Dendreon Corp fka DNDNQ

[travfourr]

Q AND A W/ ir:
1. Does the CD54 data we have seen indicate/ suggest that IMPACT is doing slightly better than the integrated data? Until the data are unblinded, we would not draw any conclusions on the current study relative to the integrated data from the previous Phase 3 studies.
2. Recently the company has been repeating the standard line that if the interim misses its alpha, then the DMC will likely just recommend moving on without relevaling anything. Can you tell me if management and the FDA have discussed and precified computations to check for it? Sections 6.5 and 6.6 of this FDA Guidance (http://www.fda.gov/cber/gdlns/clindatmon.htm#7) talk about the procedural aspect of that. The company and the FDA have had substantive discussions regarding the design of the study. This includes all aspects of design including the sections you mentioned. The current study does not have a futility analysis, and as we have indicated, if we hit the pre-specified criteria for approval, we would expect to unblind the data and would know the statistical result for the study. If not, we would expect to continue to the final analysis next year.
3. If the Interim exceeds it's alpha, but is <= the final alpha, could the IDMC recommend the trial be stopped for high probability that the final look will be successful? Our understanding from the FDA is that they would accept either a successful interim or final analysis for survival as sufficient to amend our BLA submission. If we are not successful at the interim, we would expect that we would need to wait until the final results even if we were in a position that we had a high probability of success for the final.
4.Would/ Could you please comment on this:
If the company knows the CD54 upregulation on batches released and where they sent them and they've monitored the trial sites and they say that the results of 02b are right in line or close to the combined results it seems to me it would be pretty easy to shake out the data into two reasonably accurate KM curves. I know this leaves out all the biostatistical mumbo jumbo but how difficult can it be to plot two curves especially when the trial is based on two prior phase threes and we know that CD54 correlates strongly with survival. Undoubtedly they've been doing this (exquisitely refined with an array of biostatistical tools that I have no clue about) and their corporate body language would reflect what they've seen. Based upon this and what we know about the science and the prior studies (all the prior studies) I'd say that the interim will be successful. All of the other criteria will refine these basic curves but given that the trial size is relatively large a number of incorrect or lost data points shouldn't make a substantial difference. We do not have access to unblinded data. Any data that you may have on a blinded basis is just that--blinded data. You might get concerned if you did not see blinded data behaving consistent with prior studies. You cannot be certain of anything because events are occurring as expected. As such, we do not think that you can speculate on the probability of the interim results other than what we have stated previously which is that: The interim is designed based upon two previous Phase 3 controlled studies. The powering of the interim is such that we believe it has a reasonable possibility of success. By this we mean that the interim includes more events than the 164 included in the integrated results of D-9901 and D-9902A. If we saw results consistent with the integrated results, then we would expect to achieve statistical significance and amend our BLA. The final is powered at 88% to achieve its endpoint at 304 events, which we expect to reach in 2009.