>> Of course that would ignore dozens of trials where GMCSF did nothing for cancer patients. <<
I'm not saying the DNDN procedure is not working , just that I haven't seen much that convinces me that it's acting in a prostate-specific way as opposed to as a general immune stimulant. If DNDN or others have shown that GM-CFS fused to a non-tumor-specific protein , used in the same ex-vivo stimulation process , doesn't work as well as PAP-GM-CSF against prostate cancer , fine. Cite those studies.
I may not have made myself clear in my point about frozen Provenge. With Provenge , we're talking about a treatment where the MOA is immunological , but beyond that it's poorly characterized. Now you're going to freeze/store/rethaw/infuse a cellular infusion product and assume that no unknown ( read : confounding ) immunogenic species will be generated in that process that will cause it to be different from fresh Provenge. Not a problem if the surrogate marker is the final endpoint , but IMO a bad idea if post crossover endpoints like survival are critical.