Northwest Biotherapeutics: Why the Failure of Celldex’s Rintega in Newly Diagnosed Glioblastoma Probably Has No Implications for the DCVax-L Trial
Investment Thesis
Some subscribers have asked me if the failure of Rintega in the phase 3 ACT-IV trial in newly diagnosed glioblastoma has any implications for success or failure of DCVax-L in its phase 3 trial in newly diagnosed glioblastoma. This note explains why I believe there are no negative or positive implications.
Subscribers have also asked me for my investment thinking on Northwest and a note is in progress. Northwest has been exceedingly weak over the past six months since screening for enrollment was temporarily halted in the phase 3 trial of DCVax-L in August of 2015. At the time, the Company suggested that it would submit data in just a few weeks to the FDA and investor expectation was that we would understand the reasons for the halt in perhaps October. However, the halt has now been ongoing for over six months with no word from the FDA. Throughout this period, the Company has been unable to communicate anything to investors. There has been a cone of silence about the stock for over one-half year and all data flow has ground to a halt.
I have been silent on NWBO for the reason that I have had nothing to say as I am awaiting the outcome of the screening halt in the phase 3 trial of DCVax-L which I originally expected in 4Q, 2015. I still cannot come up with a logical explanation for this halt and the investment thesis for NWBO is critically tied to the outcome of this event. See my report Thoughts on Suspension of Screening of New Patients in the Phase 3 DCVax-L Trial In the meantime, here are my thoughts on the Rintega failure.
Perspective on Rintega
Celldex announced that the ACT IV phase 3 trial of Rintega (rindopepimut) was halted because the Data Safety and Monitoring Committee (DSMC) informed the Company that the trial was not on track to meets its primary endpoint of overall survival. Rintega performed as expected and in accordance with the statistical design of the trial as survival was consistent with results seen in earlier phase 2 studies. The failure of the study was due to the standard of care (SOC) arm performing significantly better than Celldex expected. The DSMC’s interim analysis showed median overall survival of 20.4 months for Rintega and 21.1 months for SOC with a hazard ratio of 0.99. Rintega showed no improvement over SOC.
Rintega and DCVax-L are totally different products with very different mechanisms of action. Rintega is based on a peptide that targets a single cancer antigen target in glioblastomas that express EGFRvIII mutations which is about 30% of glioblastomas. DCVax-L is a living cell designed to create an immune response against all antigens present in all glioblastomas. For those who want to delve further into the very substantial differences, I have included product descriptions of both at the end of this report. The only link between the two products is that both are involved in phase 3 trials in newly diagnosed glioblastoma multiforme.
The statistical design of the ACT IV trial assumed that median overall survival for the SOC control arm would be about 16 months based on results from several previous trials. Short sellers are suggesting that the ACT IV trial shows that expected median overall survival for SOC has improved from 16.1 months to 21.1 months. They go on to suggest that this is ominous for the DCVax-L phase 3 because the SOC arm will perform better than expected as in as it did in the ACT IV trial.
I have looked carefully at the failure of the Rintega ACT IV trial and believe that it has no implications for the potential success of the DCVax-L phase 3 trial. I explain why in the first two sections of this report.
Median Overall Survival Results for SOC in other Trials in Newly Diagnosed Glioblastoma
The standard of care (SOC) in newly diagnosed glioblastoma is surgical resection followed by a regimen comprised of radiation plus the chemotherapy drug temozolomide. SOC was first established by results from the Stupp trial which compared radiation plus temozolomide to radiation alone in 578 patients. This trial was published in the New England Journal of Medicine in 2005 and showed that SOC produced median overall survival of 14.6 months.
Roche ran two major trials comparing Avastin plus SOC versus SOC. These trials showed that Avastin made no difference in outcome, but it did provide support data on what expected median overall survival could be expected from SOC. The Gilbert trial of 978 patients showed that SOC had median overall survival of 16.1 months. The Chinot trial of 458 patients showed median overall survival of 15.6 months. Results were published in the NEJM in February of 2014.
These were three large and well controlled trials that all indicated that median overall survival of the SOC regimen of radiation plus temozolomide could be expected to produce median overall survival of 14.6 months to 16.1 months. The SOC results in the ACT IV trial of rindopepimut showed much longer median overall survival of 21.1 months. I attribute this to a significant difference in the types of patients enrolled in the ACT IV trial versus those in the Stupp, Gilbert and Chinot trials
The Probable Reason for Why SOC Was So Good in ACT IV
There seems to be a straightforward explanation and this is the degree of surgical resection. The first step in treating newly diagnosed glioblastoma patients is to try to surgically resect as much of the tumor from the brain as possible and then begin radiation plus temozolomide. The more complete the resection, the better the expected outcome.
The higher survival for SOC in the ACT IV trial was probably driven by the trial protocol which required that all patients in the trial undergo gross total resection. This differed dramatically from the Stupp, Gilbert and Chinot trials in which only 40% to 60% of patients were completely resected.
The impact of resection on median overall survival was made very clear by a retrospective analysis of the Stupp trial. This analysis showed that in 91 patients who were 90% or more resected that the median overall survival was 18.1 months as compared to 14.6 months for all patients lumped together without consideration of the degree of resection.
This raises an interesting point in regard to how Celldex designed their ACT IV phase 3 trial. The design was based on the ACT III phase 2 trial of Rintega. This was a 65 patient trial in which all patients were fully resected and given Rintega SOC. There was no control arm. ACT III showed a 21.8 month median overall survival. Celldex took this as very encouraging in comparison to the Stupp, Gilbert and Chinot trials that showed median overall survival of 14.4 to 16.1 months. They apparently failed to consider that patients such as those in ACT III who are fully resected should do better. In the just halted ACT IV trial the median overall survival in the SOC arm was 21.1 months which is very close to what was seen with Rintega in the ACT III trial. It may have been the case that the encouraging results in ACT III were due to the more complete surgical resections and that Rintega produced no benefit.
In the phase 3 trial of DCVax-L in newly diagnosed glioblastoma, the trial is being done in patients with varying degrees of surgical resection which was the case in the Stupp, Gilbert and Chinot trails. Hence, the 14.4 to 16.1 months of median overall survival in the SOC arms of those trials is likely to be what might be expected in the DCVax-L trial if there was a true SOC arm. However, in the DCVax-L trial, patients on SOC are switched to DCVax-L if their cancer progresses so that there will be no true measure of median overall survival in the control arm. The primary endpoint of the DCVax-L trial is time to progression of the disease and not overall survival. The latter is a secondary endpoint, but the cross over design of the trial which means that potentially all patients in the trial will receive DCVax-L will confound analysis of overall survival.
Rindopepimut Product Description
Rindopepimut is an immunotherapy that is designed to target a mutation of the epidermal growth factor receptor (EGFR), a molecular complex that occurs in the membranes of cells. After binding to EGFR, proteins such as epidermal growth factor and TNFα create cellular signals that promote cell growth, proliferation and differentiation.
In normal cells, the EGFR receptor is sometimes turned on and sometimes off, but in some glioblastoma patients there is a mutation called , epidermal growth factor receptor variant III (EGFRvIII) that results in the receptor being permanently activated and resulting in uncontrolled cell division. It is hypothesized that this mutation through the release of IL-6 also stimulates the growth of neighboring cells that do not express EGFRvIII and that cells with this mutation may release microvesicles (cellular components that are released into the space outside the cell) containing EGFRvIII which can merge with neighboring cells.
Rindopepimut is comprised of a peptide or amino acid sequence that is specific to EGFRvIII and is not found in normal cells. This makes it an attractive target for immunotherapy. This peptide is then linked to a carrier protein called keyhole limpet hemocyanin, or KLH. Keyhole limpet hemocyanin is used extensively as a carrier protein to stimulate a response from the immune system. The peptide/ KLH linkage is administered as an injection along with the adjuvant GM-CSF. The objective is to create a powerful immune response that will kill cancer cells that express EGFRvIII. EGFRvIII is expressed in 30% of glioblastomas according to estimates provided by Celldex. I have seen other estimates that place it at 20%.
DCVax-L Product Description
DCVax-L is a totally different product form Rintega; it is actually comprised of living cells. Almost all drugs that I am aware of use a manufacturing process to produce the active pharmaceutical ingredient. In the case of DCVax-L, the manufacturing process is the product: it can be divided into three phases;
- The patient first goes to a leukapheresis center where white blood cells are collected.
- These are shipped to a manufacturing facility where monocytes are separated and given nutrients and cytokines that allow them to differentiate into immature dendritic cells.
- These are loaded with cancer antigens to produce the vaccine, which is then shipped back to the physician’s office or some other facility where it is given to the patient.
The leukapheresis process uses a tube that extracts blood from one arm and runs it through a machine that separates out white blood cells needed to produce DCVax-L. It returns the remainder of the blood, which now contains primarily red blood cells and plasma, through a tube in the other arm. Over a period of a few hours, the machine can extract a considerable number of white blood cells which are then shipped to the contract manufacturer, Cognate Biosciences.
The leukapheresis product contains monocytes, lymphocytes, granulocytes and other white blood cells. The cells are spun in a process that separates the different cellular components into gradient layers based on their density; one of these layers contains concentrated monocytes. The mixture containing monocytes is given nutrients to keep them alive and cytokines that stimulate their differentiation into immature dendritic cells. Mechanical factors such as rocking the culture are also used.
The next important step is to induce these immature dendritic cells to capture antigens from the tumor. At the time of surgery, most of the removed tumor tissue is sent to Cognate. It is lysed and then exposed to the maturing dendritic cells which capture the antigens. This process is meant to simulate the process in which dendritic cells take up antigens naturally. In the natural process, these antigens are then displayed to helper T-cells through the MHC complex to mount an immune response against tumor tissue displaying those antigens. DCVax-L is injected into the body to achieve the same effect of stimulating an immune response.
Tagged as DCVax-L, Northwest Biotherapeutics Inc., NWBO, Rintega failure in phase 3 ACT IV trial, suspension of screening in DCVax-L phase 3 trial + Categorized as Company Reports, LinkedIn
Very nice description of the processes and products. You have a real gift for explaining a technical process to the non-technical readers.
Good article Larry and GBM really seems to be a tough nut to crack! When CLDX announced the “re-act” data that was so positive, the one caveat in that trial was those that received Rintega were fully resected and the control was not which I believe created a “false positive”.
As far as NWBO goes, your reference to Get Smart and the cone of silence is comical. My thoughts is Power’s often stated that the -L trial was going to the end regardless. I believe that a second DMC review happened (Aug would have been 20 months AFTER the first DMC review) and that the DMC board halted the trial for futility and Power’s chose to continue the trial to the end without disclosing this to investors. Also considering -L per Power’s was to be fully enrolled no later than Q2 of 2014 and still wasn’t fully enrolled as of AUG 2015 is a problem in my eyes. Not to mention NWBO’s cash burn has increased as well as payments to Cognate and dilutive financing on worse and worse terms reinforces that. .Not sure why such a small company with ZERO new participants in trials rolls through cash so quickly, but the nickname of “Enron Powers” seems to be coming more and more true as the stock price surely reflects that.
Thank you for the comment. Your hypothesis that NWBO is continuing the trial even though an interim look by the DMC suggests futility is a possible scenario. The DMC is an advisor to the Company on the trial and if the DMC were to recommend halting for futility, the Company could continue the trial if there were no safety issue. It would not be illogical or sinister on the part of the Company as it consistently is the case in immunotherapy that a small percentage of patients (perhaps 30%) with durable responses drive the results and that the Kaplan Meier’s curves separate late in the trial. If there is no safety issue, there is a very strong scientific argument to be made in continuing the trial until the endpoint is reached.
I actually saw this occur in one other case. Cell Genesis halted a trial of its prostate cancer vaccine because there were more deaths in the drug arm than control at an interim look. However, Johns Hopkins continued to monitor the patients and the investigator watching the data has said that the curves have reversed and are now widening in favor of the cancer vaccine. The data has not been published so that my information which came from a presentation at a conference cannot be corroborated. However, I spoke at length with the investigator and have strong confidence that it is correct. I also don’t know if the results are statistically significant.
Also, this trial is stratified for many important variables such as MGMT methylation, degree of resection, immune status, etc. Even if DCVax-L was ineffective there is much to be learned about glioblastoma sub-populations and this is a strong reason for continuing the trial.
What your scenario fails to explain is why the screening halt and why the FDA involvement in looking at data. There is no safety issue and all patients are receiving standard of care with or without DCVax-L. Again, it there were a safety issue, the trial would have long ago been shut down for those enrolled in the trial. Perhaps you have an explanation, but why is the FDA involved and why is it looking at data (we don’t know what type) if there is no safety issue. There is something highly unusual in this situation and I think that when we eventually find out what it is, it could be shocking.
Thank you Larry for a comprehensive report on what appears to be a failure of one treatment and the possible meanings it has for NWBio….I appreciate that at some time in the near future you will again try and make sense of the “cone of silence” that is now the strategy of NWBio at this time…
I will await that report and then share my comments on it….longCLDX & NWBO
Larry- Surely you will have a comment of your own regarding the comment above? Or even better simply decontaminate the board.- Respectfully, Bill DeMott
Shocking? You mean like, massive financial malfeasance? The stock has already been decimated so please: elaborate some other shocking scenarios, hopefully among them something positive.
They say that hope is not a strategy but what else is left after what’s already happened to our investments in this area?
Just based on the data submitted to the FHA. it is reasonable to assume that NWBO knows what the FDA is considering, If their considerations spelled a bleat future for DCVax-L, Les cpuld not have written that the detractors “were trying to turn good news into bad news”.
Also it id unlikly that thr company would continue yo spend millions to prepare for commercialization
So this is more grist for the nill. inho.
Larry, With all the the patients in L trial get getting the vaccine in the end. Let’s say for argument sake that the vaccine works really well. So now the placebo and the patients in the regular trial are living longer in fact much longer. Like 10 months more than standard of care. However maybe there is only a one month overall survival difference between the placebo and non placebo group. Would the trial end in an error and fail even though the drug looks like a huge success?
Adding from my last comment. Can the L trial succeed if there is no statical difference in PFS and OS stats of the non placebo and placebo group. Even though both groups overall survival is much higher than standard of care?
The primary endpoint in the phase 3 trial is progression free survival and the secondary endpoint is overall survival. The trial protocol allows patients in the standard to care arm to be given DCVax-L once their cancer progresses (starts growing). This means that most/ many patients in the trial might will be receiving DCVax-L at the termination of the trial. Obviously, this confounds interpretation of overall survival results. Let me make a few points.
Some critics like Adam Feuerstein have wrongly stated that progression free survival is not an acceptable endpoint. This is totally wrong (rubbish) as many cancer drugs, notably the checkpoint modulators Opdivo and Keytruda, use PFS as an end point in some trials. The viability of an endpoint depends on how medically meaningful it is. In the case of glioblastoma, progression free survival is enormously important. This is an aggressive cancer growing in the confined space of the skull. As it progresses or expands, it impinges on other areas of the brain causing devastating side effects, notably convulsions. It can actually push parts of the brain down the brain stem. Any drug that can provide a meaningful lengthening of the time to progression will be quickly approved. So what is meaningful?
The phase 3 trial of DCVax-L is designed to show statistical significance if DCVax-L can improve progression free survival by 4.0 months. Let me put this in perspective. Three well controlled trials have shown that progression free survival with standard of care is 7 to 8 months. Hence, DCVax-L probably would have to show PFS of 11 to 12 months to be successful on a statistically significant basis. As a point of reference, in the phase 1/2 trial of DCVax-L in 20 patients the progression free survival was 25 months.
Let’s turn now to overall survival. The same three well controlled trials have shown standard of care produces median overall survival of about 16 months. This means that half of patients are dead in the time frame of 16 months. Two year survival is about 25% and three year is about 16%. In interpreting, the phase 3 results we would certainly want to see median overall survival of perhaps 18 to 21 months or more in the group who received DCVax-L from the start of the trial. In an aggressive cancer like glioblastoma, a four month improvement in median overall survival is considered to be a significant therapeutic advance. Some drugs have been approved for highly aggressive cancers on the basis of as little as 1.5 month improvement in median overall survival (gemcitabine in refractory pancreatic cancer).
In this study, we would expect to see the standard of care arm which was switched to DCVax-L perhaps do better than 16 months. If the primary endpoint of PFS is reached and something like these data for overall survival are achieved the drug would almost certainly be approved. As a frame of reference, in the 20 patient phase ½ trial of DCVax-L median overall survival was 36 months.
As a final note, I would point out that the statistical design of the trial calls for the enrollment of 348 patients. At the screening halt, the Company said that meaningful more than 300 patients were enrolled in the trial. There is no way of knowing but this language suggests to me that the trial has enrolled 320 or more patients. If the full 348 are not enrolled, it could have an effect on the statistical analysis, how much or how little I can’t say.
Remeber the primary endpoint of L’s PIII study is Progression Free Survival. Overall Survival is a secondary endpoint. There seems to be some concern about patients being able to cross-over to receive L and how that may skew the results. I cut and pasted the following from Wikipedia RE: PFS vs OS.
On the other hand, PFS becomes more relevant than OS when in a randomized trial patients that progress while on treatment A are allowed to receive treatment B (these patients may “cross” from one arm of the study to the other). If treatment B is really more effective than treatment A it is probable that the OS of patients will be the same even if PFS may be very different. This happened for example in studies comparing tyrosine kinase inhibitors (TKI) to standard chemotherapy in patients with non-small cell lung cancer (NSCLC) harboring a mutation in EGF-receptor. Patients started on TKI had a much longer PFS, but since patients that started on chemotherapy were allowed to receive TKI on progression, OS was similar.
Hi Larry,
Thanks for you article. If the CLDX failure has any impact on NWBO, I believe it is in the competition in the addressable market. Instead of having to compete against CLDX (and potential other competitors), NWBO if approved may become the only undisputed treatment in this unmet medical need with a terrible outlook. So I would guess the value of NWBO as a company should be now more highly valuated (possibly more than CLDX before the failure, but that’s a different story).
Thank you Larry and everyone for making the on-going discussion of what may or may not be happening with the “L” phase III trial…..I only wish I had something positive to add to it…but I don’t so I will continue to read everyone’s comments and “stay tuned” to what may or may not be disclosed by the company and anyone else that thinks they have some insight to what is actually happening….One question I have always had surrounding any Bio Tech trial is, “does the patient sign somekind of non-disclosure agreement that prevents them from commenting on their participation in the trial?” I would think that if a significant amount of patients were living longer and feeling better they would be shouting it from the mountaintops and we would be hearing this antedotal evidence of the power of their particular treatment….just a question…thank you
There’s no stop for futility. This entire episode began in Germany, I’m 95% sure of this, and not just because that’s where the website made it look like a halt. I believe they stopped screening at the rest of the sites because they didn’t want an unfair advantage at any one site. I think there’s something up in Germany; it’s a weird and difficult issue negotiating with the Sickness Funds, which NWBO can’t do directly through the HE, only through individual Hospitals who’ve applied to negotiate for reimbursement through the HE. Additionally, they’re required to negotiate through an arrangement in the HE called a NUB agreement. This only lasts for one year and has to be renegotiated each year. However, if they have a full Marketing Agreement (MA) they can negotiate to be a supplement to the DRG system of coding and billing, or get their own DRG. Both of these latter give the best chance to secure reimbursement at top billing and keep it there. The NUB route may not provide success. NWBO and/or the PEI may have felt it unfair to keep bringing in more patients and giving some people Placebo while an approval process was in the works.
I’m hypothesizing NWBO is going for a MA through the PEI and it’s the PEI that’s the “regulatory body” they’ve provided documents to. I think it’s also possible that the data was indeed looked at as a second IA and there’re too many people continuing to live to stop the trial early as now they’re comparing “early vaccination to late vaccination” per Dr. Liau. If the PFS was statistically significant and might warrant AA perhaps if they can tease out the difference between OS after progression on Placebo vs. OS at early vaccination after SOC, or find that even after progression DCVax works just as well as early vaccination they can conclude the OS is from the DCVax in both groups. This would Undoubtedly be cause for Full Approval.
My conversations with management over the past 6 months, and reading between the lines, has led me to believe they’re shooting for the whole enchilada here, PPS be damned. They’ve completely implied to me that they’re not going to settle for short term gain where it may lead to a delay in a long term home run. Hence the silence. I’m trying to think outside the box, like I think Linda Powers is.
Thank you Larry for all the dialogue and opinions on what may or may not be happening with this revolutionary science and application for many kinds of active cancer…..I appreciate your insights into the many strange things that surround this infant company and its fight for life and the deep desire to remain independent and above water for awhile longer….So, if you happen to read this, thanks….I am buying more shares at these fire sale prices and will continue to until it is either proven to be effective or a dud, as some have speculated….at these prices, it seems like a steal…however, it does keep dropping so every share I have ever brought is below what the closing price is today….so, “stay tuned and wait and see”…..I sure hope the science works and people are given a better alternative to Chemo and Radiation,,, what a day that will be….so, the “cone of silence” goes on and “out-side the box” is normal and I, for one, (maybe the only one), trust Linda and Les and Max and the BOD and JW and you, my friend….cheers