Thank you for your explanation. Still I didn't understand how it was possible for CEO to pick measures that suited him/trial. He is not a stupid guy, he is not a criminal. His decision was based on something. Maybe he demonstrated results for people with some special DNA (I mean that the patient's DNA specific properties say that this medicine can give positive results of treatment for this patient)!? Maybe Anavex is ahead of the all planet in such an approach to find the patient cohort who will be successfully treated? That is why we all need a reasonable analysis from a qualified scientist.
You do know that any endpoint has to have the statistical analysis described in the SAP prior to the data lock.
And even if pick and choose was the case, which it is not, the agencies (FDA etc.) will look at the statistical analysis and decide whether or not it is appropriate. The FDA statisticians will look at the data and do their own analysis if the do not think the analysis presented is appropriate. The FDA will decide to approve or not based on their internal statistical analysis if they think the companies analysis is not right.