Replies to post #37459 on Biotech Values

[AlpineBV_Miller]

Cox analysis that lowered the p-value to 0.02 was not fully pre-specified.

Can you expand on what you mean by "not fully" versus "fully"?

I comprehend all your other arguments on the Cox analysis of 9902a except what you believe Dendreon should have done to make this analysis "fully" pre-specified.

[gofishmarko]

Re: DNDN skepticism

I tend to agree with your analysis of DNDN's actions and timing of same. I also think the 'key question' is the one you outline here :

>> Thus, the key question is whether 9902a is sufficiently supportive of 9901 to warrant approval based on one successful study and one supportive study. My contention is that 9902a is not supportive because the raw p-value for overall survival was 0.33 and the Cox analysis that lowered the p-value to 0.02 was not fully pre-specified. <<

What I wonder about is the extent to which the FDA will deviate from stringent requirements , e.g. , that a particular type of analysis be prespecified , in indications where survival is measured in months with the best available treatments. I would hope that they maintain some flexibility in these cases particularly , though not so much as to allow bad science to prevail. My own guess, like yours ,is that they'll get the approvable letter pending 9902b , but I think DNDN is still a reasonable buy at these levels if you have a medium term investment horizon , and you could get an early surprise.

Here's a question that bugs me. Why the crossover trials ? Was this all DNDN's idea , or the FDA , or both ? Was it done for compassionate reasons , or to encourage enrollment ?

If they had run clean trials , i.e , Provenge plus SOC vs. placebo plus SOC , the analysis would have been straigtforward and the results clear by now.

[lreichel]

DNDN

I'd like to ask a few follow up questions. Thanks in advance for your response.

1. Why isn't 9901 sufficient for approval? As for 9902a analysis being modified, isn't that OK if you learned something from the results? If the company used sound statistical analysis, I would think it is acceptable.

2. The delay in filing the BLA could be caused by many factors, not the least is that they were searching for the MOA. I thought management filed after discussing further with the FDA. Could the delay have been self imposed and overturned from discussions with the FDA?

3. Manufacturing capacity reduction makes sense if you are cash strapped and can expand quickly, both of which seem to be the case for the company.

4. As a CFO, you generally leave because you are shown the door or there is something wrong with the accounting policies/financial statements. I haven't seen anything on the accounting side that seems to be a problem, but we wouldn't know unless there is an internal source. Also, there is no reason to play accounting games with a pre revenue company whose fortunes rest on their product, not their results.

You may be correct, but there is just as much anecdotal support for a positive outcome as you have laid out for a negative response from the FDA. How heavily do you weigh the fact that there isn't any real downside (except cost) and there is a potential upside for people with little or no alternative?

I'd like to see more facts on the pro and con side to get a better understanding on the prospects for Provenge. Arguments that rely on supposition as to motive and rationale are not as persuasive. Like many small early stage biotech's we have fewer facts and rely on intuition to form an educated guess.

[chessterr2010]

DewDiligence, thanks! Appreciate all your comments!

[bubba_smh]

DEWDILIGENCE >The slowness of the BLA submission—16 months from the time that the final 9902a were reported until the last BLA module was filed—suggests to me that DNDN was in no hurry to get the BLA done because they themselves do not expect approval on the first review cycle. <

So are you suggesting that DNDN filed this BLA for the veneer of it and for eye candy for the shareholders even though in their words the FDA encouraged them to do so?? I have a hard time believing they would waste millions of dollars and significant time and effort in something they didn't believe they had a decent shot at. Additionally, weren't you the one on the yahoo message boards that said DNDN would never submit a BLA based upon these two studies - that basically it was a bluff? LOL

[Gametheory101]

Barron's has a different opinion of Dendreon.

http://online.barrons.com/public/main