Maxim Jacobs

On FDA Response Letters and Dendreon's Provenge

"The company announced the receipt of a 'Complete Response Letter' which is the new lingo for 'Approvable Letter.'"

Just a minor correction, but Complete Response Letters are not new lingo. They have always been how CBER has communicated its needs for approval. Approvable and unapprovable letters are how CDER makes its decisions. So a Complete Response Letter is actually both, CBER does not make a distinction.

Update on Dendreon's Briefing Docs: Provenge Missed All Endpoints

Dan,

"It took you 24 hours to post this MISLEADING "analysis" report "

I think Seeking Alpha has simply become so popular that the editors are swamped with posts to look at. 2 months ago when I submitted a post on AGIX it took about 2-3 hours to post. The last couple have taken much longer. I submitted this before noon yesterday, just about 3 hours after the briefing docs were released.

"This is a LIE. Please check #2 in www.fda.gov/ohrms/dock... ... quote:
"Page 3, safety results, 1st paragraph, replace the phrase “compared to none in APC-Placebo treated subjects” with: “compared to 2.6% (2 out of 78) in APC-Placebo treated subjects.”"


We are referring to two different statistics. I am referring to deaths due to CVA while you, and the correction, refer to CVA events (that did and didn't result in death). Check out www.fda.gov/ohrms/dock... Table 25, and you will see the statistic that I am quoting. Given the number of CVA deaths in such a small trial, I wouldn't call it a non-issue. It is at least one that needs to be explored further. Or would you rather give someone a drug of questionable efficacy that could kill them of stroke?

I think the opinion of the clinical reviewer trumps anything from the statistician (and by the way, he wasnt exactly pro-Provenge given all the discrepancies he found).

"Maxim, one more interesting point that is worth a lengthy discussion is about the recent ILLEGAL "printing/issuing" of phantom shares (aka "naked shorts") by the shorts. Your comment about this ILLEGAL activity will be very interesting ..."

I'm not quite sure what you are referring to. I remember some stock a few years ago were listed on the Berlin Bremen exchange without their knowledge and you could naked short them there but I dont see how this naked shorting can be done on the US market, especially not en-masse (sometimes brokers print your tickets but end up taking forever to deliver your shares, but those incidents are always isolated). Everyone I know is complaining that they have no borrow even though this recent pop in DNDN is creating a wonderful shorting opportunity.

The Short Case on Dendreon Corporation

Dan:

"How can one compare AGIX and DNDN??? Was AGIX rejected by the FDA after showing survival benefit? AGIX had a CLINICAL RISK (no phase-3 data was PUBLICLY out) -- so it was anybody's guess what would be the outcome. Right? WRONG! Joe Edelman has admitted in an interview that analysts call doctors that participate in clinical trials for "inside info":"

One can compare AGIX and DNDN easily. In both cases long positions were based on flawed study data and the argument "if it works, it will be huge." And I don't understand the link between Joe's quote and the discussion we are having on DNDN. I dont know anyone who decided on their AGIX position because of "inside information" from docs. In such a large trial you can't trust a few docs anyway as they might have treated only placebo patients. AGIX, and DNDN, both had armies of people shorting them because of the publicly available information which showed pretty convincingly that neither drug is as effective as the companies want you to believe.

"Maxim, the FDA has "reserves alpha" for SURVIVAL regardless if it is an end-point. Please check the FDA's decision re carvedilol". The meaning is that if a company shows the GOLD STANDARD - the company gets this extra alpha ON TOP of the primary endpoint(s)"

I dont think a 120 person trial proves there is a survival benefit. Remember in the earlier TELK trials, the company was touting a survival benefit for Telcyta which evaported once in properly run trials which were actually targeting survival. Survival is only a gold standard when proven in a survival trial. At best this survival benefit is a signal at best and an anomaly at worst. I've seen that countless times in Biotech.

"HUH??? MORE important statistic ... in comparison to WHAT? Would one prefer to have a longer "time to tumor progression" or longer "SURVIVAL" with MINIMAL SIDE EFFECTS?"

The primary endpoint of a trial is always more important than a secondary from the FDA's point of view, especially when you have things like crossovers confounding the secondary data.

"Statistics 1.01, Max? Why do you think that BP are running huge trials? Isn't it in order to power the trial and demonstrate a useful benefit? "MANUFACTURING a benefit (ie, p=0.01 for survival, for example) in small trials is MUCH HARDER than with large trials, isn't it?"

It is not just about a certain statistic, but about convincing the FDA that the drug works. In a small trial there is a larger chance that you will have a false positive.

"WRONG ... the initial SPA was for a 250-patients 9902B study. it was later expanded to 500. But who really cares about the fine details?"

I guess you got into a groove of trying to correct me but you correct a minor detail without answering the meat of the point which was, they were initially going to run a trial to back up the 9901 data, but couldnt enroll and then decided to go through with an FDA approval. Not exactly a great reason to apply for approval, because docs don't want to enroll patients in your trial. Also, the fact that they couldnt enroll a 250 patient trial makes it even worse than if they couldnt enroll a 500 patient trial.

"I bet that some shorts spent hours speaking to doctors that will participate in the AC. I hope (and believe) that regardless of your (indirect?) offers to them, those top-notch MDs understand that there is NO REASON to wait 3 years to see the results of a powered study with 95%-chances-to-succeed... (this has been analyzed before - given that it measures SURVIVAL using Cox-regression -- p=0.0006 in the combined 9901+02A for Survival/Cox is a nice starting point) AND considering the mild side effects of Provenge AND considering that over 50% of the pts prefer NOT to take the only available drug today for AIPC."

I dont think anyone has ever tried to bribe a doc on a FDA panel. Sounds like a great way to land in jail. I'd be wary of any of these creative statistical analyses as there are numerous ways to manipulate them. Especially post-hoc.

"My last point, Maxim, assuming you had no financial interest, what would YOU vote at the AC based on the data that we know TODAY? "

I would still want to do a good job, whether or not I had a financial interest. If I believed in the FDA mandate to approve drugs that have the data proving efficacy, there is no way I would approve this drug. I would want to see the additional data first. If the FDA starts just approving every drug out there for cancer (which of course it will never do) it would probably be much harder for patients to find the right drug for them as they will be busy taking placebos.

The Short Case on Dendreon Corporation

You bring up some good points Walldiver. Let me try to answer them or at least comment:

- First, it's not an inaccuracy if I don't mention the p-value. They missed plain and simple. The fact that it was close matters to bulls, not to the FDA. Also, even if they hit and it was 0.04 I would question it. Think about it, with a p-value of 0.04 there is still a significant chance that this data is a fake. Coupled with the fact that 9901 was never meant to be a registration trial and it was very small and I just don't see how anyone at the FDA can have confidence that Provenge is definitely a real drug. They will simply make them run another trial, as one is ongoing already. And I wouldnt expect patient groups to be up in arms about Provenge if it is delayed. And as you mention the 9902A trial, I've never seen the TTP data for that one, was it ever disclosed? I'm thinking it will be disclosed tomorrow in the briefing docs and the DNDN bulls may be in for a shock.

-The Taxotere statistical analysis is very questionable. First, it was not an analysis that was planned on being done before the trial and therefore is akin to data mining. Second, the problem with subsets is that the baselines may be even less balanced than in the trial as a whole. We are talking about only around 80 patients among the two trials after all, so there can be a lot of statistical hand waving going on.

-What I would have really liked to see is the TTP and survival for every Gleason Score integer in the trial. Theyre bundling of so many Gleasons together I think makes the data confusing.