>> If a sponsor expects the FDA to allocate an implicit alpha of 0.05 for survival, then it would be dumb to allocate any *explicit* alpha for survival. Might as well use all of the 0.05 of explicit alpha on non-survival endpoints and get the implicit 0.05 on survival in addition.
The fact that trial designers do allocate explicit alpha for survival as a secondary endpoint suggests that not all sponsors are willing to take it for granted that the FDA will give them 0.05 of *implicit* alpha. <<
The FDA may have been magnanimous in the Coreg case by giving the sponsor “implicit alpha” for survival, but most sponsors evidently think this was aberration else they would not be dumb enough to design in *explicit* alpha for survival as many do.
By the way, this issue (the lack of explicit alpha for survival in 9901 and 9902a) is not the reason I think Provenge will fail to be approved based on 9901 and 9902a. Regards, Dew <<