Clinical response to ruxolitinib in CSF3R T618-mutated chronic neutrophilic leukemia

Dear Editor,

Chronic neutrophilic leukemia (CNL) is a rare myeloproliferative neoplasm (MPN) that includes only about 200 patients described to date meeting the World Health Organization (WHO) criteria and the recently reported CSF3R T618I mutation. Diagnosis of CNL using the World Health Organization 2008 criteria[1] requires a WBC count of ≥25 × 10⁹/L, ≥80 % segmented/band neutrophils, absence of dysgranulopoiesis, exclusion of genetic drivers that are known to occur in others MPNs (such as BCR-ABL1, PDGFRA/B, or FGFR1 rearrangements), and <10 % immature myeloid cells [2]. Previously, CNL was often confused with chronic myeloid leukemia (CML), atypical CML (aCML), or chronic myelomonocytic leukemia (CMML); however, this changed with the identification of oncogenic mutations in the granulocyte colony-stimulating 3 factor receptor (CSF3R) gene in approximately 83 % of WHO-defined CNL patients [3]. CSF3R mutations in CNL are either nonsense or frameshift mutations truncating the cytoplasmic tail (truncation mutations) or point mutations in the extracellular domain (membrane proximal mutations) [3, 4]. Truncation mutations are sensitive to dasatinib in vitro, while membrane proximal mutations show ligand-independent signaling through the JAKSTAT pathway and cells harboring this mutation are sensitive to Jak1/2 inhibitors in vitro. Additional observations from Mayo Clinic showed that concomitant SETBP1 and ASXL1 mutations in CNL patients with a CSF3R mutation help to further distinguish CNL from aCML and are associated with poorer outcomes [5]. There is as yet limited clinical experience with ruxolitinib in patients with CNL, given the low prevalence of the disease [6, 7]. After the initial case reported by Maxson et al., Dao et al. presented the case of another patient with CNL and a CSF3RT618I mutation who experienced reduction in hepatosplenomegaly and improvement in blood counts and constitutional symptoms with ruxolitinib, but there was no reduction in CSF3R T618I allele frequency or overall bone marrow cellularity after 3 months of treatment [6]. This argues for the inability of ruxolitinib to cytoreduce the malignant clone in the short-term. Dao et al. described a double-mutated CNL patient (CSF3RT618I and SETBP1 D868N) who was refractory to the treatment with both ruxolitinib and hydroxyurea both in vitro and in vivo [7]. The authors speculated that the co-expression of SETBP1 in these CSF3R T618I mutations in a patient with CNL might have contributed to the ineffectiveness of JAK inhibitor therapy in CNL. We report the case of a patient with a CSF3R and SETBP1 mutation, who responded to ruxolitinib, at least for a brief period of time.

Our patient is a 76-year-old man, who presented with a new sharp intermittent left upper quadrant (LUQ) abdominal pain and fatigue without other constitutional symptoms. Complete blood count (CBC) showed a white blood cell count (WBC) of 147 × 10⁹/L, hemoglobin (Hb) 9.7 g/dL, MCV 104, and platelet count 235 × 10⁹/L. The differential count included 73 % neutrophils, 4 % lymphocytes, 1 % monocytes, 1 % myelocytes, 5 % promyelocytes, and 2 % blasts. Blood smear showed dysgranulopoiesis (Fig. 1a). A CBC done 7 months earlier was normal. Examination demonstrated mild LUQ tenderness without organomegaly or lymphadenopathy. Uric acid and LDH were elevated while liver enzymes and kidney function were normal. Computed tomography (CT) scan of the abdomen and pelvis showed hepatomegaly (18.4 cm) and splenomegaly 15.5 cm), which was absent on prior CT in 2012, as well as a wedge-shaped splenic hypodensity possibly representing a splenic infarct (Fig. 2a, b, c). Marrow biopsy was 100 % cellular with profound erythropoietic hypoplasia, mild myeloid dysplasia, and diffused interstitial reticulin fibrosis (Fig. 1b). No aspirate could be obtained. Flow cytometric studies of blood confirmed the majority of cells being mature neutrophils and less than 1 % of cells being CD34+ CD117+ myeloblasts. There was no increase in monocytes and no evidence for monoclonal lymphocytosis.

Fig. 1.

a Wright-Giemsa-stained blood smear (x400) demonstrating dysgranulopoiesis. b Bone marrow biopsy with myeloid hyperplasia and multilineage dysplasia (x100)

Fig. 2.

Pre- and post-ruxolitinib therapy contrast-enhanced CT scans. a Axial view pre- and post-treatment. b Coronal view splenomegaly preand post-treatment. c Coronal view hepatomegaly pre- and post-treatment

The patient was treated with hydroxyurea while awaiting results of florescence in situ hybridization (FISH) and polymerase chain reaction (PCR) studies for t(9; 22) and BCR-ABL fusion, respectively, which were negative. Supportive care with rasburicase, allopurinol, and fluids was administered. FISH showed three copies of chromosome 8q in 10 % of the bone marrow cells. Molecular profiling demonstrated no mutations of CEBPA, NPM1, FLT3, KIT, JAK2, PDFGR, FGR1, or CALR. CSF3R T618I mutation was present, suggestive of a diagnosis of chronic neutrophilic leukemia (CNL)[5]; SETBP1 (SETBP1 G870S) and ASXL1 (ASXL1 R404 and Y700) mutations, both of which have been reported to be associated with worse outcomes and resistance to JAK2 inhibitor therapy in patients with CNL, were also present [7–9]. The patient declined allogeneic stem cell transplant. Ruxolitinib was started at 5 mg twice daily and the dose increased in 5-mg increments to 20 mg twice daily (Fig. 3). On follow-up 3 months after initiation of ruxolitinib therapy, symptoms resolved, the patient was able to resume strenuous physical activity, and the CBC markedly improved (Fig. 3). Additionally, he achieved a dramatic reduction in spleen size (11.6 cm in length from previously 15.5 cm) and moderate reduction in liver size (17.6 cm in length from previously 18.4 cm) (Fig. 2 a, b, c). Five months after initiating ruxolitinib, he lost response with increasing WBC and worsening platelet count and Hb, so hydoxyurea was added with stabilization of his WBC count (Fig. 3).

Fig. 3.

WBC, ANC, Hb, and platelet count upon presentation and during treatment with hydroxyurea and ruxolitinib

The response in our case suggests that patients with SETBP1 mutations may respond to ruxolitinib, at least for a brief period of time. An ongoing small clinical trial is evaluating the impact of ruxolitinib on natural history of CNL (Clinicaltrial.gov NCT02092324). As ruxolitinib therapy has not been shown to change the natural history of CNL, ASCT should be considered for transplant-eligible patients with CNL.