The MPNclinic Roundtable: How I diagnose MPNs.

MPN is a rare disease, debilitating and potentially deadly. There is no cure beyond risky SCT, but with proper management a normal life is possible for most patients. The key to therapy is accurate and early diagnosis.

Accurate diagnosis has become more difficult in face of rapidly changing disease definitions, advances in genetics and molecular biology and the discovery of new mutations and pathways affecting each of the myeloproliferative neoplasm phenotypes.   There is no group more qualified to review and discuss MPN diagnosis than the physicians gathered at this MPNclinic Roundtable:  “How I diagnose MPNs.” The meeting took place on June 7, 2013, at 7:00 AM PST, via conference call at the Mayo Clinic, Scottsdale facility.

I am honored and pleased to present,  Dr. Richard Silver, Weill-Cornell, Dr. Hans Hasselbalch, Roskilde Hospital, University of Copenhagen, Dr. Attilio Orazi, Weill-Cornell, Dr. Jason Gotlib, Stanford, Dr. Srdan Verstovsek, M.D. Anderson, Dr. Ross Levine, Sloan-Kettering, and our Host and first speaker,  Dr. Ruben Mesa, Mayo Clinic.

MESA

So, then, thank you very much for including me in the call with such an expert panel.  Clearly we will be making a few opening comments and then passing along to my colleague Attilio Orazi, a world expert hematopathologist involved with the diagnosis.  As a whole we utilize the diagnostic criteria of WHO from 2008 for the diagnosis of MPNs, each of the main diseases we discuss, whether they be ET, PV or MF.  We use World Health Organization (WHO) 2008 criteria, the standard guidelines of our field.  Obviously in our discussion there are times when these guidelines aren’t perfect for certain individuals/certain cases, but in general they are a useful set of guidelines.

There’s also an additional set of guidelines we use when individuals are moving from ET or PV to post ET or PV myelofibrosis. Many who are reading the MPNforum are individuals who have already been diagnosed.  These criteria rely on a couple of differing angles.

The first criterion is to try to establish that there is in fact a blood disease as opposed to a secondary cause of blood count changes.  The second is to try to determine among the blood diseases, the reason for the abnormality, whether there’s an increase in the red cell count, an increase in the platelet count, white count or spleen or what have you.

So we look at the diagnosis. What are the issues that are most pertinent to patients with an MPN?  The first step is for a clinician, a provider, a physician to suspect there is an abnormality, and that can be either from an abnormal blood test at the time of a physical exam, ER visit or unexpected event like bleeding or thrombosis, or an unexpected finding like the spleen is enlarged.  Lab studies are drawn of the blood and we look for changes in the counts of the blood in the chemistries, including markers of white blood cells, such as LDH, chemistry studies that have an impact on blood cells such as iron cells and erythropoietin levels; we look for genetic mutations that have been identified with MPNs, most specifically the JAK2 v617f mutation, and many individuals, most with PV and about half with ET and MF have it. We look for less common mutations on a more select basis.

And for MPN patients overall, at least I can speak for myself, that barring weird circumstances, a bone marrow aspirate and biopsy with cytogenetics at baseline is a very helpful test, sometimes for diagnosis, always to get a sense of prognosis and confirming the diagnosis, by assessing other aspects of the disease, to look at issues such as scarring of the bone marrow, present or absent, any chromosomal changes such as increase or decrease in blasts.

So you doctors put all these pieces together: physical exam, changes in the marrow, changes in the blood counts and how the individual feels, and are able to properly put people into the category:  1. Do they have a blood disease? and  2. If they have an MPN, which of the MPNs do they most accurately have to diagnose? Perhaps with that I can hand over to Attilio next. But Attilio Orazi is a hematopatholgist who can tell us what we look for in the bone marrow…

ZHEN:  Ruben, right before we turn it over to Attilio … What about the patient who has NOT yet been diagnosed? For a physician who sees a patient, what might alert him or her that this patient might have an MPN? … Then we can get on to the bone marrow with Dr. Orazi. Would that work?

MESA

Sure.  Serge, why don’t you jump in on that one?  A person with a possible but undiagnosed MPN.

VERSTOVSEK

Well, we would like to assess the patient for the presence of an MPN by looking at several issues. First, one looks at the blood cell count, the high or low values for hemoglobin, white blood cell count and/or platelets count. It is also informative if there are unusual white blood cell types present in the blood, for example blasts, metamyelocytes, myelocytes, or similar. These types are not normally present in the blood. In some patients one can find abnormal size and shapes of red blood cells too.

It’s not just a blood examination, however. Diagnostic criteria for myelofibrosis include abnormality in lactate dehydrogenase (LDH), which is measured in the blood as a part of a metabolic panel (blood chemistry). LDH comes out of dying cells, and is typically high in myelofibrosis, unlike in essential thrombocythemia and polycythemia vera.  Then there is the physical exam. Palpable enlargement of the spleen (splenomegaly) is one of the minor diagnostic factors for myelofibrosis, but it can also be seen in essential thrombocythemia and polycythemia vera.

A comprehensive examination of the patient by physical exam, by blood count, and by blood chemistry testing may raise a suspicion for an MPN, leading to the bone marrow biopsy that is required to make fully informed decisions on what if any of the MPNs is present.

GOTLIB

I would just add that as hematologists, we talk about what an internal medicine doctor needs to recognize, or when to suspect that an MPN may be present.  Internists should be familiar with the differential diagnosis for a patient that has a high red blood cell count, or an elevated white blood cell or platelet count.  One needs to distinguish between a reactive cause for high blood counts before settling on a primary (or bone marrow derived) disorder as a cause of elevated blood counts, since the latter are much more uncommon.  An elevated red blood cell count is a defining feature of polycythemia vera, and a high white blood cell count or elevated platelet count can be observed with any of the MPNs. This is just a general framework that internal medicine physicians can start with as they think about referring a patient to a hematologist.

ZHEN

Just a clarification.  Reactive versus essential would result from high counts deriving from some secondary trauma instead of the underlying disease, is that right?

GOTLIB

Correct.  What I’m referring to is trying to distinguish causes outside of the bone marrow that in some way contribute to high counts versus a problem arising from the bone marrow, such as an MPN.   Examples of general conditions that can lead to an increased platelet count include iron deficiency, inflammatory disorders, and infections.  These are just a few general categories that should be ruled out before rushing to the diagnosis of an MPN since these are much more common in the general population.

SILVER

You can add cancer to that. When you look at the spectrum of platelet count elevations, we find as an initial sign looking at platelet counts over 350,000, a very sharp cutoff point. Over 99% of patients would be in the non-myeloproliferative category. We have to approach the patient as an internist, not as a hematologist. Three very common causes of thrombocytosis are inflammation, blood loss, and cancer.  It’s a very good screen in fact.  Only after these are excluded should one think of an MPN from the standpoint of a platelet count.

I think what a family practitioner should look for in a blood count that might suggest an abnormality that might lead down the line to a diagnosis of an MPN might be elevated platelet count.

The other abnormalities that you read about in text books that we see regularly, like nucleated red cells and immature granulocytes are really a late manifestation of these illnesses, and then any abnormality of the white count, which is usually a later sight than an earlier one.  Even with patients with PV we might discuss later, since Hans is on the phone, many patients with PV who present with hematocrits that are not significantly increased above normal and really require sophisticated testing to make an appropriate diagnosis.

Serge also mentioned blood chemistries.  LDH and uric acid are very good screening tools .20% of PV patients present with gout and elevated levels of uric acid and that’s very helpful diagnostically, as is an increased LDH.

ZHEN:  What symptoms are a patient with an undiagnosed MPN likely to experience that brings him to a physician?   Dr. Hasselbalch?

HASSELBALCH

Thanks so much for the opportunity to join you. I think it is very important to stress that in a family practice there are some very non-specific symptoms you have to pay attention to. For instance, aquagenic pruritus with intense itching after bathing is a highly characteristic symptom in many patients – in particular in PV.  When this symptom is present in a patient with an elevated hematocrit and/or elevated WBC or elevated platelets, an MPN should be seriously considered – and add to this an elevated LDH plus or minus an elevated uric acid concentration.

Other non-specific symptoms that a GP and an internist might come upon are, for instance, headache, dizziness, and visual disturbances. And again — when combined with abnormal blood biochemistry as outlined, these symptoms should alert the GP of an MPN.  Symptoms of microcirculatory disturbances in fingers and toes are extremely common.

All these symptoms may have bothered the patient for years before diagnosis and severely impaired quality of life. The GP and the internist may not take notice of these symptoms and findings, being interpreted as “stress “ (headache, dizziness) or being “reactive” (elevated WBC or platelets). But if you have these elevated counts without obvious causes then it is of utmost importance to call the hematologist.

Fatigue in the myelofibrosis stage, but also seen in the early cancer stage (ET/PV), may be massive and in combination with night sweats, weight loss and abnormal biochemistry (immature white cells and red cells in peripheral blood) should raise suspicion of an MPN.  The elevated WBC and platelets leukocytosis should be persistent – and without elevated CRP [C-reactive protein]. I always advise my colleagues to repeat the test one or two weeks later, and if still elevated refer to a hematologist. If we can discover and diagnose the patients at a much earlier stage, we may also have a much better opportunity to reduce the risk of complications (thrombosis/hemorrhage).

It is very important to stress that we should revive the red cell mass (RCM) investigation, since without performing an estimation of the RCM we misclassify up to 60% of our ET patients.  A large number of JAK2^v617f positive ET patients actually have PV, as assessed by an expanded red cell mass.  I think it is very important to underscore this and use this investigation in all ET patients – at least those being JAK^2v617f positive. The consequences of not doing a RCM estimation is that a proportion of ET patients will not be adequately treated with phlebotomies to secure a hematocrit of <0.42 in women and <0.45 in men.

ZHEN: Dr. Orazi, how then would we differentiate prefibrotic-PMF from ET patients?

ORAZI

Before we do that, I would like to return to the important issue of how we separate ET from early phase PV. In a JAK2 positive patient with thrombocytosis, the first issue is indeed to exclude early phase of PV. Then you exclude prefibrotic-PMF.  You basically have to approach the differential diagnosis in a systematic way.

We know that ET can be confused with PV and sometimes with early phase PMF. Suspicions of PV should be particularly high in patients with low MCV, low ferritin level or lacking stainable bone marrow iron. Separation is possible by examination of bone marrow biopsy because the typical appearance of PV is different from that of ET. In ET the only lineage that is really increased is the megakaryocytic lineage. Contrarily, in PV you would see increased cellularity in the three cell lines (erythroid, granulocytic, megakaryocytic). Also the megakaryocytes look different in PV than in ET, they look more pleomorphic and if you put everything together and you have an open communication with your clinical hematologist counterpart you can come to the conclusion that your patient is not ET, but it could be early phase PV particularly if he lacks stable iron and if the ferritin is low and then we can order additional steps (e.g., red cell mass, EPO).   In this important differential diagnosis, the bone marrow biopsy doesn’t stand alone, but within a diagnostic algorithm it plays a crucial role.

LEVINE

The issue is how genetic markers help. At the current time, we have no genetic markers that let us choose between PV, ET and PMF and this remains a clinical and pathological distinction … With a lot of effort we can identify genetic markers that impact diagnosis, progression and prognosis, but we don’t have it as yet. Only with a lot more effort together can we figure out how to use genetic markers to inform clinical care.

ZHEN: How often then should a BMB biopsy be performed?

SILVER

So our philosophy is – and I’m tying my philosophy to that of Attilio since we have evolved together over the seven or eight years since he’s been here … talking over each case over the microscope.

It’s critical for the hematopathologist to sit with a clinical hematologist and review cases. Everything he says I subscribe to and everything I say from a morphological standpoint he subscribes to since we evolved this together.

It’s unique and unfortunate that in other sites this close collaboration is not standard. So having said that, I believe that every patient with an MPN should have a bone marrow biopsy at diagnosis. That’s because of two reasons:  First I think that everybody is trying to move earlier and earlier into treatment – and that makes sense. We don’t treat breast cancer when it’s metastatic; we shouldn’t be treating MF when patients present with big spleens and fibrotic marrow.  So the bone marrow establishes not only morphologic criteria that Attilio mentions but second it gives you a baseline for a foundation for how how much fibrosis exists at diagnosis when the patient is first seen, such as the degree of reticulin grading and collagen.  And what we found was that this is a relatively early manifestation, not a late manifestation.  It’s very important to recognize this because the Cervantes scoring system was not based on bone marrow findings or splenomegaly   So that’s where the money is. Willie Sutton, the famous bank robber, when asked why he continually robs banks, said, “That’s because that’s where the money is.”  You really have to look at bone marrow and study the progression of the disease and it should be studied early and frequently.

Second, I think that we feel that while morphology of bone marrow can be helpful – Attilio said this quickly so I want to reinforce it — it’s not a stand-alone.  Nothing is a stand-alone from a diagnostic standpoint.

We espouse chromium-51 red cell mass, but even in a very experienced isotope lab you can get a borderline value, so that has to be interpreted just like every other laboratory test we use in medicine.

The bone marrow is helpful in diagnosis.  I don’t think we would use it, per se, to make the diagnosis, but it can be very persuasive in a borderline case, so we could answer the question.  Basically we believe it should be done at diagnosis and then at periodic intervals during the course of the illness, much like a cardiologist or GI specialist would use their respective EKGs or endoscopies to follow the course of the illness.

ZHEN: We talked a little bit about differentiation. How can we determine if a patient you see is presenting with ET or pre-myelofibrosis?

ORAZI

We have not discussed the role of bone marrow examination really yet.  We discussed the reason to do it and its frequency but we have not told the patient the role of the bone marrow examination, what it can do for you and not do. So I think it would be important to spend a few minutes on that.  Dr. Silver has already explained the reason to do a baseline marrow and follow up marrow. There are other reasons. One is the one we have just briefly mentioned, the issue of differential diagnosis.

So the bone marrow is not only useful to confirm a suspected diagnosis, but it is also helpful to differentiate myeloproliferative disease from other conditions.  Some of these fall under the term of “unsuspected pathology.” Examples include myelodysplastic syndrome or myeloproliferative/ myelodysplastic neoplasms, the so-called “overlap syndromes.” You know there are anemic patients with thrombocytosis who have a myelodysplastic syndrome with a chromosomal deletion of chromosome 5q. Without a bone marrow examination, these patients can be confused with essential thrombocythemia but the bone marrow morphology is completely different.

There are also other entities, e.g.,  refractory anemia with ring sideroblasts and thrombocytosis and myelodysplastic syndrome associated with inversion of chromosome 3, which can rarely be confused with MPN and are separable by bone marrow examination.  The hematopathologist reviews the bone marrow at the microscope and puts everything together including clinical information and other laboratory results (e.g., LDH level) and generates an “integrated” pathology report.

Now going back to Zhen’s question in relation to how do you separate essential thrombocythemia from early phase primary myelofibrosis, I think it is important to put this question in the right clinical context.

Pathologists dealing with bone marrow really don’t function separately from the clinical service. Unfortunately there is a common misunderstanding that the WHO classification is a morphological classification. It is not.   It is based on integration of clinical findings, pathological findings including genetics and cytogenetics, and morphology.

So going back to the issue of ET versus early PMF, we pay particular attention to several parameters useful to separate ET from PMF, none of which is completely diagnostic on its own. These include marrow cellularity, its cellular composition, presence or absence of cellular atypia (e.g., in the megakaryocytes). Then we assess the bone marrow stroma by staining for reticulum to look at the amount of fibrosis and so on. All of this is integrated with clinical findings (e.g., presence/absence of splenomegaly), laboratory findings (CBC results, peripheral blood smear review, LDH level) and clinical history.  If you do this carefully and discuss your opinion with the clinical hematologist, I think consensus in the diagnosis can be reached in the vast majority of patients including the separation of early phase PMF from ET.

Dr. Silver may want to comment on this also since in our institution we do that basically on a daily basis.  With this said, realistically, there are 20% or so of patients where such separation is not possible at disease beginning. In these, obviously the bone marrow will have to be repeated – at least once a year or maybe more frequently if CBC or clinical findings worsen.

There is a misconception out there that the pathologist looks at the slide and says, “Ha! This is what it is!”  It is not that simple. You have to integrate all the data and discuss it with your clinical hematology counterpart and then come up with a consensus diagnosis.

MESA

Perhaps I can make a critical comment on this, because this is an area that can create both great anxiety and confusion   I think what Attilio describes, that clearly our colleague Juergen Thiele and our German colleagues have been at the forefront of this discussion for many years, is that within the context of what we historically used to call ET there are probably two different groups of people. And one group of those individuals probably have a bit more problematic disease more likely to change to the much more classic or overt forms of MF. They have different marrow changes that define them, in the marrow they have different clinical features.

Attaching the term early myelofibrosis creates a tremendous amount of anxiety for many. Is this a more difficult part of ET — should we call it an early form of myelofibrosis or something else? Again there are a lot of emotionally charged aspects in terms of the naming.

It is not, necessarily, the same entity as we call the classic or more problematic MF, and if patients get this title or label for this disease, they will look online or read about how patients with MF, advanced MF, behave. So in my estimation as a clinician I think the information they present is true. There are different groups there.  I think there’s a lot of emotional reaction to how it’s labeled, with how it is classified or called, but there is a distinction and it relies both on the clinical behavior as well as the bone marrow. It is really neither in isolation.

ZHEN: To move off this slightly, to Serge. How do you avoid errors in diagnosis due to the impact therapeutic agents like Jakafi or interferon may have in altering  the morphology or proliferative aspects of the case?   When looking at a patient, how do you know whether or not you’re looking at the impact of the drug itself on the condition?

VERSTOVSEK

You point to a problem that we often see in my practice. I work at the referral center and we see patients who have received therapies in the past or are on a therapy on presentation to us.  I would mention a case of hydroxyurea, which is widely used as therapy for essential thrombocythemia, polycythemia vera, and myelofibrosis, and can affect the size and shape of bone marrow cells on its own. Many times we have difficulties in evaluating a patient’s bone marrow disease status due to possible presence of “therapeutic effect” of hydroxyurea.

This is particularly a problem if one looks only at the bone marrow biopsy to detect which disease a patient has, and whether the disease has changed. What is important, and has already been highlighted, is that not just bone marrow biopsy is important but all other tests and procedures (blood tests and physical exam) which together lead to a proper diagnosis.

Many times we request from a referring doctor the initial bone marrow biopsy samples, for us to be able to properly evaluate the  patient’s current situation and advise accordingly, whether there is a change due to therapy or due to a progression of the diseases.  I think you pointed out one of the main problems that we deal with during consultations with patients who are in therapy or who have been on therapy for long time.

ZHEN: The issue for physicians when they’re seeing a patient is how do they proceed. When is it urgent to refer to a hematologist and when can they simply observe prior to a referral? Jason, what do you think?

GOTLIB

I think that internists have to recognize that high blood counts are as concerning as low blood counts. A while ago I had a patient who had a high hemoglobin and persistent headaches, lasting for many months, but apparently her internist may not have been sufficiently impressed by the abnormal labs.

For example, PV patients with high hemoglobins who are left untreated are at a higher risk of thrombotic events, and symptoms such as headaches or pruritis, associated with impaired quality of life.   So recognizing what the differential diagnosis is for high platelets, high red count, and a high white blood count is essential.  I don’t think you can always expect the internist to get the LDH and uric acid, which are more aligned with what the hematologist would obtain.

I think the important thing is to know when to make the referral.  The internist shouldn’t be expected to know the WHO criteria for an MPN.  However, they should have the basic understanding to distinguish between reactive and non-reactive causes of increased blood counts and refer to a hematologist accordingly.

We should talk about some of the major morbidities associated with MPNs, in particular, clotting and bleeding.  If a patient presents with a bleeding or clotting event and abnormal blood counts, that should be a signal for a referral to a hematologist.  Sometimes it may happen in an outpatient setting or when a patient is admitted to the hospital.  This is an example of when the hematologist needs to be consulted immediately to get a knowledgeable opinion.

ZHEN: To wrap up… circle around, any final words to the physician in family practice? Advice that might make sense to recognize and treat MPNs.

HASSELBALCH

Yes, sure, it bears much repetition.  In the family practice it is a combination of symptoms, physical findings and abnormal biochemistry, non-specific symptoms such as itching (pruritus), headache, dizziness or visual disturbances when combined with abnormal blood chemistry such as elevated white blood count, elevated platelet count and elevated hematocrit.

When you have excluded infection, inflammation and cancer, then you have to be alert that this might be an MPN. So this is my main message as also stressed by others. When a patient has thrombosis with an elevated platelet count and/or an elevated leukocyte together with a virtually normal C-reactive protein (CRP) – that’s a suspicion of an MPN.

GOTLIB

I’ve also had referrals where an internist has laid hands on the patient and detected a large spleen in the setting of normal or mildly abnormal counts.  Splenomegaly is not specific to, but can be one physical examination finding consistent with, an MPN and could also be a trigger for a referral to a hematologist. I’ve been very impressed by internists who have been able to generate the initial suspicion of such a diagnosis based on their finding of splenomegaly on examination.

HASSELBALCH

May I add just another comment which we have not stressed that much — that is the patient with an elevated platelet count above 1,000,000-1,500,000 and hemorrhage that is a mnemonic for MPN disease.

MESA

Fantastic discussion. Let me just add to that.  One of the most prominent features as the population ages and the population median age becomes older in North America, Europe and elsewhere, is that diseases of the blood like MPNs continue to increase in their prevalence. It’s important to increase awareness among physicians that a blood disease might be in their patients who are having symptoms – splenomegaly, blood count changes, whatever. It is important because their initial suspicion is vital to find such an illness … not only an MPN, but all the other blood diseases of the marrow.

 

LEVINE

I think the biggest challenge is how are we going to move the genetics as a marker of the disease to actually using it to help make decisions for our patients? I think this is something that we in the field need to do a better job, because patients in the field are going to get more and more data back and we have to be equipped to know how to educate ourselves, our patients and the referring clinician about how to use that data. We still have a long way to go.

SILVER

I agree with all the previous remarks. I’ll stand.

MESA

That sounds good.

ORAZI

One last word on bone marrow examination: The patient needs to be aware that we take this very seriously because we are aware that this is a very unpleasant experience and it keeps coming up as an issue from a patient point of view. However it provides very valuable information.

To maximize its diagnostic values, please pay attention to this. When a patient receives a pathology report, he/she should read it very carefully, ask questions if things are not clearly stated and remember that for the differentiation of early PMF from ET we rely not only on bone marrow morphology, but also on the presence/absence of key clinical and laboratory results (the so called minor criteria) and that their presence/absence should be indicated in the report.   These include hemoglobin value, LDH level, presence or absence of splenomegaly, and that the peripheral blood smear needs to be reviewed for the presence of immature myeloid and erythroid cells. Those are the four minor criteria and they must be integrated into the diagnosis. For a diagnosis of PMF it is necessary to fulfill at least two of these four criteria. Bone marrow morphology alone is not enough!  I think this is an important message to stress.

HASSELBALCH

In the context of when to do BMB:  In Denmark we treat the majority of ET and MPN patients with interferon. We know in three to five years, patients may pause interferon and they have a normal bone marrow architecture.  Another reason why we also stress doing bone marrow biopsies sequentially is to see the effect of interferon therapy.

MESA

I thought this format of focusing on a specific set of issues was probably the best format instead of trying to tackle an enormous spectrum … Very good … thank you all.

Thank you.