Market Data 
Markets
Replies to post #99348 on NorthWest Biotherapeutics Inc (NWBO)
3.1. OBJECTIVES
3.1.1. Primary objective
The primary objective of this study is to compare progression free survival (PFS) between patients in the DCVax-L cohort and patients in the placebo cohort in patients with no evidence of disease progression at baseline.
3.1.2. Secondary objective
The secondary objective of this study is to compare Overall Survival (OS) between patients in the DCVax-L group and patients in the placebo group in patients with no evidence of disease progression at baseline
DCVax-L. Both DCVax-L and the placebo are tested at the contracted manufacturer prior to release to the study site. Patients for whom sufficient DCVax-L was not
generated are not eligible to continue on this protocol.
All subjects who had a leukapheresis will undergo external beam radiation therapy (which may include intensity modulate radiation therapy or IMRT) and concurrent temozolomide chemotherapy as part of standard primary treatment, initiated as soon as possible, typically 3-4 weeks after surgery (Appendices A & B).
Two weeks after completion of radiation and concurrent chemotherapy treatment,
subjects will undergo the Baseline Visit, during which the final tests to determine eligibility are performed. Patients who do not have obvious evidence of progressive
disease at the Baseline Visit (as determined by MRI) are enrolled in the main arm of the study (intent to treat), and are randomized to receive DCVax-L in the treatment cohort or autologous MNC in the placebo cohort.
Randomization and treatment
assignment takes place within 1 week of the Baseline Visit. At the Baseline Visit, patients must be scheduled to return to the clinic 1 week later to receive their first immunization.
In the Phase III trial, PFS and survival times will be calculated from time of randomization, which is expected to occur approximately three months after initial surgery.
But there are two things to note about this patient.
1. She would not have made it to randomization in the Phase 3 trial.
2. Her "objective Response" was very likely pseudoprogression. Or, as Dr. Liau said in her paper:
One patient (patient 5) had near complete regression of residual tumor, which was seen on MRI 2 months after completion of peptide-pulsed dendritic cell vaccination and before any additional adjuvant treatment. Both the size of the areas of T2W hyperintensity and the contrast-enhancing tumor decreased in this patient (Fig. 2). Although this radiographic change is more likely related to a delayed response to radiation therapy, it is interesting to speculate that dendritic cell–based immunotherapy might have contributed to this clinical response,
Earlier I challenged MD1225 to figure out what all 4 still-living patients from the two earlier DCVax trials had in common. This is one of those 4 patients (the other 3 are from the second Phase I/II trial). For all 4 of these patients, their MRI got "worse" before it got "better". Think back to the Hoos videos recently discussed. What does that suggest for PFS? --- AVII
Pseudoprogression and Radiation Effects
Standard therapy for glioblastoma involves maximal safe tumor resection followed by radiotherapy with concurrent and adjuvant temozolomide.24,25 Twenty to 30% of patients undergoing their first postradiation MRI show increased contrast enhancement that eventually subsides without any change in therapy (Fig 2). This phenomenon, termed pseudoprogression, likely results from transiently increased permeability of the tumor vasculature from irradiation, which may be enhanced by temozolomide, and complicates the determination of tumor progression immediately after completion of radiotherapy.26–30 Pseudoprogression may be accompanied by progressive clinical signs and symptoms and seems to be more frequent in patients with a methylated MGMT gene promoter.30 This treatment-related effect has implications for patient management and may result in premature discontinuation of effective adjuvant therapy. This limits the validity of a PFS end point unless tissue-based confirmation of tumor progression is obtained. It also has significant implications for selecting appropriate patients for participation in clinical trials for recurrent gliomas. Failure to exclude patients with pseudoprogression from these studies will result in a falsely high response rate and PFS and the possibility that an agent will be incorrectly considered to be active. To address this issue, the proposed new response criteria suggest that within the first 12 weeks of completion of radiotherapy, when pseudoprogression is most prevalent, progression can only be determined if the majority of the new enhancement is outside of the radiation field (for example, beyond the high-dose region or 80% isodose line) or if there is pathologic confirmation of progressive disease (Table 2). It is recognized that the proposed histologic criteria have important limitations, but they provide guidance on the type of findings that are suggestive of progressive disease. For patients in whom pseudoprogression cannot be differentiated from true tumor progression, enrollment onto trials for recurrent gliomas should not be permitted. Patients who remain clinically stable and/or are suspected to have pseudoprogression based on metabolic or vascular imaging should continue with their current therapy.
Unlike the Macdonald Criteria, which do not take into account progressive nonenhancing disease, the new response assessment will consider enlarging areas of nonenhancing tumor as evidence of tumor progression (Tables 3 and 4). However, precise quantification of the increase in T2/FLAIR signal can be difficult and must be differentiated from other causes of increased T2 or FLAIR signal, such radiation effects, decreased corticosteroid dosing, demyelination, ischemic injury, infection, seizures, postoperative changes, or other treatment effects, before making a determination of progressive disease. Changes in T2/FLAIR signal that suggest infiltrating tumor include mass effect (as determined by sulcal effacement, ventricular compression, and thickening of the corpus callosum), infiltration of the cortical ribbon, and location outside of the radiation field. Although it would be preferable to have an objective measure of progressive nonenhancing recurrent disease similar to contrast-enhancing disease, the Response Assessment in Neuro-Oncology (RANO) Working Group felt that this was not possible at present given the limitations of current technology.
The initiation of these changes can be subtle, and convincing non–contrast-enhancing growth may require one or two confirmatory scans. If non enhancing progression is determined after retrospective review of images, the scan at which these changes were first detected should serve as the progression scan.
Thank you Renzo. It's a pleasure being here. I'm happy to make a contribution to this discussion. Actually, from somebody's perspective that has been more than a decade already on the topic of immunotherapy [inaudible 01:28:33] it's very nice to see how this is coming together, and how much alignment we already have around the issue, just from the conversation that happened this morning. I'm expecting there will be more of that as today progresses. What I'd like to do is, actually make the bridge now between the known, the conventional, endpoints. The known conventional endpoints that we we'll be talking about in the next session, tell you a bit about the history that we have had, tell you about the hypotheses we drew from the early data that told us we need new endpoints, look at the challenges for clinical trials, and how we might improve those endpoints, but not answering all the questions here because the next panel will hopefully do that, and then look at what's the ultimate goal that we're trying to achieve.
So starting that, I make the simple statement about cancer immunotherapy might be somewhat repetitive, but I think it's really important to be clear. The mechanism of action as [inaudible 01:29:30] what we do is chemotherapy a targeted agents is just not the same. It's indirect, we're targeting the immune system and not targeting the tumor. The immune response that results from that is dynamic. These are living cells that are doing things. They're not just a dying tumor, from a [inaudible 01:29:51] [inaudible 01:29:52] agent. Having said that, you achieve more than one affect when you activate the immune system. Now the consequences are there's no direct anti-tumor affect, and that might be delayed affect
Direct antitumor effect, and there might be delayed effects that actually show you biologic and pharmacodynamic activity that can be measured well before there is clinical activity and that come in all kinds of shapes and forms. The endpoints we're talking about, they're influenced by that. Having said this, this is a graph that you are all familiar with. It comes from Bob Schreiber and his group around immuno-editing. It basically makes a simple biological case for how the immune system interacts with the tumor. That can then translate to endpoints or observations you make in the clinic. What do we see here? You see the three Es of immuno-editing. Either if the immune system outperforms the tumor basically the immune response is strong, it can actually lead to elimination of the disease. If you look at an equilibrium, the immune system may keep the tumor in check but it is not eliminating it. That's a very common phenomenon with immunotherapy. The last item is of course escape. If the immune system loses the battle, then the tumor will progress and ultimately the patient may die.
If you overlay that on this biologic graph, how we actually translate this into clinical parameters, you may call elimination "response," equilibrium "disease stability," and ultimately progression be the "escape." Very simplistic way to look at it, but that's probably helping to understand the biology as we look at clinical observations.
Now what are those clinical observations? Let's start with the ipilimumab because that's where the story began. We still have a lot of the data that's being discussed, comes from the early ipilimumab observations.
We have seen some CAT scans already, so I'm not going to get into too much depth here. Larry Schwartz made a very good case for what you can observe radiologically. Here you have a case of a conventional response. There is a lesion at baseline that shrinks, and immunotherapy can do that particularly the PD-1 blockers do that very nicely.
There are other examples though. The delayed response type phenomena, or pseudo-progression, or flare, as we have named them, they can show that the tumor can get bigger before it shrinks. This is an exam that I like because it's actually not a CAT scan. It's a lesion on the back of a patient that gets very inflamed. Here you can see the quality of the lesion not just the size. The quality changes from an indurated nodule to something soft and spongy. There's detritus on the surface of the lesion. It suggests there's activity here. Something is happening to the lesion that probably is induced by the immune system. Ultimately the lesion shrinks and disappears. This patient has become a complete responder for quite a long period of time.
If you want to translate that into a radiograph, here is another example where you have a very massive metastatic lesion in the thigh of a patient that gets larger and then shrinks. Again another example of a delayed response. These are phenomena that we see and that we of course want to be able to capture systematically.
Here's another example. Not uncommon at all. Stable disease where over prolonged periods of time, the immune system may create that equilibrium with the tumor and keep the tumor in check, but the lesion remains.
Here's the last example. A new lesion appears while there's a response in existing lesion. The new lesion is indicated by an arrow here. It's a very small little lesion, but that lesion of course would under conventional criteria be progressive disease. It means treatments [inaudible 01:33:40] and then ultimately losing the opportunity for the patient to have more benefit.
If you take all of that and you put it into tumor volume graphs over time, that's what these four patterns look like. A conventional response in the left upper corner. A slowly declining tumor volume, so stable disease in a sense over long periods of time, in the right upper corner. A delayed response in the left lower corner. Tumor volume increases before it shrinks. Then, the appearance of a new lesion added to the total tumor volume in the right lower corner. If you look at the red line there, it actually doesn't start at the beginning. It starts somewhere at a follow up time point. The volume of that lesion, if you add it to the green line which is the volume of the lesions from baseline, leads to still a shrinkage of the total tumor volume or a reduction of the total tumor volume over time, meeting the criteria of a response, but now factoring in a new lesion. The new lesion itself, it may just be the immune system recognizing micrometatstatic tumor cells, an actually desired drug effect, or an initial growth of a tumor before it shrinks. In either case, if it's put together with the rest of the lesions, it can still define a response.
Here is then the question that everybody wants to answer: Are these new observations in any form or shape related to better outcome of the patient, meaning the gold standard endpoint of survival? This is retrospective data, so it was not prospectively defined because at the time of the ipilimumab program, when it started, we didn't know what we would expect or what we should expect. In other words, we had to do this retrospectively. Many new analyses have actually been done prospectively. It shows you here that the conventional responses, which are the top curve, have the best survival outcome. The progressive disease patients that are progressive using new criteria and old WHO criteria, they have the worst outcome. The line in the middle is new. That line represents the three new patterns that I've shown you. There's clearly a difference in survival outcome, and if that carries forward, it actually would make a case for why measuring these new patterns can be meaningful.
I'm adding now another graph just to show that there has actually been progress from the ipilimumab early days to today. This is from the pembrolizumab program that was shown at ASCO in 2014. It was one of the first graphs that came out after ipilimumab. It shows the same phenomenon. There is a variety of new patterns, and those new patterns do better on survival than that is the case with conventional progressive disease.
Then a few other observations that are worthwhile, and we somewhat heard that already from the FDA and actually from Larry Schwartz as well, is the kinetics that you see on survival and how do the non-survival endpoints response or progression-free survival ultimately relate to the survival outcome. We know historically from cytotoxic agents or from targeted therapies that there is often a dichotomy between the PFS and the OS endpoint. Often the hazard ratios for PFS or the benefit on PFS has been greater than that what you see for survival. With immunotherapy, we're making the opposite observation. Let's have a look at that.
This is ipilimumab. Here you see the late separation of curves. You see the plateau of the curve which shows long-term survival for a sub-group of patients. When you look at the response rate, it's 11%. Look at disease control, it's almost 30%. The long-term survival at two years as it's shown here that was one of the secondary endpoints measured, is actually 24%. There is a gap here between response and survival. If you add the stable disease patients into the mix, which disease control rate does, it helps to bridge the gap. The first indicator, you need to look at this slightly differently.
Here let's look at another example where it's not exactly the same. For the PD-1 blocking agents, we have seen that response can be high, not in every disease but pretty well across the board, and that PFS can be quite favorable. The PFS and the OS outcomes can be very similar. That uses standard criteria. The PF OS relationship is almost 1:1. We see that in a variety of settings. It's very different from ipilimumab.
Then of course, Renzo made this case already for Provenge or sipuleucel-T, that is a cell-based cancer vaccine, that has shown no benefit on PFS at all. Basically no responses, but a reasonable-sized benefit on survival. That falls more in the category of cancer vaccines where you historically have not seen responses. We're still grappling with how we make cancer vaccines work, but there is still a lot of interest in the field. If we now think at the spectrum of immunotherapy agents that are being investigated, they're not all checkpoint modulators. They're not all PD-1s. You might find within the checkpoint category, we're going from blocking agents like PD-1, PD-L1, to antagonizing agents like OX40, ICOS and other ones, and you see new dynamics. For us to be able to measure what these agents do is important. I believe for that you will need a more flexible tool than conventional response. For PD-1, you may not, but for the spectrum of immunotherapies we're dealing with, that flexibility might have to increase.
I'm drawing now the conclusions from the data that I've shown you. The unusual response patterns and delayed effects reflect the underlying biology of the immune system. This is something we thought at the time of ipilimumab, so six, seven years ago, novel patterns of response or detectable disease represent clinical benefit because they do actually correlate with better survival. Conventional chemotherapy-driven response or PFS does not account for all benefit patterns and therefore und-
... Forms of PFS does not account for all benefit patterns and therefore underestimates benefit. That might vary, based on what I just said, between different types of immunotherapy. Many evaluation of [inaudible 01:40:14] can be impacted by the late clinical effects due to non-proportion hazards, it may require modified statistics.
Think about this as something that's 6 years old. If you now take this forward, what are the clinical trial challenges that result from that, and only looking at old endpoints, not the new ones yet. Conventional overall response rate of PSF underestimates the effect and what could that lead to? It could prematurely discontinue a new therapy, especially if it's a therapy that doesn't deliver a higher response rate, but could otherwise potentially improve survival.
The potential clinical trial failure might occur if a conventional effect is insufficient to meet the study objectives. Let's say you want a 20 percent response rate, you get 10 percent conventional response and 20 percent unconventional response. Then, if you measure the unconventional response you might meet the objective. If you don't, you would not meet the objective. Of course this is widely variable, there's a variety of things we need to consider, but the concept makes sense. Having the tool that enables you to measure that should actually enable us to be better in immunotherapy investigation.
Premature treatment discontinuation due to conventional progression is a problem we've all faced with many patients, where you actually think, "This patient is benefiting from the therapy," but the measurement on the CAT SCAN tells me I actually have to stop treating now because it is conventional progression. That's something we need to overcome. We have started doing that. I'll speak to that in a second. That's from a clinical trial perspective and from a treater's perspective something very relevant for clinical practice.
Then there's another aspect here, when you're actually sequencing therapies. We're now in the era of combination therapy where this will become a much larger question. If you're sequencing therapies it could be difficult to actually discern the effect of a previous therapy, from that of a following therapy, depending upon how you determine progression in between.
Let's look at how can we improve these endpoints. There's a bit of history here, you've spent more than a decade to start this process so I started a timeline. Beginning in past century as [inaudible 01:42:43] clinical observations have been made, a lot of them, and published around unconventional and mixed-response observations. Some lesions get big, some lesions shrink. Very heterogeneous pictures. Nobody knew how to give credit to that, or what that means, ultimately it got dismissed. That was more than a decade of publications, mostly with cancer vaccines that hasn't really gotten much recognition. Now, what did we do? We started doing some workshops with the Cancer Immunotherapy Consortium and started describing the phenomena and get community buy-in to managing these new phenomena. One observation was the delayed separation of capillary markers. This was an analysis of all Phase 3 trials that have even been done with immunotherapy and were in a public domain. We did that in 2006. We came up with the conclusion, those that has a separation of curves, they have a delay. What does this mean? Do we actually need to do new trials differently. That was very important.
Then we came with the immuno[inaudible 01:43:47] response criteria that was actually done jointly between CIC and BMS using a [inaudible 01:43:53] that was published in 2009. From there we now actually have seen nice expansion on that activity going into new versions of the original IRRC. Going getting more nuanced, considering more data. They are labeled as IR-resist. The new version of I-resist, the IM-resist, and the IR-lugano. I just had to put that in here, which is for lymphoma. It was just published in blood and it uses the same basic principles, but it moves on from solid tumors to hematologic malignancies.
Then of course, besides advancing criteria, we're also getting more data. I hope the next session will actually show us a lot of that is a lot of new data that enables us to say there's some solidity to these observations.
Here we are today, hopefully this will propel us to the next level after having spent the better part of the decade trying to get to this point.
I say a few more things about the IR response criteria. As we heard from Larry Schwartz, I think he said it very well, IR response criteria were not designed to give a very, very specific way of managing this problem. They were concepts that you can apply in a variety of ways. We of course, in order to use them in a clinical trial, had to apply them in a very specific way, so we used WHO as the under-pinning system. The reason was the [inaudible 01:45:24] program had been using WHO criteria for all their trials. It made no sense to try and deviate and move to resist. We used WHO but we defined principles that could easily be translated to resist, and that has subsequently happened. Confirmation of progression was a component. Measuring new lesions was a component. Total tumor burden, including new lesions, that means measure the lesions and add them to the tumor volume. Treat post initial progression, not indefinitely, but at least that first time when progression occurs, allowed the patient to still benefit from therapy until you have really confirmed progression.
Then of course, if you take those components, you define response criteria, and apply them to the endpoints that we're interested in you would then create things like IR response, or IROR, [IIDSCA 01:46:14] including stable disease and IIPFS, which might actually become the endpoint theta we are the most interested in as we progress this story.
We were then asked by clinical cancer research when we published this the first time around, about 6 years later to give a progress update. That was published last year. It makes a few nice points. The first one is, we actually have seen transferability of IIRC concepts from WHO to resist. Almost everything we are talking about now is resist. The application f the IIRC has been done beyond melanoma. It's true most of the data was melanoma initially, but we have now seen it in lung cancer, we have now seen it in other diseases. The inclusion of the concept has been put into regulatory guidance documents for exploratory purposes so far, but that is useful. The expansion of IIRC into more concrete criteria has happened. That's what I've just listed for you. Then of course, we have seen a class effect. Once you apply this broadly, different immunotherapies can show these patterns. It's not just a [inaudible 01:47:19] it's not just a one-drug phenomena.
This is what I view as worth defining progress in the last 5-6 years. Here, if you look at the different adjustments of IIRC we have seen, IR-resist, I-resist, IM-resist, and IR-lugano, they have a lot of overlap. The original principles that I just named, they're all reflected in almost all of the criteria. There are a few nuances difference. The one that I would point out to is the IM-resist, is actually aiming to look at progression post the confirmed progression, or response post the confirmed progression, which is expanding our concept a little bit further because there are very late effects that you might sometimes see that would certainly not be captured by most of the variations that we see here.
I mentioned already in some regulatory guidances we have seen inclusion of the basic principles of IIRC. This is a cancer vaccine that the FDA had issued. This is also by now 5 years old. There is an EMA guidance which is in the process of being updated right now where the same principles have been included. We actually took a workshop of the CIC in 2014 that was 5 years after publishing the criteria to see how much further have we come. That was a bit premature compared to where we are today, but it already points us in the right direction. There is a class effect here across a variety of different immunotherapies and a class effect legitimizes the use of criteria much more so than if you look at a single agent.
I come to the survivor question. The point has been made already; if survivor gets better from any new therapy that is included in our [mememtario, 01:49:15] then the tail of the curve will go up, more patients will live longer, and any new therapy that you will introduce will have a much harder time to do a meaningful trial with a reasonable duration on a conventional survival endpoint.
We got to find a way to make this more palatable. Here is a slide that I just love to show. It's an [Epineveale, 01:49:35] Phase 1 follow-up showing a very high plateau of the survival curve in metastatic melanoma. This data still needs more follow-up and more Phase 3 extension, but it is exciting to see combinations can produce benefit and we are achieving this increase in survival for patients.
Now let's have a quick look at what are the challenges and solutions for survival, and you will hear a lot more about this when the statisticians speak later ...
And you will hear a lot more about this when the statisticians speak later. The challenge is obviously the non-proportion hazard. In order to solve that we need new statistical models, potentially in some instances it had been proposed to over- power studies. Because you can draw then from that extra power when you encounter the late effect. And that seems to dilute your original power. And then you can deal with sensitivity analysis. Which help understand things but not necessarily saving a primary endpoint in a trial.
AI
