Dendreon: Don't Forget About Risk vs. Reward

In the last few weeks, the price of Dendreon (DNDN) has increased from ~$4 to ~$23. On March 29, the FDA Office of Cellular, Tissue and Gene Therapies Advisory Committee recommended to the FDA that there is substantial evidence of efficacy and safety of Provenge for the treatment of patients with asymptomatic, metastatic, androgen-independent prostate cancer. The FDA Advisory Committee voted 17-0 in favor of the safety of Provenge and 13-4 in favor that substantial evidence exists on efficacy. The FDA will review the advisory committee's recommendation and anticipates a decision on Provenge by May 15.

Since the release of the FDA Advisory Committee's recommendation, I have read numerous articles DNDN's potential value. One article suggested DNDN would be taken out at $30 after the FDA decision is announced. Another article states "a price per share of $70 within several months post approval is not out of the question." Many believe approval of Provenge on May 15 is a foregone conclusion and the rise in the stock price would suggest so.

While I understand the widespread excitement generated by the advisory committee's vote, investors seemed to have forgotten risk vs reward. I would suggest taking a step back and analyzing the advisory committee's vote in more detail.

The 17-0 vote in favor of the safety of Provenge was no surprise. Safety has never been the real issue. The questions surrounding Provenge has centered around efficacy. While a 13-4 vote in favor of efficacy does sound promising on the surface, it is important to address the efficacy issue in more detail. According to the FDA briefing documents before the vote, the original question regarding efficacy was stated as such:

Does the submitted data establish the efficacy of sipuleucel-T in the intended population?

For some reason, the FDA reworded the original question. The new wording on the question read as follows:

Is there substantial evidence to support that Provenge works in the intended population?

It is clear that the vote benefited from the revised language in the question presented to the panel. Originally, DNDN conducted 2 clinical trials designed to show that Provenge slowed the spread of cancer in men whose prostate cancer had spread to other parts of the body. Both failed to meet the primary endpoint of the study. The first study of 127 patients showed the drug extended overall survival by 4.5 months which is statistically significant. Apart from the fact that the trial was not specifically designed to show extended survival as a specified endpoint, the small patient size of the study would suggest that more information would be needed to approve the drug. Furthermore, it should be noted that in the second trial, while overall survival was prolonged, the result was not statistically significant.

While I acknowledge the FDA typically follows the panel's recommendation, I would find it surprising if the FDA did so in this case. In the past, the FDA has rejected drugs that did not meet its primary endpoints. Based on two small trials which failed to meet its primary endpoint, and a post analysis result that was not replicated in the following trial, it would be 'out of character' for the FDA to approve Provenge on May 15.

With that said, the prolonged overall survival in the first trial and the safety of Provenge cannot be ignored. DNDN is currently under going a larger trial of 500 patients to show Provenge extends life. Interim data is expected in 2008 with completion of the trial expected in 2010. It would be more in line with the FDA' s recent history to delay a final decision and issue an approvable letter on May 15 requiring additional data before making such a definitive ruling. In fact, members of the advisory panel suggested that the definitive proof would come from an ongoing trial. This seems to be the likely approach the FDA will take.

If, in fact, the FDA issues an approvable letter, the stock will most likely sell off substantially from these levels. Some examples of companies selling off after an approvable letter are POZN, ISPH and ENCY. Not only did the companies' stock prices experience a drop, the enthusiasm of long term investors was muted by long ongoing processes to a final ruling. If DNDN is issued an approvable letter, this would be a substantial step forward in a long process for a definitive ruling. Apart from the risk of competing products being developed over the next few years, investors will have to wait till 2008 for interim data and 2010 for the completion of the study.

Needless to say, there is tremendous risk in how this study may turn out. I question whether the risk/reward is worth it at the current stock price of $23-$24 having increased ~500% over the last 2 weeks for something that potentially may or may not come for another 3 years. While I'm cautiously optimistic on the approval process going forward, the stock price has gotten way ahead of itself given the likely scenario of the issuance of an approvable letter.

DNDN 1-yr chart

Disclosure: none

Comments

[Echo To All:]

Risks to reward? I have written a piece painting the realistic risk, and it is minimal. You say you are optimistic, yet you make comments like this:

"While I acknowledge the FDA typically follows the panel's recommendation, I would find it surprising if the FDA did so in this case."

You are kidding, right?

You also mention... "For some reason, the FDA reworded the original question." That should tell you something about the FDA, and CBER. If it doesn't, then I don't know how to spell it out for you any clearer.

The stock's conservative value is around 30, w/ a ROW it is around 40. (And that is on Provenge alone, excluding the potential of the underlining Cassette Technology applied to other cancers. Which is already proven via Neuvenge for certain breast cancers.) The stock has been artificially held down for 2-3 YEARS, and you call its current movement crazy? I call its capping crazy. DNDN should never have been at 3-5 before the AC meeting. The downward pressure created this craziness, and now DNDN is merely going to where it should be.

(PS... because of all the shorting interest, and lack of sellers, it will go higher than 40 upon approval. Any one still short should keep Suppy and Demand in mind. The real risk is on the short side.)

2007 Apr 10 08:35 AM

[Wall Diver:]

Instead of wondering why the FDA reworded the original question, Lets Be Reasonable should ask why the question read "Does the submitted data ESTABLISH efficacy" in the first place. The FDA's own guidelines for drug approvals are phrased along the lines of "Does the product demonstrate substantial evidence of efficacy?" As the first three panel members all interpreted "establish" to mean "100% conclusiveness," it was understandable that the committee chairman Dr. James Mule asked the FDA officials present at the meeting if the question could be reworded. When it was reworded to conform with the FDA's own guidelines for drug approvals, Provenge passed the efficacy question with ease.

I do agree with Let's Be Reasonable that one ethical question remains, and that is the question of what to do about the remaining ~33 placebo arm patients yet to enroll in the ongoing Phase III trial, 9902B. If the FDA approves Provenge by May 15, what's to prevent these patients from remaining in the placebo arm of the trial?

Does the FDA deny the approval of Provenge in order to get a more accurate reading of 9902B? Or should the FDA withold this safe and more effective treatment from ~100,000 men over the next few years until the 9902B survival data matures? I firmly believe that the agency and the company can hash out a workable solution, such as establishing a cutoff date for US enrollment, filling the remaining spots in the trial with patients from countries where Provenge won't be approved for several years (Canada, NW Europe, Australia), and approving the treatment for marketing. There is clearly an overwhelming demand in the prostate cancer community for an effective treatment like Provenge that does not have the severe side effect profile characterized by Taxotere chemotherapy.

2007 Apr 10 09:03 AM

[Wall Diver:]

I should have added in my comment above that the placebo patients in 9902B (as well as the first two Phase III trials) have the option to cross over and receive salvage Provenge after their cancer progresses. This is a requirement of the FDA for placebo-controlled trials, and Dendreon's are no different from Onyx's trials or those of other oncology biotech companies.

2007 Apr 10 09:07 AM

[Keith Rettig:]

It should be clearly noted that the FDA did NOT change the question of efficacy. They merely defined what they meant. The first three panel members interpreted the question to mean "determined, proven, without a doubt, etc.". The first three all said they wanted Provenge in the hands of the patients but that they couldn't agreed that the efficacy of the drug was established without doubt from the current research.
It was then that the panel chair asked the FDA for clarification. The FDA clarified that they wanted to know "if the data provided substantial evidence". This is not changing the question; this is defining what they meant.
I was quite impressed with how this played out. It clearly shows that the CBER panel is thinking of the patients as well as the quality of the product; and thus performing exactly as the FDA wants them.
I strongly agree with 'echo toall' regarding that this would seem to indicate the FDA's intentions for approval.

2007 Apr 10 10:02 AM

[CrossProfit_:]

Agree.

CrossProfit

2007 Apr 10 02:03 PM

[mouton29:]

The point is clear enough but since this argument about "changing the question" is a common one, I think it is worth flogging a bit more.

The question was clarified, as Keith says, to reflect (i) the language in the FDA's statutory mandate and (ii) to reflect the phrasing that has used been in all prior examples I could find on the FDA website. Specifically, 21 USC 355(d)((5) permits the FDA to refuse an application if it finds that "there is a lack of substantial evidence that the drug will have the effect it purports or is represented to have under the conditions of use prescribed, recommended, or suggested in the proposed labeling thereof..."

Thus, the FDA seeks to determine whether the evidence establishes efficacy, but that standard for that is whether there is "substantial evidence."

What happened at the hearing might be likened to a civil trial in which the judge initially instructs the jury to determine whether the plaintiff should prevail in, say, a wrongful death action. The jury makes it clear in their deliberations that they think this means that they must find that the defendant is guilty "beyond a reasonable doubt." The judge clarifies and instructs them that no, as this is a civil trial, the legal standard is, that the plaintiff must establish his case by a preponderence of the evidence.

2007 Apr 10 06:50 PM