Efficacy and Safety of Erythropoietic-Stimulating Agents with Ruxolitinib in Myelofibrosis Patients : A Retrospective Analysis on 45 Patients. on Behalf of the French Intergroup of Myeloproliferative Disorders (FIM)

Background

Ruxolitinib is a current therapeutic option, which has demonstrated rapid and durable reduction in splenomegaly and improved disease-related symptoms in patients (pts) with primary myelofibrosis (PMF), post-polycythemia vera-MF (PPV-MF), and post-essential thrombocythemia-MF (PET-MF). Anemia is another frequent issue in MF, which may be managed by the use of ESA, leading to a 40-50% response rate in small studies. Consistent with its JAK2 signalling inhibition, ruxolitinib therapy has been shown to be detrimental on the hemoglobin level, increasing the depth of anemia or transfusion need, especially during the first 12-24 weeks of treatment in the COMFORT studies.

Despite potential antagonistic mechanisms of action on JAK2, some responses on anemia have been reported with the addition of ESA to ruxolitinib in a small subset of pts in the COMFORT II study. The present study aimed to better assess the efficacy of ESA on anemia related to ruxolitinib and tolerance of this combination in a larger cohort of pts treated for MF in general practice.

Methods

We performed an observational study on patients with MF previously or currently treated with concomitant ESA and Ruxolitinib in French centers members of the FIM. Informed consent was provided by the pts. Data collected included characteristics of the disease, treatment, responses to ruxolitinib and ESA. They are reported according to the IWG-MRT/ELN 2013 criteria.

Results

This analysis was performed in July 2016, on the 45 first consecutive pts in 11 centers. Median age at diagnosis was 73 (range 42- 89), 30 (67%) were men. Twenty-five pts (56%) had primary MF, 11 (24%) PET-MF and 9 (20%) PPV-MF, overall diagnosed between 2004 and 2016. IPSS risk categories were low/int-1 and int-2/high in 16 (36%) and 28 (64%) pts, respectively. Twenty-nine (64%) were JAK2V617F positive, 5 harbored MPL mutation and 8 had CALR mutations.

Median time between MF diagnosis and ruxolitinib was 21 (0-109) months and median follow-up from ruxolitinib starting was 13 (2 - 53) months. At time of ruxolitinib initiation 32 (71%) pts were transfusion independent and 13(29%) had transfusion need. Ten additional pts became transfusion dependent after ruxolitinib initiation. Other causes of anemia were renal insufficiency n=7, surgery n=1, 1 cytoreductive therapy with hydroxyurea.

Type of ESA were darbepoetin alfa, [n=26]; epoetin alfa, [n=3], epoetin beta [n=8], epoetin zeta [n=4], epoeitin theta [n=4], with a median duration of exposure to ESA of 15 months [1-92mo]. ESA was introduced either before ruxolitinib (n= 17), simultaneously (n= 4) or afterward (n= 24) after a median of 2 months [1-26mo].

Response rate to ruxolitinib were in accordance with previous reports: For splenomegaly, 33 (73%) of pts achieved at least a partial response, 8 (17%) were stable and 4 (9%) were progressive. Thirty pts (67%) had at least partial response on constitutional symptoms.

Response assessment of anemia according to IWG-MRT/ELN 2013 criteria: 7 pts (16%) achieved a RBC transfusion independency, 13 (29%) pts had an increase in hemoglobin level of Hb >2g/dl (2 pts achieved both criteria), which results in 40% of objective responses. The median time to best response on anemia after ESA initiation was 3 [1-84] months.

For safety, a pulmonary embolism occurred in one patient possibly related to ESA, no other adverse event occurred, in particular no spleen enlargement was described.

At time of analysis, 36/45 pts were still alive: 1 underwent allogeneic bone marrow transplant, 34 were still treated with ruxolitinib whereas 28 patients were still undergoing ESA therapy.

Conclusions

This retrospective analysis is the largest cohort describing the use of concomitant ESA with ruxolitinib therapy in "real life". We report 40 % of objective responses, consistent with ESA response rates without ruxolitinib for MF related anemia. Tolerance seemed acceptable without hampering efficiency of ruxolitinib. Our results suggest that ESA should be considered as a possible therapeutic for anemia in myelofibrosis patients treated with ruxolitinib.