Edcpf - it is all in the statistical analysis, all of which is decided and agreed upon before the start of the trial. And people wonder why it takes so long to get a phase 3 up and running! It will be understood that after X number of events, those will be unblinded and run through some statistical and safety analysis. I think of it like the KM graph with areas shaded based on the different recommendations. Once the interim data is plotted, where it falls on the graph determines the recommendation. The plot and areas for recommendations change for each interim look because there are more events and probabilities are different when projecting to the likely end result of the study.
As the guidance discusses, it is important to have all this decided before the start of the study and agreed to. The FDA is looking for more concrete results, the sponsor looking to get the drug approved as quickly as possible. Garnick, having gone throught this many times (DMCs are used in almost all oncology phase 3s) knows the ropes and I'm confident the number of events for the first look-in are towards the low end because of the results of the phase 2.
Only a matter of time.
JMHO of course
JBAIN
News 
Market Data 
Discover 
