Introgen Therapeutics: Advexin's P-Value Borg - Resistance is Futile
On May 29, 2008, Joseph Pantginis, a research analyst from Canaccord Adams who covers Introgen (INGN) with a Buy rating, was interviewed by BioWorld Today about Introgen's recently released pivotal data for Advexin (available here under the search term "Introgen").
Even though the pivotal trial's statistical analyses and data had been massaged for what appears to be several years, Introgen was unable to make lemonade out of its Advexin lemon.
However, the research analyst somehow saw success. Not only that, but he chastised investors for their inability to understand or see his point of view.
This reprimand is not the first. In an October 4, 2007 research report, Dr. Pantginis issued a warning to shorts by likening the impact of the recently reported pivotal data to a "match" that would scorch them, saying:
We consider the short position in the stock to represent gasoline and the Phase 3 data the match.
To add mixed metaphorical impact to the flaming visual, he also said:
[the shorts will be] up the proverbial creek without a paddle.
So let's look at what he had to say.
Number One: According to BioWorld, Dr. Pantginis said:
While the study of Advexin in patients with head and neck cancer met its primary and secondary endpoints, investors were focused on data that showed there was no statistical difference in survival between Advexin and methotrexate, the comparator drug, in the intent-to-treat [ITT] population.
He also said that:
Wall Street failed to understand that the Advexin trial's goal was not to show superiority over methotrexate, but an increased survival with superior safety and tumor response, which it succeeded in doing.
In other words, the analyst is essentially claiming that the primary efficacy endpoint was a composite of safety and efficacy.
While I do not find Dr. Pantginis's analysis credible, it is consistent with what was said in a presentation on the pivotal trial results at the American Society of Gene Therapy [ASGT] on May 29, 2008. That presentation identified the primary efficacy endpoint as:
Survival with superior safety in the intent-to-treat or biomarker populations.
While I have no knowledge of what the investigators who reported on the trial "thought" the primary endpoint was in the past, I do know that on December 20, 2007 Dr. Pantginis issued a report on Introgen where he explicitly disclosed his understanding of the primary endpoint.
Here is what he said then:
Recall that the prospective statistical plan agreed to by the FDA includes the overall intent to treat population of the Phase 3 (response and survival primary endpoints), but also included the prospectively defined subpopulation using the p53 biomarker analysis.
On December 18, 2007 he said this:
The endpoints of the Phase 3 study are tumor response and overall survival, which includes the evaluation of prognostic biomarker defined populations.
I could go on but the point is clear - if investors misunderstood the true goal of the study, you helped because you never mentioned or appear to have expected anything but an outright survival advantage.
In any event, Dr. Pantginis's "analysis" and the presentation at ASGT are questionable.
First, Advexin did not meet either its primary or secondary endpoint as detailed in the original protocol and even under the most liberal definition of success and even if one concedes that there is such a thing as a liberal definition of success in the context of a clinical trial.
Here is actual text from a protocol summary dated September 29, 1999 (amazing this all started nearly ten years ago):
The proposed study in reproducing the survival outcome observed in phase IIa is powered at 98% two-sided test to establish a survival advantage of [Advexin as a] single agent versus methotrexate single agent, 7 months versus 4 months. It will have a 90% power to detect a difference of 4.5 versus 7 months.
Oops! It looks like the exact opposite happened. But perhaps all of this investor confusion is mere terminological inexactitude and shorts will, in fact, be set afire one day.
Now let's go back to the "composite" endpoint of survival with superior safety.
In the near decade since the Advexin saga began Introgen has never once referred to a composite efficacy/safety endpoint and even if they forgot to or were less clear than they meant to be in their historical statements, a composite endpoint of the nature apparently being claimed by Introgen is not a regulatory endpoint and no similar endpoint has ever been a regulatory endpoint for an oncology drug. Progression free survival [PFS] and disease free survival [DFS] are valid composite endpoints but what Introgen seems to be pointing to is nothing close to PFS or TTP. What the Advexin pivotal study demonstrated in my view was a novel ADWLT endpoint, or accelerated death with less toxicity.
Getting back to the specifics of Introgen's press release, the company refers this "exceptional side-effect profile" and states that "methotrexate side effects can be life-threatening."
That's an interesting observation, but guess what? Methotrexate's life-threatening side effects appear to be less life threatening than treating patients with Advexin as evidenced by the fact that patients treated with Advexin lived a median 4.4 months compared to 6.1 months for patients on methotrexate.
This might be copacetic with patients if they experienced some other sort of benefit; say for example, a reduction in tumor pain. Alas, tumor pain was measured in the pivotal trial and was highlighted among four other grade 3/4 adverse events.
But guess what? More patients on Advexin experienced tumor pain than those on methotrexate. On two other measures, asthenia and leukopenia, Advexin was better than methotrexate but the data points were not statistically significant.
According to Introgen and the AGST presentation, Advexin did show superiority to methotrexate in regard to the incidence of pneumonia. This was highlighted in Introgen's press release and in the BioWorld article in the form of a quote from Robert Sobol, senior vice president of medical and scientific affairs at Introgen, who said:
Methotrexate has been associated with serious and life-threatening adverse events, such as pneumonia, whereas Advexin's adverse events were mostly mild, such as injection-site discomfort or pain.
According to the press release, the incidence of pneumonia in methotrexate patients was 10.9% vs. 1.6% in the Advexin arm. These figures were also highlighted in the ASGT presentation except that the ASGT presentation included a prominent asterisk indicating that the pneumonia data for the methotrexate arm, but not the Advexin arm, were based on "sponsor assessment[s]".
If this were a joke it would be funny but it's not a joke. Introgen's management team is a well-regarded trailblazer in terms of data analysis but drug developers do not generally contribute their clinical evaluations of patients to scientific literature.
In addition, how the company employees made these assessments is not disclosed. Most disturbing of all is that clinicians like Dr. Nemunaitis allowed their names to be appended to sponsor-generated data. Will the principal investigator please stand up!
All of this safety talk is essentially meaningless though because Introgen failed to provide any information whatsoever on the statistical methodology they used to evaluate what they are undoubtedly saying is a composite endpoint. The reported statistical analysis only refers to survival and tumor response. Since they lost on those measures they appear to be asking you to forgive the survival penalty on the basis of observations about toxicity.
The problem is that there is a material and in this case fatal difference between a true composite outcome measure and merely reporting worse efficacy in combination with a more favorable toxicity profile.
Number Two: Dr. Pantginis says that Advexin "is up against the Wall Street gene therapy discount". Essentially, the analyst is saying that Introgen has been tainted by the failures of others. This is nonsense. Introgen has been tainted by Advexin.
Number Three: The analyst is quoted as saying:
The true measure of Advexin . . . is that the p53 biomarker data could clearly identify patients who could benefit and guide physician treatment decisions.
What benefit is he referring to? The benefit of dying quicker than he or she would if treated with methotrexate but maintaining a full head of hair? What he really seems to be saying is that Advexin is a diagnostic tool, not a therapeutic.
Finally and most importantly, one has to ask what exactly "survival with superior safety" is. I don't even have to run a pivotal trial but I can tell you with certainty that Jelly Beans are safer than methotrexate. I can also tell you that patients given Jelly Beans would survive, at least for a while. They probably wouldn't survive as long as patients on methotrexate but they wouldn't have any adverse events other than those normally associated with Jelly Beans.
Notwithstanding this, I am highly confident that the FDA would never approve Jelly Beans for use in patients with head and neck cancer, no matter how delicious they are.
Disclosure: none
Related Articles
|
Top Rated Comment Streams:
-
1.Hedged In661
- 2.
-
3.Smarty_Pants399
-
4.cos1000291
-
5.axelrod608290
- Free E-Newsletters
- About Seeking Alpha
- About Us
- Contact Us
- What's New
- Readers Feedback
- Advertise With Us
- Contributors
- Contribute an Article
- Feature Your Book
- Our Contributors
- Anonymous Contributions
- Dispute an Article?
- Legal
- Terms of Use
- Privacy
- Copyright
This article has 6 comments:
-
candewish
-
3 Comments
Jun 05 11:57 AM-
ND-LBC
-
2 Comments
Jun 22 03:38 AMA bulletpoint at the beginning of the Introgen press release dated 5/28/08 states:
-- Primary and Secondary Efficacy Endpoints Successfully Met, Study Objective Achieved
The first line of the press release speaks to the survival endpoint:
"Introgen Therapeutics, Inc. (NASDAQ:INGN) today announced that ADVEXIN(R) (p53 tumor suppressor therapy) significantly increased survival in end-stage head and neck cancer patients with prospectively identified p53 favorable profiles (7.2 vs. 2.7 months; p less than 0.0001)."
The first line of the second paragraph of the press release notes a positive tumor response:
"The increase in survival and tumor response, among the p53 favorable profile patients, highlights the value of developing 'targeted' therapy for patients with recurrent, refractory head and neck cancer," said Jack A. Roth, MD..."
According the the MD Anderson News Release dated 5/28/08:
Patients with a favorable p53 profile who received Advexin® had a median survival of 7.2 months, compared with 2.7 months for those whose tumor expressed high levels of mutant p53 before treatment. Patients with this unfavorable profile were better off taking the chemotherapy drug methotrexate, resulting in median survival of 5.9 months.
Further, I think the author of the above blog article misinterprets the quote from the same news release:
""The important finding is that patients who benefit from treatment can be identified with the p53 biomarker. The biomarker will enable physicians to personalize treatment," said Roth, who also directs M. D. Anderson's W.M. Keck Center for Innovative Cancer Therapies."
A personalized therapy is a major breakthrough in cancer therapy. Introgen and the physicians at MD Anderson view this as a major benefit and selling point, on top of (not in place of) the positive tumor response and increase in survival over methotrexate. I believe this accounts for the strong emphasis given to Advexin as a targeted and personalized therapy in the press releases. As researchers, they are excited about this breakthrough.
The author of the blog argues that Introgen has changed the endpoints, which it hasn't and that Advexin failed to meet these endpoints, but it did as shown by the data.
I wonder why the author of the blog went to so much effort to misrepesent the data? A large short position?
-
Jobu37
-
10 Comments
Jun 27 09:29 AM-
ND-LBC
-
2 Comments
Jun 30 01:53 PM-
Biotechy
-
1 Comment
Aug 13 12:41 PMUntil this data is released, it will be difficult to determine the efficacy of Advexin.
-
the biotech onion
-
2 Comments
Sep 02 03:12 PM