iwCAR-T 2023 | Updates in AML

Hello, my name’s David Sallman from Moffitt Cancer Center in Tampa, Florida. It’s really my pleasure to be here with two colleagues, Marion Subklewe from Munich in Germany and Roman Shapiro from the Dana Farber Cancer Institute. And we are at iwCAR-T in Scottsdale, Arizona in April of 2023. And we really just wrapped up an acute myeloid leukemia session. And I, I think where CAR-T has really been that paradigm shift in hematological malignancies. It hasn’t quite gotten us there yet, as maybe sort of one common message. So maybe Marion I’ll start with you, like, what do we need to do to kind of overcome the challenges that we had? How do you see CAR-T and then bispecifics as maybe our past forward for patients that are lacking options.

Yes. Sort of frustrating that the results so far and bispecifics and CAR-T cells have been rather underwhelming if you compare to the B-cell malignancies, considering that allogeneic stem cell transplantation has been the most successful antileukemic treatment in AML. So I cannot accept that this will not work, but we have to sort of modulate further. So I think one of the challenges is identifying suitable target antigens. I mean, we know that there’s a huge antigen heterogeneity intra-individually. So that is one of the challenges. And I think why we are trying to find a more specific target antigen, we probably should possibly accept hematotoxicity. I mean, these are all mild add lineage antigens. They are also express

it was in the healthy hematopoietic system. And I think maybe if we go a step back instead of replacing allogeneic stem cell transplantation, one of the strategies might be to integrate CAR-T cells into an allogeneic stem cell approach, part of the conditioning regimen and show efficacy, reduced relapse rate after allogeneic stem cell transplantation might be one of the strategies.

Yeah. No, no, I agree. I think you know that hope for that single antigen CAR is probably not realistic. And I think you nicely showed in some of your data, how heterogeneous it is not only do you have the mark or not, but what sort of the intensity of expression? And so can we choose some combination? And I think what is that combination? How many targets that we need so that we can really try to cover over 100% blasts? Actually, I think those are key translational data that to me are not, not that clear and may be very different in molecular subset. So a patient with an IDH or p53, maybe they have a phenotypic expression that is best to target with product A, product B, you know, I think we’d love to have one product that covers everybody, but maybe that’s not necessarily realistic.

Roman, what do you think are some other challenges we have in sort of the cellular therapy, you know, challenges that we’ve had again in AML patients.

Yeah. And you know, there are many and I think we are systematically overcoming them, but it’s going to take time. But you know, one of the challenges that comes to mind even from our session earlier today are the leukemia stem cells. And I know there’s been debate in the field about whether these things really exist. But, you know, I’m in the camp that believes that leukemia stem cells exist and these are potentially more difficult to target. They don’t cycle as readily, there are fewer of them. They might not express antigens like for instance, what was discussed with NKG2D. And so the ability to target those cells that ultimately prevent even later relapses, you know, let’s say you took 100% blasts and went down to 0% blasts, but you did not really eradicate the leukemia stem cell that can perhaps later, six months or eight months or two years later result in relapse. That will be a key thing I think in AML. And it’s interesting that for NKG2D is an example, I mean, there have been studies and groups that have shown that you can up regulate ligands for NKG2D with certain drugs like PARP inhibitors. As an example, there’s a paper that’s published in Nature a few years ago, there are strategies that one can potentially implement to combine these modalities with cellular therapy to perhaps target the leukemia stem cells.

And I think another challenge that we heard is sort of, what is the T cell fitness? Of course, with autologous products, that could be an issue. So, you know, Marion, I know you’re doing a lot of work. Do you think this is the disease setting, the prior therapies, molecular subsets, I mean, what is the biggest challenge with the T cell fitness? You talked a lot, you know, if we’re just sort of constantly engaging with continuous bispecifics that, you know, we get this exhausted, you know, phenotype in the end. So what do we need to be thinking about, both maybe in CAR-T and bispecifics, thinking about the actual T cell itself. So first of all, I think we need to learn as much as possible from the B-cell field. So, I mean, it has been shown for BLIN that it works best in the MRD setting, low disease burden. So I think this is one of the things we need to improve. We are currently applying bispecifics and CAR-T cells often in patients where there was high disease burden, that had received a lot of prior treatment lines. So I think that is one of the things we can probably translate from the B-cell malignancies. But then I think there are probably differences and we know that T cells from AML patients, even at initial diagnosis are already compromised in function that’s been shown in transcriptional, but also on functional data.

And we know that the T cell fitness is further decreasing with each prior treatment line, we know the T cell fitness is decreasing at time of relapse. And at least from our group, we could show that if you’re targeting an antigen that is ubiquitously expressed and you are sort of giving the bispecific, and I think the same applies to the CAR-T cells is sort of continuously engaging a ubiquitously myeloid lineage target antigen, you’re further inducing T cell exhaustion just like in chronic viral infections. So I think in that context, the target antigen is also contributing to T cell fitness. And in the bispecifics, clearly, I would argue that you have to apply the bispecific and have that treatment free intervals. And that the way you dose your bispecific actually also makes an impact on T cell fitness.

Yeah. And I think we have so many moving parts, right? We have target, we have T cell, we have other immune micro environment issues. And I think we need to be very cognizant of that when we’re developing our trials. So are we at least in multiple different cohorts trying to answer these questions and getting the key sequential bone marrow specimens to maybe help answer those translationally? But you know, maybe, maybe CAR-T cells are not the answer. And I think, you know, maybe Roman, are NK cells the future? We’ve chosen the wrong cell with T cell, NK is the answer? Maybe if you want to comment on anything else. So gamma delta T cells or, or maybe you know, alternative approaches other than T cells.

Yeah. So, you know, I think we’re all collaborators and we have a collaborative spirit and we like to work together. And in fact, that’s my bias in this space. I think that the answer for leukemia will not be one kind of cell. I think that, you know, I’m biased towards NK cells. I work a lot with NK cells. I think about NK cells all day. And so I believe that NK cells can definitely form part of the puzzle to attain a deep remission, sustaining that remission is a challenge. And I think that ultimately, you know, we alluded to this earlier, the success of just bone marrow transplant right. The fact that we can cure 60% of leukemias, at least, you know, that’s at the, at the very beginning, if you think about it, you know, that’s a success of cellular therapy. Can we replicate that in the relapse setting? And with bone marrow transplant, if you look at the graphs, there are all kinds of cells in there, not just NK cells, not just T cells, there are monocytes, there are neutrophils, there are many other cells in the graph that might be contributing to the success of bone marrow transplant. I think that in the relapse setting, you know, if we’re going to use a CAR product, it’s going to have to go into various cell types that will have to work together. In the NK cell, in the early phase NK cell trial that we talked about in the session, you know, I that BPDCN case example was just, you know, we were so interested in this. I was trying to understand this better. Is it how is it possible that we infused a bunch of NK cells and a bunch of T cells ended up in the tumor site where there were no T cells there before? You know, clearly there is interaction between the different immune effector cells. How to take advantage of that, how to make them work together to accomplish a sustained prolonged remission, I think is going to be ultimately the way forward.

Yeah, I think we’ve heard it’s, it’s unfortunately going to be, be, be quite complicated, maybe like just taking us home, you know, Marion. Like, what is, what is your sort of message of optimism, you know, are we going to have our, you know, immune therapy breakthrough and how are we going to get there?

Yeah. So I really caution that we conclude from the data. We have to be too pessimistic, right? So I think we really have to remember the success of allogeneic stem cell transplantation. We have all the data also from the B-cell malignancies that platforms do work. So I think we are need to apply our CAR-T cells and bispecific in a different clinical scenario. And we have to move earlier into a clinical trial. And then I think we have to think of combinatorial strategies. I mean you are conducting this trial with also membrane IL-15 bound. We have to think of combinatorial small molecules with maybe CAR-T cells bispecifics, apply this in a low disease tumor burden early in treatment line before we kill a platform, just because we don’t see efficacy if we apply this now in fifth line of therapy. So and I also think AML is a heterogeneous disease, so probably one approach is not applicable to all kinds of AML. So we also have to adjust to the different genotype and phenotype of AML, making a little more complicated I must admit, but I think there’s so many aspects that are obvious for improvement that I think it would be a mistake to give up at this

point.

Yeah, I agree. I think this, it’s just highly endorsed that obviously we need clinical trials and actually to get these clinical trials going as soon as possible and maybe even, you know, reaching out to our regulatory authorities, preventing some of these, they’re kind of very harsh rules on getting these trials, getting enough patients treated quickly enough so that we can really try to see if the therapy is efficacious or not. But I think with that, you know, just thank you for listening, kind of for this AML wrap up at iwCAR-T.