iwMPN 2025 | Novel approaches to upfront management of myelofibrosis

Prithviraj Bose:

Hello, I’m Prithvi Bose from the Department of Leukemia at MD Anderson Cancer Center in Houston, Texas, and we are here at the 3rd International Workshop on MPNs and MDS in Lisbon, Portugal. I’m joined by my colleagues Donal McLornan and Haifa Kathrin Al-Ali, and we just completed two very interesting and engaging sessions on different aspects of myelofibrosis, and we will be discussing those. 

Donal McLornan:

So, my name is Donal McLornan from University College London. 

Haifa Kathrin Al-Ali:

And hi, my name is Haifa Kathrin Al-Ali. I’m a hemato-oncologist at the University Hospital of Halle in Germany. 

Prithviraj Bose:

So, Donal and Haifa, we just had two really interesting sessions, I thought, and in the first one, we discussed, you know, we started with prognostic models for myelofibrosis, of which there are many, and Alessandro walked us through those really nicely. Lucia Masarova from my institution then covered cytopenic MF. I covered early MF, and you know, maybe should we treat these patients, who should we treat, and with what, and then you, Donal, covered the very complex subject, I think, one that is not fully agreed upon, of suboptimal response. So, why don’t we start with you? What stood out to you in that session? 

Donal McLornan:

Yes, I really liked the session overall because I think it’s for all of us in the field, we see that the prognostic models that we’re using are constantly changing. And Haifa brought up a really interesting point for us. I mean, many of us in the area are using molecular-enhanced models now across our patient age ranges, whether or not they’re for intensive or non-intensive treatment, but actually, these are models that have been derived from patients for many decades ago in some cases, and also, when we think about it, we should be thinking about dynamic modeling for patients on treatment, and I think, as you suggested, as a future thinking concept, we need models that are relevant to our contemporary practice today and thinking about landmarking those models based on treatment responses, and I think, as we all know, we have the RR6 model looking at dose density of ruxolitinib, looking at spleen responses, transfusion requirements at three and six months, and how that is powerfully dichotomizing that cohort of MF patients as regards survival. And I think that’s really what we need to apply across the field. 

Prithviraj Bose:

No, absolutely. 

Haifa Kathrin Al-Ali:

Absolutely. And I think we all realized, with all the open questions and the nice discussion we had, that MF is not CML, and that we still don’t know so much, and we have to learn. And yet, I think the future is also brighter for the patients because of all the developments, which we then discussed in the second session. 

Prithviraj Bose:

Right. And again, going back to your point about response to treatment and how we should, or the lack thereof, and how we should incorporate that into our prognostic modeling. Right, so the RR6 is so far the only such formalized effort, but do you think we should be incorporating that into, you know, just our… front-line disease stratification, of course, by definition, means before treatment, and that’s where all our models have been developed, but how would you, let’s say, incorporate that? Would you use, like, the six months, like the RR6 has done, of ruxolitinib, or do you have any other…? 

Haifa Kathrin Al-Ali:

Absolutely, absolutely. I think this is really fundamental, and you know, with all the available drugs coming, and as you said, also second-line, third-line, how long should I treat? So, I think something like the RR6 model is quite helpful, not only within clinical trials but also, you know, in everyday clinical work. Whether this is the ideal model, I think it’s our duty to generate the data, but I think, yeah, we have to. I always say, the response treatment, the dose patients can tolerate, of not only JAK inhibitors, I think of all medications, even in other diseases, it’s the same situation, and the response. And I think, you know, anemia and the impact of anemia, is it really still so bad under the new treatment? You know, we prevent now, some of these drugs could prevent patients from becoming transfusion-dependent, or even patients that are transfusion-independent are turning back to transfusion independence. And you know, the weighing of anemia versus transfusion dependency, I think, needs to be incorporated in all our predictive models, particularly with younger patients, transplant, because we still have, you know, it’s not that easy to send patients with the morbidity and mortality, we still have, and so, yeah, we need to improve our stratification. 

Prithviraj Bose:

Right, you remind me with momelotinib, as I was showing, in actually the next session that we just did, where there is data that if you make them transfusion-independent, or if you improve their hemoglobin from below 10 to over 10, that could have an impact on survival, to your exact point. And Donal, just coming back to you, so you touched on a complex subject of suboptimal response, which every trial has defined differently. And then there is a lot of overlap between suboptimal and failure. Right. So, what do you think? Shall we ever have a consensus on this? 

Donal McLornan:

I think really, when we were talking about suboptimal response, as we all know, I think we definitely need a working definition for us that’s clinically relevant across our practices, whether it’s a tertiary or community practice. Because I think we know from emerging data that early identification of someone who’s not going to get a durable response, whichever JAK inhibitor we use upfront, we should be thinking about early intervention in my perspective to change and I think that’s where that suboptimal response versus it’s probably easier to diagnose your primary refractory patient to your JAK inhibitor, much more difficult on a day-to-day basis to do that with a suboptimal responder. But clearly, they’re going to get a less durable response, not often good dose density of the drug. So I think we need to collectively as a community formulate probably a working diagnosis that changes over time, that’s probably product specific as regards the JAK inhibitor and which line we use it. And that sounds complex, but I think moving forward to get the best responses for our patients, we need to do that. 

Prithviraj Bose:

And maybe it would simplify our trials if all could use the same definition of suboptimal response or failure. Like for anemia we are now maybe moving towards that. There was a paper last year in Blood that sort of tried to you know make uniform the eligibility criteria for anemia trials and the response criteria for anemia as well. 

Haifa Kathrin Al-Ali:

Right and I always give the example you know this overlap as you mentioned as well between suboptimal, refractory, resistant patients, loss of response and I think it is particularly with the combinations that are emerging to identify the real suboptimal and maybe these are the patients who are likely to benefit most from add-on strategies with whatever compared to patients that are truly refractory or resistant because the idea you know from all other diseases is not to wait until patients become refractory but to try to go upfront. The question is always in MF what is upfront from the very beginning or at the suboptimal. So the suboptimal is really a very important issue we still have to solve. 

Donal McLornan:

I agree fully and I think to link the two things we’ve talked about so far about prognostication, which is a snapshot of a signature upfront in a treatment naive group versus then defining suboptimal response, one of the things we need to do is dynamically monitor clones, whether it’s driver or non-driver, and maybe build that into the models as well. If we’re seeing a strong percentage reduction in a driver clone, that’s clearly a more optimal response versus if we’re seeing something much more sluggish or no response, should that be built in? Because I think in our younger patient population, that’ll help us think about when we intervene with a combination strategy, when we think about more aggressive therapies. 

Prithviraj Bose:

Right. So, you know, it’s a nice segue, Haifa, what you just last said into our second session where we had separate talks on pelabresib by Adam Mead, navtemadlin by yourself, selinexor by Claire Harrison, and I provided an overview of the approved JAK inhibitors. So let me start with you. With the navtemadlin, it has such a unique trial design, right? First time ever in the field. 600 patients, ambitious trial, but really very logical. I think they’re all going to start with rux and only if they’re suboptimal for both spleen and symptoms do they then randomize to navtemadlin or placebo. Your thoughts on this, especially as it compares to other drugs which are being developed more in the front line?

Haifa Kathrin Al-Ali:

I think it’s a very nice and a new and strategy, you know, challenging, we as academia definitely could not do such trials. But I think this is the right way to learn what are the patients. And even we can then maybe indirectly compare, you know, in this trial, patients have at least 18 weeks of treatment with rux. And we could, you know, then really maybe define what is definitely a suboptimal versus optimal versus refractory patient. Yeah, I like the design. I think it is challenging. 600 patients, it’s not a joke. 

Prithviraj Bose:

So maybe identify truly who needs the second drug, right? 

Haifa Kathrin Al-Ali:

And I think it will make maybe the basis for future combination therapies and the definition of suboptimal, yeah. 

Prithviraj Bose:

Now, we also had a lot of discussion about pelabresib. Obviously this is the one with where there is a completed Phase III trial as opposed to say selinexor where we don’t know the results and navtemadlin, which is a lot more early. So with pelabrasib something that came up quite a bit was the leukemic transformation and certainly it was a bit unexpected, it was surprising and concerning but at the 72 weeks at EHA Alessandro Vanucchi showed us that the difference between the arms is narrowing, It is coming down. Now, I found that reassuring. You guys’ thoughts on this important topic? 

Donal McLornan:

Yeah, I mean, I think like all of us in the field, we were quite alarmed when we first saw that signal with the combination of pelabresib and the backbone of a JAK inhibitor. But I think this 72-week data for me is much more reassuring, that we’re not seeing this strong perpetuation of transformation risk. And we’re thinking about… because that for me biologically at the start didn’t make sense, why when we’re using this synergy we’re seeing this potential increase in transformation, but I think the long-term data is reassuring. 

Haifa Kathrin Al-Ali:

It is but still it is worrying the early signal is worrying because you are getting you know early AML transformations and we don’t know why particularly always considering that 60% of the patients were intermediate one, not advanced disease, so sure it is reassuring but we have to learn and understand and we don’t have an answer yet to the reason why these early increases in leukemic transformation occurred in the combi arm. 

Prithviraj Bose:

Right. And also Haifa, let me go back to this. I know we touched on it, but you said a couple of times during the session that, if it’s fair to put it this way, that you’re more a fan of the add-on strategy than say going with the combo up front. Your overall thoughts on that, like what makes you think that way? 

Haifa Kathrin Al-Ali:

And as you said, we discussed this. What is the end point? What do I want to achieve with an add-on from the very beginning? It is always, you know, combination. We all know JAK inhibitors, particularly ruxolitinib, has particular adverse events, particularly the hematological adverse events. We need the dose modification in almost 70% of our patients. And we do all this stuff in the first three months. And if I put another drug from the very beginning, you know, we reduce both or we stop. I don’t know. We don’t really know what we’re doing and we don’t know what is the aim. And that’s why I’m, and you know, most of our patients are elderly. They have comorbidities. So maybe we are putting the advantage we know from a strong drug like rux at risk for something with a very, very big question mark. 

Prithviraj Bose:

If those reductions are necessary. 

Haifa Kathrin Al-Ali:

Also reduction, discontinuation, yeah. 

Donal McLornan:

I also agree and I think obviously clearly with focusing on the clinical endpoints with this sort of strategy but also the translational work that will come off from having that JAK inhibitor period lead in then the combination thereafter, because that’s a very different combination situation than starting both drugs up front so I really like that with the emergent translational work we’ll probably see with the immune repertoire with the clones etc. with that lead-in approach. 

Prithviraj Bose:

I think the MANIFEST-2 was probably informed by the arm 3 of MANIFEST, where they saw some amazing spleen and symptom responses by starting both together. And that’s probably what just led to the MANIFEST-2. And then the selinexor trial century has sort of followed suit. I guess the leap of faith here is that, if you get a deeper spleen response and symptom response at week 24, that will translate in the long term to greater benefits. But we don’t know that. 

Haifa Kathrin Al-Ali:

And not yet seen with MANIFEST-2. We don’t have a survival benefit. 

Prithviraj Bose:

Not yet. Correct. Well, very engaging sessions. Thank you both so much for your insights.

Donal McLornan & Haifa Kathrin Al-Ali:

Thank you.

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