My understanding is the same - neither the CEGE Vital-1 nor Vital-2 trials are non-inferiority trials. In fact, since Vital-1 does not control against the GVAX arm taking Taxotere after GVAX, in many ways it is quite similar to the Provenge 9902b trial where the combination of immunotherapy plus Taxotere is so disproprtionately superior to immunotherapy alone that even if their ITT patients using the combination are in any significant minority, their results, especially at the median survival point, will likely drive the overall results for the entire ITT group.
Results in Phase 2 trials are often poor predictors of Phase 3 results, as Provenge's Phase 2 TTP results were for its Phase 3 TTP results, especially wrt to 9902a. So the Phase 2 35 month median survival results reported for the 32 GVAX patients receiving the highest dose that is now being used exclusively in Vital-1 and Vital-2 Phase 3 studies, where some of the Ph2 patients also received Taxotere after GVAX, will not necessarily be duplicated in Phase 3. The GVAX median survival results would also be less likely to be statistically significant at the interim look than at a final unblinding. However, that GVAX 35 median survival is virtually identical to that of the 51 of 147 Provenge 9901 and 9902a patients ITT patients who took Taxotere after Provenge as reported by Dr. Petrylak last November. Since Drs. Petrylak, Small and Higano are now all GVAX investigators, I would imagine they are encouraging the ITT patients to follow GVAX with Taxotere, the present standard of care. OTOH, GVAX is not an FDA approved therapy and, unlike the 9902b Provenge trial, there is no crossover from Taxotere to the experimental immunotherapy.
The median survival for the asymptomatic subgroup in the Taxotere TAX 327 pivotal trial was apparently 22 to 23 months. Assuming the same median survival for the Taxotere arm in Vital 1 holds true, I find it difficult to understand why you believe so strongly that GVAX (+Taxotere)in Vital-1 with a possible 12 month greater increase in median survival than Taxotere alone would fail to achieve a statistically significant improvement over Taxotere in the same way that the Petrylak analysis would suggest that Provenge (+Taxotere) would prove superior to Taxotere. I'm not saying that you won't be right and anything, of course, is possible, but from all that I've read, GVAX looks to have as decent a chance for success as ultimately Provenge does.
