NorthWest Biotherapeutics Inc (NWBO)

[froggmister]

Transcript of Dr. Liau JAMA Author Interview

Lots of talk of $DNDN, and I was right in the thick of that. I bought my first shares of $NWBO in 2012 after selling $DNDN – luckily most of it sold near the high of $50 a share, although I did hold some while piece by piece the company was exposed as woefully unprepared for success (apparently ex saw this more clearly than I did - live and learn). I was definitely one of the Dendreonite “lunatics” described in the Deep Capture story referenced here lately, and yes, there are a lot of comparisons – what has caught my eye is the way skeptics of the data turn out to have conflicts of interest – that was very true in the $DNDN saga. In fairness, I also bought other immunotherapy company stocks with my $DNDN profits that did not go well ($ONTY, RIP) but we won’t dwell on those.

I held just a small stake as I watched $NWBO from afar through all the well-documented travails – the halt, Woodford, and many others. It wasn’t until I read IkeEsq’s transcript of the 2018 ASM (held in Feb 2019) that I decided I needed to do a deep dive and consider going all in. I have benefited greatly from all of the amazing DD shared here by so many, including the transcripts of IkeEsq and Senti. So to pay that back a bit, here is the transcript of Linda Liau’s Audio Author interview accompanying the watershed JAMA publication of the Phase 3 DCVax-L’s trial results.

https://podcasts.apple.com/us/podcast/association-of-dcvax-l-with-extension/id1227001070?i=1000586593291

A couple things that struck me. One is that West makes two glaring errors in his questions. One is that he refers to TTF as “Standard of Care” which it is not, and that the comparators are “historical.” Linda doesn’t call either mistake out, and finesses the TTF question graciously, especially since we know what she thinks of TTF from other comments she has made. The error regarding “historical” comparators from someone as knowledgeable as West makes it clear how hard this misconception is to move past.

Jack West: (0:16) Hello this is Dr. Jack West from the City of Hope Conference and Cancer Center, and web editor for JAMA Oncology with another audio author podcast. I’m happy to be joined today by Dr. Linda Liau, MD, PHD and MBA, who is the chair and a professor in the dept of neurosurgery as well as the director of the brain tumor program at the UCLA David Geffen School of Medicine. She’s here now to discuss a new publication in JAMA Oncology for which she is the lead author, “Autologous Tumor Lysate-Loaded Dendritic Cell Vaccination In Patients with Newly Diagnosed and Recurrent Glioblastoma, a Phase 3 prospective externally controlled cohort trial.” Thanks so much for joining today.

Linda Liau: (1:03) Thank you for having me.

JW: (1:07) So let’s just start with some context. Can you please talk about this platform of autologous tumor lysate-loaded dendritic cell vaccination; what’s the mechanism of action and what does it actually entail in practice?

LL: (1:20) Yes, so the mechanism of action for the platform really centers around the dendritic cell, dendritic cells are professional antigen presenting cells that are part of the repertoire of immune cells in the body, so the concept really is to take the antigen presenting cells and present them to antigens, particularly for brain cancer dendritic cells don’t normally circulate widely around the brain so they may not recognize tumor antigens within the brain. So the concept is really taking a patient’s own dendritic cells and pulsing them with antigens, and in this case it is derived from the patient’s own tumor tissue. So that is itself the vaccine, and that is administered to patients, and in previous pre-clinical and early clinical studies we found that when this is done we can induce T-cells infiltration into glioblastomas, which is a way to make these non-immunogenic tumors immunogenic, because we do know is that glioblastomas unfortunately do not have high infiltrations of T-cells so this is a way to induce that. So that’s essentially the mechanism of action behind this platform, and in terms of what this entails in actual practice this is individualized or personalized immunotherapy in the sense that the dendritic cells are autologous – taken from the patient via leukapheresis – and the tumor tissue is also autologous taken from the patient at the time of surgery, and it would involve basically collecting the white blood cells and collecting tumor cells, co-culturing them outside of the body in a manufacturing facility, and then when that product, that cellular product is made that is shipped back to the individual sites and for injection into the patients. The injection is just a vaccine shot given in the arm just like any other vaccine.

West: (3:19) And the leukapheresis process and then all the subsequent injections that are done, initial every few days and then weeks and then maintained over months is all after the initial debulking or full craniotomy resection that gets the tumor tissue that’s needed as well as obviously being a therapeutic procedure, and then all of these patients underwent postoperative chemoradiation afterwards, right, with temozolomide and per standard, both at the time and what would be done now, before going on to the vaccine, is that correct?

LL: (4:00) Correct, yes.

West: (4:03) So the trial itself was developed as placebo-controlled with an initial plan to have the primary endpoint be Progression Free Survival (PFS) and comparing placebo to those who received the active drug, or maybe vehicle, compared to the active vaccine, was converted to Overall Survival (OS) after the trial had been initiated. Can you talk about the process of the shift in the primary endpoint and why that came about?

LL: (4:33) So while the trial was underway the field recognized that pseudo-progression is a major confounding factor in glioblastoma, particularly with newer immuno-oncology treatments and also in the MGMT methylated population, even with the standard of care treatments, so what happened as the trial was underway was that it became apparent that it was very difficult to distinguish what was true progression vs pseudo-progression. And the way the trial was designed the determination of progression was actually made by independent radiologists at a centralized site. So all the trial sites would send in the MRI scans and the centralized radiologists would adjudicate whether there was progression or not and what we began to find was that in over 50% of the cases the radiologists did not agree – there was ambiguity in that adjudication, and then over the years in the field what we’ve learned is that the way the radiological reviews are done there’s not really a way to distinguish between progression and pseudo-progression. This trial, when it was written, the criteria used for that was based on the McDonald criteria for determining progression, which is what was prevalent in the field at the time. Since then that has been changed to the RANO criteria (Radiological Assessments for Neuro-Oncology) but when immunotherapies came onboard that was viewed as not really adequate, then there was something called iRANO, and iRANO lately was viewed was not be adequate either, and now there’s something called modified RANO, so anyway through all of this what we learned was that the field still to this day cannot distinguish between true progression and pseudo progression and that why we changed the endpoint to OS, because at least that is a non-equivocal endpoint.

West: (6:40) Just to clarify for our listeners, IRANO is specifically for immunotherapy with the potential for pseudo-progression, is that correct?

LL: 6:49) Correct, that’s true, but just earlier this year it was actually deemed iRANO probably is not a good assessment criteria either because part of iRANO requires a re-scanning in 3 months after initial visualization of tumor progression, and what was found when they applied this to actual practice is that almost 50% of patients in immunotherapy trials then became censored. So all things considered, at least in our trial, OS was the probably the best endpoint.

West: (7:28) We’ll get on to the actually efficacy endpoints, but before we do, it’s probably worth highlighting that the ultimate comparison was not internally as a randomization within the trial but to externally controlled historical data from other trials. Can you talk about that decision? And then we can get into what you actually saw.

LL: (7:50) The comparison to external controls was done based on that fact that, quite frankly, because we lost the internal controls. There was crossover arm designed into the trial – and that was actually mandated by the FDA at the time – because was a relatively invasive procedure in the sense that patients had to undergo leukapheresis and surgery in order to be involved in this trial so it was felt to be ethically best if patients were allowed the option to crossover after their tumor progressed. So with that, and the confounders of pseudoprogression vs. true progression is that what happened was about 90% of the people in the control arm had crossed over, so essentially there were very few people left in the internal controls so that’s why the decision was made to compare this group of patients to the external controls, and this decision was made prior to the unblinding and was written into the Statistical Analysis Plan that was submitted.

West: (8:53) So let's get into the findings. What did you observe in terms of the OS for both the nGBM and rGBM cohorts compared to the external populations? .

LL: (9:07) In the nGBM cohort what we found was that the mOS in our patients was 19.3 months compared to 16.5 in the external cohort when we compared it at the time of randomization. The trials that were selected were all randomized at a similar time point. In the rGMB cohort that difference was 13.2 mo in our vaccine-treated patients vs 7.8 mo in the external cohort. But what I think was even more interesting is really the long survival tail in the vaccine cohorts. So the 5 year survival in our trial was 13% vs. 5.7% in the external cohort. Interestingly, when we did subgroup analyses and parsed out the MGMT methylated patients, their mOS was almost 3 years, 30.2 months from randomization and 33 months from the time of surgery. I find that very interesting because many of the immuno-oncology trials, for instance the checkpoint inhibitor trials that have been done in recent years in the GBM population have purposely taken out the MGMT patients because it was felt that this subgroup of patients with GBM were responsive to chemotherapy so we should be testing our immunotherapies in the unmethylated group, but what our trial showed was that perhaps the added survival advantage of immunotherapy could be more profound in the MGMT subgroup. This is something that will probably warrant further study to validate the data obviously.

JW: (10:58) What about the tolerability of the vaccine? Did you see any toxicity concerns in the clinic?

LL: (11:06) Surprisingly no, it was actually very good tolerability for the vaccine, and the safety profile was exceptionally benign. Out of the over 2000 doses that were administered throughout the trial we only saw 5 SAEs that were attributable to the vaccine itself, and most of the toxicities were relatively minor – flu-like symptoms, fatigue – similar to what you would experience with any kind of vaccination.

JW: (11:40) I didn’t see the numbers for the process of screening vs patients who were actually enrolled. Were there patients for whom the leukapheresis to obtain the dendritic cells or the generation of the process externally from the tumor lysate was unsuccessful? How common was it for patients to get through from start to finish vs attrition along the way in real life?

LL: (12:05) So we actually did publish some data regarding the screening of these patients and logistical issues related to vaccine manufacturing in a 2018 publication, but eventually the percentage of patients who had screen failures based on unsuccessful manufacturing of vaccine was about 6%, so there were some patients for whom we not able to make vaccines for but overall it was a relatively low percentage.

JW: (12:35) So obviously these data are favorable compared to the external data, but this study started more than a decade ago, and obviously it’s undergone some evolution, and in that time the field has evolved so that today Tumor Treating Fields (TTF) would now be incorporated as a Standard of Care (SOC). That said, I would say that that can be cumbersome, it’s not practiced that universally, but what would you consider the current implications of this work to be, given that the external data don’t include TTF and that SOC has at least optimally evolved somewhat since the conceptualization of this trail.

LL: (13:19) That’s a very good point. The trial itself and the external controls which we compared the trial to did not include patients who underwent TTF so obviously that’s a little bit different than currently, but even with that being said, TTF is still not widely necessarily, you know, widely used in most of the larger academic centers, and I think the results of our trial in the context of the current field really hopefully will allow further studies in the immuno-oncology space. I don’t think a Mechanism of Action between dendritic cells vaccines and TTF’s are absolutely similar, and so the reality is we that don’t have enough treatments for our patients not that we have too many, and different subgroups of patients will probably respond differently to different treatments. So I think the goal in the future is to really decipher which patients respond better to immunotherapies and vaccines and which patients may respond better to TTFs and which might respond best to both used concurrently. So I think this hopefully will serve as a platform for future studies of various types of combinations.

JW: (14:40) So essentially this could be kind of orthogonal to TTFs, I mean they could be combined or approached independently; there’s no reason that it has to be some either/or approach, but potentially a both/and, yes?

LL: (14:55) I don’t thing there is a need to prove that this is better or worse, it very well could be orthogonal or even complementary. As a neuroscientist, I think scientifically what is interesting about this approach is that what we know is that it gets T-cells into the glioblastomas, and that could be what is needed, you know the activation of T-cells is \needed to allow other treatments, you know the immune modulators and the checkpoint inhibitors to really have an effect in glioblastomas because those treatments have been effective in other cancers but unfortunately as single agents we haven’t seen efficacy in patients with GBM. So hopefully this will allow a window for that type of study to occur.

JW: (15:43) Do you see the next step of this being focused on the MGMT population only or more broadly?

LL: (15:52) I think it would be focused more broadly and more narrowly, I think the answer is both. I think there is some efficacy in the unmethylated group as well, we’re just not seeing the impact quite as greatly as in the methylated group. I think MGMT methylation is probably surrogate for some other marker that we have yet to discover.

JW: (16:15) And then, finally, we already alluded to the data supporting TTFs and the underutilization in practice because of it being challenging, somewhat cumbersome in the real world, or at least in certain areas being underutilized. How does this potentially fit in? I did note that it took a long time to enroll…it was from 94 centers to get over 300 patients. Is this something that could be adopted far and wide or is this something that that is best suited for specialized centers.

LL: (16:56) So because of trials done in 94 centers around the world, many of these centers and not actually big academic centers, many are community hospitals. So I think that the completion of this trial showed was that it is feasible to be done at various types of centers throughout the world. What it involves is like I mentioned earlier, collecting the tumor tissue a the time of surgery, the leukapheresis procedure and the manufacturing, but once that is done the cumbersomeness is actually quite minimal, it just involves that the patient comes back for vaccine injections. So I think just getting this trial done was quite a feat, but the fact that we were able to complete it showed that it is feasible to have this vaccine manufactured at 94 centers and on the patient side hopefully it will be well tolerated and feasible.

JW: (17:48) Excellent. Well, Dr. Liau, congratulations to you and colleagues on this paper especially being a field where real advances are hard to come by, and it is hard to show clinically meaningful progress, so really a valuable lead here. Good work.

LL: (18:05) Thank you very much