So, next we&rsquo;re going to discuss a 67-year-old woman who goes to her primary care physician with a number of symptoms. She has fatigue, she has abdominal fullness, she has night sweats, and she has been losing weight. She has lost 17 pounds. On her past medical history, she just has hypertension on physical exam. Her spleen tip is palpable, about 7 cm below the left costal margin. Nothing else on exam. In terms of her blood count, she&rsquo;s anemic with a hemoglobin of 9.2. She has a normal platelet count of 189,000 and she has leukocytosis with a white-blood cell count of 19,000.

So, at that point, you know that the patient has leukocytosis anemia and a big spleen. There&rsquo;s a very broad differential diagnosis here. And looking at the peripheral blood smear can be very helpful because this could be a patient with chronic lymphocytic leukemia, mantle cell lymphoma, or marginal zone lymphoma. So, lymphomas can present like this. Even hairy cell leukemia can present like this. And so, the peripheral blood smear might be very helpful with that leukocytosis if there are a lot of lymphocytes there and you can perform flow cytometry on the peripheral blood to try to ascertain the diagnosis. Evidently, that was likely not the case because she has a bone marrow biopsy, and the bone marrow biopsy shows class 3, or grade 3, fibrosis in the marrow, myelofibrosis. They do immunostains for CD34 and CD117 and find 1.2% blast in the marrow, and they do a molecular analysis showing that the disease in the marrow is positive for substitution of missense mutation in

mutation. So, the final diagnosis is primary myelofibrosis.

Case Presentation: Primary Myelofibrosis

In looking at her presenting symptoms, physical exam, and blood counts, this is not at all atypical for patients. And in fact, patients like this may be seen by a number of specialists and primary care physicians before they actually get to the hematologist. So, if a blood count isn&rsquo;t checked and they&rsquo;re having abdominal fullness, night sweats, prurituswhich is quite common&mdash;or fatigue, a number of evaluations for other disorders might have ensued before finally getting to the blood count and showing that the problem is in the bone marrow. So, this is quite typical.

Now the natural history of myelofibrosis is quite varied. Some patients having a very long survival, and some much shorter. And a number of prognostic staging evaluations, or models, have been developed. The first that we&rsquo;ve been using is Cervantes&rsquo; International Prognostic Score criteria, where there are 5 prognostic factors: age over the age of 65, which she has; presence of constitutional symptoms such fevers, night sweats, and weight loss, which she has; anemia with a hemoglobin under 10; leukocytosis with a white-blood cell count over 25,000; and the presence of more than 1% circulating blast. That&rsquo;s the only one we don&rsquo;t know that she has, but she has all of the other 4.

If you have none of those factors, then you have low-risk disease with a median survival that exceeds 10 years. However, most patients, 90% of patients, will have intermediate 1risk disease with 1 risk factor, intermediate-2 with 2 risk factors, or high-risk disease with 3 or more risk factors. The median survival gets progressively worse with each of those. For patients with high-risk disease, the median survival is only about 2 years.

Presentation and Risk Stratification of Myelofibrosis

This is a very typical case of primary myelofibrosis, and it brings us to the discussion of how exactly this is diagnosis made. Now you would think that seeing fibrosis in the marrow would make the diagnosis. That is clearly not the case. There are multiple things that can cause reactive fibrosis of the bone marrow: other malignancies, carcinomas, hairy cell leukemia lymphomas, infections and granuloma in this disease, drug reactions can do it. Even autoimmune disease has been associated with an autoimmune myelofibrosis. And so, the mere fact that there&rsquo;s fibrosis doesn&rsquo;t mean that the patient has primary myelofibrosis.

In fact, morphologically what the pathologists hang their hat on in terms of this diagnosis is the megakaryocytic hyperplasia and atypia with hyperchromatic, very folded and low-beta nuclei, and the clustering of those megakaryocytes. It&rsquo;s really the megakaryocytic abnormalities that lead to the consideration of myelofibrosis as the diagnosis. And in fact, there is a prefibrotic myelofibrosis that&rsquo;s characterized by those same megakaryocytic changes, but without the grade 2 or 3 myelofibrosis that&rsquo;s seen then.

So, the second criteria then are to rule out other myeloproliferative neoplasms and other myeloid malignancies. And the third is to find some evidence of clonality. Now in 90% of patients, there will be mutation in

, the V617F, as in this patient, or in calreticulin, or

. Those are the 3 genes that are most commonly mutated in myelofibrosis. The other 10% of patients have what is called triple-negative myelofibrosis. A number of other mutations have been associated with myelofibrosis such as

, for example. That&rsquo;s important because not only can a broader myeloid gene panel make the diagnosis if you don&rsquo;t find

, calreticulin, or MPL, this more extended gene panel can help with the diagnosis by showing clonality, but also, it&rsquo;s prognostically important. So, patients with

-mutated myelofibrosis, especially if they don&rsquo;t have a calreticulin mutation, have a worse outcome.

If you don&rsquo;t find evidence of clonality based on the mutation analysis, then you have to be certain to rule out all of the reactive causes of myelofibrosis. You have to have all 3 of those major criteria: megakaryocytic hyperplasia and atypia, the ruling out of other myeloid malignancies, and finally, evidence of clonality by mutational analysis, or at least ruling out other reactive cause of myelofibrosis.

There are 5 minor criteria, you only have to have 1 of those: a leukocytosis with a white count over 11,000so, it&rsquo;s different than the staging system&mdash;anemia hemoglobin under 10, elevated LDH, splenomegaly, and the final one is a leucoerythroblastic picture&mdash;immature white cells, nuclei red blood cells, teardrop cells, and giant platelets. Those are things that you might see in a patient in the peripheral blood of a patient with primary myelofibrosis.

What&rsquo;s interesting about the morphologic classification, how you make the pathologic diagnosis, the splenomegaly is a minor criteria, but the degree of splenomegaly actually has not been associated with prognosis in the prognostic schemes. Even in the revised International Prognostic Scoring System and the revised Dynamic International Prognostic Scoring System, splenomegaly isn&rsquo;t included. What&rsquo;s included would be if a patient has thrombocytopenia, if the patient has certain poor-risk cytogenetic changes requiring transfusions. And in fact, anemia is doubly weighted in those more recent classification systems.

So, this is a challenging diagnosis. Patients with myeloid fibrosis, I truly believe their marrows need to be seen by hematopathologists who are experts in the diagnosis of myeloid neoplasms. Distinguishing the myeloproliferative neoplasms from each other pathologically is quite a challenge. Even grading the degree of fibrosis is something that is not tremendously well standardized, and there can be disagreements between pathologists. So, I do recommend getting a second opinion from a hematopathologist who specializes in myeloid malignancies.

Identifying a Diagnosis of Myelofibrosis

This patient is clearly symptomatic from her disease and the question is not whether or not you&rsquo;re going to treat her, but with what? What is the best option for her? And I think we have data from the COMFORT-II study that can help us in that decision. The COMFORT-I study, just to remind you, was a study of ruxolitinib, the JAK1/JAK2 inhibitor, versus placebo. Well, I&rsquo;m not going to dwell on that study because in that study, some patients didn&rsquo;t receive anything. And a patient who has constitutional symptoms, symptomatic from their splenomegaly, I think most of us would want to treat that patient.

In COMFORT-II, which was a study that was done in Israel and Europe, patients with intermediate 2 and high-risk myelofibrosis, either primary myelofibrosis or following prior polycythemia vera or ET, were randomized to get ruxolitinib at a dose of 20 mg twice daily if their platelets were above 200,000, or 15 mg twice daily if their platelets were between 100,000 and 200,000. They were excluded from the study if their platelets were below 100,000.

Patient were twice as likely to get that than the control arm, which was best available therapy. Patients who received best available therapy, most of them received hydroxyurea. Some received modulatory agents like lenalidomide or thalidomide. Steroids have been used, and interferon as well, if I didn&rsquo;t mention that already.

And so, there was a crossover in this study. The primary endpoint was at 48 weeks control of the splenic volume. So, a reduction of splenic volume as measured by MRI of more than 35% reduction in that volume. The primary endpoint was met. None of the patients in the best available therapy arm actually achieved that endpoint. And it was 28% of patients at 48 weeks who actually had a 35% or more reduction in the splenic volume with ruxolitinib. A key secondary endpoint was control of symptoms, which is important to consider in this patient who is symptomatic with night sweats, weight loss, and the abdominal fullness.

In that study, approximately 45% to 50% of patients had a 50% or more reduction in a total symptom score that had been validated previously in patients with myeloproliferative neoplasms such as myelofibrosis. Even when you consider each of the individual symptoms and the total symptom scorenight sweats, pruritus related to cytokines, weight loss related to cytokines, but also symptoms related to the spleen like early abdominal fullness, pain under the ribs&mdash;patients had the same type of benefit from ruxolitinib in all of those individual symptoms.

So, I think in a patient like this who is symptomatic, I would start with ruxolitinib. Her platelet count is 189,000. The label dose is 15 mg twice daily. Quite frankly, I often start lower, but remember to escalate the dose as tolerated every 4 weeks.

Now there was another benefit that was seen in that study, another key secondary endpoint was survival. One of the problems with evaluating survival in this study, and as well the COMFORT-I study, was a crossover design. And if patients had a progression of splenomegaly or an increase in their requirement for opiate analgesics for control of painful spleen, they could actually cross over. And in fact, in the COMFORT-I study at a median of 9 months, patients crossed over from placebo to ruxolitinib. In the COMFORT-II trial, at a median of 15 months, they crossed over. Most of them crossed over from best available therapy to ruxolitinib.

However, when you evaluate survival at 3 years and now at 5 years on that study, even though there&rsquo;s a crossover, there was a trend for an improvement in survival. The

value at 5 years was 0.06, but there was a trend for an improvement in survival with about a 35% reduction in the risk of death of that 5-year follow-up of the COMFORT-II study. And so, in a patient like this, you&rsquo;re not only trying to control symptoms, you&rsquo;re not only trying to control the spleen, which you can do with ruxolitinib, but you are also trying to extend survival.

Well, let&rsquo;s not forget that ruxolitinib so far has not been clearly shown to change the natural history of the disease. We have not been able to show, because of the crossover honestly in these studies, that there&rsquo;s a decrease in fibrosis, it complicated that analysis. Although you could see some decrease in the allele burden of the

missense mutation, it has not been dramatic. And so, it&rsquo;s not clear that ruxolitinib really changed the natural history of the disease.

And so, in a patient with high-risk disease, she only has hypertension, if her spleen is controlled, she gains weight, and she has improved performance status, this is somebody who I would definitely still consider for allogeneic stem cell transplant, which is the only potentially curative option. So, that&rsquo;s something we shouldn&rsquo;t forget about in our patients with intermediate 2 and high-risk disease, that a potentially curative option that can actually reverse the fibrosis is allogeneic transplant.

COMFORT Trials: Initiating Therapy for Myelofibrosis

I&rsquo;ve talked about the COMFORT-II study, but let&rsquo;s go back to the COMFORT-I study because I think there is an important message there. In that study, patients were randomly assigned to get ruxolitinib or placebo. And so, let&rsquo;s say that you have a patient who has intermediate 2 or high-risk disease and they&rsquo;re not terribly symptomatica few night sweats, let&rsquo;s say, their spleen is enlarged but it&rsquo;s not painful. Is that a patient who you should start on ruxolitinib? And I think the survival data from the COMFORT-l needs to be taken seriously and considered in a patient like that.

Patients with intermediate 2 and high-risk disease, those who were randomized to placebo, even though almost all of them crossed over to ruxolitinib, the survival curve separated after the median time to crossover, which was 9 months, suggested that by delaying exposure to ruxolitinib in an intermediate-risk to a high-risk patient, you may actually be affecting that patient&rsquo;s chance at long-term survival. Maybe they&rsquo;ve just been allowed to become more cachectic, susceptible to infections, which is cause of death in this disease, or vascular events. It&rsquo;s not really clear why it is, but nevertheless, the observation was there in the COMFORT-I study when randomized to placebo.

So, I think this risk stratification is really important. I think hematologists need to take it seriously, recognize that the majority of our patients with intermediate-risk and high-risk disease, and by delaying therapy because they don&rsquo;t feel that the symptoms are severe enough or the spleen is big enough, you actually may be doing some harm by putting that patient at risk for shorter survival.

If you are following a patient though, if they are completely asymptomatic and they have lower-risk disease, their spleen might be palpable but it&rsquo;s not symptomatic, this is a patient who you need to follow very closely and you need to ask questions that really get at whether the disease is affecting them. This can be tricky because patients can have symptoms from this disease that they are chalking up to something else&ldquo;I have aches in my bones because I&rsquo;m old and I have arthritis,&rdquo; or &ldquo;I have pruritus because my dermatologist says I have eczema,&rdquo; or &ldquo;I&rsquo;m having night sweats because I am going through menopause.&rdquo;

And so, patients may have been told that they have a whole bunch of symptoms related to a number of diseases whereas the unifying disease may actually be the myelofibrosis. So, I think the treating physician has to keep that in mind when questioning their patients that what they&rsquo;re going to be told by their patient may actually be due to the myelofibrosis and deserves attention and therapy. Likewise, by physical examination if the spleen is enlarging and grows obviously, let&rsquo;s say by 5 cm, that&rsquo;s a sign of disease progression. And that patient should be questioned very closely about symptoms from the spleen, from constitutional symptoms, and considered for therapy with ruxolitinib if that&rsquo;s the case.

MF: Undertreatment Risks and Markers for Progression

What ruxolitinib has done is really drawn some attention to these myeloproliferative neoplasms. And we now have a specific agent for not only myelofibrosis but also a certain class of patients with polycythemia vera. So, I think it behooves us as hematologists to understand the disease process and how it affects patients. Because now we have a real option for them in terms of symptom control, reducing the size of the spleen, and extending survival.

However, there are still great challenges. We know that ruxolitinib can cause cytopenias. It must be remembered that patients who went on these trials were often already anemic, and many were transfusion dependent. That is not an exclusion to giving JAK1/JAK2 inhibitors like ruxolitinib. In fact, in those clinical trials, although the hemoglobin decreased to a nadir about 8 to 12 weeks after starting therapy in myelofibrosis with ruxolitinib, it came back up close to baseline. That might be due to shrinkage of the spleen, control of cytokines, less anemia, or chronic inflammation, but for whatever reason, the anemia did get worse at the beginning but then better later on. So, the important message there is anemia should not stay the hand of a hematologist in treating a patient with this drug. However, it does bring up the difficulty of making sure our patient is aware that they may actually require transfusion after starting this therapy, and at least for some finite period of time.

The other thing is that although ruxolitinib is clearly an advance in terms of the quality of life for our patients, and I believe it&rsquo;s showing a benefit in terms of survival as well, it&rsquo;s not a therapy that apparently is targeting the very stem cell here. It has not changed the natural history of the disease.

So, I think we need to really learn a lot more about the pathophysiology and what&rsquo;s the underlying cause of myelofibrosis in order to better target it with therapies that are coming down the line.

The Clinical Management of Primary Myelofibrosis

Case Presentation: Primary Myelofibrosis

Presentation and Risk Stratification of Myelofibrosis

Identifying a Diagnosis of Myelofibrosis

COMFORT Trials: Initiating Therapy for Myelofibrosis

MF: Undertreatment Risks and Markers for Progression

Significance of Ruxolitinib and Unmet Needs in MF