Amer Zeidan:

Good morning, good afternoon, good evening, everyone. Welcome to another episode of VJHemOnc MDS sessions, where we’ll be talking to three of our top experts about the management of patients with lower-risk MDS. We’ll be focusing on anemia as the main clinical problem that manifests in patients with lower-risk MDS. The discussion will actually focus on the management of patients in a practical setting in your clinic. How do you do it? How do our experts think about their decisions as they approach patients in terms of initiation, in terms of continuation, in terms of assessing failure, and the next agents they go to?

My name is Amer Zeidan. I work at Yale University, and it’s a true pleasure to have three experts with us today. From a European perspective, we have Dr Sophie Park, who is a professor at the University Grenoble Alpes, which I hope I pronounced correctly. And a top expert in MDS focusing on lower-risk and ESA use. We also have Dr Yasmin Abaza, from Northwestern University in Chicago, and Dr Maximilian Stahl, from Dana-Farber Cancer Institute. Welcome to you all, and thank you for joining me. Maybe I can start with Dr Park. Sophie, you have worked for a very long time on MDS and ESA use. What do you see as some of the common challenges or the misconceptions about ESA use when doctors treat it, especially those who are not very familiar with MDS?

Sophie Park:

Thank you, Amer, for the kind invitation, first of all. For the first question, in Europe, you have EMA authorization for EPO alfa, and you have different kinds of erythropoietin. As you know, there are no comparison trials of different erythropoietin use. The first challenge is perhaps the homogenization of the dosages, and the use of the kind of erythropoietin. But as you do, in the literature, there were a lot of meta-analysis comparing the response rates of different kinds of erythropoietin.
Finally, it is not the kind of erythropoietin which matters, but probably the dosage. It has been shown in two meta-analysis that high doses of erythropoietin yield higher response rates. For example, for erythropoietin alfa, about 40 to 80,000 unit per week of erythropoietin is about 60% of erythroid responses. Whereas when you use about 20 or 30,000 unit per week, you yield just 30 to 40% of response rate. For darbepoetin, for the glycosylated form of erythropoietin, it is between 150 microgram or 300 microgram per week.

First of all, it is dosages. As you know, when you evaluate the response rate to ESA, you evaluate it at around three months after the initiation of ESA. It has been shown that the higher response rate you have, the higher duration of response you have, so it is important to have early response rate to ESA, and longer duration of response to ESA. Otherwise, if at three months after the initiation, you have a ESA failure, it is hard after to maintain duration of response to ESA. But, of course, you have to adapt and modify to the patient you have in front of you. If you have an elderly, very elderly patient, with renal insufficiency, you will not begin with 60,000 units of erythropoietin per week. The main message I would say is higher dosages of ESA at first, and then after you can adapt after. The kind of ESA does not matter because, mainly, the response rate should be the same between the different kinds of ESA.

Amer Zeidan:

Before I go to Yasmin, just one last point I want to check with you. Have you ever seen patients overshooting their hemoglobin in MDS? In your practice, from a practical sense, have you ever seen patients go to hemoglobin of 14 and 15, and have cardiovascular. Because this is being, I guess, used as the main reason why that you start lower, and you have to be very cautious. In my experience is not responding, rather than overshooting a response.

Sophie Park:

Yes. I’ve never seen 14g/dl of hemoglobin after one month of ESA. When I initiate, I’m taking blood counts every week, the first months, and then after every 15 days, or every month, after three months. Early, you just look after very tightly, and then after you can space, but you don’t have high levels of hemoglobin in the first weeks of the initiation of ESA.

Amer Zeidan:
Thank you. Yasmin, going to you. Just following up on the same practical approach. I agree with Sophie. One of the problems I see in the community is underdosing. Sometimes people use these renal doses, which are much lower doses of ESAs, rather than what we do to manage MDS, which is generally not a problem of ESA or of EPO production. It’s more of a problem of bone marrow resistance. This is why you need supraphysiologic doses of replacement, rather than replacement, which is what is done in kidney failure. But how do you approach the issue of how do you choose between short-acting, long-acting, and the dose? Do you start with a high dose and then go down, or do you start with a low dose and then go up?

Yasmin Abaza:

Thank you very much, Dr Zeidan, for the invitation. I actually do, because most of these patients are older, and just going weekly is not something that a lot of patients like to do. As Sophie said, there’s no difference in the type of ESA. My practice is to go with a long-acting darbepoetin every two to three weeks. I typically start with the high dose. I go with the 500 micrograms from the get-go, and I give the treatment every, about, three weeks or two weeks, depending upon how bad the anemia is. I just continue doing that. I have not really seen any overshooting, but usually, within 12 weeks to… I try to give up to 24 weeks, just to make sure they’re indeed ESA failures. But, usually, patients actually do very well after 12 weeks or so, and if they do respond, it’s adequate. I’ve had a couple of patients where I had to even stop the ESA completely because they’re above 10, and some of them had six months with no need for any growth factor and they did very well. When they respond, they really truly respond well, and then sometimes I use more, like space out, rather than decrease the dose of ESA.

Amer Zeidan:

Yeah. I think this is also a very important issue, in terms of waiting enough. I think most of the guidelines are just, and I think Sophie mentioned that as well, is around three to four months, you generally, if you don’t see a response at that point, this is the time to think about moving on now. We’ll discuss ESA failure, and the differences more a little bit later. But one of the other practical issues, Max, that I want your opinion on, and then I’ve seen it, we actually have conducted real life analysis, it’s an old paper by now, almost 10 years ago, but at least in the US, many docs were not even checking EPO level. They were just starting ESA regardless.
Maybe back then, because we’d not have good alternatives, so whatever was EPO level, what are you going to do? You probably are going to give ESA if the patient has MDS. But the other thing I have seen is people continuing the drug, again, continuously. I have seen patients who are on two years of ESA while they are still needing transfusions, because the patient says they feel better, or the doctor says, “I have nothing else to do.” I think the transfusion needs have reduced a little bit, but it doesn’t seem objectively it has reduced. What’s your sense of some of those common, I guess, misconceptions or errors that you see in the community around you for ESA use?

Maximilian Stahl:

Yeah, thanks Amer. I would say the point that you mentioned already that EPO levels are infrequently checked. When I get referrals, often the EPO level is not checked. That’s actually a little bit hard, because then once we put people on erythropoietin, of course, EPO level changes a little bit. But I think that’s actually very helpful, because this has been described, also, already in the ’90s, by Dr Hellström, who developed, actually, a score predicting of how likely it is that you even benefit from erythropoietin. What she found is that the EPO level was really important in that. Generally, if people have an EPO level of less than 100 units per liter, I’m enthusiastic about EPO. I think people will have, probably, a pretty good response with EPO first-line. But if it’s more than 500, I really don’t think we get such good response rates. Between the 100 and 500, it’s sometimes people respond, sometimes people do not respond. But I use that all the time, when I get patients newly diagnosed, and particular in the setting of now selecting patients for first-line treatment, between luspatercept and erythropoietin, which I’m sure we’ll talk a little bit about later. I think this can help in the patients who are SF3B1 wild-type to select patients based on their erythropoietin level.

Amer Zeidan:

Yeah. Before I go back to Sophie, one logistical aspect is administration. It has been very challenging for us here to give ESA at home. I know they do this routinely in Europe, from discussion with our European colleagues. That comes largely from insurance reasons and logistical reasons relating to visiting nurses. But I actually also discovered that it varies across the US. I’m not sure, have you ever been able to give it at home, Max?

Maximilian Stahl:

Yeah. I am, actually. It’s entirely dependent on insurance, so you never know beforehand, but I always try to give certain supply of erythropoietin. [inaudible 00:12:50] insurances give limited injections. For example, give five doses of Aranesp. That can be stored in the fridge, and can be administered. Other insurances don’t allow it at all, and that’s a big limitation. I agree, Amer.

Amer Zeidan:

Yeah. Yasmin, before I go to Sophie, have you been able to give it at home, or do you only give it in the infusion center?

Yasmin Abaza:

We have not been able to give any home administration. It’s only in the infusion center, so our insurances have not been allowing that at all. I’m impressed that Max has been able to do it.

Amer Zeidan:

Yeah. I initially thought it’s across the US, but actually talking to colleagues in different states, it seems like some insurances can facilitate that easier. Sophie, I know you guys do it all the time. Actually, I was giving a talk, recently, in Paris and I even discovered that sometimes even azacitidine is given at home, and I was like, “Wow.” We cannot even give ESA here.

Sophie Park:

Since several years, and even since 20 years, I think, in France we give ESA at home, by nurses that we have in town around the patient. There is no problem. The patient buys their ESAs at the pharmacy in town, and near their home, and then the nurses near their home administrate the ESA every week. We have no problem of administration of ESA in France. For azacitidine, it is quite different. Azacitidine has to be prepared in hospital, but then after the nurses… It depends of the systems, but either it could be the nurses from hospital, or the nurses around the patient, which administrate azacitidine.

Amer Zeidan:

That’s great. We certainly get… I think many patients would be happy to be able to do that here as well. I want to ask you about the EPO level situation, which Max alluded to a little bit. Historically, we thought of EPO level above 500 as a level at which trying ESA does not make a lot of sense. Although I have to say, before we had luspatercept, I, actually, was using it in anybody. Because, again, I’m not going to give HMA from the get-go just for someone with anemia. I tried it, although the response rate is probably closer to 7%. But now that we have other drugs that are active, and, actually, some of the randomized trials that were done in Europe, both for Procrit, as well as darbepoetin, interesting to me that EPO above 200 seems to predict a large degree of non-response. If you look at the paper by Dr Fenaux, I think it was 0%, that people who were above 200 responded to Procrit for transfusion independence. We have similar indirect data from the placebo arm in some of the subsequent randomized trials of the new agents. But what’s your take on the EPO level at which you would not consider trying ESA?

Sophie Park:

For me, it should not be a 200, but perhaps 500. But several years ago, we did not have any treatment for anemia for MDS patients, just ESA and transfusions. I was like you. Even if you had high levels of endogenous EPO level. We try, and then after we know that the patient will not respond very high and very long. Then, after, you anticipate, for the second-line of treatment, for this kind of patients. But it is important, those EPO level, for studies, for clinical studies, and for papers and for predictive factors. Because as you know, I worked several years on the predictive factors of response to ESA, and we have identified the EPO level, gene mutations, flow cytometry scores, mutation burdens, as a predictive factors of non-response to ESA. These biological tools can help to anticipate the second-lines of treatment for these patients, and especially for the young patients.

Amer Zeidan:

Do you do this, or do you use this in clinic, or do you just go by the Nordic score with the EPO level and transfusion dependence?

Sophie Park:

I’m not using it by the clinic in the routine manner, but then after when you see that the patient seems to lose, it’s response. I’m doing, for example, NGS, to evaluate the mutation burden. Then, after, to calculate the IPSS-M, and then after to discuss more aggressive treatments for these kinds of patients. So, yes, it helps in routine for NGS, for example, but for flow cytometry it has been evaluated, and I would like to confirm that in a prospective manner, and for iron metabolism, too. I would like to confirm that.

Amer Zeidan:

Yeah. I actually have also had faced the same dilemma because based on your, and, I think, Valeria, and other people data, that the more mutations, the less likely someone is to respond to ESA. Sometimes I see patients with three, four mutations. I think it’s probably unlikely that they’re going to respond, but I have not systematically used that to decide on skipping. I want to move from the front line initiation, to the failure of ESA. Again, you have done great work, Sophie, on this, trying to understand the primary versus secondary failure of ESA, and the outcomes of patients after ESA failure. Maybe you can overview the main findings from that big multicenter analysis that you’ve done.

Sophie Park:

Yeah. We have published in GCO, several years ago, a larger study on ESA failures. So either primary failures at month three, or secondary failures after initial response to ESA. You lose your response. We have shown that when you have early failure to ESA, these patients’ outcomes are worse. They have worse overall survival, and a higher incidence of AML. That’s why the response to ESA, and the duration of response to ESA, is important to predict the outcome of these patients.

Amer Zeidan:

Yeah. No, this was a very good paper. Yasmin, moving to you. Does that factor into your decision, about the second-line treatment, if this is primary failure or secondary failure?

Yasmin Abaza:

Not yet, no. It has not been factoring in, but now it just really looks… Because, currently, we don’t have much once they relapse ESA, the paradigm too, and paradigm has shifted quite a bit. But when you go to the second-line, what do we have? We have the MEDALIST for luspatercept, so we have to look at ring sideroblasts and SF3B1. Then you can either add… There’s some European data to add GCSF to ESA, or has changed to Revlimid plus or minus Procrit, and then you have HMAs after that. It has not factored in, but it does make me think, should I.. I have to look at the patient total, look at their mutational status. My second-line, for sure, again, that has changed quite a bit recently. If they follow the MEDALIST criteria for ring sideroblast definition, they would go to luspatercept. If they cannot qualify for luspatercept, that’s when it becomes a little bit of an issue, because then the question is should we do Revlimid, versus just going directly to an HMA? Then if they’re transplant eligible, get them at least evaluated for transplant. But, of course, that has changed dramatically with the recent data.

Amer Zeidan:

Going back to the frontline setting because luspatercept is still not yet… I guess it just got approved in the frontline setting, I understand in Europe, but I’m not sure if the reimbursement has been sorted out. I don’t think it has yet been used in the frontline sitting in practice. But in the US, it has been approved since September, and I’m interested in Europe and Max’s approach to this. How do you currently decide between luspatercept versus ESA in the frontline setting?

Yasmin Abaza:

I think there’s a lot of debate on, “How do we use ESA now?” I’m interested to know what Max thinks, but in general, I think the COMMANDS data is very compelling, especially with the duration of transfusion dependence. Compared to ESA therapy, I think it did show that it did better in most of the categories. I think one of the categories are a little bit iffy, is the SF3B1 mutant versus wild-type, where it looks at least not inferior to ESA therapy. I think in the majority of patients, I try to, especially if they are… Except in very older, although I have to say in older, I’ve been using luspatercept more in the frontline setting. I’ve used a standard dosing with one, and then you escalate every six weeks depending upon response. I’ve not done it the other way, where you do the high dose and you come down. I’ve really used a standard dose, and I pretty much, at the current time, I’ve been trying to change to luspatercept in the majority of patients. Only in patients where they’re not willing to do that, I would go with ESA. But the majority of patients, are now, we’re transitioning to luspatercept for almost everyone. I don’t-

Amer Zeidan:

Are you using RS negative?

Yasmin Abaza:

For now, yes, but that is still area of debate. Should we RS negative SF3B1 wild-type, is that the right thing to do, or should we still do ESA and then we go with luspatercept? I think that’s an area that we have not ironed out. Because the question always comes, then, “What if patients fail luspatercept, or do not respond to luspatercept?” Is ESA will work? We don’t have data to show that. We have data for the opposite, but we don’t have data to show that if you fail luspatercept, or do not respond to it, ESAs work. I think the only reason for RS negative and SF3B1 wild-type, that we use luspatercept frontline, is because they would not be eligible for the MEDALIST, so using luspatercept second-line would not be something you would do. We try to just use it upfront anyway. But again, whether or not that’s the correct thing to do, I think that’s still an area of ongoing debate.

Amer Zeidan:

That’s actually an interesting angle. I have not thought of it this way, in terms of insurance approval. Last question before I go to Max. Have you combined ESA and luspatercept off-label, whether from the get-go or as an add-on strategy at all?

Yasmin Abaza:

Not yet, but the data from Moffitt, and there’s an ongoing trial now, is quite compelling. I think that’s something that I’m thinking of doing. I have not done yet, but I need to think of doing, and we can maybe think of doing an RS negative and SF3B1 wild-type patients. I don’t know if that’s the right thing to do, but I think I need to think more about how to use the double regimen, especially in those patients.

Amer Zeidan:

Yeah. I’ll ask Sophie later about the experience in the COMBOLA trial, which is a French trial, looking at this add-on strategy as a second-line. But before we go to Sophie, Max, two questions along the same lines is how do you choose between the two drugs in the frontline setting, and your experience in, or have you tried combinations? I have to say, the combinations that I have tried, I only combined after failure of one of the drugs, ESA or luspa. My experience has not been as good as what can be, been suggested by the Moffitt group, but it’s a very anecdotal, limited number of patients. Max?

Maximilian Stahl:

Sure. Yeah. I think Dr Abaza, she mentioned already a lot of good points. Generally, how I choose is in the SF3B1 mutant patients, I use luspatercept first-line. In the SF3B1 wild-type ring sideroblasts negative patientsm I don’t use luspatercept as frontline, I still use ESA, and the reasons are the following. I think number one is the transfusion independence rates look very similar. The duration of transfusion independence look very similar in that subset, and ESAs are cheaper. I feel I still have some responsibility to not ruin our healthcare system, and giving luspatercept to everybody. I would say the debate about the sequencing is a very good debate. I think I appreciate the points being made prior. On the one end, one can argue there’s really not much data for ESA post luspatercept dose, versus the reverse is true, as you mentioned. On the other hand, the insurance reason is actually a good point. I never tried to get luspatercept approved in SF3B1 wild-type patients post ESA use, but that is relevant. Not entirely sure insurance would always pick up on that subtle difference, but they might. Then that might also be an issue. You could argue from both sides, but in the RS negative SF3B1 wild-type patients, I still use ESA as frontline.

In terms of the add-on strategy question, I think the Moffitt data was distinguished quite a bit between is it primary luspatercept failure or secondary luspatercept failure? What was the ESA level when ESA was added on? What they found is that with primary luspatercept failure, the response rates to ESA add-on was really not that good. With secondary it was better. For patients who had an EPO level of more than 500 also didn’t work. I think the add-on strategy probably only works in selective cases when you did respond initially to luspatercept, but then lost response and if your endogenous EPO level is still low. I think, for that case, it does work. I have not used it frontline in patients who are luspatercept ESA naive, but I think we might do a trial in that setting, so stay tuned.

Amer Zeidan:

I think I agree with all of the points you mentioned. My biggest struggle with the frontline use, which I still use extensively, but I think my back in the mind is the issue, that both of you mentioned, is how do ESA work after luspa failure? I’d like to have a better sense that we are not burning bridges down the road. But for the RS negative selecting, I think I look at them as not only one group as negative, but I think in terms of are they heavily transfusion dependent? Is the EPO level, is more than 200? Some of those patients I lean towards luspatercept, if they have EPO level that’s more than 200, or if they have more transfusion needs. Now, if you are RS negative, and EPO level of less than 100 or 200, and maybe minimal transfusion needs, or close to transfusion dependence, I think those are patients that are probably reasonable for ESA as well. Sophie, I don’t know what’s your take on all of this, in terms of how do you select. As I mentioned, I don’t think you have been able, yet, to give it as a standard of care approach, but I’m sure you’re going to face that question soon if you haven’t. Then what’s your sense about the combination? I know you are part of the COMBOLA trial as well, so you might have a better sense of that.

Sophie Park:

Yes, it is a very interesting question. First of all, for practical reasons, we cannot use, for the moment, luspatercept in first-line, because we had the EMA authorization very recently, in the March or April, if I remember. Then after, in France, we have to have authorization for reimbursement, and it will take quite a long time, nearly one year. For the moment, we will not use luspatercept in frontline. But nevertheless, we can think about the population we will choose to give luspatercept.
I agree with all of you, with all your reasons and data, that luspatercept seems to work more in the subset of SF3B1 patients. The advantage over ESA would be perhaps in this subset of patients. For you, you have the problem of administration of ESA in the hospital, et cetera. For us, it is very easy. I don’t think that luspatercept will challenge ESA very quickly in France. For the moment, I think perhaps all MDS patients, except a very small subset, for example, transfusion dependent, or just not transfusion dependent, but just having received one or two transfusion before MDS diagnosis, or EPO level, high EPO on the [inaudible 00:31:41] level. For these patients, perhaps luspatercept could be beneficial over ESA. For the other one, for the SF3B1 wild-type, or the MDS without excess of ring sideroblasts, I think all doctors will give in first-line ESA and on the SF3B1 mutated patients, or with excess of ring sideroblasts, it could be, perhaps, depending on the doctors, if they will give luspatercept or ESA in frontline. In second-line, we have, in France, yes, the COMBOLA trial, and there were two arms. The first arm was to just try the association of luspatercept and the ESA at increasing dosage. There was no DLT, or no problem. For the moment we try the association, but the results are ongoing. I cannot say if there is any advantage of the association of ESA to luspatercept over luspatercept alone.

Amer Zeidan:

But the randomization, just for people who don’t know the trial, the randomization is, after ESA failure, is to give luspatercept as monotherapy, versus adding luspatercept to ESA as an add-on strategy, correct?

Sophie Park:

Yes, yes.

Amer Zeidan:

Yeah, which is similar to what the eco group here did with lenalidomide, after ESA, do you switch to lenalidomide, or do you add lenalidomide to ESA? This was, actually, one of the surprises to me, that adding lenalidomide in that setting, on del(5q), after, at that point, seemed to help a little bit. But I have to say, it’s not something I commonly do on my, I generally will not add lenalidomide to ESA. If I’m going to use lenalidomide, I usually use it as single agent. This is actually a good transition, because we discussed ESA and luspatercept a lot. Yasmin, clearly we have lenalidomide, which has some activity in patients without deletion 5q. We have immunosuppressive therapy. We, potentially, in a month or so, we might have imetelstat, at least in the US if the FDA approves the drug, which seems very likely based on the ODAC recent vote. How do you see, after June 2024, how do you think about choosing your second, third, and fourth line agent? Walk us through your thinking process about choosing between these different options. I didn’t mention HMA, of course, which are also an option in the US, even for lower risk disease.

Yasmin Abaza:

I think it’s a question, I think there’s a lot that we need to think about. I think based on the IMerge data, given how heavily transfusion dependent these patients were, and heavily pretreated, and the very significant rise in median hemoglobin, I want to say it is at 3.6 or 3.9 rise in mean hemoglobin level, and the nice results with IMerge. I think, especially that these patients that respond the best, if I recall the data correctly with the SF3B1 mutated patients as well. I think if patients fail frontline therapy, given the high responses, and the dual responses with imetelstat, that I think that would be my first drug afterwards. I have not had much success with Revlimid in non-del(5q) patients. Responses are not that high, around 20% or 25%. I’ve not really seen that much success, in clinic at least. I think I would rather do imetelstat, given the nice data from IMerge, as my second-line after. Again, once we figure out what we’re going to do with luspatercept ESA sequencing, that would be something that I would do afterwards.

Revlimid is more of an approach that I try for some patients, for 12 weeks, it doesn’t work out. If I have a trial. Well, Revlimid is one of the drugs I don’t use as much. And if I have a trial, I would prefer a trial over Revlimid. I just have not much success with responses. I think imetelstat would be my next go-to drug once we have it, followed by a trial, and then Revlimid maybe as a short bridge. Then if I have to, I’ll move to HMA therapy. The problem is once you get to HMA therapy, we’re stuck. And if they fail, survival drops quite a bit, to 12 to 17 months, and then we’re stuck in these patients. I try as much as possible to delay HMA therapy, given the significant decline overall survival in patients who are HMA failures. But these are in patients who are just anemic only. We’re not going to approach those who are bicytopenias, or have more than one cytopenia.

Amer Zeidan:

In RS negative patients, after the ESA failure, would you go for, assuming imetelstat becomes available, would you go for imetelstat or luspatercept?

Yasmin Abaza:

I think I’m still undecidable about what to do about RS negative. I think that’s a very difficult population to treat. I think that in that patients, I would still maybe go all the way down to imetelstat, and I think I’ll bypass luspatercept given there’s no data, or there’s not a favorable response in RS negative on luspatercept as a second-line. I think in the RS negative population, if they received ESA therapy, and they failed, or primary, secondary resistance, I think I would just jump all the way down to imetelstat as my second one.

Amer Zeidan:

Yeah. I think all of this highlights, I think, the next phase of studies really need to focus on the sequencing of therapies, similar to what has been done with multiple myeloma and other diseases. Because I think one of the most important points, in my mind, is making sure that if you lose activity of one agent, for whatever reason after you use another agent, that probably will factor in heavily in terms of which agent to use first. Max, I wanted your opinion about the different sequencing and approach that here has been mentioned, but also immunosuppressive therapy, which you worked on quite a bit. You have a nice large… Well, I think the largest dataset, retrospectively, in immunosuppressive therapy, as well as meta-analysis. But I think most people are not using, I would say, immunosuppression that much in lower-risk MDS, but what’s your thinking about?

Maximilian Stahl:

Yeah. Maybe first, about the sequencing of the therapy, about the imetelstat. I think the one comment I would make is that at least in their subgroup analysis, it didn’t look like it was S genotype-dependent. Both SF3B1 wild-type and mutated patients responded, and the SF3B1 wild-type patient did not respond significantly less than the mutated patients. Which I think is good, because that gives you an option for those patients who are not really in the METALIST population. I think that’s a good thing. Also, there were a couple luspatercept pretreated patients, if I remember right in the IMerge trial, and there were responses in that population. It was very few patients, but there were responses, so I think that’s encouraging.

In terms of the immunosuppressive therapy question that Amer mentioned, I think, generally, there have been several predictors, or so-called predictors, described of who might benefit from it. Generally, I use it for hypocellular or hypoplastic MDS, so with less than 20% cellularity. There are other markers, such as small PNH, or LGL clones, or HLA-DR status. Those I think I use as additional information. Generally, if you have a quite hypocellular MDS, with an expansion of PNH or LGL clone, I think that biology is very similar to an aplastic anemia patient. Those patients, I typically treat with ATG cyclosporine, ATG tacrolimus, and then I usually add eltrombopag as well, for full triple therapy. I did have successes with that, but those are patients where it’s,generally, not that simple to actually tell apart, “Is this an aplastic anemia patient? Is this an hypoplastic MDS patient? Or is it an overlap between the two?” I think immunosuppressive therapy definitely has its role in that. There are other immunomodulatory therapies now in development, such as IRAK inhibitors, NLRP3 inhibitors, but those are all investigational. What their role is going to be in different subsets of MDS, I think, is up to debate.

Amer Zeidan:

Yeah. I think this was a very interesting discussion. In the last five minutes or so, I just want to go around for your final thoughts on, one, on your use of iron chelation therapy for lower-risk MDS patients, and then anything exciting in particular that you see in the next year or so. This is what I think proved to be a very important couple of years in lower-risk MDS. Maybe I’ll start with you, Sophie.

Sophie Park:

Okay. For the iron chelation, of course you use it when the patient are transfusion dependent. Then we have, in France, a trial evaluating low dose deferasirox in ESA failure patients, because we have shown that iron metabolism is quite a disturb in these patients. Perhaps by reducing the oxidative stress with low dose deferasirox, we could improve erythropoiesis. Perhaps this could be interesting results in the future. For the question of the sequence of imetelstat and the luspatercept in ESA failure, perhaps we could also add the background of the patient, the age of the patients, because imetelstat is quite toxic in the first cycles with thrombocytopenia and neutropenia. So perhaps in young patients, when you want to reach a low transmutation, and perhaps imetelstat could be treatment of bridging before allo transplantation in these low risk MDS patients with, for example, high IPSS-M. For example, for elderly patients, luspatercept could be better with less toxicity, then after you can use imetelstat. I don’t know if, in USA, you can use it now?

Amer Zeidan:

Yeah, the FDA is supposed to make a decision sometime in June, but they have oncologic advisory committee, which the FDA is not obliged to follow their opinion, but it was very favorable, 14 to 2 to approving the drug. I think many of us expect it to be approved.

Sophie Park:

Okay. Perhaps in the future, real life data on the sequencing, the use could be interesting.

Amer Zeidan:

Yeah.

Sophie Park:

Yes.

Amer Zeidan:

No, I agree. I think it’s going to be very interesting, in terms of how do we select patients, but beyond the specifics of the anemia, I do agree that starting with patients with good neutrophil and platelet count is probably one of the considerations that will be important with this drug, as well as close monitoring, and prophylactic antibiotic use, and all of these things. Just one point of clarification, before I go to Yasmin. For the low dose deferasirox that you’re using, are you using the Jadenu formulation or the old formulation, the one that’s not easy, and many patients cannot tolerate it?

Sophie Park:

I know the easier one.

Amer Zeidan:

Yeah. Yeah. Because certainly one of the problems with these iron chelation agents have been difficulty of patients to tolerate them. But Yasmin, I’m interested in your approach you have. It’s a very polarizing, I think, thing in the field. Many people don’t even use iron chelation altogether, so I’m interested in your approach.

Yasmin Abaza:

No, we do. There is a survival benefit, as you know, on two trials, I to say the TELESTO and the MDS cancer analysis for chelation versus no chelation, so we do use chelation. I think the debate that we sometimes have internally, until we came up with a consensus, is what’s the ferritin level upon which you start chelation? I think there’s studies looked above 1,000, but NCCN says above 2,500. We typically go by above 2,500ng/mL of ferritin as a cutoff to start, and they have to have received at least above 15 to 20 units of PRBCs as a transfusion. Once they hit those levels, we do use the newer version of the chelation for patients, to try to get the ferritin at least below 1,000.

Amer Zeidan:

Yeah. Yeah. I think that TELESTO provided evidence for event-free survival and improvement in cardiac issues. But I think the bigger issue continues to be, like everything in lower-risk MDS, nothing has been conclusively shown to improve survival. I think this has been, also, one of the issues with iron chelation as well. Max, your last thoughts in iron chelation, but in general as well.

Maximilian Stahl:

Yeah, about the iron chelation, we’re actually, a lot of patients I’ve put on iron chelation, they don’t like it. It’s a fair answer. We put them on it, and then they say, “No, thank you,” and go off it. I think that’s the one part about iron chelation. In terms of I think the field and lower-risk MDS is starting to get really crowded and exciting. The key question is going to be sequencing, as we mentioned it, but also combinations. What combinations make sense biologically? What combinations make sense logistically? And is it better to combine agents or sequence them? I think there are many, many different combinations one can think about, particular if one moves towards the famous last words, disease modifying agents, like imetelstat, combination with HMA, et cetera. Or do we do ESA combination with luspatercept, or ESA, luspatercept, imetelstat, triplet, quadruplet therapies. I think that needs to be finally balanced with side effects to those agents, and the ease of administration, and the cost. I think there will be trials that need to be conducted to figure all of this out. But I think, overall, it is exciting to have more agents, than just ESA and Revlimid.

Amer Zeidan:

I think that’s very good summary, and clearly very exciting days for lower-risk MDS with a lot of development. We did not even cover all these new exciting agents that are being studied, but certainly I think the field is finally moving in a direction where we are going to have more options for our patients. I like to thank our three panelists again, Dr Sophie Park, Dr Yasmin Abaza, and Dr Maximilian Stahl. Looking forward to having you again in a subsequent episode of VJHemOnc on MDS Sessions. Thank you so much.

Amer Zeidan: Consultancy: Schrödinger, Notable, Chiesi, Mendus, Otsuka, Foran, Syros, ALX Oncology, Kura, Seattle Genetics, Servier, Boehringer-Ingelheim, Jazz, Orum, Geron, Syndax, Gilead, Zentalis, BeyondSpring, Incyte, Pfizer, Celgene/BMS, Janssen, Amgen, Agios, Daiichi Sankyo, Tyme, Taiho, AbbVie, Ionis, Takeda, Genentech, Epizyme, Lox Oncology, Novartis, Astellas, BioCryst, Regeneron; Honoraria: Schrödinger, Notable, Chiesi, Mendus, Otsuka, Foran, Syros, ALX Oncology, Kura, Seattle Genetics, Servier, Boehringer-Ingelheim, Jazz, Orum, Geron, Syndax, Gilead, Zentalis, BeyondSpring, Incyte, Pfizer, Celgene/BMS, Janssen, Amgen, Agios, Daiichi Sankyo, Tyme, Taiho, AbbVie, Ionis, Takeda, Genentech, Epizyme, Lox Oncology, Novartis, Astellas, BioCryst, Regeneron; Research Funding: Foran, Shattuck Labs, Astex.

Yasmin Abaza: Advisory board/Consultancy: Astellas, Bristol Myers Squibb, Geron, KiTE, Pfizer, Rigel, Servier; Institutional trial support: AbbVie, ALX Oncology, Biomea, Biosight, Curis, Novartis.

Maximilian Stahl: Advisory board: Novartis, Kymera, Sierra Oncology, GSK, Rigel, BMS, Sobi, Syndax; Consultancy: Boston Consulting, Dedham group; CME activity: Novartis, Curis Oncology, Haymarket Media, Clinical care options.