I am curious how you reconcile the belief in the efficacy of Provenge, combined with the statement by Dr.Gold that the interim analysis is "very well powered" - not merely "well powered" -- with concluding that the chance of a successful interim look is much less than 1/3rd.
It depends on:
a) How much you trust Gold - e.g. if 9902b patient pop really is no sicker than 9901 as measured by Halabi nomogram then the odds go up - but Gold used the weasel word 'substantially'. My trust is limited for all biotech management and especially for Dendreon where the Stat Review made it clear that they very significantly overstated their whole CR process. Trust has to be earned in biotech and the only CEO I trust is CYPB's Kranzler (I am on record as trusting him before the results this week) and possibly ZGEN's Carter.
b) Whether your assumptions are the same as the one Dendreon may be making - e.g. are they assuming a constant HR through time, whereas I assume that curves have a very low HR before about 24 months and so the interim will be disadvantaged by all the late enrollees.
Finally note that companies notoriously overstate their powering by overestimating what HR they should expect in the trial population.
News 
Market Data 
Discover 
