p3analyze........................
very nice work, you have the OBF stuff down too.
my only comment:
"2. However, if using the pooled hazard ratio of 1.5 from 9901 and 9902a, to power the study at 90% would require only 256 events, which means the final analysis can be done by end of Dec 2007, and data release in 2008.
3. A conservative powering approach would be to use HR observed in 9902a, ie 1.3, but that would require 456 deaths to have 80% power, therefor this senario is not likely."
I would guess then that the HR is not 1.3 , but closer to 1.5 - the target of 75% of events is the extreme end of the spectrum, and usually just past the median is reasonable for a survival trial.
I have seen them too set up this way: expect median survival on the control arm to be X months; at that point in time, what will be the relative event level in the drug arm ? If then S(t) = 0.4 (say) then you need only 55% of events.
So I expect the target HR is closer to 1.5, and you should figure on around 275 maybe 300 events at most.
(I know others have said 75% of events, and if they have convincing arguments to make for that, that would be great, and perhaps on this board we will arrive at where it should be !)