iwMDS 2025 | iwMDS wrap-up: translating research into practice to optimize patient care

Amer Zeidan

Hi everyone, my name is Amer Zeidan. I’m from Yale University, and it’s a true pleasure to be speaking to you here from Lisbon, where we actually just concluded our third international workshop for myelodysplastic syndromes and myeloproliferative neoplasms.

Elizabeth Griffiths

Hi, my name is Elizabeth Griffiths. I’m at the Roswell Park Comprehensive Cancer Center in Buffalo, New York, and I’ve been really pleased to participate in this excellent meeting.

Lisa Pleyer

Hi, my name is Lisa Pleyer. I’m from Salzburg, Austria, from the Paracelsus Medical University, and this is my second time here at this great meeting, and it’s been a wonderful experience.

Andrew Brunner

Hi, I’m Andrew Brunner. I’m from Massachusetts General Hospital in Boston, Massachusetts, United States, and it’s been a real pleasure to chat with everybody about some of the challenges we have in MDS and some of the advances we’re making over the last few years.

Yasushi Miyazaki

Hi, I’m Yasushi Miyazaki from Nagasaki University, Japan. I really enjoyed the two-day iwMDS workshop. I would like to thank the organizers for inviting me here.

Amer Zeidan

So thank you so much for joining us here today. So maybe I will start with Dr Miyazaki. Everybody in this workshop gave a talk, and I wanted to kind of get a quick summary, like a couple of minutes’ summary, about the talks that have been given, so maybe we can start with you.

Yasushi Miyazaki

Thank you. I talked about the general situation of our transplant in Japan and our study regarding the conditioning intensity for lower-risk MDS patients. In Japan because we have a lot of cord blood sources. One-third of allogeneic transplants are performed using cord blood. This is going to be one of the specific features of transplants in Japan. In terms of the intensity for lower-risk MDS patients, we found that reduced-intensity conditioning has better GvHD-free relapse-free survival compared to myeloablative conditioning. This information will be helpful for patients who are going to receive our transplant in the world. However, considering the number of transplants or the proportion of patients who can receive transplants is still limited. So we need to explore the safer and easier way for transplantation.

Amer Zeidan

Thank you so much. That’s great. Andrew?

Andrew Brunner

Yeah, it was a real pleasure to be here and get to talk over the last few days about MDS and many new therapies that we’re developing, as well as understanding some of the studies that have just been completed. I think that what I was asked to speak about was really how we understand our communication about efficacy of trials. And this is really relevant. When we use a new drug or when we’re trying to develop a new drug, one of the first things you’re trying to do is show that there is potential, that this can beat our current standard of care. And then we want to show that this is meaningful in some later comparison. And in MDS, often we’re not always speaking the same language. And I think that we, as a group, are in a unique position to try to advance both some aspirational goals. How do we find activity of a drug? How do we find activity of a compound that is really meaningful? Maybe enhancing blood counts or looking at novel response mechanisms. At the same time, when we talk about trials, we need to make sure that they can talk to each other. It’s a little bit like the Rosetta Stone of MDS. We need to be able to say that one trial can relate to another. And sometimes the way we’ve published our studies and the way we’ve reported out our outcomes hasn’t always allowed a very easy comparison between trials or to historical controls. And so one of the goals that we have from this conference is to develop a set of minimal criteria that we can use in the future as sort of advice for when you’re running a trial. Sure, we want to know exciting potential activity of a drug. We want to know ways in which it might change. Responses might change our perception of what we historically perceive. But we also want to be able to relate these drugs to existing criteria. And so can we, as a group, come together to say, here’s some minimal criteria that we should use so that we can kind of interrelate drugs, find things that both beat historical expectations and have promise for the future. And so hopefully, that’s a real tangible endpoint that we can develop out of our meeting here this weekend.

Lisa Pleyer

Yes, so I was asked to talk about response criteria, and in the first part of my talk, I focused on comparison of response criteria in patients with MDS, AML, and CMML. And we’re in the lucky position to have a large database that currently has more than 2,800 patients, and we’re about to include data from a further 500 patients. So we’re able to directly look at these comparisons on a patient level. And what our data showed was that IWG composite CR from the 23 criteria published by Dr Zeidan’s group is basically not only able to distinguish patients with the best prognosis in MDS or high-risk MDS, but this goes through for low-risk MDS, CMML, and AML patients as well. The second part of my talk focused on trying to find a new response criterion where you don’t necessarily need bone marrow evaluation parameters termed peripheral blood complete remission. And we have already published this data, but we want to move on to the third part of my talk where we focused on guaranteed time bias or immortal time bias and ways how to eliminate this. So what we are going to focus on in the next weeks and months is eliminating this bias with AI methods and using all of the sequential data that we have with LSTM models. So we have about 45,000 differential blood counts entered into our database and a further 40,000 data points of other lab values of quality of life measures and other things, ECOG, that we want to all merge into this model to find out which values, which parameters and which values thereof are really important and essential for prognosticating outcome in these patients. And that’s work in progress.

Amer Zeidan

Yeah, and that’s actually very important because we had a session during this meeting with the FDA, and actually they did emphasize the importance of validating these response criteria. So I think your work will go a long way, along with some of what was suggested by Andrew about validating these, and hopefully within the context of the multiple Phase III trials that have been conducted. Elizabeth, you covered oral hypomethylating agents, which I think are a very patient-friendly approach of delivery of drugs.

Elizabeth Griffiths

Yeah, so I was invited again to speak about an update on where we stand with oral hypomethylating agents. And I think a lot of us were hoping that with this new data coming, that we would have a new standard of care, which would maybe replace single agent hypomethylating agents. And I think unfortunately what we found now with Validate reporting negative data is that hypomethylating agents remain our standard of care at this point. And so I think it’s really incumbent upon us to optimize utilization of hypomethylic agents for our patients in order to minimize toxicity, minimize time toxicity, as well as actual toxicity. And so we touched a little bit on some of the data using these agents, both in the first line and the real world setting, which demonstrates them to be, I think, at least as good as our standard parenteral approach. And potentially implementing these therapies in the post-transplant setting as a mechanism of for stalling relapse, which is really our long-term problem. And these patients hoping to get a large percentage of our patients to transplant.

Amer Zeidan

I think that’s great. And just to give additional context for this, for people who don’t follow the field very closely, just before this meeting, like a couple of weeks before we came, there is also Verona trial, which is a randomized Phase III trial of Aza-Ven, which was the last standing Phase III trial that everybody was waiting the results for were announced to be negative. We still have not seen all the details, just the press release that the primary endpoint, which is overall survival, was not met, which clearly was frustrating to many of us who treat patients with MDS and particular high-risk MDS. And this was extensively discussed in the meeting. Of course, I think once we see the full data, it’s gonna be much more informed discussion. But despite the frustrations, I think there is a lot of excitement about what we can do to move the field forward in high-risk MDS. And in that context, actually, I gave a talk about some of the immune checkpoint inhibition and some of the new directions that we are seeing beyond the traditional PD-1, PD-L1, CTLA-4, the traditional immune checkpoint inhibitors that have not done very well in the context of myeloid malignancies, but some of the novel agents looking at TIM3, a new way of targeting CD47 as well as a new immune checkpoint inhibitor going after cleaver one, which is a macrophage, basically receptor are showing actually promising data, including some very interesting six differential impact of sabatolimab, the anti-TIM3 agent, which we are trying to understand better and how can we use this data to kind of move the field forward. So, I think more to come hopefully in the near future.

Elizabeth Griffiths

Well, I’d love to comment on that if that’s okay. You know, I think there were some of us in the field who were very enthusiastic about doublet therapeutics with venetoclax combinations, and I think some of us who were a little bit less enthusiastic. And I have to put myself among the less enthusiastic group, in part because I saw a lot of toxicity associated with venetoclax implementation early on in MDS therapeutics. And I think fundamentally it speaks to the differential biology of the MDS disease state compared with the AML disease state. And I think, however you want to think about it, venetoclax plus hypomethylating agent is really a cytotoxic regimen, right? It’s predicated on the ability to induce apoptosis. And I think when we think about disease status and MDS, these patients are perforce individuals with a slower disease cadence and therefore less proliferative underlying disease biology. And so it’s not surprising in my view that this sort of cytotoxic nuclear bomb type approach, as we’ve shown before with chemotherapy, with other agents, is less effective in this disease state. And so I think instead of pushing for these very cytotoxic upfront therapeutics, I think an approach that really capitalizes on the things we’re starting to understand about underlying disease biology and the immune dysfunction within that context are more likely to show us improvements. I mean, I think the fact of the matter is that allogeneic transplant remains the only potentially curative strategy for this disease, says that immunotherapy is likely to be the key there. It’s the tickle, right? It’s the small signal that we have for how to cure this disease. And I think what we can say is that that’s the only therapeutic modality that’s shown real promise, real success. And so we have to build on that.

Lisa Pleyer

But I would also like to comment on that. I mean, I agree with everything you have said, but I think the results of the trial, and that was also discussed in this meeting, which I thought was very interesting, they don’t really give us the final results yet because a third of the patients that were included had less than 10% blasts. And also perhaps the dose reduction scheme. So I think in MDS, most regimens are more toxic. So maybe you need to reduce the dose even more when you start. And when they did the dose reductions, they mainly reduced the Aza backbone in the three steps first. And I think for the azacitidine treatment, you should not, well, that’s what the data from our registry show, we haven’t published it yet, but is that you shouldn’t really reduce to less than five days, full dose, and then you should stick with the four weeks. So I think that perhaps that was also, so we don’t really know quite yet. So not all of it. So that would be interesting for me to know in addition to what the trial has shown.

Amer Zeidan

And I think this is very important because, you know, we still have not seen, you know, the data of, you know, in terms of what led to the failure of the trial. But I think, you know, in my opinion, I think it’s actually a quote from, I understand that, doing the same thing or the same experiment multiple times and expecting different results is the definition of insanity. So I think we really have to learn from these negative trials about what was the problem. And from each negative Phase III trial, I think we have learned something. So especially that there’s actually what we have learned over the last couple of months, that there’s already a Phase III trial of another PCL2 inhibitor with an overall survival endpoint that seems to actually have a sample size even smaller than Verona. So I think it’s very important to understand exactly what went wrong with Verona and how that would inform the other trial, because certainly nobody wants another Phase III that will be negative. But maybe I’ll go to you, Andrew.

Andrew Brunner

No, yeah, obviously we could all talk about this quite a bit. I think it’s a great reminder that MDS is extraordinarily heterogeneous and that different patients need different things. And some patients do need blast reduction, and that will yield blood count improvement. And some patients really just need transfusion independence, and some patients really need more platelets. And I think that a challenge, and this is a kind of a good reset maybe in the field to say, how are we designing our studies? Is risk, which is largely defined outside of therapy, should that be more affiliated with therapeutic intervention? Because when we assign risk, we kind of are saying, I’m going to treat you one way or another. So I’m either going to put you on a strategy of chemotherapy, potential transplant, I’m going to try to modify your disease, or I’m going to let you live with your disease, and we’re going to try to support your blood counts. And maybe there’s some people with molecularly high-risk disease that would really benefit from luspatercept or imetelstat and not trying to do chemo and transplant. And maybe there’s some people with lower risk disease, which really we should do a different approach where they would be more responsive to traditionally high-risk approaches. And I think that realigning to try to identify maybe in the future, who best responds to a given treatment strategy rather than what’s likely to happen if we let people just coexist and not do anything. That may be a way to really kind of advance forward and how do we design trials next? Because if we keep doing the same thing, I agree with you completely, we’re choosing a pretty broad group of people to put on to one treatment, we put them on that treatment forever, and we see what happens in a few years. And I think that we can be a little bit more selective if we think about it in a slightly different angle.

Yasushi Miyazaki

Yeah, I agree. Through talking about the Verona study, we had a lot of discussion regarding study design, as you see, selection of patients, and how about the endpoint, overall survival. Did we have enough number of patients? Many, many discussions we had. And as you said, MDS is a very, very heterogeneous disease. And giving cell therapy policy is not a good one, of course. We have to change disease history. So for example, azacitidine is a disease history modifying agent. And now we understand that imetelstat also can change the history of the disease. With the high risk molecular pattern, but we understand, we know that in the higher risk MDS molecularly, they are still heterogeneous. So I think we need to learn more about the biology of MDS. We just separate higher risk and lower risk, but that will not be enough. We have to separate them based on not only by genetics, probably we need methylation status or others. But I agree with all of you, we need to learn more about MDS itself.

Amer Zeidan

Dr Miyazaki, I actually wanted to follow quickly on that point, because you asked a very important question during the workshop. And this is something that I have seen actually come up more and more in AML trials about, you know, they have this risk stratification by geography. And I think it reflects both understanding that maybe the underlying biology and genetic differences between, you know, different continents, Asia versus, you know, highly Caucasian population, maybe in the U.S. and Europe, but also different, like, practices in terms of how you treat patients and different approaches we still have not seen like this kind of play out in MDS because largely the tries have been kind of negative but I think what happens with many of the Phase I tries is that they kind of primarily accrue from the US and maybe somewhat from Europe. And then once they go to the Phase III this is when you have like significant accrual from East Europe as well as China and potentially Japan etcetera. So I wonder how in your opinion this plays out? So not only for higher risk but I would even add for lower risk because in lower risk MDS we had a lot of discussion about the transfusion thresholds and you know people were arguing eight versus nine versus ten and what I actually discovered during a visit to China a couple of years ago is that they actually transfuse patients for hemoglobinopathies and six or six and a half so they are even much more restrictive than us here in I don’t know what’s the threshold in Japan, but how do you view these differences between the continents?

Yasushi Miyazaki

Okay, first I talk about the CML. CML is a BCR-ABL disease. And all over the world, imatinib TKI works very, very well, and we can get the same results.

Elizabeth Griffiths

Because we have an effective drug.

Yasushi Miyazaki

But the transfusion things were those not completely defined by only one genetic mutation, for example. It’s very difficult to deal. We didn’t discuss much about the social economy situation of patients, which can affect the results of clinical outcome. So we have a lot of factors which affect those things. In Japan, actually, we transfuse hemoglobin around 8 or something like that. And Dr Sekeres told their study about the academic institution start to transfuse around hemoglobin level 8 in United States, I think. So that’s a very similar one. So of course it depends on the condition of institutions or hospital, academic or community hospital. Community hospitals start to transfuse around 10 or something. So that’s the difference. That is not for patient factor. That’s a factor of institutions or hospitals. So there are many confounding factors, I think.

Lisa Pleyer

I would like to briefly follow up on your statement or the question regarding why do Phase III trial results not reflect Phase II trial results. Well, I think that’s a classical example of guarantee time bias because in Phase II trials, you often have overall survival, I mean, overall response rates or CR rates as end point where you have the classic lead time bias. And then when you look at overall survival in Phase III trials, in the Phase II trials due to the guaranteed time bias, you’re overestimating the effect of the drug. That means you’re underestimating the number needed to treat for the Phase III trials, and that’s perhaps why they are negative, is because you’re not including enough patients based on the Phase II trial results.

Amer Zeidan

Yeah, and I think all of us are looking forward to seeing kind of the full results of Verona, but I think there will be a lot of what I would call post-mortem, like in terms of Verona and all the other trials and how we can move like going forward. But thank you so much. These are wonderful insights. And I would like to kind of thank the panel and thank everybody who participated in this largely like really international workshop that initially started with a smaller group of U.S.-based physicians and then extended to Europe. And now we have representation from Asia and from Australia. And we are hoping to extend to South America and Africa. and I think it’s very important to understand globally for MDS and MPN. And actually this year we started also live streaming of the workshop. We had many people who watched remotely and the sessions have been recorded. And at one point they will be put online for those who are interested to kind of see the full discussions that took place. And I would end that our next workshop is going to be actually next year and October 2nd through the 4th. And we hope to continue to grow and maybe see some of you over there. Thank you so much.

This transcript is AI-generated. While we strive for accuracy, please verify this copy with the video.