Restoration of response to ruxolitinib upon brief withdrawal in two patients with myelofibrosis

In 2005, an activating mutation in the Janus Kinase 2 gene (JAK2 V617F) was found in the majority of patients with primary myelofibrosis, which led to the development of specific JAK inhibitors 1. One of these, ruxolitinib, a selective inhibitor of JAK1 and 2, has recently been shown to improve splenomegaly, disease symptoms, and occasionally disease burden in myelofibrosis 2-4. JAK2 mutant clones were found to persist in the presence of chronic JAK2 kinase inhibition in mouse models 5. This persistence was shown to occur via the trans activation of JAK2 by other JAK kinases rather than the emergence of a compensating JAK2 mutation or an activating mutation in a pathway component downstream. Importantly, this meant that the observed persistence was reversible upon temporary withdrawal of the JAK2 inhibitor, at least in vitro 5. This reversibility suggests that patients who develop resistance to JAK2 inhibitors may also become resensitized upon temporary withdrawal of treatment 6. Here, we describe two case studies that address this hypothesis. The first patient was a 59-year-old female, diagnosed with primary myelofibrosis. Ruxolitinib treatment was initiated with an oral 20 mg dose bid (Fig. 1A). At study entry, the patient had palpable splenomegaly of 19 cm below the costal margin (cm b.c.m, Fig. 1B). After 3 weeks of treatment, an excellent response was observed in terms of a reduction in spleen volume to 8.5 cm b.c.m. (Fig. 1B), and the loss of all constitutional symptoms as well as a 2 kg weight gain. After 6 weeks, the dose was reduced to 10 mg bid due to a decrease in platelet levels to 81 × 109/l and an increase in transfusion frequency caused by the presence of transfusion-dependent anemia which led to an increase of the spleen size (Fig. 1A–C). After continuous treatment of 15 mg bid for around 15 months, the spleen had again increased in size to 20 cm b.c.m, (Fig. 1A,B), and constitutional symptoms started to reappear. To counteract this, the dose was raised to 20 mg bid (Fig. 1A). However the subsequent response was marginal and after monitoring over a further 16 months of treatment, the patient was eventually classified as a non-responder. To prepare for an alternative therapeutic strategy, ruxolitinib treatment was gradually withdrawn over a period of two weeks (Fig. 1A, Supporting Information Appendix Fig. 2). However, this resulted in a dramatic reappearance of more severe constitutional symptoms, as well as an increase in leukocyte and platelet counts, lactate dehydrogenase (LDH) levels, and a spleen size of 25 cm b.c.m (Fig. 1B,C, Supporting Information Appendix Fig. 2). Therefore, after 2 days without treatment, and due to the unavailability of the alternative therapy, ruxolitinib treatment was resumed at a reduced dose of 5 mg bid (Fig. 1A–D, Supporting Information Appendix Fig. 2). Unexpectedly, even with this low dose there was an impressive response, particularly of spleen size and LDH levels. Within 11 days, spleen size was reduced to 14 cm b.c.m. There was also an increase in haemoglobin levels and a decrease in transfusion dependency (Fig. 1B,D, Supporting Information Appendix Fig. 2). These effects lasted around 8 weeks. We also observed similar albeit less pronounced results with a second patient. This patient, a 73-year-old female diagnosed with post-polycythemia vera myelofibrosis, had palpable splenomegaly (15 cm b.c.m) and excessive constitutional symptoms. The patient was treated with a 15 mg bid dose of ruxolitinib and after two months the spleen had reduced to 6 cm b.c.m, and constitutional symptoms had improved. Two months later, hemoglobin levels had decreased to 8.6 g dl−1 and the spleen increased to 11 cm b.c.m. Upon a further increase in spleen size to 16.5 cm b.c.m, the patient was classified as a non-responder and ruxolitinib treatment was withdrawn. Again, upon reintroduction of ruxolitinib, the patient regained response, albeit to a lesser extent than previously seen. Spleen size decreased from 16.5 to 13.0 cm b.c.m, and hemoglobin levels increased from 9.2 to 12.8 g/dl. In conclusion, two patients who initially responded to ruxolitinib but developed resistance regained response upon temporary ruxolitinib withdrawal. This supports previous results in vitro, and demonstrates that intermittent ruxolitinib treatment, at least for some patients, may lead to an enhanced therapeutic response. The article was drafted by Helen Pickersgrill, a professional medical writer funded by Novartis Pharmaceuticals. Heinz Gisslinger,1* Martin Schalling,1 Bettina Gisslinger,1 Cathrin Skrabs,1 Leonhard Müllauer,2 Robert Kralovics1,3 1Department of Hematology, Medical University of Vienna, Vienna, Austria 2Institute of Clinical Pathology, Medical University of Vienna, Vienna, Austria 3CeMM Center of Molecular Medicine of the Austrian Academy of Sciences, Vienna, Austria Additional Supporting Information may be found in the online version of this article. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.