<This is incorrect. The cases where the measured HR is worse than the assumed HR, but the trial is still stat sig, are part of the power percentage.>
ooops Yes, absolutely my bad .... rusty here.
One point I would make though is that the underlying assumption (with the formulae for size N etc) is somewhat of an estimate.
<Biotechs as a group clearly overestimate the TRUE efficacy...>
I think it's readily a trade-of with Costs. After all, they have the same statistical consultants who have the same trials, you can bet they are pleading for larger trials. Especially when they have a drug which works to some extent, so why have to dilute hugely ....
But that's what's so rosy about Proellex - it works by a mile and I dont think trial size is constrained by cash - actually the one time I spoke to JoeP over the phone (apart from the recent NYC show) I had suggested he double the numbers in UF and he said that although they could easily do that, there was no need.
I have to admit though - I do wish the Anemia trial was larger.
Market Data 
Markets 
AI
