AV.
Nothing that you have asserted, demonstrates that a 2015 IA for efficacy actually took place.
It might have, but you don't know that, and nor do I.
All we know from references in a couple of places, is they planned for one.
No more, no less.
You keep making this huge leap, that because they planned for one, they must have had one, and not only are you assuming they had one, you are also assuming that it happened before the implementation of the hold, and that one led to the other.
Although the 'numbers of events' trigger might have changed with trial resizing, the planned efficacy IA was probably still scheduled to happen based on an event number threshold.
And as such, if events were slower or faster than expected, the efficacy IA would be later or earlier than anticipated.
I note that that you now describe your futility theory as a 'theory', rather than asserting strongly that that is what happened, as you did previously.
So it's gratifying that you are now acknowledging less certainty.
I wouldn't accuse you of 2 + 2 = 5.
Imo, it's more a case of ? + ? = f (for futility).
You are coming up with an 'answer' or 'solution', if you like, without knowing the variables involved in the calculation.
As I've indicated elsewhere, the possible causes of the partial hold are manyfold.
As time goes by and the more I think about this, I am less inclined to think that the hold had anything to do with a specific recommendation from an efficacy IA.
With a possible Protocol violation coming more into the frame.
And a technical, dosing, or manufacturing, or shelflife, or documentation issue, could cause a protocol violation.
Mostly opinion.
News 
Market Data 
Discover 
