Bootstrap f2 - which software? [Software]

posted by Shuanghe  – Spain, 2019-01-07 13:15 (653 d 13:00 ago) – Posting: # 19742
Views: 2,867

Dear MSimoes,

» I have been using DDSolver, but recently I heard that the Spanish agency replied to a dossier submission saying that it is not acceptable because of miscalculations.

From the information I got from one meeting with the agency, they use R that was developed by a professor in certain university. I suggested that they should publish the code with the seed number so everyone can use and reproducibility wouldn't be an issue. Haven't seen any thing published yet.

» do you know why the DDSolver is not recommended?

I've never used DDSolver's calculation for f2 or bootstrapped f2 for regulatory submission. I have no idea how internal code of DDSolver works but from the f2 calculation I saw, it seems that the code doesn't evaluate if your number of points is valid according to EU or US guidance (By that I mean to say the CV% of early points, how many points > 85% release etc). the results of f2 depends on how you select the points. If bootstrapped f2 also works similarly, then I wouldn't be surprised if the agency refuse to accept it.

» What about Ph-Eq, any feedback?

Well, the result of this program is not reproducible. You'll find Helmut's about this in this forum. however, some articles were published using this program and one of the authors is from Spanish agency, so I guess that you can try it. :-D

All the best,

Complete thread:

 Admin contact
21,174 posts in 4,412 threads, 1,477 registered users;
online 18 (0 registered, 18 guests [including 6 identified bots]).
Forum time: Thursday 03:16 CEST (Europe/Vienna)

It is not so much that I have confidence in scientists being right,
but that I have so much in nonscientists being wrong.    Isaac Asimov

The Bioequivalence and Bioavailability Forum is hosted by
BEBAC Ing. Helmut Schütz