
Note that Phoenix / Win­Non­lin (if steady state is selected) would estimate AUC_τ. You’ll notice that, since if t_last ≠ τ then AUC_last ≠ AUC_τ. Would be nice if Pharsight implements C_τ in a future version of Phoenix / Win­Non­lin. If I have spare time I’ll check the latest beta-release of v6.4 (can’t promise).

---

• Only if you would have no missing last concentrations, the con­centration at τ > LLOQ, and no time deviations (t_last = τ).