Assessing efficacy in myelofibrosis treatment: a focus on JAK inhibition

Abstract

Myelofibrosis (MF) is characterized by splenomegaly, anemia and a debilitating symptom burden (e.g., fatigue, night sweats, pruritus, bone and muscle pain, undesired weight loss). Moreover, these symptoms impair activities of daily living and quality of life. Until recently, there have been no approved therapies for MF, and management of MF has been predominantly palliative. Dysregulated JAK-STAT signaling is associated with the pathologic MF disease state. A novel class of therapies, the JAK inhibitors, offers the potential to abrogate this pathologic signaling pathway. In clinical trials of patients with intermediate- and high-risk MF, JAK inhibitors have demonstrated efficacy in reducing splenomegaly and MF-associated symptoms. Evidence from ruxolitinib trials also suggests that JAK inhibitors may improve survival outcomes.

Keywords::

• CYT387
• JAK
• JAK-STAT
• Janus kinase
• LY2784544
• myelofibrosis
• pacritinib
• quality of life
• ruxolitinib
• SAR302503

Figure 1. The JAK-STAT signaling pathway.

Cytokines and growth factors bind to the extracellular receptor and induce a series of intracellular changes: (1) JAK proteins associate with the intracellular portion of the receptor and are phosphorylated; (2) phosphorylated JAK proteins in turn phosphorylate STAT proteins, which then phosphorylate other downstream proteins and (3) phosphorylated STAT translocates into the nucleus and promotes the transcription of genes. Other proteins also regulate this signaling pathway, such as SOCS1, which prevents activation of this pathway.

Reproduced with permission from [41].

Myelofibrosis (MF) is a BCR-ABL-negative myeloproliferative neoplasm (MPN), that includes de novo disease, or primary MF (PMF), and MF secondary to polycythemia vera (post-PV MF) or essential thrombocythemia (post-ET MF [1,2]). The hallmark of these diseases is an increase in mature blood cells, which arise from clonal expansion driven by genetic mutations in pluripotent cells of the bone marrow compartment. The polyclonal response to this myeloproliferative process eventually results in bone marrow fibrosis. MF is typically associated with progressive splenomegaly [2], which can lead to other complications such as portal hypertension or splenic infarcts [2]; cytopenias [3]; and an increased likelihood of transforming to a blast phase (i.e., secondary acute myeloid leukemia [sAML])[4]. sAML has a poor prognosis (i.e., median survival is 2.6 months after transforming to a blast phase) and there is limited responsiveness to standard intensive chemotherapy [5]. MF is also characterized by a debilitating symptom burden, including but not limited to fatigue, pruritus, night sweats, fever, muscle/bone pain, abdominal pain and unintentional weight loss [6,7]. These chronic symptoms can significantly impair quality of life (QoL) [6].

The median survival for patients with MF ranges from approximately 2–11 years depending on risk stratification [8]. Causes of death are multifactorial and include comorbidities, serious infections, transfusion dependency-related sequelae, organ dysfunction and sAML [9,10]. Approximately 15% of deaths in patients with MF have been attributed to sAML [9,10]. Prognostic factors associated with shortened survival at initial presentation include the following: older age (>65 years old); anemia (hemoglobin <10 g/dl); high white blood cell (WBC) counts (>25 × 10⁹/l); constitutional symptoms (fever, night sweats, weight loss); and blood blasts ≥1% [8]. These prognostic factors are the basis of risk stratification systems: the International Prognostic Scoring System (IPSS) [8] and the Dynamic International Prognostic Scoring System (DIPSS) [4]. Patients with PMF and 2 IPSS prognostic factors (defined as intermediate risk-2) and ≥3 IPSS prognostic factors (defined as high risk) have a median survival of approximately 4 years and 2 years, respectively [8]. The DIPSS is a dynamic model used for risk assessment at any point during the history of the disease; it captures the ‘tempo’ of the disease over time and represents the progressive nature of this malignancy [4]. While the DIPSS is based on the same prognostic factors as the IPSS, the factors are not weighted equally; anemia is weighted more than other factors [4]. The DIPSS-Plus model incorporates the prognostic factors from the DIPSS in addition to unfavorable karyotype, low platelet counts (<100 × 10⁹/l) and transfusion dependency [11], and as such may provide more detailed and specific delineation of patients at risk for early death, which may prompt treating doctors to implement the proper therapies at the appropriate moment in the disease course. Limitations of the DIPSS-Plus model include the need for cytogenetic information that is not readily available for many patients with MF in clinical practice due to poor bone marrow sampling, and the lack of experience with the model in the clinical trial setting.

Treatment goals in MF include reducing the signs and symptoms of MF, improving QoL and, if possible, prolonging survival [12]. The majority of the traditional therapies for MF have demonstrated limited efficacy in the management of splenomegaly, symptoms or cytopenias [13,14]. A new class of agents, the JAK inhibitors, has been shown to improve splenomegaly as well as the symptoms of MF [15–20]. One such JAK inhibitor under evaluation, ruxolitinib, became the first-in-its-class to be approved by the US FDA for the treatment of intermediate- or high-risk MF [101]. The objectives of this review are to describe meaningful end points in the evaluation of MF disease severity and the efficacy of JAK inhibitors on these measures.

Response criteria & assessment tools

Response criteria

There are numerous ways to assess disease involvement and the burden of signs and symptoms associated with MF, in addition to evaluating the responsiveness and efficacy of therapy. The International Working Group for Myelofibrosis Research and Treatment (IWG-MRT) developed criteria for response to therapy in MF, modeled after other criteria, such as the IWG criteria for myelodysplastic syndrome [21]. Treatment responses are categorized as complete remission (CR), partial remission (PR) or clinical improvement (Box 1). The criteria for ‘clinical improvement’ incorporate reduction in spleen size and improvements in anemia, platelet count and neutrophil count [21]. Such criteria perform an essential role in providing consistent measures of clinically relevant end points [22]. However, the CR and PR as surrogate markers of survival have not been validated, and it would be difficult to achieve CR or PR with currently available therapy apart from allogeneic stem cell transplantation (alloSCT [21]). Also, these criteria were published in 2006, prior to advanced clinical development of JAK2 inhibitors. Therefore, patient evaluation should encompass clinically relevant disease manifestations regardless of whether or not these are part of published response criteria.

Splenomegaly

Splenomegaly is present in the majority of patients with MF [2]. Some clinical trials define spleen response as a ≥50% reduction in splenic size (length) as assessed by palpation (consistent with IWG-MRT criteria) or ≥35% reduction when assessed by imaging (corresponding to an approximate ≥50% reduction in palpable spleen length, as shown in the Phase I/II trial of ruxolitinib [2,23]). Evaluating changes in spleen size by palpation, however, may not be as objective as evaluating changes in spleen size by MRI. In a Phase II study evaluating the JAK inhibitor pacritinib (formerly known as SB1518), a 50% reduction in spleen size as assessed by palpation correlated with a 25% reduction in spleen size as assessed by MRI [19]. Thus, use of objective measurements such as MRI or computed tomography (CT) to evaluate changes in spleen size may be more useful in evaluating response to therapy, especially in clinical trials.

MF-associated symptoms

The diagnosis of MF and other MPNs remains challenging; in particular, MF resembles a few other hematologic disorders and often requires hematopathology expertise to make a diagnosis [24]. Therefore, in clinical practice, the first step in managing MF is confirming the diagnosis. The next step is to evaluate prognostic factors and categorize the patient according to a risk stratification system, such as the IPSS or DIPSS [12]. Although alloSCT is associated with mortality risk [25], patients who are categorized as having intermediate-2 risk or high risk may be considered for alloSCT treatment [12]. The major symptoms should also be identified and an appropriate treatment regimen should be selected; the majority of patients should be treated based on symptoms, which can be anemia-based and/or splenomegaly-based and/or general systemic-MF-related and on the presence of transfusion requirement [12].

Constitutional symptoms (i.e., fever, night sweats, weight loss) are a de facto prognostic factor for shortened survival [8]. Indeed, these parameters are included in the IPSS [8] and DIPSS [4] risk-stratification systems. The MF-related symptom burden, however, is not limited to these constitutional symptoms; MF-specific symptoms also include pruritus, bone/muscle pain, abdominal pain/discomfort, left subcostal pain, early satiety, inactivity and fatigue [6]. The symptoms related to splenomegaly (i.e., early satiety, abdominal pain), can also impair physical function, such as walking, bending, eating or breathing [6].

The Myelofibrosis Symptom Assessment Form (MFSAF) is a tool specifically developed to evaluate the presence and severity of MF-associated symptoms [7]. The modified MFSAF v2.0 and earlier versions have been used in Phase II and III clinical trials [15,26]. In a recent version of the MFSAF (v2.0), patients rated the presence and severity of splenomegaly-related symptoms (abdominal discomfort, pain under the ribs on left side, early satiety), systemic symptoms (night sweats, pruritus, bone/muscle pain) and inactivity using a scale of 0 (symptom not present) to 10 (symptom present and worst imaginable [15]).

Fatigue is a problematic symptom and is pervasive, persistently severe, and almost universally present in MF patients [6]. Moreover, it interferes with activities of daily living and QoL. Tools to evaluate fatigue include the Functional Assessment of Cancer Therapy Fatigue scale (FACT-F), the Brief Fatigue Inventory, the Cancer Linear Analog Scale and the Patient Reported Outcomes Measurement Information System (PROMIS) Fatigue scale [27,28]. The FACT-Lym, which was developed for assessment of symptoms in lymphoma patients, includes an evaluation of constitutional (or ‘B’) symptoms, such as pruritus, bone/muscle pain and fatigue [16,29].

QoL

Improving QoL remains one of the primary goals for the management of cancer [12]. One common QoL instrument used in oncology is the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire-Core 30 (EORTC QLQ-C30), which incorporates functional assessments across several domains (i.e., physical, cognitive, role, emotional and social), in addition to several individual symptoms (including fatigue [30]).

Anemia

Anemia, which is prognostic of worsened survival [8], is prevalent among those with MF; approximately 30% of patients with PMF initially present with anemia [31]. The IWG-MRT includes improvement of anemia in its MF treatment-response criteria (Box 1) [32]. Red blood cell (RBC) transfusion status is also prognostic of disease outcomes; RBC transfusion dependence at diagnosis or acquiring RBC transfusion dependence ≤1 year after diagnosis is associated with worsened survival [32]. Transfusion independence is a component of clinical improvement according to the IWG-MRT criteria [21]. There is debate within the field as to the relevance of hemoglobin values in defining RBC transfusion dependence and RBC transfusion independence. An expert panel recommended modifying the definition of RBC transfusion dependence and RBC transfusion independence; although RBC transfusion dependence is defined as ≥2 U/month, it does not include hemoglobin values (Table 1) [33]. Despite these published reports, there remains no universally accepted definition of transfusion dependence/independence used consistently in clinical trials or clinical practice.

Traditional therapies for MF

The majority of the traditional therapies are palliative [14] and predominantly used for the management of splenomegaly or anemia [13]. Hydroxyurea (hydroxycarbamide) has frequently been the first-choice chemotherapy for the management of splenomegaly. In addition, it is often used for the management of other hematologic problems such as leukocytosis and thrombocytosis [3,13]. In one study, approximately 30% of patients receiving hydroxyurea achieved a reduction in spleen size and 12% of patients achieved improvements in hemoglobin levels [34]. Nonetheless, the benefits achieved with hydroxyurea are transient, with a median response of approximately 13 months. Adverse events associated with hydroxyurea include the development and/or worsening of anemia and ulcers (i.e., oral and leg).

Other therapies have been used for the management of splenomegaly: splenectomy and irradiation [35]. Splenectomy is associated with significant mortality and morbidity (9 and 31%, respectively [36]). Moreover, hepatomegaly, which is frequently associated with MF, is especially common in patients with a history of prior splenectomy [37]. Irradiation provides a short-term alleviation of symptoms but is also associated with significant morbidity (i.e., cytopenias [13,35]).

Symptomatic anemia is typically managed by blood transfusion, although formal efficacy data from randomized clinical trials are lacking [38]. Anemia treatment recommendations put forth in recently published UK consensus guidelines for MF diagnosis and management include the use of recombinant human erythropoietin in patients with inappropriately low levels of erythropoietin (<125 µ/l) and androgen therapy with danazol [38]. Recommendations from European Leukemia Net include corticosteroids and immunomodulatory drugs (e.g., thalidomide and lenalidomide) in addition to erythropoiesis-stimulating agents and androgen therapy [12]. Each of these options may induce treatment-limiting side effects, and no comparative trials have been conducted [12].

The only potential curative therapy is alloSCT [12]. AlloSCT is associated with high treatment-related mortality and relapse [25], and only a small proportion of patients are reasonable candidates [14]. Ideal candidates should be young, without comorbidities [39], with a performance status ≥90% and without peripheral blood blasts [25]. There remains, therefore, an important need for novel therapies that reduce the severity of MF symptoms and splenomegaly, and hopefully improve overall survival.

JAK-STAT signaling

The JAK-STAT pathway plays a pivotal role in the differentiation and development of hematopoietic cells and the functioning of the immune system [40]. The JAK family comprises JAK1, JAK2, JAK3 and TYK2 [41]. Dysregulation and constitutive action of this pathway has been implicated in MF and is therefore a target for MF treatment [41].

Normal JAK-STAT signaling

Cytokines and growth factors activate the extracellular portion of their cognate receptors [41,42], which in turn promotes the recruitment of JAK proteins to associate closely with the intracellular portion of these receptors and the activation of the JAK proteins via phosphorylation (Figure 1). Phosphorylation of JAK proteins, in turn, leads to the phosphorylation and activation of several intracellular downstream signaling proteins, such as STAT proteins [42]. Phosphorylated STATs translocate to the nucleus and act as inducible nuclear-transcription factors [42], which lead to transcriptional modulation and eventual expression of cellular, molecular and (patho)-physiologic actions that were promoted by the initial signal (ligand).

JAK signaling does not always require STAT; for instance several hematopoietic responses to G-CSF have been shown to be independent of STAT3 [43]. Other signaling pathways downstream of JAK include the PI3K-AKT-mTOR-Forkhead transcription factors signaling proteins and Ras-Raf-MEK-ERK pathway [41]. AKT is required for differentiation induced by erythropoietin [44], while erythropoietin-induced cellular proliferation is dependent on MAPK [45].

JAK family signaling is not entirely redundant, and JAK2 plays the essential role in mediating signaling by erythropoietin, thrombopoietin, GM-CSF, IL-5 and IL-3 [46]. As a result, JAK2 signaling has a pivotal role in determining the fate of hematopoietic progenitors. Mice lacking JAK2 ultimately maintain the ability to generate both lymphoid and erythroid progenitor cells, but fail to develop definitive erythrocytes [46,47].

The JAK-STAT signaling pathway is tightly regulated; without the presence of ligands and activation of their cognate receptors, the associated intracellular signaling proteins and transcription factors do not become phosphorylated, or activated, and the JAK-STAT system remains in a quiescent state [41]. There are also numerous negative regulators of the JAK-STAT signaling system, such as suppressors of cytokine signaling (SOCS [48]), protein inhibitors of activated STAT (PIAS [49]), and the intracellular JAK2 inhibitor LNK^-/- that negatively regulates JAK2 activation through the SH2 domain [50]. Negative regulation of JAK plays an important role in controlling hematopoiesis and preventing the development of myeloproliferative diseases as is evident in LNK ^-/- mice, which form MPNs associated with thrombocytosis, splenomegaly, fibrosis and B-cell overproduction.

JAK-STAT signaling dysregulation & the pathogenesis of MF

The dysregulation of the JAK-STAT signaling pathway in hematopoietic progenitor cells has been implicated in the pathogenesis of MF. Some of the somatic mutations that contribute to this dysregulated JAK-STAT activity include gain-of-function mutations directly in JAK2, upstream signaling mutations such as the mutation of the thrombopoietin receptor (MPLW515L [51]), and loss of JAK regulation by mutations in LNK (LNK exon 2 mutations [50]). Mutations in JAK that lead to the constitutive activation of JAK2, namely, the JAK2 V617F [52–54] and JAK2 exon 12 mutations have been identified, although this latter mutation was primarily identified in patients with PV [55]. JAK2 V617F is the most common mutation occurring in a large proportion (50–60%) of patients with PMF [53,55]. With JAK2 activating mutations, the pathway becomes cytokine independent and growth-factor-independent; therefore, even in the absence of these ligands, the intracellular signaling proteins are constitutively active [54–56]. Clonal expansion leads to increased JAK2 V617F allele burden and homozygosity influencing disease phenotype and differentiation of PV and ET. Further involvement of JAK-STAT signaling in the pathogenesis of MF relates to its role in mediating signaling from proinflammatory cytokines. These cytokines are elevated in environments of myeloproliferative disease and contribute to the debilitating symptoms of MPNs [23].

The identification of dysregulated JAK-STAT activity regardless of mutation status in MF provides a rationale for the development of therapies that abrogate this signaling pathway. JAK inhibitors block a catalytic ATP-binding domain within the JAK enzyme, which in turn prevents the phosphorylation of downstream signaling proteins (Figure 1) [41,42]. Importantly, the activity of JAK inhibitors, such as ruxolitinib, is not contingent on JAK2 V617F mutational status [16].

Treatment of MF with JAK inhibitors

There are numerous JAK inhibitors that are approved or undergoing evaluation in clinical trials for the management of MF, including ruxolitinib, SAR302503, CYT387, pacritinib and LY2784544 (Table 2) [15–20]. In November 2011, ruxolitinib was approved by the FDA for treatment of patients with intermediate- (both intermediate-1 and intermediate-2) or high-risk MF, including PMF, post-PV MF and post-ET MF [101].

Ruxolitinib

Ruxolitinib, formerly known as INCB018424, is a JAK1 and JAK2 inhibitor [57,58]. Ruxolitinib treatment demonstrated efficacy irrespective of mutation status in two large Phase III studies in patients with PMF, post-PV MF and post-ET MF and intermediate-2 or high-risk disease.

COMFORT-I was a double-blind, placebo-controlled Phase III clinical study (N = 309) in which patients were randomized (1:1) to ruxolitinib or placebo [15]. Significantly, more patients in the ruxolitinib arm achieved at least a 35% reduction in spleen volume, as assessed by MRI or CT, versus placebo at week 24 (41.9 vs 0.7%, p < 0.001) [15]. Importantly, while the majority of patients in the ruxolitinib arm had some reduction in spleen volume, almost all patients in the placebo arm had increases in spleen volume [15]. Improvements in MF-associated symptoms were also achieved with ruxolitinib therapy. Symptom burden was assessed with the modified MFSAF v2.0. The total symptom score was the sum of the individual symptoms (minus inactivity) reported by the patients: night sweats, pruritus, bone/muscle pain, abdominal discomfort, pain under the ribs (left side) and early satiety; 45.9% of patients in the ruxolitinib arm achieved a 50% or greater improvement in total symptom score by week 24 versus 5.3% in the placebo group (p < 0.001). Individual symptoms improved in the ruxolitinib arm and worsened in the placebo arm. Moreover, consistent with improvements in MF symptoms, patients also experienced improvements in fatigue and QoL, as assessed by the PROMIS Fatigue scale and EORTC QLQ-C30, respectively. Evidence also suggests that ruxolitinib may be associated with improved survival outcomes. At the time of a planned 4-month follow-up (after the primary analysis), ruxolitinib was associated with a significant survival advantage over placebo, with 8.4% deaths in the ruxolitinib arm and 15.6% deaths in the placebo arm (HR, 0.50, 95% CI: 0.25–0.98, p = 0.04).

In a separate open-label, randomized Phase III study (COMFORT-II; n = 219), patients were randomized (2:1) to ruxolitinib or best available therapy (BAT) [16]. Significantly, more patients in the ruxolitinib arm (28%) achieved at least a 35% reduction in spleen volume versus BAT (0%) (p < 0.001) at week 48. While most patients in the ruxolitinib arm had a reduction in spleen size, the majority of patients in the BAT arm had worsening splenomegaly. The EORTC QLQ-C30 was used to evaluate QoL. Patients in the ruxolitinib arm experienced improvements in global health status and role functioning, as well as individual MF-related symptoms captured by the QLQ-C30 instrument, while those in the BAT arm experienced worsening by most of these measures. FACT-Lym scores also improved with ruxolitinib treatment.

The ruxolitinib adverse event profile reported in the blinded COMFORT-I study was consistent with that observed in the open-label COMFORT-II study. In COMFORT-I, the most common adverse events with ruxolitinib therapy were hematologic; a greater proportion of patients in the ruxolitinib arm versus the placebo arm had grade 3 or 4 anemia (45.2 vs 19.2%) and grade 3 or 4 thrombocytopenia (12.9 vs 1.3%) [15]. The most common nonhematologic adverse events that occurred more frequently in the ruxolitinib group were ecchymosis, dizziness and headache.

SAR302503

SAR302503, formerly known as TG101348, is a JAK2 and FLT-3 inhibitor [41]. This agent is currently under evaluation in a Phase III study (JAKARTA) [102].

A Phase I study evaluated SAR302503 in patients (N = 59) with intermediate- or high-risk MF, post-ET MF and post-PV MF [17]; JAK2 V617F-mutation positive and negative patients were enrolled. Per the IWG-MRT criteria, 45% of patients achieved a spleen response with SAR302503. Patients also categorized their symptoms as absent, mild or moderate by using a scale of 0 (no symptoms) to 10 (worst). Among the patients who experienced symptoms at baseline, resolution of these symptoms occurred for the following proportion of patients: early satiety (56%), fatigue (25%), night sweats (89%), cough (67%) and pruritus (50%). Among patients who reported leukocytosis and thrombocytosis at baseline (defined as WBC count >11 × 10⁹/l and platelet count >450 × 10⁹/l, respectively), leukocytosis was resolved (achievement of normal WBC counts) in 56% of patients and thrombocytosis was resolved (achievement of normal platelet counts) in 88% of patients across doses. SAR302503 treatment led to a reduction in the JAK2 V617F allele burden in mutation positive patients (n = 51), although whether this corresponds to clinical response has yet to be determined.

The most common treatment-emergent hematologic grade 3 and 4 adverse events were anemia and thrombocytopenia; diarrhea was the most common grade 3 or 4 nonhematologic adverse event, most likely due to the FLT-3-inhibiting activity of this agent [41]. Hyperlipasemia was the most common grade 3 or 4 laboratory abnormality [17].

CYT387

CYT387, a JAK1 and JAK2 inhibitor [58], is undergoing evaluation in Phase II studies [102]. In a Phase I study evaluating CYT387 in patients (n = 36) regardless of mutation status with intermediate- or high-risk PMF, post-ET MF and post-PV MF, CYT387 was associated with improved splenomegaly for 37% of the patients in the study according to IWG-MRT criteria. In addition, among patients who were categorized as either having anemia (hemoglobin <10 g/dl) or being transfusion dependent at baseline (n = 22), approximately 63% of the patients improved or achieved stable anemia according to IWG-MRT criteria. The following proportions of patients achieved ‘complete resolution’ of symptoms: fatigue (16%), itchiness (57%), night sweats (75%), cough (50%), bone pain (44%) and fever (100%). During this Phase I study, grade 3 or 4 thrombocytopenia occurred in 22% of patients whereas grade 3 treatment-emergent anemia occurred in 3% of patients [18]. In an expanded Phase I/II multicenter study, CYT387 was also associated with improved splenomegaly and improvements in constitutional symptoms [59]. The most common treatment-related grade 3 or 4 adverse events were thrombocytopenia and hyperlipasemia.

Pacritinib (SB1518)

Pacritinib (SB1518) is a JAK2 and FLT-3 inhibitor [58]. In a Phase II trial evaluating pacritinib in JAK2 V617F-mutation positive and negative patients (n = 34) with PMF, post-ET MF and post-PV MF, pacritinib was associated with a reduction in spleen size, with 44% of patients achieving a reduction in splenomegaly as assessed by palpation (≥50%) and 32% achieving a reduction in splenomegaly as assessed by MRI (≥35%) [19]. Pacritinib was associated with a reduction in MF-associated symptoms, with reductions in abdominal pain, bone pain, early satiety, fatigue, inactivity, night sweats and pruritus as assessed by a >2 point reduction in the MFSAF score [19]. The most common adverse events were gastrointestinal, which would be expected due to the associated FLT-3 inhibitory action of this compound [41].

LY2784544

LY2784544 has been described as a JAK2 V617F (mutation)-selective inhibitor. A Phase I trial was undertaken to determine the maximum tolerated dose in patients (N = 19) with JAK2 V617F-positive MF, PV and ET (level of MF risk per IPSS or DIPSS not reported). Seventy-six percent (13 of 17 evaluable patients) achieved ≥35% spleen reduction as assessed by palpation and a reduction in fibrosis grade was seen in three of five patients with available data. Finally, LY2784544 was associated with an improvement in MF-related symptom burden, including pruritus, bone pain and night sweats. An maximum tolerated dose of 120 mg was estimated, based on the incidence of tumor lysis syndrome postexposure, but this was considered at the lower boundary of the biologically efficacious dose range; therefore, clinical studies with this agent are now incorporating a ‘lead-in’ period with a lower dose followed by further dose escalation [20].

Expert commentary

In the clinical setting, after first confirming diagnosis, the patient’s risk should be assessed according to the IPSS or DIPSS risk stratification system [12]. Additional aspects of the patient’s clinical profile (such as their full range of symptoms) should also be identified [12]. For patients who are categorized as being intermediate-2 risk or high risk, alloSCT therapy should be considered if eligible for this procedure based on age, performance and comorbidities [12]. Ruxolitinib, a JAK1 and JAK2 inhibitor, may be considered for patients with intermediate- or high-risk MF, including PMF, post-PV MF and post-ET MF [101]. This JAK inhibitor has demonstrated reductions in splenomegaly and improvements in the MF-associated symptom burden and QoL. In addition, data suggest that ruxolitinib may be associated with improved overall survival.

Ruxolitinib is associated with hematologic adverse events, including grade 3 or 4 anemia or thrombocytopenia. Therefore, if this agent is selected for the management of MF, complete blood counts and platelet counts should be collected before and during therapy, and precautions should be implemented to mitigate the risk of hematologic adverse events [15,16]. The starting dose should be determined by the baseline platelet count: 15 mg twice-daily ruxolitinib for a platelet count between 100 × 10⁹/l and 200 × 10⁹/l and 20 mg twice-daily for a platelet count exceeding 200 × 10⁹/l [15]. There are preliminary data to suggest that a starting dose of ruxolitinib 5 mg twice-daily with subsequent optimization based on safety and efficacy up to final (eventually attained) doses ≥10 mg twice-daily may be an effective dosing strategy for patients with baseline (pretreatment) platelet counts between 50 and <100 × 10⁹/l [60].

Other JAK inhibitors in development also show promise for the treatment of MF. Collectively, this new class of therapies has illuminated treatment benefits not captured in the IWG response criteria (developed before the availability of JAK inhibitors or knowledge of their clinical benefits). Reductions in spleen size and symptom improvements can have meaningful benefits on patient QoL. Moreover, MF-related mortality is often attributed to causes associated with massive splenomegaly and complications, such as portal hypertension or end-organ dysfunction, especially as MF patients also suffer from multiple comorbidities (unrelated to MF [9,10]). Therefore, it is important to incorporate a full range of disease assessments and measures of treatment response into patient evaluation, from both the physician and patient perspectives, based on our current and evolving understanding of MF. Equally important is to offer patients relief from florid and debilitating chronic complaints, as well as progressive splenomegaly, aiming to at least partially restore their functional abilities and improve health-related QoL and possibly long-term overall survival in this highly symptomatic malignancy.

Five-year view

Until recently, there have been no approved treatments for MF. Moreover, the traditionally available therapies for MF have been predominantly palliative, with only transient benefits. The availability of ruxolitinib and development of other JAK inhibitors in parallel has broadened our understanding of the disease and increased treatment expectations for patients and their physicians.

Other promising therapeutic targets have also been identified, such as the intracellular mTOR and histone deacetylases. Therapies that modulate these signaling pathways are under evaluation in patients with MF. New treatment strategies, such as combination therapy (e.g., ruxolitinib and panobinostat, a pan-histone deacetylase inhibitor), may further improve patient outcomes.

The only curative therapy will remain hematopoietic cell transplantation (HCT) in the foreseeable future. Since poor performance status is a predictor of poor outcomes with HCT, the reduction in symptomatic burden of MF with JAK-inhibitors may have the potential of improving results with transplant [61]. How to incorporate JAK inhibitors with HCT and when in the disease course to implement these modalities will be important questions to be addressed in upcoming clinical trials.

Table 1. Delphi expert consensus panel definitions of red blood cell transfusion dependence and red blood cell transfusion independence.

	RBC transfusions	Surveillance interval
RBC transfusion dependence	≥2 U/month	3 months
RBC transfusion independence	None	3 months
Reduced RBC transfusion dependence	50% decrease	3 months

RBC: Red blood cell.

Reproduced with permission from [31,33]

Table 2. JAK inhibitors approved and in development.

JAK inhibitors	Stage of clinical development
CYT387	Phase II
LY2784544	Phase I
Pacritinib (SB1518)	Phase I/II
Ruxolitinib	Approved, Phase III
SAR302503	Phase III

Data taken from [102].

Box 1. International Working Group consensus criteria for treatment response in myelofibrosis with myeloid metaplasia.

• • Complete remission

  • – Complete resolution of disease-related symptoms and signs including palpable hepatosplenomegaly.

  • – Peripheral blood count remission, defined as hemoglobin level of at least 110 g/l, platelet count of at least 100 × 10⁹/l and absolute neutrophil count of at least 1.0 × 10⁹/l. In addition, all three blood counts should be no higher than the upper normal limit.

  • – Normal leukocyte differential, including disappearance of nucleated red blood cells, blasts and immature myeloid cells in the peripheral smear, in the absence of splenectomy^†.

  • – Bone marrow histologic remission defined as the presence of age-adjusted normocellularity, no more than 5% myeloblasts, and an osteomyelofibrosis grade no higher than 1^‡.

• • Partial remission

  • – Requires all of the above criteria for CR except the requirement for bone marrow histologic remission. However, a repeat bone marrow biopsy is required in the assessment of PR and may or may not show favorable changes that do not however fulfill criteria for CR.

• • Clinical improvement

  • – Requires one of the following in the absence of both disease progression (as outlined below) and CR/PR assignment (CI response is validated only if it lasts for no fewer than 8 weeks).

  • – A minimum 20 g/l increase in hemoglobin level or becoming transfusion independent (applicable only for patients with baseline hemoglobin level of less than 100 g/l)^§.

  • – Either a minimum 50% reduction in palpable splenomegaly of a spleen that is at least 10 cm at baseline or a spleen that is palpable at more than 5 cm at baseline becomes not palpable^#.

  • – A minimum 100% increase in platelet count and an absolute platelet count of at least 50 000 × 10⁹/l (applicable only for patients with baseline platelet count below 50 × 10⁹/l).

  • – A minimum 100% increase in ANC and an ANC of at least 0.5 × 10⁹/l (applicable only for patients with baseline ANC below 1 × 10⁹/l).

• • Progressive disease

  • – Requires one of the following^¶:

  • – Progressive splenomegaly that is defined by the appearance of a previously absent splenomegaly that is palpable at greater than 5 cm below the left costal margin or a minimum 100% increase in palpable distance for baseline splenomegaly of 5–10 cm or a minimum 50% increase in palpable distance for baseline splenomegaly of greater than 10 cm.

  • – Leukemic transformation confirmed by a bone marrow blast count of at least 20%.

  • – An increase in peripheral blood blast percentage of at least 20% that lasts for at least 8 weeks.

• • Stable disease

  • – None of the above.

• • Relapse

  • – Loss of CR, PR or CI. In other words, a patient with CR or PR is considered to have undergone relapse when he or she no longer fulfills the criteria for even CI. However, changes from either CR to PR or CR/PR to CI should be documented and reported.

^†Because of subjectivity in peripheral blood smear interpretation, CR does not require absence of morphologic abnormalities of red cells, platelets and neutrophils.

^‡In patients with CR, a complete cytogenetic response is defined as failure to detect a cytogenetic abnormality in cases with a pre-existing abnormality. A partial cytogenetic response is defined as 50% or greater reduction in abnormal metaphases. In both cases, at least 20 bone marrow-derived or peripheral blood-derived metaphases should be analyzed. A major molecular response is defined as the absence of a specific disease-associated mutation in peripheral blood granulocytes of previously positive cases. In the absence of a cytogenetic/molecular marker, monitoring for treatment-induced inhibition of endogenous myeloid-colony formation is encouraged. Finally, baseline and post-treatment bone marrow slides are to be stained at the same time and interpreted at one sitting by a central review process.

^§Transfusion dependency is defined by a history of at least two units of red blood cell transfusions in the last month for a hemoglobin level of less than 85 g/l that was not associated with clinically overt bleeding. Similarly, during protocol therapy, transfusions for a hemoglobin level of 85 g/l or more is discouraged unless it is clinically indicated.

^#In splenectomized patients, palpable hepatomegaly is substituted with the same measurements.

^¶It is acknowledged that worsening cytopenia might represent progressive disease, but its inclusion as a formal criterion was avoided because of the difficulty distinguishing disease-associated from drug-induced myelosuppression. However, a decrease in hemoglobin level of 20 g/l or more, a 100% increase in transfusion requirement, and new development of transfusion dependency, each lasting for more than 3 months after the discontinuation of protocol therapy, can be considered disease progression.

ANC: Absolute neutrophil count; CI: Clinical improvement; CR: Complete remission; PR: Partial remission.

Reproduced from [21] with permission of American Society of Hematology.

Key issues

• • Myelofibrosis (MF)-associated symptom burden and progressive splenomegaly have a deleterious impact on quality of life.

• • Traditional therapies have been predominantly palliative and have historically offered limited benefit in improving the signs and symptoms of MF.

• • The discovery of the JAK-STAT pathway has improved our understanding of the pathogenesis of MF and led to the development of JAK inhibitors.

• • Along with the investigation of these new therapies, the identification of outcome measures (beyond the IWG-MRT criteria) and incorporation of these measures into clinical trials has led to better and more comprehensive assessment of treatment efficacy. In addition, the recent and ongoing clinical trials question the validity of proposed response criteria and provide an opportunity to better define criteria such as transfusion dependence/independence.

• • Ruxolitinib, a JAK1 and JAK2 inhibitor, was recently approved by the FDA for the treatment of patients with intermediate or high-risk MF.

• • Ruxolitinib treatment resulted in significant and clinically meaningful reductions in spleen volume and marked and sustained improvements in MF-related symptoms and quality of life measures.

• • Data available to date also suggest that ruxolitinib may improve survival versus placebo.

• • Other JAK inhibitors are in development and preliminary studies show promising results in terms of spleen reduction and symptom improvement.

Acknowledgements

We would like to thank Nicholas J Sarlis (Incyte Corp.) for helpful scientific exchange and thoughtful discussions on recent clinical study data.

Financial & competing interests disclosure

Srdan Verstovsek has disclosed the following relevant financial relationships: grant support from Incyte Corp., Lilly Oncology, Bristol-Meyers, AstraZeneca, Geron Corp., YM Biosciences, Celgene, Gilead, Roche, Infinity Pharmaceuticals, S*BIO, NS Pharma and Exelixis. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Editorial support was provided by Susan Kralian of The Curry Rockefeller Group and funded by Incyte Corp.