Autolus Therapeutics plc (AUTL) Earnings Call Transcript & Summary

Great. Welcome, everyone. Let's go ahead and get started. My name is Vikram Purohit. I am one of the research analysts with the Morgan Stanley biotech team, and I have on the line with me Christian Itin with Autolus Therapeutics. Christian, welcome. Thank you for joining us.

Thanks for having us.

Sure. And before we get into our conversation, I just need to read a brief disclosure statement. Please note that this webcast is for Morgan Stanley's clients and appropriate Morgan Stanley employees only. This webcast is not for members of the press. If you are a member of the press, please disconnect and reach out separately. For important disclosures, please see the Morgan Stanley research disclosure website at www.morganstanley.com/researchdisclosures. If you have any questions, please reach out to your Morgan Stanley sales representative. With that, let's go ahead and get started. Christian, thank you for joining us. Maybe to level set for those joining us on the line, who may not be fully familiar with the Autolus story, maybe you can walk through some opening remarks that give us a sense of some of the recent developments for the company's platform and where some of the key pipeline programs currently stand.

Happy to do that, and again thanks for having us at the conference. Autolus is working on CAR-T cells with a focus initially with our lead programs in the B-cell malignancies. What we're looking to do with every one of our programs is to tailor it to the specific disease setting, the specific challenges that the biology of the disease actually represents. And so our first program that we're moving or has been moving into pivotal stage is called AUTO1. It's a program that is designed for the treatment of patients with acute lymphoblastic leukemia. The second program is called AUTO3, designed very differently from a structural perspective and also what the key attributes are that we were designing into the product for the treatment of patients of -- with diffuse large B-cell lymphoma. And then behind that, we have a set of programs that are in early stage of clinical development, the T-cell lymphoma program called AUTO4; sister program to it, which is in late-stage preclinical called AUTO5; and a GD2 targeting program called AUTO6; and its sister program, AUTO6 NG, which is coming behind it, will be in the clinic during the course of next year for the treatment of neuroblastoma as well as of GD2+ tumors. So there's quite a range of products at that stage. There are further products in the pipeline, where we use programming technology that allows us to actually adapt the cells to the quite challenging microenvironment that many tumors actually create around themselves and are designed to actually make sure that the CAR-T products are resilient, have an ability to withstand those hostile environments and continue to be active. And there's quite a suite of technologies that are sort of used for that. But let me go back to the lead programs, just illustrate some of the thinking that has gone into those and into the design of them. Acute lymphoblastic leukemia, the disease setting that we chose is the first one to go after, is a particularly challenging disease. It's localized in the bone marrow. Disease crowds out the marrow and as a consequence of that actually has an enormous immunosuppressive effect because it sort of blocks the regeneration of a lot of the blood cells that we all rely on. So typically, patients in late-stage disease actually succumb to sepsis because they basically cannot cope from an immune system perspective with pretty much any type of infectious disease. So in order to sort of actually be able to capture that leukemia and be able to sort of eradicate it, you have to sort of bring together 3 properties that are actually very difficult to combine in a single product. The first is you need to have a very high level of clinical activity because the cells that are driving this leukemia are almost like stem cells but very close to stem cell in terms of their makeup. And they have an enormous proliferative potential. So you need a very, very high level of potency. The second aspect is because these cells have this high proliferative potential, you have to be able to put pressure on that compartment for a long period of time to make sure you've got every one of those cells and indeed have an ability then to convert this patient into long-term remission. That actually, when you think about the cell product, requires those cells to persist for a long period of time. So you need persistence that is in the range of 18 months, maybe longer, which is very difficult to engineer. And the third aspect is that the population that we're actually going in are the adult patients, and adult patients tend -- actually, many of them tend to be elderly and as a consequence of both the disease as well as the very intense high-dose chemotherapy regimens that they've been exposed to, tend to be actually very fragile. And the problem with that is that they have a hard time tolerating the high-intensity immunological toxicities that, in fact, CAR-T cells can induce. And so what you have to combine is a very high level of potency with very long persistence with a good safety profile. And that is what we designed into the product. And what we realized was that the way to get there was to actually create a product that behaves as physiologically as possible as close to a normal T cell as possible. And that's, in essence, what we designed into the receptor. The way we actually engage the target is quite different from the first-generation product. We do have a very transient engagement, highly specific but transient, and that allows us to actually get a lot of kills down with a very limited amount of cytokines being produced at the same time. And of course, in the aggregate, that means that the product actually has a limited cytokine release. And in fact, what we have seen both in pediatric patients as well as in adult patients is that we do not induce high-grade cytokine release syndrome typically referred to as grade 3 or higher, so a very unusual profile there. When we look in terms of the actual potency, we see that we have activity that are about twice the level of the activity that we see with blinatumomab, which is the standard care for relapsed/refractory adult ALL patients. And we can show that we at least double the molecular complete response rate in these patients in the range of above 80% versus about 28% for blinatumomab. We also show that, in fact, the event-free survival is doubled. We go from 31% with blina to 62% with our product. And we also have been able to show that we get there with an adverse event profile that is comparable to that of blinatumomab and obviously much more favorable compared to the first-generation CAR programs that have been tested in acute lymphoblastic leukemia. So that's the first program. So it's really tailored around a very specific set of quite challenging properties that have to be combined in a single product. We're now at the stage where we are moving the program in a pivotal study. We expect top line data at the end of next year. And what we have been able to show at EHA, where we gave the last update, is that we have now patients that have reached 18 months [ above ] complete remission without actually receiving a second transplant. So this is the activity of the CAR-T product of AUTO1 on its own that give us a long-term molecular remission in these patients. We're planning for additional update at ASH, which gives us pretty much an additional 6 months of observation on these patients. Now the second program in DLBCL called AUTO3 has to tackle quite a different set of challenges. DLBCL is obviously a disease that typically resides in the lymphatic tissue, but there's also infiltrating organs in later stage and can be present obviously across a lot of different stations across the patient's body. Very different from ALL. Typically, the DLBCL cells are not present in the marrow. And so when you actually infuse the CAR-T product, it has to actually first find the tumor, and they actually start lymphoma and then start to actually work on it. What we look to achieve in DLBCL is complete remissions. And what we did see is that, typically, those complete remissions actually, if they can be sustained, can actually lead to long-term benefit. Now when we looked at the first-generation products, again, what we've seen and what was the remarkable thing about those products is that they, for the first time, in third-line settings actually showed long-term remissions. But there was a difference between the maximum line to CRS they could induce versus the CRS that actually sustained over time. And somewhere between 1/4 to 1/3 of those CRS got lost actually over time. And when a group at Stanford, another group at MD Anderson just looked at, well, what is it about this lymphoma at the time of relapse that sort of allowed it to escape the CAR-T therapy? They found 2 key mechanisms. The first is that between 1/3 to half of the patients have lost CD19 antigens. In other words, the CAR-T cells lost their particular target they can actually hold on to and actually attack. The second mechanism that was in 2/3 of the samples actually was upregulation of PD-L1 checkpoint inhibition. Those were the primary 2 drivers that actually drove these lymphoma from a CR back into relapse because they allowed to actually circumvent the activity of the CAR-T cells. So what we designed with AUTO3 is actually 2 features into the program. One is we're using a dual targeting approach. So we have 2 receptors in -- types of receptors in each of the CAR-T cells. One is the CD19 targeting CAR, and the other one is the CD22 targeting CAR. Now the way we designed those receptors is different from what actually has gone into the clinic before. We use different binders to recognize the target antigens. We actually also use different ways of co-stimulating on the CD19 side. We have an OX40 signal. And we have a very different structure overall for the CAR on the CD22 side. So they're structurally very different. And what we basically screened for is we screened for products that in our preclinical work will give us a high level of activity without producing a high level of cytokines, again, to also actually not only get a good level of activity but also give us a good adverse event profile. What we then do now in -- from a clinical perspective is we actually, as we precondition the patients, we include a single dose of pembrolizumab. And obviously, the pembrolizumab allows us to counteract the upregulation of PD-L1 on the lymphoma and provides us with a window of approximately 3 months within which the CAR-T cells can do the job uninhibited by the checkpoint sitting or being expressed de novo or already preexisting on the lymphoma cells. And so using that approach, what we've been able to show is that we get somewhere in the range of about the recommended Phase II dose, somewhere in the range of 60% complete remissions. And when we look at the complete remissions that we have been able to induce, it looks like these remissions are persisting. We don't see early relapses, which is the hallmark that we've seen with all the first-generation product. So it gives us a very unique property, and obviously, the goal is to have a high level of sustained complete remissions; and secondly, to be able to get there without inducing a lot of toxicity. The reason why that matters is, is that the majority of DLBCL patients are not treated in academic centers at the centers of excellence, but they're treated actually in non-academic hospitals where they treated in specialized oncology work clinics. And in fact, in those settings, either inpatient or outpatient. In fact, 80% of the patients are treated in that segment. Only 20% would ever go to a center of excellence. And so good safety is important because that allows you to actually have a product that can actually be used across all settings of care and has an ability to reach to patients where they actually are. So those are kind of 2 examples of some of the properties that we engineer into our products, some of the thinking behind those products and obviously, now also the having the clinical data to actually show that, indeed, those desired properties actually are the properties that we actually now do see in our clinical trials in those patients. I'm happy to talk about sort of some of the technology aspects for newer programs maybe a bit later in the conversation.

Great. Thank you for the overview. So maybe let's start with AUTO1. Could you walk us through the design of the pivotal study for that program?

Right. So as I mentioned, this is a program that has a very significant level of activity and that allows us actually to run this program in a single-arm design with approximately 100 patients in the study. And what we're looking at is the complete remission rate, the CR rate in that study. As the primary end point, we're also going to be looking at event-free survival, duration of response and sort of the key additional end points to follow within this population on this trial. So it's a very straightforward design, similar to the trial that was designed for blinatumomab or has been designed in similar settings for CAR-T programs as well.

Understood. And how would you frame the commercial opportunity in the U.S.?

So when we think about the adult ALL space, we do have between the U.S. and the top 5 European countries about 3,000 patients in the last line setting. And that gives you sort of a size of the population, which is about 3x the size of the pediatric population in the last line setting. So it's the size of a population with obviously a very high medical need and absolutely need to get to transformational activity, which blinatumomab couldn't deliver. And obviously, none of the other therapeutic modalities today actually can deliver for these patients.

Understood. And understanding that it's relatively early days to be thinking about this. But how are you thinking about building commercial infrastructure in the U.S.? What do you envision being the team in terms of size and scope of the commercial team here?

Well, one of the interesting things about acute lymphoblastic leukemia is that, that [ concentrate ] to DLBCL is actually a disease that tends to be managed at centers of excellence. And the patients do get highly concentrated at centers of excellence. And that has a lot to do with the fact that these patients in last line setting have an enormous amount of adverse events related to the disease itself and do require very intense patient management. Patients tend to actually have episodes of sepsis, need access to ICUs, et cetera, to manage the disease. And so they tend to get actually highly concentrated in the academic centers, the transplant units. And that actually allows you, when you think about your commercial approach, actually to work with a very focused commercial footprint, a very focused approach and one where, obviously, you can operate with a relatively small commercial operation. Quite similarly, that also holds true for Europe. In Europe, the countries actually are organized in the way that for each country actually has -- tends to have a cooperative group for the management and treatment of patients with ALL, particularly the adult ALL population. And so there's a referral network that you have, again, to a relatively small number of transplant centers that do manage the patients and again, allows you to actually work also in Europe with a very focused commercial organization. So it is one of those settings where you can operate with a rather small focused footprint, obviously, very well tied into the respective centers but also tied into respective physician as well as patient organizations.

Understood. And how do you think about pricing at this point both in terms of kind of absolute price? And any sort of thought you might have given to different pricing models for the therapy?

Right. I think what's important is that this is a disease setting that, because of the significant amount of support that the patients need at the back end, the disease is challenging from that perspective in terms of the cost per se for that patient population. When you look at the cost of a patient at the final year of life with a certain outcome being death in the U.S., you're approaching about $1 million in terms of cost to manage that patient. So it is a very significant cost with obviously a very, very poor outcome. So it is a setting where, if you actually have a therapeutic approach that allows you to get at least a proportion of these patients into long-term remissions and cure that, that actually would be a very valuable therapeutic for that particular patient group. And in that sense, I think it is one of those places where, with the appropriate profile, you do have, I think, a very good ability to determine the value and actually also, frankly, have a fair value that is in a sizable range. So at this point, to sort of guide you in absolute, I think, is certainly not the right time to do that. It depends on the actual outcome that the study delivers and obviously, the impact it has, both for the patients as well as for the health care system in their respective country. But when we look across the board, you see the current ranges in terms of pricing. And there is also, obviously, particulars in the context of the pediatric ALL group, was also a very interesting case study done by NICE to sort of actually evaluate the value to the system, which, in fact, that particular value exceeded the price point that, in that example, Novartis was choosing for the product, Kymriah, that they have active in the space for pediatric patients. So I think there is a good ability there to actually articulate value, but we obviously want to do that with the actual final data and presented at that point.

Understood. Maybe we can now move on to AUTO3. So you gave us a recap of the construct in your opening remarks, but maybe we could talk a bit more about the outpatient cohort that has been initiated there. Talk a little bit about the rationale driving that decision and kind of what you think the significance of that data set could be.

Right. So first off, the observation with the program is and when we look at the patients that actually achieve a CR in the trial, we have not seen high-grade cytokine release syndrome in these patients achieving a CR, nor did we see signs of neurotoxicity. So it's a very unusual, very favorable adverse event profile. And when you look at the reasons for why the initial product sort of got confined to the inpatient segment, it was really about the requirement to sort of have intense patient management, particularly through the immunological adverse events that tend to occur early on or can occur early on in CAR-T therapy. And it is really that the intensity of patient management required to get those patients through that actually leads these therapies to be predominantly given in the inpatient academic centers and unfortunately, not where those patients actually, the majority -- the vast majority of those patients are normally treated. And that also actually is one of the key reasons why the sales numbers are where they are for the product. So what we want to do with the outpatient cohort is actually to -- actually provide answer to several questions. The first is we want to actually give the centers an ability to actually gain experience with the product in outpatient setting. Obviously, their current experience is with the first crop of CAR-T products and of course, there is a certain level of concern to go outpatients. So one of the key reasons for doing this now is really for the centers to gain experience ahead of us actually considering outpatients in a potential pivotal study. The second aspect is we want to understand what are the -- what is the adverse event profile, obviously, that we see in that setting, corroborate, obviously, the efficacy data. But then also, we want to understand what proportion of the patients get admitted and what is the reason for admission, how long do the patients actually have to be admitted and what is the intensity of patient management required upon admission. Those are key questions that we're obviously -- we'd like to get data for and to start to get a feel for the program and how it actually behaves and how we should think about in actually rolling this product out and making it available not only in the inpatient but also in the outpatient segment.

Understood. And when is the next update for AUTO3 and DLBCL expected?

So the next update is at the end of this week on Friday at ESMO. We'll provide an update on the experience with the recommended Phase II dose regimen. And the next update after that is going -- is planned for ASH. And obviously, for ASH, we would hope to have also the majority of the outpatient experience collected at that point so that there is a substantial update and substantive update over and above the ESMO data by that point in time. And of course, for all the patients that we already have treated, this is obviously a significant level of follow-up and a good understanding of the durability of the CRS that we can induce.

Understood. And before I go on, let me take a pause and see if we have any questions in the portal here. [Operator Instructions] Okay. I don't see any yet. So let's just keep moving on. Maybe we can pivot to manufacturing, talk a little bit about what your current capacity is and how you see manufacturing needs evolving over the coming months.

Right. So the manufacturing setup we're using is semi-automated, fully enclosed system, which allows us to actually, in a single room, operate a large number of products at the same time. So it gives us an ability to actually have economies of scale and obviously work with a system that is the commercial at the commercial level during our clinical trial conduct. At this point, we're operating at a facility north of London in the U.K., and it has a capacity that allows us to actually run our pivotal studies. We're in excess of 200 products a year can be produced in that particular facility at this point. We're expanding that. We're doubling that capacity, actually, which should sort of come online beginning of next year so that there is a very significant amount of clinical trial capacity available for our -- the conduct of our trials. As we go forward, what we expect to do is build out the capacity in the U.K. to support a limited -- to support launch for a limited period of time and then actually transition to a facility that we're building out in collaboration with Alexandria Real Estate in Rockville, Maryland, which will be the full-scale commercial facility that will carry the loan forward from there from a commercial perspective. That's the way we set it up, and that's the trajectory that we're taking.

Great. Understood. Now looking at some of the earlier programs that we didn't get a chance to discuss beyond AUTO1 and AUTO3. Maybe you could just briefly recap some of the key milestones expected for those programs throughout 2020 and 2021.

Right. So the program on the hematological side that we're working on in addition to the 2 lead programs is AUTO4 in T-cell lymphoma. That program, we expect to have data during the course of next year from our initial clinical experience. Obviously, it's a novel approach. It's a unique approach to targeting T-cell lymphomas and obviously a program we're very excited about. There is a sister program to that, which we're developing to enter the clinic during the course of next year. And that gives us a program for the treatment of T-cell lymphoma in combination with a companion diagnostic that allows us to actually select the patients for the respective program be it AUTO4 or AUTO5. When we look into the earlier pipeline, a next-gen version for -- on the -- building on the AUTO1 backbone for the treatment of pediatric patients with ALL, that program is moving into clinical development towards the end of this year. And when we then look further into -- onto the solid tumor side, the 2 programs that we're moving, are particularly excited about, the first one is the next-generation program for our GD2 targeting approach called AUTO6 NG. What we do there is we build on the clinical experience we gained with the initial AUTO6 program, where we actually could demonstrate clinical activity and a good safety profile with the particular chimeric antigen receptor we're using. And then we're adding to that additional protein modules that actually render the cells insensitive to checkpoints, insensitive to TGF-beta as well as actually providing a survival signal, which is a cytokine-like signal. Now it's important all of these signals act on the inside of the cell, so there's no systemic toxicity that builds up through that. And we're planning to get that program back into the clinic during the course of next year. So it's a very exciting program, very significant level of technology over and above of what we've been seeing and so far been putting into the clinic and we hope gives us the right set of properties to crack this particular solid tumor setting. In a substantially larger indication, we're working on with a program called AUTO7 where we use a similar approach. We use targeting on the antigen being PSMA. But then we're adding additional modules as we're using with the AUTO6 NG program, and one element we put on top of the ones we're using also with AUTO6 NG that give us resilience and that additional element that we're adding on top actually changes the immunological milieu within the tumor by having very low quantities of interleukin 12 that actually can be secreted into the tumor environment without actually developing systemic activity. So quite an exciting range products with very significant additional technology that we're moving forward. And then there's also the final program is a multi-myeloma program that we designed to address some of the key challenges that we've seen in that space. And that program is also expected to get back into the clinic early part of next year.

Okay. Great. Well, thank you. That's time for us, but thank you very much, Christian. Appreciate your taking the time to give us this update, and we'll talk to you soon. Everyone on the line, thank you for your time.

All right. Thanks a lot for having us.

Okay. Thank you. Buh-bye.

Buh-bye.