Cellectis S.A. (ALCLS) Earnings Call Transcript & Summary

Okay. Great. Welcome, everyone. I'm Yigal Nochomovitz. I'm one of the biotech analysts here at Citi. Welcome to the panel on autologous, allogeneic CAR-T, CAR-NK and more exploring the state of play in cell therapy. And it's my great pleasure to have with me today from Atara Bio, Pascal Touchon, who's the President and CEO; from Caribou Biosciences that recently went public, Rachel Haurwitz, President and CEO; from Cellectis, Andre Choulika, the CEO; and from Gracell Bio, William Cao, Chairman and CEO. So welcome all of you today. Thank you very much for taking the time. I think as a starting point for those less familiar with each of your companies, maybe we could just go through and do a brief introduction to the company and the pipeline and maybe comment as well on what you see as the strengths and differentiating qualities of your cell therapy platform. So maybe, Andre, you want to kick it off?

Sure. Well, thank you very much for the nice introduction, Yigal. Appreciate it. Cellectis is a gene editing company at the basis. We've been founded on the concept of gene editing at start. We had the first wave of the gene-edited product that we started developing back like a few years ago, I would say like almost 8 years ago, on the concept of using gene editing to make allogeneic CAR-Ts. And since then, Cellectis has been developing a series of products as allogeneic CAR-T. We started with the first product that was a CD19 that we licensed out to Servier that is currently -- between Servier and -- actually, when we started with Servier, it was with Pascal. Pascal was at Servier at this time. So that was a very exciting time. And we licensed their U.S. rights to Allogene so that's currently under the name of ALLO-501 or ALLO-501A. And then we had a lot -- a series of other CAR-T in the space. Recently, we've announced also the launch of a new platform that is a gene editing platform for hematopoietic stem cells. And we're tackling the first type of diseases -- or disease that will be sickle cell disease. But there are a series of other type of diseases, ranging from lysosomal storage diseases to immunodefficiencies. And we have, since the summer, full manufacturing ranging from buffers,immunodeficiencies so DNA, RNA vectors and cell therapies that are manufactured internally. And that's quite a quick overview of Cellectis.

Thank you very much, Andre. Rachel, tell us about Caribou in 1 minute.

Thank you, Yigal, for the invitation. Caribou is a CRISPR genome editing company. We were spun out of Jennifer Doudna's lab at UC Berkeley about 9 years ago. And during this time, we've actually invented our own proprietary next-generation CRISPR technology. We call it the chRDNA platform, and it has far more specific genome editing than first-generation CRISPR-Cas9. We use this technology today with a focus on oncology. We're advancing 2 kinds of therapies in parallel. We're developing allogeneic CAR-Ts for hematologic malignancies and allogeneic CAR-NKs for solid tumors. A key theme that runs across all the work that we do is persistence. We believe that enhancing persistence is key to unlocking the broader opportunities for allogeneic cell therapies, and I'd be happy to dive into more of those details as the conversation progresses today.

Perfect. Hi, William.

Gracell is founded in China, and we have been developing 2 platform technology trying to resolve the industrial sort of bottlenecks. One is for autologous CAR. It takes weeks, and that's sort of in the first norm. And obviously, that creates not just cost -- high cost, but also the time to deliver is an issue. And some of the patients could not wait. Or some of the patients, the disease progress to next stage. So we developed this FasTCAR, which is next-day manufacture. They're great to resolve some of these issues. And the trio CAR, it's a stand-alone UCART, gene edited, but we don't use anything else to suppress the allo reaction. Instead, we apply the dual CARs. So one goes to kill the target cell and the other one to prevent rejection, sort of dual-CAR approach. We have multiple pipelines in clinical trials or IIT trial, IND trials and sort of late stage of early developments. And we also have enhanced CAR for solid tumor, and that is based on gene editing. We have published the first IIT studies. That is a PD-1 knockout, whether it's TCR plus PD-1 knockout, which is autologous CAR or mesothelin CAR-T. And the result is encouraging in the sense of potential of autoimmune disease, and that's the most worrisome potential side effects. And it turned out, we didn't see any severe side effects. So it gave us sort of baseline safety tolerability. So now we have a second-generation sort of enhanced CAR and flipped negative signal around and targeting certain solid tumors and also adapting the Claudin 18.2 as a new target. So yes, we know -- most of the people are in China, but we are spreading to South California.

Thank you. And Pascal?

So Atara Bio is located in California. Atara Bio is a leading allogeneic T cell immunotherapy company with a fully differentiated allogeneic cell therapy platform relying on EBV for Epstein-Barr virus T cells from healthy donors. And we have now preprogrammed in the clinic, including what is arguably the most advanced allogeneic cell therapy in Phase III CAR T-cell, the first-in-line BTD in oncology, soon to be submitted for regulatory approval in the EU and the U.S. We have also in Phase II a potentially transformative therapy for multiple sclerosis, which is now in a randomized controlled trial and has the potential to be the first-ever therapy in progressive MS to reverse disability. And then we have a pipeline of allogeneic CAR-T in liquid and solid tumors, the 2 most advanced in terms of allogeneic CAR-T going to the clinic next year: one targeting CD19 as a potential best-in-class for B-cell malignancies, the other one targeting mesothelin for a number of solid tumors.

Perfect. So let's get into a little bit more of a specific discussion about cellular therapy. I'd like to get an understanding from the panel as to what you see as the key hurdles and challenges associated with allogeneic therapies, allogeneic cell therapies that is, and how you're thinking about edits to your cell therapies to overcome these challenges. So curious, Andre, do you want to start?

Sure. Well, there is always pros and cons in any type of approach. And of course, one of the strengths of autologous therapies is that the cells are not rejected unless the protein gives -- really immune responses that reject it. But the fact is that there is a longer persistence due to the fact that it's autologous. But this is a double-edged sword. It's a double edged-sword. It's good if you're working, for example, in B-cell malignancies because you cannot -- without B cells, it's fine. But finally, it reduces the speed of the use of these type of technologies essentially for like NHL or AML, et cetera, which remains something that is of high value, but very limited. A lot of different targets, you cannot live with the CAR remaining in the body for years and years. And then the CAR remains the target and this -- it has to be something that persists enough to cure all the tumor and has to go afterwards. For example, if you're targeting targets such as [indiscernible] that would be familiar to Pascal or if you're targeting targets such as CD123 which is expressed in myeloid progenitors. You cannot survive without myeloid progenitors. So you have to hit the target and go. So it make more sense if you want to expand this type of technology to a series of different type of tumor-associated antigens and applications to something that would be turned off at a time. And that's why I firmly believe that there is a space for autologous therapies in general. But the more it will grow, the more, I think, autologous therapy are going to be very powerful in hospitals. But the real industrial product that will be commercialized, regulated, et cetera, makes it difficult to spread this through a series of different type of indications and an industrial process at the end. It makes more sense to try to edit the cells, try to work on their fitness state, try to have them extended over a period of time and potentially develop an application of redosing person, the patients, such as a consolidation therapy instead of thinking of what has happened in the beginning of, like, let's say, CTL019, which is today generally a single shot. People get out from like the disease and walks away and everything is fine. This is not the way these type of therapies are wanting to evolve in the future. It's going to be more sophisticated than this, and I believe also it's going to be probably also a potential of combo therapies at the end.

Okay. Rachel, do you want to give us your perspectives on what attributes are important to consider in designing an allogeneic cell therapy?

Yes, absolutely, and I'll hit on a similar theme as Andre did. Of course, one of the key differences for allogeneic CAR-Ts relative to autologous CAR-Ts is that they are foreign and they will be subject to this fairly rapid immune-mediated rejection. And so we are really focused at Caribou on enhancing the persistence of these cells. To Andre's point, it's not about keeping them forever, but it's about keeping them long enough to have sufficient and appropriate antitumor activity. At Caribou, we think about persistence in multiple orthogonal ways. Our lead program, CB-010, is in a Phase I study now, and it's an allogeneic anti-CD19 CAR-T. To the best of our knowledge, it's the first allo CAR-T into the clinic with a PD-1 knockout. And our reason for removing PD-1 using our genome editing technology is to prevent premature T cell exhaustion to maintain these allo CAR-Ts in a high antitumor activity state for a longer period of time. Of course, this is one approach for persistence. An orthogonal approach is to actually prevent that rapid rejection by the patient's immune system. And that's the strategy we take with our second program, CB-011. It's an allogeneic anti-BCMA CAR-T on track for an IND filing next year. For this program, we manipulate HLA Class I presentation through multiple genome edits. We actually removed all of the endogenous Class I presentation by knocking out a gene called beta-2 microglobulin or B2M. And then additionally, we site-specifically insert a transgene that encodes a fusion of B2M with HLA-E. And this ensures that our product candidate is decorated only with HLA-E with an eye to preventing both the patient's T cells and importantly, their natural killer cells from rapidly clearing these therapies. These 2 strategies are, of course, distinct and orthogonal and could potentially be additive depending on the tumor target, and we're today exploring the opportunity to actually combine both of these and a single product candidate in the future.

And William?

So we are developing both autologous and allogeneic. And to be very frank, at this moment, we do see and enjoy both -- advantage of both platforms. And it's not conclusive based on our clinical experience. We have 15 products going through clinical IIT trial to derisk the innovative designs. But what we see with each platform, there are clear need for each platform, autologous, for obvious reasons, right? And also for very clear reasons. I agree with both of you, the convenience and redosing as allogeneic. And especially for some patients, you just can't have enough white cells from these patients for autologous CAR-T, but also the cost and the waiting period and everything that goes to support allogeneic. But again, as Andre pointed out, the persistence, and that is the whole industry. This is the challenge. And whether you want to be long persistence or you want to deal with potential issues or you're putting a switch in and deal with potential CMCs, there's a lot of issues. But I think the future definitely is allogeneic. But the question is still there, how long you want to keep these as persistent and especially moving to solid tumor site. And I don't believe that with a few shots of allogeneic CAR-T you can clear up the tumor mass or in my phasing tumor lysis syndrome at that kind of fast speed. But we're testing like everybody else. We are moving UCART for multiple myeloma in addition to autologous. So we are testing both to find their fit indications in the situation, and that's where we are.

I don't know, Pascal, did you want to add anything to that?

Maybe 2 things to add. One is I certainly agree with William that allogeneic is the future as well as with Andre and Rachel that persistence and the adapted persistence of what you want to have in terms of durability of response is a key aspect there. Now we're looking at that with a very different technology based on clinical experience because we've already treated more than 300 patients with our platform, and we've been proving that it is safe to use our platform in these patients, that there is no safety issue with these EBV T cells coming from healthy donors. And we've also proven that the cells are persistent enough for durable response. Now what we're doing in terms of additional gene edit is really to further stimulate what I call the functional persistent, which is the ability of these cells not only to persist but to persist in a state of activation and memory that allows them to be functionally persistent for long. And we do that not only in leveraging EBV T cells from healthy donors but also in having a new co-stimulatory domain called 1XX, that has been invented by Michel Sadelain at Memorial as well as a PD-1 dominant-negative receptor for solid tumors on these cells. And we have data there at the preclinical level on the PD-1 DNR. And soon, we will present that in a few months at a clinical stage that shows that this is helping the cell not only to persist functionally, but also to resist the immunosuppressive environment in solids tumors. And the final point I'd like to make is the importance of robust scale-up in manufacturing. And we've proven that already. We are making now with a new technology some of our product in stirred-tank bioreactor, where depending on the dose, we can make like with ATA188 or MS product. We know we can make up to 20,000 doses from one leuko part . So we are so advanced being in Phase III and Phase II and close to submission from one of our products that we've been able to develop a very robust manufacturing process that is scaled at a sufficient size to be able to treat patients in a very simple way. And again, we have a very large clinical experience with more than 300 patients treated already.

Okay. And you all sort of touched on this theme already in terms of the interplay between autologous and allogeneic. And as you mentioned, William, allogeneic is the future. But that being said, I'm just curious more specifically if any of you wanted to comment on whether autologous -- whether there'll still be a role for autologous in certain scenarios clinically.

Maybe I can start there and saying -- because I work on both, as you know, Yigal. I think certainly, the autologous are today approved product. Non-allogeneic product is approved. So certainly, they are in place today and they are saving lives today in a CD19 and BCMA space there. Now what's happening is that we see academic teams exploring new targets with autologous, and I think that's great that academic teams can do that. It's very easy for them to do so, and that's going to derisk a number of targets for many aspects there. But I don't see a future for autologous moving forward for CAR-T. There might be a future when you need a kind of oligoclonal TCRT response there, where it might be today with our current technology still a bit challenging to make a product made of very different type of TCRTs there. But we will find a way to get to that point. But that's certainly where the allo space today is if you don't have a CAR-T or one TCR only but you want to have a multiple TCR to address a particular tumor there.

I cannot agree more with Pascal actually on what he just said. Autologous CAR-T today saves lives every day with fantastic products such as Kymriah, Yescarta, great products. There is like also 2 other products that came out, one from BMS, one from Johnson & Johnson with BCMA CAR-T. It is saving lives. But is it sustainable over time to do it in a way that's done industrially is a complication. And I also agree very much with Pascal on one point. It will remain the therapy. But I think at the time, it will loop up and go back to the hospitals and academic centers where it was born. Actually, allogeneic -- autologous CAR-T were essentially developed by like University of Pennsylvania, the NCI, MSK, Sloan-Kettering, et cetera. And it will probably go back again in hospitals when you have a single very specific type of treatment, and it will be slowly replaced completely by allogeneic. So it's kind of a wave, but this wave will disappear. And it will remain, but it will remain at bone marrow transplant as a business for hospitals such as MBM. It's a huge business for them. But it's not the business for pharma companies or biotech companies on the long run. It could be on the short run. Now when you discuss with most of the physicians that do cell therapies, one of the things they will tell you on a regular basis, it's becoming a total nightmare for most of them to treat patients for autologous therapy. Saturated -- totally saturated. The number of clinical trials ongoing are huge. All over United States, all over the rest of the world, it's becoming a complexity in logistics, and market access for these things is impossible. So it's not sustainable if you want to expand this to more than what it is today, and you will have to find a replacement to try effectively to tackle cancer. This is not the way to go.

Very interesting. Rachel, did you want to add anything?

I think my co-panelists have hit the highlights.

Okay. Very good. All right. Well, moving on then. I'm curious how each of you are thinking about the relative interplay between CAR-NK versus the CAR T-cell therapies. Where does CAR-NK have an advantage over CAR-T and vice versa? Are there settings where potentially you could use both in a combo strategy? Curious, anyone's thoughts on that?

Yes. I will be happy to dive into that one, Yigal. But at Caribou, our mindset is twofold as we think about these 2 different kinds of therapies. We focus on hematologic malignancies using allogeneic CAR-Ts, and we focus on solid tumors using allogeneic CAR-NKs. So why? As we look at the CAR-T field and especially the autologous CAR-Ts that we've just been discussing, there are obviously tremendous proof of concept in terms of the power of the immune system to have anticancer activity. I think that emerging field has also demonstrated how challenging it is to harness T cells, in particular, to have sufficient and appropriate antitumor activity in the solid tumor setting. Hematologic malignancies are challenging. Solid tumors are even more challenging as we think about the need for trafficking and targeting, penetrating the tumor, overcoming the immunosuppressive tumor microenvironment and dealing with the underlying heterogeneity of solid tumors. And so at Caribou, our perspective is that natural killer cells are a better start point than T cells to try to tackle that litany of challenges. And we see the natural killer cell as the start point and not the end point. We're actually able to use our genome editing to make a variety of different edits to try to tackle a number of these different issues, including introduction of a CAR for antigen-specific targeting. In order to do all this, we actually start with iPSCs. So at Caribou, we take iPSCs, edit them in a variety of ways to tackle these challenges and then differentiate them using proprietary protocols that we've developed into natural killer cells with antitumor activity. So we believe that this will be a really exciting and compelling approach to address the solid tumor compartment, which is obviously a huge unmet need as we think about the role of cell therapies in cancer care today.

Maybe I can point a different view, and I think that's what the panel is for probably, is to have divergent views as well as some time line in . I mean what we believe at Atara, the T cell or the T cells even in solid tumors. I mean the reason why anti-PD-1 and anti-PD-L1 are working is because they allow T cells to do their work. And I think there have been so much experience now with all these checkpoint inhibitors about allowing the T cells to really do their work and combat the cancer cells that we still believe that T cells are the key for both solid and liquid tumors. The challenge we see with NK cells is that they are fully differentiated cells that are short-lived in vivo. They're not supposed to persist. I mean T cells are there -- are the memory cells. They are the one having -- especially when you have a phenotype in your product that is more into the central memory cells, you have cells that are supposed to persist long to be able to fight infection. And you use that to make sure that they can persist long enough for addressing the tumor there. NK cells are not supposed to stay and they are not supposed to persist there. And that's why there is a lot of engineering techniques there. But if you look at what has been one of the key study to support NK cells, which was the great study by Katy Rezvani at MD Anderson, what people tend to forget is that in that paper, if you read the paper carefully, you understand that these cells that were used in patients and were targeting CD19 CAR were, in fact, HLA-matched, partially matched, for most of the patients, 9 out of 11. Anyway, KIR mismatch -- and KIR mismatch is important for NK cells to be fully put on there. So they were not fully allogeneic from that point of view and even though they were not adding that durable response that is being seen with CAR T there. So we strongly believe that NK cells are possibly interesting and might give some short-term impact, but not to address the importance of durable remission in solid and liquid tumors. Sorry, back to you.

Let me kind of switch back a little bit to defend the autologous CAR a bit since we're doing both. There are reasons. We do have strong conviction with the autologous CAR. You've got to give any technology product life cycle sufficient time. If you look, any new technologies come and go. And what you get from clinical is real. It's a solid. So what we're seeing autologous CAR, especially I'm not promoting FasTCAR, but autologous CAR does give you long persistence without concern of safety issue. And so what we see from my -- our patients, a CAR-T can persist 1 year and 1.5 years, and sometimes they can pop up again. Maybe responding to antigens, maybe not. But there is no off target. There is no other concerns, and this is the beauty. Now CRS and neurotoxicity can be resolved. And how -- this dual CAR for multiple myeloma, no neurotoxicity. And one injection -- you think about it, one injection in 5 days, the issue has gone. Most standard of care for CRS 1 to 2 grade, it's gone, and patients have recovered. 95% of patients achieved CR and for a long time. That's a miracle. Now we cannot cover it up with issues. Autologous does have issues. But right now, this is a proven platform. Now again, like everybody else here, we embrace the allogeneic. What we see from the program of TIL targeting T cell leukemia cells, and our goal was to try to one shot and hopefully last long without stem cell transplantation. So that was the design for that study. But it turned out only one patient last -- I mean CR without any events more than 18 months, but only one. And the rest of patients, 6 months, 8 months or 3 months and then relapsed. Now for TIL, in such aggressive disease, this is remarkable because there is no other standard of care, effective care. But this -- let us think that, well, the design, maybe the persistence. Maybe we need to put certain signal transductions, maybe make it more persistent and so on. There's a lot of work we need to do. So our second-generation UCART for B-cell malignancy is hitting clinical trials. And this is the hope. But I would not hope that much that this is going to resolve, we'll achieve the same consistence as autologous CAR-T. We need -- we have a few years to go to manipulate the molecules that are important for persistence without increase the side effects. And I believe this is where we are. Now back to the NK. The attractiveness, I think, is really drive by the hope of safety. The NK is really safe regardless of cord blood with little mismatch or iPS derived. And I'm really excited about the future of IPS because you can manipulate so much at this master bank without affecting the cargo side, without affecting the gene transduction for CAR-NK or anything that you want to add on at a later stage. That's the beauty. One batch, much bigger than UCART. Right now, we can make 300 doses of UCART, but that's not ideal, right? We should make more to make a more batch, less batch variation, but IPS is attractive. But the current data, at least leading programs by other companies, that the unique actually combination therapy, either additional antibody or IL-2, and efficacy is kind of at the par at the current autologous CAR-T therapies. It is not sexy yet. There's a hope. You can read those, but then lymphodepletion is trick. How many times you are looking for depletion? And can we live without antibody assistance? And if this is the hope for solid tumor, what kind of antibody combination with NK would pull the trick? So there are unanswered questions or need -- a lot of work need to be done in testing clinical. That's our view.

Okay.

Just a little word about like NKs. Just to come back to what you just said, it's combo -- and like what William said, those combos definitely -- the enemy here is definitely cancer. So anything good to try to tackle cancer is great. I agree with Pascal and also with Rachel. I agree with everyone because everyone has an angle. But finally, the answer is not always about the ability to try to blow out the tumor itself. So for example, if you want to go against a solid tumor, what is interesting is the T cell, they're super expandable. So they can expand very powerfully. And what Pascal said definitely makes sense in like the PD-1, the checkpoint inhibitors are very active on T cells. They're not active on NK. On the other side, you see that NK can be very good in cleaning out some part of the solid tumor. So why the concepts are trying to exclude one or the other, et cetera, and not trying to include everything in a potential therapy? For example, we're developing a CAR that targets fibroblast-associated proteins and the cancer-associated fibroblasts that makes this protection for the tumor. You blow it out with T cells. You can have checkpoint inhibitors. It depends the way also in designing your T cells. It could attack the patient T cells after once the immune system goes back and can turn to the hot -- like cold tumor into hot tumor and potentially have a very strong effect on some -- most of the solid tumors that are not responsible to these type of checkpoint inhibitors. And then finally, you can potentially have an NK CAR injection to clean out the situation afterwards. And we have to keep a focus, not on the technology we're using but really on how to try to eradicate tumors cells of the patient as cleanly as possible. And I think all these approaches are very comprehensive. Not that [indiscernible], but that's the concept that we're loving. And I think that's really why NK are really interesting, the single fact that it's more easy to differentiate them from like IPS cells under access, et cetera. So you can do all the engineering you want to do before it can access this. But definitely, T cells are the gate to these solid tumors.

And William, you brought up lymphodepletion, which was on my list of questions. I'm curious how each of you are approaching lymphodepletion question, what strategy or strategies you are using to do the lymphodepletion to improve engraftment. Curious if you could comment on that.

I can start. So we have done lots of things . So of course, we have like the standard, which is Cy/Flu from all autologous space that have been very extensively developed. It doesn't work without it most of the time because you have to create space for the cell to extend. Even if you have autologous cells, if you don't do Cy/Flu for -- the engraftment will not happen so well. So we've done Cy/Flu. But obviously, we have a second protocol. So we made a comparison in the program for UCART22 and UCART123, where you do Cy/Flu with and Cy/Flu on alemtuzumab, which is a monoclonal antibody, that will maintain the immune system for a long period of time to -- for the T cell to extend. And we've seen with the -- like your CAR-T 19 -- or CD19 result, BCMA result that you can have long persistence using the alemtuzumab. You can see the CAR like even 3 months after injection and pretty long duration over 6 months to 1 year of complete remission post the use of alemtuzumab. So alemtuzumab brings a lot to the lymphodepletion regimen. And the way to dose this is to prepare for the Phase II and the expansion capability of the trials with our partners. On the other side, [indiscernible], also a new approach that we've also presented at the last ASGCT, which is cells with lymphodepleting CAR. So CS1 -- the target CS1 is spread on all immune cells, NK, T, macrophages, whatever, B-cells, et cetera. So we do Cy/Flu first. But dialing up or down the Cy/Flu to have more cells in order to enhance the cells to start the expansion, and the general effect makes it extremely interesting. And you can see over 100, 104, 105 days after the injection of UCART CS1, the cells persist for a long period of time. So cells don't actually deplete in CAR. Even though it's not easy to wheel, it's something that is conceptually very interesting. And then finally, what would be interesting is, like, for example, the approach that was being described by Rachel that would also be developing very extensively, which is to knock out the beta-2 microglobulin, which could open to the NK attack versus beta-2 microglobulin replaced by HLA-E, which could block also NKs with HLA-E. And here, it's interesting to see if you would lymphodeplete or not, and that's something that should be done in -- I guess that you have to look to. But it's a very, very important part of the secret sauce behind the CAR-T development journey.

We have, Yigal, a slightly different experience there because we have with Tab-cel in particular and with ATA188, but with Tab-cel specifically on hundreds of patients, an experience of expansion and persistence without lymphodepletion. Because the patients are being treated with no pretreatment, 5 to 10 infusions and that's it. Only as well monitoring. And that experience shows that the product is still able to expand and persist despite the fact there is no lymphodepletion. Of course, these patients that we are treating, most of them are immunocompromised so it might explain what's happening in this patient there. But it's also linked to the fact that we have this partial HLA matching for the product there and to the patients that might also allow more persistence. Now the other experience we had is coming from Memorial who did this clinical proof of principle that using an allogeneic CD19 EBV CAR-T, you then have sufficient persistency to have very long durability of remission. In the patient they treated, they had an 83% complete remission that was maintained for more than 2 years. So -- and there has been no loss of response over 26.9 months median follow-up there. So that's really the 2 experience. But in that case, they had so many for the patient with Cy/Flu a typical one. So what we think, that it depends very much on the status of the patient and how long you want them to persist. There are some of our product allogeneic T cells where we don't need it for the patient, and some others where we'll need some level of it for the patient, certainly not too long and not too heavy.

And maybe I'll dive in here as well, maybe a viewpoint in between these 2. So for our product candidates, we believe lymphodepletion is critical. It's a message we've heard from our SAB, from KOLs time and time again. And the perspective we hear from them is that if you don't sufficiently lymphodeplete for these kinds of allogeneic CAR-Ts, it really doesn't matter what bells and whistles you put on the product candidate using your genome editing if you don't create that window of opportunity for sufficient engraftment. So our strategy is to use the Cy/Flu combo. However, it is a more significant, more stringent approach than others have used in the allogeneic setting. So it leads to a deeper lymphodepletion. It's a protocol that was developed about a decade ago at the NIH. It's been used for autologous CAR-Ts and for TILs so there's quite a lot of safety data collected on it. And we believe it will be important and impactful in the allogeneic setting in order to create that, that we know that we believe will be necessary for sufficient antitumor activity.

Our experience echoes the importance of lymphodepletion. But also, we clearly see the correlation. The level of lymphodepletion versus expansion of UCARTs, it's clear. More you do, then it's definitely creating environment in favor of expansion of UCARTs. But then the cost may be potential infection, and that is a very clear clinical science there. The key question is, I think, whatever we do, the goal is we want to win over autologous CAR-Ts with persistent durations and CRS side effects. And I think that's the key, and we are working on it. And I think what we're trying to do is how long we need? What is the potential downside for the persistence we may face? We may need to explain to regulatory bodies. And I think the heme -- I think in a few years, we'll have a more perfect solution for heme -- for solid tumor. I think even for autologous, we need to break through these areas that Rachel mentioned for infiltration, particularly in tumor marker environment. How do we let the UCART to -- in harmony with the tumor marker environment? I think that's something we need to work on. And we are working on that and hopefully, we have data in 12 months.

Okay. Now all of you have run or are running Phase I studies in -- with cell therapy. So I'm curious if you could talk about some of the pitfalls for running early-stage Phase I trials in -- with cell therapies. What are some of the pitfalls that you would advise someone entering into the same development strategy? Any thoughts there in terms of what to avoid with the early-stage clinical trial with cell therapy?

Maybe one thing to highlight is I think site selection is key, right? As we think about actually deploying these kinds of therapies, and as you and others have highlighted in the discussion today, the potential risk for CRS and other side effects, there are locations across the United States and now across the world who have deep expertise in using these kinds of cellular therapies and being able to quickly detect the kinds of side effects that can be well managed and well treated. So certainly, our experience has been that it's very beneficial and helpful to work with sites who are deeply expert in these arenas.

I couldn't agree with Rachel any more. Where we have -- doing this clinical IIT in China, the variation among hospitals is significant, especially the experience dealing with CRS. And with the same degree of CRS, hospital A could -- who have more experience who can deal with the CRS in a standard way. And then a hospital with less experience may escalate the therapy easily so that you easily rate Grade 3 because if you use it, there's pressure. So that needs to be -- I think experience of the doctors in the hospitals is the key to give you quality data on Phase I. But I think more challenging to ask, and it is our experience, is when you have a normal design, for example, FasTCAR, the dose required for -- to achieve similar efficacy is about 50x lower or at least 10x lower. And what we do in the beginning, trying to -- have submitted applications through Ethic Committees, and they are strongly against from -- we start from low dose. Typical dose escalation. That is a challenge. And they do not want to let the patient miss the opportunity of the benefit of potential therapy, and they want we start at middle dose. They don't want to start at low dose. They think it's not beneficial to patients. That is our challenge. So -- and after first one patient, if they are convinced to see a decent proliferation, and they would agree to escalate higher dose or vice versa. And those are our challenges.

I think I fully agree with Rachel and William there. We have a different set of challenges because we are running a Phase III study with allogeneic cell therapy. We're also running a placebo-controlled Phase II outside of oncology and neurology. So that's also leading to very specific challenges. And what you need, of course, is safety. But you need also to be able to treat a patient in outpatient clinic. Like in a multiple sclerosis trial, we're not treating patients only that are hospitalized. They are outpatient. They come and go for their treatment like for monoclonal antibody. And this is a new experience for many of these doctors so we have to train them. So that's very important because we believe it is the future of allogeneic cell therapy that this therapy will be used like monoclonal antibodies in the future, depending, of course, on the type of patients. But in some patients, it could be used in outpatient clinic with just 2 hours monitoring due to the safety profile that we can achieve there. So moving to Phase II and Phase III is leading to the need to go beyond expert centers and to be able to have a safety profile that is remarkable enough to be able to be used in many, many sites across the world. And we're running these studies across 3 continents in the U.S., in Europe and in Australia. And we've seen that with good training of these investigators, they are able to use cell therapy. Even neurologists are able to use cell therapy in a very safe and efficient way. And of course, oncologists and hematologists are more used to that. But they really can treat patients in an outpatient setting with allogeneic cell therapy.

Okay. Great. We're running out of time, but I'll do a quick lightning round if you'll allow me. Just go around the horn here and ask some company-specific questions. Maybe starting with you, Pascal. Could you just help set expectations for the data for the allogeneic mesothelin CAR-T, the ATA2271? I believe you're going to have data in the fourth quarter of this year. Can you just give a quick overview of what we should expect?

Yes. So this product, ATA2271, is targeting mesothelin. It's still the autologous version. The allogeneic goes to the clinic next year. But this is going to be the first time ever in patients that somebody has used the PD-1 dominant-negative receptor CAR-T as well as 1XX as a co-stimulatory domain. So data will be about safety. Some will be about efficacy, but that will be just the first 2 cohorts. So that's rather limited. But very importantly, we'll have data on expansion and persistence of the cells. So that will be important because of the PD-1 dominant-negative receptor and 1XX as a co-stim domain.

And Rachel, you've just started recently the Phase I ANTLER study in B-NHL for your lead product, CB-010. Curious what you would see as the criteria that you need to see to achieve proof of concept in that Phase I trial.

Yes. Great. Great question. So much like Pascal, this is a first example of a program with a particular kind of modification for persistence. In this case, this is the first allogeneic CAR-T to the best of our knowledge with a PD-1 knockout. And so obviously, we'll be learning not only specifically about the impact of this program in the B-NHL setting, but more generally, the impact of the PD-1 knockout and understanding in what other product candidates it might be important in the future. This is a Phase I study. So of course, it's officially about safety and tolerability. But clearly, the benefit of working in the cell therapy space is the opportunity to collect and ultimately disclose emerging efficacy data as well. So certainly excited to share the first kind of data from this ongoing study sometime next year.

And Andre, for you. Obviously, your pipeline has been heavily focused on hematologic malignancies. But I know you've said in the past, solid tumors are part of your long-term plan. So could you talk a bit about transitioning the pipeline for your cell therapy to solid tumors and your strategy there?

Yes, sure. Actually, we have presented 3 products that are going to go into the clinic starting next year probably on stages because it's going to be a lot. Well, we have our first allogeneic mesothelin CAR-T that we're pushing into the clinic somewhere like in 2022. It had like a specific interest in there, which is the knockout of TGF beta-2R receptor to try to remove all the negative feedback loop that you can have in the tumor -- the solid tumor microenvironment. So the knockout of TGF beta-2R can provide some insights on the way the cells will behave in this type of solid tumors. It's a very exciting CAR. We have excellent results so far in preclinic and it will be pushed forward. The next CAR for solid tumor that we're pushing forward is this fibroblast-associated protein trying to turn cold tumor into a hot tumor that could be done potentially like in clinical trial in combo that could unleash a lot of different targets of tumors that do not respond to checkpoint inhibitors and, in this case, could potentially become checkpoint inhibitor potents. And that could open the gate for a lot of different type of application in this array of different type of solid tumor where fats are really playing some key role. And finally, the last but not least, which is a super interesting -- or very sophisticated that we're going to push in triple-negative cancer, which is [indiscernible] that have 3 knockouts, 2 knock-ins and the TGF beta-2R dominant-negative that will be expressed there. And that's probably for later. It's still early stage. These CARs are in the pipeline in terms of whether it will be pushed into the clinic starting next year.

Okay. Great. And finally, quickly for William. Just talk about the next-day manufacturing. And does next-day manufacturing, in your view, render the off-the-shelf therapies as potentially unnecessary? Or will there still be a role for off-the-shelf therapies in a world where we have next-day manufacturing?

First of all, the FasTCAR or the next-day manufacture still require -- it is person to person. So the QC is each person. It is unlike the UCART. One batch QC, that's it, right? So there is a clear difference here. And of course, FasTCAR is still autologous and it really carries the advantage of the autologous CAR-Ts. And -- but we are very excited to see the safety profile of a dual CAR and it gets so much to manipulate to make a safer but highly efficacious dual CARs. And as our data -- clinical data shows, our high dose now so far has been 11 patients 100% sCR and with a sound safety profile without any neurotoxicity. And we decided to move into frontline therapy for high-risk population. And that's a sort of game changer if that happens. So we got a license to move on frontline, and we're very excited. And hopefully we can present data very soon because this is the representation of CAR-T therapy in frontline, and that's exciting.

Well, thank you all, Pascal, Andre, Rachel and William. Thank you all so much for a very lively discussion. We appreciate it very much. Good luck with the rest of the conference, and we'll be in touch.

Thank you, Yigal. Thank you.

Thanks for the opportunity.

Thank you. Bye.