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.
Thanks Biobonic,
Interesting : another trial stopped by the IDMC
Worth noting this is a preplanned efficacy and futility
The key word different from CVM trial is preplanned efficacy analysis, whereas when talking about CVM IDMC meetings one has to assume that they are running efficacy analysis, and again assume that the IDMC is not ran by idiots that just look at safety (nobody being dosed anymore their meeting would be without purpose but to get a paycheck if they would only do this) and give a free pass to the next step.
That is the "little" uncertainty that makes the charm of this stock and also allow crazy interpretations by the horrible short lobby.
I did
Sushi is getting emotional and I hope he has news
I am getting emotional too,
I mean those who shorts and bash biotechs and companies aiming at curing people in order to make some profit should be banned from the human specie
I do not see that as a bad either. It will dilute where it would not have diluted, just in case it goes way above 13.75 by end of June
I am assuming the following from this :
- there will be no news before 28/05 (this one was obvious, but still is true ... )
- there will be likely be no news before end of june, otherwise he would just have changed the date without the strike price, therefore Geert is confident pps will be north 13.75 but not so confident north 19.75
- by setting this strike price, it gives an incentive for the fund to trade it and push it higher than 13.75$ which is good for everybody to have someone with deep pockets pushing it higher
- he is confident there will not be shorting collateral
- Apart from pure friendship I imagine that an undisclosed win win deal has been agreed upon
I am curious about who own those series V.
There is MMcap for sure with 100k of warrants, but for the remaining 700k I do not know and I do not think Sushi knows
Good call Sushi !
Have no idea who has the remaining 700k warrants
Guys
Have a look at NVAX
Since I wrote this post: 300M market cap to now 2,5B
Ok ok they are covid best in class with 388M funds from NIH and and announced trial on humans soon (phI)
But that's what this crazy period is all about right now
Good day and GL to all
Fosco
this was a fake
I saw that link on Yahoo and responded the following :
While it appears as a balanced opinion, the mentioned biotech does feed the bear case. So worth analysing it to kill this in its nest
The author states that "The bears contend Multikine will likely end up in history as another cancer immunotherapy that failed phase 3 due to better than expected performance from the SOC cohort. One such example will be discussed below. Newlink Genetics"
I have seen quite a few tentative comparisons between CVM clinical trial duration and Biotechs whose CT lasted “longer than expected” and finally failed, in the same fashion than CVM clinical trial as per the bearish thesis. Each time I have investigated the details, both cases did not compare. The devil is in the details, but details do count when investing a lot of money.
For instance I have seen something that compare this trial to Vical’s Allovectin-7. Vical’s delay in trial was due to dropouts. We have factored dropouts in Multikine study.
So let us have a more precise look at this Newlink Genetics trial that ended in 2016
First I looked at inclusion criteria
I had a brief look at Seer Data for Pancreas, following inclusion criteria (resected tumor)
CT
“ Patients must have undergone surgical resection for the tumor and extent of resection must be either R0 (complete resection with grossly and microscopically negative margins of resection) or R1 (grossly negative but positive microscopically margins of resection).”
There has been literature on resected pancreas cancer survival such as
this one from 2007
“ Patients who underwent an R1 resection had a median overall survival of 21.5 months compared with 27.8 months in patients who underwent an R0 resection. “
The 2007 study says resected pancreas median survival was shown between 22 and 28%. (Have a look at table 6)
Then, as for the survival trend, I had a look at Seer database and posted survival in this Table 1 in this sheet
The pancreas SEER data showed SUSTAINED increases EVERY YEAR, some small, some in large amounts. And while the pancreas statistics fluctuated (see years 2006-2007, 2010-2011), they still showed SUSTAINED increase in survival over time. From 28% in 2005 to 46.2% in 2013 Y3.
Knowing that overall median was 29.3 months in this trial I am not impressed by how 29.3 compared with this 2007 study (22 R1/28 R0), which showed huge variance in data as well. With MK we are talking of 24 months at minimum median improvement, and variance/SE is lower as per seer database,
So combine 2007 study results with yearly increase from the past & fluctuations ... then 29% overall was very achievable in the control arm.
Bottom line : Failure of Newlink Genetics trial was predictable back in 2016 and also from the interim endpoints measured extrapolation and I would not have recommended to buy this stock back in 2013 or 2016 only judging by overall survival which reached 29% median in both arms
By contrast, The SEER data for Oral Cavity cancer do NOT show SUSTAINED yearly increases in survival for our subset. see
table 3 in this sheet
Indeed the devil is in the details as this study, such as the previous ones I have seen, do not compare to what we are witnessing at Cel SCI .
Love answering my own post
All "vaccine drugs" have been going up big
Look at NVAX trajectory since my last post, now it's 1 B market cap
Look at INO, now it's a 2 B m cap
If CVM gets a grant from NIH, expect another big uptick.
Ideal would be announcement of start of ph I trial, like NVAX
Dear Sab
I love when you use the word "parabolic"
I have not been following very well all your explanations, but at this point in time what is your overall analysis regarding price action / volume and what to expect ?
Thank you
Thanks for the update and kind words Proxima,
I do not know much what to think of this IDMC statement
To me it does not make sense to have them work on data at this point in time, except than having an unblinded view before the unblinding which purpose would be weird
I did attend the call,
I was somewhat frustrated it ended so quickly, I thought there would be lots of questions
I never had the sense Geert was trying to manipulate
He seemed frustrated as well to have so many "guests" and so few "shareholders" logged in
He reiterated that 298 was very close, eg not there (that he knew)
He said that they would disclose, or not, when 298 happens, but anyway it is not material because Cel SCI will still be blinded.
So like s/o I know would say, "298 would be a nothingburger" at this point, and I confirm, as for me maths do not change much whether 298 happens in 05/15 or 05/20, what will be important to know is the survival in the soc arm and the dropouts numbers which are blinded.
I believe unblinding will happen right after datalock, datalock is required to clean the data (by a specialized company) and store it in its final, non-modifiable state in the database. Only after data is locked, unblinding can happen to the sponsor, eg it will be known which patient took the MK and which did not. He said datalock could be delayed by Covid situation.
So bottom line:
- 298 is not here
- Cel SCI is still blinded
Knowing Geert, there is no question that he is not skilled enough to be able to answer without breaching the law (He is a lawyer, ins't he ?). I understand things could be different if he had some knowledge of the trial outcome, but as 298 is not there yet he is still fully blinded and has nothing (negative or positive) to hide.
So no concern at all on my side, just frustration like everybody else. I think they should re-plan a Q&A session. With a working system this time, just to calm down the nervous shareholder base.
Fosco
PS : By the way did he mention IDMC ? Will there be an IDMC meeting like someone wrote ? I doubt very strongly there should be one.
Between us : I volunteer to be prime to having the raw data
It won't take long for me to make an investment decision afterwards (a few minutes... ok may be hours)
Fosco
BTW : I checked the fresh new SEER data (2019 file that includes up to 2017 data) and there is no much change in survival unfortunately for the patients
I have put the latest table in this tab (you have to scroll down)
google sheet tab
As you can see, patients treated in 2016 did not have much 1 year survival expectancy than the previous years. Same for 2015 patients Y2, 2014 Y3 and so on
GL all
@Sab
Hoping you, your wife and daughters are going well
Here is the impact, it is helpful for assessing the Standard of Care baseline
Previous data could give at best five year survival for patients cured by US SoC system in 2011 :
This one will give :
- 5Y survival for those treated in 2012
- 4 Y OS for the 2013
- 3 Y OS for 2014 patients
- 2 Y OS for 2015 patients
- 1 Y OS for 2016 patients
This helps building the SoC baseline that serves to compare current survival pattern (linked to the "298 not happened yet") and provide an estimate benefit in the MK arm
It does not predict endpoint as such, endpoint was predicted by :
- leaked progress data (such as Ergomed 96% or Taiwan 208)
- 298 statement :"it has not happened yet" or "it is expected by 15th of May"
It could have predicted the theoretical endpoint, eg the date that 298 should have happened under normal circumstances, or the one used for "median survival improvement" measurement, which is when 50% of patient are deceased : you compare overall timing (since treatment) required to reach those 50% between soc and test arm.
Take care
Fosco
Indeed
decision not made easy for the governments
nobody anymore (see Sweden) dares defy the lockdown method except Iran who is opening businesses again amid economic catastrophy.
That tells you lots
On another topic, I just received new data Nov 2019 from SEER (includes 2017, previous only include 2016). A few paperworks and I will be able to access it
Will let you know the outcome
Fosco
New article out !!!
Nothing new at all, but I wanted to write something out of boredom
Article SA Blog
GL all
Fosco
LR : Currently in France, a self sustainable producing country, farmers need 200 000 workers to help them produce as a result of supply chain disruption and foreign workers ban.
I do not think we will lack ability to feed ourselves (not true in some countries) however those who were mocking when I bought 100 pounds of rice 1month ago are not anymore !
@Robot
No pb I did not take it harshly
I have just updated the PIII chart assuming 298 has not happened yet as per the latest filling. We were at 20% one month ago or so. Now it should be above 23% with what I believe are conservative assumptions (20% dropouts, better SOC OS), so 30% or more would not be surprising.
Take care
Fosco
@Robot : You said "It is therefore very early to make any conclusive statements about what the overall mortality rate will be for the novel coronavirus, according to the World Health Organization"
... I fully agree with this statement but at least one knows which countries test a lot and one knows they have less more deaths per cv positive than countries who only test the most severe cases, therefore your death rate should be mimicking the countries who test a lot (less than 1%)... all this is, of course, "Hopefully".
@LR: putting the economy as a priority VS confinement. Don't get me wrong, I do not have an opinion on this as I do not have a crystal ball but just a few remarks :
1- I do not think any democratic leader in the world can sustain this stance, even your president.. see Boris Johnson. The worse thing would be to go through this hazardous path and then changing mind : you would have much more deaths, collapse of healthcare system and finally confinement. Think of president strong supporters who see relatives starting dying : uncle, brother, mother, wife etc... they will urge him to step back... too late. I do not think he will not change his mind and this is the worse that can happen
2- On the other hand, doing a too short confinement will not solve the problem. It has to be long, followed by everyone to smooth the infection rate or it will be useless.
3- If real death rate happens to be above 3%, wow, you'll have your economic disaster anyway
JMO of course
Take care of yourselves
Fosco
Let's keep it clear, coronavirus death rate is probably less that 1% worldwide, with higher % in older population, lower in younger
I believe that what changes is the number of tests carried out in Germany VS Italy. Italy has probably a huge cv positive,untested, population. Germany tests many people which reflects something close from the true infection rate, and therefore the death toll seems lower. What is lower is the infection of the population. Whether they want to keep this infection low will depend if they implement a successful containment strategy
So it will be the same for the US, 100 new deaths mean that about 15 days ago 10 000 new persons were infected and if these folks have not been contained then they might have contaminated in the meantime 80 000 new folks. Whether you want to keep this infection low will depend if you implement a successful containment strategy.
Trump says "it has to be over by Easter". We do not really know what it means : containment or contagion.
If you think that Trump will lift containment by Easter to save the economy, then be ready for piles of bodies all around the corner. Not much in % (less than 1%), but still a lot and brutal (1 to 3 millions americans, shortening their life expectancy by 1 to 10 years).
If you think that Trump means contagion, than this is pure wishful thinking. It won't be over until a successful vaccination as we can see that even countries which managed it indoor have to face now more imported cases.
Sorry but this is the grim reality
not totally OT but great article from collegemate
https://medium.com/@tomaspueyo/coronavirus-act-today-or-people-will-die-f4d3d9cd99ca
Circulate if you will
Fosco
yeah good idea Chuckster Or Courchevelle
omg
Because of Trump won't be able to make it to the Vegas Party, unless I use a British Airways flight. (May-be he's got BA shares)
Hi Spidey,
Thanks a lot, finally I gave up, as she did not seem to want to understand, or at least acknowledge all her dd was poor.
The good thing about it is that I was so P/O that it gave me time, ideas and energy to produce a new article so that it can benefit a wider audience
Take care
Fosco
Well, thanks to recognize my merit Sab, ahaha !
Frankly I don't know if the jump above 10 was due to this, but the coincidence/correlation with INO and NVAX was too tempting not to note it,
In any way it will help us, for the moment being to use the bad market tide to our advantage. I guess other timely realeased news will follow (I am longing to learn about those "partners" mentionned in the release).
Expect the 1B market cap sooner rather than later
On my side I sold all my other holdings, kept only real estate and put some funds in bearish contrarian indexes trackers
GL 2 all today
Fosco
Fair enough Static,
All your reasons seem very valid
Just wondering where Leaps has gone as Coronavirus epidemics is just about to hit all of us, and the only stocks going OK are vaccines.
GL to all, in all senses,
Fosco
Answering my previous mail dated 01/25
So we have NVAX climbing to new heights (downs being only due to tens of millions of dollars cash raised by dilution)
INO surged today as well on COVID vaccine news
So still nobody thinks this rally could be due, at least partially, to LEAPS ?
Good Luck to you too Cotton
Hi
Was quiet upset by last BS article in SA, so I updated a bit my CVM analysis if some want to have a look
Article in SA
Thanks
Fosco
Thanks Sushi
I guess this is the last expected date to get the 298th' patient data collected by the CRO.
Therefore :
1) The 298th event might occur before this date
2) Last possible date for data gathering & analysis (material event) might add a few weeks to this date. Material event might happen in April or May 2020 latest
As a side note, only the dates have changed
(see history)
This is intentional, they must have had some updates from the CRO
We are getting closer
Take care
Fosco
Thanks and congrats for your successful call !
Fosco
thanks przgwxl
I like the fact that you confirm that it is logical for the IDMC not to have stopped the trial at this point, but still the overall survival is really good.
I am not sure of the HR computation as well, it should have been around 80%
But overall I enjoy the way you compute data with your tools
Take care
Fosco
@Sab
Calling your wise advice in here
what do charts currently say ?
I found today's rebound quite bullish on my side, but no science to substantiate
Fosco
Let's meet on the slopes then
Where are you skiing ?
Currently in the french alps
Fosco
@lightrock
I understand your method. I understand that as logrank is measuring global separation of the curves, timing when both curves separate is not important as long as they separate throughout the clinical trial. I am surprised though, intuitively both curves should separate more are more as time increase, shouldn't they ? Because the more time passes the more test arm is supposed to be efficient, shouldn't it ?
@przgwxl
I understand you found 25% risk of failure with my conservative assumptions, even if drug is 20% superior. I am also a bit surprised, because these assumptions still imply at least 20% survival superiority. From Cel SCI clinical trial, power of the 298 sample is 80% based on 10% survival superiority, meaning exactly that the 298 sample has 80% of chance of success if drug is 10% superior : so both figures don't match. I am thinking that it might be due to the fact that the variance you are using around the mean is too big: I think you are using the binomial law. Have you thought of using the normal law with the standard error provided by the Seer Database (for instance from this link , Standard Error is between 0.6% and 0.8%, so I think a normal law can be computed with this SE around the mean... just an idea).
@prezgwxl
@lightrock
Thanks, interesting, so yes, if sample is bad, we still have probability of failure with "too good" survival figures.
Out of curiosity (not 100% sure how montecarlo works),how did you compute survival values : "random" shots around the seer mean (provided by log regression) following a normal distribution with the known standard error from Seer ?
Sab
Thanks For the science
Looks encouraging, in particular while the « virus » stocks are down
Fosco
Agreed !
That's what I tried to explain in this table
https://docs.google.com/spreadsheets/d/1iVkDqatmv_R9kQgKPYy-dLizhwAlaq-LQbD5T1venV4/edit#gid=452440789
Indeed thanks LR and PRZ !
I do retrieve the 201, 173 etc... as well depending on control arm survival. I am glad different methods seem to be converging.
Prz, your model does confirm dropout do not affect statistical significance as I always thought (which is unfortunately an open door to higher dropouts than expected) but confirms that it does affect potential efficacy if too high (I would say over 30%) and 298 should happen now, and therefore IDMC would have stopped for futility.
I liked very much that with my "conservative" model p was >0.05 in March'19 when IDMC met (around 152 events in control) and probably very close from 0.05 in September (163 events) and << 0.05 with 167 events. I am amazed how p value gets closer to 0 (5e-4) with just a few events (8) from your 159 threshold.
Thank you all
@przgwxl
Very interesting, very very interesting indeed !!!!
Thanks !!!
A few remarks questions (and I am feeling excited writing them)
You said : "The actual number of predicted events in the control arm is 201" . Where do you take this ? I am not finding 201 events in any model. May be in the model with SEER data for stage III+IV, but then I don't think it is realistic (in particular in the early years as too weak patients are being discarted in the study)
If I consider my latest model with 18.5% cumulated dropouts, then :
1) We are currently around 298 in total with 167 events in control arm : not so far from your "detectable" 159 !!!
2) In Early October 2019,when the IDMC met, we had around 163 events in my model
So in conclusion, p might have been may-be <0.05, but still very close from 0.05 which justifies a "trial should continue" in both senses : significance is good, but not so outstanding as to stop the trial
A question of curiosity : if you increase dropouts rates (say 25%, 30, 35), does it change the level of events required in the control arm to get p<5% ? I am in the belief that censorship does not affect significance as long as 298 is reached but not sure.