Selinexor Shows Modest Spleen Response in JAK Inhibitor–Refractory MF

Once a Janus kinase (JAK) inhibitor stops working in myelofibrosis, few options remain that can shrink an enlarged spleen without pushing blood counts lower. Single-agent selinexor, an oral inhibitor of exportin-1 (XPO1), produced a 35% or greater spleen volume reduction (SVR₃₅) at week 24 in 17% of heavily pretreated patients, and it did so with no meaningful myelosuppression, according to phase 2 results published in Blood Cancer Journal.¹

Median overall survival reached 35.6 months (95% CI, 23.9 months to undefined), and median progression-free survival was 28 months (95% CI, 4.4 months to undefined). The activity was modest, but it came in a population with little left to try.

Why Myelofibrosis Needs Options Beyond JAK Inhibition

Aberrant JAK/STAT signaling driven by mutations in JAK2, CALR, or MPL sits at the center of myelofibrosis pathogenesis, and JAK inhibitors have been the standard of care since ruxolitinib was approved. Ruxolitinib reduces spleen volume and symptoms in about 40% of patients, but many of those responses are transient; myelosuppression frequently forces dose reductions that compromise symptom control, and the malignant clone survives.

The consequences of losing that option are steep. Survival after ruxolitinib discontinuation has ranged from 11 to 16 months across several studies.² Allogeneic stem cell transplant remains the only potentially curative approach, but it carries substantial morbidity and mortality, and many patients are ineligible.

Single-Center Trial in Patients Who Had Already Failed Ruxolitinib

The ESSENTIAL trial (NCT03627403) enrolled 17 patients at a single center between May 2019 and March 2023.¹ Nine were men and 8 were women, with a median age of 66 years (range, 43-80). Dynamic International Prognostic Scoring System risk was low in 1 patient, intermediate-1 in 7, and intermediate-2 in 9. JAK2^V617F was present in 11 patients (65%), CALR mutations in 5 (29%), and MPL^W515L in 1 (6%), and 10 patients (59%) carried at least 1 high-molecular-risk mutation.

All 17 had failed ruxolitinib, and 1 each had also failed fedratinib or pacritinib, after a median 13 months (range, 0.5-96) of prior JAK inhibition. Fifteen patients were resistant to JAK inhibition and 2 were intolerant. Median baseline spleen volume was 1300 cm³ (range, 410-5790), and median total symptom score was 28 (range, 0-100). Selinexor was started at 80 mg (n = 6), 60 mg (n = 6) or 40 mg (n = 5) weekly. The study was terminated by investigator and sponsor decision after low accrual, with the data judged sufficient to detect a response.³

Spleen, Symptom, and Anemia Responses With Selinexor

Of 11 patients evaluable at week 24, 3 reached SVR₃₅ and 5 met a 25% reduction threshold (SVR₂₅).¹ In the intention-to-treat population, the week-24 SVR₃₅ rate was 17%, and the rate at any time was 23%. Symptom score reductions of at least 50% were seen in 11% of that population at week 24 and in 29% at any time; 9 patients were unevaluable. Three patients stayed on selinexor up to 96 weeks with sustained spleen responses.

Blood counts held stable, and no patient discontinued for hematologic toxicity. The most common grade 1/2 treatment-emergent adverse events were nausea, fatigue, anorexia, diarrhea, and weight loss. Bone marrow fibrosis fell by at least 1 grade in 6 of 14 evaluable patients by the end of treatment, while 1 patient progressed. Among 8 patients with non–transfusion-dependent anemia, 3 achieved a major response and 1 achieved a minor response under 2024 International Working Group-European LeukemiaNet criteria. Hepcidin fell during treatment, and the authors reported a borderline significant correlation between that drop and hemoglobin improvement.

“Within the limitations of a small trial with considerable early drop-out, the ESSENTIAL study suggests that selinexor is safe and tolerable in JAKi-resistant MF patients, with modest clinical activity,” the researchers wrote.

Small Sample and Reduced Dosing Limit the Conclusions

The trial was small, single-arm, and marked by early attrition: all 17 patients stopped selinexor, 5 for toxicity and 6 to move to another treatment. Only 37 of 60 planned plasma samples were available for the cytokine analysis, a gap the authors attributed to disruption from the COVID-19 pandemic. Karyopharm Therapeutics, which markets selinexor, funded the trial and employs 2 of the authors.

Dosing was the other constraint. Because selinexor toxicities overlap with myelofibrosis symptoms, patients received less than the 100 to 160 mg weekly approved in myeloma or 120 mg weekly in lymphoma, which the authors said compromised efficacy. None of the patients who started at 40 mg weekly reached SVR₃₅.

Against other second-line options, selinexor landed in a familiar range: week-24 SVR₃₅ and symptom response rates, respectively, were 36% and 34% with fedratinib, 9.3% and 7.4% with pacritinib, and 25% and 9% with momelotinib. For clinicians weighing salvage options in patients whose counts cannot absorb more suppression, tolerability may matter as much as response rate. The phase 1 portion of the SENTRY trial (NCT04562389), which paired selinexor with ruxolitinib in patients naive to JAK inhibition, settled on 60 mg weekly as the phase 3 dose, and the authors said the combination proved superior for spleen volume reduction and survival.

References

1. Tantravahi SK, Ellero AA, Patel AB, et al. Single agent selinexor is active in patients with myelofibrosis refractory or intolerant to JAK inhibitors. Blood Cancer J. 2026;16:120. doi:10.1038/s41408-026-01571-2

2. Mascarenhas JO, Verstovsek S. The clinical dilemma of JAK inhibitor failure in myelofibrosis: predictive characteristics and outcomes. Cancer. 2022;128(14):2717-2727. doi:10.1002/cncr.34222

3. Selinexor in myelofibrosis refractory or intolerant to JAK1/2 inhibitors (ESSENTIAL). ClinicalTrials.gov. Updated May 12, 2026. Accessed August 27, 2026. https://clinicaltrials.gov/study/NCT03627403