Emerging Data Offer Alternatives for Patients With Lower-Risk MDS

Frontline treatments that achieve transfusion independence are of paramount importance for patients with lower-risk myelodysplastic syndrome (MDS). Progress in the past few years has led to long-term management of the symptomatic events that may cause progression to high-risk MDS; however, emerging therapies change the history of lower-risk MDS.¹

In a recent OncLive The Talk™ video program, Rami Komrokji, MD, led a discussion about practice-affirming updated findings with luspatercept-aamt (Reblozyl) as well as early impressions of the data on novel agents presented at the 64th American Society of Hematology (ASH) Meeting and Exposition^2-5.

Komrokji: Regarding treatment for lowerrisk MDS, when do you start, what’s your first-line treatment, and is that changing?

Amy Elizabeth DeZern, MD, MHS: Usually, the presenting symptom or blood count irregularity for lower-risk MDS is predominantly anemia. Most people believe that erythropoiesis-stimulating agents remain the first-line therapy in most patients who do not have a deletion 5q [del(5q)]. Red blood cell [RBC] transfusions are always helpful in patients who have anemia from lowerrisk MDS. We all have varied thresholds to start therapy for our patients who present with anemia. I, personally, usually use a hemoglobin [level of] around 9 or 9.5 g/dL and certainly vary that depending on how symptomatic a patient is from sideropenia.

Komrokji: There’s no magical threshold or cut-off, but what you suggested is very reasonable. I follow the same in practice because that gives you the 1 g before getting transfusion-dependent [disease].

At [ASH] we heard the updates also from the SINTRA-REV study [NCT01243476]6 where they did low-dose lenalidomide for a couple of years in patients who have milder and non-transfusion-dependent [disease]. Are you changing your practice to start lenalidomide earlier in patients with del(5q)?

Guillermo Garcia-Manero, MD: [We] saw these data a couple of years ago because what we saw at this past ASH meeting was the final output of that study. But the [presentation was] very compelling 2 years ago. This is not the most common presentation [of MDS] we see at MD Anderson, but there were a handful of patients with transfusion-independent del(5q) low-risk MDS where I started this low-dose lenalidomide, this 5 mg. And I have to tell you, the results are really par to what they showed at the presentation. Their hemoglobin [levels were raised], and these patients now for 2, 3 years, remain transfusion-independent. Of course, this is uncontrolled data, so I don’t know what would have happened otherwise. And you see other things that are quite interesting, such as normalization of the mean corpuscular volume.

I think it’s a paradigm change. [This study] in Europe, led by María Díez-Campelo, MD, PhD, in Spain, was a randomized placebo study of early intervention in del(5q) MDS. These were patients with MDS, some anemia, but not enough for transfusions. And what investigators did was they started the patients on low-dose lenalidomide vs placebo and showed that those that were started on the lenalidomide earlier had a significantly longer period of transfusion-free survival. That is a new term in MDS, we never really had this before; it was quite significant.

The study did not have a crossover design, but of course, those patients who became transfusion-dependent who were on the placebo arm, they went on to get lenalidomide later. So, it’s difficult to really get if this translates into OS. The questions become: toxicities of the lenalidomide and the possibility of acquiring a TP53 mutation. But from the data presented there was no evidence of that.

This is important, not just for those with del(5q), but it may suggest that early treatment in low-risk MDS, even if [the patient is] transfusion-independent, if you had the right drug and the right target, is the way to go. So, I’ve done that now in maybe 4 or 5 patients with very nice results.

DeZern: I started changing my practice patterns somewhat based on [the interim analysis], and I have a very eyes-wideopen conversation with the patients about where the data is [from]. But I share your belief that it may possibly be truly changing the natural history of the disease. And at a slightly lower dose, I have, in a handful of patients, found the toxicity [with lenalidomide] quite manageable and the rise in hemoglobin to have quality-of-life benefits for the patient without much detriment.

Amer Zeidan, MBBS: I’m impressed with the data. I cannot say in terms of my practice that it has been fully practice-changing, but it is intriguing, and I consider it on a case-by-case basis. I think my main concerns are that we did not have—or we did not see—quality-of-life data from the study. As Dr Garcia-Manero hinted we also don’t see a clear OS advantage, although it could be there.

Guillermo Garcia-Manero, MD: [For] patients with what I call low-risk, anemiconly disease with a low EPO [erythropoietin] level, the standard of care is an erythropoiesis-stimulating agent [ESA]. For patients with sideropenia, meaning significant anemia, significant thrombocytopenia, we tend to use very low-dose, hypomethylating agent [HMA]–based approaches. But I think that’s controversial, although it’s in the [National Comprehensive Cancer Network] guidelines.

But for a prototypical patient who received an ESA and now is not benefiting from that intervention, the question is, what to do? Of course, in this setting of SF3B1-mutated, ring-sideroblastic anemia, we have a drug indicated—luspatercept—based on the MEDALIST trial. And we’re all very eager to see what happens with the COMMANDS trial [NCT03682536].⁷ There was a news release that this is a positive frontline randomized study, but we have not seen the data. That will be for sure one option that you will choose.

If patients are not SF3B1 mutant or they don’t have ring-sideroblastic anemia, then, in my practice, the next step will potentially be an HMA. Occasionally, I may use lenalidomide off-label, for a non-del5q patient. I tend to do this in patients with a decent platelet count—let’s say more than 100,000—following the observation that they tend to respond very well to lenalidomide. If you look at the original phase 1 trials, it was already clear there was [benefit] not just [in] the presence of the del5qs, but in patients with a platelet count over 100,000.

To summarize: If a patient has anemia and meets the indication for luspatercept, you will go for that. If they have more than 1 cytopenia, maybe consider low-dose HMA. If they only have anemia and failed DSA [anti-HLA donor-specific antibodies], sometimes pushing for lenalidomide, if you can get that for your patient, may be useful. And the responses are fast, so you can do lenalidomide for a couple of months, and if there is no benefit, you move on. That’s basically, in a nutshell, what I would think of.

Komrokji: That sets the stage to go into the exciting data presented at 2022 ASH. We saw a few updates from the MEDALIST study at the past ASH meeting.^2-5 Dr Zeidan, what did we learn about use of luspatercept?

Zeidan: MEDALIST has set luspatercept as a standard-of-care option in patients who have ring sideroblast anemia, who need transfusions after a failure of ESA, or [for whom] ESAs are unlikely to work. There was some concern in the community about the duration of transfusion independence, the median of which was close to 32 weeks.

Luspatercept was approved in 20208 and was presented at ASH were several secondary analyses looking at how to use the drug in the community setting. One of the main findings of those secondary analyses is that luspatercept is usually started at 1 mg/kg and then you escalate the dose to 1.33 mg/kg, and then to the maximum dose, which is 1.75 mg/kg. This was done as part of the clinical trial guidance: if a patient still needed transfusions, you must escalate. What we saw is that most patients needed to escalate, meaning that the 1 mg/kg for most patients does not work, especially for those with a high burden of transfusion at baseline, and for patients who have high EPO levels. This is important because we see that in the real world, where we see patients starting on low doses and community physicians often forget to escalate and the patient continues needing blood transfusions. I think it’s very important for individuals who are using the drug to escalate the dose not only if the patient is not responding but [if] they have a partial response. I will still escalate because you can get transfusion independence out of this.

Another analysis on MEDALIST looked at OS [and] the main conclusion is that patients who responded had better survival, and this is probably not a surprising finding. Of course, improvement of survival with an intention-totreat intervention vs placebo has been quite difficult in lower-risk MDS in general. I do think over time with earlier use and optimizing these therapies we might hopefully get there.

Another important presentation from MEDALIST was regarding multiple transfusion-independence episodes. Why is that important? Because some people will look at the 30-week median transfusion independence with luspatercept and say, Why would I treat someone to get 7- or 8-weeks median transfusion independence?

What the secondary analysis has shown is that most patients who respond to luspatercept, 80%, will have multiple transfusion-independence periods. Meaning that you can get more than 8 weeks, even if they need 1 unit of blood for whatever reason—it could be incidental, a gastrointestinal bleed or something else—that breaks your transfusion-independence period. Then you go onto another one. One patient had 10 different transfusion-independence periods, which is quite impressive. We are still trying to figure out from a risk-score criteria how to combine these together. The bottom line is that saying 32 weeks is the median transfusion-independent period somewhat underestimates the benefit patients get with this drug.

Komrokji: There are a few presentations of real-world data complementing what you discussed and data from the MEDALIST. A very consistent message to what’s published.

Dezern: It is fun to think about having novel options here. Historically, one of the reasons providers have pushed off the use of a new agent is because we have so few. That is changing and if we’re able to use luspatercept earlier, we can then bring in novel agents as they become available. I’m hopeful we can improve outcomes for our patients with MDS.

Imetelstat may be just that. It’s an interesting drug, with a different mechanism of action than what we’ve spoken about so far. It’s a telomerase inhibitor, and we believe [it] targets cells with high telomerase activity as well as those that have increased expression of the telomerase reverse transcriptase expression. This is a drug we’ve known about for a little while, and data from the phase 2 portion of IMerge [NCT02598661] has already been published a few years ago. At ASH, Uwe Platzbecker, MD, and colleagues went through to describe a very specific subset of those original patients and we can learn a fair amount from these data.^9,10

What they looked at was the clinical benefit in patients who were non-del(5q), had not had thalidomide, and had not had an HMA. These patients became transfusion-independent for greater than 1 year on the trial receiving imetelstat. Ultimately, this was only 11 out of the original 38 patients who were in the early portion of the trial and met those very specific criteria. They fall into a group who many would consider having a high transfusion burden. They were getting approximately 6 units in the 8 weeks prior to the time of enrollment. All previously had ESAs, which is part of how they became eligible for the original trial. The data showed that the [median] duration of transfusion independence was approximately 92 weeks. The OS was almost 5 years, at a little less than 60 months. Thankfully, in terms of safety with this mechanism of action for imetelstat, none of the 11 patients progressed to acute myeloid leukemia. This may not surprise you given that they’re low risk and they responded to a drug.

What ties together many of the themes that we’ve been speaking about is that it looks like, at least from the data presented, that when these patients became transfusion independent, they had a reduction in the mutational burden of their SS3P1. This is a lower-risk group of patients, but if you’re truly changing the amount of that mutation that is causing the disease, that may be important. In over half of those who had mutational data, there was a greater than 50% reduction in that varietal frequency. It’s compelling data, hinting at a changing natural history.

I’m optimistic that this is an active drug.

Komrokji: The question would always be how do we think of sequencing those? We have now 2 active therapies, luspatercept and imetelstat.

Zeidan: That’s a great question and a very good problem to have. If you asked me 2 or 3 years ago, I would tell you 2023 was going to be the year of high-risk MDS. In turns out, 2023 is actually the year of lower-risk MDS. We have to see the full data, but COMMANDS’ primary end point has been met.7 We know the drug is very active in the second-line setting. With imetelstat, the data have been shared in quite a bit of detail in the news release and are quite compelling.9 It’s not common to see a transfusion-dependence rate that holds in the phase 3 at 38% with a compelling duration. The high-level summary of the safety was consistent with what all of us have seen before. If these data pass scrutiny and get approvals, that could be a major change in how we treat MDS. If we have the same discussion in 2024, it would be a very interesting discussion.

Assuming the trials are reviewed and are positive in terms of the regulatory assessment, luspatercept could potentially become a firstline option. I don’t think ESAs are going to go away. Certainly, they are going to be [useful in] patient subsets. One question that many of us will ask is how does ESA work if you give imetelstat? We don’t cure these patients; they are going to go through several therapies. One of our goals is to think of the total therapy. The second thing to think of is imetelstat use in the second-line setting, if that’s approved. It would be established as a post-CsA [cyclosporin A] failure. The question is going to be, do you use it before or after luspatercept?

Lower-risk MDS is looking like I think very exciting in the next few years. Patients are going to benefit from all these results.

Komrokji: We must start changing that paradigm in lower-risk MDS. It’s always palliative [treatment to achieve] transfusion independence [and] one-third progress to high-risk MDS, and the others, even if they stay in lower-risk category, half die directly from the disease or the interaction of the cytopenias with other comorbidities.

We could do better, and we are starting to introduce treatments earlier that are active and find those patients at risk Another investigative agent, a pyruvate kinase activator, AG-946, seems a little bit out of the norm [of what] we would think of for MDS.¹¹

Dezern: There have been some historical data that show an acquired pyruvate kinase deficiency in patients who with MDS. The hypothesis is that there’s a red blood cell–specific glycolytic enzyme of pyruvate kinase that may be responsible for the anemia pathogenesis in some patients with low-risk MDS. AG-946, is a new small molecule, an allosteric activator of wild-type and mutated forms of that very specific red cell enzyme for pyruvate kinase. It’s possible that if you can use AG-946 as a treatment for anemia that could enhance the red cell functionality and survival of those red cells by increasing the glycolysis and the differentiation of the retroid cells in the marrow.

There’s some reasonable preclinical data, and what the abstract at ASH displayed was that investigators have a planned phase 2a/2b multicenter study specifically for lower-risk MDS, which is a thoughtful study design because they really want to understand proof of concept. They’re being quite prescriptive in their categories of eligible patients based on transfusion burden: non-transfusion dependent group, a lot of transfusion burden, and a high transfusion burden. Then they’re going to give a lower dose of AG-946 in the phase 2a portion to 20 patients based on their current proposal and then see if they get a hemoglobin response or if they have patients become transfusion-independent by this mechanism. If the proposed trial meets the criteria for hemoglobin response, then the phase 2b portion will enroll nearly 100 patients in a randomized fashion at different doses. Interestingly, the phase 2a portion is the highest dose that you could potentially get randomly assigned to if it proceeds to the phase 2b. There’s also a placebo in the 2b part.

Komrokji: There was also a phase 1 study [NCT05030675] about the SYK [spleen tyrosine kinase] inhibitor in patients with MDS.

Garcia-Manero: We know that there is quite a bit of innate immune dysregulation in myeloid cells in these patients. The reality is that there is cytokine deregulation. There are probably alterations in the microenvironment in those patients with lower-risk disease and a lot of this signaling through toll-like receptors, etc.

Fostamatinib [Tavalisse] is a drug that has an indication for ITP [immune thrombocytopenia]. [It] has a proven track record of safety and efficacy in that context. The phase 1 study was designed for patients with low-risk disease with this idea that inhibiting SYK may have kind of a role in the autoimmunity pathway in these patients.

Komrokji: There is also the development of oral HMAs.

Garcia-Manero: This story starts many years ago when we were developing oral azacitidine. This drug is now approved for AML as postconsolidation therapy. In those studies we saw that the pharmacokinetics [PK] were very low, meaning that investigators were giving these oral azacitidine, but the exposure was kind of almost subliminal. However, patients were responding with rates of response of approximately 40%. There was this perplexing dichotomy.

We started thinking that perhaps we could use even lower doses of HMAs in patients with low-risk MDS…. This includes approaches developed to fully absorb these HMAs, which resulted in combining the HMA, decitabine or azacytidine with the drug cedazuridine, a cytidine deaminase inhibitor. The drug that inhibits the enzyme that inhibits absorbing this drug.

The first drug approved that you mentioned is this drug called ASTX727, an oral formulation of decitabine/cedazuridine. This comes in a tablet with the 2 drugs in a fixed ratio: decitabine is 35 mg and cedazuridine is 100 mg. It’s not weight-based and this pans out to be identical PK to intravenous decitabine. The drug is very effective at the beginning.

What they have done [and presented this year] is a new formulation of this oral decitabine/ cedazuridine. There were 6 different doses of the decitabine with the cedazuridine, which is very important because it gave us a true different formulation of this oral decitabine/cedazuridine for low-risk MDS. The bottom line is I think there’s going to be a role for these oral HMAs in low-risk MDS either with oral decitabine/ cedazuridine with a different dosing schedule or maybe a few years from now the oral azacitidine that we used to give.