Skip to Main Content


Skip Nav Destination

You Make the Call: How would you treat a patient with an underlying bleeding disorder and CSF3R-mutated MDS/MPN-U?

December 2, 2022

December 2022

David Steensma, MDDavid Steensma, MD
Global Hematology Head
Novartis Institutes for BioMedical Research
Cambridge, Massachusetts




A 78-year-old male was found to have a leukocytosis with left shift (76% neutrophils, 9% bands, 1.7% monocytes, 1% eosinophils, 6% metamyelocytes, 7% myelocytes, 3% promyelocytes), first noted during an admission for atrial fibrillation with rapid ventricular response. His white blood cell (WBC) count was reported to be 300 × 109/L. After undergoing a bone marrow biopsy, he was started on hydroxyurea (HU) and discharged to a rehabilitation center. About a month later, he was hospitalized for a large thigh hematoma, but no intervention was taken. He was found to be anemic and required red blood cell (RBC) transfusion. His bone marrow results came back as myelodysplastic syndrome/myeloproliferative neoplasm-unclassifiable (MDS/MPN-U). The MDS/MPN-U has not been treated, and his WBC count is stable around 100 × 109/L.

The initial report from the biopsy noted that morphology and immunophenotypic features were most consistent with chronic myeloid leukemia (CML). However, subsequent tests were negative for BCR-ABL1. No overt dysplasia in erythroid or myeloid precursors was noted. Megakaryocytes appeared adequate with occasional smaller (relatively hypolobated) forms. Next-generation sequencing (NGS) identified the following variants: ASXL1 (31.4% variant allele frequency [VAF]), CSF3R T640N (40.1% VAF), CSF3R Q776 (14.7% VAF), FLT3 (4% VAF), GATA2 (46% VAF), SETBP1 (51% VAF), SRSF2 (47% VAF), and CUX1 (48% VAF). The metaphase karyotype was normal.

The patient now requires an RBC transfusion every two to three days. He recently presented with worsening thigh pain, was found to have an active bleed on CT angiogram, and underwent embolization. He is being worked up for an underlying bleeding disorder and his platelet counts are normal. Given the CSF3R mutations present in his disease, would you treat this patient with ruxolitinib or a hypomethylating agent (HMA)?


The first question here is: what is the diagnosis? The overall picture, including the NGS results, are typical of MDS/MPN, a group of disorders characterized by overlapping MDS- and MPN-like pathological and clinical features.1 Importantly, CML has been ruled out with BCR-ABL1 testing. The case does not fit any of the classic BCR-ABL1-negative MPNs, and the three classical MPN-associated driver mutations (i.e., JAK2, CALR, MPL) are absent. Absolute monocytosis is present due to the extreme leukocytosis, but there is not enough relative monocytosis (≥10% blood monocytes) to meet diagnostic criteria for chronic myelomonocytic leukemia (CMML). There is no eosinophilia, so chronic eosinophilic leukemia is ruled out, and there is too much immaturity in the myeloid series to fit with chronic neutrophilic leukemia (CNL), which is characterized by at least 80% neutrophils and bands and less than 10% earlier myeloid series cells in the blood.

Therefore, I think this case fits best with what the 2016 World Health Organization (WHO) revised fourth edition classification termed “atypical CML” (aCML), given the persistent leukocytosis, greater than 10% circulating neutrophil precursors, lack of basophilia and monocytosis, granulocytic proliferation and dysplasia with absent or minimal erythroid and megakaryocytic dysplasia, and absence of tyrosine kinase fusion genes that would point to another disorder.2 In the fifth edition WHO reclassification proposal published in June 2022, aCML was renamed “MDS/MPN with neutrophilia” to avoid confusion with BCR-ABL1-positive CML.3 In contrast, in the competing International Consensus Classification (ICC) proposal published in Blood by Arber et al., the aCML designation was retained, but minor changes were proposed, such as noting that SETBP1 mutations in association with ASXL1 mutations – as seen in this case – are strongly suggestive of an aCML diagnosis.4 (As an aside, it is problematic to now have two competing hematologic malignancy classifications, but that’s a discussion for another day. Editor’s note: See “Breaking Up Is Hard to Do” for more about this.)

In 2020, Palomo et al. showed how even MDS/MPN-U cases often cluster into molecular profiles that more clearly mimic well-defined phenotypes.5 In that series, ASXL1, SETBP1, GATA2, and SRSF2 mutations fit within the aCML cluster. CSF3R is uncommonly mutated in aCML compared to CNL but is still recurrently mutated in aCML.6 The two CSF3R-activating mutations observed in this patient are less commonly observed in MDS/MPN than CSF3R T618I. In this patient, the CSF3R-mutant cells represent the bulk of the circulating white cells, given the mutations’ high VAFs.

Now that we have assigned a diagnosis of WHO 2016/ICC aCML (also known as WHO 2022 MDS/MPN with neutrophilia), how should this man be treated? He is symptomatic and has developed hematologic complications, so treatment is clearly indicated. Since cytoreduction with HU has failed to adequately control the myeloproliferation, other options would include ruxolitinib, an HMA, or cytotoxic chemotherapy. There’s also limited experience with pegylated alfa interferon.7 Jason Gotlib, MD, from Stanford University published “How I treat atypical CML” in Blood in 2017, which is a helpful guide.8

To date, the largest series of ruxolitinib use in patients with aCML with CSF3R mutations comes from the 2020 multicenter study by Dao et al.9 Among 44 patients, 23 of whom had aCML, 35% responded to ruxolitinib. Many patients were not able to complete six cycles of therapy, responses were almost all partial and transient, and the patients with CNL had better outcomes than those with aCML. Most of the patients who did respond had CSF3R T618I, not other CSF3R variants, so the likelihood this patient would have a durable, deep response to ruxolitinib is low.

I’m not aware of any series of HMA therapy for aCML like the one with ruxolitinib, just anecdotes. My personal experience is that HMAs don’t work very well in this setting. If we extrapolate from HMA series in another MDS/MPN overlap condition, proliferative CMML, a response rate of less than 30% would be expected, and most responses would be partial and not durable.10 However, azacitidine or decitabine is worth considering.

Finally, “old-fashioned” cytotoxic agents such as etoposide, busulfan, 6-mercaptopurine, or thioguanine can be used for palliation. I once was able to manage a patient with highly proliferative aCML with busulfan for a number of months, and the Mayo Clinic’s MPN group led by Ayalew Tefferi, MD, has published their experience with this approach.11 New treatments are definitely needed.

The bleeding event in this patient is worrisome. With a normal platelet count, something like acquired von Willebrand disease, which we can see with extreme thrombocytosis, would not be expected. It is possible that instead there is platelet dysfunction as part of the myeloid clonal process.12 It would be best to rule out an acquired coagulation factor inhibitor or another hemostatic disorder.

Disclosure: Dr. Steensma works at a biomedical research institute owned by Novartis. No other conflicts of interest were reported.


  1. Thota S, Gerds AT. Myelodysplastic and myeloproliferative neoplasms: updates on the overlap syndromes. Leuk Lymphoma. 2018;59(4):803-812.
  2. Arber DA, Orazi A, Hasserjian R, et al. The 2016 revision to the World Health Organization classification of myeloid neoplasms and acute leukemia. Blood. 2016;127(20):2391-2405.
  3. Khoury JD, Solary E, Abla O, et al. The 5th edition of the World Health Organization Classification of Haematolymphoid Tumours: Myeloid and Histiocytic/Dendritic Neoplasms. Leukemia. 2022;36(7):1703-1719.
  4. Arber DA, Orazi A, Hasserjian RP, et al. International Consensus Classification of Myeloid Neoplasms and Acute Leukemias: integrating morphologic, clinical, and genomic data. Blood. 2022;140(11):1200-1228.
  5. Palomo L, Meggendorfer M, Hutter S, et al. Molecular landscape and clonal architecture of adult myelodysplastic/myeloproliferative neoplasms. Blood. 2020;136(16):1851-1862.
  6. Maxson JE, Gotlib J, Pollyea DA, et al. Oncogenic CSF3R mutations in chronic neutrophilic leukemia and atypical CML. N Engl J Med. 2013;368(19):1781-1790.
  7. Jabbour E, Kantarjian H, Cortes J, et al. PEG-IFN-alpha-2b therapy in BCR-ABL-negative myeloproliferative disorders: final result of a phase 2 study. Cancer. 2007;110(9):2012-2018.
  8. Gotlib J. How I treat atypical chronic myeloid leukemia. Blood. 2017;129(7):838-845.
  9. Dao KT, Gotlib J, Deininger MMN, et al. Efficacy of ruxolitinib in patients with chronic neutrophilic leukemia and atypical chronic myeloid leukemia. J Clin Oncol. 2020;38(10):1006-1018.
  10. Coston T, Pophali P, Vallapureddy R, et al. Suboptimal response rates to hypomethylating agent therapy in chronic myelomonocytic leukemia; a single institutional study of 121 patients. Am J Hematol. 2019;94(7):767-779.
  11. Begna K, Abdelatif A, Schwager S, et al. Busulfan for the treatment of myeloproliferative neoplasms: the Mayo Clinic experience. Blood Cancer J. 2016;6:e427.
  12. Jones E, Dillon B, Swan D, Thachil J. Practical management of the haemorrhagic complications of myeloproliferative neoplasms. Br J Haematol. 2022;199(3):313-321.


A 49-year-old female received her second dose of the Moderna COVID-19 vaccine, and later that night, she experienced a sudden-onset and very severe headache. She had a prior history of migraine headaches, so she took some migraine medication she had available at home. The next day, her neurologist advised that she take sumatriptan; however, her headaches persisted. A few days later, she went to the emergency room, was given additional medication for her headaches, and was discharged.

Her headaches continued, and she returned to the emergency room. On imaging, she was found to have a thrombosis of her superior sagittal sinus, right and left transverse sinuses, and thrombus into the right sigmoid sinus. She was initially anticoagulated with heparin, then transitioned to rivaroxaban.

She also has a long history of ulcerative colitis (UC) that has been treated with prednisone and mesalamine. Her UC has been in remission, and she has been off UC treatments for the past five months. She currently does not have any gastrointestinal symptoms. She has normal hemoglobin levels and no iron deficiency.

I usually stop anticoagulation in COVID vaccine-associated cerebral sinus thromboses after 18 months to two years. However, patients with UC have a risk for spontaneous cerebral sinus thromboses. Can I stop her rivaroxaban at this time, or does she need to continue long-term rivaroxaban? If so, would you recommend a 10 mg or 15 mg dose?

How would you respond? Email us at

Disclaimer: ASH does not recommend or endorse any specific tests, physicians, products, procedures, or opinions, and disclaims any representation, warranty, or guaranty as to the same. Reliance on any information provided in this article is solely at your own risk.


Connect with us:

May 2024


Close Modal

or Create an Account

Close Modal
Close Modal