The evolving treatment landscape for myelofibrosis is providing greater precision for JAK inhibitor selection, with patient-specific factors such as anemia, thrombocytopenia, and spleen and symptom control influencing treatment decisions. Emerging post hoc analyses suggest that transfusion independence, hemoglobin improvement, and spleen response may be associated with improved survival outcomes. Therefore, individualized treatment selection and dosing considerations are increasingly important when choosing between and switching JAK inhibitors.

It is probably fair to say that all patients with myelofibrosis will need JAK inhibitor therapy at some point. Some patients may initially present only with cytopenias, and, in those instances, other treatment options can be used. However, most patients with myelofibrosis will eventually develop splenomegaly and/or symptoms and will then need a JAK inhibitor.

We are starting to develop more granularity in selecting among the 4 available JAK inhibitors (ie, momelotinib, ruxolitinib, fedratinib, and pacritinib), but there is not always one right answer. There are several factors that may guide you in one direction or another. For example, I think that most clinicians would choose momelotinib for a JAK inhibitor–naive patient with severe anemia. However, if the anemia is mild to moderate, you could potentially consider either ruxolitinib or momelotinib.

We are also learning more about the differences between these drugs. It is my perception that ruxolitinib and fedratinib are the more potent JAK2 inhibitors. However, potency comes at a cost, as more potent JAK2 inhibitors can also contribute to cytopenias. On the other hand, momelotinib and pacritinib can ameliorate cytopenias, or at least not worsen them. Momelotinib also inhibits ACVR1, which helps with anemia, and it is possible that it is not inhibiting JAK2 as potently.

Post hoc analyses have also taught us a lot in myelofibrosis. For example, a post hoc analysis of the SIMPLIFY-1 and SIMPLIFY-2 trials demonstrated that transfusion independence at 24 weeks was an important clinical end point associated with improved survival among those taking momelotinib. Post hoc analyses of the SIMPLIFY-1 and MOMENTUM trials showed that patients with anemia (ie, those with a baseline hemoglobin level of <10 g/dL) who were treated with momelotinib and were able to get their hemoglobin level up to 10 g/dL or higher also had a survival benefit, although this association did not reach statistical significance. Another post hoc analysis of patients randomized to momelotinib in the SIMPLIFY-1 and MOMENTUM trials conducted by Palandri and colleagues showed—perhaps not surprisingly—that the dual response of transfusion independence and spleen volume reduction (SVR) correlated with improved overall survival in these patients. But, interestingly, transfusion independence seemed to be a more important driver of survival benefit than the degree of SVR, as overall survival was improved in patients achieving transfusion independence across SVR thresholds (10%, 15%, 25%, or 35%).

Regarding ruxolitinib, we should remember that it is a dose-dependent drug. If you are not able to deliver the optimal dose, you may not get the results you expect. The Response to Ruxolitinib After 6 Months (RR6) model identified a dose below 20 mg twice daily at baseline and at months 3 and 6 as one of the factors associated with poorer survival. In the SIMPLIFY-1 post hoc analysis evaluating transfusion independence and SVR, patients with baseline anemia (hemoglobin <10 g/dL) who were randomized to ruxolitinib had better rates of spleen response when their baseline platelets were 200 × 10³/μL or higher, whereas momelotinib performed numerically better when platelets were below that threshold. One explanation for this may be that patients with lower platelet counts typically receive lower doses of ruxolitinib (per its label), which would limit its efficacy.

Something to remember when looking at these post hoc comparisons is that, for many of the trials with ruxolitinib, investigators were likely underdosing the therapy due to patients’ low platelet counts. Typically, outside of a trial setting, a higher dose is often used (eg, 10 vs 5 mg twice daily in the 50-100 × 10³/μL platelet subgroup), depending on patients’ baseline platelet counts and tolerance.

When looking at the data, I would caution against automatically concluding that every patient with myelofibrosis who has anemia and platelets under 200 × 10³/μL should automatically receive frontline momelotinib. For a transfusion-independent patient who develops anemia on ruxolitinib, if ruxolitinib is doing a good job controlling the spleen and symptoms, it would be understandable for one to consider continuing the ruxolitinib and adding on a therapy to support the anemia. However, if the patient becomes transfusion dependent, I would generally switch to momelotinib.