Mesa Explains Risk Assessment For Patients With PMF and How it Affects Treatment

During a <em>Targeted Oncology </em>live case-based peer perspectives discussion, Ruben A. Mesa, MD, discussed risk assessment and treatment options available based on these assessments for patients with primary myelofibrosis with a group of physicians. Mesa, director of the UT Health San Antonio Cancer Center, explained these treatment options based on a case scenario of a patient with PMF.

Ruben A. Mesa, MD

During aTargeted Oncologylive case-based peer perspectives discussion, Ruben A. Mesa, MD, discussed risk assessment and treatment options available based on these assessments for patients with primary myelofibrosis (PMF) with a group of physicians. Mesa, director of the Mays Cancer Center, home to UT Health San Antonio MD Anderson Cancer Center, explained these treatment options based on a case scenario of a patient with PMF.


January 2016

A 59-year-old man presented to his physician with symptoms of fatigue, night sweats, weight loss and abdominal pain with early satiety past 6 months. On physical exam, his spleen was palpable, 10 cm below the left costal margin. Genetic testing revealed aJAK2V617F mutation and anASXL1mutation. A bone marrow biopsy revealed megakaryocyte proliferation and atypia with evidence of reticulin fibrosis. Additionally, a blood smear showed leukoerythroblastosis. His laboratory values were notable for red blood cells, 3.20 x 1012/L; hemoglobin, 9.7 g/dL; hematocrit, 34%; mean corpuscular volume, 93.1 fL; white blood cells, 12.1 x 109/L; platelets, 247 x 109/L; and peripheral blood blasts, 0%.

Targeted Oncology:What are the diagnostic criteria for PMF?

MESA:The diagnosis for PMF is a little complicated. It’s a bit like the diagnostic criteria for lupus. There isn’t a single test available. There are consensus criteria, notably the World Health Organization criteria, that bring together several factors.

One, there is the clinical suspicion. Does the patient have a molecular mutation? Over 90% of patients will have aJAK2mutation, about a third will have a calreticulin [CALR] mutation, and a small percentage will have anMPLmutation. There is a small percentage of patients who do not have 1 of those 3 driver mutations. Notably, these mutations are mutually exclusive, and patients typically only have 1 of those 3 mutations.

The other mutations that you receive on a myeloid panel are referred to as additional somatic mutations. Some [of these mutations] may have pathogenetic implications, and some may not. But these are different from the driver mutations.

What other diseases should be ruled out when making this diagnosis?

Before we could test forJAK2, CALR, orMPL, there was some uncertainty regarding the diagnosis of PMF. There are several secondary causes of fibrosis: rheumatologic diseases, infection, and bone marrow damage. I would say the other diseases in the diagnostic differential are other myeloid disorders. The trickier diagnoses to exclude are chronic myeloid leukemia [CML]. I’ve seen patients who were diagnosed as triple-negative myelofibrosis, but what they really had was CML. So we must be certain with these patients. Another tricky diagnosis is myelodysplastic syndrome [MDS] with fibrosis. If the patient has no splenomegaly, a lot of cytopenias, but lacks any constitutional symptoms, I’m much more suspicious: Is this really an MDS with a secondary fibrosis? These diseases clearly overlap, but they have a different biology and require different therapies.

Another factor are the overlap symptoms; for example, chronic myelomonocytic leukemia. This can give the patient fibrosis. You can exclude this based on histology and the monocyte component.

How important is the peripheral blood smear?

It is completely appropriate to order a peripheral blood smear before you move to next-generation sequencing.

The peripheral drug tells us quite a bit. You have dacrocytes demonstrated, a leukoerythroblastic picture [emerges], you might see an elevation in the lactate dehydrogenase. There are many things in terms of the peripheral blood and the appearance of the smear that are important factors to consider in the differential.

Myelofibrosis has a broad range of natural history. What are some of the risk factors that should be considered?

The degree of the severity of cytopenias is important. So is anemia, which is worse if the patient is transfusion dependent, particularly before they have cytoreductive therapy. Likewise, if the patient has leukopenia. Now, this isn’t very common, but if the patient has leukopenia, that suggests a worse prognosis.

Other factors include splenomegaly. This clearly adds to the burden of myelofibrosis. Interestingly, splenomegaly is not a factor when developing myelofibrosis scores; however, I think that’s a reflection of the data. All of these studies that yield identifying prognosis are based on retrospective chart stud&shy;ies. Many of these chart studies are inadequate. In the setting of a clinical trial, there are volumes of notes [that are clear]. But looking back at old charts, you would look at the notes and it would say, “spleen big,” “spleen huge,” or “spleen down to the pelvic bridge,” and that can be difficult to interpret and this affects the data.

As we assess risk, over time a number of scoring systems have evolved. Please share your thoughts about this evolution.

Scoring systems, traditionally used in myeloproliferative neoplasms [MPNs], have now evolved to include molecular information to supplement the [scoring] that these systems provide. If we try to come up with a treatment plan for patients with myelofibrosis, there are several components to consider: First, do they have cytopenias? And if so, are they problematic? Second, is there splenomegaly and is that problematic? How big is it to be problematic? Well, that’s very much in the eye of the beholder. Is it causing pain or discomfort? Is the patient experiencing difficulty sleeping because they can’t find a comfortable position? Is the patient experiencing difficult symptoms?

The fourth component is the issue of progression. What level of progression impacts intervention? Most patients who present with myelofibrosis will die only if the disease progresses. If the disease worsens, that’s the cause of death. The path of myelofi&shy;brosis resulting in a bigger spleen or debilitating symptoms could cause death from potential complications that occur. Or myelo&shy;fibrosis could transform to acute myeloid leukemia, which turns out to be a very refractory disease that can be difficult to treat and is a different biology from myelofibrosis.

Now there have been almost an absurd number of publica&shy;tions about prognostic scores in myelofibrosis. I usually tell our trainees to not bother memorizing these prognostics because (1) trying to keep track of multiple prognostic scores in multiple disease settings, especially hematology malignancies, let alone hematologic and solid tumors, is very difficult; and (2) it’s more important to understand where these scores are helpful; and (3) there is some question about what is prognostically relevant that tells us a bit about the biology of the disease. And these scores [from different systems] clearly overlap.

So if we think about the prognostic scores for CML, like the Sokal Index or the myelofibrosis prognostic score like MIPSS [Mutation-Enhanced International Prognostic Scoring System], these all overlap. Intellectually, these all overlap. It’s a spec&shy;trum of diseases.

There are clinical variables: anemia, leukocytosis, thrombocy&shy;topenia. So cytopenias are a big variable. Is the patient moving toward acute leukemia? Are there increasing circulating blasts? Or do they have difficulty symptoms?

Prognostic scores are an assessment of the disease itself, but a disease doesn’t exist in a vacuum. It exists in the patient. I can put the disease in 2 different patients, but it will have 2 differ&shy;ent natural histories because of the patient’s fitness, age, and comorbidity status.

What’s changed over the past 50 years is something referred to as other molecular features and how they fit into the [management] mix.

Where are these scores helpful?

They are the most impactful in terms of helping the clinician make the most difficult decisions, which is the consideration of stem cell transplantation. How long is the patient going to live? But the scores don’t tell you how long the patient is suffering from the disease. These scores measure prognosis; they don’t measure disease burden. They are the most helpful when it comes to the issue of transplantation.

The MIPSS70, which applies to individuals under 70 years, considers age as a contributing factor to prognosis and as it relates to the decision of if a patient is a candidate for stem cell transplantation.

How should treatment be considered according to these risk assessments?

As someone who has cared for a large number of people with myelofibrosis, the one thing that I have learned with transplant is that we can clearly wait too long [before offering it].1If anything, over the years I’ve become more aggressive in terms of the timing. For the most part, we used transplantation as a salvage therapy, and it doesn’t work that well as a salvage therapy in myelofibrosis.

Without question, for patients with high-risk disease, we should be thinking about allogenic stem cell transplant. Now, almost everyone who has a transplant has been on ruxolitinib [Jakafi] or is on ruxolitinib. So there’s probably not a lot of soul-searching to determine if the patient should be on ruxoli&shy;tinib. If they are a candidate for transplant, they should be given ruxolitinib.

The timing of when they’re on ruxolitinib and when they should go to transplant is what we’re learning more about. My sense, and it’s difficult to prove the data, is that the patient with an opti&shy;mal response to ruxolitinib probably does best on transplant. The flip side is that’s when the patient wants to have the transplant the least because they’re feeling well and their spleen is smaller and they don’t anticipate a negative outcome.

For patients who are nontransplant eligible, the NCCN [National Comprehensive Cancer Network] guidelines suggest for us to start the patient on ruxolitinib. If there’s an appropri&shy;ate clinical trial, meaning a frontline clinical trial, then ruxolitinib should be started.

Patients with intermediate-risk disease [are] the trickiest [to treat] because it depends on how the patient is affected. The patients tend to be younger and have high-risk molec&shy;ular features but low disease activity. The question is who do we treat? It comes down to how the patient is being affected. Are the patient’s symptoms problematic enough to warrant therapy?

The low-risk end of the spectrum depends on how much the patient is impacted. You can consider nonpharmacologic treatments, especially if the standard therapies have too much expansion toxicity associated with them. There’s a range of nonpharmacologic studies ongoing.

What other factors trigger you to start therapy for myelofibrosis? When anemia is the driver, what sort of therapies do you consider?

It’s tricky, and it’s one of the more frustrating things about treatment. We now have 2 approved drugs in myelofibrosis— ruxolitinib and fedratinib [Inrebic]—and neither are effective in improving anemia. We use erythropoietic-stimulating agents if they have some baseline renal insufficiency, but that’s really it. For older patients who might have hypertension, the immunomodulatory agents have shown some activity, including thalidomide [Thalomid], which is the least expensive.

How important is it to start therapy early in patients with myelofibrosis?

At this point, part of our scientific gap is that we don’t know why patients progress. So it’s a hard thing to monitor. And that’s why we’ve had difficulty creating a therapy that limits progression. I think, in responding patients, there’s no question that JAK inhibitors, and perhaps pegylated interferon, may have an impact on progression.

What we can say, based on the COMFORT trials2involving ruxolitinib, is that individuals who had less advanced disease had better and more durable responses. Those patients with advanced disease with pelvic involvement, ascites, clearly had benefit. Patients in the middle of the pool—that is, intermediate risk—did much better.

There are still some patients in the phase I study started in 2007 at MD Anderson who are still on the drug. That’s 12 years now; some patients had life expectancies of 3 years when they started.

Are there risks with undertreatment of patients in the this setting?

Sure there are. [Patients] can do poorly while you’re not treating them. With myelofibrosis, while we’ve been focusing on symptoms, the patient can develop thrombotic events. We sometimes see significant thrombocytosis, significant thrombotic risk, and this is a second level of risk. Additionally, with the impact observed with JAK inhibitors, we have more advanced disease. You’re probably better off being on therapy. Ruxolitinib is helpful as a main therapy and less as a salvage therapy. If the patient is already having symptoms, there’s probably no benefit in holding onto ruxolitinib as a salvage therapy.

Please discuss the MPN symptom assessment form and its purpose.

My colleagues and I developed the 10-item symptom assessment form.3This form is easily obtainable and can be useful at baseline and for tracking these symptoms going forward. We tried this in 10,000 patients, and it has been translated into 30 different languages.

I would say that we have seen cultural differences in how populations rate their symptoms but not their prevalence. So, the prevalence and ranking are remarkably consistent.

Is there a survival benefit associated with the use of ruxolitinib?

Over time, we have found that patients live longer. The 2 randomized clinical trials for ruxolitinib did not use survival as the primary endpoint.2That being said, there is no question that patients live longer. Now, it’s not a cure, and there is not a plateau, but any way we look at the data, such as through pooled analysis, there [is] evidence of improved survival.

The issue of dose and the issue of the quality of the response [do] matter for survival. We have seen that the improvement in the spleen does correlate with improvement in survival. I think this is because the spleen acts as a biomarker of response to JAK inhibitors. I don’t think there’s a biomechanical reason for improvement in survival, I think it’s a biomarker of response.

As we think about dose, the target you want to reach is 10 mg twice a day. Looking at improvement in symptoms, the 10-mg dose has an impact. Doses lower than that have less efficacy and probably less likelihood for an improve&shy;ment in survival.4

Now, some people point out that ruxolitinib can cause anemia in some patients, which is a negative risk factor. By giving the drug, does that make their prognosis worse?

That doesn’t seem to be the case. There’s a difference between baseline anemia and therapy-associated anemia.5

What is the regulatory status of fedratinib for this patient population?

Currently, it’s approved, but it was on hold by the FDA for some time before approval. This drug was all ready for approval back in 2013, but it was found that 8 patients out of the 900 patients treated had a neurologic event that was suspicious for Wernicke’s encephalopathy. The FDA put the drug on clinical hold and provided the original manufacturer instructions about how to continue. The original manufacturer decided regulatory approval was too restrictive, so the drug just sat in limbo.

Another company reacquired the rights to the drug, and the 8 patients were examined by independent neurologists, and they said maybe only 1 patient developed Wernicke’s during the trial and the others might have had Wernicke’s at enrollment.

The other 7 patients were determined to have had mild neuro&shy;logic events. Eventually, the FDA approved the drug with a black box warning. A subsequent study is looking at this in greater depth.

I think we’ll get a much better sense of the drug by actu&shy;ally using it. It could be used in the frontline, but it will require better characterization; that is, who is responding or are there molecular profiles that are different. We don’t have data on that yet. There are some data that suggest that some patients can use fedratinib as salvage therapy.


  1. Tefferi A, Guglielmelli P, Pardanani A, Vannucchi AM. Myelofibrosis treatment algorithm 2018.Blood Cancer J. 2018;8(8):72. doi: 10.1038/s41408-018-0109-0.
  2. Verstovsek, S, Mesa RA, Gotlib J, et al; COMFORT-1 investigators. Long-term treatment with ruxolitinib for patients with myelofibrosis: 5-year update from the randomized, double-blind, placebo-controlled, phase 3 COMFORT-I trial.J Hematol Oncol. 2017;10(1):55. doi: 10.1186/s13045-017-0417-z.
  3. Emanuel RM1, Dueck AC, Geyer HL, et al. Myeloproliferative neoplasm (MPN) symptom assessment form total symptom score: prospective international assessment of an abbreviated symptom burden scoring system among patients with MPNs.J Clin Oncol. 2012;30(33):4098-4103. doi: 10.1200/JCO.2012.42.3863.
  4. Verstovsek S, Gotlib J, Gupta V, et al. Management of cytopenias in patients with myelofi&shy;brosis treated with ruxolitinib and effect of dose modifications on efficacy outcomes.Onco Targets Ther.2013;7:13-21. doi: 10.2147/OTT.S53348.
  5. Al-Ali HK, Stalbovskaya V, Gopalakrishna P, Perez-Ronco J, Foltz L. Impact of ruxoli&shy;tinib treatment on the hemoglobin dynamics and the negative prognosis of anemia in patients with myelofibrosis.Leuk Lymphoma.2016;57(10):2464-2467. doi: 10.3109/10428194.2016.1146950.