Adaptimmune Therapeutics plc (ADAPY) Earnings Call Transcript & Summary

Hi. I'm Ad Rawcliffe, Adaptimmune's CEO. Before I start, I want to touch on 2 things. Firstly, I'll refer you to this disclaimer slide as our presentation today contains forward-looking statements. And secondly, I want to remind everyone we have a Q&A session at the end of this virtual event. You can use the Q&A function in your Zoom meeting portal to send through questions as shown here by the arrow to the right. I'd like to have a warm welcome to our very first live virtual event, which gives us the opportunity to deep-dive into specific elements of the 2-2-5-2 strategy I laid out last year and explore particular topics in a level of scientific and operational detail that we can't often get into but which forms the lifeblood of the company. We plan on having more of these types of events in the future, but what better place to start than with our allogeneic strategy given the recently announced collaboration with Genentech. Today, you'll hear from Jo who'll present an in-depth review outlining what has been achieved to date with our allo platform. Then you'll hear from Helen on our collaboration with Genentech, followed by Q&A. What I want you to take away from the next hour is the following. Firstly, Adaptimmune has developed a leading allogeneic T-cell platform over the last 5 years based on edited stem cells. This platform is flexible, scalable and produces functional T cells that kill tumors. Secondly, we're combining this platform with everything we've learned from decades of autologous T-cell therapy R&D into a pipeline for Adaptimmune and its partners. And lastly, that we'll be able to deliver this to patients, thanks to the cell therapy, CMC and supply chain capabilities that we've built. All of this makes us uniquely placed to drive the future of T-cell therapies. Last year, I outlined our 5-year strategy or the 2-2-5-2 plan. Allogeneic cell therapies, particularly those derived from stem cells such as ours, are a key part of Adaptimmune's future and the future of the industry. And you may remember me highlighting our desire to develop therapies that are both curative and mainstream. Allogeneic approaches may well play a part in developing curative therapies, but they are a key component of the mainstream, of making benefits of cellular therapy available to people broadly and cost-effectively. Today, we can confirm that we intend to file an IND for our first allogeneic TCR against MAGE-A4 in late 2023. Next in the clinic will be our HLA-independent TCR or HiT, targeting mesothelin, which is partnered with Astellas. The strategic collaboration with Genentech further expands our pipeline of allogeneic therapies. The universe of oncology cell therapy is growing and evolving every day. And I want to take a moment to put what we are doing in context of the broader landscape and to show how we and in particular our allogeneic platform fits in. I like this slide because it shows the sheer variety of cell therapy approaches. But 2 things are true whenever you try and capture something like this on one slide: there'll be things missing, and it will fail to convey every possible level of complexity. Nonetheless, it does show some things very clearly. The landscape remains weighted towards autologous therapy, shown here in blue. But there are increasing number of allogeneic approaches also being explored, shown here in red. Across both platform approaches, CAR targeting of T or NK cells dominate. And this isn't surprising because CAR-T therapies are the most advanced against hematological malignancies, with marketed autologous CAR-T therapies and late-stage autologous and allogeneic CAR-T therapies in development. However, the focus on non-CAR branches is increasing as a way to address solid tumors. Obviously, solid tumors are a much larger opportunity than hematological malignancies and significantly more challenging. CAR-T in this setting have to date proved less impressive. And this is likely to be partially target-based and partially the mechanics of CAR-T binding and signaling and partially things we don't understand yet. I think it will be unwise to say CAR-T will never be effective in solid tumors. But the focus is increasingly on TCR and TCR signaling-driven approaches to get to cells that persist, migrate to a tumor, overcome the tumor microenvironment and mount a sustained antitumor assault. And here, Adaptimmune is the established leader. In the autologous setting, running parallel with Iovance, we're planning for one of the first 2 T-cell therapy BLAs for solid tumors next year. And we've enabled our partner, GSK, with their late-stage cell therapy program. We have a multi-tumor study with our next-generation approach targeting MAGE-A4 with data at ESMO next week. And we have exploratory programs in TILs and in HLA-independent TCRs. And in the allogeneic setting, we have our allogeneic alpha beta TCR-targeted platform. And this is not a crowded space. It's also true that we could use our allogeneic platform with other warheads and potentially to develop other immune cells. But given our history with TCR T cells in the autologous setting, we are uniquely positioned to exploit the space. And these therapies are uniquely positioned to be able to address large numbers of patients worldwide across a broad range of solid tumors. And this strategically is what makes our approach differentiated. We're combining the allogeneic technology platform together with our decades of cell therapy expertise together with integrated and experienced CMC capabilities. Only Adaptimmune combines all of these 3 elements. Our technology platform derived from stem cells is designed to be flexible, scalable and commercially viable, enabling virtually unlimited genetic enhancements to produce the best cell therapies possible. Our T-cell therapy expertise with a toolkit of targeting and other genetic edits to make effective T cells that we can use in a plug-and-play fashion in our allo platform. Together with translational learnings from real-world clinical experience with T-cell therapies against solid tumors, and this enables a vast potential pipeline for Adaptimmune and for our partners. And our CMC and supply chain capabilities, which is second to none when it comes to making T-cell therapies, including agile, integrated process development, all specifically dedicated to making TCR T-cell products. This adds up to a company that is uniquely placed to deliver effective, consistent products on demand. And with that, I'll turn it over to Jo who will take you through a deeper dive on our platform, what distinguishes it and the progress that we've made.

Thanks, Ad. Hi. I'm going to take you through our allogeneic cell therapy platform and some of the latest progress that we've made. We really want you to have a good understanding of our platform, and that means there's going to be some fairly heavy science to get through here. I'll make no apology for that, but I'll be very happy to answer any questions that you may have in the Q&A session later. We're using induced pluripotent stem cells or iPSCs for this platform, and there are plenty of reasons why we've chosen to do that. iPSC have many unique characteristics that make them suitable for cell therapies. One of the most important things about this platform is that we use a single source of cells. All of the cells come from one original donor. The iPSC lines that we use were made from umbilical cord blood, and those cells were reprogrammed back to the completely pluripotent state, and single cell clones were isolated to generate a starting line. When cells are in a truly pluripotent state, it means that they have the potential to turn into any cell type that's found in the human body, that is if they're given the right cues and signals to differentiate down the relevant developmental pathways. Another brilliant feature of iPSCs is that they have a really high proliferative potential. Theoretically, they can expand indefinitely in the pluripotent state. And because of that, we can make very large banks of those cells, and that gives us reproducible starting material for our platform. We can take a vial from a bank and know that we are starting our process with the same material every time. And that's a really important feature for product consistency. This is really only possible with a stem cell-derived platform because batch-to-batch variation cannot be fully controlled when using healthy donor-derived cells. And this capacity for expansion and the ability to bank the cells also allows for a very flexible platform. We're using gene editing here. And that allows us to do multiple different edits either sequentially or in parallel at the same time so that we can modify multiple genes. Using editing technology means that we don't have issues with lentiviral vector capacity limits that could restrict the changes that we want to make. Editing gives us the flexibility to really build our future programs however we like based on translational learnings. So when we find out what happens with these cells in the clinic, we're able to apply those learnings back to research and make further edits and bring forward new products to address the things that we need to. We also think that this is a scalable platform. Because we use a single source, there's a single line to fully characterize for each product. Our platform uses defined media as we go through this differentiation process to make T cells from the edited iPSC. We don't use serum, and we don't use feeder cell lines. Those are both really important factors to enable GMP manufacturing to work in a sustainable, controlled way. All of these components are vital to generate an adaptable cell therapy platform that can be used to treat different cancers by plugging in different TCRs or CARs. Stem cells are the only cell type that enables this kind of approach. Because of their key characteristics, they're highly proliferative, they can generate any cell type, and they're derived from a single source. There are 3 key areas of focus or 3 buckets, if you like, when you think about building a stem cell-derived allogeneic platform. And so those are, number one, gene editing. So which genes are you targeting? What modifications are you going to make? Are you knocking out? Are you knocking in? Are you truncating? Are you specifically mutating? And most importantly, does that result in the biological effect that you are aiming for? And then number two, the differentiation process itself. So how do you tweak that process to give you the right sort of cell with the desired phenotype? Several different cell types have a range of anticancer activities that could be augmented by a TCR or a CAR. And which one is the best option? And then number three, for scale-up. How do you move from research-scale plates to manufacturing-ready vessels and maintain the right cellular characteristics from complicated processes so not to be underestimated? All 3 have different biological challenges associated with them, but all 3 are critical to making good products. And you'll see that I've used this numbering throughout the coming slides to help make it clearer which of these areas that I'm talking about. So let's start with number one, the gene editing side of things. You can either do all your edits in one step or you can do multiple rounds of different edits, and we've used both approaches before. So this is the process that we go through. We start off with a single iPSC claim, and then we edit it. But we go through a single cell cloning step after each round of editing. And that single cell cloning step is really important. Going back to a single cell, we know that all future cells come from that original cell, so they will all carry the same modifications. And then we look at the genetics to make sure that we've got the right edits. We can check whether there are any off-target edits, look for mutations, et cetera, basically check that nothing's gone untoward, and we can screen out anything that we don't want in the final product. And if we want to add anything extra or tackle second copies of certain genes, then we can go around that cycle again. The ability to quality-control the clones and understand the modifications we've made is critical. That single-cell cloning process allows us to select the right line to take forward based on the genetic profile. And once we've chosen that best line and we're sure that it's got the right genetic and growth characteristics, we can expand them up to make a stem cell bank. Because of the inherent proliferative potential of iPSCs, they can be expanded to make hundreds or thousands of vials in a master cell bank. And then you can take a single vial from that master cell bank and generate a working cell bank, just like you would do with producer cell lines for viruses or antibodies. Very careful culturing methods are required to maintain the pluripotency of the stem cells and to keep the genetic profile of the bank in good shape. You need to minimize the introduction of unwanted mutations. And afterwards, once the banks are frozen down, we can perform quality control tests on them to give us confidence in our GMP starting material. And we know that we can go back to that bank and start the differentiation process with the same high-quality material for every batch of T cells that we're going to make. We need multiple rounds of editing to make a safe and efficacious product. So firstly, we need to address potency. And for that, we are inserting the same TCRs that we use in our autologous process because we've got clinical data with those, but you could just insert a CAR just as easily if that's what you wanted to do. And we're also looking at other next-gen modifications to improve the quality of the cells after differentiating. And secondly, it's really important to engineer for safety as well. And that's where we need to remove the natural TCR to prevent graft-versus-host disease. We've recently made progress with the RAG knockout to address the native TCR, and I'll come back and explain a little bit more about what RAG is and why we've chosen to knock that out instead of the TCR genes in the next slide. So just bear with me. And then thirdly, we're also looking to make these cells persist for longer because we know from our autologous therapy that people who do well on T-cell therapies, their cells hang around for a long time to do the work. Cancer is not cured in a couple of days. So we're looking at making modifications to hide our allo cells from the patient's immune systems so that they can persist for longer in the patients, and there is always the option for repeat dosing if necessary. So that single cell cloning step that I mentioned at the beginning, that makes sure that every single cell in this product contains every edit. So you don't have a mix of cells where some cells have got some modifications and not others, which is what you get if you do the editing in bulk. Because of that single cell cloning step, every single cell contains every single edit. And this feature sets stem cell-derived products apart from autologous products and other allogeneic approaches where healthy donor cells are edited in bulk. So here, this is an example of editing for safety to remove the natural TCR. We haven't edited the TCR genes themselves here, but we've knocked out the right gene instead, so why did we choose to do that? The line that we're using at Adaptimmune was not made from a T cell, so the TCR genes haven't been rearranged to make functional TCRs already. And that TCR gene cluster is huge with multiple variable bits that must be spliced together to make the working TCR. And that process is known as VDJ recombination. And this is responsible for the hyper-variability that's critical to how the Adaptimmune system works. And there are 2 copies of the TCR alpha chain, 2 for the beta and gamma and delta chains as well. RAG encodes a protein that's required to enable that VDJ recombination and make those gene rearrangements to produce working TCRs. So if we knock out both copies of RAG, we prevent the cells from making any of the TCR chains. So that's only 2 edits rather than tackling 8 if we were to take out all of the TCR chain constant regions. Knocking out RAG activity and stopping the TCR genes rearranging means that these cells can't differentiate the T cells unless you insert a separate TCR. And we've done it this way because we use a TCR platform at Adaptimmune, which requires a different editing strategy than you might want to use for a CAR. It's just one of the differences in the way that TCRs and CARs work. So the flow cytometry plot that you're looking at here shows cells that have been differentiated to T cells, and they now express both CD3 and TCR. Those markers are both hallmarks of T cells. And the clone here hasn't been edited for RAG, so it still has the ability to make its own T-cell receptors. You can see our MAGE-A4 TCR staining in the blue box or staining with dextramer there, so we know that's specific for our MAGE-A4 TCR. But in the red circle on the left, you can see that there are other TCRs present. So those cells aren't expressing the MAGE-A4 TCR, but they're still TCR positive. And those TCRs are formed without specific family selection, and we have no idea what they might recognize. It could be absolutely anything. So we've got to get rid of them. There's no way that you would want to give that to a patient. However, in the plots on the right, we have knocked out RAG this time, and we've managed to prevent that natural TCR expression. You can see that all the cells are really staining in the DEX from a positive blue box, so we know we've got the right TCR present. And we're not seeing cells in the green circle, so we've lost the expression of those natural TCRs. And we can confirm that in other assays as well. So this flow cytometry assay here measures surface protein by active body binding. But we can also confirm TCR expression or the lack of it by PCR-based repertoire analysis, looking at the actual gene rearrangements to see whether functional TCRs have been made or not. It's important to know that the edits do have the expected functional outcome. Then next, I'm going to move on to the second area of focus or bucket #2, and that's our differentiation process, which is the core of the platform. What we've created here is a way of making those iPSCs become T cells by making sure that they follow the right developmental pathways. I'll take you through what that looks like in more detail in a moment. So we start with one of those vials from the working cell banks. Remember, that's the fully edited GMP starting material or the output of bucket 1, and then we sort this out. And we put it into our differentiation process where we guide the stem cells through multiple intermediate stages to become T cells. And once they are T cells, we harvest those cells just like we do in our autologous process. We wash them and we freeze them down in patient-ready doses. And because we'll be making a large batch with multiple doses, we can then release-test the whole batch at once instead of releasing individual products for individual patients as we do now with the autologous products. And we can do all of that quality testing and know that all of those doses that we've made are good to go to patients and then get stored in a centralized inventory ready for shipping. And one single vial of iPSC will make multiple doses of T cells. Exactly how many doses we'll be able to make depends on our third area of focus, working on the scale-up, which I'll come on to later. This method of manufacturing in batches is key to making cell therapies a mainstream option for patients as it enables easier access. So this gives you an idea of what that iPSC to T-cell journey looks like in a dish as performed in our research labs. Our process is really trying to mimic how these cells develop in the human body in a lab setting. So we start out with stem cells on the left, though remember, at this point, those cells could become any sort of cell that you find in the human body. And as they start down the differentiation process, they start to form complicated 3-dimensional structures that you can see there in that second panel. Those self-assemble. They all come from the original iPSC. The differentiation process actually generates its own support cells in a complicated mix of cell types. And then they actually form into tubular structures that look like blood vessels in the well. So in the third panel, you can see cells which have been stained red. And that red staining is an endothelial marker, which you would normally see lining blood vessels. And the green cells that you can see inside those red tubes are stained for CD45, and CD45 is a marker of blood cells. So you have a blood vessel-type structure where you're starting to make blood cells. What you're seeing here is hematopoiesis or blood generation in a dish. This is a critical step and is mimicking how blood first forms in humans. Blood production first starts out in the aorta before it moves to its final home in the bone marrow much later on in human development. Those green CD45 cells are where the T cells will come from, if you push them even further along. And you can see that in the far right-hand panel. The green staining on the final image is CD3, which is an archetypal T-cell marker. And you can see there that the T cells are growing inside the vessel along the top of the picture. And you can also see them bursting out and migrating out into the culture further down, which is another component of natural T-cell behavior as they need to escape from blood vessels and go into tissues under certain conditions. So this shows that we're on the right track to making the right sort of cell with T-cell markers and T-cell behavior at the end of our differentiation process. What's actually happening in that culture system? I'm going to take you back to the developmental pathways now, and I'll go through the fate decisions that the cells made on their way to becoming T cells. So we start with stem cells, and they can become anything at this point. First 3 stages of the process pushed the cells to become hematopoietic stem cells. And by this point, they're restricted to making blood cells, nothing else. So no neurons, no muscle, no liver or kidney, et cetera. All of those pathways have been cut off, and the cells can't go back. They can just go forward to become blood cells. But there are still an awful lot of different cells in blood to choose between. And the next fate choice is between myeloid and lymphoid cells, so we need to push those cells to make the lymphoid choice. That myeloid arm leads to things like dendritic cells, monocytes and macrophages, amongst others, and that's not what we're aiming for. And once they've been pushed down the lymphoblast branch, then the next choice is between becoming an NK cell or sticking with the lymphocytes. As we're focusing on T cells, we don't want them to take that NK cell branch. Then next, the B cell split-off, and we don't want those either, so we need the cells to choose the T-cell path. We're getting close to now, but we still need to push the cells to take the alpha beta path rather than the gamma delta path. And then finally, we get to the mature alpha beta T cells. You can see that it's quite a complicated process to generate T cells. As the cells are dividing, there are lots of fate choices that they need to go the right way to get the cells that we want at the end. We've deliberately chosen to steer away from NK and gamma delta T cells for now, but we know where those branch points are in our process, and we can go back and look at them in the future if we want to expand the platform. Our current focus on alpha beta T cells comes from the clinical success that we've seen in our autologous programs with afami-cel in synovial sarcoma and our SURPASS next-gen trial. We're feeding as much of that information as possible into our allogeneic platform, hoping to achieve similar successful outcomes for patients by aiming for the same cell type. As I said before, the ability to learn from our autologous cell programs is really valuable for the allogeneic platform. We use information on how the TCRs work in patient cells and what the outcome was for those patients to guide our allogeneic platform development. Our translational team does a lot of single cell gene analysis on autologous patient products, and we can use that to test how similar our allo cells look in their gene expression profile. What I'm going to take you through now is a bioinformatic rendering of the total gene expression of a whole load of cells plotted onto the 2D plot. And each dot that you see here represents a single cell, and they're color-coded by sample. The analysis looks at the expression level of every gene and assigns a position on the graph to gauge differences. It's trying to highlight the biggest differences between the cells without caring what those genes are or what they do. So the closer the dots, the more similar the genic expression profile. We'll go through it bit by bit, and it's underpinned by some very complicated maths. But the important thing to remember is that dots close together means that the cells are very similar and dots far apart means that the cells are quite different. So what you see here are cells from patient-derived products. They've been split between CD4 and CD8 cells. And they're lenti transduced, so TCR positive, or non-transduced, so TCR negative, which gives you the 4 colors. So if we start with the stem cells, and that's the yellow blob, you can see that they sit in a very different part of the profile, quite a long way away from the autologous process. And after the cells have been through stages 1 to 3, they've become hematopoietic stem cells, so those are the blood precursors. And you can see that the pink blob is again in a completely different spot. So it's not like the autologous product, and it's not like the stem cells. Their gene expression profiles have changed an awful lot in those first 3 stages. And next, we're going to stage 5 where we have immature T cells in the green blob, and they express both CD4 and CD8 markers at this point. And you can see now that actually those cells are closer to the autologous product than the cells in stage 3. Stage 6 is about maturing into CD3, CD8 single positive cells. And you can see now that stage 6 in the turquoise is actually cutting the autologous product, and there's a little bit of overlap between the turquoise and the red. So they're looking fairly close to the same CD8 cells that we have in the patient-derived products in this analysis. And they don't look exactly the same, and that's okay. We'll see how they do in patients and bring that information back for comparison in the future. This sort of analysis and our ability to bridge between autologous and allogeneic is really useful because it helps us see changes in the process to tweak cell phenotype, and we can look at this sort of gene expression information to see if we're getting something that looks very similar to our autologous products, which we know works in solid tumors. It really is our clinical success with the autologous T cells that's driving where we take our allogeneic platform. And that's one of the main reasons that we're really focused on T cells rather than NK cells because we want to build on that clinical success. What you're going to see in this video when we play it in a moment is that the cells that we're making in the allogeneic products have the ability to kill more than once. So that's something that's really important. We know that it's a long battle in the patient against their tumor. And these things don't just get cleared within a couple of days. So it's important that the cells are able to survive and continue to kill multiple targets. So we'll play the video now. And what you're seeing is the allogeneic T cells stained in blue, and they're in a well with antigen-positive tumor cells stained in red. When those T cells recognize and kill those antigen-positive tumor targets, the cells flash green. You can see that the cells are pretty active. Some of them move around a lot during the course of this video, which is an image taken every 5 minutes for 48 hours. That green signal is [ Cas-based ] signal, and that's a sign that those tumor cells are dying. You can see that there's one individual cell here which kills up to 6 target cells, and you can also see how some of those blue T cells are really active and really kicking that tumor. And now we're going to move on to bucket 3, and that's the scale-up. To make a good allogeneic product, we need to be able to get economies of scale. Where the program is at the moment is that we've got a research-scale process that's in our Milton Park facility near Oxford, and we're feeding that into our allogeneic process development team. So we've got a dedicated team working on this, but they also work very closely with our other PD teams who have developed our lentivector and autologous T cell processes. Leveraging learnings about the use of large-scale bioreactors from our lenti work and the impact that different culture technologies have on T-cell phenotype from the autologous T-cell team. So it's a combination of understanding conditions that are good for T cells combined with the scale that we use for lentivector, looking at those bioreactors and how to scale things up because there's no well-trodden path to tap into here. This sort of equipment has not been designed with our application in mind. So we're trying to learn as much as we can from as many different areas as possible and apply it in a new way. And this is exactly what was done for the first autologous T-cell processes. We're also using this team's knowledge of how to successfully tech-transfer new processes into new custom-built manufacturing facilities. We already have 2 manufacturing facilities, which has been critical to the success of our autologous platform. And we're now building an allogeneic facility. We know from experience that being able to control your manufacturing is a critical component of clinical success. So we're basically putting everything we've learned from the autologous side of the business, both scientifically and operationally, into our allogeneic program. From the patient's point of view, this is a much easier process. Once they're diagnosed with cancer, they work with their clinician at the hospital. They still need a blood test to screen for HLA, and they will need a tumor biopsy to test for antigen. And if both of those are positive, the patient and the clinician now just request the cells from us at Adaptimmune. That request will go to the centralized inventory, and we'll be able to ship a sample straight away. So the patient has a much easier time of it. They don't have to go into hospital for a leukapheresis like they would for an autologous cell treatment. It's less time in hospital for them, one less procedure, and they get the cells faster. It's easier for the clinicians, too, as there's one less visit to schedule, and it's much more like a normal medicine. As soon as those cells arrive, they'll be ready for infusion at the hospital whenever the patient and clinician are ready. So to finish off this section, I'm going to take you through our allogeneic pipeline. Our lead allo program uses the same MAGE-A4 TCR that we're using in afami-cel and our SURPASS next-gen trial. So we know a lot about the safety of that TCR already and its efficacy in 2 different autologous products. We plan to file an IND for our first allogeneic TCR against MAGE-A4 in late 2023. Other proprietary TCRs can be plugged into the platform, and we're looking at next-gen modifications in early research. Our partnership with Astellas is progressing well, and we have 2 projects live now with them. The first nominated collaboration target is mesothelin, and we're using Adaptimmune's HLA-independent TCR or HiT platform for that so patients will not be restricted by their HLA type. And this will be the second allo product to enter the clinic under our 2-2-5-2 strategy. Work on a second target with Astellas is also underway. And as we've just announced, the deal with Genentech, a second partnership that really adds weight to what we're doing with our allogeneic platform. We'll be working with them to plug their TCR technology into our stem cell platform to create a range of new off-the-shelf TCR products and a novel personalized product to help more patients. It's a real boost to our team to have them on board, a strong validation of what we've achieved to date, and I'm very excited about what we hope to achieve together in the coming years. So that's the latest on the platform for now. Happy to take any questions when we get to the Q&A section, and I'll pass this on to Helen. Thank you.

Thanks, Jo. So before I start, I want to acknowledge the great work of the Adaptimmune allogeneic team, leveraging as it does a strong scientific pedigree in producing optimized autologous T-cell products into the complex science of generating off-the-shelf T cells. And secondly, I want to say how very excited we are about this new strategic collaboration with Genentech announced earlier this week, which validates what we have believed and been working on for several years now, namely that it is possible to generate allogeneic T cells able to mimic our autologous products for the attributes necessary, we hope, to have a long-term effect on solid tumors. We believe we have built a world-leading allogeneic platform that puts us and our partners in a strong position to deliver T-cell therapies to cancer patients on demand. With this new collaboration, we will combine Adaptimmune's leading allogeneic platform with Genentech's expertise in editing T-cell receptors into autologous T cells in addition to its clinical and commercial capabilities. The collaboration has 2 components: first, the development of allogeneic T-cell therapies for up to 5 distinct shared cancer targets in a similar way to those we are developing for our MAGE-A4 program. And second, the development of personalized allogeneic T-cell therapies where any patient can receive a T-cell product for their cancer customized with their own T-cell receptors. For each component, Adaptimmune will develop the cell lines to produce clinical candidates using our iPSC-derived allogeneic platform to produce T cells or as we refer to them, iT cells. In each track of the collaboration, there is scope to produce next-generation versions at the starting cell lines as we learn which improvements will be important clinically beyond the edits that Jo described earlier. Genentech will provide the input TCRs and take the T-cell therapies into subsequent clinical development and commercialization. Of note, the collaboration excludes targets for which Adaptimmune has an active ongoing program, including our MAGE-A4 targeted therapies or targets for which we have already licensed to a third party. This is a very attractive collaboration for us, for the science but also financially as we may receive up to approximately $3.7 billion in payments. In addition to the upfront payment of USD 150 million, we will receive additional payments totaling $150 million over 5 years unless the agreement is terminated. We may also be eligible to receive research, development, regulatory and commercial milestones in payments potentially exceeding $3 billion in aggregate value as well as tiered royalties on net sales in the single -- mid-single to low double-digit range. Finally, we could opt in to a 50-50 U.S. profit and cost share on the 5 off-the-shelf products, meaning that we will be eligible to share 50% of profits and losses from U.S. sales on these products as well as the ex U.S. regulatory and sales-based milestone payments and royalties. And so to conclude, I want to say again how very excited we are about this collaboration. So thank you for your attention today, and I hand over to Bill for the Q&A section.

Thank you, Helen. Very happy to be here. I'm Bill Bertrand, the Chief Operating Officer here at Adaptimmune, and I have the honor of spearheading the questions that have come in during this webinar today. [Operator Instructions]

So we will kick this right off. Helen, the first question is to you. Can you please provide more details on the financials of the deal?

Thanks, Bill. So I can provide a little bit more granularity under what we've disclosed publicly. So apart from the upfront and the payments over the first 5 years, we are also able to receive up to $1.85 billion in terms of research, development and regulatory milestones and $1.5 billion in sales milestones. So that's some additional granularity on the milestone breakdown in terms of how we get to the over the $3 billion component. And those details are within our 8-K filings.

Perfect. And Helen, maybe just a slight follow-up to that. What are the financials and profit share details for the personalized products? Is it similar to what you told us for the off-the-shelf targeted products?

So we don't have the option. We're already potentially opting into the targeted products, products similar to the MAGE-A4 in terms of what they look like as products for delivery to patients. We will be participating in terms of development, commercial milestones and royalties on the personalized product stream.

Perfect. Thank you, Helen. Jo, a question for you. Can you speak to the manufacturer in terms of time relative to the autologous product and how it compares from a cost perspective? And maybe after you've addressed that, Ad, we can ask you to talk a little bit about the cost of goods.

Okay, Bill. So I think the key for -- as far as patients are concerned about the speed is that the manufacturing is obviously completely separate from the patients. So the speed of delivery from a patient point of view is very different from the speed of delivery of how long it actually takes to make a single batch of product. I can't go into any more detail on the scale that we're at or the cost of goods yet. It's still too early for us to talk about that in detail. But the very fact that we're building our own manufacturing facility, we'll have the ability to control the number of batches when we schedule them, how many we want to run, that's all a key part of being able to deliver this at the scale that we need to. But I can't give any more details today.

Perfect. Thank you, Jo. And since you did address what we can say on the cost of goods, we will give Ad a brief reprieve on that. So Jo, next question is also for you. What distinguishes Adaptimmune from other allo platforms, like Fate's, for instance?

So I think what distinguishes us is we're stem cell-derived going for T cells, specifically alpha beta T cells. So we're not the only people in that space. As you said, Fate are also doing that. I think what differentiates us from Fate in particular is the ability to learn from our autologous products. So the fact that we have both streams in-house is something that's very helpful to us and we think is a key advantage. There are other people out there in the stem cell space, but they're generally starting with CARs, and they're starting on the NK cells because it's easier to get to an NK cell than it is to a T cell. So I think there will be more people joining us in the TCR T space over time. But right now, they're focusing in different areas than we are.

And let me ask this follow-up question, although I think you did hit a bit of it in the answer you just gave, is why are NK cells easier to make? And are T cells really a better cell for therapies than NK? And what are some of the key differences? I don't know if there's anything you want to expound on from your last answer relative to that.

So I think if you remember the differentiation pathway slide that I showed you, the NK cells have to make fewer fate choices before they come to that phenotype. So they come out earlier. The differentiation process is a little bit shorter, which obviously helps with manufacturing. And there are plenty of companies who are saying that they can do that at a good scale as required for clinical use. But I think when you look at what's out there in the broader space for NK cells versus T cells, NK cells do have innate antitumor activity. They do work in some circumstances, but they don't last a long time like T cells do. And that's really where we focus on T cells, is that we know we've got good clinical data with T cells. We know that patients who do well have their T cells stay around for a very long time. And that's always been our aim, is to go for that. So it's not that NK cells won't work, and we may end up in a hybrid as these cells develop over time. It's an early stage for the field. Nobody knows what the right answer is yet. And I think we're going on, alpha beta cells are the prize for us. And there will be other people looking at NK cells and gamma deltas, and we will see who wins in patients in real life.

Great. Thank you, Jo. Ad, the next question is for you. Can you give some color on your cash runway and how the deal with Genentech impacts it?

Thanks. The deal with Genentech extends the runway from previous guidance, which was early 2023 through into 2024. So we are, relatively speaking, well capitalized at this point in time.

Thank you, Ad. Jo, the next question is for you. Have you been able to benchmark the serial edit process versus batch editing to understand the difference in quality and viability of the product?

So I think the way that we work, everything done upfront in the stem cell part rather than in the bulk editing part is because it gives us that ability to control our starting material. So when it comes to working in batches, I think there's plenty of evidence out there that the more -- the higher the number of edits that you want to achieve, then the lower the efficiency of getting all the edits in. So if you look at what we're doing, we're knocking out 2 copies of RAGs, we're knocking out 2 copies of beta 2Ms. We're doing some other knock-ins. And by the time you get past 2 steps -- when you get past -- so if you get to, say, 90% efficiency, which is hard to do with a single edit, by the time you've added 2 edits, then you're 0.9 times 0.9, you're down at 0.81. And every time you add on another edit and if you're looking to knock out 2 copies of the gene, your number of cells that carry everything goes down really quite substantially. And even with the best editing technology in the world, you're never going to get to a product that is homogeneous. And that's why the stem -- that's what I think the advantage of stem cells is in that single cell cloning step means you end up with a much more homogeneous product where you know exactly what editing you've done because you've done it upfront. So it's a completely different way of looking at this process and this problem. And certainly, when you want to go to large scale, we -- Adaptimmune, we really feel that that's the right way to go.

Thank you, Jo. I'm sorry, just trying to scroll through and organize some of these questions that are coming in fast and furious. So Jo, the next 2 questions are also for you, and I'll sort of combine these 2 and allow you to kind of answer in the manner you see fit. First, how does your differentiation process differ versus competitors? And is there anything that can be patented to create a barrier of entry? And in follow-up to that, what is the main manufacturing challenge that you face in the development of the cells?

Okay. So we have filed IP on the process along the way, and we do always look for any kind of area that we can protect. So I think one of the key differentiators is not using -- supporting cell external lines. We -- everything is self-supporting. But it is a complex field, and there is -- IP is an area that everybody is trying to protect everything as much as they can. So we are always looking at that, and we're always looking at what we can protect. And we have filed multiple patents as we've gone through this process. When it comes to scale-up, it really is about taking what started out as a 2-dimensional, so a flat culture process, and I showed you in those pictures where you can see those vessel structures and moving that into 3 dimensions. That's where it becomes difficult. So the scale-up is around how you make a process that is closed and amenable to manufacturing at scale rather than something which is an open research-scale process. So it's always a challenge to do that process development work. It's always something that you need to work with equipment manufacturers in because as soon as you get to scale, just physics comes into a process as well as biology. But the really important thing is to check that you've maintained the same biology at the end of the process. And with these differentiation processes, which do take a long time, they're very complicated processes. Ours is -- as are our competitors as well as far as we can tell, they are very similar in terms of length, so the number of days that you need to get these cells to go through these different fate decisions. That all requires very careful control. And one of the biggest issues, in fact, is being able to sample cells from bioreactors because bioreactors are generally made for you to sample supernatant end product. They're not made for you to get the cells back out of. So that's something that we're working with providers on at the moment.

Great. Thank you, Jo. And one more, and then I promise I will give you a brief break, and we'll get Helen up there to answer a couple of questions. So do you think you will need Cy/Flu preconditioning? Or could you use a different lymphodepleting regimen?

So I think we're basing our first steps on what we already know from autologous. So it's very likely that we will look at something similar as we go forward. But I think what -- the beauty of a platform like this is that it enables us to make many more changes and modifications. So we are already -- it's been a long-term goal at Adaptimmune to see if we can get away from that at some point in the future. And as I mentioned, we're already working on next-gen approaches in the allogeneic team. That's certainly one of the factors that we're considering of whether we can get away from that in the future for patients.

Great. Thank you, Jo. Helen, a couple of questions for you. The first, the collaboration with Genentech is confined to solid tumors only. Or is there any future plan to expand in hematological cancers?

Thanks, Bill. We haven't actually mentioned indications or target indications at all in anything that we've disclosed, and we won't at this point. And we see it's a confidential discussion with Genentech in terms of where the targets they selected will go. So we may be able to talk about that more in the future, but right now, we're not saying.

Okay. Perfect. Thank you, Helen. And another question for you. Are the 5 targets chosen by Roche already known? And do they compete with the longer-term pipeline of Adaptimmune's selection?

So obviously, I can't answer the first question. I can't answer with any specific granularity. Clearly, we have a plan to move forward with. So obviously, some of them are known. All I would say is that we're really excited that they're actually complementary to what we are doing within Adaptimmune. So we have our own portfolio and programs, and this collaboration brings in new and complementary targets and programs alongside what we're doing. So I think overall, we have an expanded portfolio, which is great.

Perfect. Thank you, Helen. Jo, a couple of questions, again, not surprisingly, for you. So first, beyond the global gene expression profile, how do your iT cells compare in terms of functional assays like cell expansion, interferon production, polyfunctionality, et cetera, once exposed to antigen?

So those are questions that we are looking at. Absolutely, I don't have all of the answers, and those will be something that we would want to put out, I think, in the future. Obviously, we know what we're aiming for with the autologous cells. We always have that as a benchmark that we compare ourselves to. The allo products will be different. I mean we are making CD3, CD8 single positive cells. There's no CD4 component. We're very mindful of that, knowing what we know with our autologous programs, and we are looking at all the factors. So I can't give you any more information today. Certainly, a lot of it does depend on the actual clones, come through editing and its comparison in terms of screening the number of clones. It's not just about the genetics and their growth characteristics at the iPSC level. It's also how well they differentiate and the quality of the cells that you get at the end. So it is about clone selection always. And so at the moment, it's not something that we're prepared to disclose.

Perfect. And Jo, maybe one follow-up, and then I have another question for Helen after that. So Jo, do you see any differences in the CD4/CD8 ratio for your iT cells versus your clinical experience with autologous products? And do you plan to control this ratio for your allogeneic products?

So right now, we're only making CD8 cells, as I just said. There are no CD4 cells in the product. But we do know that CD4 cells can play a really critical role in terms of patient responses. So we're looking at next-gen approaches to -- as well as process changes to maybe look at a CD4 component. But the first products will be CD8 only.

Perfect. Thank you. And Helen, will the personalized platform choice of Roche stop Adaptimmune's own work on TILs?

Thanks, Bill. The short answer is no. Different sorts of products. TILs, obviously autologous, not isolating patient-specific TCRs. So rather different but same idea. So no.

Okay. Perfect. And Jo, actually, I was incorrect. Another question has come in. One more for you. Is the personalized medicine basically just TILs?

So I think you can think of it as an off-the-shelf version of TILs. I think it's designed for the patient specifically, and that's really where we're trying to get to how to make TILs off the shelf because at the moment, they're so personalized and autologous, yes.

Okay. And I know I keep teasing you saying we're not going to have more questions, but another one came in. And this will be the last question. Is it easier -- this is for Jo. Is it easier to make iPSC CAR-T cells compared to iPSC TCR cells?

So I mean we don't have any -- I don't have any past experience of CAR iT cells. I think our learnings of how the TCR drives the differentiation means that for me personally, I am intrigued as to how you qualify a T cell when you've actually stopped its TCR. I mean if they remove -- they use TCR alpha chain to put the CAR load -- the CAR in instead, I think what you end up there with is its differences in biology between CARs and TCRs. How the CAR drives the differentiation, it will be slightly different to how the TCR drives the differentiation. So there may well be qualitative differences at the end of those processes. But I have no evidence for that. We've done no experimentation with CARs internally. So it's -- that's just speculation on my part at the moment.

Perfect. Thank you, Jo, and thank you, Helen and Ad, for answering the questions. And thanks, all of the attendees, for submitting these very interesting and exciting questions relative to our allogeneic programs and platform. And with that, I will turn it over to Ad for some closing comments.

Thanks, Bill, and thanks, Jo and Helen. And thanks, everyone, again, for your questions. Cell therapy for cancer is in its infancy. And if you've listened to me previously, you'll know, I believe that cell therapy now is where antibodies were perhaps in the early '90s. And the potential of cell therapy, we believe, is obviously significantly more profound than the potential of antibodies. We're already delivering for people with cancer. We're poised to file a BLA next year for a rare unmet medical need, and we have a rich pipeline of further candidates and enhancements to continue the TCR evolution. And today's virtual event was about understanding the role that our allogeneic platform plays in enabling Adaptimmune and its partners to treat more patients with more effective cells on the road to making cell therapies that are both curative and mainstream. And here, we're in an excellent position. We have an industry-leading stem cell-derived platform. We have deep expertise in developing and manufacturing T-cell therapies. We know what it takes to make effective T cells against solid tumors. And together with our partners, Astellas and Genentech, we're committed to transforming the lives of people with cancer with allogeneic T-cell therapies. Thanks, everyone, for joining us, and please reach out to our IR and comms colleagues if you'd like to speak further. Have a good rest of the day.