Expert Financial Analysis and Reporting

Introduction

The catalyst for writing this update was the MHRA certification of Northwest Biotherapeutics’ Sawston, UK plant for GMP production of DCVax-L for compassionate use. This was a critical step on the path to approval and commercialization of DCVax-L. Having taken pen in hand, I decided to update my reporting on other issues relating to commercialization and approval. I believe that we may be approaching the publication of results for the phase 3 trial of DCVax-L in newly diagnosed and recurrent glioblastoma multiforme in the form of a peer reviewed article in a major medical journal. This, if it occurs, will be an absolutely critical event that will shape the attitude of the physician community, regulators and investors as to whether DCVax-L will become a new component of standard of care in newly diagnosed and recurrent glioblastoma multiforme. Because of this, I expanded this report to hypothesize what the results might be, possible resultant regulatory action and potential commercial issues if DCVax-L successfully achieves its primary endpoint in the trial as I expect. This report builds on numerous prior reports, that I have written, some of which are referenced by links.

As I was in the final stages of finishing this report, 15 key FDA officials, including Richard Pazdur, published a manuscript in the Annals of Oncology (a highly regarded and  innovative peer reviewed medical journal in Europe) declaring  that  in some cases the use of external controls in clinical trials could be appropriate and defining the circumstances under which this would be so. Because Northwest is using external controls in its Statistical Analysis Plan, there had been some concern that FDA regulators might reject its BLA out of hand.  The publishing of this manuscript could not be more timely and meaningfully reduces the chance of the BLA being rejected out of hand. Aside from the results of the phase 3 trial-GMP capability at Sawston and the FDA position on external controls were two major issues/uncertainties related to the  potential approval and commercialization of DCVax-L. These developments were two major wins for NWBO. Now we await the all-important results from the phase 3 trial.

Organization of This Report

As I got into writing this report, the number of issues that I addressed kept expanding and expanding. As a result, the report has become lengthy and complex. To aid you in reading, I have listed below the issues addressed in this report in the order in which they appear. So, if you don’t have the time or interest to read the whole report at once, you can fast forward to issues that most interest you. For those of you who just want to see key conclusions, I have written the section called “Key Points Raised in This Report” that allows you in a couple of minutes to see my conclusions on key questions related to the NWBO investment thesis. Here then are the sections of this report.

Glossary of Abbreviations Used in This Report

Key Points Raised in This Report

MHRA Approval for GMP Manufacturing at Sawston, UK  Is a Very Important and Remarkable Achievement

Approval for Commercial Use is Remaining Hurdle; Understanding the Statistical Analysis Plan (SAP) is First Step in Gauging Chances for Approval of DCVax-L in GBM

    Primary Endpoint is mOS

    External Controls Are Designated as the Control Arm

   The Trial is Also Designed to Determine Efficacy in rGBM

Will the FDA and Other Regulatory Agencies Accept This SAP?

Will The Phase 3 Trial of DCVax-L Be Successful?

Why the Delay in Releasing Topline Data?

The First Approval Could Come in the UK

Will DCVax-L Become Part of SOC for Glioblastoma Multiforme?

The Survival Tail Could Be the Most Important Data from the Trial

Development of the Sawston Facility to Meet Commercial Needs

What Might Be the Price of DCVax-L?

Estimated Worldwide Incidence of Newly Diagnosed Glioblastoma Multiforme

Addressable Markets As Measured in Sales for UK, US, Canada and EU (ex UK)

Potential of Dendritic Cell Vaccines Beyond GBM

Glossary of Abbreviations Used in This Report

GBM- Glioblastoma multiforme is the most aggressive type of brain cancer. About 50% of patients die within about 17 months of diagnosis.

ndGBM-Newly diagnosed glioblastoma multiforme.

rGBM-Recurrent glioblastoma multiforme. Cancer recurs after initial response to treatment.

SOC- Standard of care. Currently for ndGBM, this is surgical resection followed by radiation and then the chemotherapy drug temozolomide.

DCVax-L- A dendritic cell therapeutic vaccine developed for the treatment of GBM.

MHRA- The Medicines and Healthcare products Regulatory Agency regulates medicines in the UK comparable to the FDA in the US.

EMA – European Medicines Agency is the European Union equivalent to FDA.

Health Canada- The Canadian equivalent to the FDA.

GMP- Good manufacturing practice is a methodology through which regulatory agencies ensure that drug products are consistently produced and controlled in accordance with specified quality standards. GMP approval from regulatory agencies must be obtained in order to commercialize a drug.

Sawston- This is Northwest Biotherapeutics GMP facility in the UK for DCVax-L production capable of supplying world markets.

mPFS- Progression free survival (PFS) is the length of time during and after the treatment of a cancer that the cancer does not progress or worsen.  The median is the middle of a list of numbers with half above and half below.

mOS- Overall survival (OS), the duration of patient survival from the time of beginning treatment. mOS is the gold standard endpoint for clinical trials of cancer drugs. At mOS, half of the patients in a trial have died.

SAP- Statistical Analysis Plan is a document containing a detailed elaboration of the principal features of the analysis described in a clinical trial protocol; it includes procedures for statistical analysis of the primary and secondary end points and other data. It must be specified prior to data lock for a clinical trial.

Key Points Raised in This Report

• The data base for the phase 3 trial of DCVax-L was locked on October 5, 2020. For reasons discussed in this report, there has not yet been a release of key results. I am hopeful that we may be approaching the release, but management has given no guidance and I have no basis for judging when this might occur.
• I hypothesize that the trial met its primary endpoint of mOS for ndGBM and possibly mOS for rGBM. This is based on my analysis of blinded trial results (see this link), conjecture by the lead investigators in the trial that is also based on blinded data (see this link) and anecdotal reports regarding compassionate use (see this link and this link).
• Results will almost certainly first be released in a peer reviewed, major medical journal. This will be critical in forming opinions by the physician community, regulators and investors on the potential for DCVax-L to become a component of SOC in both ndGBM and rGBM. This is potentially a major (earth shaking) catalyst.
• If results are clearly positive as I expect, this should lead in coming months to regulatory submissions in the UK, US, Canada and the EU (ex UK), These were regions in which the trial was conducted.
• Northwest has been producing DCVax-L for compassionate use in the UK for several years using a GMP approved facility in London which could produce product for up to 4 to 5 patients per month.
• The MHRA has just certified NWBO’s Sawston, UK plant for GMP production of DCVax-L for compassionate use that will take the place of the London facility. It has the capacity to produce product for 45 to 50  patients per month or after allowing for downtime due to plant maintenance, perhaps 500, or so, an annual basis. This certification is a remarkable accomplishment for such a small company.
• The MHRA will need to certify Sawston for commercial GMP production if DCVax-L is approved. The successful certification for compassionate use raises my expectations that the agency will also certify Sawston for commercial use, but this is not guaranteed.
• The MHRA approval of the Sawston plant in the UK was a critical step toward the commercialization of DCVax-L. Assuming that it receives commercial GMP certification, it can then be expanded to supply worldwide demand for DCVax-L and support several billion dollars of sales.
• I believe the first approval of DCVax-L for commercial use could be in the UK and, if accelerated, might occur this year.
• In this report. I present a plausible scenario (one of many possible) that suggests that the annual addressable market for ndGBM is $265 million in the UK, $1.3 billion in the US, $145 million in Canada and $1.5 billion in the EU (ex UK). The market opportunity is huge.
• I hypothesize that DCVax-L could be approved as an addition to SOC for ndGBM and also rGBM. There has been no meaningful improvement in drug therapy since 2005 for ndGBM in which 50% of newly diagnosed patients die within 17 months or so of diagnosis. Physicians and patients are desperate for any treatment that can prolong survival. Note that DCVax-L would be added to current SOC so that it would not have to replace some current component. These factors along with a remarkably benign side effect profile and easy to administer intradermal injections argue for swift and meaningful penetration of the ndGBM market.
• The last patient in the DCVax-L trial was enrolled in November of 2015 so that we will have at least five years of data on that patient and even longer on all other patients. About 95% of ndGBM patients are dead at five years so that we will have unprecedented and extremely valuable data on actual survival, not just mOS. This allows the calculation of a survival tail.
• It is the survival tail that has been instrumental in the success of other immune therapies, notably checkpoint inhibitors like Opdivo and Keytruda. In this report, I compare the actual survival tail of Opdivo shown in phase 3 trials in second line non-small cell lung cancer to an estimate based on blinded data of the survival tail of DCVax-L in ndGBM. These two cancers are both aggressive cancers with five year survival rates of 3% for second line non-small cell lung cancer and 5% for ndGBM.
• The survival tail of Opdivo in second line non-small cell lung cancer showed that 13.4 out of every 100 patients treated with Opdivo survive for five years versus 2.6 of every 100 patients treated with chemotherapy; the survival tail is the delta (difference) of about 11. Based on estimates derived from blinded data in the DCVax-L trial in ndGBM, I estimate that 15  of every 100 patients treated with DCVax-L might have survived five years versus 5 of 100 with SOC. If this proves to be the case, the survival tail or delta of 10 could be viewed with great excitement by the oncology community.
• As Winston Churchill so famously said “Now this is not the end. It is not even the beginning of the end. But it is, perhaps, the end of the beginning”. There remain many twists and turns, but this could be one of the most dramatic, positive stories in the history of biotechnology.
• The current market capitalization of NWBO is approximately $1.0 billion if one accounts for potential dilution from options and warrants. If the story unfolds as I have laid out in this report, comparison with peer oncology companies suggest to me that the market capitalization could possibly reach $5 to $10 billion or more in coming years. Of course, this all depends on the phase 3 data and subsequent regulatory actions by the MHRA, FDA, EMA and Health Canada.
• I want to emphasize that this is a high risk investment and that current or potential investors must recognize this. I am basing my analysis on blinded data and there is no certainty as to whether the unblinded data will confirm my conclusions. NWBO is a small but meaningful portion of my stock portfolio. I think that there is the potential for a very strong upside move in the stock, but I also accept the risk that the unblinded data could be disappointing and result in the loss of most of my investment as is the nature of almost any biotechnology investment. Any potential investor should understand and be willing to accept this risk/ reward profile.
• In the event that DCVax-L gains approval for ndGBM, it validates the dendritic cell therapeutic vaccine technology. The mechanism of action in ndGBM might also work for any solid tumors that can be surgically resected. So, there might be the potential for expansion of usage over a broad range of solid tumors resulting in commercial potential several magnitudes that of GBM. This would take place over years or decades.

MHRA Approval for GMP Manufacturing at Sawston, UK  Is a Very Important and Remarkable Achievement

NWBO’s Sawston UK manufacturing facility has just received MHRA approval for GMP manufacturing of cell therapy products (i.e. DCVax-L)  for compassionate use. Northwest has been producing DCVax-L for compassionate use in the UK for several years using a GMP approved facility in London which could provide product to treat up to 4 to 5 patients per month. Note that the COVID pandemic has negatively affected this number in some months over the last two years. I am guessing that perhaps 50 patients, more or less, have received DCVax-L on a compassionate use basis in the UK in recent years.

Sawston is the designated site for commercial production of DCVax-L. If the product receives regulatory approval, the plant will need to be certified for GMP production of commercial DCVax-L. The successful certification for compassionate use raises my expectations that it will also be GMP certified for commercial production, but this is not guaranteed.

It is a major achievement on a technical level to gain GMP certification of any manufacturing facility and especially for a new facility seeking its first certification. This is even more the case for cell therapies like DCVax-L whose manufacturing processes are much more complex than for most drugs, in large part because production is labor intensive. Also, remember that manufacturing complexity is enhanced because cell based products are living organisms, not organic molecules.

GMP manufacturing compliance is a sine qua non for commercialization. I can’t recount how many times failure to achieve GMP compliance has delayed and sometimes prevented a product from gaining regulatory approval. This seems more often than not to be a high hurdle which emerging biopharma companies trip over, certainly more so than established companies with long term experience in manufacturing. This approval speaks volumes about the competency of NWBO management. It is a remarkable achievement.

I would point out to you that even mighty Novartis had GMP issues that significantly slowed the launch of its cell therapy (CAR-T) product Kymriah. I want to emphasize and re-emphasize that the MHRA approval for GMP manufacturing of cell therapies at Sawston would be a proud achievement for even a large biopharma company and much more so for a development stage biopharma.

Approval for Commercial Use is Remaining Hurdle; Understanding the Statistical Analysis Plan (SAP) is First Step in Gauging Chances for Approval of DCVax-L in GBM

With GMP manufacturing likely in place I am breathing a huge sigh of relief. I was concerned that DCVax-L might not be approved because it would not have an approved GMP manufacturing facility. I re-emphasize that commercial approval will require a separate facility approval, but it is extremely encouraging that NWBO now will actually be producing vaccines in real world circumstances. The next critical step, of course, is gaining commercial approval for DCVax-L. Let’s move on to discuss how investors can gauge the potential for approval. In considering the probability, let’s start with the SAP that will be used to analyze the results of the phase 3 trial.

Primary Endpoint is mOS

When the trial was designed and initiated over 14 years ago, management thought that the primary endpoint of the trial would be mPFS with the key secondary endpoint being mOS. These were customary endpoints for cancer trials at the time because clinical development was almost entirely focused on chemotherapy drugs. If they are effective, chemotherapy drugs quickly shrink the size of the tumor as measured by CTI and MRI imaging. Progression free survival refers to situations in which the tumor is eliminated, shrunk or is not growing. Hence mPFS as judged by imaging is a valid primary endpoint for chemotherapy although mOS is the gold standard.

In 2013, six years after the DCVax-L phase 3 trial began, important new immunotherapy drugs began to emerge as effective cancer therapies including but not limited to the checkpoint inhibitors, Yervoy, Opdivo and Keytruda. It soon became apparent that these drugs act differently than chemotherapy. Their mechanism of action induces the body’s natural immune cells to attack the cancer which can lead to inflammation of the area comprised of cancer and surrounding cells. Imaging technologies may then show an enlargement of tissue which mistakenly can be interpreted as the tumor getting bigger or progressing even if the immunotherapy drugs go on to effectively shrink or eliminate the tumor. This is known as pseudo-progression. At the time of the initiation of the DCVax-L trial in 2007, the phenomenon of pseudo-progression in brain cancer was virtually unknown.

The thinking at the beginning of the DCVax-L phase 3 trial was that mPFS would be the primary endpoint, but based on the clinical experience with checkpoint inhibitors this came to be viewed as problematical because pseudo-progressors could not be easily differentiated from patients whose cancer had actually progressed. While the trial was ongoing and blinded, it became apparent to NWBO based on the accumulating evidence from trials of other immunotherapy drugs that mPFS was an inappropriate primary endpoint.

The SAP must be determined and specified prior to data lock which in this case was October 5, 2020. So, before this date, the SAP to be used to evaluate the trial was specified and mOS was designated as the primary endpoint. This is a much more meaningful endpoint than mPFS in almost all cancer trials and unquestionably is almost always the appropriate endpoint for immunotherapy drugs like DCVax-L.

External Controls Are Designated as the Control Arm

There remained another issue with the original phase 3 trial design that needed to be addressed before the SAP was finalized. There were 331 patients enrolled in the trial; they were randomized 2:1 to DCVax-L plus SOC in one arm and SOC plus a placebo made to look like DCVax-L in the other arm. Patients randomized to SOC were allowed to cross over to DCVax-L if their cancer progressed. In order to retain blinding in the trial, patients who progressed on DCVax-L were also offered DCVax-L even though they actually were already on the drug. This was necessary to maintain blinding of patients and physicians. Overall, there was a smaller number than 99 patients started on SOC who received only SOC. We won’t know this number until the trial is unblinded.

The aim of all clinical trials is to show that a new drug meaningfully improves outcomes when compared with the control arm which is usually current SOC. With an mPFS endpoint, the statistical analysis in the DCVax-L trial would have been straightforward in comparing 232 DCVax-L plus SOC patients versus 99 on SOC plus placebo. However, the analysis for mOS was more complex owing to the cross over design. Due to the crossover, the control arm was depleted from 99 to a substantially lower number which could not provide statistical power allowing for comparison of DCVax-L plus SOC to SOC plus placebo.

Northwest opted for a different approach. As a control group against which to compare the results of patients treated with DCVax-L added to SOC, they decided to use patients treated with SOC from several concurrent ndGBM trials that were conducted in the same timeframe as the DCVax-L trial. These are referred to as external controls. In the aggregate, when I looked at concurrent trials, I found external controls that could amount to as many as 1,200 to 1,500 patients which could be compared to the roughly 232 patients who were randomized to DCVax-L plus SOC.

Short sellers have argued that the FDA would never allow one trial to compare results to those from others so that the DCVax-L trial results would be rejected outright. Several years ago many people, including me, would have agreed with this point of view. However, times have changed. Beginning in 2017, the first CAR-T drugs were approved for certain hematological cancers on the basis of single armed phase 2 studies that compared them to SOC results from other trials. Just as I was finishing this report, there was a very positive development. The FDA publicly stated that external controls could be used in clinical trials; Northwest could not have wished for more. I will discuss this shortly.

So, here is how the DCVax-L trial results in ndGBM will be analyzed. The roughly 232 patients who received DCVax-L plus SOC will be compared with SOC results of somewhere around 1,200 to 1,500 patients drawn from other drug trials in ndGBM, that were conducted roughly concurrently with the DCVax-L trial. The primary endpoint is mOS. There likely will also be examination of numerous secondary endpoints such as mPFS, methylation status, age, degree of resection, IDH status etc.

The Trial is Also Designed to Determine Efficacy in rGBM

The design allowed patients in the phase 3 trial randomized to SOC to cross over to DCVax-L if their cancer progressed. This forms the basis for determining the potential efficacy of DCVax-L in rGBM as well as ndGBM. Patients who progressed on SOC are obviously experiencing cancer progression. Until the trial is unblinded, we won’t know how many patients met this criteria, but per the previous discussion, this might have been a significant percentage of the 99 or so patients. Importantly, the design of the trial did not allow NWBO management to know if a patient who progressed was on DCVax-L or SOC thus retaining blinding for the rGBM statistical analysis.

Associated with the phase 3 trial, there were also 51 patients whose cancer progressed before they could be randomized into the trial; these are rGBM patients. NWBO decided to treat these patients with the same drug regimen employed in the phase 3 trial. These patients can’t be included in the analysis of efficacy in rGBM, but the FDA will likely want to see the results to determine if they are supportive. Likewise, for the 50 or so compassionate use patients treated in the UK.

Hence, the trial is designed with the goal of gaining approval not only in ndGBM, but also rGBM. Results of DCVax-L in rGBM will be gauged against a number of contemporary trials carefully selected to  provide a patient  population that matches the metrics of the DCVax-L trial. Note that no drug has ever been shown to prolong survival in rGBM.

Will the FDA and Other Regulatory Agencies Accept This SAP?

Short sellers have ardently maintained that there is no way that regulators will accept this SAP. They have argued that the regulatory agencies will interpret the selection of mOS as the primary endpoint rather than mPFS as data dredging and accordingly will reject the trial without even considering the results as submitted by NWBO. But wait a minute, NWBO established mOS as the primary endpoint before the data lock in October 2020, without seeing unblinded results from the phase 3 trial. Some short sellers have disingenuously claimed that NWBO was able to cheat by peeking at the data in the ongoing trial. This is blatantly wrong as NWBO has publicly made clear that it never had access to trial data which was collected, stored and handled by independent CROs. It is also important to emphasize that mOS is a much, much better and harder endpoint than mPFS. The difference between life and death cannot be misinterpreted or statistically manipulated. I can’t see how any objective observer can argue that the FDA would have any issue with mOS as the primary endpoint.

Short sellers also maintained that the FDA would not accept external controls. The next two paragraphs were written before 15 key senior officials of the FDA (last week) authored a manuscript in the Annals of Oncology (a highly regarded and  innovative peer reviewed medical journal in Europe) stating that  they are now open to the use of external controls in measuring trial outcomes in some cases and that this could be a preferred approach in some circumstances. Prior to this, it was uncertain how the FDA officials would regard the use of external controls. This publication reads like a roadmap on how and why to use external controls. Coming from many of the key leaders at the FDA, this clearly answers the question on potential use of external controls in the affirmative and is a huge win for NWBO. Nevertheless, I decided to leave in the next paragraphs as originally written for perspective as to why hedge funds argued that the FDA would not accept external controls and therefore that the trial was doomed to failure. Consider this argument to have been significantly weakened.

Before the FDA authored manuscript, I had written that the use of mOS as a primary endpoint and external controls made sense other than for hedge funds who have complied massive short positions. But was there any evidence that regulatory agencies would accept the use of mOS and external controls? There was one clear example that they would. After receiving the SAP, the regulatory agency in the United Kingdom (MHRA) published on its website that the endpoints of the DCVax-L trial were mOS (for both ndGBM and rGBM patients) with the control group being external controls. Until last week, there was no clear public indication as to how other regulators like the FDA, the EU’s European Medicines Agency or Canada’s Health Canada would view these aspects of the SAP.

Having the MHRA agree on the SAP and grant a GMP license to produce DCVax-L under a compassionate use program (called specials in the UK) was certainly encouraging. So it seemed likely that the SAP plan accepted by the MHRA would also be accepted by other agencies. Still, there was no clear public indication from the FDA, EMA and Health Canada on how those regulatory agencies would view the use of external controls. However, based on the attitude of the FDA, Health Canada and EMA in approving the CAR-T drugs, it seemed probable that they would. The publication by much of the FDA leadership clearly shows a willingness to accept external controls and gives a roadmap of how to do it. With its publication in a major peer reviewed European medical journal, it removes uncertainty about the use of external controls.

Here is a link to my report that discusses the statement by the FDA

Will The Phase 3 Trial of DCVax-L Be Successful?

Now let’s turn to the critical question of whether the trial will be successful. I have written extensively that analysis of blinded data from the phase 3 trial strongly suggests that DCVax-L is having a meaningful effect on survival. We know that the two distinguished lead investigators for the trial, Dr. Linda Liau (in the US) and Professor Keyoumers Ashkan (in the UK and Europe) have publicly stated that patients in the phase 3 trial seemed to be living meaningfully longer than expected. This observation was based on blinded data.

Professor Ashkan may also have been influenced by his extensive experience with compassionate use in the UK. I am guessing that 50 patients, more or less, have been treated with DCVax-L on a compassionate use basis in the UK in recent years by Professor Ashkan and colleagues. Dr. Liau also had a meaningful and quite positive experience with about 15 patients treated with DCVax-L in its phase 2 trial.

On October 5, 2020, Northwest announced that it had locked the data base for the phase 3 trial. It was my expectation that the results would be announced by year end 2020 or early 2021. Here it is, early 2022, and no results have been announced. What is going on? Short sellers have alleged that management has seen that the trial has failed to reach its endpoint and is withholding the data from investors in order to unload their stock holdings. This is utterly absurd and the delay actually increases my confidence that the trial was successful.

Top management may or may not be still blinded to the data at this point in time, but it is almost certain that at least one member of management or an outside member of the statistical analysis team is unblinded. If the trial had failed, they would have been obligated to inform top management who would then have the legal obligation to inform investors that the trial had failed to reach its primary endpoints as outlined on the UK website. Failure to announce this material information publicly would have been securities fraud. This is strongly suggestive to me that the trial did achieve statistical significance for mOS in both ndGBM and rGBM patients.

Why the Delay in Releasing Topline Data?

Why then the interminable delay in releasing topline data. I think that there are several things that we can point to. First of all, this is a very small company with only a few employees to supervise a gigantic data gathering and analysis operation by numerous independent CROs and service firms. This was made difficult by the long duration of the trial so that the data was not always immediately available. It was also done in the time of COVID when many of the health care professionals involved in the trial were locked down outside of their labs or offices or were too busy treating COVID patients to focus on checking data points from a many years long clinical trial. Also, the many physicians participating in the trial had to affirm via affidavit that they agree with the conclusions drawn by NWBO about the phase 3 trial. This must have caused innumerable Zoom calls and paperwork.

You might respond OK, I can understand those reasons but why not just go ahead and release the topline data and publish it later in a medical journal. The reason is that it is extremely, extremely important to have the data published first in a peer reviewed journal. A stamp of approval from such a respected outside medical source is essential and customary practice in order to gain credibility and broad acceptance of the results by the physician community. Importantly, most medical journals would almost certainly refuse to publish study results if they have already been publicly announced or presented.

The First Approval Could Come in the UK

As I previously discussed, the UK seems to have accepted the SAP which uses as a control group patients treated with SOC from concurrent ndGBM trials (external controls) and uses mOS as the primary endpoint. My analysis of blinded data is strongly suggestive that the trial will succeed. And to repeat, I believe that if the trial had failed to reach the specified endpoints of the phase 3 trial that management would have already disclosed the failure. So, I conclude that the DCVax-L clinical trial met the primary endpoints as laid out on the MHRA website and that the use of external controls is acceptable.

I speculate then, that DCVax-L will first be approved in the UK. As previously mentioned, there has been a compassionate use program going on in King’s College for several years led by Professor Ashkan, the clinical lead for neuro-oncology at King's College Hospital in the UK. He has an active research interest in brain tumors and movement disorders and heads the Neuroscience Clinical Trial Unit. Professor Ashkan is one of the most respected neurosurgeons in the UK and Kings College is the premier teaching hospital in the UK. He is a respected and listened to key opinion leader for the MHRA.

Professor Ashkan believes that immunotherapy is the way forward in the treatment of brain tumors. He has publicly stated that DCVax-L, as a personalized immune therapy, may be an important advance in the treatment of brain tumors. He says that there is nothing cleverer than the ability of the immune system to cope with variation. He has further stated that brain tumors are notoriously hard to treat because they are extremely heterogeneous. Because of rapid mutations that characterize GBM, no tumor in one individual is the same as a tumor in another. Indeed, the mutations in one part of a tumor in the same person may may be very different from some other part. He is very hopeful about DCVax-L because it picks up the several antigens specific to each patient’s tumor. It them primes the immune system to attack the numerous antigens characterized by these different mutations.

As one of the most respected neurosurgeons in the UK, regulators are likely aware of Professor Ashkan’s conviction and his enthusiasm for results seen in the blinded phase 3 trial and compassionate use patients. As previously mentioned, there may have been 50 or so patients treated in the UK on a compassionate use basis. I presume that results in compassionate use were encouraging, otherwise it would have been shut down. This might have resulted in familiarity that predisposes regulators to a favorable view of the drug. Of course, I have no way to assess if this is correct.

All of this leads me to believe that DCVax-L will likely be approved first in the UK if trial results are positive. Very importantly, the UK has made a conscious effort to lead the world in expediting the approval of breakthrough drugs as they were the first to approve Pfizer/ BioNTech’s COVID-19  vaccine Comirnaty, and just recently Merck’s molnupiravir, the promising COVID oral therapeutic. DCVax-L could be viewed in the same light.

Will DCVax-L Become Part of SOC for Glioblastoma Multiforme

The publishing of phase 3 clinical trial results in a peer reviewed medical journal will be critically important in determining if physicians, regulatory agencies and investors will come to view DCVax-L as an important new component of SOC in ndGBM. Consider these points.

• Will the topline data be compelling and the SAP plan accepted by the key opinion leaders who are selected to review a manuscript submitted by NWBO to assess that conclusions drawn from the trial by NWBO and participating physicians are correct and warrant publication in a peer reviewed, major medical journal? This would validate the results of the phase 3 study for practicing physicians by highly knowledgeable, key opinion leaders with no ties to Northwest. I think we should know the answer to this critical question in the not too distant future.
• Will regulatory authorities then approve DCVax-L on the basis of the phase 3 trial? It is difficult to imagine, although not impossible, that the results of a clinical trial published and presented as successful in a peer reviewed, major medical journal would then be viewed unfavorably by regulators. So, publication should give investors strong conviction that it will be approved by regulators.

DCVax-L was studied as first line therapy in ndGBM and the company is in a position to also seek approval in rGBM. With approval, I think that there would be a rapid uptake for each indication. There has been no new drug for ndGBM that has been shown to extend survival since temozolomide in 2005 and there is no established SOC for rGBM. Note that the Optune medical device was approved as an addition to SOC for rGBM in 2011 and for ndGBM in 2015.

GBM is a very aggressive cancer that grows in the confined space of the skull squeezing the brain and leading to many dangerous medical conditions so doctors must move quickly to treatment after diagnosis. The mOS for ndGBM is roughly 17 months from diagnosis; half of patients have died at this point. Only about 5% of patients survive for five years.

It remains to be seen how much DCVax-L might increase survival. I would note that the Stupp trial in 2005 showed that adding the chemotherapy drug temozolomide to the then SOC in ndGBM of surgical resection followed by radiation increased mOS by 2.5 months. Temozolomide then became part of SOC. I hypothesize that the DCVax-L trial was successful, but how much so? If it improves mOS by 2.5 months, I would guess that oncologists would consider this to be a meaningful advance. Generally, a 4.0 months improvement in mOS for an aggressive cancer like ndGBM is considered a major advance. I think that oncologists would be enthusiastic about using DCVax-L in most patients if it improves mOS by 3 to 4 months.

Also important in the decision to use DCVax-L is that its side effects are remarkably benign, typically being mild fevers and chills and some modest injection site reactions. These are treatable with Tylenol and antihistamines. This is a unique and very favorable aspect of DCVax-L as almost all other cancer drugs have highly toxic side effects. Uptake of DCVax-L would not be hindered by concern about its side effect profile. Equally important, administration is quite simple, requiring an easily administered intradermal injection. These arguments suggest that it would be quickly taken up across a broad spectrum of ndGBMs and rGBMs.

The Survival Tail Could Be the Most Important Data from the Trial

Northwest Biotherapeutics began the phase 3 trial of DCVax-L in ndGBM over 14 years ago.  New enrollment was temporarily put on hold from 2008 through 2011 due to the inability to access funding and then restarted. Median enrollment was reached in May 2014 and the last patient was enrolled in November 2015. This means that for the last patient treated in the trial there will be at least five years of data and for the median patient treated there will be nearly six years. Remember that with SOC, only 5% of patients survive for five years so we will get a firm indication of the extent to which it may extend survival perhaps never seen before in a cancer trial. This actual survival data for each patient could be more meaningful in judging the trial than the traditional endpoint of mOS.

The era of immune therapy came to the fore of cancer treatment in 2014 with the introduction of the checkpoint inhibitor Yervoy which was quickly followed by Opdivo and then Keytruda. As experience grew with the checkpoint inhibitors, the medical community became excited about one key aspect of these drugs. It became evident that a small, but significant percentage of patients experienced much longer survival than those treated with chemotherapy. This has come to be known as the survival tail and this perhaps has been the most important factor that has led to Keytruda achieving estimated sales of $17.1 billion and Opdivo $7.4 billion in 2021, some 5 years after their introduction.

My guess is that NWBO could have unblinded the trial as early as 2018 if it had decided to evaluate the trial on the basis of mOS. However, management recognized that the importance of the survival tail for checkpoint inhibitors was also likely to be true for DCVax-L (as is probably the case in all immune therapy) and that this could perhaps be the most critical aspect on which the medical community would evaluate DCVax-L. They made the very difficult but absolutely correct decision to not unblind the DCVax-L trial and let it run for another two years in order to determine if there is a survival tail.

So what is an example of a survival tail? Let’s look at the data from  Bristol-Myers Squibb’s CheckMate-017 and CheckMate-057 phase 3 trials in which Opdivo was tested as a single therapy in treating second line non-small cell lung cancer in comparison to docetaxel, then the SOC. They reported pooled results for the two trials which showed that 13.4% of patients treated with Opdivo were alive at five years as compared to 2.6% for docetaxel. This means that 13.4 out of every 100 patients treated with Opdivo could be expected to be alive at five years versus 2.6 patients out of every 100 patients for docetaxel. This difference of 10.8 is the survival tail.

Obviously, second line non-small cell lung cancer and ndGBM are very different cancers so why discuss these Opdivo results? My reason is that they are comparably aggressive cancers. We can see from the CheckMate trials data that about 3% of patients with non-small cell lung cancer that were previously treated with chemotherapy live for five years. There is less precise five year survival data for ndGBM treated with chemotherapy, but data from other GBM trials and the general opinion of most key opinion leaders indicates that the five year survival is 5% or less. It is my hypothesis based on blinded data that DCVax-L possibly could show five year survival around 15%. This would result in a survival tail delta of 10, comparable to Opdivo in second line non-small cell lung cancer.

See this report for more detail.

Development of the Sawston Facility to Meet Commercial Needs

The Sawston facility is being developed in phases, both to optimize the timing of capital requirements and to enable each phase to be state of the art when developed because manufacturing technologies are advancing rapidly in the field of cell therapies. To date, the Company has developed Phase 1a of the facility, utilizing approximately 4,400 square feet of the overall 88,345 square feet in the plant. The Company anticipates that Phase 1a alone will be able to manufacture DCVax-L products for 45-50 patients per month. Because of the need for plant aintenance, this might be 500 or so patients per year. This is a significant increase from the current manufacturing capacity of four to five patients per month which, to date, has been taking place using a GMP clean room facility in London, UK.

If demand expands rapidly following approval of DCVax-L as I think possible, the Sawston facility can be significantly enlarged as only 5% of available space will be initially used. Not many details have been released but NWBO has publicly expressed excitement with the progress it is making with its Flashworks acquisition which goes a long way toward automating what is now a highly labor intensive manufacturing process. So looking down the road, if demand explodes for DCVax-L, Sawston is probably capable of producing product for several thousand patients. It can supply the UK, US, Canada, EU and other world markets in the early years of commercialization.

What Might Be the Price of DCVax-L?

Let’s start with thinking about pricing by considering the CAR-T drugs. These are cell based immunotherapies like DCVax-L that were first introduced to the US market in late 2017 for the treatment of certain relapsed/refractory hematological cancers. These cancers are roughly as aggressive as GBM as measured by expected survival following SOC therapy. Hence, their pricing might be a good benchmark against which to measure the potential price of DCVax-L. The pricing of CAR-T drugs is about $300,000 or more for a single injection which constitutes the entire course of therapy. In addition, the cost of drugs used to condition the patient for CAR-T transplant and additional drugs and hospital stays frequently needed to manage side effects can in extreme cases bring the total cost of therapy to as much as $1 million. NWBO has not said anything about pricing, but I think they could reasonably ask for a comparable price of $300,000 for a course of therapy. Extremely importantly, payors would also not be bothered with the cost of other drugs and medical care needed to prepare patients for treatment or to treat side effects.

The dosing of DCVax-L is different than for the CAR-T drugs. In the clinical trial, treatment started with injections at days 0, 10 and 20 followed by booster injections at months 2, 4 and 8 in the first year. After year 1, treatments were twice per year or roughly months 12, 18, 24, 30 and so on. The pricing of DCVax-L is likely to begin with a meaningful upfront payment when treatment is initiated with three injections followed by a payment at the time of each subsequent injection. This should be received very well by payors because if the drug does not work, payment can be stopped. Compare this to CAR-T drugs which are given as one injection so that the cost to the payor is the same if the drug works or fails.

So, revenues would be realized over time starting with the initial treatment of three injections and then realizing incremental revenues with each additional injection. This makes for difficulty in modeling. Just for the sake of illustration, let’s assume that DCVax-L initially costs $75,000 for the first three injections. Then assume that each subsequent injection given over the next few years is each priced at $30,000. Here are the resultant calculations:

• All patients would likely receive the three initial injections at a price of $75,000.
• In the event that a patient progressed before month 2 and was taken off therapy, revenues for that patient would be $75,000.
• If the patient did not progress before month 2 but did before month 4 and was taken off therapy, the revenue per patient would be $75,000 plus $30,000 for one booster or $105,000.
• Similarly if the patient progressed after month 4 but before month 8, revenue per patient would be $75,000 plus $30,000 plus $30,000 for a total of $135,000.
• If the patient did not progress before month 8, the revenue per patient would be $75,000 plus $30,000 plus $30,000 plus $30,000 or a total of $165,000 in year 1.
• If the patient did not progress in year two the revenue per patient in year two would be $30,000 plus $30,000 or $60,000 for cumulative revenues over two years of $225,000.
• If the patient did not progress in year three the revenue in year 3 would be $30,000 plus $30,000 or $60,000 making for cumulative revenues over three years of $285,000.
• In the phase 3, all patients who progressed were given the option to receive DCVax-L whether they were already on DCVax-L or crossed over from the SOC arm. When the trial is unblinded, we might have an indication as to whether a patient being treated with DCVax-L might receive benefit by staying on the drug even if the disease progresses.

You can see that estimating revenues per patient can vary widely. In the event that the patient is taken off treatment after the initial three injections, revenues per patient would be $75,000. If the patient was successfully treated over three years the revenue per patient would be $285,000. There are numerous in between scenarios. These numbers are extremely arbitrary and must be taken with a grain of salt, but they are indicative of the possible pricing matrix for DCVax-L.

Estimated Worldwide Incidence of Newly Diagnosed Glioblastoma Multiforme

The annual incidence of ndGBM in the US is 3.21 cases per 1 million population which calculates out to 10,700 patients based on a population of 331 million. I don’t have an estimate of incidence for the UK, Canada and the EU (ex UK). However, let us assume the incidence of ndGBM is the same 3.21 per 1 million population for each region. Based on 68 million people in the UK, 38 million in Canada and 378 million in the EU (ex UK), the respective annual incidence of ndGBM would be 2,195 in the UK, 1,225 in Canada and 12,200 in the EU (ex UK) for a total annual incidence of 25,095. There are also sizable populations in China, Japan and many other regions of the world which could be addressed over time.

Addressable Markets As Measured in Sales for UK, US, Canada and the EU (ex UK)

In my discussion of pricing, I came up with a scenario in which NWBO would realize $75,000 if a patient failed therapy after three initial injections and $165,000 if the patient remained on therapy for one year. I want to emphasize that I have arbitrarily come up with these numbers. NWBO has not commented on their possible pricing strategy.

Based on these pricing estimates and the estimates of ndGBM incidence just discussed, the total addressable market for all annual, ndGBM patients in the US would be somewhere between $800 million ($75,000 times 10,700 patients) and $1.8 billion ($165,000 times 10,700). The low end of this range is based on the estimate that every possible patient tries DCVax-L and it fails in each case. The upper end implies that every possible patient tries DCVax-L and it works in each case. Obviously, neither of these numbers are remotely possible, but the actual number is somewhere in the range. The comparable range for the UK would be $165 million to $360 million; for Canada, $90 million to $200 million; and for the EU (ex UK) $900 million to $2.0 billion.

Just as a benchmark, I am going to suggest that the addressable market in the US for DCVax-L is halfway between $800 million and $1.8 billion or $1.3 billion. With similar reasoning, the addressable market in the UK would be $265 million, $145 million in Canada and $1.5 billion in the EU (ex UK). Obviously, DCVax-L is not going to capture 100% of the addressable market. However, the desperation in the ndGBM market for treatments that can prolong life would suggest that if DCVax-L becomes part of SOC, it would make a major penetration of the annual addressable market. Remember that DCVax-L would not be replacing any part of the current SOC treatment; it would be added on. The benign side effect profile and easy administration (an intradermal injection) would also meaningfully boost market penetration.

There could then be additional revenues for a successfully treated patient over several years as they receive additional injections. Treatment of rGBM could produce an additional revenues stream as could revenues from other world markets. Currently, about 50% of ndGBM patients treated with current SOC relapse within one year. This could be an additional and meaningful opportunity in rGBM in the early years of DCVax-L commercialization to the extent that current SOC continues be employed. I am not going to attempt to estimate the potential market opportunity for these additional uses. Nor, am I going to throw out annual revenue projections for DCVax-L. However, I hope that this gives you a sense that there is a huge commercial opportunity if the data warrants the addition of DCVax-L to SOC for ndGBM.

Potential of Dendritic Cell Vaccines Beyond GBM

In the event that DCVax-L gains approval for ndGBM, it validates the dendritic cell therapeutic vaccine technology. The mechanism of action in GBM is the same for all resected solid tumors. Remember that DCVax-L manufacturing requires tumor tissue obtained from surgical resection. So, there might be the potential for expansion of usage over a broad range of solid tumors that might have commercial potential several magnitudes more than GBM. This development would take place over years or decades.

 

 

 

 

 

 

 

 

Tagged as Northwest Biotherapeutics Inc. + Categorized as Company Reports, LinkedIn