There is a lag between performance of a study and preparation of manuscript and then another lag between submission to a journal and publication. Therefore, most phase 3 studies end up getting published about 1-2 years after they are completed. Therefore, some phase 3 will be published after FDA approval.
Fact remains: Aricept was approved based upon Rogers et al, 1996 study I cited of 436 PARTICIPANTS IN THE US. THIS IS NOT SPECULATION NOR MARKETING SOURCE (package label...)
The 1996 Rogers et al study that you refer to was a phase 2 of 161 patients (PubMed 8915035). It was not a > 400 patient study.
Per the Centerwatch link in your post "Controlled clinical trials in over 900 subjects demonstrated...." If only the 1996 paper was looked at by the FDA, then only 161 patients were evaluated. Even if you think there were > 400 patients in that study, I believe 400 is far less than 900.
I relooked at the pertinent publications. You are correct that Burns et al. was not part of the submission (It was used to get the label expanded to include severe while the original label was mild and moderate). I made a mistake there. Rogers seems to be lead author for many of the studies including the 1996 phase 2. She published two distinct Phase 3 studies in January 1998 (submitted March 2017) and in May 1998. Both studies state that they were phase 3, the second one stating "This phase 3 study was 1 of 2 pivotal trials undertaken to establish the efficacy and safety of using donepezil in patients with mild to moderately severe Alzheimer disease". One study had 468 patients and the other had 473 (468 + 473 = 941 which is greater than 900 and agreeing with the Centerwatch statement).
I stand behind my original post rebutting the fictitious claims of “boogey-trials” needed for approval
I stand behind my original assertion. The FDA did look at two phase 3 studies. There were a total of over 900 patients not just one study with >400
Not that it matters for this post, but I agree with the majority here that Aricept is not a very good AD medication. In a couple years we will see if A273 is better.