SOHO State of the Art Updates and Next Questions: Identifying and Treating “Progression” in Myelofibrosis

Abstract

Over the last decade, the Janus kinase (JAK) 1/2 inhibitor ruxolitinib has become widely established as the cornerstone of pharmacologic therapy for most patients with myelofibrosis (MF), providing dramatic and durable benefits in terms of splenomegaly and symptoms, and prolonging survival. Ruxolitinib does not address all aspects of the disease, however; notably cytopenias, and its ability to modify the underlying biology of the disease remains in question. Furthermore, patients eventually lose response to ruxolitinib. Multiple groups have reported the median overall survival of MF patients after ruxolitinib discontinuation to be 13–14 months. While consensus criteria only recognize splenic and blast progression as “progressive disease” in patients with MF, disease progression can occur in a variety of ways. Besides increasing splenomegaly and progression to accelerated phase/leukemic transformation, patients may develop worsening disease-related symptoms, cytopenias, progressive leukocytosis, extra-medullary hematopoiesis, etc. As in the frontline setting, treatment needs to be tailored to the clinical needs of the patient. Current treatment options for patients with MF who fail ruxolitinib remain unsatisfactory, and this continues to represent an area of major unmet medical need. The regulatory approval of fedratinib has introduced an important option in the post-ruxolitinib setting. Fortunately, a plethora of novel agents, both new JAK inhibitors and drugs from other classes, e.g., bromodomain and extra-terminal (BET), murine double minute 2 (MDM2) and telomerase inhibitors, activin receptor ligand traps, BH3-mimetics and more, are poised to greatly expand the therapeutic armamentarium for patients with MF if successful in pivotal trials.

Introduction

The International Working Group for Myeloproliferative Neoplasms (MPN) Research and Treatment (IWG-MRT) criteria (2013 revision)¹ define “progressive disease” in patients with myelofibrosis (MF) as defined in Table 1. These criteria only recognize and address new or worsening splenomegaly and leukemic transformation (LT) as modes of disease progression, although “relapse” is defined separately and more broadly, as shown. Disease progression may take the form of worsening anemia and/or thrombocytopenia, progressive MPN symptoms or leukocytosis, extra-medullary hematopoiesis compromising organ function or causing pain, etc., in addition to the more readily measurable splenic or blast progression. It follows, therefore, that progressive disease in MF needs to be treated according to the patient’s clinical presentation. Allogeneic hematopoietic cell transplantation (allo-HCT) remains the only known cure for MF and should be considered in all eligible patients, although it is best performed at the time of optimal response to Janus kinase (JAK) inhibitor therapy.² Albeit rarely needed today, splenectomy can occasionally be very helpful in ameliorating disabling symptoms or severe cytopenias.³

Table 1.

Definitions of progressive disease and relapse in MF per IWG-MRT 2013 criteria.¹

Progressive disease	Appearance of a new splenomegaly that is palpable at least 5 cm below the LCM or
	A ≥100% increase in palpable distance, below LCM, for baseline splenomegaly of 5–10 cm or
	A 50% increase in palpable distance, below LCM, for baseline splenomegaly of >10 cm or
	Leukemic transformation confirmed by a bone marrow blast count of ≥20% or
	A peripheral blood blast content of ≥20% associated with an absolute blast count of ≥1 × 109/L that lasts for at least 2 weeks
Relapse	No longer meeting criteria for at least CI after achieving CR, PR, or CI, or
	Loss of anemia response persisting for at least 1 month or
	Loss of spleen response persisting for at least 1 month

Abbreviations: MF, myelofibrosis; IWG-MRT, International Working Group for Myeloproliferative Neoplasms Research and Treatment; LCM, left costal margin; CR, complete response; PR, partial response; CI, clinical improvement.

The JAK inhibitor ruxolitinib ushered in the era of targeted therapy in MF and has revolutionized the care of patients with MF over the last decade. Best known for its robust and sustained benefits in terms of reducing symptoms and spleen size through its anti-inflammatory (anti-cytokine) and anti- proliferative actions, ruxolitinib has also been shown to improve survival of patients with intermediate-2/high risk MF.⁴ Furthermore, there is mounting evidence that the benefits of the drug could be greatest when used earlier in the disease.^5,6 However, patients eventually lose response to ruxolitinib (median duration of spleen response in the pivotal COMFORT trials was about 3 years^7,8 and patients with ≥3 non-driver mutations have a shorter time to ruxolitinib discontinuation⁹), whether due to the phenomenon of JAK2 inhibitor “persistence”¹⁰ or other mechanisms, and prognosis after ruxolitinib discontinuation is poor (median survival 13–14 months).^11–13 Thus, “progression” today in MF mostly means disease progression on ruxolitinib, also referred to as “ruxolitinib failure”,^14–16 although worsening anemia, for example, in a patient without significant MPN symptoms or splenomegaly receiving anemia-directed therapy would also certainly constitute disease progression. In this article, we focus on novel therapies being developed in the post-ruxolitinib setting for resistant/intolerant patients or as rational “add-ons” to ruxolitinib.

Fedratinib

Fedratinib is a JAK2 inhibitor approved in the US in 2019 for the treatment of intermediate-2/high risk MF and now also authorized in the European Union for patients with MF-related splenomegaly and/or symptoms. In the placebo-controlled JAKARTA trial¹⁷ conducted in JAK inhibitor-naïve patients with baseline platelets ≥50 × 10⁹/L, fedratinib, 400 mg daily, exhibited comparable efficacy in terms of ≥35% spleen volume reduction (SVR35) and ≥50% reduction in total symptom score (TSS50) at 24 weeks to that observed with ruxolitinib in the COMFORT-1 trial;¹⁸ rates of anemia and thrombocytopenia were similar as well. JAKARTA-2 was a non-randomized, single-arm trial in 97 ruxolitinib-exposed patients with MF (median duration of prior ruxolitinib therapy, 10.25 months) and platelets ≥50 × 10⁹/L in which fedratinib, 400 mg/d, produced an SVR35 rate of 31% and a TSS50 rate of 27% at 24 weeks on intention-to-treat (ITT) analysis.¹⁹ Re-analysis of these data using stringent criteria (Table 2) to define ruxolitinib failure (79 patients met this definition: 65 resistant and 14 intolerant) yielded virtually identical results: a 30% SVR35 rate and a 27% TSS50 rate at 24 weeks.²⁰ The median duration of prior ruxolitinib therapy was 10.7 months. A pooled analysis of data on patients with baseline platelets in the 50–99 × 10⁹/L range from JAKARTA (n=14) and JAKARTA-2 (n=33) showed preserved efficacy of fedratinib in this subset of patients (as compared to those with baseline platelets ≥100 × 10⁹/L), supporting 400 mg daily as the starting dose across baseline platelet counts, as long as ≥50 × 10⁹/L.²¹ Fedratinib does cause gastrointestinal toxicity, possibly due to its fms-like tyrosine kinase 3 (FLT3) inhibitory action; prophylactic anti-emetic therapy should be considered and diarrhea promptly treated. Although very rare,²² Wernicke’s encephalopathy (WE) is the subject of a black box warning on the US label for fedratinib: thiamine levels must be checked prior to fedratinib initiation and periodically during treatment, and any deficiency corrected before beginning fedratinib. Long-term follow-up of participants on the JAKARTA and JAKARTA-2 trials is not available as all patients had to come off owing to a full clinical hold being placed by the Food and Drug Administration (FDA) on the fedratinib development program over WE-related concerns that was lifted after careful review of the putative WE cases showed only one confirmed (in a severely malnourished patient) and two likely cases of WE (both had sustained nausea and vomiting preceding the development of neurodeficits, which resolved with thiamine repletion despite continuing fedratinib) among 670 patients receiving fedratinib across the clinical development program.²² The FREEDOM studies (NCT03755518, NCT03952039) are now underway to collect data on fedratinib in the second-line setting, and use the stringent criteria outlined in Table 2 to define ruxolitinib failure.

Table 2.

Stringent criteria for ruxolitinib (RUX) failure used in the reanalysis of JAKARTA-2²⁰ and in PAC203.³⁷

Relapsed	RUX ≥ 3 mos with regrowth (defined as < 10% SVR or < 30% decrease in spleen size from baseline following an initial response)
Refractory	RUX ≥ 3 mos with < 10% SVR or < 30% decrease in spleen size from baseline
Intolerant	RUX ≥ 28 days complicated by development of RBC transfusion requirement (≥ 2 units/mos for 2 mos); or grade ≥ 3 thrombocytopenia, anemia, hematoma/hemorrhage while on RUX

Abbreviations: RUX, ruxolitinib; SVR, spleen volume reduction; RBC, red blood cell; mos, months.

Momelotinib

Momelotinib is a JAK1/2 inhibitor that also improves anemia in patients with MF,^23,24 possibly via inhibition of the type 1 activin receptor (ACVR1, also known as ALK2) and suppression of hepcidin production by the liver.^25,26 Hepcidin levels are known to be elevated in patients with MF and correlate with worse outcomes.²⁷ Momelotinib was non-inferior to ruxolitinib in terms of SVR35 at 24 weeks in a head to head trial (SIMPLIFY-1) in 432 JAK inhibitor-naïve patients with intermediate/high risk MF and baseline platelets ≥50 × 10⁹/L, but inferior in terms of TSS50.²⁸ SIMPLIFY-2 randomized 156 ruxolitinib-pre-treated patients with intermediate/high risk MF 2:1 to receive momelotinib, 200 mg daily, or best available therapy (BAT).²⁹ Importantly, there was no minimum platelet count required for eligibility. Ruxolitinib was allowed as BAT, and 89% of patients on the BAT arm received ruxolitinib. Patients had to have received at least 28 days of prior ruxolitinib and either have required RBC transfusion on ruxolitinib or had their ruxolitinib dose adjusted to less than 20 mg twice daily because of grade ≥3 anemia, thrombocytopenia or bleeding. Seventy two percent of momelotinib patients and 64% of BAT patients had received ≥12 weeks of prior ruxolitinib. This trial did not meet its primary endpoint of SVR35 at 24 weeks: 7% for momelotinib and 6% for BAT (p = 0.9), but 26% of patients in the momelotinib group achieved TSS50 from baseline to week 24, as opposed to 6% in the BAT group (nominal p = 0.0006). In both trials, anemia-related endpoints, e.g., RBC transfusion rate through week 24, transfusion independence (TI) and transfusion dependence (TD) rates at week 24, favored momelotinib, but statistical significance could not be claimed because of the sequential testing procedures adopted in the statistical design. For example, in SIMPLIFY-2, the TI rate at week 24 was 43% in the momelotinib group and 21% in the BAT group (nominal p = 0.0012). In SIMPLIFY-1, these proportions were 66.5% for momelotinib and 49.3% for ruxolitinib (nominal p < 0.001), and 30.2% of momelotinib patients were transfusion-dependent at week 24, compared with 40.1% of ruxolitinib patients (nominal p = 0.019). Momelotinib, 200 mg/d, is now being compared 2:1 against danazol, 600 mg/d, in a pivotal phase 3 trial (MOMENTUM, NCT04173494)³⁰ in 180 anemic (hemoglobin < 10 g/dL) and symptomatic (TSS ≥10), JAK inhibitor-pre-treated patients with MF and baseline platelets ≥25 × 10⁹/L, with the primary endpoint being TSS50 at week 24 as assessed by the myelofibrosis symptom assessment form (MFSAF),³¹ version 4.0. SVR35 and TI rate at week 24 constitute key secondary endpoints.

Pacritinib

Like fedratinib, pacritinib is an inhibitor of JAK2 and FLT3, and also inhibits interleukin-1 receptor-associated kinase 1 (IRAK1) and colony stimulating factor 1 receptor (CSF1R).³² Pacritinib is relatively non-myelosuppressive, and the early phase trials of this agent in patients with MF enrolled patients with any degree of cytopenia.^33,34 Pacritinib beat BAT (not including JAK inhibitors) in 327 JAK inhibitor-naïve patients with higher risk MF and any degree of anemia or thrombocytopenia in the PERSIST-1 trial in terms of both SVR35 (at week 24, primary endpoint) and TSS50 (at week 48 by ITT analysis and at week 24 by per protocol analysis).³⁵ PERSIST-2 compared 2 doses of pacritinib, 200 mg twice daily and 400 mg once daily, against BAT (ruxolitinib in 45% of the patients) in 311 patients with intermediate/high risk MF and platelet counts ≤100 × 10⁹/L.³⁶ This trial was impacted by the placement of a full clinical hold on the pacritinib development program by the FDA owing to concerns over excess mortality from cardiovascular events and bleeding, and the ITT efficacy population included 75, 74, and 72 patients randomized to pacritinib once daily, twice daily, and BAT, respectively. Prior JAK2 inhibitor therapy was allowed, and 44%, 45% and 47% of patients, respectively, in the 3 arms had received the same, the vast majority ruxolitinib. Pacritinib (arms combined) was superior to BAT for SVR35 at week 24 (18% versus 3%, p = 0.001) but not for the co-primary endpoint of TSS50 at week 24 (25% versus 14%, p = 0.08). The 200 mg twice daily dose, however, was superior to BAT for both SVR35 (22% versus 3%, p = 0.001) and TSS50 (32% versus 14%, p = 0.01) at week 24. After the lifting of the clinical hold, a randomized, phase 2, dose-ranging safety study, PAC203, incorporating a number of risk mitigation measures, was conducted.³⁷ One hundred and sixty one patients with advanced MF resistant to or intolerant of ruxolitinib per the criteria set forth in Table 2 (50% met both criteria) were randomized to receive 100 mg daily, 100 mg twice daily or 200 mg twice daily of pacritnib. Efficacy was modest across the 3 doses, but highest for the 200 mg twice daily dose (9.3% for SVR35 and 7.4% for TSS50 through week 24). No excess incidence of grade ≥3 hemorrhagic or cardiac events was observed at this dose. Interestingly, among 24 patients with severe thrombocytopenia (baseline platelets <50 × 10⁹/L) receiving 200 mg twice daily of pacritinib, the 24-week SVR35 rate was 17%. It has been speculated that pacritinib, perhaps because of its IRAK1 inhibitory action, may be more efficacious against the “myelodepletive” phenotype of MF, characterized by smaller spleens, lower blood counts and lower mutant JAK2 allele burdens.^38,39 The latter has long been known to be associated with worse overall survival (OS) and leukemia-free survival (LFS) in primary MF (PMF).^40,41 In contrast, one study found ruxolitinib to be more efficacious in the presence of a mutant JAK2 allele burden ≥50%.⁴² Pacritinib, 200 mg twice daily, is now being studied against BAT (low dose ruxolitinib, steroids, androgens, hydroxyurea) in a pivotal phase 3 trial (PACIFICA, NCT03165734) in intermediate/high risk MF patients with <50 × 10⁹/L platelets at baseline and no or a limited duration of prior JAK inhibitor therapy. The primary endpoint is SVR35 from baseline to week 24, while TSS50 and OS are secondary endpoints; crossover is not permitted.

Imetelstat

While a number of non-JAK inhibitor agents representing diverse drug classes have displayed clinical activity in the setting of ruxolitinib failure, the suggestion of a robust survival benefit in the post-ruxolitinib setting for the telomerase inhibitor imetelstat in the IMBARK^™ trial (albeit cross-study comparisons between this trial and other datasets, i.e., historical controls) has generated substantial enthusiasm, and a pivotal phase 3 trial (IMpactMF, NCT04576156) of this agent compared to BAT (excluding JAK inhibitors) with OS as the primary endpoint is now underway.

Imetelstat was first studied in a small, pilot study at the Mayo Clinic, and led to complete and partial remissions in 7 of 33 patients (21%), unusual in MF clinical studies.⁴³ The multi-center IMBARK^™ trial then evaluated both 4.7 mg/kg (n=48) and 9.4 mg/kg (n=59) of imetelstat, administered intravenously every 3 weeks, in 107 patients with MF who had failed JAK inhibitor therapy (Table 3). The lower dose arm was subsequently closed due to insufficient activity, and the patients on active treatment allowed to cross over to the higher dose arm. While the rate of SVR35 at 24 weeks in the 9.4 mg/kg arm was a modest 10.2% and the rate of TSS50 was a reasonable 32.2%, median OS was a striking 28.1 months,⁴⁴ although of note, median OS for patients who stopped ruxolitinib while still in chronic phase (n = 167) was 27.5 months in an Italian “real world” study.¹² Furthermore, greater than 40% of patients experienced ≥1 grade improvements in bone marrow fibrosis and ≥25% decreases in driver mutation variant allele frequency (VAF). Reduction in bone marrow fibrosis significantly correlated with OS improvement, while spleen and symptom responders showed trends towards survival benefit compared with non-responders. Additionally, ≥20% VAF reduction correlated with improved clinical benefits, including higher rates of spleen and symptom responses, bone marrow fibrosis improvement and longer OS.⁴⁵ Intriguingly, imetelstat appeared to particularly benefit patients with “triple negative” disease, a known poor-prognosis subgroup. In these patients at the 9.4 mg/kg dose, median OS was 35.9 months, the 24-week rate of SVR35 was 18.8% and that of TSS50 was 50%.⁴⁶ Shorter telomere length at baseline and higher baseline expression levels of human telomerase reverse transcriptase (hTERT) predicted for improved clinical outcomes with imetelstat at the 9.4 mg/kg dose, and these characteristics were enriched for among the patients with triple negative disease. Finally, the pharmacodynamic effects of imetelstat (decrease in telomerase activity and/or hTERT expression level) were dose- and exposure-dependent, and ≥50% decreases in one or both of these parameters correlated with better clinical response rates and longer OS.⁴⁷

Table 3.

Major inclusion criteria for the IMBARK^™ trial of imetelstat in MF.⁴⁴

Patients with Intermediate-2 or High-risk MF (Int-2/High-risk) patients who have relapsed after or are refractory to prior treatment with a janus kinase (JAK) inhibitor Relapsed or refractory to JAKi defined as documented progressive disease during or after JAKi: Patients must have worsening of splenomegaly-related abdominal pain at any time after the start of JAKi therapy and EITHER: No reduction in spleen volume or size after 12 weeks of JAKi therapy, OR Worsening splenomegaly at any time after the start of JAKi therapy documented by: Increase in spleen volume from nadir by 25% measured by MRI or CT, or Increase in spleen size by palpation

Abbreviations: MF, myelofibrosis; MRI, magnetic resonance imaging; CT, computed tomography.

Pelabresib

Pelabresib (formerly CPI-0610) is an orally administered, small-molecule bromodomain and extra-terminal (BET) inhibitor that has demonstrated promising clinical activity in the phase 2 MANIFEST trial (discussed below) and is now being studied in combination with ruxolitinib in the phase 3, placebo-controlled MANIFEST-2 trial in the JAK inhibitor-naïve setting in patients with intermediate/high risk MF (NCT04603495).

While combined JAK and BET inhibition has been shown to be synergistic in MPN models, both in vitro and in vivo,⁴⁸ and highly encouraging results of the combination of ruxolitinib and pelabresib in the JAK inhibitor-naïve setting (arm 3 of the MANIFEST study)⁴⁹ set the stage for its developmental path forward, pelabresib can be a very useful single agent after ruxolitinib discontinuation, as well as an “added on” partner for ruxolitinib in patients having a suboptimal response to ruxolitinib monotherapy. Pelabresib, at a starting dose of 125 mg daily for 2 weeks out of a 3-week cycle, was studied as a single agent in patients with intermediate/high risk MF who were no longer on ruxolitinib due to resistance/intolerance or were not candidates for it in arm 1 of the MANIFEST study.⁵⁰ Either RBC TD or a baseline spleen volume of ≥450 cm³ (for non-TD patients) and a minimum baseline platelet count of 75 × 10⁹/L were required. Of 46 patients enrolled, 19 were RBC transfusion-dependent and 27 not so. In the transfusion-dependent cohort (arm 1A), 21% of evaluable patients achieved TI. In the non-TD cohort (arm 1B), 30% of evaluable patients achieved SVR35 at week 24, while only 8% did in the TD cohort. Symptom improvement was also particularly pronounced in the non-TD cohort, with 48% of evaluable patients achieving TSS50 at 24 weeks, as opposed to 8% in the TD cohort. Importantly, 50% of evaluable patients in the non-TD cohort experienced a ≥1.5 g/dL improvement in hemoglobin level. Common (≥20%) treatment-emergent adverse events (TEAEs) included thrombocytopenia, nausea, diarrhea, taste changes, asthenia, respiratory tract infections, cough, constipation and weight loss. Promising clinical activity was also observed in arm 2 of the MANIFEST study,⁵¹ in which pelabresib was “added on” in patients who had been on ruxolitinib for at least 6 months with a stable dose for the preceding ≥8 weeks and were either transfusion-dependent or had a spleen volume ≥450 cm³. Fifty two TD patients and 26 non-TD patients were enrolled to arms 2A and 2B, respectively. Among evaluable patients, the rate of conversion to RBC TI for TD patients was 36%, and the rate of SVR35 at week 24 for non-TD patients was 29% (21% for TD patients). TSS50 at week 24 was achieved by 46% and 38% of evaluable TD and non-TD patients, respectively. The profile of TEAEs was similar to that in the monotherapy arm.

Luspatercept

Anemia eventually develops in all patients with MF and, as noted above, worsening anemia can be a manifestation of disease progression or relapse, whether or not accompanied by worsening in other aspects of the disease. Treatment of anemia in MF remains unsatisfactory and constitutes an area of unmet medical need. Anemia is not addressed by either of the two currently available JAK inhibitors and, while not adverse prognostically,^52,53 ruxolitinib-induced anemia is a significant problem in clinical practice and a frequent cause of ruxolitinib discontinuation.¹³ Proper management of anemia is essential for maintaining JAK inhibitor dose intensity, so as not to deprive patients of the undeniable benefits of these drugs.

Luspatercept is an activin receptor type IIB ligand trap currently licensed for the treatment of anemia in beta-thalassemia and lower risk myelodysplastic syndrome (MDS) with ring sideroblasts.⁵⁴ Luspatercept is administered by subcutaneous injection every 3 weeks and is believed to work by sequestering stromal ligands belonging to the transforming growth factor beta (TGF-β) superfamily, thus preventing their interaction with the activin receptor and consequent activation of downstream Smad signaling.⁵⁵ Ultimately, this removes the inhibitory effects of these ligands on erythroid maturation, although the precise target ligand is not known with certainty.⁵⁶ Luspatercept was studied in an open-label, phase 2 trial (n = 79) in patients with MF and anemia; patients were divided into 4 cohorts based on whether or not they were RBC transfusion-dependent and were receiving ruxolitinib (stable dose for ≥16 weeks).⁵⁷ Although clinical activity (TI or a sustained ≥1.5 g/dL hemoglobin improvement in non-TD patients) was observed in all four cohorts, the highest response rate was seen in the cohort of patients receiving ruxolitinib and RBC transfusion-dependent (n = 22), 36% of whom achieved TI, with a median cumulative duration of 55 weeks.⁵⁸ Accordingly, luspatercept is now being studied in a placebo-controlled phase 3 trial (INDEPENDENCE, NCT04717414) in patients with MF on JAK2 inhibitor therapy (for at least 32 weeks, with a stable dose for the last ≥16 weeks) who have required 4–12 RBC transfusions in the 12 weeks immediately preceding randomization.

KRT-232

JAK2 V617F leads to overexpression of murine double minute 2 (MDM2, the physiologic negative regulator of p53) in MPN cells.⁵⁹ Furthermore, TGF-β, implicated in the pathogenesis of MF on multiple levels, induces MDM2 expression.⁶⁰ Finally, TP53 mutations are rare in chronic phase MF.^61–63 Taken together, these observations support pharmacologic inhibition of MDM2 as a therapeutic strategy in MF as a means to activate wild type p53 and induce apoptosis of the neoplastic cells.

KRT-232 (formerly AMG-232) is an orally administered MDM2 inhibitor that is being studied in multiple different settings in patients with MF. In the first part of a phase 2 study in patients whose disease had relapsed after or was refractory to JAK inhibitor therapy, a dose of 240 mg of KRT-232, administered daily for 7 days every 28 days, emerged as a safe and effective dose/schedule to be taken forward into phase 3 testing, now underway in the second part of the study (NCT03662126) in comparison to BAT (excluding JAK inhibitors). In the first part of the study, four of 25 (16%) evaluable patients receiving KRT-232 at this dose and schedule obtained a best spleen response of SVR35.⁶⁴ The best rate of TSS50 at this dose and schedule (27 evaluable) was 30%. Timing of the baseline spleen imaging (while patients were still on ruxolitinib) and lack of a ruxolitinib washout might have contributed to the rather modest SVR35 rate because of an undetected “flare” that may have occurred after ruxolitinib discontinuation. KRT-232 is also being studied as an “add on” to ruxolitinib in patients with a suboptimal response after at least 18 weeks (stable dose for ≥8 weeks) of ruxolitinib monotherapy (NCT04485260), in combination with TL-895, an investigational Bruton’s tyrosine kinase inhibitor, in patients with JAK inhibitor-relapsed/refractory MF (NCT04640532), and alone in MF patients who are intolerant of JAK inhibitors (NCT04640532). A recent report demonstrating the transient expansion of mutant TP53 clones during therapy with another MDM2 inhibitor, idasanutlin, in patients with PV is of some concern; this clearly warrants vigilance in the KRT-232 studies.⁶⁵

PRM-151

PRM-151 is recombinant pentraxin-2 (serum amyloid protein), an endogenous substance that localizes to sites of tissue injury and promotes repair.⁶⁶ PRM-151 is being developed as an anti-fibrotic agent and promising results have been reported in patients with idiopathic pulmonary fibrosis.⁶⁷ In the first stage of a two-stage trial in patients with MF, 18 of 27 patients derived durable benefits, staying on study for a median of 30.9 months.⁶⁸ Nine of these patients received PRM-151 alone, and nine in combination with ruxolitinib. Spleen and symptom improvements were observed both in patients receiving PRM-151 monotherapy and in those receiving it in combination with ruxolitinib. Mean hemoglobin levels and platelet counts improved over time and generally correlated with reductions in reticulin and collagen grade.

Three doses of PRM-151, 0.3, 3, and 10 mg/kg were tested in randomized fashion in the second stage of the study in 97 patients with intermediate or high risk MF, grade 2 or 3 bone marrow fibrosis and anemia and/or thrombocytopenia who had previously received or were ineligible for ruxolitinib.⁶⁹ The primary endpoint was reduction in bone marrow fibrosis by ≥1 grade at any time during the study. PRM-151 was administered intravenously once every 28 days for 9 cycles after administration on days 1, 3 and 5 in cycle 1 only. There were a substantial number of discontinuations, with only 51 patients completing 9 cycles. The overall response rate (ORR) in terms of bone marrow fibrosis grade decrease was 27.8%, with no significant differences between the 3 dosing cohorts. Five of 31 (16%) and 6 of 13 (46%) patients achieved TI for RBC and platelets, respectively. Thirty four percent of patients achieved TSS50 at any time; the rate of SVR35 was not reported. Details of the development plans going forward for PRM-151 in MF are awaited.

Navitoclax and parsaclisib

Although studied in limited numbers of MF patients to date and only in “add on” fashion in the setting of suboptimal responses to ruxolitinib, multiple phase 3 trials have been announced or are underway for the B-cell lymphoma-2/-extra long (Bcl-2/-xL) antagonist navitoclax and the phosphatidylinositol-3-kinase delta isoform (PI3Kδ) inhibitor parsaclisib, all in combination with ruxolitinib (NCT04472598, NCT04468984, NCT04551066, NCT04551053). Both combinations are supported by preclinical findings of synergism.^70,71

Navitoclax was studied in 34 patients with MF who were deemed to have had a suboptimal response to ruxolitinib after ≥12 weeks of therapy (median duration of prior ruxolitinib was 20 months).⁷² On-target thrombocytopenia from Bcl-xL inhibition was managed by close monitoring of the platelet count and dose adjustment, with 68% of patients being able to reach the target 300 mg/d dose. At week 24, the rate of SVR35 was 27%, and six of 20 evaluable patients (30%) had achieved TSS50.

Parsaclisib was studied in 53 patients with MF; suboptimal response to ruxolitinib was defined as in Table 4.⁷³ Thirty three patients received parsaclisib weekly after an initial 8-week period of daily dosing, and 20 patients received continuous daily dosing of parsaclisib. Only 1 patient in each dosing cohort (out of 30 and 14 evaluable, respectively) attained SVR35 at week 24. In terms of symptoms, out of 27 and 12 evaluable patients in the daily/weekly and all daily dosing cohorts, only 1 and 2 patients, respectively, achieved TSS50 at 24 weeks. Parsaclisib was well-tolerated, with no colitis or pneumonitis reported, and low incidences of rashes, diarrhea and transaminitis. Interestingly, continuous daily dosing was associated with a better safety profile than daily followed by weekly parsaclisib dosing.

Table 4.

Definition of suboptimal response to ruxolitinib used in the parsaclisib trial.⁷³

•Treated with ruxolitinib for ≥6 months with stable dose for ≥8 weeks immediately prior to enrollment

AND

•Palpable spleen >10 cm below left subcostal margin on physical examination at screening OR

Palpable spleen 5–10 cm below left subcostal margin on physical examination AND active symptoms of MF at the screening visit defined as 1 symptom score ≥5 or 2 symptom scores ≥3 each, using the Screening Symptom Form*

Abbreviations: MF, myelofibrosis.

^*Screening Symptom Form: 10-point scale for each of the following 7 symptoms: night sweats, pruritus, abdominal discomfort, pain under left ribs, early satiety, bone/muscle pain, inactivity.

Other agents

A multitude of other new drugs, some JAK inhibitors and others not, have been studied in patients with MF in recent years, predominantly in the second line setting after ruxolitinib, often also including patients not eligible for ruxolitinib, e.g., due to cytopenias, lack of splenomegaly/symptoms or both. Some of these have been investigator-initiated trials and others industry-sponsored studies. Overall, results with these agents have been modest, and their continued development for MF is less certain. Examples include the lysine-specific demethylase-1 (LSD1) inhibitor bomedemstat,⁷⁴ the JAK1 inhibitor itacitinib,⁷⁵ the JAK2 inhibitor NS-018,^76,77 the second mitochondrial activator of caspases (Smac)-mimetic LCL-161,⁷⁸ the aurora kinase A inhibitor alisertib,⁷⁹ and the CD123-targeted antibody-drug conjugate (ADC) tagraxofusp.⁸⁰ The results of clinical trials of these agents that have been published or presented are summarized in Table 5. Others, such as the anti-programmed death 1 (PD1) antibody pembrolizumab,⁸¹ the anti-lysyl oxidase like 2 (LOXL2) antibody simtuzumab,⁸² the anti-TGF-β antibody fresolimumab⁸³ and the heat shock protein 90 inhibitor AUY922⁸⁴ either demonstrated no clinical activity or had to be terminated due to toxicity or other reasons. Results are awaited from ongoing trials of numerous novel agents being studied either as monotherapy or in combination with JAK inhibitors – a partial list is provided in Table 6.

Table 5.

Selected single agents studied in recent clinical trials in patients with MF.

Agent (class)	Phase of study, population studied	Route of administration, main efficacy results	Safety concerns, if any, other comments	Reference
Bomedemstat (LSD1 inhibitor)	2, int-2/high risk, R/R to, intolerant of, or ineligible for rux/fed, n = 49	PO daily, SVR35 rate 14% and TSS50 rate 25% at wk 12	Dysgeusia common (35%), dose titrated to target platelet range in each patient	74
Itacitinib (JAK1 inhibitor)	2, int/high risk, previously treated or treatment-naïve, n = 87	PO daily or BID, TSS50 at wk 12 (primary endpoint) 35.7% for 200 mg bid and 32.3% for 600 mg qd; 28.6% and 35.5% at wk 24	SVR35 at wk 24 rate 16.7% across all 3 doses but 100 mg bid cohort not expanded as only 20% met primary endpoint	75
NS-018 (JAK2 inhibitor)	1/2, int/high risk, n = 48 in phase 1, 23 treated previously with JAKi, n = 29 in phase 2, all with prior JAKi treatment, 62% intolerant, 38% R/R	PO daily or BID, 300 mg qd chosen as RP2D, 36% spleen CI (by IWG-MRT) in phase 1, 4 anemia CI and 1 platelet CI, SVR35 rate at wk 24 12% in phase 2	Anemia (21%), thrombocytopenia (17%), nausea (14%)	77
LCL-161 (Smac-mimetic)	2, int/high risk, R/R to, intolerant of, or not candidates for JAKi, n = 50	PO weekly, ORR 30% (by IWG-MRT) comprising CI-symptoms, anemia, spleen, CCyR	Fatigue most common cause of dose reduction, ameliorated by dex pre-treatment	78
Alisertib (AURKA inhibitor)	1, int/high risk, R/R to, intolerant of, or unlikely to benefit from rux, n = 24	PO BID on d1–7 q21d, response rates in evaluable patients: spleen 29% (by palpation), anemia 11% (TI), symptoms 32% (TSS50)	Restoration of GATA1 staining and improvement in megakaryocyte morphology observed	79
Tagraxofusp (CD123-directed ADC)	1/2, R/R to or intolerant of JAKi, no minimum platelet count, n = 35	IV on d1–3 q21d (C1–4), q28d thereafter; 60% had SD, 2 spleen and 3 symptom responders (by IWG-MRT)	Capillary leak syndrome a unique toxicity of this drug	80

Abbreviations: LSD1, lysine-specific demethylase-1; JAK, Janus kinase; Smac, second mitochondrial activator of caspases; AURKA, aurora kinase A; ADC, antibody-drug conjugate; int, intermediate; R/R, relapsed/refractory; SVR35, ≥35% spleen volume reduction; TSS50, ≥50% reduction in total symptom score; IWG-MRT, International Working Group for Myeloproliferative Neoplasms Research and Treatment; JAKi, JAK inhibitor; rux, ruxolitinib; fed, fedratinib; qd, once daily; bid/BID, twice daily; CI, clinical improvement; PO, oral; IV, intravenous; CCyR, complete cytogenetic response; TI, transfusion independence; C, cycle; d, day; wk, week, n, number of patients; ORR, overall response rate.

Table 6.

Selected novel agents currently in early phase clinical trials for MF.

Agent	Drug class	Clinicaltrials.gov identifier
AVID200	TGF-β1/3 trap	NCT03895112
PU-H71 (add on to rux)	HSP90 inhibitor	NCT03935555
Pevonedistat (add on to rux)	NAE inhibitor	NCT03386214
PRT-543	PRMT5 inhibitor	NCT03886831
TL-895	BTK inhibitor	NCT04655118
Mivebresib (alone or with rux or nav)	BET inhibitor	NCT04480086
ABBV-744 (alone or with rux or nav)	BET inhibitor	NCT04454658
Pelcitoclax	Bcl-2/-xL inhibitor	NCT04354727
Elotuzumab	Anti-SLAMF7 monoclonal antibody	NCT04517851
TP-3654	PIM kinase inhibitor	NCT04176198
INCB053914 (alone or with rux)	PIM kinase inhibitor	NCT02587598
INCB000928 (alone or with rux)	ACVR1/ALK2 inhibitor (for anemia)	NCT04455841
Siremadlin (add on to rux)	MDM2 inhibitor	NCT04097821
Crizanlizumab (add on to rux)	Anti-P-selectin monoclonal antibody	NCT04097821
Sabatolimab (add on to rux)	Anti-TIM3 monoclonal antibody	NCT04097821
LTT462 (add on to rux)	ERK1/2 inhibitor	NCT04097821
NIS793 (add on to rux)	Anti-TGF-β monoclonal antibody	NCT04097821
PXS-5505	LOX inhibitor	NCT04676529

Abbreviations: MF, myelofibrosis; rux, ruxolitinib; nav, navitoclax; TGF-β, transforming growth factor beta; HSP90, heat shock protein 90; NAE, NEDD8 activating enzyme; PRMT5, protein arginine methyltransferase 5; BTK, Bruton’s tyrosine kinase; BET, bromodomain and extra-terminal; Bcl-2, B-cell lymphoma-2; Bcl-xL, B-cell lymphoma extra long; ACVR1, activin receptor type 1; MDM2, murine double minute 2; ERK, extracellular signal-regulated kinase; TIM3, T-cell immunoglobulin and mucin domain-containing protein 3; PIM, proviral integration site for Moloney murine leukemia virus; SLAMF7, signaling lymphocyte activation molecule F7; LOX, lysyl oxidase.

Conclusion

Drug development for MF is now an extremely active area of both preclinical and clinical investigation. Besides the many agents discussed or mentioned in this article, there is also interest in bringing back established drugs in conjunction with ruxolitinib to deepen responses and counteract cytopenias; examples include low dose thalidomide⁸⁵ and interferon.^86,87 Disease progression on ruxolitinib is probably the greatest area of unmet medical need in MF and is, appropriately, the focus of most early drug development efforts, although intervening earlier in the disease process clearly has merit and continues to be explored in combination studies. In the coming years, it is likely that multiple new agents will gain regulatory approval and significantly expand the array of therapeutic options for our patients.