Real-World and Post-Hoc Analyses Support Luspatercept Use in Low-Risk Myelodysplastic Syndrome

Guillermo Garcia-Manero, MD

In an interview with OncLive^®, Guillermo Garcia-Manero, MD, highlighted 3 key analyses of luspatercept-aamt (Reblozyl) in low-risk myelodysplastic syndrome (MDS) presented at the 2024 EHA Congress, and expanded on their implications for the treatment paradigm.

According to data from a real-world study investigating the clinical benefit of luspatercept dose escalation among patients with low-risk MDS in the United States, more patients achieved red blood cell transfusion independence (RBC-TI) with luspatercept after undergoing their first dose escalation. Data presented at the meeting by Kashyap Patel, MD, of Carolina Blood & Cancer Care Associates, also showed that 47% and 43% of those who were previously RBC transfusion dependent (RBC-TD) achieved TI for at least 8 weeks following their first and second dose escalation, respectively.¹

Additionally, findings from a post-hoc analysis of the phase 3 COMMANDS trial (NCT03682536) were presented by Esther Oliva, MD, of Massachusetts General Hospital, and demonstrated a correlation between increasing hemoglobin levels and clinically significant improvements in quality of life (QOL) for RBC-TD patients with low-, very low–, or intermediate-risk MDS. Moreover, patients who were RBC-TI and achieved a hemoglobin level of 10 g/dL or higher were more likely to experience a meaningful improvement in patient-reported outcomes (PROs) for all primary domains, including anemia-related toxicities and fatigue.²

“We’re starting to gain momentum [in low-risk MDS],” Garcia-Manero stated during the interview regarding this research. “To see real-world data from Dr Patel [showing] this kind of activity [with luspatercept] is quite satisfactory, [as is] seeing the data from Dr Oliva showing an improvement in QOL.”

Lastly, results from a study conducted by Garcia-Manero and colleagues showed that luspatercept significantly improved erythroid (HI-E), neutrophil (HI-N), and platelet (HI-P) lineages in the COMMANDS study population. More patients achieved HI-E with luspatercept vs epoetin alfa, and a numerical improvement in platelet and neutrophil counts was observed during the initial 24 weeks of treatment.³

Together, these studies underscore the safety and real-world potential of luspatercept in lower-risk MDS, substantiate the current practice of luspatercept dose escalation, and support re-evaluation of target hemoglobin levels in this population, Garcia-Manero stated.

Garcia-Manero is a professor, chief of the Section of Myelodysplastic Syndromes, deputy chair of Translational Research, and the fellowship program director in the Department of Leukemia in the Division of Cancer Medicine at The University of Texas MD Anderson Cancer Center in Houston, where he is also the chair of the Faculty Senate.

OncLive: What is the importance of assessing the impact of dose escalation on real-world clinical benefit with luspatercept? Please highlight the key findings reported from the study.

Garcia-Manero: The way we administer luspatercept and the way it is approved follows a dose-escalation schema. You start with 1 mg/kg, and the drug is given every 3 weeks. You give a couple of doses and if there’s no response, then you can increase the dose to 1.3 mg/kg and then 1.75 mg/kg, [which is] the cap dose. My understanding is that this was determined by the FDA. It is a bit of a safety issue, saying that if we go too high with hemoglobin which we would like to see in the majority of our patients, this could be detrimental because if you have uncontrolled erythrocytosis with a hemoglobin level of 20 g/dL, our patients may experience complications. Fortunately, or unfortunately, we are never able to achieve levels that high in MDS, but this is a safety concern, and we need to follow the FDA guidelines.

Here, the questions are: How many patients are dose escalated and what is the impact? This real-world analysis shows that the majority of patients were dose escalated. This means that the first 1-mg dose may not be the most effective way to start, and that perhaps, at some point, authorities should reconsider this issue. Should we start at 1.3 mg/kg? Is 1.75 mg/kg actually the highest dose level we can administer? This is being evaluated in a couple of clinical trials looking at luspatercept up front. It will be interesting to know whether instead of 1 mg/kg we could go directly to 1.3 mg/kg, and whether we could escalate beyond that in the future.

What are the implications of these data for patients with LR-MDS who are RBC-TD vs RBC-TI?

What these data are telling us is that this may not be a very effective way to use this drug. One has to think about cost of the drug, cost of transfusions, and whether we could maybe skip this and go directly. We need more data, but these findings strongly suggest that this [dosing scheme] needs to be investigated [further]. These data will not support us switching our practice and starting at 1.3 mg/kg. However, it gives ideas to people who do clinical research in this area in terms of new trial design and [could] perhaps [engender] discussions with authorities [regarding] the most efficient and safe dose of this compound. It also shows that dose escalation [of luspatercept] is important not only in transfusion-dependent patients, as there was a cohort of patients that were RBC-TI. This is interesting because the level of the drug [was determined] for patients who are candidates for transfusion. So [a patient] could have a hemoglobin level of 8 g/L and never receive a transfusion but be a candidate for transfusion. In a way, the drug could be started early. It appears that those patients who were RBC-TI also benefited from the dose escalation. That is quite intriguing, and something that we need to understand better.

What knowledge gaps did the post-hoc analysis of the COMMANDS study aim to address, and what did it reveal about the relationship between hemoglobin and QOL in patients with lower-risk MDS?

Over the past couple of years, we’ve seen 3 new indications for patients with lower-risk MDS, 2 of them for luspatercept. The main goal or primary end point of these clinical trials that led to the approval of luspatercept was improvement in hemoglobin. The question is, what happens when you achieve that? Do the patients feel better? Do they have any improvement in QOL? This was a very important presentation by Dr Oliva, who’s an expert in this area and has been working on this for a long period of time. She utilized the data from COMMANDS, [which is a] phase 3 randomized study comparing luspatercept vs an erythropoiesis-stimulating agent [ESA].

She used these PRO tools that have different domains and related this to increases in hemoglobin. What she showed is that those patients who had an improvement in hemoglobin experienced a benefit in terms of their reported PRO issues. What was more interesting was that this [benefit] was clearer for those who have a hemoglobin increase over 1.5 g/dL and for those who achieved hemoglobin levels over 10 g/dL. What this suggests is that designing this trial with a goal of increasing hemoglobin should improve the QOL and wellbeing of our patients, which is what we want as clinicians.

How might these findings inform clinical decisions regarding the target hemoglobin level for anemic, RBC-TD patients within this population?

This sounds easy to do, but it’s quite complex. There are many variables that affect how we analyze these QOL end points in clinical trials. When we do this in a randomized study, we have strict guidelines for transfusion. For instance, in the prior phase 3 MEDALIST trial [NCT02631070], we compared luspatercept vs placebo, but then we were trying to achieve levels of hemoglobin that were predetermined on the trial. In a way, we were achieving that with transfusions, meaning that affected hemoglobin levels that we were trying to achieve with the drug. Sometimes this is not as straightforward in clinical trials. [However,] this [study is a] successful example [suggesting] that it can be done. It also suggests that we should incorporate this clearly into all our trials for lower-risk MDS, and that we need to learn about these implications. What is the [target] hemoglobin level? Is it just [reaching RBC-TI]? Is it [a level] of 9, 10, or 11 g/dL? This has implications for the use of luspatercept, where there [could be] a dose-escalation schema in which one could aim [to achieve] higher hemoglobin levels. Also, the analysis is agnostic for the drug, so it was not specifying whether this was for luspatercept or an ESA, and patients were treated with either drug. Understanding the difference between luspatercept and ESAs and understanding these issues related to dose escalation will be quite important. Overall, it’s a very important first step.

What was the rationale for assessing the impact of luspatercept on erythroid, neutrophil, and platelet lineages in ESA-naive low-risk MDS?

When these drugs were designed, they were thought to be erythroid-maturation agents, and were thought to act in the later steps of blood cell formation. I was not anticipating any effect on white blood cells [WBC] or platelets. In one of the SOHO Annual Meetings a few years ago, someone asked leaders in the field whether you could see [those kinds of] responses, and I was curious about this question. The reality is that when we did the MEDALIST trial and we saw those responses. Now, that study compared [luspatercept with] placebo, and we saw that there was not only an increase in hemoglobin, but also an increase in platelets and WBC. It’s, [therefore,] logical to ask the same question in the phase 3 COMMANDS trial.

How did hematologic improvement rates and safety profiles compare between patients treated with luspatercept and those treated with epoetin alfa in the study you presented?

If you look at the HI-E [rate], it was 74.2% with luspatercept and 53.0% with an ESA; the HI-N [rate] was 33.3% vs 25.0%, and HI-P [rates] were 42.3% vs 30.0%, [respectively]. The HI-E is very significant, because that’s the erythroid level, which is the end point of the study. However, these patients are not always cytopenic. This means that they always have anemia, but they may or may not have neutropenia and may or may not have thrombocytopenia. The number of patients may not be enough to show statistical significance. However, there is a clear numerical trend [indicating] that luspatercept improves platelet counts and may improve neutrophil count in patients with lower-risk disease. [These data show that] this drug doesn’t cause cytopenia, so it’s quite safe. That is very clear, even if these P values are not significant. [The agent] does not induce neutropenia, and it will not induce thrombocytopenia. The next question is, why does this happen? Also, do these drugs have another role beyond erythropoiesis? The overall message is that this drug is safe and may improve platelet and neutrophil counts. Hopefully, in the future, we’ll better understand why this happens.

Are any next steps planned based on this research?

We have a study ongoing here, at The University of Texas MD Anderson Cancer Center. In this investigator-initiated trial, we are enrolling patients with low-risk MDS who have significant cytopenias, such as thrombocytopenia or neutropenia, in addition to anemia. We want to understand the dynamics of changes in platelet count and WBC count, so we selected a population that is more cytopenic than the typical population [enrolled onto the] COMMANDS trial or other studies [in this space]. That will allow us to better visualize this bi- or tri-lineage effect. Each cohort will accrue approximately 20 patients. We also have designed an ambitious correlative analysis looking at single-cell RNA sequencing profiling, to understand what happens at the molecular level when we administer these drugs.

Looking ahead, where do you see the treatment paradigm heading in MDS?

We’re seeing a lot of progress all of a sudden. A lot of it is occurring in low-risk disease, but I’m seeing some improvement in [other areas,] as well. We saw luspatercept approved [by the FDA] in the second line [followed by the] first line, and we just had the FDA approval of imetelstat in the second line. One potent drug, KER-050, is going into phase 3 trials. We saw the phase 3 SINTRA-REV trial [NCT01243476] in Spain [evaluating] low-dose lenalidomide [Revlimid} in [patients with] RBC-TD MDS and deletion 5q [mutations, as well as] data with oral hypomethylating agents [HMAs.]

In high-risk disease, we are waiting for something that will improve the survival of our patients. We’re particularly waiting [to see more from] the phase 3 VERONA trial [NCT04401748] of [azacitidine (Vidaza) in] combination with venetoclax [Venclexta], but we’ve done quite a bit of work with oral HMAs. My group led the development of oral decitabine and cedazuridine, which has been approved now for MDS [by the FDA] and for acute myeloid leukemia in Europe. We have also shown very potent combinations within oral decitabine, venetoclax, and [others], and we’re starting to see quite a bit of improvement in transplant outcomes in MDS. There was also the recent FDA approval of the IGF1 inhibitor ivosidenib [Tibsovo] in relapsed/refractory MDS. These drugs are being tested in frontline trials, and we have similar data with IDH2 inhibitors.

[Overall,] we are moving in a good direction [in MDS], certainly better than we were 3 or 4 years ago. We have unmet needs, [including efforts to] target p53, HMA failures, and we saw a couple of disappointing studies with sabatolimab and magrolimab, but [overall], we are making quite a bit of progress.

References

1. Oliva E, Yucel A, Lord-Besse J, et al. Relationship between hemoglobin and quality of life in RBC-TD patients with lower-risk myelodysplastic syndrome receiving luspatercept or epoetin alfa. Presented at: 2024 EHA Congress; June 13-16, 2024; Madrid, Spain. Abstract P774.
2. Patel K, Chatterjee D, Hughes C, et al. Real-world dose escalation and outcomes among patients with lower-risk myelodysplastic syndromes receiving luspatercept in clinical practice. Presented at: 2024 EHA Congress; June 13-16, 2024; Madrid, Spain. Abstract P768.
3. Garcia-Manero G, Della Porta MG, Santini V, et al. Multilineage and safety results from the COMMANDS trial in RBC-TD, erythropoiesis-stimulating agent–naive patients with very low-, low- or intermediate-risk myelodysplastic syndromes. Presented at: 2024 EHA Congress; June 13-16, 2024; Madrid, Spain. Abstract P780.