Iwfal on this board and Clarksterh on the I-Village board have posted that a Cox regression analysis performed on a trial whose arms are either well balanced or slightly favoring the placebo arm improves the p value by much, much more than one would expect. Basically, the phrase used has been "decreasing the noise." From what I've read over the last two years, including today's slides, there was a very slight bias in favor of 9901's placebo arm. Dr. Small characteristically does not pump up the Provenge metrics, so he didn't describe the Halabi model as slightly favoring the placebo arm. In another example, even though a 33% higher percentage of 9901's placebo arm patients took post-Provenge Taxotere than the patients in 9901's Provenge arm, Dr. Small has for the past year described the chemo use in each arm as being well balanced, deliberately de-emphasizing the actual percentages, which favored Provenge.
As for the placebo crossover vs pure placebo median survival breakdown, we already know the 24-month results for 9901: 26.3 months crossover vs 19.3 months pure placebo. I'd really like to see the final 36-month results of this metric, as I believe continued positive numbers would be one more brick in the wall to convince the advisory panel.
