Amer Zeidan:
Hi everyone. Welcome to this new episode of MDS Sessions with VJHemOnc. So today it’s a pleasure to be looking at the main MDS presentations from the American Society of Clinical Oncology meeting, as well as the European Hematology Association meeting. My name is Amer Zeidan from Yale University, and it’s a pleasure to be accompanied today by Dr Andrew Wei from Alfred Hospital, as well as Dr. David Sallman from the Moffitt Cancer Center, both very well known experts in MDS. So we’ll be discussing some of the most important abstracts and foresee how things are going to look in the next two years in the clinical research and translational research for MDS arena. So maybe I can start with Dr Wei. So from looking at the list of presentations for both meetings, what caught your attention in MDS?

Andrew Wei:
So from a therapies point of view in terms of more high-risk disease, there are a number of abstracts presented looking at the role of pevonedistat, venetoclax with azacitidine and also sabatolimab plus hypomethylating agents. And there was also a presentation of IDH inhibitors such as enasidenib in combination with azacitidine albeit the frequency of IDH mutations in MDS is quite small.

And there at least one new agent caught my eye on that was the IRAK4 inhibitor presented by Garcia-Manero. I thought this was particularly intriguing, even though it was a very early stage study and it dovetails a little bit with a study that I presented with regards to sabatolimab and that is the growing role of the inflammatory and inflammasome pathway in MDS. And there was a more summative presentation given by Julie Schneider from Leipzig, which was on the topic of the inflammasome in MDS and so she found that high-risk disease was associated with more inflammation activation in MDS, and so this perhaps gives some biological rationale to IRAK4, which is targeting the inflammasome in more the earliest stages, and with respect to sabatolimab being shown to reduce IL-1b, one of the key effectors of the inflammasome pathway, after sabatolimab in responding patients, I found quite intriguing.

Amer Zeidan:
That’s a very nice connection on the translation a little bit with multiple agents being studied. How about you, David, what caught your attention?

David Sallman:
Yeah, just to piggyback, so we’ve had a lot of interest in inflammasome targeting now for a while. So for example, we do have a canakinumab low-risk MDS study, it’s just initiated accrual in 2021. And to go along with the interesting mechanism where TIM3 may impact this as well, there is a basket trial that has been launching where IL-1beta and TIM3 and then other combinations are being looked at. I think, one thing to step back, a lot of studies actually have looked at in lower-risk MDS, some of the inflammasome components are significantly over expressed versus higher risk, but clearly they may play a role in both and I think it’s an interesting space where it’s easier to test agents earlier, of course, in HMA failure MDS where we don’t have anything, but thinking about potential disease modification in lower risk and how these things play together, I think are important since we really don’t have any disease modification.

And I think these translational correlators are really very, very key. I think there’s still a lot to understand from the MOA perspective and we can comment with sabatolimab, magrolimab and others, is it the immune changes or this case inflammasome, is it LSC targeting? And I think we’re a little bit stuck with the fact that these trials are all single arm studies and so what is really different between response and change in the immune microenvironment with response, and then what’s truly occurring with these novel agents? Because then I think it can help us direct the next set of studies that make the most rationale, just to comment on that.

To me, I thought some interesting abstracts are looking at how we utilize HMA-venetoclax, so for example Jacqueline Garcia had presented one abstract looking at dose modifications in MDS. So as we know, we already use a 14-day regimen instead of the continuous regimen in the beginning. But what I thought was intriguing is whether or not you dose adjusted the HMA, you change to seven day, it really didn’t make a hill of beans of difference as far as the ultimate outcome. So I think dose adjustments and really very close monitoring with these patients with early bone marrow biopsies is key, but I think still, is one way better than another? Potentially not, I think, based on that.

I think similarly, what I thought was interesting is Keith Pratz. MRD has not been as much of a thing in MDS, but I really think particularly with all these novel doublets and then maybe as we think about the path forward and triplets, ideally looking at MRD in our MDS patients probably by high sensitivity next-gen sequencing via whatever technology, I think will ideally give us a lot of insight. But you’re looking at MRD flow, and I think it was potentially an AML abstract, but less than 10^-3 three versus still predicted further stratified prognosis long-term. I really think these studies are going to be key in our MDS populations, particularly with these doublets going forward.

Amer Zeidan:
Yeah, and to take this same issue forward and also touch basing on some of, I think, the drugs that are making it forward, I just want to get Andrew’s thoughts on the pevonedistat. So pevonedistat is one of the drugs that have really completed randomized Phase II now which has been reported. The Phase III has completed accrual and results are expected in the near future, and it could be one of the first drugs that go before the FDA, depending on the results of the Phase III. So Dr Sekeres presented an update looking at some of the molecular clearance and the reduction in variable allele frequency that occurs with this combination. And it’s an interesting first-in-class agent, which is always nice to study in myeloid diseases. So I want to get your thoughts, Andrew, about how do you see the clinical benefit of pevonedistat in what has been presented so far from the Phase II trial, of course, while we wait for the Phase III results.

Andrew Wei:
Yeah. So Mikkael Sekeres presented an oral at this year’s EHA Meeting, looking at pevonedistat with respect to mutational clearance using a NGS-based approach and also there was a paper published in Leukemia recently, which actually looked at some elements of the Phase II randomized trial of aza plus pevonedistat versus aza alone. So I think the highlights of those studies are first, the response rate is quite impressive in terms of the combined response rate. The durability is also quite impressive, however the survival curves are, from the leukemia paper, come together over time. So that study was a little bit different from other MDS studies in that it merged patients with CMML, with oligoblastic AML, and also high-risk MDS. And it does beg the question as to whether, in fact, MDS with high levels of blasts and AML with low levels of blasts is a real distinction between those entities or in fact, is it just a continuum of a similar disease, but just more of a spectrum rather than a dichotomous split between MDS and AML. And I think it’d be really interesting in the next iteration of WHO where patients with blasts above 10% go because there was a recent abstract by Elihu Estey arguing that MDS with EB2 should in fact be classified or treated as either MDS or AML, but basically both and not really distinguishing one from the other.

Amer Zeidan:
I agree, and they actually just published a letter in Blood making the same argument that a blast count of 10% to 30%, that those patients should be considered eligible for both AML and MDS trials when they are unfit for intensive chemo, which I think, for many of us who treat these patients day in and out, this is our clinical sense that there is a lot of overlap and the 20% cutoff is somewhat artificial.

Building on the same theme, David, you talked about aza-ven and the dosing, and now we know that aza-ven is approved in the frontline setting for older unfit patients with AML and there has been a lot of data presented in the last few years from the ongoing studies for high-risk MDS. So since venetoclax is available already in the US, and you mentioned the issues that come in terms of how do you manage the dosing and the differences in the dosing and the drug interactions and the myelosuppression, all of these, nuances. Are you seeing a lot of off-label use basically in Florida or around your area, basically with venetoclax and MDS?

David Sallman:
Yeah. So I think it’s an important point. I think we can all, I think, definitively agree that single-agent HMA is a quite underwhelming therapy for frontline high-risk MDS. To get a complete remission in 15 to 20% and less than half of patients having some response and you also have to wait for a very long time. Three to six months with coming to centers twice a week and transfusions and all of this is quite burdensome on the patients as well. So particularly for patients that have excess blasts, I really don’t utilize single-agent HMA ever, with maybe one caveat that I’ll talk about in a second.

So clinical trials need to be highly prioritized. All of the agents, we really need to definitively answer these questions and so all our 500 patient Phase III studies. So that really needs to be of paramount importance going forward, really across the globe so that we can really, again, answer what the durability of outcomes, molecular subsets, is survival truly being improved. So that’s the top priority. I think though, it’s clearly not possible for everybody and I think we’ve learned a lot. So I use HMA-venetoclax quite extensively for p53 wild-type patients. So that’s my one caveat, given that the group has not appeared to do any better with p53, and I do find that the cytopenia prolongation, and maybe it’s just a factor that the quality of response is not as good, is quite marked in that patient group. So if they are p53 wild-type with excess blast off study, I use HMA-ven and pretty much all of our patients approval is quite easy.

I do a day 21 bone marrow biopsy. We then hold to allow count recovery and often within less than two months of time, you can actually achieve your response, and again, I think similar to what the data that we’re seeing. And even if patients have been exposed to HMA for a short amount of time and then they come to our center, we find that adding venetoclax back, and hopefully we’ll maybe present some more data future, but we published just a smaller paper with Sloan Kettering Group that can also be effective, again, probably in the setting of p53 wild-type.

Amer Zeidan:
And it’s going to be interesting to see how the landscape changes in the real life practice as some of those agents become more available. So I guess now to dig more into some of your own presentations and I’ll start with you, Andrew. So you presented an update from the ongoing Phase I study of sabatolimab, which is I guess, a targeting agent that affects both the myeloid cells, but also as an immune checkpoint inhibitor, and we are still trying to fully understand the mechanism of action. The clinical data has been presented in combination with HMA more than once now, but your update was more focused on trying to understand at the cellular level what’s actually happening with the agent, but I also thought it was interesting to see some of the molecular subset data like the TP53 data that you presented. So maybe you can give us some overview of what presented and your thoughts on that.

Andrew Wei:
Yeah. So in addition to the sabatolimab paper, we also had a poster on ven-aza, which David mentioned, and so I think it’s interesting to actually compare those two regimens because the common question we’re asked is which one do you think is going to be the future? And I wonder, in fact, it’s not necessarily going to be one or the other and that a variety of different therapies will be appropriate for different patient situations and clinical goals. So for instance, if the goal is to take someone to transplant and you want to get a remission quickly, then I think it’s hard to go beyond ven-aza because the response rate is 80% and they happen very quickly. With respect to if you don’t have a transplant as your end game and you want something which is more tolerable and you want to keep the patient on it for a long period of time, then some of these other options might be, in fact, are more amenable, for instance, magrolimab or sabatolimab or even pevonedistat.

And with respect to p53 mutation, the data is still emerging. Obviously we have, I think, the last presentation on MDS from David, there was, I think, four patients with p53 mutation in the MDS group. Response rate there was 75%. We now have some data from ven-aza where the response rate is 80%, but CR rate 16%, but obviously there is the concern that the durability of response is still unknown. And from the sabatolimab experience, there were 14 patients with p53 mutation. The mCR rate was 50% but the duration of response was 14.7 months, which is interesting. And one, I guess, unusual aspect of the sabatolimab study was that the responses, as you mentioned in these higher-risk groups, didn’t appear to be lower than the overall group, which is usually not the case for conventional therapies. Whether there’s something biologically relevant with respect to targeting TIM3 and preservation of response in higher-risk disease I think remains to be proven.

The only other thing I’ll mention is that with the ven-aza study, it’s important to note that almost two thirds of the patients actually had EB2, which is an unusual feature for an MDS study and I think it’s because our experience in AML with obviously more blasts was obviously very strong, and I think physicians probably preferentially enroll their patients to MDS studies with EB2 because they know it’s almost AML and more likely to work and to work quickly.

Amer Zeidan:
Very good points. David, you also presented a magrolimab over the last one to two years now, and I don’t think there was a presentation about APR-246 this year, but there has been an interest since for a few years now and the publication of the Phase II data that you and the French group have done in the last couple of years. What’s your sense of this TP53-targeting type of approach with both of those drugs and where do you see this field going forward?

David Sallman:
Yeah, so I think the start is I really think we need to subdivide these patients out and we really need to think about p53 focused studies and not… And I know it’s tough when companies want to go for broad label, irrespective of molecular subset, but I think for better or worse, this group really needs to be targeted specifically. I think the things that we’ve learned to date is that response rate by itself is probably a useless criteria, at least from just a pure CR or composite CR response rate. We know that obviously a lot from the AML despite 55-60% response rates with HMA-ven for this group, you saw the median OS of around six months.

And so thinking about this group separately, I would say the largest cohort period in AML would be the magrolimab data that we presented last year with data now of over 40-something patients. The follow-up is too short, so I think the next presentation of that data will be extremely clear. Follow-up was a little bit over five months with a median overall survival of around 13 months. But we know, particularly with magrolimab, that responses do continue to deepen over time and thus that actually may improve. And if we’re really starting to be past 12 months, that seems to be, at least in AML patients, quite strikingly of a difference.

In MDS, as Andrew mentioned, the p53 cohort is small, but I think, to our earlier discussion, this is the 10 to 30% blast group. I mean, 80-85% of these patients probably fit there. They’re very high blast p53 is somewhat uncommon and often associated with other co-mutations, so I think this as a group, it’s a biologically, probably singular entity where high VAF across both subsets is associated with poor overall survival, they typically have a lack of other mutations. So I think we can think about our MDS and AML studies going forward.

I think also too what both of you mentioned is the path forward is going to be multiple combinations, and so how do these drugs work specifically in the p53 subsets? So at least with the magrolimab, the leading theory right now is the balance of pro-eat me signals is what leads to the ultimate synergy with very low, if any, activity with single agent. And so we know that HMA upregulates pro-eat me signal calreticulin and now with multiple groups, not just with magrolimab, but with ALX and I believe another agent, they’ve shown that for example, venetoclax can also tip that balance. So in HMA-ven 47, now I already said, is that a good idea in p53? Maybe, maybe not. I think if that balance is critical, I think still would be worthwhile testing. At the same time could we have combined agents like an HMA, APR-246, and 47 agent? How does APR impact pro-eat me signals? We don’t know. But could we go through sort of a dual pronged approach? Because again, a median OS of 12 months, even if that’s a significantly better, it’s nothing to take home and we need to really think about the next level for these patients. APR and magrolimab have the most data, I think, from that perspective, but we need to think about sabatolimab. We know that there are clear immune differences and so would adaptive and innate immune activation have a dual pronged benefit in this group? We need to test it. So I love a lot of different triplet trials and maybe we can have basket studies if we can get companies to work together, specifically for molecular subsets like p53.

Amer Zeidan:
So before I go to Andrew, I wanted to follow up on one point, David. So there has been a lot of talk about TP53 mutation targeting with anti-CD47 agents and there has been a lot of concern about lack of clarity of why this subset would be particularly sensitive, or is it just because those patients do very poorly that any drug that generally works for MDS and AML in a very good way could potentially show effects in a more prominent way? So what’s your current thinking about this issue?

David Sallman:
Yeah, it’s a great question. The latter is very easy to say, “Hey, this is a bad group that now is appearing to do better.” We and others have published that they are quite dramatic differences in the immune microenvironment in p53 versus wild-type, so that alone may have differential sensitivity to immune modulating therapy, I think is a question. Even could PD-L1, PD-1 therapy, although those have not been overly impressive, could this subgroup potentially have a benefit? That’s a little bit off-topic. I think there’s some thought, although not proven, is the cytogenetic complexity of these patients. Could that tip to a higher pro-eat me signal? I would say we’ve tried to look at it a little bit and we don’t think 47 is the biomarker. From calreticulin expression in a small subset, I would say was not dramatically different, but that’s just one pro-eat me signal.

So I do think there needs to be continued work. Is there something truly novel about sensitivity with p53 or it’s a group? I think in the end, the clinical data trump everything else and we will learn on the backend as far as all approvals in MDS, how they truly work or how they may truly work for subsets. But clearly the signal is there and I think, especially in the randomized trials, we can start to understand better what’s different and single cell RNA sequencing may be very insightful from that perspective, which is easy enough to do nowadays.

Amer Zeidan:
Yeah, so Andrew, in terms of thinking about the combination therapy, you and David both brought up the combination and although we still don’t have even a doublet approved at this point, we are already thinking about triplets, maybe quadruplets in the future and following that multiple myeloma paradigms. And I keep thinking that MDS patients are generally much older than most of hematologic malignancies. Most of them are in their seventies and the myelosuppression with the bone marrow environment being quite diseased and not a lot of residual hematopoiesis. So how do you see the feasibility beyond the appeal in some selected younger patients who are very good fit? How do you see for the overall population of MDS? Do you think these will have overall implications for the entire MDS population or are we just going to be talking about selected subsets who can tolerate such triplets or more intensive therapies?

Andrew Wei:
Perhaps for us in the last couple of years, the biggest clinical change with these new therapies is the perception of patients, perhaps not so much as these older patients who are really just going into a palliative course for a period of time and there’s no long-term future. We’re now looking at every MDS patient that’s coming through and the first question we’re asking ourselves is can we do a reduced intensity transplant on these patients? Because we now have agents which can give us really high response rates that we’ve never seen before and I feel the discussion is not so much a matter of doublets, triplets and which agents, but it’s which combination to get us to a transplant and when to do the transplant. For instance, do we just wait for blast clearance if EB1 or 2 was the problem, or should we try and go for cytogenetic remission? And as David mentioned before, should we even try and go for MRD remission, which could be, in fact, NGS-based before we go to transplant to basically make our big shot, which is the transplant, the most effective as possible.

And so I think for me, this is the central set of questions. Which new agents will get us to transplant? When to do the transplant? And maybe like in AML, should we be doing something after transplant, potentially MRD-guided as well? What we really need to know, I think is for instance, is just blast clearance enough, because we usually get that more quickly? If we also have to wait for cytogenetic and other MRD clearance, then that’s going to delay the transplant. Does that come at a better cost of progression? Maybe not so much in MDS as opposed to AML and so I think these questions are really critical, which hopefully future studies might address in a more systematic manner, because at the moment, a lot of it’s just based upon institutional and individual practice.

David Sallman:
Yeah. I think that’s really, really key. I would love to start to see some of the transplant outcomes from some of the early phase patients, because that ultimately are, can we cure a significantly increased proportion of patients? I think that’s a real hope. There’s been anecdotal discussion across that, but really seeing these data, and then as you said, looking at MRD pre, post. Of course, you’ve done a lot with oral hypomethylating agents. Can we think about those strategies with some of these agents in the maintenance approach?

And then maybe just to throw a wiggle in it, what about induction chemotherapy in high-risk patients? There are several trials with big COs going on, with MDS EB. There’s not really data to date. Again, I think we would not do that in the p53 setting, but what about in the wild-type patients? I think it’s something for us to at least to think about. We have this different paradigm but we already said we have the same question in our fit 60-year-old. Is HMA-ven or chemo a better route for a fit patient going to transplant? To me, that’s maybe the biggest unknown question in older, but fitter acute myeloid leukemia. But I think we’ll have the same question. The paradigm has never been there, but maybe as some of these trials with big COs or other combinations are looked at, I think they’re really important.

Andrew Wei:
Definitely. Even historically I think when we’ve had patients with EB2 and obviously MDS, it’s always been a little bit less attractive to go with intensive chemotherapy because one, patient is going to be in hospital, two, complications is going to be high and three, the patient’s condition for transplant is going to be sub-optimal. Whereas I must admit, ven-aza with the 14-day shoot just hits that sweet spot where we’re going to get a response rate as pretty comparable to intensive chemotherapy but I think without all of the downsides of hospitalization and high infection rates. So I think for us, it’s hard to go past that at the moment, but I think the key question is for p53 mutant patients, is magrolimab, for instance, the way to go? And I’m really interested in your thoughts in terms of when do you take patients, because we’re doing the magrolimab pivotal study at the moment and obviously the question is, do we just go for blast clearance or do we wait for CR? I know the study wants that, but what’s best for the patient? Do we wait for cytogenetic clearance, MRD clearance, but as you know, that’s going to take more time to achieve. What would be really interesting what you do for your patients?

David Sallman:
I think what’s nice, at least with the magrolimab data is, although the durability may not be forever, it’s not like these rapid progressions. So you have time to get them through, let’s say, six cycles or even somewhat more. I don’t even know if we reported a patient that has progressed more quickly than that, in this setting of p53. So it’s not like, I think with ven, that’s our real fear in p53, things can change that quickly. I think with magrolimab that not been the case. And we just published a paper in Blood Advances. I think the cohorts are small, so we really try to look just with commercial NGS, 5% VAF, not MRD. So I think that’s a real critical question going forward. When we look at the p53 VAF, how does that impact?

So if you did not clear p53 mutation, your outcome was identical to just continuing HMA monotherapy. No difference whatsoever. Whereas if you cleared prior to transplant and again, not at MRD, that’s the only group that had benefit. And it’s still there. The tail of the curve is not good, but those were the long-term survivors were patients with clearance. So that’s actually how we’re practicing essentially. If we do not get at least a high level of p53 clearance, then we’re thinking of other novel therapies and not taking them to transplant. I think that’s probably the most important marker to predict outcomes. Again, the cohorts were in the seven to nine patient realm, but if you look at the New England 10 day decitabine, which was somewhat intriguing for the transplanted group, the same number of patients.

Amer Zeidan:
Actually, I take you up on that thought and ask Andrew for his thoughts about some of the new agents. There are so many drugs in advanced Phase III testing, it’s quite amazing and in high-risk MDS. But I wanted to get your thoughts, Andrew. We had the MCL-1 inhibitors basically come as part of interest in targeting apoptosis pathways and there was a lot of interest and then things seem to become not as active because of the concern about some of the toxicities. But there still a very good preclinical rationale and I know you’ve done a lot of work in this area for myeloid malignancies. So do you see that there could be a resurgence of those agents and how do you see they contribute to management of MDS patients?

Andrew Wei:
So as you’ve mentioned, the first thing we learned with MCL-1 inhibitors, despite incredibly strong preclinical rationale, the combination with BCL-2 inhibitors was going to cure lots of diseases. I think what we’ve found is that the therapeutic window is much more challenging than we had expected and that was partly because all the preclinical testing was done obviously in murine models, but murine MCL-1 is not as well targeted as human MCL-1 by the drugs and so there was a one log difference in binding potency. And the other thing is that when we went into patients, obviously a lot of these patients had received a lot of anthracyclines, a lot of patients are older and one of the key issues of targeting MCL, which we know again from basic biology, is that there is going to be a cardiac signal, there will be a gastrointestinal signal and there could also be a liver signal.

So what we need to do now is to work out, can we lower the dose of the MCL-1 inhibitor to a level where we can stay below the threshold of causing complications to those critical organs, then combine it with the most synergistic partner possible such that we can get the clinical efficacy without having to raise the dose of an MCL-1 inhibitor? And is intermittent schedule and picking the right patient that doesn’t have a huge anthracycline burden, cardiac co-morbidities, liver tox, et cetera going to give us that therapeutic window?

Some other interesting approaches to try and come to the same outcome is to use things like nanoparticles, which are thought to concentrate into the bone marrow more effectively than just giving the drug as a naked drug systemically. Can you encase MCL-1 inhibitors in nanoparticles, which can then concentrate into the bone marrow and then target the leukemic cells or MDS cells more than the heart, liver and gut? And another mechanism or approach, which will be a technological hurdle, but if it can be surmounted could be a massive breakthrough, is can you conjugate an MCL-1 inhibitor to an antibody which is now directed towards the target of interest? If you could do that, then you could literally deliver an MCL-1 payload to any tissue you like, as long as you’ve got a fairly narrow spectrum antigen. So I think MCL-1 inhibitors remain biologically strong. It now becomes a clinical art of trying to find a therapeutic window with a naked drug or a technological hurdle to overcome. Can we deliver it where we want to, away from critical tissues?

Amer Zeidan:
So David, on the same line, we have, I guess, other targets within the apoptosis pathways and the CD47 targeting with magrolimab was the first in class. Clearly there are a number of other agents targeting both CD47 and the SIRPα interaction. So what’s your view on those agents and the differences that they have from the magrolimab?

David Sallman:
Yeah, so I think in the beginning, there’s been a lot of concern, obviously with RBC binding and anemia, and we know we get that with magrolimab, although with the priming strategy and this process of RBC pruning where magrolimab-exposed RBCs actually shed CD47 and not really an impact over time, I think it’s been less so. My some somewhat concern with some other agents is the lower binding to RBC going to potentially decrease potency CD47 receptor occupancy on the leukemia cells? So I have that concern. I think that being said, lemzoparlimab, which is also an IgG4 monoclonal antibody, which may have differential RBC binding based on glycosylation differences, that’s gone full dose escalation, has not had anti-drug antibodies as you presented issues with the Celgene compound in the past, it has gone through full dose escalation and already plan to the frontline trials, given partnerships with venetoclax, will be ongoing.

I think the SIRPα fusion proteins such as TTI-621, TTI-622 are interesting. I think the IgG1 domain is probably not optimal because you can have a dose-dependent thrombocytopenia and they probably did not reach their full PK potential based on that. So TTI-622 being an IgG4, I think is interesting, to go forward. They have dosed significantly higher and there’s some hint of monotherapy activity. What does that mean? It hasn’t really, at least in the lymphoma trial to date, shown them robust increased synergy with rituximab, but I think it’s something to look at.

And then I think thinking of the way of the future, can you target CD47 and other? So there is, for example, the CD47-CD40 ligand inhibitor that’s going to be entering trials. There are potentially bispecific agents, although that can be a little bit of challenge given the ubiquitous expression, but potentially 47 plus other, I think is intriguing. And potentially, like with an LSC eradication, you’d be able to partner with another LSC targeting, even in the setting of CAR-T, like a bispecific CAR could be something of the future. But I think for a frontline, probably maybe looking at magrolimab, lemzoparlimab, I think thinking in the relapsed or novel-novel combinations, some of these other agents may have interest to be looked at.

Amer Zeidan:
Yeah, this has been a theme, I guess, across immunotherapy where monotherapy might not necessarily give you all a lot of activity in myeloid malignancies, but the combination seemed to go much beyond the monotherapy. So thank you so much to both of you. This was a great discussion. A lot of exciting developments in MDS and the field is going forward at a very rapid rate. It looks like we are going to have some major change in the landscape over the next few years. Thank you so much and I look forward to talking to you on subsequent sessions in the near future.