Monitoring for Signs of Progressive Disease and Response in Myelofibrosis

Andrew Kuykendall

What are you seeing among your patients with myelofibrosis when they're having progressive disease, from a clinical standpoint?

Natasha L. Johnson

When I am seeing a patient in clinic, at every single visit, I am assessing their symptoms and trying to have them complete the symptom assessment form to see if their symptoms are worsening. I’m assessing the spleen with a physical exam to determine whether the spleen is growing. Are they having signs of worsening splenomegaly, including early satiety, weight loss, abdominal pain, discomfort, or fullness? And then I look at their counts. Are they developing new or worsening cytopenias, or have we changed the dosage of their drugs recently, and despite that, the cytopenias haven’t improved? I look at their differential blood count—do they have new peripheral blasts present? All those would be signs that a patient could possibly be progressing.

Andrew Kuykendall

There's a difference between signs of clinical progression and how we define progression. Currently, we define progression as either a progression in the spleen size or a transformation into acute leukemia—obviously, that's not all-encompassing. However, we know that there is progression of the disease that is not captured by that limited definition.

We can have patients whose symptoms get worse; we can have patients whose counts change and who get massive leukocytosis. That happens occasionally, where we see patients for a long time and their white count is consistently 15 to 20, and then all of a sudden, it's 45, 60, or 75. Our current definition of progression doesn't capture that, but clearly that patient has changed, and their disease has changed. Similarly, we may have someone who's always had a hemoglobin level of 10 to 11, and now it's 6 or 7, and they're getting transfusions. That, again, is progressive disease.

When I think about progression, I think about it in all the different ways that myelofibrosis can affect a person. It can affect a person's symptoms, it can affect their counts, it can affect their risk of developing leukemia, and it can affect their spleen size and liver size. If they're having worsening of any of those features, to me, that's disease progression. I often like to correlate clinical progression—worsening of anemia, leukocytosis, worsening platelets, more blasts, worsening spleen size—with genetic progression. That's where we're often doing bone marrow biopsies, repeating next-generation sequencing panels to look at gene mutations, or looking at cytogenetics. Often what we're able to see is that they've acquired a new mutation, the disease has become more complex, or the JAK2, MPL, or CALR driver mutation they had that was previously present in 30% of their cells is now present in 100% of their cells. That's going to give you a more aggressive phenotype. The normal cytogenetics they had on the past bone marrow biopsy now show the addition of trisomy 8 or show some new translocation or deletion such as monosomy 7 or del(7q). All these things are often seen when you have clinical progression. When you can correlate the occurrence of clinical and genetic progression, you know that the disease is changing. That would be a time where alternative treatment options are considered, or where the benefit of transplant may now outweigh the risks.

Natasha L. Johnson

What would you say is the typical response time that you should see in a patient who has started receiving momelotinib in terms of their spleen, their symptoms, and anemia? When would you say that a patient has experienced momelotinib failure, and what would you consider for a second line of treatment?

Andrew Kuykendall

Those are good questions! With ruxolitinib, we have some data to look at as far as JAK inhibition. We know that when we're thinking about symptom responses, those typically occur quite quickly with ruxolitinib. We generally get a sense of spleen responses at around 12 weeks, where we're starting to see changes in the spleen volume; maximum spleen response may not occur until 24 weeks. With anemia responses in patients taking ruxolitinib, worsening of the anemia tends to happen in the first 4 to 8 weeks. So you can get a sense for anemia over the course of 2 months, spleen response at 3 months, and symptom response anywhere from days to weeks to months.

Momelotinib is going to be a bit different because it has a longer half-life, so I don't think we can expect the same timelines. If we're looking at symptom improvements, it may not be something we're seeing over the course of days to weeks—it may take a little bit longer. With momelotinib, these symptom scores may be helpful in that regard. When we started using ruxolitinib, we often didn't use the symptom scores in practice, because patients would come back in talking about how much better they felt immediately afterwards. But with more gradual or subtle symptom improvement—like we see with momelotinib—going forward, these symptom assessment forms may be more useful. Spleen responses, though, I would expect the same trends with both ruxolitinib and momelotinib—I would still expect to see those over the course of 3 to 6 months. Anemia responses may take longer. We know red blood cells last 90 days in circulation, and so seeing changes in anemia or hemoglobin is going to take a longer period of time. Even looking back at the pacritinib data, we saw that some of their achievement of transfusion independence was seen at 12 to 24 weeks, not within the first 12 weeks; I think we can expect similar trends with momelotinib, where we see a general, gradual improvement in hemoglobin and transfusion requirements.

To move onto your next question: when would I consider a momelotinib a “failure?” We don't know yet. I think that I would consider the same factors as ruxolitinib to make that assessment. I also think it depends on why the patient is starting the agent. If they’re starting it because they are transfusion-dependent or anemic and their transfusion requirements stay the same or worsen, then I would argue that I'm not sure exactly what it's doing there. The problem is you don't know whether they would have worsened more if you hadn't used momelotinib.

If they're still feeling well, we're in a similar situation as where we were with ruxolitinib before, where we don't have a great alternative for a second line of treatment. Pacritinib has an anemia benefit, but it is via the same mechanism of action as momelotinib, so I don't know if switching to that option is going to make a ton of sense. We also don't know about add-ons to momelotinib—that's really untested. The idea of adding danazol or an erythropoiesis-stimulating agent onto momelotinib is certainly something we have not tested, but it may be something we utilize down the road. We've kicked the can a little bit down the road to help with anemia—now we're going to have to kick it a little further if we can think about combinations with momelotinib.

If you switch to ruxolitinib or momelotinib and immediately, a patient’s spleen volume got worse or their symptoms got worse, I might call that a “failure” and go back to a JAK inhibitor that had given them better benefit. We've seen that with pacritinib and fedratinib when patients switch from ruxolitinib over to one of those agents—sometimes itching worsened, and so we had to switch back—but we haven’t had much experience with momelotinib yet. I will add that it’s really important to understand how patients are feeling on ruxolitinib before you switch therapies, and then to quickly assess if they feel worse, better, somewhat better, and look at counts as well. It’s important to really take the whole picture into account when you're considering continuing vs changing therapies.

Natasha L. Johnson

Speaking specifically to momelotinib, we know that side effects seen in the studies included some fatigue, dizziness, thrombocytopenia, and neutropenia with bacterial infections. Is there a point that you would look at a patient and say they require a dose reduction or interruption due to adverse effects?

Andrew Kuykendall

For cytopenias, I don’t think that we're going to consider dose reductions or interruptions for momelotinib, because I don't think that's necessarily going to be the cure there. I think you can consider dose reduction or interruption if you're talking about other, nonhematologic adverse events. We certainly didn't see a ton of those on the clinical trials. We saw some slight increase in gastrointestinal toxicity. In a patient who's having a lot of diarrhea, nausea, and vomiting, before completely discontinuing the agent, it may be reasonable to hold it for a little bit and try at a lower dose if you think it's providing benefit in other ways.

The label does suggest considering a dose reduction if you get to really low platelet counts. On the trial, patients could start with a platelet count of 25,000, but dose reductions were recommended if they went to less than 20,000. I don't know in practice if that's going to be something that we necessarily need to do. We'll have to get a feel for that, because I think if something's safe to be used at 25,000 platelets, it seems safe to be used at 15,000 platelets. Certainly, you could hold the momelotinib, and if the platelets came back up rapidly, you could consider that the low platelet count was related to that; dose reduction could be reasonable in that patient. However, I wouldn't want to sacrifice the clinical response of a suboptimal dose if the dose was reduced for reasons that weren't necessary.