Updates in low/intermediate-risk MDS

Visual Abstract

Abstract

Low- and intermediate-risk myelodysplastic syndromes (LR-MDS and Int-MDS, respectively) are characterized by ineffective hematopoiesis, along with the presence of at least 10% dysplasia in one cell line, accompanied by a low number and depth of peripheral blood cytopenias, a low bone marrow blast percentage, and a score of ≤0 on the Molecular International Prognostic Scoring System (IPSS-M). The information gleaned from mutational profiles at the time of myelodysplastic syndrome (MDS) diagnosis and over subsequent time points help with classification and prognosis, guiding therapeutic decisions. In LR-MDS, these decisions are initially focused on improving symptom control and optimizing hematologic parameters. New therapeutic options to reduce the red blood cell (RBC) transfusion burden have emerged since 2020 and include luspatercept and imetelstat. Erythropoiesis-stimulating agents and lenalidomide also address anemia and are generally recommended to start at the time of transfusion dependency, although emerging data suggest that an earlier start of these interventions might offer clinical benefits. Patients can derive years of benefit from these approaches in LR-MDS, but despite these therapies, ultimately MDS will evolve into higher-risk MDS (HR-MDS)/acute myeloid leukemia. Even though most LR-MDS patients present with anemia, patients can have isolated thrombocytopenia for which thrombopoietin receptor analogues can be used if blasts are low. Immunosuppressive therapy such as antithymocyte globulin is favored in the hypocellular MDS setting. Dose-modified hypomethylating agent use can be considered for LR-MDS, although neither overall survival (OS) nor progression-free survival (PFS) has been shown to improve with this approach. Targeted therapy directed to the presence of an IDH1 mutation is U.S. Food and Drug Administration (FDA) approved for the rare IDH1 mutated MDS (<10% of the time) and consideration to use an IDH2 inhibitor for IDH2 mutated MDS (<5% of the time) is reasonable. Interestingly, IDH mutations seem to appear with increased frequency in older patients and in patients with underlying autoimmune/rheumatological disorders.¹

Learning Objectives

• Describe the expected clinical outcomes for patients with LR-MDS based on genomic profiling with hematologic and cytogenetic parameters

• Assess treatment in the clinical management of LR-MDS to optimally tailor patient care for anemia, isolated thrombocytopenia, and pancytopenia

• Review novel and emerging therapeutic approaches for patients diagnosed with LR-MDS

CLINICAL CASE

A 75-year-old woman with a history of irritable bowel disease on mesalamine therapy was referred to the hematology oncology department after routine bloodwork showed macrocytic anemia. She had previously taken oral copper and vitamin B6 supplements, and these nutritional values were in the normal range. She consumed 1-2 glasses of wine at least 2 times per week. The complete blood count showed hemoglobin (Hb) of 7.3  g/dL, mean corpuscular volume of 104 fL, a white blood count of 2.0  ×  10⁹/L, an absolute neutrophil count of 1.2  ×  10⁹/L, and a platelet count of 140  ×  10⁹/L. Iron, vitamin B12, and folate levels were normal, and the erythropoietin level was 110  U/L. A bone marrow (BM) aspirate and biopsy showed dysplastic changes in 25% of the erythroid cells, 1% blasts, and 50% ring sideroblasts, as well as <10% kappa predominant plasma cells. Next-generation sequencing (NGS) revealed mutations in DNMT3a p.R882H VAF 46.5%, SF3B1 p.K700E VAF 48.2%, TET2 p.K1243 VAF 31%, and TET2 p.A1379 VAF 32%. Cytogenetic testing was normal. Bone marrow biopsy reported a diagnosis of myelodysplastic syndrome (MDS) with low blasts and SF3B1 mutation (MDS-SF3B1) per ICC 2022 and WHO-5.^2,3 The calculated IPSS-R was 2.5 (low risk) and IPSS-M was −0.35 (moderately low risk). She started on luspatercept (LUS) at 1  mg/kg subcutaneous injections once every 3 weeks, reached transfusion independence by the third dose of LUS, and continued therapy at the same dose/frequency. She has a noted kappa-predominant monoclonal gammopathy, and her creatinine is normal. Annual skeletal surveys are normal.

Introduction

Myelodysplastic syndrome (MDS) is a result of dysfunctional production of cells from the bone marrow, whereby the presence of >10% dysplasia in any cell lineage is a standard requirement for the diagnosis of this cancer, with few exceptions. The bone marrow dysfunction can be due to the low quantity and/or low quality of cells, which for most patients leads to poor red blood cell (RBC) production and subsequent clinical symptoms of fatigue and other issues (ie, chest pain, dyspnea on exertion, and headaches).

The incidence of MDS is increasing due to our aging population and also due to patients surviving their first cancer diagnosis and exposure to chemotherapy, radiation, or both. Therapy- related MDS is at higher risk of progression to acute leukemia, whereas lower-risk MDS can often be managed by watchful waiting and/or supportive measures. At the time of diagnosis, about two-thirds of patients will have low-risk MDS (LR- MDS), with the majority over the age of 60 years. There have been impressive strides in the treatment of MDS with 4 drugs approved by the US Food and Drug Administration (FDA) in the last 5 years (Figure 1). Prior to this, there had been advancements through the adoption of genomic testing using NGS and improvements in the classification and prognostication systems. Furthermore, understanding of the pathogenesis of myeloid malignancies with cardiovascular disease and rheumatological/inflammatory diseases is being advanced.

Figure 1.

Progress in myelodysplastic syndromes over the past three decades. CHIP, clonal hematopoiesis of indeterminate potential; ICC, international consensus conference; IPSS, International Prognostic Scoring System; IPSS-R, revised IPSS; IPSS-M, molecular IPSS; NGS, next-generation sequencing; WHO, World Health Organization.

Because of the significant variability in the clinical outcomes for patients, prognostic scoring systems are a central part of assessment to guide the selection of clinical treatment. The Revised International Prognostic Scoring System (IPSS-R) is a commonly used tool, and a cutoff of ≤3.5 defines the LR-MDS population.⁴ The integration of molecular information into the International Prognostic Scoring System (IPSS-M) is the most accurate scoring system for MDS, and a cutoff of less than zero identifies patients with LR-MDS (moderately low, low, and very low), with a predicted median overall survival (OS) rate of 4.6 years.⁵ There are challenges that remain when using the IPSS-M to prognosticate risk for patients with MDS since this scoring system does not account for specific hot-spot mutation sites that may confer increase risks, germline predisposition states (ANKRD2, DDX41, others), nor coexisting conditions like rheumatologic disorders (eg, vasculitis, lupus, VEXAS [vacuoles, E1 enzyme, x-linked, autoinflammatory, somatic syndrome], and others) and/or overlapping bone marrow disorders (monoclonal gammopathy and large granular leukemias/lymphoma), which could equally affect treatment decisions.

The main approach to LR-MDS is aimed at improving anemia (with or without additional cytopenias) and the attendant clinical symptoms that affect quality of life (QoL), such as physical fitness and personal independence. Symptomatic anemia can lead to long-term transfusion dependence (TD), which leads to dependence on frequent and often lengthy clinic visits, possible transfusion reactions and alloimmunization, and concerns about iron overload. Treatment goals for MDS patients include transfusion independence (TI), decreased symptom burden, decreased infection and bleeding rates, and improved QoL.

Management of iron overload in patients diagnosed with LR-MDS is an important aspect of their medical treatment. Progressive iron accumulation due to red blood cell (RBC) transfusions and/or defective iron utilization can result in end organ damage from the iron-related oxidative stress (particularly in the liver, heart, and endocrine organs).⁶ Given the overlap of comorbid conditions seen in elderly patients with that of iron overload, it can be challenging to attribute the cause of organ dysfunction (eg., vascular damage, heart disease, or renal or hepatic dysfunction). Serum levels of ferritin and transferrin saturation are used to help quantify the burden of iron overload or encourage the use of magnetic resonance imaging (MRI) for liver or cardiac deposition. Ferritin levels >1000  ng/mL showed a dose- dependent impact on OS for patients diagnosed with LR-MDS and had an independent impact on OS in transfusion-dependent LR-MDS in the European LeukemiaNet prospective MDS registry.^7,8 The goal of iron chelation is to prevent the long-term complications and improve the QoL for patients. Indication for starting chelation therapy is if a patient has received over 50 units of packed RBCs in their lifetime or if ferritin levels are persistently over 1000  ng/mL, although these general guidelines vary and can be context dependent. Furthermore, the tolerability of the oral agents (deferasirox or deferiprone) challenge compliance, and alternative options may be needed (subcutaneous deferoxamine), particularly if oral therapy is not tolerable or if elevation of liver function tests preclude continuation. A more comprehensive review/primer on iron chelation in MDS was recently published.⁹

Ideally, the goal is to cure this myeloid neoplasm. Hematopoietic stem cell transplant (HSCT) is the only curative therapy, but it is rarely indicated for patients in the LR-MDS setting, given their long median OS, their average elderly age at diagnosis, and the typical comorbidities that increase morbidity and mortality from such a procedure. Nevertheless, HSCT can be offered for some LR-MDS scenarios that are deemed concerning. For example, consultation for HSCT is recommended for patients with (1) continued transfusion dependence despite exposure to standard therapy for lower-risk disease, (2) severe cytopenias, or (3) high-risk mutations and/or clonal/cytogenetic progressions. For those patients that have a low blast percentage in the marrow and continued unremitting transfusion dependency, referral for HSCT should be pursued since they have a significantly inferior survival rate, increased hospitalization rates, and poorer health-related QoL.¹⁰ In the BMT CTN 1102 study, a 21.3% improvement in OS was seen with the use of reduced-intensity HSCT for patients with higher-risk MDS (HR-MDS) aged 50 to 75 years if a donor is available (P  =  0.0001), and this benefit did not come at a cost of reduced QoL.¹¹

Current Treatment Strategy for LR-MDS and Int-MDS

The management of LR-MDS has focused on supportive care, including watchful waiting, transfusion support for anemia and/or thrombocytopenia, and prevention and management of infections (Table 1 and Figure 2). The cornerstone of pharmacologic therapy has been the use of erythropoiesis-stimulating agents (ESAs), particularly if the erythropoietin level was <200  U/L along with a low transfusion burden (less than 2 units per month). Overall response rates are 20% to 40%, with a duration of benefit of up to 1.5 years. Other agents considered in the management of LR-MDS include lenalidomide (LEN), luspatercept (LUS), immunosuppressive therapy, hypomethylating agents (HMAs), and imetelstat (IMET). The management approach to isolated thrombocytopenia, isolated neutropenia, and hypocellular MDS has not changed in some time.¹² This update will focus on management of MDS with anemia.

Table 1.

Available therapies in LR-MDS and Int-MDS^{14-16,19,20,36-46}

Agent	Mechanism of action	Population	National Clinical Trial (NCT) number	Phase of study	Status	Reference
Epoietin alfa	ESA	LR-MDS w anemia	NCT01381809 and EPOANE3021	3	EMA approved	37
Darbepoetin alfa	ESA	LR-MDS with anemia	NCT00095264	2	Off-label	38
Lenalidomide	Immune regulatory agent	LRMDS del(5q)	NCT00179621	3	FDA/EMA approved	39
		LR-MDS without del(5q)	NCT01029262	3	Off-label	40
Luspatercept	TGF-β inhibitor	MDS-RS or with SF3B1m	NCT02631070 MEDALIST	3	FDA/EMA approved	14
		LR-MDS with anemia	NCT03682536 COMMANDS	3	FDA approved	15,16
Imetelstat	Telomerase inhibitor	TD non-del(5q) R/R or ineligible to ESA therapy LR-MDS	NCT02598661 IMerge	3	FDA approved	19
Ivosidenib	IDH1 inhibitor	R/R IDH1m MDS	NCT02074839	1	FDA approved	41
Olutasidenib	IDH1 inhibitor	R/R IDH1m MDS and tx naive IDH1m MDS	NCT02719574	1 + 2 w/and w/o AZA	Off-label	20
Enasidenib	IDH2 inhibitor +/− HMA	R/R IDH2m MDS and upfront IDH2m HR-MDS	NCT06577441 and NCT03744390	both 2	Off-label	22,23
Azacitidine (AZA)	HMA	MDS with pancytopenia	NCT01720225	2 + 3	FDA approved	37
Decitabine (DEC)	HMA	MDS with pancytopenia	NCT01720225	2 + 3	FDA approved	37
Oral decitabine	HMA	MDS (RA, RARS, RA-EB, Int-1, Int-2, CMML)	NCT02103478 and NCT03306264	2 + 3	FDA/EMA approved	43,44
Eltrombopag	TPO mimetic	MDS without excess blasts, with thrombocytopenia or pancytopenia	NCT02928419 EQoL-MDS	2	Off-label	45,46
ATG/CsA	Immunosuppressive	LR-MDS with pancytopenia	Retrospective study	NA	Off-label	47

+/−, plus/minus; ATG/CsA, antithymocyte globulin/cyclosporine A; CMML, chronic myelomonocytic leukemia; ESA, erythroid-stimulating agent; HMA, hypomethylating agent; IDH1m, mutated isocitrate dehydrogenase; LR-MDS, low-risk myelodysplastic syndromes; RA, refractory anemia; RA-EB, refractory anemia with excess blasts; RARS, refractory anemia with ring sideroblasts; R/R, relapsed refractory; TD, transfusion dependent; TPO, thrombopoetin.

Figure 2.

Algorithm for treatment of lower-risk MDS. ATG, antithymocyte globulin; EPO, erythropoietin; ESA, erythropoietin- stimulating agent; HMA, hypomethylating agent; IPSS, International Prognostic Scoring System; IPSS-M, molecular IPSS; IPSS-R, revised IPSS; IST, immunosuppressive therapy; MDS, myelodysplastic syndromes. Figure created in BioRender. Carraway, H. (2025) https://BioRender.com/ltfyk9ot.

Transfusion dependent del5(q) MDS is treated in the upfront setting with a LEN-based approach yielding a median 2-year duration of response in over 70% of patients with an oral medication taken twice daily.¹³ It has been a challenge to identify agents in the management of MDS that have been able to meet response criteria like that seen with LEN. However, in 2019, the FDA approved LUS for patients with MDS with ring sideroblasts with or without SF3B1 mutation who were relapsed/refractory (R/R) or ineligible for ESA therapy.¹⁴ A broader approval was granted in 2023 for very low-risk to intermediate-risk MDS (Int-MDS) patients who may require RBC transfusions in the first-line setting, regardless of ESA use or SF3B1 status^15,16 LUS acts as a ligand trap for transforming growth factor beta and inhibits downstream signaling by SMAD2/3, leading to late-stage erythroid maturation. In the COMMANDS study,¹⁶ RBC-TD non-del5(q), and ESA naive LR-MDS pts (IPSS-R defined) with EPO levels <500  U/L were randomized between LUS and epoetin alpha. LUS showed a 59% RBC TI rate for at least 12 weeks in duration, with a concurrent mean Hb increase of 1.5  g/L compared with 31% in the epoetin alfa group (P  <  0.0001) with a significantly longer duration of response with LUS (P  =  0.005). Interestingly, there was a high number of SF3B1-mutated MDS patients (63%) and/or the presence of 73% ring sideroblasts (RS), which was higher than the expected rate of ~35% for the general LR-MDS population. It is not yet known why there is an improved response rate to LUS in MDS with RS and/or SF3B1 mutations. LUS is FDA approved for all patients with MDS regardless of the NGS mutation profile and/or the presence/absence of RS. Nonetheless, given the clinical trial demographics in the COMMANDS study, the evidence to support the use of LUS in non-RS and/or non-SF3B1-mutated patients was not empowered to show a difference, with outcomes appearing equal for TI at 53% vs 50% for non-RS, so questions remain about the best upfront therapy for these cohorts. Recently, a post-hoc analysis has been completed and shows that there is a trend for improvement in OS for those patients who received LUS [median OS not reached for LUS compared to 46.7 months for epoetin alfa arm (HR, 0.33; P  =  0.02)].¹⁷

CLINICAL CASE (continued)

After 2 years of response to LUS, the patient unfortunately had a new transfusion requirement, needing 6 units of packed RBCs every 2 months and Hb was 7.1  g/dL. A repeat bone marrow biopsy showed similar findings except for 3% blasts and stable NGS results consistent with MDS-SF3B1. Cytogenetics had a new del(12p). Absolute neutrophil count was 1.1  ×  10⁹/L, and platelets were 110  ×  10⁹/L. RS were present at 25%. Erythropoietin level was 277  U/L. The features of the current time point remain consistent with moderate low risk (score −0.35) per IPSS-M, although this is not a validated tool for dynamic application. She started therapy with IMET at 7.5  mg/kg IV every 4 weeks and became TI after 10 weeks.

IMET is a first-in-class telomere inhibitor that was approved by the FDA and the European Medicines Agency (EMA) for patients with LR-MDS who are relapsed/refractory to, or ineligible for ESA-based therapy. The phase 2 portion of the IMERGE study examined patients with LR-MDS who had RBC TD anemia (>4 or more units over 8 weeks) and had MDS that was R/R to, or ineligible for ESA treatment. Almost 40% of patients treated with IMET achieved the primary end point (8-week RBC-TI), and 23% achieved 24-week TI. Reversible neutropenia and thrombocytopenia (generally grade 3 or 4) occurred in two-thirds of patients as the most frequent treatment-emergent adverse events.¹⁸ These data led to the opening of the phase 3 portion of IMERGE, where LR-MDS (RBC-TD, non-del5q MDS that are R/R or ineligible for ESA therapy) patients were randomized (2:1) to either IMET or placebo. This phase 3 study also had a high number of MDS patients with RS (62%) and/or SF3B1 mutation. Compared to 15% of patients on the placebo arm, 40% of patients on IMET achieved the primary end point (RBC-TI for a duration of 8 weeks), with a median response duration of 51.6 months. Patients with RS had a higher likelihood of response to IMET compared to RS-negative status (other subgroups also showed benefit with IMET vs placebo). Impressively, among the IMET responders, the median Hb increase at 8 weeks, 24 weeks, and 1 year was 3.6, 4.2, and 5.2  g/dL, respectively. Notably, there were responders to IMET in both low (≥4 unit but ≤6 unit transfusions in 8 weeks) and high (>6 units per 8 weeks) TD patients.¹⁹ Challenges with pancytopenia from IMET caused dose interruptions and reductions and/or delays in therapy but were manageable. Treatment discontinuation occurred in less than 5% of patients.

Targeted therapies: IDH inhibition

Ivosidenib, an oral IDH1 inhibitor, was approved by the FDA for R/R IDH1 mutated LR- and HR-MDS. Results from a single-arm phase 1 clinical trial led to its approval due to a 38% complete remission (CR) plus partial response rate and an 83% objective response rate of (CR plus marrow CR). Median duration of treatment was 9.3 months with a median follow-up time of 27.1 months. At least 20% of the patients had a high or very high IPSS-R risk MDS. The IDH1 mutation is seen in fewer than 10% of MDS patients, but ivosidenib is an oral agent with a reasonable side effect profile and excellent clinical responses, and is an attractive option for R/R MDS patients. Further studies (IDIOME for IDH1m and IDEAL for IDH2m) using ivosidenib or enasidenib (ENA), respectively, for upfront treatment of IDHm LR-MDS (failed ESA), or upfront and/or relapsed IDHm HR-MDS with or without azacitidine (AZA) therapy are ongoing. NCT03503409 and NCT03744390. Other novel IDH inhibitors are being investigated and could be incorporated into the future management of MDS patients (ie, olutasidenib: NCT02719574; and ENA: NCT03744390 and NCT03383575). Notably, olutasidenib has shown promising responses (CR 27%, with duration of remission at 14.6 months and a high TI rate) in IDH1m MDS, although the number of patients (N  =  22) with long-term follow-up was low.²⁰

ENA, an oral, selective IDH2 inhibitor, has not been approved by the FDA for IDH2-mutated MDS but should be considered as an option based on promising response rates and a reasonable safety profile of this agent. IDH2 mutations occur in about 5% of patients diagnosed with MDS. ENA has been studied in the upfront setting in combination with AZA for HR-MDS (N  =  27) and as monotherapy in patients with MDS that was refractory to hypomethylating therapy (N  =  23). The overall response rate in the upfront combination arm was 74%, with 26% CR and a median OS of 26 months. The overall response rate was 35% in the R/R monotherapy arm with 22% CR and a median OS of 20 months. In this study, there were 11% and 13% LR-MDS patients in the respective cohorts.²¹ The GFM and EMSCO groups led a study using ENA with one of the three cohorts, including LR-MDS, for whom hypomethylating therapy has failed them, and went on to receive monotherapy with ENA once daily. The CR rate was 55% (6/11 patients) with a median time to response of 3 cycles and a median OS not reached.²²

Novel treatment options in LR and int-Risk MDS

Elritercept (KER-050), a novel agent targeting the transforming growth factor-beta (TGF-β) pathway, is under investigation (Table 2). The study is using a 4-week interval for subcutaneous dosing (as opposed to every 3 weeks with LUS). Elritercept inhibits select TGF-β ligands, including Activin A and B and Growth Differentiation Factor-8 and 1 (GDF-8 and GDF-1, respectively). Preclinical data show that elritercept acts on both early and late states of hematopoiesis and has an impact on thrombopoiesis. Preclinical data also show increased bone formation and improved iron metabolism/utilization, all of which support a different mechanism of action than LUS.²³ A Phase 2 study evaluating elritercept in more than 80 patients with LR-MDS (very low, low, and intermediate) plus anemia has shown hematological responses in 55% of patients over the first 24 weeks (with best responses in RS+ and EPO <500  U/L). TI was achieved in 39% of patients. Median duration of TI response has not been reached for patients who achieved TI >8 weeks from baseline through week 24.²⁴ Given these promising results, the phase 3 randomized double-blind study is evaluating LR-MDS patients that are refractory, intolerant of, or ineligible for ESA therapy to either elritercept or placebo. Exclusion criteria include prior LUS, LEN, IMET, or HMA therapy, and the primary end point is >8-week RBC-TI by week 24.²⁵ It is exciting to see alternative TGF-β agents, and it remains to be seen if there will be similarly high response rates in RS+ or SF3B1-mutated MDS.

Table 2.

Ongoing studies in MDS^{22,24,25,34,47-50}

Mechanism of action	Population	Trial name/NCT	Phase	Status	Reference
IDH1 inhibitor	(1) Treatment-naive IDH1m HR-MDS (2) R/R (HMA) IDH1m HR-MDS (3) R/R (ESA) IDH1m LR-MDS with anemia	IDIOME: NCT03503409	2	Ongoing	47
IDH2 inhibitor	(1) Treatment-naive IDH2m HR-MDS (2) R/R (HMA) IDH2m HR-MDS (3) R/R (ESA) IDH2m LR-MDS with anemia	IDEAL: NCT 03744390	2	Ongoing	22
IDH2 inhibitor + HMA	MDS, RA-EB, AML, CMML all with IDH2m	NCT03383575	2	Ongoing	NA
IDH2 inhibitor + HMA	High-risk MDS	NCT06577441: MyeloMATCH	2	Ongoing	NA
TGFB inhibitor vs ESA	NTD LR-MDS	ELEMENT: NCT05949684	3	Ongoing	NA
TGFB inhibitor + ESA	LR-MDS, non-RS, nondel5(q), failed ESA	COMBOLA: EudraCT number 2021-000596-37	1  +  2	Ongoing	34
TGFB inhibitor	R/R (ESA) LR-MDS (NTD, LTB, HTB) and CMML	NCT04419649	2	Ongoing	24
TBFB inhibitor	R/R (ESA) very low-, low-, or int-risk MDS (TD only)	RENEW: NCT06499285	3	Ongoing	25
IRAK4 inhibitor	Treatment-naive and R/R or ineligible for ESA therapy LR-MDS	NCT05178342	2	Ongoing	48
Dual IRAK1/IRAK4 inhibitor	Treatment of R/R LR-MDS	NCT05308264	1	Ongoing	49
Pyruvate kinase activator	LR-MDS	NCT05490446	2	Ongoing	50
CXCR1 and CXCR2 inhibitor +/− HMA	Nondel(5q) LR-MDS R/R (ESA/LEN/HMA)	NCT04245397	1	Ongoing	NA

AML, acute myeloid leukemia; CMML, chronic myelomonocytic leukemia; ESA, erythroid-stimulating agent; HMA, hypomethylating agent; HTD, high transfusion dependent; LEN, lenalidomide; LR-MDS, low-risk myelodysplastic syndromes; LTD, low transfusion dependent; NTD, non-transfusion dependent; RA, refractory anemia; RA-EB, refractory anemia with excess blasts; RARS, refractory anemia with ring sideroblasts; R/R, relapsed/ refractory; TD, transfusion dependent. TPO, thrombopoetin.

IRAK inhibition

Patients with hematological malignancies, particularly myeloid neoplasms, can show considerable evidence of inflammation, either due to their underlying disorder, or concurrent inflammatory disorders such as vasculitis with concurrent hematological malignancy (VCHM).²⁶ Emerging data demonstrate that there can be somatic mutations/genomic phenotypes (eg, TET2, SF3B1, U2AF1, IDH mutations) that result from and perpetuate an inflammatory state. These can drive clonal evolution due to the overactive microenvironment, as well as inhibit normal hematopoiesis. Interleukin-1 receptor-associated kinase 4 (IRAK4) is a key kinase downstream of the interleukin-1 receptor (IL-1R) and toll-like receptor signaling pathway, whose overexpression and hyperactivation are associated with cancer and inflammatory disorders. Preclinical data demonstrate that mutations of SF3B1 and U2AF1 can lead to a preferential production of the longer isoform of IRAK4, which causes maximal activation of the innate immune signaling pathways in myeloid malignancies.²⁷ Thus, the use of IRAK inhibitors and degraders to target IRAK4 has emerged as a novel therapeutic strategy given the importance of IRAK4 in the activation of nuclear factor kappa-light-chain- enhancer of activated B cells (NF-κB). (Figure 3). Clinical trials using emavusertib (CA-4948) are ongoing in MDS and acute myeloid leukemia (NCT05178342 and NCT04278768). Interim data from the TakeAim Leukemia Trial on 14 HR-MDS patients with spliceosomal mutations (SF3B1, U2AF1) receiving 300 mg BID for 7-14 days in a 28-day cycle, reported 4 patients achieving marrow CR.²⁸ R289 is an oral agent that is a dual inhibitor of IRAK1/4 is also being studied in R/R LR-MDS patients (NCT05308264) and has encouraging preliminary efficacy and has been granted fast-track and “orphan drug” designation by the FDA for the treatment of MDS.

Figure 3.

The role of IRAK4 in the inflammatory cascade and sites of inhibition. Figure created in BioRender. Carraway, H. (2025) https://BioRender.com/a5qh6gd.

Activation of inflammasomes such as NLRP3 can be a result of the epigenetic dysregulation in myeloid neoplasms due to mutations in TET2, ASXL1, and SRSF2. This leads to excess IL-1b and IL-18 release, creating a proinflammatory marrow milieu that results in ineffective hematopoiesis and cytopenias due to pyroptosis.²⁹ Early results from a phase 1 study using DFV890 in LR-MDS and chronic myelomonocytic leukemia (CMML) show a favorable safety profile with this oral agent.³⁰ (NCT0555269).

Future questions

Further questions have emerged due to the introduction of these agents into the paradigm for management of patients with MDS. There is evidence that TD is associated with a decreased OS, and particularly those MDS patients with high (and low) TD have a trend toward worse OS and PFS compared to those without TD.^7,31 Investigators have examined whether the introduction of therapy before patients have TD may be beneficial. In the phase 3 Sintra REV study, non-TD del5q LR- or Int-MDS patients had delayed time to TD and improved responses with low-dose LEN therapy compared to placebo.³² In addition, it is recognized that patients exhibiting a hematologic response to first-line therapy have a better OS compared to no response or even second-line response. The first response to therapy is often the most durable, with subsequent therapies having lower response rates and shorter durability.³³ Key questions surrounding the use of LUS include whether there is a benefit to starting it prior to when patients are TD, and whether this might be able to delay the progression of disease (improve OS and/or PFS). The phase 3 ELEMENT Trial is trying to answer this very question NCT05949684. Other questions exist regarding the optimal sequencing of agents: (LUS then ESA) v (ESA then LUS), or in combination in the upfront setting (LUS plus ESA). The COMBOLA study is hoping to identify the safest dose and best response rate in non-RS and wtSF3B1 LR-MDS patients receiving either combination LUS/ESA or LUS monotherapy after having lost response to prior ESA therapy.³⁴

Answers to these questions will likely come from large retrospective data or real-world data analyses rather than upfront phase 3 randomized studies. Finally, the COMMANDS study that led to the FDA approval of LUS was not empowered to determine the best upfront therapy for non-RS and/or wtSF3B1 MDS, so questions remain as to what therapy (ESA or LUS) is best to use in the frontline setting. The addition of IMET has been exciting due to its impressive HI-E and rate/duration of TI. However, the use of IMET is challenged by the side effect profile with pancytopenia and management thereof. Investigations focused on dose modification and/or combination therapy with marrow supportive therapy (GCSF or LUS/ESA) are an area of keen interest, with the hope of similar efficacy.

Regarding the sequencing of agents, the phase 3 IMERGE study did not permit patients with prior exposure to LEN/LUS/HMA, but it appears that similar response rates were seen to IMET therapy despite prior exposure to these agents in the phase 1 and 2 portions of the studies, although the numbers evaluated are small.³⁵ Additional data to support this will be needed for validation and better understanding of the likelihood of response to IMET with prior exposure to other therapies.

Finally, a key area of interest is trying to safely combine therapy in the upfront setting (ie, LUS/HMA) with the hope of improving OS, PFS, and duration of remission (DOR). Since there is off-label use of low-dose HMA (3- or 5-day regimen) for patients with LR-MDS with multilineage cytopenias, or as second-line therapy in the US,³⁶ future investigation will likely be aimed at combining this approach with any of the aforementioned agents. An essential component to any of the studies for LR-MDS and Int-MDS is to identify and unify the approach to measuring QoL end points for patients with MDS in prospective clinical trials.

Conclusion

New agents have been approved for the management of LR-MDS and Int-MDS, which have expanded the options for treatment in this stage of disease. The incorporation of these agents has been exciting, although it has led to more questions regarding earlier introduction of therapy, appropriate sequencing of agents, and future combinations. Investigation with novel therapies is led by improvements in targeting the TGF-b pathway and/or the inflammasome, and the mutational landscape of myeloid diseases may help optimally target disease in individual patients depending on the disease phenotype (SF3B1 vs TET2 vs IDH mutated). Understanding the response of TGF-b1 inhibitors to SF3B1m (and RS-positive) myeloid diseases remains elusive. We are eager to elucidate biology underlying the overlap of inflammatory diseases (ie, vasculitis, lupus, VEXAS, or others) with myeloid neoplasms.

Conflict-of-interest disclosure

Hetty E. Carraway receives research funding for an investigator-initiated trial from Celgene/BMS. She is a consultant for Agios, AbbVie, Daiichi, Genentech, Geron, Jazz, Novartis, Servier, and Stemline. She has been on the Speaker Board for BMS, Geron, Jazz, Stemline, and Novartis. She has served on the DSMB for Syndax, ASTEX, and Takeda.

Off-label drug use

Hetty E. Carraway: none to disclose.