Pacritinib for the treatment of patients with myelofibrosis and thrombocytopenia

ABSTRACT

Introduction

Myelofibrosis (MF) is a rare myeloproliferative neoplasm characterized by a complex symptom profile, cytopenias, splenomegaly, and potential for leukemic progression. Severe thrombocytopenia is common in patients with MF and correlates with poor prognosis; however, until recently, treatment options for these patients were limited. Pacritinib, a potent Janus kinase (JAK) 2/interleukin-1 receptor-associated kinase 1 (IRAK1) inhibitor, has demonstrated significant reduction in splenomegaly, improved symptom control, and a manageable safety profile in patients with MF regardless of the severity of thrombocytopenia.

Areas covered

This review will outline the pacritinib drug profile and summarize key efficacy and safety data, focusing on the 200 mg twice daily dose from phase 2 and 3 studies that formed the basis for the recent US Food and Drug Administration approval of pacritinib in patients with MF and severe thrombocytopenia (platelet counts <50 × 10⁹/L).

Expert opinion

Pacritinib, with its unique mechanism of action targeting both JAK2 and IRAK1, offers patients with MF and severe thrombocytopenia a new treatment option, providing consistent disease and symptom control. Adverse events are easily manageable. Further analyses to identify ideal patient characteristics for pacritinib and other JAK inhibitors along with studies of pacritinib combinations are warranted, including in related myeloid malignancies.

KEYWORDS:

• Cytopenia
• JAK inhibitor
• myelofibrosis
• pacritinib
• thrombocytopenia

1. Introduction

Myelofibrosis (MF) is a rare myeloproliferative neoplasm (MPN) characterized by aberrant inflammatory cytokine expression, cytopenias, splenomegaly, progressive bone marrow fibrosis, extramedullary hematopoiesis, the potential for leukemic progression, and shortened survival [1,2]. MF can arise de novo (primary MF) or secondary to other MPNs, including post-polycythemia vera (PPV) or post-essential thrombocythemia (PET) [3,4]. Symptoms of MF, many of them constitutional, may go unrecognized in early disease, but with progression can lead to reduced quality of life (QoL), functional status, and activities of daily living [5–8]. The most frequent symptom is fatigue, experienced by nearly all patients [6]. Cytokine-driven constitutional symptoms are also common (e.g. night sweats, itching, bone pain, fever, and weight loss), as are those related to cytopenias (e.g. bleeding) and splenomegaly (e.g. early satiety and abdominal pain/discomfort) [2,6]. Up to 20% of patients with MF die after MF transforms to acute myeloid leukemia, while other causes of death include cardiovascular complications and consequences of cytopenias, such as bleeding and infections [1,9].

Virtually all patients develop anemia, and most experience thrombocytopenia during the course of their disease or treatment [10–13]. Cytopenic MF presents as a bone marrow failure state, and patients often need transfusions and have elevated risk of bleeding and infection [14]. Approximately one-fourth of patients with MF have moderate to severe thrombocytopenia (platelet counts ≤100 × 10⁹/L) at diagnosis [12], with the frequency rising over time (~45% one year post diagnosis and ~70% at any time during their disease course/treatment) [13,15]. Severe thrombocytopenia (platelet counts <50 × 10⁹/L) is estimated to be present in ~11% to 16% of patients with MF at diagnosis [13,16,17]. Incidence-based modeling, however, does not account for patients who develop severe thrombocytopenia due to treatment or natural disease course [13], and thus the reported frequency may be underestimated. Indeed, a recent survey of >800 hematologist-oncologists from 12 countries indicated that 35% of their patients with MF had severe thrombocytopenia, and 34% had moderate thrombocytopenia (50–100 × 10⁹/L) at the time of their last treatment initiation [13].

Moderate to severe thrombocytopenia has a profound negative impact on disease burden and prognosis in MF, correlating with reduced QoL, greater symptom burden, anemia, transfusion dependency, peripheral blasts, and risk of leukemic progression [12,18]. Importantly, it is also associated with shorter survival among patients with MF. In a retrospective cohort analysis, median survival was only 15 months for patients with a baseline platelet count <50 × 10⁹/L and 44 months for those with 50 to 100 × 10⁹/L, compared with 57 months for patients with platelet counts >100 × 10⁹/L [12]. Moderate to severe thrombocytopenia, along with unfavorable karyotype, is also an independent risk factor for poor prognosis according to the Dynamic International Prognostic Scoring System (DIPSS) Plus, a refined version of DIPSS that combines previously identified risk factors (age >65 years, hemoglobin <10 g/dL, leukocytes >25 × 10⁹/L, circulating blasts ≥1%, and constitutional symptoms) with prognostic information on karyotype, platelet count, and transfusion status [19]. Survival varies greatly by risk group, with common prognostic systems predicting survival of <2 years for patients with high-risk MF [11].

2. Overview of the therapeutic landscape

Patients with MF and limited symptom burden are commonly managed with a watch-and-wait strategy [20]. As symptom burden increases with disease evolution, supportive care is introduced, including transfusion support (with or without iron chelation), antifibrinolytic agents, hematopoietic growth factor support, and cytoreductive therapy if warranted [1,20]. For patients with intermediate-2 or high-risk MF, allogeneic stem cell transplant (SCT) is the only potentially curative treatment, although many patients are either ineligible for or unwilling to undergo this procedure [1,20,21].

For patients with higher-risk MF who are not SCT candidates, presence of severe thrombocytopenia is currently the only delineator for treatment recommendations from the National Comprehensive Cancer Network (NCCN) [20]. Based on the recent update, pacritinib is now recommended for patients with platelet counts <50 × 10⁹/L, as well as for those with platelet counts ≥50 × 10⁹/L who have previously received ruxolitinib or fedratinib [20]. Both ruxolitinib and fedratinib are associated with myelosuppression and require dose modification or hold in the event of severe thrombocytopenia [20,22,23]. Therefore, enrolling in a clinical trial has long been the only non-SCT treatment recommendation for patients with MF and platelet counts <50 × 10⁹/L. This finally changed on 28 February 2022, when pacritinib was approved for the treatment of patients with MF with severe thrombocytopenia [20,24].

Several therapies for MF are currently under investigation [25]; however, only Janus kinase (JAK) inhibitors have been approved or entered late-phase development. See Table 1 for an overview of approved JAK inhibitors and those approaching approval. The JAK inhibitor class has immunosuppressive properties that may explain their benefit for cytokine-related MF symptoms, as well as for rheumatologic conditions. However, these properties may be associated with an increased risk of infections or secondary malignancies, as several JAK inhibitors indicated for rheumatologic disease have required black box warnings for infections and lymphoma [26].

Table 1. Currently available and emerging JAK inhibitors for the treatment of MF.

Compound	Source	MoA/target	Regulatory status in the US
Pacritinib	[24,27–30]	Inhibitor of JAK2, IRAK1, ACVR1^a, and FLT3	Approved on 28 February 2022 for the treatment of patients with MF and platelet counts <50 x 10^9/L. A phase 3 trial assessing pacritinib vs physician’s choice in patients with MF and severe thrombocytopenia is ongoing
Ruxolitinib	[22,25,31–34]	Inhibitor of JAK1 and JAK2	Approved on 16 November 2011 for the treatment of adult patients with intermediate or high-risk MF or polycythemia vera who had inadequate response to or are intolerant of hydroxyurea, and for certain patients with graft-versus-host disease. Ruxolitinib is also under investigation in several combination regimens
Fedratinib	[23,35,36]	Inhibitor of JAK2 and FLT3	Approved on 16 August 2019 for the treatment of adult patients with intermediate or high-risk MF
Momelotinib	[37–39]	Inhibitor of JAK1, JAK2, ACVR1, and FLT3	In phase 3 clinical trials

ACVR1: activin A receptor type 1; FLT3: fms-like tyrosine kinase 3; JAK1/2: Janus kinase 1/2; MF: myelofibrosis; MoA: mechanism of action; US: United States.

^aPacritinib inhibits ACVR1 with a mean IC₅₀ of 16.7 nM [29].

2.1. Currently available JAK inhibitors

2.1.1. Ruxolitinib

Ruxolitinib is a JAK1/2 inhibitor approved in 2011 for the treatment of adult patients with intermediate- or high-risk MF, including primary MF, PPV-MF, and PET-MF [22]. The approval was based on data from two phase 3 trials that assessed ruxolitinib vs placebo (COMFORT-I; N = 309) or best available therapy (BAT) (COMFORT-II; N = 219) in patients with intermediate-2 or high-risk MF and baseline platelet counts ≥100 × 10⁹/L [22,33,34]. Patients with platelet counts <100 × 10⁹/L were specifically excluded [33,34] based on data from previous investigations of ruxolitinib demonstrating myelosuppression and treatment-induced cytopenias [40]. Median baseline platelet counts in patients treated with ruxolitinib were 262 × 10⁹/L (COMFORT-I) and 244 × 10⁹/L (COMFORT-II) [33,34]. The ruxolitinib starting dose in the COMFORT trials was 15 or 20 mg twice daily (BID) depending on platelet counts (100–200 × 10⁹/L vs >200 × 10⁹/L) [22]. After this, dosing was further individualized based on tolerability and efficacy, ranging from 5 mg to 20 mg BID [22].

Across the two studies, ruxolitinib demonstrated improved efficacy vs placebo or BAT; ≥35% spleen volume reduction (SVR) was achieved by 42% of patients treated with ruxolitinib (vs <1% with placebo; intent-to-treat [ITT] population; P < 0.001) at Week 24 in COMFORT-I and by 28% (vs 0% with BAT; ITT; P < 0.001) at Week 48 in COMFORT-II [33,34]. Symptom improvement as assessed by ≥50% reduction in modified Total Symptom Score (mTSS) at Week 24 was observed in 46% of patients treated with ruxolitinib (vs 5% with placebo; P < 0.001) in COMFORT-I [34]. A more recent exploratory analysis of pooled data from COMFORT-I and COMFORT-II also suggested a survival advantage associated with ruxolitinib vs control (BAT or placebo) [41]. However, aligned with previous studies, ruxolitinib was associated with treatment-induced cytopenias in the COMFORT studies, with most patients experiencing grade ≥3 anemia and/or thrombocytopenia [33,34,40].

The association with treatment-induced myelosuppression has led to safety concerns with ruxolitinib for patients with thrombocytopenia and limited efficacy with lowered dosing in this patient population [42–44]. In a phase 2 trial in patients with MF and moderate thrombocytopenia, ruxolitinib was initiated at 5 mg BID and at Week 24, approximately one half of patients (27/52) were still receiving <10 mg BID, 39% (20/52) were dosed at 10 mg BID, and only 10% (5/52) received >10 mg BID [45]. In the JUMP trial, assessing the potential of low-dose ruxolitinib (≤10 mg BID), patients with MF and moderate thrombocytopenia experienced worsening of thrombocytopenia and reduced overall survival compared with patients who did not have thrombocytopenia [43].

2.1.2. Fedratinib

Fedratinib is a JAK2-selective inhibitor that also exhibits off-target inhibitory activity against fms-like tyrosine kinase 3 (FLT3) and low inhibitory activity against JAK1 [35]. Fedratinib was approved in 2019 for the treatment of adult patients with intermediate-2 or high-risk primary or secondary (PPV or PET) MF [23]. Fedratinib may be used in the front-line setting or following ruxolitinib treatment [20]. The approval was supported by the phase 3 JAKARTA trial, and the recommendation for use following ruxolitinib was based on data from the phase 2 JAKARTA-2 trial [23,46]. In JAKARTA, ruxolitinib-naïve patients with baseline platelet counts ≥50 × 10⁹/L were randomized to receive fedratinib 400 or 500 mg once daily (QD) or placebo. Median platelet counts at baseline were 221 × 10⁹/L and 241 × 10⁹/L for the fedratinib 400 mg and 500 mg QD arms, respectively [47]. Later, the JAKARTA-2 trial assessed fedratinib 400 mg QD in patients with intermediate- or high-risk ruxolitinib-resistant or -intolerant MF, excluding those with baseline platelet counts <50 × 10⁹/L [46]. At the beginning of JAKARTA-2, 33% of patients had platelet counts between 50 × 10⁹/L and <100 × 10⁹/L, and 66% had platelet counts ≥100 × 10⁹/L [46].

Fedratinib improved the ≥35% SVR rate compared with placebo (36% with 400 mg QD and 40% with 500 mg QD vs 1%) in JAKARTA [47]. Most patients in this trial, however, experienced treatment-related myelosuppression, and four cases of encephalopathy were reported [47]. The phase 2 JAKARTA-2 trial was terminated due to these findings and data from other fedratinib trials [46,48]. Following the initial ‘per protocol’ analysis of JAKARTA-2, an updated ITT analysis reported an ≥35% SVR in 31% of patients in the ITT population [48]. Fedratinib now carries a black box warning for encephalopathy, including Wernicke’s, requiring assessment of thiamine (vitamin B1) levels before and during treatment and replacement of thiamine as indicated [23].

Two ongoing phase 3 trials, FREEDOM and FREEDOM2, are currently evaluating risk mitigation strategies for gastrointestinal (GI) toxicities and encephalopathy with fedratinib [49,50]. Recent data release from FREEDOM demonstrated lower rates and severity of GI adverse events (AEs) vs previous fedratinib trials and no cases of Wernicke’s encephalopathy [50].

2.2. Emerging JAK inhibitor

2.2.1. Momelotinib

Momelotinib is a JAK1/2, activin A receptor type 1 (ACVR1), and FLT3 inhibitor currently in late-stage clinical development for MF [38,39,51]. Two phase 3 clinical trials have assessed momelotinib vs ruxolitinib (SIMPLIFY-1) and BAT (SIMPLIFY-2) in JAK-inhibitor-naïve patients and those previously treated with ruxolitinib, respectively [52,53]. Patients had to have platelet counts ≥50 × 10⁹/L to enroll in SIMPLIFY-1 while no minimum cutoff threshold in platelet count was set for SIMPLIFY-2. Mean baseline platelet counts among patients receiving momelotinib were 301 × 10⁹/L (SIMPLIFY-1) and 171 × 10⁹/L (SIMPLIFY-2) [52,53].

The primary but not secondary endpoint of noninferiority was met in the SIMPLIFY-1 trial; ≥35% SVR was achieved by 26.5% vs 29.5% of patients treated with momelotinib 200 mg QD and ruxolitinib, respectively (P < 0.011), whereas ≥50% reduction in mTSS was reported in 28% vs 42% (P = 0.98) [53]. In SIMPLIFY-2, 7% vs 6% of patients in the momelotinib 200 mg QD and BAT arms, respectively, achieved ≥35% SVR (P = 0.90), indicating that the primary endpoint of noninferiority was not met, and 26% vs 6% achieved ≥50% reduction in mTSS (nominal P = 0.0006) [52].

Similar to ruxolitinib and fedratinib, momelotinib is associated with treatment-induced myelosuppression. In the SIMPLIFY studies, patients treated with momelotinib had a higher rate of all-grade treatment-induced AEs compared with those treated with BAT, with anemia and thrombocytopenia being the most common grade ≥3 AEs [52,53]. In addition, in SIMPLIFY-1, treatment-emergent peripheral neuropathy occurred in 10% of momelotinib-treated patients [53], and despite excluding patients with grade ≥2 peripheral neuropathy from SIMPLIFY-2, 11% of momelotinib-treated patients experienced peripheral neuropathy on trial [52].

The phase 3 MOMENTUM trial is currently investigating clinical benefits of momelotinib vs danazol in symptomatic patients with MF who have anemia and were previously treated with an approved JAK inhibitor [37]. Patients were randomized 2:1 to receive momelotinib 200 mg QD or danazol 600 mg QD [54]. Primary data release included 195 patients (ITT; momelotinib n = 130; danazol n = 65), with median baseline platelet counts of 97 × 10⁹/L (momelotinib) and 94 × 10⁹/L (danazol) and median hemoglobin of 8.1 g/dL and 7.9 g/dL [54]. A total of 25% of patients in the momelotinib arm vs 9% in the danazol arm achieved ≥50% reduction in TSS (primary endpoint; P = 0.0095) [54]. Statistically significant improvements were also observed in all secondary endpoints, including transfusion independence rate (31% vs 20%; one-sided P = 0.0064) and ≥35% SVR (23% vs 3%; P = 0.0006) [54]. During the randomized period, grade ≥3 treatment-emergent AEs (TEAEs) were reported in 54% and 65% of patients in the momelotinib and danazol arms, respectively, with thrombocytopenia (22% vs 12%) and anemia (8% vs 11%) as the most common grade ≥3 TEAEs [54,55]. A separate analysis of data from a patient subset with baseline platelet counts of ≤150 × 10⁹/L (median 67 × 10⁹/L for momelotinib and 64 × 10⁹/L for danazol) showed consistent efficacy results with the ITT analysis set [56].

3. Introduction to pacritinib

Pacritinib (200 mg BID) was approved by the US Food and Drug Administration (FDA) on 28 February 2022 for the treatment of adult patients with intermediate or high-risk primary MF, PPV-MF, or PET-MF and severe thrombocytopenia (platelets <50 x10⁹/L) [24,27].

3.1. Pacritinib mechanism of action

The JAK/signal transducer and activator of transcription (STAT) pathway plays a central role in cell proliferation, differentiation, and survival, and the four JAKs (JAK1, JAK2, JAK3, and TYK2) in discussion bind various receptors, impacting signaling of >50 cytokines [57,58]. In MF, the constitutively active JAK/STAT pathway has long been considered the classic driver of the disease course, with most patients harboring acquired driver mutations in JAK2 (JAK2^V617F in ~60%), calreticulin (CALR exon 9 indels; ~30%), and/or thrombopoietin receptor (also called MPL) (~5%), all of which contribute to the JAK/STAT-mediated inflammatory cascade and dysregulated hematopoiesis [59]. Of note, lower JAK2^V617F allele burden has been associated with primary MF (vs secondary MF), lower platelet counts and hemoglobin, increased transfusion dependency, and worse prognosis [60–62].

Recent studies also implicate additional, differentiated biologic drivers in the pathogenesis of MF via activation of the toll-like receptor (TLR)/Myddosome/interleukin-1 receptor-associated kinase 1 (IRAK1) inflammatory pathway [63–65]. The TLR ligands, interleukin (IL)-33 and other alarmins (specifically S100A8/A9) activate signaling through this pathway [66–68], which may in turn stimulate an unrestrained cascade of inflammatory cytokines through nuclear factor kappa light chain enhancer of activated B cells (NFκB) transcription and activation of p38 [63], as well as phosphorylation of STAT3/5 [69]. These cytokines are not extinguished by JAK1/2 inhibition [69], and IRAK1 therefore represents a distinct pathway to target in the pathogenesis of MF, independent of JAK signaling.

Pacritinib is thought to exert clinical activity by inhibiting two distinct pathways, JAK/STAT and TLR/Myddosome/IRAK1, leading to suppression of NFκB and downstream inflammatory cytokine cascade, reduction in splenomegaly, and MF symptom control (Figure 1) [28,69,70]. Pacritinib is a potent competitive reversible kinase inhibitor with high specificity for JAK2 (IC₅₀ = 6.0 nM), JAK2^V617F (IC₅₀ = 9.4 nM), and IRAK1 (IC₅₀ = 13.6 nM) [27,28]. At physiologically relevant doses, pacritinib induces dose-dependent inhibition of JAK2/JAK2^V617F and downstream STAT phosphorylation [71] but has negligible effect on JAK1 (IC₅₀ > 100 nM), the dominant JAK for interferons and IL-2/6 signaling [28]. In contrast, ruxolitinib and momelotinib are potent JAK1 inhibitors (IC₅₀ = 3.3 nM and 11 nM, respectively) while fedratinib is more selective for JAK2 than JAK1 (IC₅₀ = 3 nM for JAK2 and 46–105 nM for JAK1) [72,73]. As fedratinib achieves mean steady state C_max value of 1,804 ng/mL (~3,440 nM; free fraction 275 nM) following multiple administration of 400 mg QD [23], fedratinib can be expected to potently inhibit JAK1 at clinical exposure levels. Since JAK1 inhibition is known to impair megakaryopoiesis and, further, platelet production, this negligible JAK1 inhibitory activity may contribute to the hematologic stability observed in pacritinib investigations [28,74–76]. Computer modeling and site-directed mutagenesis studies also validated high-affinity binding of pacritinib to the IRAK1 kinase domain [77]. Through inhibition of IRAK1, pacritinib suppresses NFκB, p38, STAT3, and downstream inflammatory cytokine cascade in a JAK-independent manner, resulting in anti-inflammatory effects [28]. Furthermore, STAT3 suppression may reinforce the effects of JAK2 inhibition.

Figure 1. Pacritinib mechanism of action [1,69,71,78–83].

Pacritinib has the potential for off-target effects via inhibition of FLT3, including internal tandem duplication (ITD) and D835Y mutations, as well as colony-stimulating factor-1 receptor (CSF-1R). The impact of these targets in MF is unknown, though these targets may have therapeutic relevance in other myeloid malignancies [28]. Pacritinib is also a potent inhibitor of ACVR1 (IC₅₀ = 16.7 nM) [29].

3.2. Pharmacology of pacritinib

Orally administered pacritinib is readily absorbed regardless of food intake, has a mean terminal elimination half-life of 27.7 hours, and is predominantly metabolized by the CYP3A4 isozyme and biliary excretion [27,84]. Pacritinib achieves maximum plasma concentration at approximately four to five hours post dose and has an apparent volume of distribution of 229 L with 98.8% plasma protein binding [27]. Pacritinib is a time-dependent inhibitor of CYP1A2 and CYP3A4 isozymes and a reversible inhibitor of CYP3A4 and CYP2C19 isozymes based on in vitro metabolism data [27]. Moreover, with multiple dose administration, pacritinib is an inducer of CYP1A2 and CYP3A4 isozymes [27]. It also inhibits BCRP, OCT1, OCT2, and P-glycoprotein transporters [27]. In patients with moderate and severe hepatic impairment, pacritinib exposure is decreased by 36% and 45%, respectively, compared with healthy volunteers [27].

The recommended dose of pacritinib is 200 mg BID based on population pharmacokinetic (PK)/pharmacodynamic (PD) modeling and the totality of data from all clinical studies of pacritinib in patients with MF [85]. Population PK modeling of data from 16 pacritinib studies (11 phase 1 and 5 phase 2 or 3 studies) included 630 patients who received pacritinib at doses of 100 mg to 600 mg QD and predicted high steady-state exposure and SVR with doses of 200 mg BID and 400 mg QD [85]. Specifically, population PK/PD modeling of data from the phase 2 PAC203 dose-finding study and phase 3 PERSIST-1 and PERSIST-2 studies found that pacritinib 200 mg BID and 400 mg QD were associated with greater SVR and reduction in TSS compared with lower doses [85].

4. Clinical efficacy of pacritinib

4.1. Overview of pacritinib development

Following preclinical and early-phase pacritinib studies, late-phase development of pacritinib for use in MF initially consisted of two phase 3 trials, PERSIST-1, initiated in 2013, and PERSIST-2, initiated in 2014 [86]. The 400 mg QD dosing for PERSIST-1 was based on dose-finding investigations and population PK modeling of doses ranging from 100 to 600 mg QD as described in section 3.2 above [87]. A 200 mg BID arm was added to PERSIST-2 for potentially improved tolerability, based on PK modeling data demonstrating increased daily systemic exposure with lower maximum concentration vs 400 mg QD [88].

In February 2016, a full clinical hold was placed on pacritinib by the FDA due to concerns over a potential detrimental effect on survival and deaths in pacritinib-treated patients due to intracranial hemorrhage, cardiac failure, and cardiac arrest [70,75]. Final data from PERSIST-2 did not, however, demonstrate poorer survival associated with pacritinib, as mortality was lowest with pacritinib 200 mg BID (hazard ratio [HR] 0.68 [95% CI, 0.30–1.53] vs BAT) [70]. A comprehensive clinical review of high-grade bleeding and cardiac events across the PERSIST studies identified potential contributing factors, including selection of patients with active or recent bleeding and/or cardiac events, as well as limited dose modification guidance to prevent and manage hemorrhage [85]. The clinical hold was removed in January 2017 after complete study reports were submitted for the PERSIST studies and an agreement was reached on the design of the phase 2 dose-finding PAC203 trial, which incorporated additional risk mitigation strategies [70,85]. Table 2 provides an overview of the completed phase 2 and 3 trials of pacritinib. A phase 3 trial of pacritinib 200 mg BID vs physician’s choice of therapy (PACIFICA) is currently recruiting patients with primary or secondary MF and severe thrombocytopenia with limited or no prior JAK2 inhibitor therapy [30,89].

Table 2. Summary of phase 2 and 3 trials assessing pacritinib in MF.

Study name (ID), phase, and N	Study design and key eligibility criteria	Inventions	Outcomes	Source
PERSIST-1 (NCT01773187) Phase 3 N = 327	Randomized, controlled study Intermediate- or high-risk^a MF and palpable splenomegaly ≥5 cm below the left costal margin; no prior JAK2 inhibitor exposure	Patients were randomized 2:1 to: Pacritinib 400 mg QD or BAT (excluding JAK1/2inhibitors)	Primary: SVR Secondary: TSS^b	[75]
PERSIST-2 (NCT02055781) Phase 3 N = 311	Randomized, controlled study Intermediate-1 or 2 or high-risk^a MF with platelet counts ≤100 × 10^9/L and palpable splenomegaly ≥5 cm below the left costal margin	Patients were randomized 1:1:1 to: Pacritinib 400 mg QD or Pacritinib 200 mg BID or BAT (including JAK1/2 inhibitors)	Coprimary: SVR, TSS^b	[70]
PAC203 (NCT04884191) Phase 2 N = 161	Open-label, randomized, dose-finding study Intermediate-1 or 2 or high-risk^a MF that is resistant or intolerant to ruxolitinib	Patients were randomized 1:1:1 to: Pacritinib 100 mg QD or Pacritinib 100 mg BID or Pacritinib 200 mg BID	Primary: Confirm recommended dose Secondary: SVR, TSS, PK/PD	[85]

^aAs assessed by Dynamic International Prognostic Scoring System.

^bMyeloproliferative Neoplasm System Assessment Form TSS v2.0.

BAT: best available therapy; BID: twice daily; JAK1/2: Janus kinase 1/2; MF: myelofibrosis; PK/PD: pharmacokinetics/pharmacodynamics; QD: once daily; SVR: spleen volume reduction; TSS: Total Symptom Score.

4.2. PERSIST-1

PERSIST-1 was a phase 3 study assessing the efficacy and safety of pacritinib vs BAT in adult patients with intermediate- or high-risk primary or secondary MF and no prior JAK2 inhibitor treatment [75]. Patients were randomized 2:1 to receive pacritinib 400 mg QD or BAT, where BAT could be any physician-selected treatment, excluding JAK2 inhibitors but including watchful waiting (no treatment) [75]. The BAT approaches could be used alone, in combinations, sequentially, or intermittently, as clinically indicated by standards of care [90]. Randomization was stratified by DIPSS risk category, platelet count, and geographical region [75]. Unlike most other trials of JAK inhibitors, patients were not excluded from PERSIST-1 based on platelet or hemoglobin levels or red blood cell (RBC) transfusion dependency [70,75]. The primary endpoint of the trial was ≥35% SVR at Week 24, assessed by blinded, centrally reviewed computed tomography (CT) or magnetic resonance imaging (MRI). The secondary endpoint, the proportion of patients with ≥50% reduction in TSS at Week 24, was initially measured using the original Myeloproliferative Neoplasm System Assessment Form (MPN-SAF) TSS; however, MPN-SAF TSS v2.0, which was developed to more accurately reflect the symptom burden of MF, was introduced following the protocol amendment on 15 August 2013 [75]. Of note, TSS endpoints in pivotal studies for ruxolitinib and fedratinib used mTSS, which excludes the individual symptom score for tiredness [34,47].

A total of 327 patients were included in PERSIST-1, 220 in the pacritinib arm and 107 in the BAT arm [75]. The most frequently administered BAT was hydroxyurea (57% of patients), and 26% of patients in the BAT arm received only watchful waiting [75]. Approximately one-half of patients in both arms had intermediate-2 or high-risk disease (43% in the pacritinib and 54% in the BAT arm). A total of 16% and 15% of patients receiving pacritinib and BAT, respectively, had severe thrombocytopenia at baseline, with the mean platelet counts estimated at 30 × 10⁹/L in this subgroup for both arms [75,90]. Pacritinib was administered for a median duration of 15.6 months and BAT for 5.9 months; after a median of 6.3 months, 84% of patients in the BAT group crossed over to the pacritinib arm [75].

Pacritinib 400 mg QD demonstrated significant and durable reduction in splenomegaly in comparison with BAT [75]. By Week 24, 19% and 5% of patients in the pacritinib and BAT arms, respectively, achieved ≥35% SVR (P = 0.0003; ITT population). The difference remained statistically significant across baseline platelet groups, with 23% (pacritinib) vs 0% (BAT) of patients with severe thrombocytopenia achieving this endpoint (Table 3) [75]. The proportions of patients with ≥50% reduction in TSS at Week 24 were comparable between the two arms overall; however, by Week 48, 15% and 0% of patients in the pacritinib and BAT arms, respectively, achieved ≥50% reduction in TSS (P = 0.0027) [75]. For key hematologic parameters, in patients with severe thrombocytopenia at baseline, pacritinib treatment resulted in increased platelet counts at Week 24 (mean platelet count 44 x 10⁹/L; P = 0.06 vs baseline), while there was no significant improvement in patients receiving BAT [75,90]. In addition, among patients who were RBC transfusion-dependent at baseline, 25% of those receiving pacritinib vs 0% of those on BAT became RBC transfusion independent during the study (P = 0.043) [75].

Table 3. Summary of efficacy outcomes from pacritinib phase 2 and 3 trials.

Study population	SVR^a at Week 24	TSS^b at Week 24	Hematologic parameters at Week 24	Source
PERSIST-1/Pacritinib 400 mg QD vs BAT
ITT	Pacritinib vs BAT: 19% (42/220) vs 5% (5/107); P = 0.0003	Pacritinib vs BAT: 19% (19/100) vs 10% (5/48); P = 0.24 Week 48 Pacritinib vs BAT: 15% (15/100) vs 0% (0/48); P = 0.0027	Proportion of pts converting from RBC TD to TI: 25% (9/36) vs 0% (0/16); P = 0.043	[75]
Pts with platelets <50 x 10^9/L	Pacritinib vs BAT: 23% (8/35) vs 0% (0/16); P = 0.045	Pacritinib vs BAT: 27% (3/11) vs 0% (0/5); P = 0.51
PERSIST-2/Pacritinib 200 mg BID vs BAT
ITT	Pacritinib vs BAT: 22% (16/74) vs 3% (2/72); P = 0.001	Pacritinib vs BAT: 32% (24/74) vs 14% (10/72); P = 0.01 Median change in TSS: −41% vs −15% mTSS Pacritinib vs BAT:35%(26/74) vs 14% (10/72); P = 0.004	Proportion of pts^c with reduced RBC transfusion burden: 22% (8/36) vs 9% (3/35) Proportion of pts with anemia^d achieving clinical improvement in Hb: 25% (11/44) vs 12% (5/41)	[70,91]
Pts with platelets <50 x 10^9/L	Pacritinib vs BAT: 29% (9/31) vs 3% (1/32)	Pacritinib vs BAT: 23% (7/31) vs 13% (4/32)
PAC203/Pacritinib 200 mg BID
ITT	Pacritinib: 9% (5/54)	Pacritinib: 7% (4/54) Median change in TSS: −27%	Proportion of pts with ≥50% reduction in RBC transfusion burden: 15% Proportion of pts converting from RBC TD to TI: 10%	[85]
Evaluable	Pacritinib: 19% (5/27)	Pacritinib: 17% (4/24)
Pts with platelets <50 x 10^9/L	Pacritinib 31% (4/13)	Pacritinib: 15% (2/13)

^aReduction of ≥35% at Week 24; ^bReduction of ≥50% at Week 24; measured by Myeloproliferative Neoplasm Symptom Assessment Form TSS v2.0; ^cPatients who were not RBC transfusion-independent at baseline; ^dHemoglobin <10 g/dL at baseline.

BAT: best available therapy; BID: twice daily; Hb: hemoglobin; ITT: intent-to-treat; mTSS: modified Total Symptom Score; pts: patients; QD: once daily; RBC: red blood cell; SVR: spleen volume reduction; TD: transfusion dependency; TI: transfusion independency; TSS: Total Symptom Score.

4.3. PERSIST-2

The phase 3 PERSIST-2 trial enrolled adult patients with primary or secondary intermediate- or high-risk MF and moderate to severe thrombocytopenia (≤100 × 10⁹/L) [70]. Patients were randomized 1:1:1 to pacritinib 200 mg BID, pacritinib 400 mg QD, or BAT (any physician-selected treatment including JAK1/2 inhibitors), and stratified by DIPSS risk category, rebound platelet count (defined as recovery of platelet count between informed consent and randomization), and geographical region [70]. Enrollment was restricted to patients with baseline platelet counts ≤100 x 10⁹/L, but patients were not excluded on the basis of hemoglobin levels, RBC transfusion dependency, or prior treatment with JAK inhibitors [70]. The coprimary endpoints for the pooled pacritinib and BAT arms were the proportion of patients in the ITT efficacy population achieving ≥35% SVR (via CT/MRI) and ≥50% reduction in TSS via MPN-SAF TSS v2.0 at Week 24 [70]. In addition to pooled pacritinib data, pacritinib 200 mg BID and 400 mg QD were also assessed separately vs BAT [70]. Exploratory endpoints included hematologic parameters and patient-reported outcomes. The 200 mg BID dose was selected for further clinical development and is now the approved dose of pacritinib [24,27,70]; therefore, this review of PERSIST-2 data will focus on this dose.

A total of 311 patients were enrolled in the study, 107 and 104 in the pacritinib 200 mg BID and 400 QD arms, respectively, and 100 in the BAT arm [70]. Efficacy ITT population included all patients randomized ≥22 weeks before the clinical hold, allowing for Week 24 data [70]. Most patients had intermediate-2 or high-risk disease (81% in the pacritinib 200 mg BID and 82% in the BAT arm) and nearly one-half had received ruxolitinib previously (42% and 46%, respectively) [70]. Mean baseline platelet counts were 71 × 10⁹/L and 63 × 10⁹/L in the pacritinib 200 BID and BAT arms (efficacy ITT population), with severe thrombocytopenia reported in 42% and 44% of patients in these treatment arms [70,92]. Ruxolitinib was the most commonly administered BAT, received by 45% of patients in the BAT arm, 39% of whom had severe thrombocytopenia and would not have been candidates for ruxolitinib per approved label or PERSIST-2 protocol [22,70]. A total of 19% of patients in the BAT arm received only watchful waiting without any active therapy [70]. Patients received pacritinib 200 mg BID and BAT for a median duration of 25 and 21 weeks, respectively. A total of 51% of patients in the BAT arm crossed over to receive pacritinib at or after Week 24 [70].

The percentage of patients in the pooled pacritinib arms (200 mg BID and 400 mg QD) who achieved ≥35% SVR at Week 24 was significantly higher than for BAT (18% vs 3%, P = 0.001) [70]. Similarly, 25% vs 14% of patients in these arms had a ≥ 50% reduction in TSS (P = 0.08) [70]. The TSS response was also recently reassessed using mTSS, which excluded tiredness from the response analysis, to better align with standards in the treatment landscape [91]. In this retrospective analysis, the pooled pacritinib arms performed significantly better than BAT in the percentage of patients achieving a symptom response (31% vs 14%, P = 0.008) [91]. Pacritinib 200 mg BID demonstrated significant and durable reduction in splenomegaly and symptom control vs BAT. At Week 24, 22% of patients receiving pacritinib 200 mg BID vs 3% of those receiving BAT achieved ≥35% SVR (P = 0.001, efficacy ITT population) (Table 3) [70]. Similarly, 32% vs 14% of patients in these arms had a ≥50% reduction in TSS (P = 0.01), corresponding to median percentage changes of −41% and −15% from baseline [70]. Treatment with pacritinib 200 mg BID led to ≥50% reduction in mTSS in 35% of patients (vs 14% of those receiving BAT; P = 0.004) [91], confirming symptom improvement irrespective of the TSS version. In patients with severe thrombocytopenia at baseline, no significant improvement or worsening in platelet counts was observed with either arm at Week 24 [70,92]. However, among patients who were not RBC transfusion-independent (defined by the Gale criteria [93]) at baseline, a larger proportion of those in the pacritinib 200 mg BID arm (22%) experienced a reduction in RBC transfusion burden than those in the BAT arm (9%) [70]. RBC transfusion independency and dependency were defined as no RBC transfusions or an average of ≥2 RBC units/month, respectively, within ≥3 months prior; reduced RBC transfusion dependence was defined as a ≥50% decrease in the average number of monthly transfusions over the previous three months [92,93]. Using Patient Global Impression of Change (PGIC) scoring, patients in the pacritinib 200 mg BID arm were more likely to self-score as ‘much improved’ or ‘very much improved’ than those receiving BAT (57% and 28%, respectively) [70].

A retrospective head-to-head comparison between pacritinib 200 mg BID (n = 57) and ruxolitinib (n = 12) focusing on patients in PERSIST-2 who had not been treated with ruxolitinib previously (ie, ruxolitinib-naïve subset) was recently conducted [94]. Based on this analysis, most patients receiving pacritinib maintained full dose intensity throughout the study (median total daily dose 400 mg). In contrast, ruxolitinib was administered at low doses at baseline (median total daily dose 10 mg [interquartile range (IQR): 10, 10 mg]), as well as at Week 12 (10 mg [IQR: 0, 10 mg]) and Week 24 (10 mg [IQR: 0, 20 mg]) [94]. Pacritinib treatment vs ruxolitinib was associated with higher numerical rates of ≥35% SVR (28% vs 11%) and mTSS response (37% vs 11%). Rates of cytopenia AEs on study were similar between pacritinib and ruxolitinib, likely reflecting low potential for myelosuppression for either full-dose pacritinib 200 mg BID or for heavily dose-reduced ruxolitinib 5 mg BID [94].

4.3.1. Efficacy analyses on pooled data

In an analysis of patients from PERSIST-1 and PERSIST-2 by JAK2^V617F allele burden quartile, pacritinib treatment led to superior spleen and symptom burden reduction vs BAT in patients regardless of allele burden quartile, including patients with wild-type JAK2 or low (<50%) JAK2^V617F allele burden [95]. For comparison, in a single-center analysis of 69 patients treated with ruxolitinib, spleen response was 5.5-fold lower in patients with low vs high allele burden (<50% vs ≥50%), even after controlling for other disease features and ruxolitinib dose [96].

In another retrospective analysis of data from PERSIST-1 and PERSIST-2, treatment with pacritinib was associated with improved SVR rates and symptom response compared with BAT in patients with MF and severe thrombocytopenia, irrespective of prior JAK2 inhibitor exposure or MF subtype (primary or secondary) [97].

4.4. PAC203

PAC203 was a phase 2 dose-finding study of pacritinib in adult patients with intermediate- and high-risk MF who are intolerant of or resistant to ruxolitinib [85]. Patients were randomized 1:1:1 to receive pacritinib 200 mg BID, 100 mg BID, or 100 mg QD, and stratified by baseline platelet count and geographic region [85]. The primary objective was to confirm the recommended dose of pacritinib. Dose-response relationship for efficacy and safety of pacritinib was further evaluated as a secondary objective [85]. As discussed above, risk mitigation measures were also implemented in PAC203, including enhanced eligibility criteria, patient monitoring, and dose modifications, following the removal of the clinical hold [85].

A total of 161 patients were enrolled in PAC203, 54 in the 200 mg BID, 55 in the 100 mg BID, and 52 in the 100 mg QD arms. In the pacritinib 200 mg BID arm, 44% of patients had severe thrombocytopenia (median platelet count 59 × 10⁹/L) at baseline, 78% had intermediate-2 or high-risk disease, and 39% were RBC transfusion-dependent [85]. In this arm, prior ruxolitinib failure, ruxolitinib intolerance, or both was reported in 78%, 70%, and 48% of patients, respectively. Patients received pacritinib 200 mg BID for a median duration of 21 weeks.

Results from this dose-finding study, as well as pooled exposure-efficacy and -safety analyses, confirmed the efficacy and tolerability of pacritinib 200 mg BID in patients with intermediate- and high-risk MF [85]. The proportion of patients achieving ≥35% SVR at Week 24 in the overall ITT population was highest with pacritinib 200 mg BID (9% vs 2% or 0% for 100 mg BID or QD, respectively) (Table 3). Among evaluable patients, pacritinib 200 mg BID was associated with an SVR rate of 19% (vs 4% or 0%), and among evaluable patients with severe thrombocytopenia, 31% of patients (vs 0% or 0%) achieved ≥35% SVR [85]. There were no significant differences in TSS responses in the ITT population; however, the greatest median reduction in TSS occurred on the 200 mg BID arm (−27%) compared to lower or intermediate doses (−3% and −16%, respectively) [85]. Among evaluable patients with severe thrombocytopenia, 15% of those receiving 200 mg BID (vs 0% and 22%) achieved ≥50% reduction in TSS [85], though differences between intermediate and high-dose arms should be interpreted with caution due to smaller sample size in this subpopulation.

5. Safety of pacritinib

Across the trials, pacritinib has demonstrated an acceptable safety profile in patients with MF, regardless of platelet count [70,75,85]. Most patients experience hematologic stability and low-grade GI toxicities, with diarrhea and nausea representing the most common nonhematologic AEs [70,75,85]. For pacritinib to BAT comparisons overall, it is important to note that the time at risk varied greatly between treatment arms in PERSIST-1 and PERSIST-2 due to crossover to pacritinib; AEs with pacritinib were recorded throughout treatment, whereas in the BAT arm, AEs were recorded until crossover, leading to shorter treatment durations [70,75,90,92].

The most common AEs reported with pacritinib 200 mg BID in PERSIST-2 were diarrhea (48%; 4% grade 3 or 4), thrombocytopenia (34%; 32% grade 3 or 4), nausea (32%; 1% grade 3 or 4), and anemia (24%; 22% grade 3 or 4) [27]. Diarrhea (typically grade 2) most often occurred during Weeks 1 to 8 of treatment, was manageable with standard antidiarrhea agents, and resolved within one to two weeks [70]. Other common grade 3 or 4 AEs included neutropenia (7%), pneumonia (7%), epistaxis (5%), and fatigue (3%) [70]. Serious AEs were reported in 47% of pacritinib-treated patients, most commonly anemia (8%), thrombocytopenia (6%), and pneumonia (6%) [27,70].

A potential for increased risk of grade 3 or 4 bleeding events (pooled per standardized Medical Dictionary for Regulatory Activities query; 14% with pacritinib 200 mg BID vs 7% with BAT), independent of platelet count and rarely leading to discontinuation, was initially noted in PERSIST-2 [70]. However, when adjusting for time-at-risk (event rates per 100 patient-years), rates of overall AEs, fatal AEs, all bleeding AEs, all cardiac AEs, major cardiac events, infections, thromboses, and secondary malignancies were comparable between the pacritinib 200 mg BID (n = 106) arm, pooled from PERSIST-2 and PAC203, and the BAT arm (n = 98), including ruxolitinib (n = 44), in PERSIST-2 [98].

For the subset of patients in PERSIST-2 not previously treated with ruxolitinib, the rates of common AEs, except for diarrhea, were similar with pacritinib 200 mg BID and low-dose ruxolitinib (e.g. thrombocytopenia 33% vs 33%, anemia 30% vs 25%, and grade 3 or 4 bleeding 19% vs 17%, respectively); fatal AEs were more common with ruxolitinib (25%) than with pacritinib (7%) [94].

In PAC203, treatment with pacritinib 200 mg BID was not associated with an increased risk of grade 3 or 4 cardiac and bleeding events, experienced by 4% and 6% of patients in this treatment arm, respectively (vs 6% and 0% for pacritinib 100 mg BID and 6% and 8% for pacritinib 100 mg QD), and the results confirmed the selection of dose and schedule for further clinical development [85]. Rates of AEs with pacritinib 200 mg BID in PAC203 were generally lower than those reported with pacritinib 200 mg BID or BAT in PERSIST-2 (e.g. grade 3 or 4 bleeding 6% vs 14% and 7%; grade 3 or 4 cardiac events 4% vs 7% and 9%, respectively), likely due to enhanced patient selection, patient monitoring, and dose modification guidelines [70,85]. Rates of diarrhea were also lower with 200 mg BID in PAC203 (30%; 6% grade 3 or 4) vs PERSIST-2 (48%; 4% grade 3 or 4) [70,85], a potential consequence of improved education efforts and more efficient management and prevention of diarrhea since there were no changes in recommendations regarding the use of anti-diarrhea medication.

In a retrospective safety analysis of pooled data from PERSIST-2 and PAC203, including patients with MF and severe thrombocytopenia, the safety profiles of pacritinib 200 mg BID (n = 71) and BAT (n = 42) were comparable, confirming the tolerability of fully dosed pacritinib in this patient population [99].

Safety data on the approved pacritinib dose of 200 mg BID from PERSIST-2 and PAC203 are presented in Table 4. Safety comparison between pacritinib 200 mg BID and the approved ruxolitinib and fedratinib doses is presented in Table 5, highlighting the lower rates of hematologic AEs associated with pacritinib even in patients with more profound thrombocytopenia.

Table 4. Summary of pacritinib 200 mg Bid safety data from phase 2 and 3 trials.

AE	PERSIST-2 Pacritinib 200 mg BID n = 106 [70]	PAC203 Pacritinib 200 mg BID n = 54 [85]
Discontinuation due to AEs, n (%)	16 (15)	10 (19)
Any AE^a, n (%)	100 (94)	NR
Diarrhea	51 (48)	16 (30)
Thrombocytopenia	36 (34)	22 (41)^b
Nausea	34 (32)	15 (28)
Anemia	25 (24)	13 (24)
Peripheral edema	21 (20)	9 (17)
Vomiting	20 (19)	NR
Fatigue	18 (17)	13 (24)
Dizziness	16 (15)	NR
Abdominal pain	10 (9)	13 (24)
Decreased appetite	NR	10 (19)
Constipation	NR	10 (19)
Any grade 3/4 AE^c, n (%)	74 (70)	NR
Thrombocytopenia	34 (32)	18 (33)^b
Anemia	23 (22)	11 (20)
Pneumonia	7 (7)	5 (9)
Neutropenia	7 (7)	3 (6)^d
Diarrhea	4 (4)	3 (6)
Epistaxis	5 (5)	NR
Abdominal pain	NR	3 (6)
Hyperuricemia	NR	3 (6)
AEs of special interest, n (%)
Any cardiac event, n (%)	34 (32)	22 (41)
Grade 3/4 AE	7 (7)	2 (4)
Any bleeding event, n (%)	45 (42)	23 (43)
Grade 3/4 AE	15 (14)	3 (6)

^aOccurring in ≥15% of patients receiving pacritinib 200 mg BID in either study; ^bIncludes platelet count decrease; ^cOccurring in ≥5% of patients receiving pacritinib 200 mg BID in either study; ^dIncludes neutrophil count decrease. AE: adverse event; BID: twice daily; NR: not reported.

Table 5. Comparison of most common AEs observed with currently approved JAK inhibitors.

Compound, approved dose, and data source	Trial	Most common hematologic AEs (≥15% of patients; any grade)	Most common nonhematologic AEs (≥15% of patients; any grade)	Source
Pacritinib 200 mg BID	PERSIST-2 (n = 106)	Thrombocytopenia 34%, anemia 24%	Diarrhea 48%, nausea 32%, peripheral edema 20%, vomiting 19%, dizziness 15%, pyrexia 15%	[27,70]
Ruxolitinib 5–20 mg BID^a	COMFORT-I (n = 155)	Anemia 96%, thrombocytopenia 70%, neutropenia 19%	Fatigue 25%, diarrhea 23%, bruising^b 23%, peripheral edema 19%, dyspnea 17%, dizziness^c 18%, nausea 15%, headache 15%	[22,34]
Fedratinib 400 mg QD	JAKARTA (n = 96)	Anemia 99%, thrombocytopenia 63%, lymphopenia 57%, leukopenia 47%, neutropenia 28%	Diarrhea 66%, nausea 64%, vomiting 42%, fatigue or asthenia 16%, abdominal pain 15%	[47]

^aDosing is based on platelet count.

^bIncludes contusion, ecchymosis, hematoma, injection site hematoma, periorbital hematoma, vessel puncture site hematoma, increased tendency to bruise, petechiae, purpura.

^cIncludes dizziness, postural dizziness, vertigo, balance disorder, Meniere’s disease, and labyrinthitis.

AE: adverse event; BID: twice daily; JAK: Janus kinase; QD: once daily.

6. Conclusion

Severe thrombocytopenia is a common feature of MF and associated with shortened survival [12]; yet, until very recently, treatment options in this patient population were extremely limited as ruxolitinib and fedratinib, the standard of care JAK inhibitors in the treatment of MF, are not indicated for use in patients with platelet counts <50 × 10⁹/L [20,22,23]. Pacritinib is a potent kinase inhibitor with a high specificity for both JAK2 and IRAK1, but not JAK1, and has potential to exert clinical activity via inhibiting two distinct pathways, JAK/STAT and TLR/Myddosome/IRAK1 [28,69,77]. In clinical trials, pacritinib has demonstrated significant and durable reduction in splenomegaly and improved symptom control in comparison with BAT in patients with MF, including those with severe thrombocytopenia, and has an acceptable safety profile in patients with MF regardless of platelet count [70,75,85]. As of 28 February 2022, pacritinib is approved for the treatment of patients with MF and severe thrombocytopenia and fills the unmet need in this patient population [1,20,27].

7. Expert opinion

The phase 2 and 3 data for pacritinib, including significant SVR and improvement in symptom control, clearly demonstrate that pacritinib has the potential to expand the current benefits of JAK2 inhibition to a greater proportion of patients with MF – those limited by thrombocytopenia – in either the JAK2 inhibitor-naïve or -exposed setting. The pacritinib kinome profile reveals IRAK1 as a novel, relevant, and previously underappreciated target in MF [69,86]. IRAK1 inhibition likely provides two benefits: 1) reduction of NFκB signaling through upstream targets such as alarmins and IL-1, and 2) less reliance on JAK2 inhibition since the IRAK1 pathway is an independent driver of MF disease course [69]. Pacritinib ameliorates the disease via two distinct pathways and, combined with the negligible JAK1 inhibition by pacritinib, ultimately results in potent anti-inflammatory and clinical activity with lower levels of myelosuppression [28,69,77].

Based on the pacritinib 200 mg BID PERSIST-2 data, I would favor the use of pacritinib in patients with platelet counts ≤100 × 10⁹/L. Given the PAC203 data supporting second-line pacritinib use irrespective of the platelet count, this would also be an alternative NCCN-endorsed option to fedratinib with obvious consideration in the setting of thrombocytopenia [20,85]. I would not recommend pacritinib dose reductions from 200 mg BID for baseline thrombocytopenia at treatment initiation. Instead, I would monitor complete blood count weekly for the first month and every 2 weeks for the second month, and then tailor the process based on the trajectory of the platelet count. I would transfuse platelets in the first four to 6 weeks before deciding to hold the dose or reduce to 100 mg BID. Based on my clinical experience, when switching a patient from >10 mg BID ruxolitinib, I would recommend titrating down to 10 mg BID first and then switching to pacritinib at the next dosing time. Patients receiving ruxolitinib ≤10 mg BID can be switched to pacritinib immediately. I would also switch from fedratinib 400 mg QD to pacritinib without titration and without washout. However, since pacritinib has a long half-life and can take days to build up to steady state [27], patients switching to pacritinib from another JAK inhibitor should be advised that their symptoms may increase over the first week on pacritinib. Closer monitoring and attention to GI toxicity and blood count monitoring are still advised when sequencing JAK2 inhibitors, and I would not recommend a washout period or overlaps between the agents.

Despite the manageable and consistent safety profile of pacritinib, it is essential that prescribers are aware of the potential GI toxicity observed with pacritinib and that they inform patients accordingly before initiating treatment. Pacritinib-related GI toxicities are easy to manage with antiemetics and antidiarrheal medication (e.g. loperamide) and are rarely a reason for discontinuation – except in instances when they occur unexpectedly soon after treatment initiation and alarm the patient. Early intervention upon the first signs of GI toxicity is key to successfully managing GI AEs and potentially preventing them from worsening. In practice, I would either advise prophylaxis with an antidiarrheal medication or use of such a medication at the first sign of any change in stool consistency or frequency, ideally prior to the onset of frank diarrhea.

Moving forward, and as new JAK inhibitors enter the MF space, a driver mutation analysis of PACIFICA study data would be of high interest and somewhat provocative, as it might inform the potential to guide JAK inhibitor selection based on molecular biomarkers. For example, identifying the most appropriate therapy based on the MF phenotype (proliferative vs cytopenic) or JAK2 allele burden (high vs low or JAK2 wild-type) could lead to improved clinical outcomes. Also, if momelotinib gains approval, it will be interesting to see which cytopenia type (anemia or thrombocytopenia) will drive decision-making in clinical practice. In most cases of concurrent anemia and thrombocytopenia, thrombocytopenia tends to be prioritized over anemia. However, in the PERSIST trials, pacritinib treatment led to significant RBC transfusion independency and reductions in the RBC transfusion burden and was also associated with improvements in tiredness/fatigue [70,75], suggesting a potential for anemia-specific activity and symptom control in patients with anemia.

The potential to combine pacritinib with other rational therapies, such as bromodomain and extra-terminal (BET) protein inhibitors and mouse double minute 2 homolog (MDM2) inhibitors [100,101], represents an important avenue to be explored, as pacritinib would likely allow for a less myelosuppressive partner in these combinations. In addition, the potential of pacritinib to be a rational therapeutic consideration outside of MF warrants evaluation in related myeloid malignancies such as chronic myelomonocytic leukemia and myelodysplastic syndrome (MDS)/MPN overlap syndromes.

Article highlights

• Myelofibrosis (MF) is a rare myeloproliferative neoplasm with a complex symptom profile, cytopenias, splenomegaly, and potential for leukemic progression

• Although severe thrombocytopenia is common in MF and correlates with poor prognosis, treatment options for this patient population have been limited since they are not eligible for the standard of care Janus kinase (JAK) inhibitors

• Pacritinib is a potent kinase inhibitor with a high specificity to JAK2 and interleukin-1 receptor-associated kinase 1 (IRAK1) that exerts clinical activity in MF via inhibiting two distinct pathways, JAK/signal transducer and activator of transcription and toll-like receptor/Myddosome/IRAK1

• Across phase 2 and 3 trials, pacritinib demonstrated significant and durable reduction in splenomegaly and improved symptom control in comparison with best available therapy and had a manageable safety profile in patients with MF regardless of the severity of thrombocytopenia

• As of 28 February 2022, pacritinib is approved by the US Food and Drug Administration for the treatment of patients with MF and severe thrombocytopenia

• Gastrointestinal toxicities can occur with pacritinib but are easy to manage; in practice, I would recommend prophylaxis with an antidiarrheal medication or use of such a medication at the first sign of any change in stool consistency or frequency, ideally prior to the onset of frank diarrhea

• The phase 3 PACIFICA trial assessing pacritinib vs physician’s choice of therapy in patients with MF and severe thrombocytopenia is currently ongoing and enrolling globally

• Future studies will further examine the clinical benefit of pacritinib in MF, either as monotherapy or in combination with emerging therapeutic options, as well as in other related disease states

Declaration of interests

J Mascarenhas receives research funding paid to the institution from Incyte, Novartis, Roche, Merck, Geron, CTI BioPharma, Bristol Myers Squibb, AbbVie, Celgene, Kartos, and PharmaEssentia and consulting fees from Incyte, CTI BioPharma, Constellation, Geron, Kartos, BMS, Celgene, Novartis, Karyopharm, Sierra Oncology, AbbVie, PharmaEssentia, and Galecto.

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.

Reviewer disclosures

A peer reviewer on this manuscript has received research support and honoraria for advisory board and satellite symposium participation from CTI Biopharma, the manufacturer of pacritinib. Peer reviewers on this manuscript have no other relevant financial relationships or otherwise to disclose. CTI Biopharma provided a scientific accuracy review at the request of the journal editor.

Acknowledgments

Medical writing support in line with Good Publication Practice guidelines was provided by Katri Niemi, on behalf of Nexus GG Science LLC and funded by CTI BioPharma.

Additional information

Funding

This paper was supported by CTI BioPharma.