Bioequivalence and Bioavailability Forum

Hi Angus,

❝ I wonder if the FDA approach has been evaluated in comparison with SAS?

Yes.*

❝ How about dropout subjects?

No problem.

❝ I thing there was a progesterone example from the FDA and it used SAS code, but I do not remember if input data was provided.

The FDA didn’t publish example data sets – only the SAS-code. Therefore, in validating the templates we used the data sets published by the EMA. You can download both in Excel-format here.

1. 4-period 2-sequence (RTRT | TRTR) fully replicated, imbalanced (77 subjects), incomplete (missing periods: one period in six cases, two periods in two cases).
   
2. 3-period 3-sequence (TRR | RTR | RRT) partial replicate, balanced (24 subjects), complete (no missing periods).

Results of both in PHX6.3 matched the ones obtained in SAS9.2. Additionally I removed period 3 of subject 24 in the second data set in order to get an incomplete design. Again, same results in PHX and SAS.

---

• Henry A, Schütz H, Hughes L, Davis S. Performing Reference-Scaled Average Bioequivalence (RSABE) in Phoenix^® WinNonlin^®. AAPS Annual Meeting and Exposition; 2013:Poster T2350. doi:10.13140/RG.2.1.1008.0800.  free resource.