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.
You are correct. Entire population that is under 70 years old is included. When I say entire population, I mean that it is a wide range, and not including "outliers" is part of statistical analysis, and should not be included in the ENTIRE POPULATION.
Take your 3 70 year old patients. The efficacy on "older" patients is not going to show efficacy as much, if those "older" people are dying of "old" age, or "other" common illnesses since they are more susceptible to other illnesses at that age.
Therefore, using your 30% argument just shows how much YOU DONT UNDERSTAND STATISTICS. Read up on Outliers!
Exactly, placebo and vaccinated group are randomized.
AVII Not argueing about steroid efficacy. IT DOESNT MATTER! (and why you really don't understand statistics).
The trial is comparing Placebo VS. Vaccine. BOTH HAVE STEROID USE REMOVED, SO BOTH ARE EFFECTED THE SAME! ALL THAT MATTERS IS THE DIFFERENCE BETWEEN THESE TWO POPULATIONS. Its called Statistical Analysis of Populations.
Your concerns about not using steroid use, or biopsy or pseudoprogression patients, it doesn't matter. Because they are removed from the Placebo population as well. ALL THAT MATTERS IS THE DIFFERENCE BETWEEN THESE TWO POPULATIONS.
When one says the entire population, it is meant age, gender, etc... Not specific subsets.
NO MATTER IF STERIOD OR BIOPSY INCLUDED OR NOT, MEDIAN PFS VALUES OF 15 MONTHS HAS NEVER BEEN SEEN BEFORE IN ANY GBM TRIAL!!!!
1) Okay now I understand what you mean about biopsy. However you couldn't include them. The reason they can not treat biopsy patients is that they do not have enough tumor tissue to create the vaccine. So they need to be removed. I am sure that would be the case with any dendric cell immunotherapy trial.
2) The removal of steroid use, even if it is negative effect, and should slightly increase the OS of the entire treated population, doesn't matter. The reason being is that, BOTH THE PLACEBO AND VACCINATED GROUPS HAVE STEROID REMOVAL! Therefore both of those populations will see the same slightly increased OS effect.
That is why you have a placebo group. The trial can only show efficacy if there is a difference in the two groups.
I will say, none of those subset groups you show, (except maybe the MGMT group), show a median PFS of 15 months. That is unheard of, and we know from the information provided since the screening, that there is an extremely high probability > 50% of the trial will show a PFS of 15 months or greater.
I actually don't understand why you are concerned, (other than the fact that the price is 0.16 a share). No doubt I do not like that the share price fell. No doubt I would like more information from them.
However, we can surmise all we want about what if they gave us more information, but one could also say that with more information, the "shorts" would have just had more information and fuel to twist and turn negatively. The proof will be in the results. On a side note, I wouldn't doubt that certain Big Pharma players would have had a turn, or supported the large shorting that took place to get the price to fall. It is well documented that small biopharma companies with no products and heresy trials are easily manipulated, and it looks like the same happened here. I agree NWBO management's subdued communication fueled the fears that helped short the stock.
However, the fact that it is taking considerably longer for the events to occur, actually plays in their favor. I have crunched the data since the screening halt in Aug 2015. Over 30% of the trial patients will have shown a PFS > 15 months. That is unheard of even for Optune. And its most likely ~ 50% > 15 months PFS.
So what you are saying actually strengthens DCVax-L Trial. By allowing patients to have steroids, how do we not know that the steroids are effecting the results. Not sure what you mean by a biopsy. All patients have their tumor cells removed, not sure what a biopsy would do, if they are looking at different cells other than the tumor.
The DCVax-L vaccine is manufactured using the removed tumor cells.
Those are the same protocals that Stupp used, IMUC P2 & P3 (except the P3 is actually looking at a subset with specific antigens), and Optune.
So NO, those aren't subsets, THOSE ARE OUTLAYERS THAT WOULD EFFECT THE DATA RESULTS, BUT WOULDNT MATTER IF THERE WAS EFFICACY OR NOT!
The reason those factors are outlayers, is that they make up a small portion of the entire population. When one says Entire Population, it is meant versus subsets, like what IMUC's P3 is looking at. That is a typical subset of a population. Or if you looked just at <50 yrs old, or > 50 yrs. old.
If you do not understand statistics, then you wouldn't understand the terminology and how it is being used. You and AVII obviously do not understand STATISTICS.
Is there different treatment protocol between recurrent and non-recurrent GBMs? The fact that they are named differently, and are treated differently, have different OS curves, etc.... means that recurrent GBM is NOT a subset of non-recurrent GBM.
The trial uses the entire non-recurrent GBM population. You are just trying to poke holes in an argument that you lost long ago!
OMG! Yes, they excluded recurrent GBM and early progressors, AS DID DCVAX-L. As did Stupp et. al., as did IMUC, as did every other GBM trial out there.
Optune is showing 13% living 5 years or more. And that is ENTIRE population! Are DCVax-L P3 is already showing better PFS results than Optune!
It is being tested with non recurrent GBM patients and ones that don't have "other" diseases. How are they treating a subset? Please define SPECIFICALLY what subset? Non-recurrent GBM subset????
These graphs are all highly selective, just as you stated. So for Dr. Liau's statement that over 25% can achieve 5 years or greater, she is speaking about the entire population, not a selective group. When you look at all the data out there on overall data, (Optune P3 trial, IMCU's trial, Termozolomide P3 trial; Stupp et. al.) , nothing shows anything close to 25% at 5 years. Moreso, the majority of GBM patients are > 50 years old, so to use that graph and say you don't hold much weight in what Liau says shows your comment doesn't hold much weight in the GBM conversation.
Blinded means that they dont know who has placebo and who has vaccine. However, they would know the total population, so if the median value PFS of the total population is ~ 15 months, then they know that there is a high probability that the PFS increase will be met.
300 patients were enrolled in Aug 2015 - Stated in a news release. Using Jan 1, 2017, That means 52 patients were PFS free for ~ 17 months.
31 were enrolled/randomized between end of Aug 2015 and Nov 2015. Using the October 2015 you get over 15 months, using Nov 1, 2015, you get over 14 months. Yes some of the patients would be at 14 months, and others at 16 months. The shorts will try and argue that all 31 patients were enrolled in Nov. 2015.
The fact of the matter is, 83 patients or (25% of the ITT) is over 14 months. And again, that assuming all 83 were randomized in Nov 2015, the fact of the matter is many of these 83 patients will be 14 - 36 months or higher, and there will be some amount of PFS > 14 months from the 248 that were enrolled throughout 2013-2015. The fact that there is significant number we know to be > = 14 months, shows that there is a very high probability that there is significant efficacy in PFS.
I quoted Optune's but I also quoted ICT-107's results, and Stupp's results. There is not one study that shows mPFS > 10 months.
In Aug 2015, it was stated that at least 300 patients were enrolled. Jan/Feb 2017 was when the announcement was made. That means that between 83 - 31 = 52 patients had ~ 17 - 18 months PFS free, and 83 patients were 14 - 16; (not all the 31 patients were enrolled/treated in Nov 2015, many of that 31 would have been spread out between end of Aug 2015 and Nov 2015.
Even 10% LTFO would not drastically change the results. you have to remember, there will be a significant amount of patients in the 248 that were treated between 2013 - 2015 that will also show a PFS of > 12 - 15 months.
The fact of the matter is, if there was no efficacy, then there should have been only 5 patients of the 31 treated after 15 months. Also, inversely, 305 patients would have had to been treated between July and Aug 2015, in order for ~ 50 patients to show 15 months PFS free, using SOC statistics. So maybe July and Aug was just one busy month for Cognate. Maybe all 300 patients were treated in those two months. Maybe, but logic tells me that wasn't the case. And even then, the math wouldn't add up, as there would only be ~55 - 60 patients PFS free at month 15.
Also, I believe there are stringent SEC rules and regulations about how long company news may be held before announcement, and it is not more than a month, so to say the 248 events occurred in Nov/Dec 2016 is a little absurd. But if you want to keep justifying why the math shouldn't be right, keep pulling at hairs. But your comments have little weight, mathematically that is.
Math does not lie; I am still compiling everything but here is some numbers:
25% of all 331 patients have PFS >= 15 months. That is unheard of from any trial. AND THAT DOESN'T INCLUDE THE DATA FROM THE OTHER 248 patients, surely some of those 248 patients will have also had PFS > 15 months as well even if you use SOC 17% of patients that are Progression free at 15 months, that's another 42 patients, making the 1/3 of the trial showing a PFS of ~ 15 months or greater.
All values below are from randomization so they can be compared.
(using avg 83 days from surgery to randomization, ICT-107 P2 trial)
Optune P3 Trial
mPFS control = 4.0 months
mPFS treated = 6.7 months
Stupp et. al
mPFS = 9.67
ICT-107 P2
mPFS control = 9 months
mPFS treated = 11.2 months
Graphically, there would have to be a drastic change in the slope at the 50% value of the PFS graph for the graph to be able to show a difference of 6 months between 50% and 35% of the population. In laymen terms, that would mean that on average 18 patients a month would have tumor progression until month 9, then, all of a sudden, only 20% of the patients would see tumor progression between 9 and 15 months, so now, it would drastically fall to an average of 11 patients per month.
Now as we know from the past, the PFS and OS GBM graphs are exponentially decreasing, therefore there are more patients eventing at the beginning, and this slows down as time increases.
Therefore, mathematically, it makes sense that the PFS median value should be significantly > than 9 months, and up to and possibly higher, than 15 months, as there is a good chance (mainly because we know atleast 83 patients had PFS >= 15 months, and ~20% (83-31 = 52 patients) were PFS free up to 18 months.
Its not open label at crossover BECAUSE... if a patient has tumor progression, then they are given vaccine no mater what, so if that is what you consider open label, then I guess, but the information about what patient was given placebo and vaccine initially, that data is not known nor given, and is only known by the third party that is adminstering this blinded data. And that is never known until after data lock. So NWBO will know how many crossover, and they could know what the overall OS is, but they wont know what the difference is between the two groups, because it is double blinded.
No, the cross-over means that when a placebo patient's tumor progresses, labeled a primary event, then that patient then is given the vaccine. Since it is double-blind, neither the researcher nor the patients know what they are being given. Therefore it is not open label at all.
Dang, I was hoping that this was some information from the inside somewhere that made its way out. The distribution that I was imagining would be similar, in that the 98 would be distributed, with a concentration around the times when the most patients were treated; which is 2014 - 2015 I would imagine.
But this does provide a good representation of what I am calculating, in that this doesn't include patients that lived ~ >20 months but unfortunately passed away, as this would increase the amount > 20 months OS even more.
PS THIS HELPS HUGE! THAT GRAPH LITERALLY PROVES MY CALCULATIONS!!!
So does that mean that we have reached the 233 events? Or are these a graph of just the longest OS 98 ?
If this is the actual distribution then it is exactly what I was saying, that the 98 alive would see the majority of their OS number well > than 24 months. This shows that there is significant amount > 36 months; approx. ~ 98 - 33 = 65 patients which is ~20% of the total 331, and then add to that the number of the 33 patients that are < 35/36 months that will also live past 36 months. If you look at the tails, once you survive into the tail, you have a good probability of living longer. So that could mean as many as 3/4 of the 33 will hit the 36 month mark as well, and that would give: 65 + 25 = 90 patients, which is ~30% (actually 27%) of the entire population living past 36 months.
And that doesn't include any of the other 233 patients.
Now here is the kicker, these numbers are all well above SOC, and so its most likely not the 110 living this long. Therefore, assuming the majority of these were non-placebo patients, nearly 50% of them are over 3 years!
Where did you get this?
All patients vaccinated from Aug 2015 were randomized. Meaning all those patients were randomized and treated at t = 0 sometime between Aug 2015 and early Nov 2015.
Now, IF, all 331 patients were randomized at the same time, THEN YES, the 98 patients that we know are living > 20 months, would be the same as the 30%.
This should explain it. SOC treatment shows the following:
30% live 20 months or more
24% live 24 months or more
~ 15% Live 36 months or more.
We know that the 15% that live 36 months or more are half of the patients from the 30% that live 20 months or more.
NOW, we know that the 98 patients (~30% of the 331) all have lived 20 months or more, HOWEVER, common sense tells us that SINCE THESE 98 PATIENTS WERE NOT ALL TREATED AT THE EXACT SAME TIME, then, the majority (AND WE KNOW THAT ~31 patients weren't enrolled Aug 2015) of those patients are > 20 months.
So that means that the majority of these patients will be distributed > 20 months.
SECONDLY, DO YOU HONESTLY THINK THAT ONLY THE PATIENTS THAT WERE TREATED IN THE SUMMER OF 2015 ARE THE ONES THAT LIVED 20 MONTHS OR MORE?????
In order to match SOC percentages, 0 of the 233 patients would show an OS > 20 months. Use some common sense!
Going to try and do most of it this weekend and next week while on holiday.
Percentages can be applied to subsets; YES here is the math:
331 x 0.30 = 99.3
110 x 0.30 = 33
221 x 0.30 = 66.3 Add those together and we get 99.3 Same number.
So listen to the justification. Take only the 98 patients that we have some information about. We know that with the screening halt, these 98 patients will have 20 months OS or higher.
Now take the other 233 patients. If SOC normally shows that 30% live 20 months or more, then of the 233 patients; ~70 patients should live 20 months or more (Additionally, we should see ~ 99.3 patients of all the 331 in the trial living, and we already know that 98 have an OS of 20 months or more).
NOW, HERE IS WHY THE 98 ISN'T THE 30% OF THE ENTIRE 331; BECAUSE ALL 331 PATIENTS WERE RANDOMIZED, AND WERE NOT ALL TREATED AT THE SAME TIME! That is why of the 98 patients, many of these will be well over 20 months, most likely greater than 2/3 of them.
A perfect example; Stupp showed 30% OS of 20 months (or ~26% of 24 months). Using the 98 (331-233 events that have not yet happened), gives us ~ 30% at ~20 months. Now this doesn't include the 233 which the majority of were randomized well prior to Aug - Oct 2015. So, using the 30% of Stupp, we should also see ~ 70 patients of the 233 live up to 20 months as well.
Add those to the 98 WE KNOW WILL BE AT 20 MONTH AT THE END OF APRIL, and we get 168 patients of this trial living 20 months or longer, which is 50%. Using math, we can estimate that there will be ~ 50% or more from the 331 patients, with an OS of 20 months or more.
Nowhere on the Dec 8 Press Release Link (see below) does it say they do not know the number of events of PFS events. Actually it does state:
"The Company announced that 331 of the planned 348 patients have been enrolled in the Trial, and that data “events” have been accumulating towards the endpoints of the Trial"
https://www.nwbio.com/nw-bio-provides-update-about-phase-3-trial-of-dcvax-l-for-newly-diagnosed-glioblastoma-multiforme-brain-cancer/
Meaning at DEC 8th, 2016, THEY STILL HADNT REACHED 248 EVENTS. THAT MEANS THAT WE KNOW THAT ON DEC 8TH, 2016, MORE THAN 83 PATIENTS WERE ALIVE, AND DID NOT SHOW ANY TUMOR PROGRESSION!!!
Okay so reread this very carefully. Use mathematics, solve for X, and it will become apparent why we know that 25% of the total 331 patients have a better PFS result than what is seen for the last 25% of the Optune and NUMEROUS SOC trial arms.
248 events: An event is tumor progression, or death before tumor progression. So that means they may not known how many tumor progressions there were in the 248, but they do know how many there weren't in the 248 (83).
Funny, because Dr. Liau showed on her presentation that 60 placebo patient "crossed over". THAT MEANS 60 placebo patients had a PFS event. They knew that on Dec. 15th, 2016.
Also, in the 8-K in Aug 2015, NWBO states over 300 patients were enrolled. Meaning only 31 patients weren't enrolled yet, and that would have happened between Sept 2015 - Nov 2015 for those patients. SO WE ALSO KNOW THAT OF THE 83 PATIENTS THAT HAD NO TUMOR PROGRESSION, 83 - 31 = 52 OF THOSE HAD A PFS VALUE > 18 MONTHS. Again, and that doesn't include any patients in the 248 that could have had an event, but still had a PFS > 18 months. SOC shows ~13%, so add another 32 patients at minimum, and the value jumps to 25% minimum show a PFS > 18 months.
Wrong - Dr. Liau stated that they hadn't reached the Primary endpoint yet, which is 248 PFS events. That was Dec. 15th, 2016.
They then stated in early Feb, 2017 that over 248 PFS events had happened. Therefore 83 patients at minimum were still PFS free in Dec. 2015. EVEN YOU TAKE 2 LESS MONTHS AND SAY OCTOBER 2015, THAT'S 13 MONTHS WHERE 83 PATIENTS HAD A PFS > 13 months.
The screening halt provides us with a Mathematical evaluation of 83 (and 98 for OS) events that had not occurred.
Has nothing to do with 10% being lost, the 10% that is lost within months of enrollment as shown on SOC, would be included in the 248. So ~31. ALso, I believe that the 10%, or 31, is part of the 50 Placebo patients that did not cross over to the vaccine, meaning, only there is less than ~19 patients of the 83 that could be SOC. That makes sense because ~ 15% of patients on SOC are PFS free at ~15 months, which would be 0.15 * 110 = ~ 17 patients which fits the 19 patients of the 50 not yet having a PFS event.
Actually, it is a fact.
1) We know when the screening halt happened.
2) We know when the 248 endpoints occurred.
Even if you take another 2 months off the 15 months, and say 83 patients had PFS > 13 months, that is WAY more than SOC would see for 13 months PFS.
Again, IT IS A FACT!
FACT: NO PATIENT WAS SCREENED AFTER SEPT 2015, AND THEREFORE NO PATIENT WAS ENROLLED IN DECEMBER 2015.
FACT: DECMEBER 15th, 2016 PRESENTATION; 248 EVENTS HAD NOT OCCURED YET!
Do the math! AND, those 83 patients are going to have PFS times between 15 months - > 24 months. BECAUSE THEY ALL WERENT ENROLLED AT THE SAME TIME!
If all the patients were enrolled at the same time, all 331 patients, then you could say, those 83 patients are the tail. BUT THEY ARE NOT. Because of the staggered enrollment over 2 - 3 years, there is going to be patients within the 248 that WILL ALSO HAVE PFS OF > or = to 15 months. STANDARD OF CARE SAYS ~15% OF PATIENTS REACH 15 MONTH PFS, WE ARE ALREADY AT 25% WITH THE 83!
That's why it insane to think no one has done the math on this!! That's the great thing about MATH, it doesn't lie!!
FACT: DCVax-L Trial Tail is Fatter than SOC & Optune
Since we know 83 patients are > 15.3 months & 31 patients are > 18 months PFS, and that does not include the results of the other 248, which will should have some amount > 15% of the 248 which is another 37 patients, so 120 patients. That's 36% of the 331 will have a PFS > = 15 months. What I don't understand is how and why investors/statisticians, haven't identified this yet. Compare that to the Optune SOC results or any other SOC trial results, and its an increase of 20% (from 15% to 35%), and compared to Optune it's an increase of 10% (from 25% to 35%). That means the DCVax-L tail is as thick as Optune's tail using just the 83 patients we know have a PFS of atleast 15 months.
The only way that the the DCVax-L trial would have less than ~35% of its overall patients with a PFS > 15 months, is if the rest of the 248 patients respond worse than what Standard of Care currently sees.
From randomization. On their graph they also state from diagnosis which is:
Median Optune PFS 6.7 months (11.2 months from diagnosis)
Median SOC PFS 4.0 months (7.8 months from diagnosis)
https://www.optune.com/hcp/newly-diagnosed-glioblastoma/efficacy
Yeah, I have 83 days on average from several different links/studies.
https://virtualtrials.com/trial_results_-_glioblastoma_treatments.cfm
So if mid Aug 2015, then that would be beginning of Nov. 2015.
However, the more I read up on the process, and FDA holds/partial holds, the more I feel that new screening was considered at the Leukapheresis stage, not Surgery. PLUS, the FDA would have called NWBO on a teleconference prior to any news release, and that would have meant that the actual hold would have been done closer to the beginning of August. Meaning that even if taken from surgery, the last patients would have been late Oct. 2015.
However, it doesn't matter if its late Oct or late Nov. What I am calculating is a mathematical fact, stating that 83 patients had a PFS of > 15 months. However, in truth, the overwhelming majority of those 83 patients would actually have a PFS value > 15 months, more like 17 or 20 months, because enrollment was being done continually, and not all at the same time.
What I will use now is the fact that 306 patients were enrolled Aug 16, 2015, and 25 patients between then, and beginning of Nov. were then enrolled.
I can then show data points on any PFS SOC graph to show that ATLEAST 25% of the 331 P3 trial patients show PFS of 15.3 months and ~18% of the 331 show a PFS of 17.8 months. AND THAT IS NOT INCLUDING ANY OF THE OTHER 248 PATIENTS. If you included them using SOC statistics, then this would increase to:
36% w/ PFS 15.3 months
34% w/ PFS 17.8 months
Optune showed
25% w/ PFS 15 months
20% w/ PFS 18 months
So if we assumed all the other 248 patients had PFS < 15 months, then the entire 331 patients in the P3 trial did better the Optune results, because we know from the screening halt that 83 patients had a PFS ~ 15.3 months (at least; conservatively).
Ok, now we are getting somewhere.
In this filing it states 300 patients at minimum were enrolled Aug 16, 2015. That means 31 could see randomization between Sept 2015 to early Nov 2015.
https://www.sec.gov/Archives/edgar/data/1072379/000114420415051646/v418946_8k.htm
That means that 52 patients were progression free for ~18 months (17.83) and 31 patients were progression free for ~ 15.3 - ~17 months (taking a Nov. 2015 randomization start time). I can therefore use 15.3 months for the 31 patients. Therefore, adding them all up, we know that there will be at least ~ 40% of the 331 patients (assuming only 18% of the 248 reach a PFS of 15 months).
Using Optune's trial information, and numerous other GBM SOC trial data, if there was NO efficacy at all with DCVax-L, there should only be ~ 60 patients in total reaching progression free survival of 15 months. We know for sure that there is atleast 83 of them, not including any from the other 248 patients.
Where is the link showing the 306 were enrolled?
It specifically states randomization occurs at the first vaccination for the NWBO phase 3 trial.
It still says Pre-Screening occurs at surgery. And the 3 month from surgery timeline (actually 83 days on average) is from numerous other trials, but even if we take 90 days, that's beginning of Nov 2015.
Also, 248 patients did not progress in Nov 2016, as Dr. Liau presented Dec 15th, 2016 and stated in the presentation that they were still waiting for that endpoint. It didn't occur until late January, early February 2017.
Moreover, only a small portion of the 83 patients would have started vaccines in Oct/Nov with the majority of those patients being > 15 months PFS. This is easily correlated by the exponential tail, in that once patients are PFS free after ~15 months, only a small portion of them see their tumors progress over time.
All I am asking for is the facts to provide me a drop dead start date, either its Sept, Oct, or Nov, 2015. That way my calculations are bullet proof. Not trying to say its one way or another, but it would have to go without saying that the beginning of Oct 2015 would be extremely conservative starting point, which would still provide a PFS over 15 months for > 83 patients, and ~19 months for OS for >98 patients.
I did, The PR you are speaking of on Aug 16, 2015 states:
Over 300 patients have been recruited for the trial. The total anticipated enrollment is 348 patients.
https://www.nwbio.com/nw-bio-confirms-phase-iii-trial-of-dcvax-l-for-gbm-brain-cancer-is-ongoing/
It was The Street that stated "just over 300", no NWBO press release stated that. And it
states that the anticipated enrollment is 348 as they submit certain information. That means they were still in the 30 days that the FDA placed the hold. See FDA Clinical Hold guidelines below:
https://www.fda.gov/RegulatoryInformation/Guidances/ucm127537.htm
EXACTLY! Screening begins at the Pre-L phase not at surgery. It also states final screening tests occur during the baseline visit. Those are 4 - 6 weeks apart.
Your quote below:
"During the evaluation phase you will have some testing noted below to indicate whether you can participate."
Testing doesn't begin until the pre-L visit, and testing is not concluded until the baseline visit. Therefore once you pass the screening, you are enrolled.
So now we need to find out when does the FDA enforce the screening halt. Meaning, do they allow patients that have not completed the screening phase to continue. Would love to see that link.