Momelotinib, a small-molecule Janus kinase 1 and 2 (JAK1/2) and activin A receptor, type 1 (ACVR1)/activin receptor-like kinase 2 (ALK2) inhibitor, represents an important addition to the treatment armamentarium for myelofibrosis (MF), due to its ability to not only address splenomegaly, and constitutional symptoms but also alleviate anemia [1, 2]. Mechanistically, momelotinib-induced inhibition of JAK1/2 -mediated cytokine signaling is primarily responsible for reduction in spleen size and improvement in symptoms, while its erythropoietic activity has been attributed to inhibition of ACVR1-mediated hepcidin production [3]. In preclinical studies conducted in a rat model of anemia of chronic disease, momelotinib-induced inhibition of ACVR1 lowered hepcidin production and resulted in increased mobilization of iron to facilitate erythropoiesis [4]. Similarly, in a phase II study of momelotinib in MF patients with transfusion-dependent anemia (TDA), treatment with momelotinib was associated with marked reduction in serum hepcidin levels, and increased markers of iron availability and erythropoiesis [5]. Over the last decade, early and late phase clinical trials have consistently underlined the anemia benefits of momelotinib [6,7,8,9], which formed the basis for its regulatory approval for MF-related anemia, regardless of prior JAKi exposure. Complementary to clinical trial data, recent real world studies have also confirmed the efficacy of momelotinib. In the current commentary, we appraise data from the use of momelotinib in the real-world setting and offer our insights to inform JAKi selection in patients with MF.

Real-world studies

In the current issue of the journal, Perez-Lamas and colleagues share important findings from the MOMGEFIN study on the real-world use of momelotinib [10]. The study was led by the Spanish Group of Philadelphia-negative Myeloproliferative Neoplasms (GEMFIN) and included 154 patients with JAKi-exposed (n = 118; 99% ruxolitinib) and naïve (n = 36) MF treated with momelotinib 200 mg daily under a compassionate access program in 74 centers in Spain [10]. Major confounders of anemia response in the study were i) lack of JAKi washout in 59% of patients with previous exposure to JAKi therapy, and ii) use of momelotinib in combination with erythropoietin stimulating agents (ESA) and/or danazol in 72 (59%) patients [10].

The study by Perez-Lamas et al. graded anemia response per the 2024 International Working Group-European LeukemiaNet (IWG-ELN) criteria [11], and separately reported responses for TDA and non-TDA and JAKi-exposed vs naïve patients. A total of 122 patients displayed anemia defined as Hgb below 11 g/dl in males and 10 g/dl in females, and 74% were TD ( ≥ 3 units red cell transfusions in the prior 12 weeks) [10]. Among TD patients (n = 90), TI was documented in 27% (at 3 months) and 31% of patients (at 6 months); response rates were numerically higher in JAKi-exposed (n = 71) compared to JAKi-naïve cases (n = 19) with respective 6-month response rates of 32% vs 25% [10]. Among patients with non-TDA (n = 32), major response was achieved in 48% (at 3 months) and 36% (at 6 months); interestingly, major anemia response rates were two-fold higher in JAKi-exposed (n = 24) when compared with JAKi-naïve cases (n = 8) (6-month rates, 58% vs 25%) [10]. However, in subgroup analysis by prior JAKi exposure which took into consideration overall (major + minor) anemia response, response rates were similar at 3 months (60% vs 50%) and 6 months (68% vs 63%) in JAKi-exposed vs naïve patients, respectively [10]. Of note, patients receiving momelotinib in combination with ESA or danazol displayed higher anemia responses compared to momelotinib monotherapy at the initial 3-month assessment (72% vs 50%), while response rates were found to be similar at 6 months (68% vs 71%). Beyond improved erythropoiesis, symptom and spleen response were documented in 92% and 24% of patients, respectively. Common treatment-emergent adverse events included thrombocytopenia (10%, ≥grade 3 6%), diarrhea (12%), infections (9%), hepatotoxicity (6%), and peripheral neuropathy (mainly grade 1–2) (5%) [10].

In a separate real-world study from Germany, entitled MoReLife (Momelotinib in Real-Life), 60 MF patients (n = 33 JAKi-exposed) received momelotinib, of which 52 (87%) displayed Hgb below 10 g/dl, and 38 (63%) patients were considered TD based on treating physician discretion without specified Hgb cut-off levels [12]. No washout for previous JAKi treatment was required. Median treatment duration was 12 weeks, and transfusion-independency (criteria not specified) was achieved in 8 (21%) patients [12]. By contrast, in a UK-Wide study which investigated the real-world efficacy of momelotinib in 85 patients with MF (78% JAKi-exposed with unspecified washout period), anemia response defined by a ≥ 2 g/L increase in Hgb or TI, was achieved in 37% and 44% of patients at 6 weeks and 3 months, respectively [13]. Overall, all three real-world studies had several shortcomings with respect to inclusion of a limited number of JAKi-naïve cases, omission of prior JAKi washout, and non-uniform criteria for anemia response assessment.

Comparison of momelotinib with other JAKi

Treatment with all four JAKi (ruxolitinib, fedratinib, pacritinib, and momelotinib) invariably results in spleen reduction and improvement in symptoms. In general, ruxolitinib and fedratinib exacerbate anemia; in the COMFORT-1 (ruxolitinib vs placebo) [14] and COMFORT-II trials (ruxolitinib vs best available therapy) [15], anemia occurred in 45% (≥ grade 3) and 42% (all grades) of patients receiving ruxolitinib, respectively. Although pacritinib has shown potent ACVR1 inhibition in preclinical studies and the PERSIST-2 trial (pacritinib vs best available therapy), showed TI rate of 37% (at week 24) [16], our personal experience has been underwhelming with hardly any anemia responders among 47 evaluable patients who were treated outside of clinical trials [17]. By contrast, momelotinib has demonstrated consistent erythropoeitic activity in the phase 3 SIMPLIFY 1 and 2 [7, 8], and MOMENTUM trials [9], which was corroborated in recent real-world studies (Table 1). Nonetheless, a comparison of anemia response across studies is challenged by the application of heterogenous response criteria. Also, the majority of prior studies recruited JAKi-exposed patients, with limited data available in JAKi naïve-cases. In this regard, the SIMPLIFY-1 study (momelotinib vs ruxolitinib), which studied JAKi naïve patients, showed higher week 24 TI rates of 67% vs 49% (baseline 68% vs 70%) and lower TDA rates of 30% vs 40% (baseline 25% vs 24%) in the momelotinib vs ruxolitinib arms, respectively [7]. In a phase 2 study of momelotinib which included 41 MF patients with TDA (n = 36 JAKi naïve), TI was achieved in 34% (at week 24) [5]. Additionally, in a retrospective analysis of JAKi-naïve patients receiving momelotinib in a phase 1/2 trial (NCT00935987) (n = 72), anemia response rate based on 2013 International Working Group-Myelofibrosis Research and Treatment (IWG-MRT) criteria [18] was 46% (in non-TD) and 48% (in TD patients) [19].

Table 1 Summary of anemia response in real-world studies and clinical trials with momelotinib for myelofibrosis (MF).
Full size table

Taken together, momelotinib is currently our preferred first-line choice regarding JAKi therapy in MF patients presenting with anemia accompanied by symptomatic splenomegaly and/or constitutional symptoms [1]. However, it should be noted that treatment with momelotinib, unlike the case with ruxolitinib, can be associated with gastrointestinal toxicity, dizziness, peripheral neuropathy, and elevated liver function tests/amylase/lipase [1]. Therefore, in the absence of anemia, we prefer ruxolitinib as the JAKi of choice for controlling splenomegaly or constitutional symptoms [1]. On the other hand, in patients with anemia as the sole clinical manifestation, we recommend anemia-directed therapies such as ESA, danazol, steroids, immunomodulatory agents (thalidomide and lenalidomide), and transforming growth factor beta (TGF-β) ligand traps (luspatercept) rather than JAKi [1]. Of note, we seldom use fedratinib or pacritinib in the front-line setting, and their utility in relapsed/refractory disease is hampered by limited efficacy in patients refractory to adequately-dosed ruxolitinib (i.e., ≥20 mg twice daily) [17, 20].

An important consideration in younger transplant-eligible patients is recognition of the limited long-term survival benefit from JAKi therapy. In this regard, a retrospective comparison of 183 patients with JAKi naïve MF that were enrolled in consecutive phase 1/2 JAKi clinical trials with momelotinib (n = 79), ruxolitinib (n = 50), fedratinib (n = 23) and BMS-911543 (n = 31), showed similar overall survival with momelotinib, ruxolitinib, fedratinib and BMS-911543 (3.5, 4, 4.4, and 5.9 years, respectively) [21]. In the particular study, spleen and anemia responses were independently associated with improved short-term survival, while long-term survival was secured only by allogeneic stem cell transplant (ASCT) (5/10-year survival rate 91%/45% vs 47%/19% in non-transplanted patients) [21]. Accordingly, we advise early rather than later referral for ASCT in younger/fit patients with treatment-requiring intermediate or high-risk MF [1].

Concluding remarks

JAKi therapy significantly improves quality of life in MF; in this regard, momelotinib, as opposed to other FDA-approved JAKis, appears to have erythropoietic activity, in addition to the class-wide efficacy in reducing spleen size and improving constitutional symptoms. Regardless, treatment of MF-associated anemia remains an unmet need that is currently pursued by ongoing clinical trials with drugs that target the TGF-b/BMP/SMAD pathway [22], including TGF-β ligand traps (e.g., luspatercept, elritercept) [23, 24] and hepcidin suppressors, such as anti-hemojuvelin antibody (DISC-0974) [25]. However, none of these drugs are considered to be disease-modifying and should not distract from pursuing ASCT in high or intermediate risk patients.