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.
FeMike - Good points. In the data you showed, look at the confidence intervals (CI) - they almost totally overlap. This indicates PFS was nearly identical, not better in the placebo group as AF has claimed.
That is a total misunderstanding/mistatement of a tiny difference on PFS (they were statistically identical), which is NOT the primary outcome anymore. Just a way to twist the results negatively.
Buried in the NYAS communications was the mention that the talk today could not be posted for on demand download because they have a publication under review. This is the first confirmation that a paper has been submitted. So, hopefully an accepted publication will be announced in the fairly near future. The reception at ASCO in a month will be very interesting (I would think oncologists would be very impressed by the doubling in rates of extended survival in this deadly disease).
Totally consistent with prior projections. All good news.
The difference is that CVM does not have approvable results, whereas DCVAX results totally fit FDA approval criteria for a deadly orphan condition.
It means tumor recurrence time was exactly the same with DCVAX or without, but with the caveat that "recurrence" in the DCVAX group may have simply been the immune response to treatment rather than actual recurrence. If it was a tumor recurring, DCVAX made twice as many people live 5 years.
Right on.
mOS does not capture the long survival tail. It is a totally different thing (i.e., mOS = 50% survived this long, rather than 15% survived 5 years).
Nice confirmation of manuscript submitted at least.
They could have just pulled her talk entirely. The fact that they bothered to get an alternate speaker (collaborator) on the same topic underscores the importance of the talk.
Good point - Thanks.
Ok - Got it. With the recent FDA comments clearly in favor of use of ECAs in some settings (and a deadly orphan condition with inadequate treatments like GBM seems to qualify), significant benefits relative to ECAs should be sufficient for approval. Chances that the ECA comparison will not be significant are virtually nil based on publicly available data (as I am sure you know).
Paranoia running rampant - Do we really know that all of the NWBO and flaskworks people who are supposed to be attending the talk tomorrow are actually signed up? Given the Elon Musk listing, seems like someone with money could do a lot of fake registrations under any name they like to raise expectations. Has Brad Silver posted anything himself about this? I don't really think this paranoia is justified given the description by the NYAS of LL's presentation as "exciting," but all the negative posts are making me question my rational assumptions.
Define mediocrity. I will consider a highly significant OS improvement vs. ECAs a true success regardless of the other findings. I do not have a strong sense for how the market will take that though. Could result in little change to the SP (everyone who has seen the info available already expects this) or a nice increase if viewed as unexpected by some or seen as a real success. We shall see.
TDD -
Regarding "...and let LL pass on bad data": LL is a scientist and will just present the data, good or bad. Given what we know, I expect results to be: 1) significant OS improvements relative to ECAs, 2) technically a failure on the original endpoints, but with compelling reasons provided for why they changed the endpoints. This is a conference on challenges to these types of trials after all, and no one knows better than LL.
Given how many claims of the negative impact of shorts on NWBO stock the company has made over the years, maybe they are intentionally keeping things vague to keep shorts guessing and limit their impact.
Not 100% sure but I think I recall hearing of a few such cases in the context of somewhat equivocal positive results from the P3 trial for severe conditions.
There is a middle ground - I think the FDA has the option of approving DCVAX for commercial use based on existing data, with the requirement of a confirmatory trial. The fact that GBM is an orphan disease helps.
A successfully treated DCVAX patient
Think about it - If LL is talking about TLD, then the company really has to do an official PR before that (and everyone could be in their pajamas for that). Why would most of the company, Brad Silver, etc even attend LL's talk virtually if it was about "background" on ECAs in the Phase 3 trial? I guarantee you they have not attended all of LL's talks where she has discussed these issues previously, and what would be "exciting" (as NYAS said) about that?
PS - Love your name. Best lawyer ever.
FeMike - I get your argument. However, I find it difficult to come up with any topic LL could talk about that is Phase 3 trial specific and "exciting." She has given numerous talks in the past year about external controls, etc., none of which were PR'd. I have been pretty pessimistic about the company's long delay in announcing TLD, and am wary of a rug pull (again). That said, this situation really seems different, and my best prediction would be a formal company TLD announcement on Tuesday morning with reference to LL's talk shortly thereafter, with the talk addressing the findings and the complexities of the trial. I think we could see a publication soon after as well. It would make little sense given the long delay in announcing TLD to do a PR of this talk presumably discussing TLD unless they knew that additional detail would soon be available in print to address the complexities not covered by a simple TLD announcement that will get the AF's of the world writing critical hit-pieces.
The NWBO website lists all of their prior press releases. They have never sent out a company PR about a talk by LL previously (although LL obviously has given many over the years). All prior PRs refer to talks by company personnel (LP, MB), not academics. I think LL's Tuesday talk is really about the TLD. Makes little sense to do a PR for what NYAS calls an "exciting update" unless this were the case.
I started wondering that myself but Annals of NY Academy of Sciences is more of a basic science than a clinical care journal. I would never say it is impossible though.
NY Academy of Sciences is one of the most prestigious academies in the US. They have their own very well respected journal.
Interesting - This PR notice was provided by NWBO. Do they normally promote LL's talks?
When I said "“The recent info about the Advisory Panel may help explain the delay,“ I was referring to Stupp's conflict of interest forms that indicated he participated in a one time advisory panel for NWBO after the data were unblinded. Journals work entirely separately from the FDA, so it is very unlikely that any pending FDA decisions would directly influence publication timing. I think it is at least possible the company itself decided to wait to publish until FDA guidance on external controls became more clear, but no way to know this.
PM -
Does a journal restrict a company from releasing TLD prior to publication? No, as long as it is a brief description.
If the answer is no despite what the company has put out, then why withhold TLD? My working assumption is that the advisors felt the TLD results could not be adequately appreciated without more context than could be provided in a normal brief TLD release. I think they felt that releasing TLD simultaneously with a publication that provided a high level of detail would help the results be received more favorably. That is not a bad thing.
Is the FDA influenced by a journal publication? By how much would the FDA be influenced? The FDA does not really care about the publication itself, but only the underlying data. The FDA is providing its own "peer review," and in much more detail than an journal reviewer could provide (a reviewer would never see the underlying data that statistics are based on, but the FDA would).
Senti - All good points.
DD - Re: publication timeline, if you get your foot in the door with a request for a revision, you are almost guaranteed to get a publication accepted. A "bad" journal will take 3 months to get you a response to your initial submission. If negative, you submit somewhere else and again at worst wait 3 months for a negative response. Repeat if necessary. If they submitted to a journal like JTM (where blinded data were published), it would have been quickly accepted. So, we should have had a publication by now if they had written it in the 3 months following unblinding.
My guess regarding the delay in publication is they submitted a lot later than we thought and they are aiming high by sending initially to a good journal. The recent info about the Advisory Panel may help explain the delay (I've served on these before). These are usually one or two day events, invitation only ("key opinion leaders"), everyone signs an NDA, and they show the actual data and discuss it and how best to proceed. The consensus must have been that the company needs to do X, Y, and Z before announcing TLD and publishing. The fact that the company took this advice from the experts is not a bad thing. Even though the timing of TLD release by NWBO has been WAY outside the norm, I think it is being driven by the best information they have.
Nope - Did not sell. I think they did not submit a manuscript until MUCH later than we had hoped, whether that is due to reanalysis using new GBM definition, some problem with data that required going back to individual sites (and covid impacts), wanting to wait until FDA formalized guidance on external controls, or whatever. If they submitted along the optimal publication timeline I had suggested, they would have had this published somewhere by now even under a worst case scenario (e.g., sending to 3 different journals). I still think the data will look good whenever they do come out, at least compared to external controls.
I loved the post recently about the guy's wife who saw a psychic saying TLD in March - that is as good a source of information as any!
Gary - Most medical researchers I know (MDs) rely on statisticians to not only do the analyses, but also prepare the tables and figures and write the "Analysis" section of the manuscript. LL would most likely write the Introduction and Discussion, and parts of the Methods, and collaborate with the statisticians on writing the Results section. I would think the contract statisticians would have prepared all necessary tables and figures fairly early on in the process - that is what they are supposed to provide to the company as a first step for evaluating TLD. There may be additional secondary analyses they decided to include for publication, but doing these should not take a long time. The rate limiting factors in getting this publication submitted are: 1) Whether the company wanted to pause publication to reconsider or re-analyze particular issues (unknown, but the new GBM definition might have led to a re-analysis, for example), 2) How quickly LL can write the main manuscript (could be a few weeks if she had some dedicated research time, but could be delayed by travel, clinical, or administrative duties), and 3) How long it takes all co-authors to review the draft manuscript and get LL feedback (hit or miss, but in my experience, the more authors you have the longer this takes). I think we are far enough out after data lock that I see no obvious reason why they could not have submitted a manuscript yet, but who knows where they are in this process. As I have noted before, they may have aimed high initially (e.g., NEJM), and have had to revise and resubmit for publication to a different journal, OR they may simply have taken a very long time to get the manuscript done for a first submission.
PS- To Foxhound's post - I agree it is very unusual to publish simultaneously with announcing TLD. But, this is an unusual situation (major change in SAP) that requires significant explanation to provide context for the TLD. I would not put it past the company to do things differently to try to address this issue.
Doc - These days, many journals (except a few NEJM level journals) publish articles online (accessible via PubMed) as soon as the article has been accepted and copyedited (a process of a few weeks after acceptance in my experience). So, in these circumstances, there is no waiting for publication - it happens online as soon as the article is available, and then comes out in printed form later. I have never heard of authors trying to schedule a specific publication date with a journal, and suspect that would be quite hard due to the mechanics of publishing paper copies of a journal (there is usually a backlog of accepted articles waiting for publication). Doesn't mean it can't happen, but I have never heard of it. Journals like NEJM routinely schedule publication dates (not in response to the authors' wishes though), because they need to arrange all the PR that goes out with their top articles.
Hi Anders -
FWIW - Time to first response to a journal submission depends on the specific journal. Two - three months does happen with some journals, and I have seen as long as 4-5 months in rare cases if they cannot get a reviewer to respond. However, some journals make rapid turnaround a priority. Some I review for ask reviewers to respond within 2 weeks, and start pestering you if you are late. If you fail to respond quickly enough, they will terminate your review assignment and have someone else do it or just go with the reviews they have received. Other journals shoot for one month turnaround on getting reviews back. A guy I work with who published in NEJM a couple of years ago had a fast turnaround at all stages. So, depression is not necessarily warranted yet. I just hope they have already submitted something - I cannot imagine why they wouldn't have so long after data lock.
Senti - I was seeing this a possible reason for why TLD has already taken so long, not something that is new and will delay TLD further. Surely LL would have known that the GBM definition change was coming before it was actually announced, and she could have incorporated it into the paper they (hypothetically at this point) have been writing. Just a guess, and I totally agree that they should not be doing anything new at this point that delays TLD further.
Thanks Dr. B - I do not see the new definition as a bad thing based on what has been posted. I do wonder though whether they will have enough information from published historical control trials to make legitimate comparisons of "real GBM" (new definition) in those trials to "real GBM" in the DCVAX trial in their planned primary analyses. I suppose if prior trials reported percentage of patients with different GBM subtypes, they might be able to weight each trial for "real GBM," but not sure how that would work.
Thanks Anders. Just been lurking with little to add. Waiting for TLD just like everyone.
Viking - I agree. I seem to recall that IDH mutation status was the last piece of data that they stated they were able to obtain (and was probably legitimately delayed by COVID). This has now been available to analyze for quite a while, but maybe (?) reanalyzing results in light of the new GBM definition may be one contributor to how long this process has taken.
Maybe, but seems like a stretch. The results they need for writing the publication will simply be a subset of the comprehensive analyses performed as part of TLD determination. It is hard for me to believe that he has seen the comprehensive TLD but would then say in the interview that they don't have the final data. Would make a lot more sense in that case for him to just refuse to comment on the data since it is not yet public knowledge. He sounded to me like he is still blinded in the interview, which makes sense if a publication draft has still not made it to him and he has not yet seen TLD because it is still internal to the company leadership and the SAB. Obviously, this is all guesswork though and you could be correct.