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.
ligthrock
lol, probably the universe had a glitch
As I was myself in the doubt (because it didn't make logical sense why wait for 298 and compare 3years survival while 3 years survival is already measured for 90% of the patients) I asked "at a median of 3 years" meaning to Dr Talor and he told me it was an estimate for the duration of the time it takes to get to 298.
And he repeated :
"Primary enpoint is event driven not time driven" which means, we need to wait till 298, not compare things at a specific year.
I am 100%, 1000%, 10 000% sure right now. So you can safely bury deeply this "3 years" thing.
I am sure also other posters will raise it again and again though
Exactly Mogles
In addition Dr Talor told me that they have tried to take homogeneous sets of patients to avoid discrepancies as much as possible
Agree but why would you need year 0 for ?
Last patients were treated in 2016
Exp model does fit well between Y1 and Y10, what is required for simulation, and is simple (to me at least) !
Oh yes I agree
Except on the 3 years mention.
Nothing, absolutely nothing, relates “3 years” to primary outcome measures (sucess criteria)
Static
I think cvm stat is too optimistic
i believe we are around 60% at year 3 with less than 10% dropouts
This will make at least 20% efficacy and 298 reached in October
Fosco
Static
He seemed convinced it would be around 50% at Y3
If this is the case, than obviously we'll be much higher than 30% improvement.
Fosco
You are basically correct
I will cut / past what I have written to make things clearer !
Like Dr Talor told me, primary endpoint is event driven and not a time driven, eg they need to have enough survival data to be able to draw conclusions. Enough data means 298 deaths and once there they compare survivors in each arm. If there are 10% More in MK arm, then bingo.
I believe they run also a logrank test, which is a method for comparing survival curves. If survival in MK arm is not the same as survival in control, then they can state that MK shows efficacy.
What is important to understand is that this does not compare survival benefit in terms of average years gained by patients, rather it's comparing those who resisted better to illness between test and control arm once 298 of them passed away.
Remember I wrote "I believe they run also a logrank test". That's what they have written they would do (See CT site).
Logrank test assesses differences in survival curves no matter which year they took the drug and which year the event took place. If more people have died after Year 1, Year 2, Year 4 etc... in SoC arm than in the other arm in total since the beginning of the trial, Logrank test will validate that MK is more effective. For instance the number of people who will survive from 2012 till 2016 will be cumulated to the people who made it from 2013 till 2017 ("n people survived 4 years after taking drug in SoC arm" and compares to "m people who survived 4 years after taking Multikine")
Another thing : forget about those 3 years numbers. 3 years relates to nothing. I am 1000% sure as I asked Dr Talor. Primary endpoint is event based and not time based : In the surving groups (800 - dropouts - 298) you will compare number of survivors, not dead people. The logrank test will compare difference in EVERY YEAR from Year 1 to Y 9 (for the 2011 people) not privileging a single year like year 3.
If you want to read more about it : you can take a little bit of time and read the following, it is well explained here with a good example in 2 groups :
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1065034/
NB a possible outcome measure (Not the PRIMARY outcome which is used to declare success or failure of the trial) derived from the Data will be comparing median survival in both group (eg time when 50% of patients die within
the 298) which is still not the average survival years after treatment, but gives a good indication of drug efficacy. The average survival years after treatment will only be determined after all patients die
I believe he programmed in Python and this is not available to public
I could do such thing, but I would need more time that I can spend right now. I have a 100+ p company to run. I could ask some of the programmers to do it in Python too but they are all busy doing something else
Rogue seems busy too, may be he 'll come out with some outcomes soon, but I rely 100% on his findings so far which seem coherent with what I thought
Fosco
Exactly Kpauliukonis Those journalists don't take any risk !
While there is "no guarantee that its phase 3 study will be successful" there is no guarantee that its phase 3 study will NOT be successful either.
Would there be guarantee that its phase 3 study will be successful,pps would superior or equal to 100 !!!!
(But in any case we got some exposure)
you are right, in the SA article "remaining" column shows -10%
I don't know why that has happened as the original sheet showed correct numbers. May be when I cut and pasted.
However the calculation shows a results for -5%
(Sorry for that)
Fosco
@rogue / lightrock
Regarding log curves. What is important is that they match Y3 (2016 patients) till Y8 (2011 patients), lets say Y9 if you want to extend till 2020
Take for instance UK original data for Oral Cavity
https://www.cancerresearchuk.org/health-professional/cancer-statistics/statistics-by-cancer-type/head-and-neck-cancers/survival#heading-Zero
it gives 3 points Y1: 78.4% Y5: 56.1% Y10: 45.2% plus a 95% confidence interval
log curve : b=-13.86 ; b=78.40 matches as : Y1:78.4 Y5 :56.1 Y10 : 46.5.
So we're pretty much at it AND, most importantly, within the confidence interval
I did the same for Taïwan with 4 points (Y1,3,5,8). For CVM I had only two points but overall we should be between confidence intervals, that's why I have chosen the log curve, it was the best match to those statistics within the Y1 till Y10 interval. Linear interpolation is wrong as it assumes a fixed number of events (not a fixed rate) while the population is decreasing, eg an increasing proportion of patient dying while in real survival analysis the proportion of patient dying is decreasing.
Rogue,
Thank you for you answer and again congrats for this fantastic piece of work
To answer in disorder to your questions.
Am I sure about those datapoints, the answer is that I am 90% confident as :
- it comes from an IRB minutes from which we had the links and original document
- I run my model with given efficacy to match Feb 3 2018 and when it did, it matched almost magically Feb 4 2017
- so far it gave me _ since January _ a predictive date for 298 (Q3/Q4) that has not been proven wrong so far while people were expecting early Q1 from Cel Sci or H2 from Ergomed
Also, to match those end point, I believe you would have to change your protocol a little bit, here is one method. It might be a bit lengthy though, so up to you if you want to spend time on this !
I don't know if you would like it as it has to assume some efficacy (ps I'd rather use Taïwan soc as in my static simulation it matches pretty well) :
1- consider two KM Life tables, one with SOC (Taïwan) and the other with soc + x % and 5% dropouts.
2- Setdate to endpoint date (for intance 3feb 2018 first)
3- run the stuff 10 000 x to compute most probable number of total (in the two arms) events in that date
4- adjust x so that most-probable number of events falls around 208
5- perform a reality check on 4 feb 2017 : run the stuff and calculate most probable number of events. Adjust x a little bit if required
6- Run the stuff and see which dates 298 is falling in 10 000 runs
7- Publish to us probability of having date between for instance D-15 days D+15 days, D being most probable date
8 - This should give us x expected efficacy and D date with a % of certainty
9 - Re-run at will with different dropouts rates etc... have some fun !
Regarding your last sentence, yes you are right, but considering that we don't know the reality, there are still three parameters that can lengthen the end date : soc OS, efficacy, dropouts. I'd rather prefer efficacy than the 2 others. On soc I believe we are close to reality. On Dropouts, its complete darkness except that we have an upper limit.
welcome Sab, and in addition to helping us watch paint dry, each month gaining 2% means at least a 10% increase in probability of success (with the key threshold of 8% to 10% which probably happened in the past)
Sab
Under a given Soc and Dropouts assumption, each new month that passes by without notice of 298 means an increase in MK potential benefit.
For Taiwan Soc, its between 1 and 2 % per month
For CVM given SoC , its around 2.7% better according to rogue's stats
So when we reached 10%, we can add those % each month to be able to assess MK efficacy... Pure bonus for us (as we only need 10%) but above all great news for the patients
"why wouldn't the approval group just stop the 298 events early and approve it promptly? " : feb might have been too early, but August ... who knows... few events left and an obvious outcome...
GO CVM !!!!
@Lightrock
Re : "That being said the simple math now is evidently just as good as all our efforts because of how far into PIII we are. "
Let's do it for the intense intellectual satisfaction it provides then lol
Take care and have a good sleep, looks like it's late for you
Fosco
ligthrock
Now with our 3 different angles for looking at it, I am getting more and more confident that we have a correct vision of the truth : don't you think so ?
Hopefully we could meet soon wherever in Austin, Paris or midwest cotton fields !
Take care
Fosco
Rogue2.
Wow, that's a fantastic job you have done.
For two reasons:
- First of all I understand it so I can validate it !
- Secondly it supplements perfectly my xl work and soc curves by giving probabilities of success ! Haven't programmed stuff for years I would not have been able to supplement my static XL with such macros or code to be able to run thousands of tries as you have done.
My remarks :
1) Intuitively, I would have put some 70% on Taïwan with end of trial by October. No surprise CVM gives 100% and is always above 10% in 2)
3) Agreed, I had similar finding.
4) Interesting
Now to go further on numbers :
- Have you tried to compute probabilities around a date for 298 ? Having in mind 133 events in 4 Feb 2017 and 208 events in 3 Feb 2018 will help.
- Have you read my interview of Dr Talor. There is a + and a -. The plus si that SoC OS is probably worse than litterature on oscc because most of patients have floor of the mouth cancer and likely more at STIV than STIII: on the low end of the survival curve (I would lower Taïwan by 3% to get 50% by Y5. Also UK soc is probably too good as it shows net survival and not overall survival. Dr Talor often talked about 50% at Y3, even lower than CVM's 55%). The - is that, by design, the study could have up to 15% of dropouts. I could not get the actual number of dropouts by Dr Talor, but I had the original study assumption from an old publication : 784 eligible patients are required to preserve the power of the study. That does not mean by any mean that it will be 15%, that only means that it could potentially hit that number. 5% is my hypothesis of work, but I could be totally wrong, the real max is 15%
Kindest regards and looking forward to your reply,
Fosco
@LR
Sure, should have figured out you had gotten that, sorry.
Not following much US politics right now, sorry for lack of culture.
In the meantime, let's enjoy the nice little fireworks kindly sponsored by Shorts & Co
6B sounds reasonable too, LR
Anything above 10 will content me for the moment being, I am a man of little needs in life
The cure stuff, is a counter speech to the Keytruda stuff. Keytruda extends your life for a few months : applied on metastatic, kills the tumor, but not the metastasis : you're doomed anyway. The cure eliminates micro-metastasis on a non yet invaded body, and cures you for good from the cancer before it starts spreading. "Intent", because it does not work 100% of the time (may be 10%), but yet even if not totally cured, you can extend your life much more than with Keytruda.
If you're cured, you don't need surgery, you can keep your tongue, face, jaw and most importantly, life. Time will be needed before you assess so,but it might be a possible phIV outcome.
Doctors and insurances should buy it for SoC.
Rather 300 pps fully diluted but still good
He mentioned several times in yesterday's conference this specific figure 12B$ as if he had it in mind...
There are many potential reasons why Geert would want to publicize and raise awareness at this point in time but I fail to see atracking Big Pharma attention as one of them. It's too late for them to purchase CVM prior approval and too early for them to make a post-approval offer.
This shall come only if PhIII is successful and he won't need to publicize : they will all know about it.
He already has set the price tag though : $12B
I gave a call to Cel Sci today (Same post on Yahoo MB)
I was lucky enough to be able to talk to Dr Talor and he has kindly offered to answer some of my questions.
So I am happy to share with the community today with the best of my understanding and my own words, sorry M. Talor if the wording is not as precise as it should be.
First of all he confirmed that despite the study was open labelled they were completely blinded.
He said IDMC should meet in general every 6 months, provided that there is enough to be discussed /analysed between two meetings.
Regarding 3 Years survival, he said that the nature of the tumor being treated (most in % is dealing with oral cavity) plus the staging (Inclusion criteria) made it eligible for the low-end survival numbers for Head and Neck cancer (in general H&N survival statistics given in literature talk about relative survival for a blend of lips, tongue, soft palate etc…which is better then what it should be from this Clinical Trial). He confirmed again that 298 includes all kind of deaths, therefore survivors are “overall survivors” and not “relative survivors”
He said :
- Survival hope at 3 years has not increased much for this specific cancer location over the years
- HPV do not impact much this zone of the mouth and therefore should not influence survival either
- No chance than Checkpoint inhibitors have had time to improve SOC from 2011 to 2016 in particular in the regions where the CT took place
He said dropouts should be contained as no more patients than the initial design were required to get statistical significance
Having all this in mind, he said that the fact that most of patients have been treated more than 3years ago (some more than 7 years) and 298 not being reached shows a survival pattern that is in not in accordance with what should be expected from SoC alone. (Personal note : This point is the main evidence given by the models that we have run so far that tend to prove that the drug is efficient)
On the question of Primary Endpoint : he said it was Event driven and not Time driven (as I suspected). They have to wait for 298 and then they compare survival in two arms and there is the need to show 10% Overall Survival improvement in MK arm. Average 3 years simply means the time they though 298 would be reached when they designed the study. Timeframe is required for Secondary endpoints such as Progression free survival, QoL etc…
Finally on the question on why this Science would work while every research seems to focus on Treg inhibidors : he simply said because he saw it work on more than 200 cases in prior studies ! He said that the fact that they treat prior SoC on an intact immune system without delaying SoC is a key point in the success of the drug that is being considered seriously now by some research oncologist and should be an eye opening if CT is successful.
Bottom line and my personal conclusion :
---------------------------------------------------
All the models ran so fast, based on time it takes to reach 298 event indicate that there is some efficacy shown in the MK arm. Most models show a strong efficacy over 10%. This discussion with Dr Talor only reinforces this feeling as it emphasizes the fact that survival expectancy is still poor for this category of patients treated with SoC.
I hope my transcript is a faithful reflection of what has been discussed, overall, do your own DD, make your own decisions.
Fosco
Hey rogue
I guess you have answered to the wrong post, haven ‘t you ? At which point was I talking about spreadsheets ?
All I said which could be related was : “All statistical models taking into account current observed survival in the PhIII clinical trial (less than 298 events so far June 2019) show that Multikine improves survival more than 10%”
So can your model prove me wrong with given SOc and Dropouts ? May be it will give a probability of success ... well, indeed all depends on things happening like they should be likely to happen if they don t then we are screwed by the lack of luck.
This being said, i found your method as you described it extremely interesting ... and am eager to get the end result when you run it many times with various assumptions ... Could we have some surprises or just confirmations ? Now I do have a big question : what does determine “success” of the study in your method ? The fact that there are 10% more survivors in mk group at endpoint ? Well that s not success, that would just mean that at some point in time there is more survivors in mk group. I believe that Success is really determined by comparing two survival curves and both are considered different by log rank test whereas your method measures success by comparing soc vs 298-soc. That s why the basic and a bit trivial method of 10% improvement in a mk curve does insure success when endpoint is reached. Food for thoughts
Survival models : focus on dropout rates
=========================================
Many of you have expressed interest regarding survival models for Cel SCI.
-How is Standard of Care Overall survival ?
-How much % could be dropouts ?
About SOC survival :
--------------------------------
We have expressed the firm belief that Oscc observed in most of the sites of the study _ from which 62 out of 101 are in developing countries_ are mostly not HPV related and mostly affecting a population which has low socioeconomic standards and poor ability to pay for insurances. (HPV related oscc have a much better cure rate)
We have seen from recent studies how disastrous survival was in Bulgaria/ Thaïland / Mainland China : heavy smokers and drinkers, likely little HPV prevalence, less private insurances.
We do therefore believe that observed OS in the clinical trial treated by SOC only is far below the observed survival in North American statistics (SEER Db) which is biaised by HPV occurrences and closer from what Cel SCI states as an unchanged survival of near 50% OS @ Y5 in absolute numbers
Regarding Dropouts rates :
--------------------------
We explained in the following thread why we believed it is low :
https://finance.yahoo.com/quote/CVM/community?p=CVM&messageId=9808e8bc-d35c-49de-9f48-1139bb3551fe&replyId=b49ce6bb-0461-4ec7-bb5a-d3cbcb45416f&bcmt=1
We can also use this Taïwanese study as a benchmark :
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5384671/
Excerpt :
“There were 24,335 newly diagnosed OCSCC patients from 2004 to 2010 in Taiwan. Carcinoma in situ was found in 1,864 patients (7.7%) whereas 469 patients (1.9%) had multiple primary cancers or distant metastasis. In addition, 362 patients (1.5%) who did not receive any treatment during the follow-up period and 397 patients (1.6%) who had incomplete records were excluded. Consequently, 21,263 patients (87.3%) were included in our final analysis.”
What this does mean : once screening is done (inclusion criteria applied in the case of cell sci) they were left with 22002 elligible patients (24 335 less 1 864 less 469 patients). On those eligible patients 362 (1,6%) did not receive SOC treatment 397 (1,8%) had incomplete records.
So we can consider that Dropouts rate in this example of SOC for OCSCC treatment for 22000 patients was of 3,4%
In conclusion :
--------------
We are in line with our models.
- With 5Y overall survival for SoC close from 50% (lets take 53% as a conservative assumption)
- With dropouts rates likely low (lets take 5% as a conservative assumption)
All statistical models taking into account current observed survival in the PhIII clinical trial (less than 298 events so far June 2019) show that Multikine improves survival more than 10%
Kpauliukonis
While your post has passed almost unnoticed, it should be of utmost importance for those who should think about investing in Cel SCI
I don't know if your questions was inspired from my original ideas regarding HPV prevalence and developing countries SOC
(here : Fosco View on trial Population)
but the answer of your ENT goes exactly where my conclusions go as well
This is all good for the trial outcome
I am not expecting a five years SoC observed OS above 55% in the best case in the blend of patients of this pIII trial
Hi Static
Another interesting study, around 3000 patients in 50 years
To be able to use as benchmark to CVM study, the first criteria to consider would be, in my opinion, the cancer stage, the second the dates : the sooner the better as they introduced more systematic S-CRT
So the interesting table is table 3.
Stage shall be "regional"
So here's how it went :
1960-1969 (%) 1970-1979 (%) 1980-1989 (%) 1990-1999 (%) 2000-2009 (%)
Regional 28.9 32.6 27.8 38.5 31.2
As you can notice soc is disastrous, never above 40% at Y5, most time around 30% and best years were 90/99 with 38.5
Well, while SOC OS might have improved in the 2011/2016 period, it is still probably as bad as possible and below 50%. This kind of bad prognosis should balance countries were OS is better in our study.
Room for optimism for Cel SCI PIII
Static
What I thought but a confirmation is a good news.
They count all deaths, hence disease related events are even lower and survival better
It does help thanks a lot !
Hi Static
Question for Gavin : in the 298 count is it relative survival (survival related to the disease) or just followed deaths. Eg if someone dies from car accident is he counted or discarted (eg considered as droppedout) ?
Yep
Uk model is 10% by now
Fosco
welcome Sab
No, that's the amount not exercised (exercisable) to date
not necessarily 10,x, but probably not 30 or even 40% or more !
for instance in a 19% MK benefit scenario, back in Feb when IDMC met, we could have had 157 events in SOC and 114 in MK arm, 271 events in total (And 27 remaining). If all the 27 would fall in MK arm, then benefit would fall to around 7%
Jmho
Cash burn and warrants
======================
I updated the warrants sheet
Conservatively 1,3 per month with a bit more of $ 11M cash on hand beginning of May that's 8 months from now so early January like Sushi said
Let's not forget as of 05 / 13 approx $ 12 M can still be raised from In The Money warrants (exercice price equal or below 5$).
Interestingly $ 6M has been raised between 1st and 15th of April , may-be within the low strike price Series WW, RR, VV, MM. That corresponds to the first rush up of pps.
Then between 15 April and 13 May $ 1,3M "only" has been raised from warrants. Enough to compensate cash burn though.
https://docs.google.com/spreadsheets/d/1ulpVesaigAtoU_rXoZ8t-A3sonbKHs6T8YV95lGMYBw/edit#gid=369549476
Hi Robot
Nice research
Indeed Asia is a dream continent for a working cancer therapy
This said I am surprised by so much HPV-related HNSCC cancer prevalence in India. Thought is was more a western world specificity (HPV related HNSCC cancers show a better survival)
Take care
Fosco
That's my reading as well,
Cel SCI has little interest now as for pps moving up till readout
Fundings is secured
That's an open message for his followers who have endured those terrible years with him, lost faith, lost a lot of money, regained faith and are back with him. Only fools and shorts can't understand this.
So be it.
Fosco
@george
4 or 5.
5 is even more conservative as it includes some drop-outs, but we should be good to both now
But what do I know ? I am just a bean counter....
Fosco
Yes very true.
While success/failure of this clinical trial will be assessed using statistics in deaths counts in both arms ("Primary outcome measure")
QoL is indeed a secondary outcome measure : For the patients QoL will NOT be only "interesting", having a better QoL due to MK is a real benefit and also a motivation to be treated longer, to live longer that should be reflected in primary outcome measure. But statistics also will be used to measure QoL improvements in both arms (patients had to fill a questionnaire that will be processed/compared using statistical methods)
@Sab
Thanks
I would love though that CVM could give a hint (eventually indirectly or in a hidden fashion) on current count of events : are we closer from 298 or from 280 ? That's something they know for sure and something that would enrich current models and in a very positive way by giving guidance for the investment community (if good, and I do consider the figures found last year from Taïwan as excellent). People just don't have a clue when it's gonna end and if they learn it'll happen in H2 they'll certainly be much more likely to invest and less warry with their wallet than they are currently. Of course time will tell what we don't know yet in any case.
Static
I agree with your statements and analysis.
"Such a process and certification and would no doubt take over a year to do"
Time at this point in time, won't be the problem, in the sense that it will play with the investor, not against.
I am thinking that they will need more cash on hand and they will have the choice between a big pharma buy out (Which might not propose as much as Geert would like if the Pharmas wants to bargain) or go it alone with a huge facility yet to build, like a huge Coca Cola factory. This will require funds and I understand why Geert is promoting right now his stuff. He wants funds, with the warrants alone they can get 60M$, more than enough to start with a strong product and a strong cash position to start a big bargain or a big facility construction
BTW if successful, our small Riva boats will be dwarfted by Cotton huge Yacht. I hope that we will get free entry to the cruiser.