Register for free to join our community of investors and share your ideas. You will also get access to streaming quotes, interactive charts, trades, portfolio, live options flow and more tools.
Register for free to join our community of investors and share your ideas. You will also get access to streaming quotes, interactive charts, trades, portfolio, live options flow and more tools.
I was not judging Cytodyn’s CEO to not follow the DSMB recommendation. It is another topic.
I was saying the DSMB recommendation to have a new interim for « sample size reassessment » and extending the mortality measurement from 28 to 43 days contradicted the scenario exposed by the previous poster.
These are facts.
Excepted Cytodyn's DSMB did request another interim review at 75%.
I quote Nader Pourhassan (Cytodyn's CEO) 's October 20th conference call, they requested this new interim to perform a "sample size reassessment". Which is precisely what the DSMB needs to do if they doubt that they will have enough patients to detect any hypothetical treatment effect In such case they want another opportunity to reestimate where they are with more data, before the clinical trial disbands.
And if they see a risk that the data may still not be sufficient to make a better guess, they also try to gather more signals, by extending the death measurement time frame. That's exactly what they also did, asking to be able to measure mortality at 42 days (was 28 days previously) after initial treatment.
It is Nader who has refused to follow their recommendation, not the DSMB.
Excepted it's HGEN who got a (well deserved) approval and Cytodyn is absolutely nowhere on the statistical, regulatory and financial fronts. But they can surely sell some dreams about the next delusion, the long haulers. Merry Go Round.
In the real world, they have by law 5 business days left (AT MOST) for the next toxic loan repayment, $7.5 millions. And they are not in a position where they can negotiate at arms' length with the toxic lender (Fife). Anything he asks, they will have to provide. Last time these were shares at $3.4, a nice discount. This times may be much worse. Implosion coming. I will write about that in detail, the clock is ticking and Cytodyn's financial bomb seems armed.
I did not produce the original Cytodyn analysis. I only corrected its 2 most striking points. I could not agree more with you: LA hospitalized numbers, as they stand today, are not good proxies for Cytodyn's CD12 scope. Arrived too late, do not represent correctly the 18 centers, bad proxy (hospitalized vs severe/critical but excluding the dying ones).
And yes, disclosed numbers will settle every analysis out there. Just like the previous COVID trial (CD10) which was a failure and has proven the statistical-oriented approach was spot on. Just like Theranos.
My analysis of both Cytodyn points reaches the opposite conclusion
(1) Probability that the trial will be successful (statistical power)
Refuting claims of low power, a colleague handles stats just fine - he back-calculated (if I recall) initial 80% power against low-30%s death reduction to 90% against high 30s.
Note the number of patients. More than x20 times more then Cytodyn’s CD03 HIV study. Legit work, more advanced, better funded, legit firm.
And no patients have mysteriously disappeared from the study like Cytodyn’s CD02 HIV study’s 4 patients that are mysteriously missing from the (suggested as stat. sig.) computation when you compare the various tables figures (more on that another time, another Cytodyn gem). This is real work, with a well deserved approval.
Press release from the FDA itself:
FDA Approves First Extended-Release, Injectable Drug Regimen for Adults Living with HIV
For Immediate Release:
January 21, 2021
The U.S. Food and Drug Administration today approved Cabenuva (cabotegravir and rilpivirine, injectable formulation) as a complete regimen for the treatment of human immunodeficiency virus type 1 (HIV-1) infection in adults to replace a current antiretroviral regimen in those who are virologically suppressed on a stable antiretroviral regimen with no history of treatment failure and with no known or suspected resistance to either cabotegravir or rilpivirine. This is the first FDA-approved injectable, complete regimen for HIV-infected adults that is administered once a month.
The FDA also approved Vocabria (cabotegravir, tablet formulation), which should be taken in combination with oral rilpivirine (Edurant) for one month prior to starting treatment with Cabenuva to ensure the medications are well-tolerated before switching to the extended-release injectable formulation.
“Currently, the standard of care for patients with HIV includes patients taking daily pills to adequately manage their condition. This approval will allow some patients the option of receiving once-monthly injections in lieu of a daily oral treatment regimen,” said John Farley, M.D., M.P.H., director of the Office of Infectious Diseases in the FDA’s Center for Drug Evaluation and Research. “Having this treatment available for some patients provides an alternative for managing this chronic condition.”
The safety and efficacy of Cabenuva were established through two randomized, open-label, controlled clinical trials in 1,182 HIV-infected adults who were virologically suppressed (HIV-1 RNA less than 50 copies/milliliter) before initiation of treatment with Cabenuva. Patients in both trials continued to show virologic suppression at the conclusion of each study, and no clinically relevant change from baseline in CD4+ cell counts was observed.
Thats why I did a sensitivity analysis. To make sure that my statistical results were not broken if, let's say, we were to reduce these 87 Cytodyn's trial deaths to 83.
I already wrote about the fact that I did some sensitivity analysis, I am repeating myself. As I said in the beginning, it may be useless to discuss mathematical details. I happily accepted based on 2 posters' request but if the questions loop, I will just send back posters to the previous answers
For 3 reasons:
(1) Calendar shift: I write that today but could not have written that before interim deaths were revealed. These professionals could not know either upon trial initial design in March. This is not a matter of skills but of available data. I would have made, I believe, the same mistakes back in March 2020 because, based on available data, we could only assume a very high mortality. They acted reasonably.
(2) The DSMC statisticians did request something during interim. It is wrong to say they "saw nothing". They obviously saw something. They requested, I quote Nader Pourhassan (October 20th conference call), a new 75% interim to perform a "sample size reassessment". This is the technical word for saying they want to reevaluate the number of patients needed, dont have enough certainty at 50% to do so, and so request a new interim at 75% to do so before the party disbands.
(3) As for the other scientists, they are all paid, remember, by Nader Pourrhassan. Pattersson has shares. Amarex is paid by the company to put out favorable posters (like other biotechs). The CMO is paid by Nader - the previous CMO resigned precisely because he accused Nader of lying and exagerating results with his investors (lawsuit, read the docket). My belief there is that Cytodyn is a Merry Go Round fraud, corrupting everythinig it touches. You could say this is not a statistical rationale, I agree but, ironically, there is a statistic: I went through all the Press Releases of Nader over the last 3 years and counted the number of failures (promise delivered on something announced with great enthusiasm but that is not trivial to pull of and was not delivered/postponed until now) vs success. Guess what, 0% success, 100% failure. Another user calls that the "Nader constant failure model". It is so true
Yes, there is a method - you consider ALL possible values for SOC mortality (eg 0%, 0.1%, 0.2%, ... 20%, 20.1%, ...) within the range indicated in these research papers (that I enlarged to make things more favorable to Leronlimab) times ALL possible values for Leronlimab mortality ((eg 0%, 0.1%, etc..). That gives you an infinity of combined scenarios. You fit each of these scenarios to the disclosed data (number of patients per arm, combined deaths at interim, combined deaths at completion) to compute the scenario likelihood. For example, the extreme (ridiculous) scenario of 0% SOC mortality / 0% Leronlimab mortality has a 0% likelihood, because it cannot explain the deaths observed. A 20% SOC mortality / 10% Leronlimab mortality has a non-zero likelihood, because it could explain the 45 and 87 deaths.
That gives you a range of scenarios and probabilities.
You can then compute for each of these scenarios its statistical power (the probability that Cytodyn will "win" the trial primary endpoint).
You end up with millions of scenarios, and their individual likelihood and statistical power. You can from that infer, from the initial death numbers, the overall probability that Cytodyn "wins".
Based on trial completion deaths (87) I ended up with less than 1%
Increase the number of deaths and the situation becomes even worse for Cytodyn.
(1) I never wrote that or assumed that. You got confused between 2 different things: reverse engineering Cytodyn's initial mortality assumptions, and simulating all the potential scenarios that could have led to the interim and completion numbers of deaths.
(2) Not relevant due to (1) above
(3) The same, you are conflating 2 separate topics. I never said that or made that assumption.
(4) Not relevant due to (3) above
(5) The same, you are conflating 2 separate topics. I never said that or made that assumption.
(6) Not relevant due to (5) above
(1) Nope upon interim what you wrote is wrong- in many cases the DSMC cant make a decision at interim, statistics in the grey area. Sorry this is the case of so many trials
(2), (3) and (4) Based on the difficulties around (1) I maintain the points
This is a theoretical rationale that is not specific to Cytodyn and applicable to all models and the bayesian approach in general as well as a large chunk of the frequentist approach. Yet this doesnt prevent Google, oil majors, Criteo, Facebook and many other firms to use these extensively. And biotech firms and the FDA for many, many trials. Dismissing this model based on a general rationale is not sufficientt.
The proper solution to manage this risk is
(1) to make minimal assumptions in the model (Ockham razor) and chose common sense assumptions (real world physics). This was the case here.
(2) run sensitivity tests regarding the assumptions and verify the conclusions still hold with alternatives. This was the case.
This Cytodyn trial is facing an uphill battle.
It’s a bayesian model fitted on the disclosed data (nb of patients in each arm, combined deaths) with large ranges for Leronlimab mortality and SOC mortality and flat priors, followed by power calculations for each of the multitude of potential alternatives.
Synthesis is here:
link
This is not appeal to authority...just that it is impossible to discuss this Cytodyn statistical analysis here if I am alone in that discussion because the skills needed for understanding each other are specialized...have a nice day
This Cytodyn fact was already known last August. So this is far from new. It is just an artifact due to NASDAQ's website data feeds...try it with other stocks, in doing so you will realize they integrate many non-NASDAQ stocks in their website...
All points are wrong
(1) As I already explained, Cytodyn's DSMC must stop a trial early (for efficacy or inefficacy) but only when he can. And here lies the subtlety: in many cases the DSMC doesn't have enough accumulated data to make such a strong and definitive decision. They have the obligation to favor doubt in cas of doubt. Otherwise (think about it), all trials across the planet would always be stopped early. If this is not the case, that must mean something is wrong with the rationale you used.
(2) Yes I reverse-engineered Cytodyn's numbers to determine their initial mortality rate assumptions, which led me to realize the trial is under-powered now (given current knowledge about COVID mortality rates that have fallen considerably since March). Sorry explaining the details would help none of us - due to the statistical skills needed for such a discussion
(3) I insist, the higher the combined mortality rate, the lower the probability that Leronlimab meets the primary endpoint of the study. It's statistics again
(4) Knowing the distribution of deaths is great but not required to infer that the study has most probably failed. Because we already know the total (combined) number of deaths. Again, it's statistics. But I can illustrate with an extreme example: if all patients had died, you would already know the trial is a failure. No need for unblinding for that. Now, if 99% patients had died, the same. If 98% had died, the same...etc...and at some point of course, we don't know. So there is, between 0% and 100%, a frontier where we can start infering that the trial is a failure. I computed it based on an exhaustive review of the litterature for severe/critical patients. The reported combined deaths (87+ deaths) is above that frontier. Statistics.
I have built a statistical model based on Cytodyn's disclosed data (patients, combined mortality at interim and completion) as well as exogenous data from 50+ worldwide research papers. The probability infered by this model that Cytodyn's reaches statistical significance for its primary endpoint is < 0.5%
It's over.
I am not saying this is not Cytodyn's DSMB responsibility - I am saying this is often not possible to make such an early decision (whether it is for early efficacy or for early inefficacy) because there are not enough patients to prove anything at interim. Think. Otherwise if that were always possible, ALL trials would be stopped upon interim review. Statistics are hard, particularly when you don't have a lot of patients. 195 is not a lot given the SOC mortality rate we know now.
So how did the media take the picture of Cytodyn's "miracle" patient - and who told them? This is truly a miracle if a journalist could get a picture of her and write an article without even being aware of her existence. My bet (Ockham's razor) is that this was orchestrated by Cytodyn's CEO. As everything else
I am just trusting the best in the biotech field to judge Cytodyn's events and leveraging their vocabulary, approach, criteria for a trial's success - I invest with their methodologies - they are experts. Have a nice day
I understand. You are using your own custom definition of Cytodyn's trial success. While this researcher uses a standard definition used by experts in the field. Explains the discrepancy. Subject closed.
Oups Cytodyn's stock just lost 10 cents while I was reading this. Clearly one of the enumerated points must be wrong.
So now we have just learned that Cytodyn's "miraculous" patient leaves a couple miles away from Cytodyn's CEO (out of the entire North America continent! what a coincidence) and had violent conduct. Not exactly someone "independent" from the company nor that can be trusted.
Question for the SEC: does she have any financial interest in having the company succeed.
Sorry, as the specialist's research paper I posted previously clearly says, statistical efficacy is an integral part of a trial "success". Cytodyn completed its trial but, no, this is not a success
So we both agree this was only Nader Pourhassan (Cytodyn's CEO) 's decision, against the DSMC analysis.
He projects confidence, against his own specialists. Exactly like the previous COVID trial (CD10). He was very confident and it became the total failure that it is today. Zero FDA approval, or as the Christmas Eve press releases says on the very last line, "won't support a request for an eIND".
Cytodyn is following the Merry Go Round trick, as I already wrote several times.
It's time for the long hauler trial to take over and soon the CD12 trial will be history.
Wrong, the Cytodyn trial can not be deemed a "successful" just because it got completed.
The definition of "success" for a trial encompasses efficacy, among other things, as illustrated by
this paper from a specialist in the field.
Wrong
- The Cytodyn trial would not have been necessarily stopped if there was no separation from Placebo in terms of efficacy. It is equally hard to stop early for inefficacy than it is to stop early for efficacy. You need quite a lot of patients and, as I already wrote, my analysis is that the Cytodyn trial is under powered, it doesn't have enough patients to prove any hypothetical effect (which anyway I think doesn't exist). The reason it is underpowered is because I believe Cytodyn designed the trial back in March based on much higher mortality assumptions. That's what I found out reverse-engineering their trial parameters with a statistical model
- The more deaths, the lower chance to have a positive outcome. Because Leronlimab had 2/3rd of patients and they already are, across arms, in the top-range of ALL worldwide studies of SOC mortality for Severe+Critical patients. More deaths means Leronlimab is not pulling down the mortality as it is supposed to
- There is no fact to support this: "They know it is very favourable to the drug and that's the reason they gave out that number in the last conference call.". If I were them I would have avoided disclosing that number if I could have done so because this is a bad number knowing SOC has low mortality and getting lower every month. Hence I believe they only disclosed it because they had to for obvious legal reasons (material piece of information for shareholders, hiding it would be a SEC crime).
- Billing codes have no relationship with future FDA approvals. Same for open label cohorts.
- For CD10 you wrote a post-hoc rationale but the ex-ante rationale (that I didn't believe in) was exactly the opposite. It said that it would be a tremendous success. My point about Leronlimab to the contrary hasn't changed before and after CD10: I believe Leronlimab acts like saline against COVID, and in fact against most diseases. Whether it's mild, moderate, severe or critical COVID, same failure against SOC. Plus, SOC has been greatly improving, so I view this battle that was already doomed as hopeless.
No confusion I was just saying that the trial was not successfully completed
These are not subtleties. Only detailed published data or a FDA approval can demonstrate that a trial is successful.
One example: Cytodyn claimed that its precious COVID trial had “impressive results”. Excepted... it has still not published the detailed data and, 6 months after, the FDA has demonstrated the results were not impressive at all, a failure.
The secret of a Merry Go Round stock fraud (which, in my opinion, Cytodyn is) is to claim wins everywhere while everything fails and introduce new shiny things to distract shareholders from the staggering failure rate. Merry Go Round.
Nope Cytodyn didn't "successfully complete" the pending trial. To the contrary, it is not disclosing the key information, the mortality rate that I predict is really bad for Leronlimab, based on what has been disclosed previously.
And no these manuscripts don't prove what you wrote, despite their title. Because they don't prove the causality, that Leronlimab is responsible for these decreases. Only a comparison with non-Leronlimab patients could prove that, and it's precisely what the past CD10 trial and the pending CD12 trial have been designed to measure. Excepted...the company has still not disclosed the CD10 detailed data, 6 months after touting "impressive results" followed on Christmas Eve by an admission of failure. Katy bar the door...
It’s over. I am a statistician and for a specialist who knows how the numbers relate to each other, what are the crucial pieces of informations and what is not, and how to drive the DSMC decision process, the numbers and explanations published by the company reveal a very bad picture of their trial.
I also think as a specialist that their FDA issues with the HIV BLA are not minor, contradictions in the numbers there too.
Last, beyond trial statistics, my opinion is that this company is a Merry-Go-Round fraud. There will always be “one last step” and yet “another promising application” for Leronlimab.
Wrong. The DSMC HAS requested trial changes to Cytodyn
If that were true we would not have this (quote):
"The rate of viral load suppression post-10 weeks of monotherapy was 68%, 94%, and 85% with 350 mg, 525 mg, and 700 mg, respectively."
Wrong - the DSMC HAS requested changes to the trial. Too bad...
It has requested them to, I quote Nader Pourhassan, Cytodyn’s CEO, perform a “sample size reassessment”. Specialists know what this means and this simple fact destroys any hope we could have had the Leronlimab would be the miracle drug whose trial is successful without any change.
No. More. Effective.
In Europe it is pretty common to use “base points” in structured finance. it was everywhere, from contracts to discussions to specifications for new products Again in a global context, judging based on your culture and where you used to work can be very misleading. Cytodyn’s CEO is Iranian and speaks with a strong accent. Still, he must be skilled (for frauds I believe) to be so successful. A $3 billion market cap is something few people have achieved. You would not be able to judge that from his vocabulary or accent.
As your question regarding NP, sorry I lost the point/question
The probability that, out of pure luck, Samantha Mottet lived in the same place than Cytodyn’s CEO is 0.01%.
That’s as bankers call it, one base point. 1% of 1%. A razor thin, incredibly low, probability. Highly significant by any standard. Clinical trials get approval for 500 times that. I hope the SEC is investigating their relationship.
Maybe she has been approached by the SEC?
Did you know that she lives nearby Nader Pourhassan (Cytodyn CEO)?
I find it amazing that out of more than 300 millions inhabitants, an entire continent that is North America, the one “miracle” patient (not my belief but that’s how she has been presented) that has been hailed as THE proof of Leronlimab’s alleged miracle happens to live very close to the CEO. Makes me wonder
(A) if they have agreed on anything
(B) if she has any financial incentive in Cytodyn’s success
This coincidence is, again, incredible. I have a higher probability to win a fortune at the casino in one night than them both living a couple miles away from each other.
No, again it was a response to a one legal hypothetical scenario raised by the poster.
Now as you ask for my opinion about Cytodyn, it is that both the primary endpoint and all secondary endpoints will be total failures, and I believe the company is a total fraud whose only objective is to enrich its insiders through a merry go round scheme. I hope the SEC will investigate this company and I will do everything I legally can to facilitate that.