There is no hard and fast rule that you need two trials.
As for the sufficiency of endpoints, you had both the primary ADAS-Cog and secondary CDR-SB each come in here with a p value below .025.
Of course, now we are also anticipating that the FDA, like the EMA before it (per the most recent Anavex slide show), will not require the ADL endpoint in this type of early stage AD trial. The success of a trial is measured by what the regulatory agency determines to be sufficient at the time of decision, and not whether an overly ambitious primary endpoint set years before was or was not met.
You can still argue whether Anavex will be able to satisfactorily demonstrate safety, or that they may have some other defect in the reported result that's currently hidden. But you are overstating your case about the risk of disapproval with the EMA.