And what you are leaving out is that the total enrollment for both of DNDN's 2 trials was less than 348. --notbrad
Sorry, what?! The first P3 randomized 127 patients and the second (IMPACT) randomized 512. That's 639 patients buddy.
There was also a large P2 that randomized 98 patients. However those data didn't look so good:
14.3 Study D9902A Study D9902A was a random ized, double blind, placebo- controlled, multicenter trial in patients with asymptomatic metastatic castrate resistan t prostate cancer. A total of 98 patients were randomized 2:1 to receive PROVENGE ® (active treatment, n=65) or placebo (control, n=33). Study D9902A was identical in design to Study D9901. Baseline demographics and disease characteris tics were generally well balanced between the treatment groups. Several prognosti c factors, including the percen tage of patients with more than 10 bone metastases, and median values fo r PSA, LDH and alkaline phosphatase favored the placebo arm. Of the 33 patients randomized to placebo, 22 patients (66.7%) crossed over to receive antigen loaded (PAP-GM-CSF) APCs pr epared from autologous APC precursors that had been cryopreserved at the time of placebo ge neration as part of the salvage study. Following disease progression, 38.6% of patients in the PROVENGE ® arm and 34.4% of patients in the placebo arm received subsequent treatment with docetaxel. There was no significant delay in time to disease progression (HR = 0.921 [95% CI: 0.588, 1.443]; P = 0.719, log rank) and no significant difference in overall survival for patients treated with PROVENGE ® compared to placebo (HR = 0.786 [95% CI: 0.484, 1.278]; P = 0.331, log rank).
They also conducted lots of other studies with over 100 patients before they went belly up like these: