Replies to post #492187 on NorthWest Biotherapeutics Inc (NWBO)

[sentiment_stocks]

I will present your previous post in quotes, and then respond.

What I think is that those 35 patients were in the initial control arm of the trial (part of the 99) and at some point, after randomization, they were thought to be progressors from SOC, but then it was realized they were late pseudoprogressors from SOC.

If these 35 (or 33) control arm patients were thought to have progressed (mistakenly), they would have had the option to cross over to DCVax-L treatment. If at some point after they’d crossed over their so-called progression had gone away or disappeared, it could have also been considered to be a reaction to the DCVax they were now receiving, and not that they had previously developed psPD from chemo/rad.

If this is the case, because of being initially thought to be progressors, they received DCVax after pseudoprogression. As a consequence of that, these patients were not classified in any of the two trial arms. They were not progressors when they received DCVax, so don't correspond to the rGBM arm and they didn't receive DCVax at the beginning of the trial, so they also were not included in the nGBM arm, but benefited from DCVax.

If a patient was in the control arm, and was considered to have evented, then received DCVax, and following this, the clinicians had decided the patient had actually developed psPD from chemo/radiation, these patients would NOT have been removed from the trial. These same patients would have instead been part of the 64 that crossed over. Period. Presumably, if there were some late psPD patients (from chemo/rad), because the trial was randomized, there would have very likely been some late psPD patients in the treatment arm… and yet, no treatment arm patients were removed for late psPD due to chemo/rad.

The main reason why I think that is the case is because in slide 11 of the final presentation there are 81 PFS events (this is PFS, not cPFS or confirmed progression-free survival events) coming from the 99 placebo arm patients, but in the rGBM trial there are only 64 patients. There is a difference of 17. How would you explain that difference if it is not the reason I mention? I'm open to read any possible reason for the difference, but I think that is. I would like to know what you and others think about this difference.

Yes, I agree that this chart reflects the original PFS, and NOT so-called “confirmed” PFS.

And, it is unusual that there were only 81 control PFS events of 99 by about 30 months, leaving 18 (right? Not 17: 99 - 81 = 18?) patients not having reflected a PFS event, and yet there are ZERO patients remaining at risk. If those 17/18 patients had died, their death would have also counted for the primary endpoint (according to the protocol).

I also think those patients could be the reason for having 13 left censored patients (lost to follow) in the JTM interim publication for the first 19 months from surgery and then zero in the treatment arm for nGBM from the final presentation, as I explained in previous posts.

A few of those patients (around 4) could have been control patients who were enrolled prior to 2008, before the better placebo injection was developed. AFTER entering the main arm of the trial, they would have quickly found out that they had been randomized to the control arm, and so perhaps chose to leave the trial, after having already being enrolled in it. It appears from the quote below that the trial was not set up to cross over patients to treatment.

In December 2006, we commenced recruiting patients with newly diagnosed GBM in a 141 patient Phase II DCVax®-Brain clinical trial. We planned to carry out the study at 12 to 15 clinical sites. The study was designed as a randomized study in which patients received either DCVax®-Brain in addition to standard of care or standard of care alone. To date, almost 50 patients have been screened at 4 clinical sites. However, patients have been reluctant to enroll in the study when faced with a 33% chance of being randomized into the control arm of the study under which they will receive standard of care alone. In order to address this issue we redesigned the study as a randomized, placebo controlled, double blinded study with a cross-over arm allowing control patients to be treated with DCVax®-Brain in the event that their cancer progresses. The study size has been increased from 141 to 240 patients and is designed to enable us to petition the FDA for accelerated approval if the study generates results similar to those achieved in earlier Phase I studies. In order to enable rapid enrollment, we are in the process of enrolling 45 to 50 additional clinical sites for this trial. As of January 1, 2009, thirteen  sites are active and a further 31 sites are at various stages of the start-up process. We are engaged in discussions with the FDA concerning the study design and end points. Depending on trial results, we plan to seek product approval in both the U.S. and the European Union.
https://www.sec.gov/Archives/edgar/data/1072379/000114420409020743/v146005_10k.htm

Please note that the above section indicates that there had been 50 patients “screened” (not enrolled) by the date of the above cited 2008 10K. However, according to the 2008 Q3, cited below, there were only 11 patients enrolled in the trial as of the 2008, 3rd quarter. If 1/3 of those were control, that would have numbered about 4 patients. Of course, IF these eleven patients were instead all enrolled in the treatment arm, then none of these patients would have left the trial due to discovering they were control, and instead, no control patients would have even been enrolled by that point.

We plan to rely on our current DCVax®-Brain Phase II clinical trial as a single study in support of regulatory approval. However, to date, only eleven patients have enrolled in the clinical trial, which is designed to include 240 patients.
https://www.sec.gov/Archives/edgar/data/1072379/000136231008007500/c77426e10vq.htm

So again, those 17 or 18 control patients who did not progress before leaving the trial remain a mystery, but I do not believe that they were late pseudoprogressors due to chemo/rad and so were removed from the trial, AFTER having been enrolled.

What adds to this mystery, is that there were an additional 17 missing patients from the control arm at the end of the trial. (The trial was supposed to enroll 348 patients - one third for the control arm would have represented about 116 patients. After the screening halt, the trial fully enrolled the treatment arm to represent 2/3 of 348- so 232 - but only enrolled 99 control patients - so there were 17 less control patients than the planned for 116). But while that is odd, company management insist that this is just coincidental.

And finally, I find possible that some SOC pseudoprogressors where not excluded from the trial because they pseudoprogressed after randomization.

The protocol did not have any provisions whatsoever for removing late pseudoprogressors from either the control or treatment arm. So although you may think that it is possible these patients were removed, this suggestion is not supported by the protocol. Additionally, it would hardly have been fair to remove any late psPD patients (who usually live longer) from the control arm, and then not correspondingly remove them from the treatment arm. And from the looks of that PFS chart shown on slide 11, there were ALOT of treatment arm pseudoprogression patients.

Sluicebox mentioned in his/her reply to you that Doctor Ashkan said that less than 10% of the patients didn't crossover. That would reinforce this theory, because we know that at least 89% received DCVax-L ( (232+64)/331 = 89%). So maximum 11% didn't receive it. If the number that didn't receive it is less than 10%, that would mean that part of the 35 patients received DCvax at some point and could correspond to the group of pseudopregressors that received DCVax at pseudoprogression.

We’ve known for a long time (many years) that about 90% of those enrolled in the trial crossed over. LL pointed that out on a slide that dates back to maybe 2016?

The odds of 17 or 18 patients in the control arm having never evented, and who went on to live many happy years of life without having receiving the DCVax treatment are very, very low. All of the doctors that have spoken thus far, seem to indicate that really, these patients have about a 1 to 3% chance of living to 5 years. And I think it’s worth pointing out here that the 14.6 median from Stupp would have included IDH mutated (which they weren’t measuring at the time) who live longer. When you remove those patients from the definition of GBM, leaving only wild-type, well, one can see why they might think that these SOC patients odds are so low. But even if we were to accord the control SOC patients in the DCVax-L trial a 5% chance of never having PFS evented, and thus counting towards those 17/18 patients we’re discussing, that would equal about 5 from 99 control patients.

IMO, thinking that these 17/18 control patients were late psPD who received DCVax but were not allowed into the main arm of the trial, AFTER having been enrolled in the main arm of the trial simply doesn’t fit the facts that we do know. You are likely familiar with the old adage, “once randomized, always analyzed”, and so, IMO, there’s has to be some other yet unknown reason. I would think that when the journal is eventually published, a more comprehensive presentation of the full data will explain this mystery.

[exwannabe]

The main reason why I think that is the case is because in slide 11 of the final presentation there are 81 PFS events (this is PFS, not cPFS or confirmed progression-free survival events) coming from the 99 placebo arm patients, but in the rGBM trial there are only 64 patients. There is a difference of 17. How would you explain that difference if it is not the reason I mention? I'm open to read any possible reason for the difference, but I think that is. I would like to know what you and others think about this difference.

To be in the rGBM sub-study the patient had to "cross-over" to it upon progression.

Any patient (from the N=99) who was either not eligible or elected not to cross would be in your N=17. These include:

Any patient who no longer meets the enrollment criterion at time of progression. It is very possible that the progression causes the patient's health to decline. But even unrelated issues could crop up. The patients are not in the best of health and are on chemo.

Any patient who just throw in the towel and elect not to cross.

Any patient who opted for a different trial figuring this trial did not work (remember, patient us blinded at time of election).

These are reasons explicitly called out by the trial protocol.

Your explanation OTOH is not in the protocol and frankly makes no sense. Once you change treatment protocol you cannot know if any remission is new treatment or psPD.

And do not forget, it has been stated numerous times that 90% received DCVax-L.