Autolus Therapeutics plc (AUTL) Earnings Call Transcript & Summary

Good afternoon. My name is Jackson Lowrie, and I'm a member of the JPMorgan health care investment banking team. Welcome today, and it's my pleasure to introduce our next speaker, Christian Itin, from -- the CEO of Autolus. And just before I let Christian come up here, afterwards, we'll be having a breakout session in the Yorkshire room, which is just outside and then to the left. So now I'll hand over to Christian.

Well, thank you very much. Thank you very much for the invitation. Thanks to all of you for joining us at a late afternoon on day 3 of the JPMorgan conference. I salute your stamina. And I'm looking forward to, obviously, sharing with you what we believe will be an exciting year for the company. Obviously, we're a publicly listed company, so please have a look at our SEC filings for the risk factors of the business. We are currently active in a number of clinical programs. We have currently 4 product candidates in clinical development. We're developing in 4 hematological indications and 1 solid tumor indication. We have multiple upcoming milestones. As we look to the middle of the year, we're going to have updates on our lead program in adult ALL as well as updates on diffuse large B-cell lymphoma, and then towards the end of the year, on all our hematological programs, updates at the ASH conference in December. The way we actually create our product is using a modular approach to programming them and changing the properties of the patient's own T cells, and as you'll see at the very end of the presentation, we also use the same technologies to start moving towards an allogeneic type of feature as well. The portfolio is owned entirely by the company at this point. We also do have a very significant IP estate around the various technologies. At the end of September, we had $229 million in cash. Worldwide rights retained, as I mentioned. And a cash runway through the second half of '21. So a brief look at the pipeline. We have decided, as we started the company 5 years ago, to initially focus on B cell malignancies because it's a disease setting we collectively knew a lot about. We're active in the space for a long time with redirected T-cell therapies and that gave us a very nice set of insights that we then translated into our lead product candidates. You'll be hearing mostly in this presentation about our product, AUTO1, which is our lead program, which is moving now into its first pivotal study as well as AUTO3, which will reach a proof-of-concept stage at the middle of this year. We're also active in other hematological malignancies, including multiple myeloma, but very interestingly, also in T-cell lymphoma, which is a significantly underserved disease setting and we have here a unique approach to tackle this devastating disease. Now in the progression of the pipeline, we also have started working on the solid tumor side with the first experiences in the clinic targeting GD2, an antigen that is present on a set of diseases, including neuroblastoma, sarcoma as well as additional indications, including small cell lung cancer and melanoma. And then we're also active and starting to build out that portfolio into additional areas of solid tumors. So the way we actually do our programming in cell program is using protein modules that confer properties to our CAR T cells. It includes obviously the ability to redirect the C -- T cell to the cancer cell. We typically do that with a chimeric antigen receptor. They can come in, in various flavors in terms of the actual molecular design, and you'll see some of those examples as we go through. We have control elements that we're using in some of our programs and are focusing more going forward on enhancing modules that actually help the CAR T cells withstand a very hostile environment, either from the side of the tumor or the micro environment that the tumor resides in. And as mentioned before, we're starting to move also towards allogeneic modules that give us actually activities that address questions related to graft-versus-host disease, but also immune rejection. So when we think about the way and how we actually deliver the product, we do manufacture with a closed manufacturing system. This is, on the left-hand side, our facility in North London, the cell therapy catapult that we're operating in, and it is actually a picture of our suite where we currently manufacture for our clinical trials, both in Europe as well as for the U.S. On the right-hand side, there is a rendering of our facility that we're in the process of building together with Alexandria Real Estate in Rockville, Maryland for U.S. commercial supply. This facility, obviously, is a rendering. It's actually off the ground. The roof is on, it just doesn't look as green around as the rendering would suggest at this point. Moving to our lead program and introducing you first to the indication that we chose as our lead indication. We're working in acute lymphoblastic leukemia, which is a very aggressive form of leukemia. It's a very fast-developing disease. It's a devastating disease, particularly for elderly patients. The current standard of care starts out with chemotherapy regimen. These are combination chemotherapy regimens. You have an induction chemotherapy and then you have typically several type sequences of consolidation chemotherapy. We manage today to get somewhere in the range of 30% to 40% of the patients into long-term remission based on the chemotherapeutic approach. Many of these patients, however, do relapse. The next option is you again try to induce a response and then do actually deliver a transplant to these patients. Now the older you are, the more challenging it is to actually tolerate the transplant. And as you look at the various age groups, that mortality associated with the transplant can be significant. Still, most of the patients where you do a transplant continue to relapse. And you then actually are at a place where you receive either blinatumomab or inotuzumab. Blinatumomab is currently the standard of care in that setting. It's a redirected T cell therapy, where the sales are programmed from the outside using a bispecific T-cell engager. But still there, almost all patients do relapse. So in other words, we're buying time, but continuously, progressively, are treating these patients, but the time that we buy these patients is progressively shorter. And so this is the population that we're looking to actually treat with the current therapeutic option, in this case, AUTO1. Now when you look at the outlook of a patient that relapses, even on the first relapse in ALL, typically, your outlook is that within a year, only 20% of these patients are alive. So it is a truly fast-progressing disease. When you look at the patient numbers, there are about 3,000 patients between the U.S. and the top 5 European countries in that stage of the disease that are in medical need and treatment, or you translate that into a potential from a commercial perspective, it's about $1 billion opportunity all in. Also, as a point of reference, the final year of life of an adult ALL patient costs in the U.S. about $1 million with a certain outcome being death, just to give a bit of a reference point here of the significance and the challenge in this space. So what we designed here is a product that's different from the first generation of CAR T therapies that have entered the market. Actually it's designed to behave as physiologically as possible. The reason why we realized that, that mattered is that when we looked at the current products that are on the market or getting close to the market, they had a feature that actually got them basically stuck on the target cells once they deliver the kill and had a very difficult time to dissociate from the target cell. What that actually triggered was a very high degree of activation of the CAR T cell, a lot of cytokines that are being produced per kill and ultimately resulted in many of these patients experiencing high-grade cytokine release syndrome that, particularly when you're an elderly patient, is very challenging to tolerate. It also drove in a significant way neurotoxicity in these patients as well. So what we created here is a product that actually has different properties. What we see on the right-hand upper side is the binding properties of the product that we're using. The standard product used in the space is referenced here as a product called a reference FMC63, which is the binder present in all commercial products, which is characterized by the binder that can actually bind very quickly, but then does not let go. It's a true high-affinity binder and that is the cause and the reason why these CAR T cells are not physiological in their behavior and get stuck on the target cells and ultimately deliver these types of toxicities. Now what we designed is a product that could bind with the same speed, but then could also dissociate rapidly once the kill occurred. And with that binder, what we hypothesized is that we should get actually better -- a higher degree of kill. We should get cells that are less exhausted, produce less cytokines, but also could proliferate much more. So you would actually have more CAR T cells in the patients that could take on the leukemia. We managed to actually demonstrate that in a number of preclinical models. This is published In Nature medicine in August last year. And we tested them, first, in pediatric patients. Based on the positive experience in the pediatric patients, we then moved into a study in adult patients with ALL. What we first look at here is the behavior of the product in the adult patients, and one of the things that you want to see is you want to see that the product stays active and present in the patient for a long period of time. This is referred to as persistence. And what you see on the left-hand side is that these curves all go flat, which means that those CAR T cells persist, they remain in the body, they remain active over a long period of time. When we compare it to the product from Novartis, which is tested -- has been tested in kids, we actually get a more active product, we get more than double the CAR T cells formed in the body as they actually proliferate. And we got about twice as long half-life, which obviously helps a lot in terms of long-term presence [and pressure on the tumor. Now when we looked at -- in a swim plot at how patients fared, the initial patients at the bottom of the chart were manufactured with a conventional process similar to what was originally used at University of Pennsylvania and also still used for Kymriah commercially. And you see there's a variable outcome that we did see, but we got 4 out of the 6 patients into a molecular remission. Now that molecular remission didn't last that long with the exception of a single patient that got transplanted, which is the third from the bottom. We then transitioned on to the commercial manufacturing process, which is the enclosed process on the machine-based setup that the picture referenced, that you saw before. And now you see that all patients managed to achieve a molecular complete remission. And most of these patients remained in molecular complete remission throughout the observation period. We have now, obviously, patients that have crossed a year in molecular remission. And remember, these patients, you'd expect only 20% to be alive. These are patients that have no sign whatsoever of a presence of a leukemia cell. So this is a very unusual finding. When we put this in perspective and put it against a current product that is used as the standard of care, Blincyto or blinatumomab, we are about twice as active on the complete remission rate. We have about 87% overall. In the closed process or commercial manufacturing process patients, everybody had a -- CR. If we look at the inventory survival at 6 months, Blincyto is at 31%. Our product overall in the blended analysis is at 68%. The closed process got to 100%. So this is at least twice as active, clinically active, as Blincyto or blinatumomab in patients that actually 60% of the patients in our trial had failed already blinatumomab or Blincyto. So in other words, this is in a worse population, we're twice as active as Blincyto was. The neurotox level is comparable. And in fact, overall, the overall adverse event profile is very comparable to Blincyto. It's important to also know that Blincyto is actually, given in an outpatient setting, it's an outpatient therapy and having a similar profile gives us a lot of confidence that we have an attractive product, not only from an activity perspective, but also from a patient management and handling perspective. Now when we compare the data to our pediatric experience, we actually see that the experience we had in the pediatric study, which is the CARPALL study and the experience that we had in the adult study, which is the ALLCAR study are actually highly comparable. We have the same level of CR rate, we have the same level of event-free survival of around 70% at 6 months, we have no high-grade cytokine release syndrome and we have a comparable level of neurotox. Now cytokine release at Grade 3 or 4 requires you to be not only in a hospital, but you actually end up in an ICU and many of these patients may actually require also ventilation, et cetera, to be managed. So avoiding that type of adverse event has a huge impact in terms of the management of the patients. Now the confidence we take from this data -- from the pediatric data is significant because what we have seen in the pediatric patients is that, of the patients that responded, half of them actually are in long-term remission. We have now a follow-up with up to 3 years in molecular complete remission. And we could still, at that point, see the CAR T cells in circulation by FACS analysis. So it gives us an awful lot of confidence around the product and tells us that we have a very attractive proposition here. Now where we are with the program is that we're moving into its first pivotal study in adult patients with ALL. We're looking to treat approximately 100 patients. These are patients that have undergone the current standard of care and failed, the same population that you've seen in the previous study. It's a single-arm study. The primary endpoint is CR rate. Secondary endpoints include molecular complete remissions, event-free survival and duration of response. We expect that this study will start enrolling in the U.S. in the second quarter and will complete enrollment first half of next year. And have complete data with follow-up at the end of '21, and we're also targeting the BLA filing at that point in time. We're also active on the pediatric side, and we'll have a second-generation program that addresses antigen loss, which was the key driver of relapse in the kids, which is going back into the clinic during the course of this year. And as we go through the course of next year, we'll make a decision whether it's AUTO1 or its next-gen brother program that will actually go into a late-stage study. So with that, I'd like to switch gears and talk about diffuse large B-cell lymphoma. This is an indication that is at least 3x as big as adult ALL. In the back end of the disease, we're talking about 10,000 patients at the highest -- in the highest medical need group in that disease setting. We have 2 CAR T products approved. They show long-term benefits in the range of 30% to 40%, depending on the program. Now what we have designed here is a program that addresses the 2 key routes of escape that we have seen with the initial products: one is loss of antigen, so basically the tumor makes itself invisible by losing CD19 expression; and the second is upregulation of checkpoint. And so our product actually has a dual targeting approach. Each cell has 2 independent firing receptors: 1 recognizing CD19; the other one, CD22. And we also give a single shot of pembrolizumab, a PD-1 monoclonal, to actually counteract the checkpoint upregulation. So what was surprising and was actually a true surprise to us as we went through the data was the safety profile for the product. We did know that we should have a good profile with regards to cytokine release. And indeed, no patient had actually high-grade of cytokine release syndrome. When we looked at the patients that actually were treated in the presence of a single or 3 doses of pembro, all of those patients were free of neurotoxicity. They were not only free of high-grade neurotoxicity, they were free of neurotoxicity of any grade. That is unusual. There is no CD19 program to date that has ever shown that. And I have seen with my old program and my old team as in back in 2002, the first neurotox, it has basically followed us all the way through. We've never seen something like that. So the remarkable thing is, we have a product now that has no high-grade CRS, no neurotox. When you think about the use of this type of a product, it is exceptionally well-suited for an outpatient setting. That matters in this disease setting because DLBCL patients, different from ALL patients, tend actually to be treated mostly in the peripheral oncology settings, but not in the high -- in the top medical schools or university settings where you typically find the treatment as an example for ALL patients. Now what is very encouraging is that we were able to induce complete remissions at every dose level that we have treated to date. We are currently at 450 million cell dose. We have also not seen any form of early relapse of CR. That is different. The current programs that are in development or on the market typically see about 1/3 of the CRs relapse within 3 months. We haven't seen that. Now what we're doing now is we're actually expanding the cohort at 450 million cell dose to understand what is the overall CR rate that we can actually induce. We believe that based on the safety profile that we've seen, that way the CR rate that is sustainable at the level of about 50%, we have a highly attractive program for this disease setting. To give you a sense that this is not only a consequence of just having patients that did very well and didn't have much tumor load, I would like to show you 2 examples. The first on the left-hand side is a patient that had a massive tumor load on the collar bone and also actually on the upper part of the lung, in the lymph nodes that are surrounding the upper part of lung. Now this particular patient was refractory to every line of therapy. That patient never responded to any therapy. But the patient responded to AUTO3. The patient had no CRS, had no neurotox, achieved a complete remission and is in complete remission now more than 18 months. The second patient on the right-hand side had tumor load actually in the lower abdomen and that was completely cleared. This patient again had 8 lines of prior therapy. So this is a truly end-of-the-line patient, had a grade 1 CRS, no neurotox and is in remission for more than 9 months at this point. So it is truly a very significant profile that we're starting to see. And obviously, by middle of this year, we will understand the overall rate of complete remissions we can achieve, and at that point, we can make a decision whether we can move the program into pivotal stage. So with that, I'd like to switch gears and briefly introduce you to some of the additional programs we're working on. One of the interesting programs we're working on is in T-cell lymphoma, which is a very tough disease because the tumor actually stems from a mature T cell, and it looks in every way like a normal T cell. That makes it very difficult to target. We've come up with a very unique way of targeting using a very minute difference in the TCR, in the T cell receptor beta chain, which has a constant domain, and each cell chooses either one or the other form of their constant domain. And there is a tiny change in inversion of 2 amino acids at the end terminus. That can actually be seen by monoclonal antibody that we can work and have been able to work into a CAR approach. And we also then, through crystallization, solving the structure and then actually running through a lot of modeling work, created the corollary counterpart that could see the other sub form of the receptor. The lead program now is targeting [ code ] called TRBC1. This program is now in Phase I, and we expect early data -- clinical data at the end of this year at the ASH meeting. This is a complete unique program. There's nothing like it in the industry. Give you just a quick flavor of where we're headed on the solid tumor side. The first question you always have to ask with solid tumors is, do you have a therapeutic window? And it is a tricky question because many of the antigens that we know and are present on solid tumors are also present on normal tissue. And so you have to find a way to differentiate between the tumor and the normal tissue and avoid hitting it. Now in this case, the target we're going after is called GD2. It's a very well-characterized single-lipid source. It's a lipid structure that sits on the plasma membrane, highly upregulated in tumors and -- but also present at lower amounts in nerve fibers. And you can imagine hitting nerve fibers is a real problem. The current crop of GD2 antibodies that are on the market, when you infuse them and do some major pain syndrome, so the kids with neuroblastoma have first to receive a morphine drip before you can even deliver the monoclonal. So you can imagine going in there with a highly potent cell-killing approach could be detrimental. So what we designed is a product that can differentiate between the 2 forms or levels of antigen and should only have seen the ones on the tumor, that was the hypothesis. And so we did a very careful testing of the program. What we have been able to show at the highest dose level is that we can actually achieve partial responses without inducing any neurotoxicity in these children. So in other words, we managed to get a therapeutic window. What we now do is we basically are adding to the product an ability to withstand the difficulty in the environment and the defense mechanisms that the tumor might use. So we start with already what worked in the clinic and we add to that. So we're adding modules that render the cells insensitive to checkpoints, to TGF-beta and give the cell a survival signal in the type of cytokine-like signal. All of these signals happen and work on the inside of the cell. There's no systemic toxicity actually that you will induce using these modules. It just changes the susceptibility of the CAR T cell to these defense or attack mechanisms of the tumor. Now we tested this in a very difficult model. We actually created a model that we designed so that the normal product, which was active in the clinic would stop working. And this is what you see on the top. In fact, what was in the clinic and worked had no impact on this tumor. When we then actually added the additional 3 modules, which is what's shown in the bottom, we can now actually completely clear the tumor and we go from a setting where the product stopped working to a setting where it now works again, all using and seeing the very same structures on the same tumor cells that before couldn't be tackled. Now I'd like to wrap up just to give you a quick sense of where we might be headed on the allogeneic side. The key topic here is you have to make sure that your CAR T cells, your allogeneic cells don't kill the host. That's a good idea. So you don't want graft-versus-host disease. And then you also want to make sure that the CAR T cells don't get killed by the immune system of the patient. And so there are 2 questions that you have to solve. I'll show you actually what we're doing currently with regards to graft-versus-host. What we're doing here is actually creating and using a module that disrupts the T cell receptor. In fact, it traps the T cell receptor in the secretory pathway and leads to degradation of the receptor. And what that does is that over a short period of time, these CAR T cells have no T cell receptor on their surface. And without that, they shouldn't be able to recognize any tissue and attack any tissue that is present in the host. You can see at the bottom that, indeed, all of the T cell receptors can be cleared. We then actually tested this in an animal model. This is a graft-versus-host disease model. And you can see in blue, the CAR T cells on their own, within 60 days, killed every mouse they were in. So you have graft-versus-host disease that actually completely kills the animals. However, if we actually eliminate, with this module, the T cell receptor and actually get it rid of it on the surface, now all of these mice survive. The way the technology we're using is the same way on how we program every other aspect also in our autologous approaches, which means we have a completely unique approach to do this and can take everything we learn from an autologous and apply it. The place where we see actual use of allogeneic is you have patients that don't have enough time to get product or patients that have so bad conditions with regards to their T cell compartment that you actually need an alternative way to give them some therapy even if that therapy [ won't ] be inferior to an autologous approach. So with that, I'd like to wrap up. There is a significant amount of news flow that we expect for the course of this year. Obviously, key updates, middle of this year as well as, obviously, at the end of the year. And obviously, from a fundamental perspective, moving our lead program into pivotal stage, which is a fundamental value step we're running through with this business. And with that, I'm happy to wrap up, and I'm happy to take questions in the breakout room. Thank you very much.