kyoto27

FDA Turns Over Dendreon Documents - Here They Are

Q to User 161198:

Provenge had a p-value of .052.
Are you really suggesting that a p-value of .052 is showing the treatment ‘failed to slow progression?’ To the contrary, a p-value of .052 clearly showed that the treatment slowed progression of symptoms of the disease and, in fact, most biostatisticians would call a p=0.052 level of significance “nominally statistically significant.”

Why the FDA did not think so in this case is unknown. Especially when you add the stellar survival log-rank p = .01 for D9901 (which was on the entire trial, not some subgroup) the FDA decision becomes what in fact it really was –unprecedented in this terminal disease setting.

And let’s not forget that the FDA, by virtue of its own revised trial design guidance, has demoted the importance of traditional markers such as TTP, subordinating them to overall survival. Nonetheless, Provenge naysayers continue to trot out TTP like a champion mare when it suits them, knowing that it is an unacceptable endpoint for making an approval decision.

I’ll let the FDA have the last word:

“When a study fails to meet its primary endpoints, there's no alpha left for other endpoints analysis. So literally, means from pure statistical point of view, the difference in other endpoints should not be considered statistically significant. Therefore, it is very difficult to interpret the hypothesis test result for overall survival in Study 1.

So if overall survival is one of the many unspecified endpoints, under testing it is very possible that a p-value of 0.01 was observed just by chance. However, survival is not one of the many, many endpoints that can be randomly selected for testing. Survival is a preferred endpoint for cancer trial. As Dendreon and Dr. Liu just mentioned, this endpoint is reliable, clinically meaningful. This is why we are here seeking advice from the (Provenge) advisory committee meeting.”

Dr. Bo-Guang Zhen, Ph.D (Division of Biostatistics, FDA )