I may be in the minority on this,
but I think the additional statistics laid out in last night's presentation bode well for the 1st interim results. I am not going to pretend that I am a statistician, but I look at it this way. The primary endpoint is, by definition, the metric that is used to end the trial (early or late). Interim analysis that includes an efficacy analysis allows a trial to be ended before the total number of agreed upon events. The primary endpoint hurdle is 6 month extension of progression free survival. Historically, the Stupps protocol has at best improved PFS to 9 months on average. This means that DCVAX-L must at least show 15 months survival. We have now been told that, for over the last 10 years, in over 200 patients, PFS is "consistently" extended by 18 months. 18 months + (the lowest pfs from the stupps protocol 6.9 months) = 24.9 months.
Again, I am no statistician, but if we are seeing the same results in the phase 3 trial as we have in the other 200 patients (albeit some are not GBM), I have to believe the trial will be unblinded on the 1st interim analysis.
I know their criteria, but the statistical algorithms make my head spin. I can only use other prior drugs that were unblinded early as my guide, and if DCVAX-L has the above results, it would blow the other drugs out of the water.
The bottom line is DCVAX-L has to show 15 months PFS, if it continues to perform the way it has over 10 years and 200 patients, it will instead show 24 months PFS. If that is a given, the only question remaining is whether statisticians can extrapolate enough data with these astounding results to demonstrate the 1st interim results are not the product of chance (beyond a certain probability).
