JAK Inhibitors, Mutations, and Transplantation in Myelofibrosis Management

ONCOLOGY® CompanionONCOLOGY® Companion, Volume 37, Supplement 10
Volume 37
Issue 10
Pages: 18-23

John O. Mascarenhas, MD, led a panel discussion on myelofibrosis management including the use of JAK inhibitors and transplantation.

John O. Mascarenhas, MD, led a panel discussion on myelofibrosis management including the use of JAK inhibitors and transplantation.

John O. Mascarenhas, MD, led a panel discussion on myelofibrosis management including the use of JAK inhibitors and transplantation.

At an Around The Practice program hosted by CancerNetwork, a panel of experts discussed advances in the testing and treatment of patients with myelofibrosis. The panel was led by John O. Mascarenhas, MD, a professor of medicine at the Icahn School of Medicine and director of the Center of Excellence for Blood Cancers and Myeloid Disorders at Mount Sinai in New York, New York.

The panelists included Gaby Hobbs, MD, the clinical director of the Leukemia Service at Massachusetts General Hospital in Boston; Kristen Marie Pettit, MD, a clinical associate professor of medicine at the University of Michigan in Ann Arbor; and Andrew Kuykendall, MD, an assistant member in the Department of Malignant Hematology at Moffitt Cancer Center in Tampa, Florida.

Disease Presentation and Symptoms Associated With Thrombocytopenia

Mascarenhas: How do patients present with myelofibrosis?

Kuykendall: The interesting thing with myelofibrosis is that they don’t have to present in 1 set way. You have patients with classic myeloproliferative neoplasms [MPNs] who present with a lot of constitutional symptoms, such as fevers, chills, night sweats, bone pain, weight loss, or abdominal pain related to splenomegaly or organomegaly. You could also have patients who just present with abnormal lab counts that were done at their primary care office with unexplained anemia or maybe some immature blood cells floating around. That might clue physicians in to do a more extensive workup and try to figure out exactly what’s going on.

Mascarenhas: There were some abstracts at recent meetings, [such as] the 2023 American Society of Clinical Oncology Annual Meeting, [that] elucidated or identified clinical features and symptom burden. What is the association between thrombocytopenia and symptoms?

Pettit: Not surprisingly, both anemia and thrombocytopenia have been shown to be associated with different aspects of myelofibrosis. Both have been shown to be associated with worse overall survival [OS] for patients with a higher risk of leukemic transformation. Thrombocytopenia is also associated with worse symptom burden, which is not surprising, given what we see clinically in our patients on a [daily] basis.

Hobbs: One of the questions I get all the time from patients is: “If my blood counts look better, why don’t I necessarily feel better?” or “If my counts are controlled, why are my symptoms still present?” One of the things that we know, both in clinical practice and from some of the abstracts that were highlighted, is that blood counts don’t always tell the whole story. It’s interesting to think about patients who aren’t transfusion dependent but still have tremendous fatigue. There’s a lot that we need to understand [regarding which] cytokines are driving those symptoms and how we make those better. Do we need to target the anemia so they’re not short of breath and have those classic symptoms of anemia resolved, or is there something else contributing to those symptoms? We don’t have great measures other than the MPN Symptom Assessment Form, which is a way of making it less subjective. The burden of symptoms is sometimes out of proportion to what’s going on from a CDC perspective.

Cytogenomics and Mutational Profiling

Mascarenhas: What happens to patients with myelofibrosis? Are they just getting polymerase chain reaction [PCR] tests for a driver mutation? Are they getting next-generation sequencing [NGS]? How often do you do it? When you see your patients with myelofibrosis, what’s your approach in terms of getting their genetic data?

Pettit: These diseases are very heterogeneous in the clinic as far as how patients present. They are also very heterogeneous as far as the outcomes and what happens to patients over time. One thing that’s less heterogeneous is the driver feature of the disease, which is the overactivation of JAK/STAT signaling. We see that occurs most commonly through mutations in JAK2, CALR, or MPL. We always check for those 3 mutations at baseline. What’s also important, prognostically, are additional somatic mutations that can be present on top of those JAK/STAT activating mutations.

My personal practice in the clinic is to test for these things at the time of diagnosis for all my patients with myelofibrosis. Depending on what’s been done already, we sometimes already know the JAK2, CALR, or MPL status when patients come to us. If we don’t know those already, I sometimes do single-gene PCR tests for those because they can be done in [approximately] a 3-day turnaround time, which can help diagnostically. Sometimes it is important to get those done quickly. At the same time, I always send a NGS panel as well. We use an internal panel that we’ve developed and continue to refine over time that has all the high-risk mutations of interest here, including TP53, ASXL1, SSF2, IDH1/2, and EZH2. That turnaround time tends to be a little longer, though. It’s very important, prognostically, to know at baseline, but it does take a little bit of time to get back. I routinely do that for all my patients with a new diagnosis of myelofibrosis.

As far as serial monitoring over time, though, that’s an interesting question and something that has yet to be determined as to how often we should be testing over time. I always repeat NGS if I see a change in the disease behavior. If a patient goes from having more proliferative myelofibrosis with a high white blood cell count, large spleen, and high inflammatory symptom burden before developing some more cytopenic disease behavior with lower blood counts and more fatigue, then I’m always repeating NGS. I [also] do it if there are any signs of disease progression toward acute leukemia at that point.

Mascarenhas: What are some concrete examples of where mutational profiling might have a direct impact on therapeutic decision-making?

Hobbs: When I first meet a patient in the clinic, and I may not have all their information, there can be a lot of risk stratification without the mutations using the Dynamic International Prognostic Scoring System Plus Score, for example. Having those molecular markers is important when you first meet a patient to risk stratify them. It centers that discussion on what to expect for the patient, and it helps in terms of making the decision of whether this patient should undergo a transplant sooner rather than later. It’s always difficult in practice when you have a patient who may look like they’re lower risk—maybe their blood counts are OK—but then you do the NGS and find a lot of these high-risk mutations like ASXL1, TP53, or others. You may not necessarily act on that immediately, but at least it heightens your awareness about that patient, and you may follow that person more closely and refer them for a transplant. For prognostication and referral to transplant, having that additional information is key.

Mascarenhas: Does the presence of driver mutations help you in deciding the use of a JAK inhibitor?

Hobbs: My decision to initiate JAK inhibitor treatment is based primarily on whether that person either has systemic symptoms or symptoms from their spleens. What driver mutation they have doesn’t influence the decision to use a JAK inhibitor. The additional mutations also don’t factor into that. It does factor into how I phrase the conversation to patients, where I may say, “I’m going to start you on a JAK inhibitor to help with your spleen and symptoms, but based on these mutations, the end goal is still to get you to transplant.”

Mascarenhas: When you put someone on a JAK inhibitor, do you do any serial monitoring of that driver mutation? Is there any value in doing that?

Hobbs: Patients have started asking more and more for repeat testing of their driver mutations. I do not do that routinely for myelofibrosis. If there’s a change in the behavior of the disease, the patient’s going to transplant. If I think they’re developing leukemia, there’s some value to repeating that testing. Honestly, I find it hard to interpret different levels of those mutations, and it leads to a lot of anxiety. I’ve had patients who looked like they’re doing wonderfully on a therapy only to check the variant allele fractions again to see that they went up. It’s hard to act on those changes because we still don’t know how to interpret them, and our treatments aren’t made to target those mutations specifically. Even though we may see that JAK inhibitors may lead to a decrease in variant allele fractions in JAK2, we still don’t know what to do with that information.

Other Disease Markers and the Triple-Negative Population

Mascarenhas: Moving away from molecular markers, are there any other biomarkers that might be helpful?

Kuykendall: There is lactate dehydrogenase [LDH], this generic marker that could be looked at as a measure of a hypoxic state or cell turnover to some degree. It’s interesting within myelofibrosis and within MPN because LDH can sometimes help differentiate between things like essential thrombocythemia and myelofibrosis. When you get more hypercellular marrows, you get more cellular turnover and start to get elevated LDH levels. Across the board, LDH by itself doesn’t help too much. However, there are times when you see a baseline LDH that’s very high, which makes you think there is a lot more going on with this disease at the mere level than you otherwise thought. You may see a change in LDH levels in a specific patient, who’d been running at an LDH that is 2 or
3 times the upper limit of normal and goes to 6 or 7 times the upper limit of normal. It may make you think that maybe the disease is either changing or there’s something else going on. We saw this a lot during the era of COVID-19, where LDH levels would go up transiently soon after vaccination before tapering back down toward a more normal level. When the numbers changed, you had to realize that maybe there was some other influence that was going on. LDH can be helpful to clue you into some change either in the disease or some exogenous factor that’s causing it, although it certainly must be taken into context.

Mascarenhas: What about a biomarker like bone marrow fibrosis? Is there any value in repeating that bone marrow after a certain interval on a JAK inhibitor? If that fibrosis grade goes down, does that have any actual clinical impact?

Kuykendall: There are a lot of problems with bone marrow fibrosis. When we look at large studies, we can retrospectively see that there is some prognostic significance to fibrosis. In large populations, the more fibrosis you have, the worse the overall prognosis would be. On an individual patient level, it’s very difficult to understand the value. If someone goes from having moderate to mild fibrosis, is that a change in their underlying disease, or is that just a sampling change or something that isn’t truly related? Stephen T. Oh, MD, PhD, from Washington University School of Medicine in St Louis in Missouri, presented data on fibrosis changes in response to JAK inhibitors, suggesting that they don’t correlate with outcomes.1 On the flip side, we see fibrosis go away when patients get stem cell transplants, and that’s certainly something that reflects the disease being gone in those patients. It may come down to the type of treatments or the way we’re getting improvements in fibrosis. When you look at the paper presented by Oh, highlighting how you’re getting fibrosis changes over the course of 6 months, it might not be a relevant timeline to look at fibrosis changes. Maybe JAK inhibitors are not what we’re expecting to see as true disease-modifying agents. But maybe if we got more disease-modifying agents and we looked over time frames of 1 to 3 years, we could see a clinical relevance to a change in fibrosis over that period.

Mascarenhas: Is there anything that should be highlighted about the

Pettit: Treating the triple-negative [disease] is tricky overall. When I see a patient with triple-negative myelofibrosis, my first step is to always make sure I’m clear on the diagnosis and it’s not something else that’s causing fibrosis in the marrow. Not everything that causes marrow fibrosis is myelofibrosis. For example, there is myelodysplastic syndrome with fibrosis, chronic myeloid leukemia with fibrosis, and autoimmune myelofibrosis. Those things are important to rule out and can sometimes be easier said than done. Based on the bone marrow morphology and the presence of other cytogenetic or molecular mutations, if we’re confident about the diagnosis of triple-negative myelofibrosis, those patients tend to have a bit of a rougher course. They tend to be a little more difficult to treat and have a bit more aggressive disease. That is one finding in triple-negative myelofibrosis that does make me think more about the possibility of transplant earlier rather than later for patients, even though triple-negative status isn’t necessarily worked into most of our prognostic scoring systems.

Mascarenhas: Similarly, with the driver mutations and the use of JAK inhibitors, is there any reason why you would not use a JAK inhibitor in a patient with triple-negative disease?

Pettit: No, there’s no reason why you would avoid a JAK inhibitor. I made my decision to use a JAK inhibitor based on the presence of symptoms and symptomatic splenomegaly, not the driver mutation status. However, there are some factors that can correlate with response or lack of response to JAK inhibitors. For example, some of the RAS pathway mutations have been associated with resistance to JAK inhibitors. It doesn’t necessarily play into my decision of whether to use the agent, but it does slightly change my monitoring and counseling of the patient before we get started with the agent.

Spleen Reduction and National Comprehensive Cancer Network Updates

Mascarenhas: Let’s say you have a patient with 12% blasts in the peripheral blood or in the bone marrow. Do you and your transplant colleagues want to reduce the situ of those patients before transplant, or do you get them right into their transplant?

Hobbs: On paper, I would say the transplant doctors often do want to see some situ reduction. It also relates to what’s going on with the patient and what else is going on other than the 12% blasts. We’ve all seen patients [who] come in one day with 12% blasts, then 2% blasts on the next day. They fluctuate wildly. It’s a matter of seeing what’s going on in the bone marrow and what other mutations that person has. Does that person also have splenomegaly, things that correlate with worse outcomes after transplant? For the most part, at least 1 or 2 cycles of some form of cytoreductive therapy—whether that be adding a hypomethylating agent by itself or adding some venetoclax [Venclexta]—might be something that we would consider for that type of patient prior to going to transplant.

Mascarenhas: Do the transplanters want a certain spleen volume or spleen length before they take the patient to transplant?

Pettit: It’s a controversial topic, and I am always going back and forth about this one. We have good data highlighting that the earlier we do transplants, the lower the grade of disease there will be, the smaller the spleen will be, and the better patients do overall.2 Does that mean we should be shooting for the smallest spleen possible before transplant? Does that improve things, or does the risk of splenectomy outweigh the potential benefit? I don’t know the answer to that yet. I believe there is a trial going on trying to answer that question. Hopefully we will have some more guidance there.

Right now, we’re just going based on retrospective and observational data. Our approach that our center has come up with is to do everything medically that we can to shrink the spleen before transplant. I often do that even at the expense of blood counts for some patients, too. In that situation, I will accept a bit more anemia and thrombocytopenia if it’s with the goal of getting someone to transplant with as small of a spleen and as good of disease control as possible. But for patients with very massive spleens, we end up occasionally doing splenectomies before transplant. It’s not a common thing to do. Our soft cutoff has been a spleen size of [approximately 20 cm] by imaging. That is a very soft cutoff. We’ve made exceptions on either side of the spectrum there, but I hope we have more hard data to go off of soon.

Mascarenhas: What are some relevant myelofibrosis National Comprehensive Cancer Network [NCCN] updates that others should be aware of?

Kuykendall: There haven’t been substantial updates to the NCCN guidelines. There’s been tweaks of some things as we get new approvals, but in myelofibrosis, we’re still stratifying patients as higher or lower risk, and we’re treating patients who are lower risk with symptoms as we would for patients who are high-risk. It’s important to note that these risk scores identify patients [who] should be considered for transplant; that’s the purpose of the risk scores. It’s not something that should factor much into our current treatments. Some trials enroll patients with higher-risk disease, but the treatments are very much aimed at symptoms and enlarged spleens. It’s not so much driver mutation or risk as much as symptoms that drive what we’re trying to accomplish.

Within the NCCN guidelines, we’re favoring JAK inhibitors for patients with higher-risk or symptomatic lower-risk disease. For the most part, we differentiate by platelet count. For those with greater than 50,000 platelets, we consider ruxolitinib [Jakafi] or fedratinib [Inrebic], which are approved there. For those with fewer than 50,000 platelets, we favor pacritinib [Vonjo], which has an accelerated approval in this markedly thrombocytopenic patient population.3 One caveat that’s overlooked a lot in the NCCN guidelines is it makes it seem like any patient who is high-risk should receive ruxolitinib or fedratinib, but that excludes those who are high-risk where anemia is their driving issue. If their main issue is anemia and they don’t have too much in the way of constitutional symptoms or symptomatic splenomegaly, then they are probably not a great candidate for a JAK inhibitor as an up-front treatment. That sometimes gets lost in the guidelines, but there is a drop-down as far as where to go when you’re considering directing your therapy at myelofibrosis-associated anemia as the prominent issue.

When you look past the first-line setting in the NCCN guidelines, anything is open. At that point in time, platelet thresholds don’t matter as much, and pacritinib can be considered as an option in the second-line setting. It has this accelerated approval for fewer than 50,000 platelets, but there has yet to be a completed trial that was focused on fewer than 50,000 platelets as the inclusion criteria. It’s been studied in any platelet count of fewer than 100,000, and then there’s an ongoing study looking at those with fewer than 50,000 platelets, but we know it’s an effective agent regardless of platelet count. Those with fewer than 50,000-platelet count is where the specific agent could be utilized. In the second-line setting, it can be used there as well, just like fedratinib. If you started with fedratinib or pacritinib, dose-modified ruxolitinib could be also a consideration in the second-line setting.

Pacritinib and Other JAK Inhibitors

Mascarenhas: Why is pacritinib different?

Kuykendall: Pacritinib is somewhat more of a JAK2 selective inhibitor, which differentiates it from ruxolitinib. Fedratinib is also somewhat selective toward JAK2. Pacritinib has this additional inhibition of IRAK1 and maybe ACVR1. The early trials with pacritinib were interesting in the fact that it didn’t seem to cause the same degree of myelosuppression that was seen with ruxolitinib and fedratinib. Some of this had been attributed to IRAK1, an NF-κB pathway target that may be involved in some other symptoms, some cytopenias, and a different inflammatory pathway. More recently, building off the knowledge that momelotinib is an ACVR1 inhibitor, it was seen that pacritinib may be a more robust ACVR1 inhibitor. This is something that is a relevant target in the treatment of anemia and cytopenias. Perhaps this is why we can safely leverage pacritinib into this more cytopenic myelofibrosis population and target splenomegaly symptoms with full doses of this JAK inhibitor in a patient population that is more difficult to provide fedratinib or ruxolitinib to.

Mascarenhas: Do you ease patients into pacritinib, or do you use the full dose?

Pettit: It’s tough, and how we do this is evolving. Personally, it depends on a case-by-case basis. If the patient is on ruxolitinib at 20 mg twice a day, I would consider a taper or maybe a steroid overlap because of the risk of the JAK inhibitor rebound as you’re switching from one agent to the other. If they are on a small dose of a JAK inhibitor, I would just switch them right over from ruxolitinib to pacritinib.

Mascarenhas: What would you do at the full dose?

Pettit: One of the benefits of pacritinib is being able to start the full dose of 200 mg twice a day regardless of the platelet count, hopefully maximizing that JAK2 inhibition that you’re getting into somebody because you don’t have the JAK1 inhibition that’s going to lower their platelet count more.

Mascarenhas: How closely do you follow someone when you’re transitioning them from ruxolitinib to pacritinib?

Pettit: I follow them closely for several reasons: symptom burden, worsening of spleen size, any toxicities from the drug, and cytopenias.4 I typically monitor complete blood counts [CBCs] depending on the situation. If their platelet count is getting low, I’d probably check their platelet count and CBC maybe once a week or every other week and follow up with them clinically within 1 to 2 weeks of starting the drug. One of the things to keep in mind with pacritinib is that the toxicity profile is a little different than it is with ruxolitinib. There are some things in patients with myelofibrosis that we need to think about proactively with some of the newer agents [such as] fedratinib or pacritinib. The main thing is gastrointestinal [GI] toxicity. Diarrhea is very common early on. Nausea and vomiting can [also] happen early on. Those are things I generally tell patients to expect. It’s better to be proactive and be prepared for it rather than to be surprised by it, because it’s harder to get on top of it once it’s already developed.

I make sure in advance that patients have loperamide [Imodium] on hand at home. I [also] often make sure they have an antiemetic, or are at least very easily able to get an antiemetic from our clinic if they need it once they start the agent. The good news is that these GI toxicities with pacritinib do tend to get better over time, not worse, [and] they don’t tend to be long-lived. They tend to be things that occur within the first couple weeks then improve over time. If we’re proactive about it, there are things we can get through without too much difficulty in the first couple of weeks.

Mascarenhas: With low platelet counts, what is your comfort level in terms of dosing ruxolitinib to manage thrombocytopenia?

Kuykendall: It depends on what other options you have. Before pacritinib was available, we tried to slowly bring patients along on ruxolitinib to keep them on a JAK inhibitor. A lot of it depends on what you’re aiming for with the treatment. If patients are benefiting from a substantial dose of ruxolitinib, they can keep their spleen volume and symptoms in check, and their platelets are running at 50,000 or maybe 40,000 consistently, I’m not necessarily switching them to pacritinib. They are very different patients than the ones whose platelets are a lot more volatile; those are patients who could switch to pacritinib.

Mascarenhas: Why is it possible to sequence JAK inhibitors? Is there any evidence that suggests there may even be molecular predictors?

Hobbs: To some degree, yes, we can sequence JAK inhibitors, and it is surprising. We’ve learned that the JAK inhibitors can hit the kinases a little differently. In that regard, it makes sense that you could sequence and block the JAK/STAT signaling pathway that you weren’t able to block with the prior JAK inhibitor. It also makes sense when you’re thinking about switching from a JAK1 or JAK2 inhibitor to one of the newer JAK inhibitors, [such as] pacritinib, that target IRAK1 [and] ACVR1, which may yield a new response where you lost a response previously. Understanding each JAK inhibitor’s mechanism of action can help you understand why you can “salvage” some of those patients or have a response in the second- or third-line setting. If the treatment goal is to improve splenomegaly and to not get to transplant, it’s nice to know we can do that. However, if the goal is to get to transplant, knowing that you can successfully sequence one patient from one JAK inhibitor to another shouldn’t make you lose track of your objective. If a patient is losing their response to a JAK inhibitor, their spleen is growing back, their symptoms are coming back, or their platelet counts are looking much worse, that’s not a time to switch from one drug to the other but a time to get that person to transplant if that hadn’t been part of the conversation.

Closing Remarks

Mascarenhas: Is there anything you want to share from this discussion that you think is important to remember?

Pettit: We need to raise the bar for response in myelofibrosis. Now that we have all these available agents with many more coming, our antiquated 35% spleen volume reduction goal and even 50% symptom score reduction goal are low bars for us to be shooting for. Now that we’ve scratched the surface of the disease, we should be trying to get deeper into the disease in quality-of-life improvements, OS improvements, prevention of progression, and treating the underlying disease.

Hobbs: I agree completely. It’s time to raise the bar on our end points, and that’s going to require more targeted agents and having the regulatory agencies revisit what a meaningful end point is. We all did a nice review of what therapies are currently available. The way that we think about when to use JAK inhibitors is important. Another point we all made was that an early transplant referral is important, even though those discussions can often take many visits until you finally determine whether that patient will go to transplant and when they’ll go to transplant. Outcomes for patients with myelofibrosis have improved significantly with transplant, with reduced intensity conditioning, and with the use of ruxolitinib and other JAK inhibitors before and perhaps after transplantation. Remembering that patients can make it through transplant is important. While all these other therapies are being developed and we get better end points, we do have a therapy that does work and is curative. Helping to increase the use of that therapy is an important goal.

Kuykendall:We have all these options that make us think we have something for everyone now. However, we should be considering clinical trials in every aspect of this disease. Whether it’s in the frontline or second-line setting, or we see low, high, or normal platelet counts, we’re still getting clinical improvements. When the best we can do is make [patients] feel a little better and make them live a little longer, then we need to leverage that and try to do better in all aspects of the disease. What I hope doesn’t happen with the approval of a third and a fourth JAK inhibitor is that we’re seeing patients in the academic setting after they’ve received 4 JAK inhibitors. We have the option of getting [patients] on good, novel, meaningful trials in the frontline setting, where we can build on the success we’ve had with JAK inhibition and then see [whether] we can do a little more as well.


  1. Oh ST, Mesa RA, Harrison CN, et al. Pacritinib is a potent ACVR1 inhibitor with significant anemia benefit in patients with myelofibrosis. Blood Adv. 2023;bloodadvances.2023010151. doi:10.1182/bloodadvances.2023010151
  2. Polverelli N, Malagola M, Comini M, et al. Allogeneic hematopoietic stem cell transplantation for myelofibrosis: role of disease status and chimerism on long-term outcomes. Blood. 2022;140(suppl 1):6870-6871. doi:10.1182/blood-2022-166212
  3. CTI BioPharma announces FDA accelerated approval of VONJO (pacritinib) for the treatment of adult patients with myelofibrosis and thrombocytopenia. News release. CTI BioPharma Corp. February 28, 2022. Accessed August 23, 2023. https://prn.to/3ppk5Cm
  4. Bose P, Gagelmann, N, Gupta V, et al. Consistency of pacritinib for spleen and symptom reduction in patients with myelofibrosis regardless of cytopenias. J Clin Oncol. Published online May 31, 2023. doi:10.1200/JCO.2023.41.16_suppl.7068
Related Videos
Expert on myelofibrosis
Expert on myelofibrosis
Experts on myelofibrosis
Expert on myelofibrosis
Expert on myelofibrosis
Expert on myelofibrosis
Expert on NSCLC
Experts on myelofibrosis
Expert on myelofibrosis
Related Content