Current Management of the Myeloproliferative Disorders: A Case-Based Review

Current Management of the Myeloproliferative Disorders: A Case-Based Review

Rice, Lawrence

Context.-Properly managed, the myeloproliferative disorders are generally compatible with prolonged survival. Challenges to the hematologist include knowing when and how best to intervene to prevent and manage complications. The cytoreductive agent of choice for these disorders is currently hydroxyurea, emerging from randomized trials beginning with those of the Polycythemia Vera Study Group.

Objective.-To examine the roles and shortcomings of interventions (including hydroxyurea, antiplatelet agents, anagrelide, interferon, thalidomide, alkylating agents, cell cytopheresis, erythropoietins, splenectomy, bone marrow transplantation, and imatinib) for myeloproliferative disorders.

Data Sources.-This report uses actual case histories to illustrate the roles and shortcomings of these interventions.

Conclusions.-Beyond phlebotomy for polycythemia vera, patients with polycythemia vera and essential thrombocythemia can be stratified by their risk for thrombosis, which guides the institution of cytoreductive therapies. High-risk patients generally benefit from cytoreductive therapy, and hydroxyurea has emerged as the agent of choice, because alkylating agents (and P32) have high leukemogenic potentials. Anagrelide and interferon are second-line agents. The addition of low-dose aspirin is beneficial for most, helping to prevent arterial thrombotic complications. Therapy in any of these disorders should be tailored to the unique characteristics of the individual patient. With myelofibrosis, therapeutic options run the gamut from observation, erythropoietic stimulators, cytotoxic agents, splenectomy, and bone marrow transplantation. Thalidomide and imatinib have shown some utility. Future challenges are the refinement of individualized treatment strategies and the development of targeted therapies based on rapidly expanding understanding of the molecular perturbations in these disorders.

(Arch Pathol Lab Med. 2006;130:1151-1156)

The management of patients with myeloproliferative disorders (MPDs) can be challenging to the clinical hematologist. Properly managed, these disorders are generally compatible with survival of many years: Survival may be nearly equal to that of age-matched normals with essential thrombocythemia (ET),1 and median survival greater than 15 years may be expected with polycythemia vera (P vera)2 (contrasting to survival of 18 months to 4 years reported in untreated patients). The median survival of patients with myelofibrosis (MF) has been shorter, but this disease is more variable in its presentation and rate of progression, so, here too, many patients enjoy prolonged relatively asymptomatic survival even without therapy. Challenges thus include knowing whether and when to intervene and then to consider which agent is most likely to accomplish the goals in a given patient with the least risk for early or (particularly important with these diseases) remote harm.

Goals of treatment are to prevent serious complications, to treat complications when they occur, and thus to improve quality of life and survival. The magnitude of predisposition for any specific complication varies with each specific disorder, yet these MPOs share propensities for the same type of complications. For example, thromboembolic complications (arterial more than venous) are the leading cause of morbidity and mortality in P vera and ET, and these complications also pose some threat in MF. All MPDs share some predisposition for bleeding complications, whether because of platelet dysfunction, thrombocytopenia, extreme thrombocytosis, or anatomic problems such as esophagogastric varices or portal gastropathy. Complications related to progressive marrow fibrosis and pancytopenia can occur in all these disorders. Evolution to acute leukemia is more common with MF, but the small intrinsic tendency for P vera and ET to undergo such transformation can be multiplied by applying leukemogenic therapies, and this phenomenon has major bearing on current treatment choices.

In this review, accepted principles and unresolved issues of management are discussed using actual patient histories for illustration. Although the following discussion is subdivided by diagnosis, it is necessary to read all sections to get the full flavor of treatment issues in any individual disease. As examples, splenectomy is discussed in detail under MF, but many of the same issues apply in the spent phase of P vera; leukemogenicity of cytotoxic therapies is discussed under P vera, and this applies to all MPDs. The scope of this review does not permit detailed discussion of management of specific uncommon complications such as portal vein thrombosis, pleural/peritoneal metaplastic implants, or issues surrounding pregnancy or surgery.


Case 1

A 49-year-old woman presented in 1987 with an acute inferior myocardial infarction. Her hematocrit was 51%, leukocytes’15000 cells/mm1, and platelets 1240000/mm^sup 3^. P vera was diagnosed after a work-up that demonstrated elevated red cell mass, elevated leukocyte alkaline phosphatase score, hypercellular bone marrow without iron, and cytogenetics revealing trisomy 12. The patient was treated with phlebotomies and briefly uracil mustard, later hydroxyurea. Pruritus, initially occurring only after warm showers, became severe and constant 3 years after diagnosis. Attempts at control included trials of 6-mercaptopurine, interferon alfa, 6-thioguanine, benzodiazepines, antihistamines, phenothiazines, and cyproheptadine. After 2 years, back on hydroxyurea, pruritus ceased to be a major problem. The patient experienced ischemic cardiac pains whenever platelets rose to 600000/mm^sup 3^, necessitating adjustments to cytotoxic drug dose. She suffered herpes zoster in 1996, then for 5 years blood counts remained satisfactory without therapy. Methotrexate, prednisone, and later etanercept were instituted in mid-2002 for worsening rheumatoid arthritis. At that time, teardrop poikilocytes were appreciated on blood smear, and the spleen was first palpable in early 2003. By mid-2004, red cell transfusion was needed for symptomatic anemia (hemoglobin 8.2 g/dL), leukocytes were 58000 cells/mm1 with 79% neutrophils and 1% blasts, and there was significant discomfort from the spleen, now 10 cm below the costal margin. Resumption of hydroxyurea proved ineffective. By March 2005, hemoglobin was 12.2 g/dL, platelets 111 000/mm^sup 3^, and leukocytes 25800 cells/mm^sup 3^ with 48% neutrophils, 3% myelocytes, 17% basophils, 20% eosinophils, and 6% blasts, and the spleen was 15 cm below the costal margin. Thalidomide and prednisone were started, with marked improvement in spleen size and symptoms. In late 2005, hemoglobin was 11.3 g/dL, leukocytes 4800/mm^sup 3^, and platelets 70 000/mm^sup 3^.

Phlebotomy.-Therapeutic blood-letting has been used since ancient times. The father of the United States, George Washington, was bled at least 5 pints of blood for a throat infection on the day he died; it might please him that this therapy lives on as a cornerstone for modern treatment of polycythemic disorders (as well as for iron overload). The rationale for phlebotomy in P vera is obvious, and it quickly became the therapy of choice once the disease was recognized in the early 20th century.3 Confirming its value, clinicians could observe rapid clearing of central nervous system symptoms, and scientists could point to rapid changes in cerebral blood flow as whole blood viscosity is rapidly reduced.4 It is interesting to note that there were skeptics who argued against long-term phlebotomy, believing that symptoms of tissue iron deficiency (apart from anemia) and the emergence of hypermetabolic symptoms would become disabling. We now know that the former problem is rarely clinically important and the latter is part of the natural evolution of the disease and not related to phlebotomy.

In the landmark first study of the Polycythemia Vera Study Group (PVSG 01)/431 patients were randomized between 1967 and 1974 to phlebotomy alone, phlebotomy plus radiophosphorus, or phlebotomy plus chlorambucil. Patients in the phlebotomy alone arm enjoyed the longest median survival of 12.6 years (statistically superior to the chlorambucil arm), owing to the emergence of late leukemias and other cancers in the cytotoxic treatment arms. The thrombosis rate, however, was highest in the phlebotomy arm in the initial 3 years of follow-up (about one third of patients); this was influenced by an initial phlebotomy target of less than 52% hematocrit, later changed to 45%. Thromboses were most often arterial (especially cerebrovascular accidents). The main risk factors for thromboses were age greater than 60 to 65 years and prior thromboses (present in up to 40% of newly diagnosed patients). The importance of the degree of thrombocytosis as a risk factor remains somewhat controversial. Other known risk factors for cardiovascular disease, such as diabetes and smoking, also come into play in defining a patient’s degree of risk.6 The current consensus is that lowrisk patients can be managed with phlebotomy alone to maintain hematocrit less than 45% in men and 42% in women; high-risk patients benefit from the addition of cytoreductive therapies. Ordinarily, 1 pint of blood is removed weekly, but intensity can be increased or decreased depending on the degree of erythrocytosis and symptoms.

Cytoreductive Therapies.-Many, probably most, P vera patients require cytoreductive therapy at some time in their disease course, most often because of risk factors for thrombosis on phlebotomy alone or to control extreme thrombocytosis. The decision in case 1 was made easy by her extreme thrombocytosis and cardiac ischemic problems, but decisions in other patients are often less clear. The first cytoreductive agents used for P vera were P32 and alkylating agents (especially busulfan)/just as uracil mustard was first used in case 1. These are quite effective in reducing platelet count, reducing phlebotomy requirements, and reducing leukocyte count in the few patients in whom this is desired (generally, white counts greater than 50000 cells/mm3 with hypermetabolic symptoms). In the PVSG randomized trial referred to previously, death because of acute leukemia emerged in 10% and 13% with P32 and chlorambucil, respectively, after median followup of 16 years. Because of this substantial leukemogenic potential (the rate was less than 2% with phlebotomy alone), these agents are now rarely used for P vera (or other MPDs), reserved for unusual cases (refractory to preferred cytotoxic agents, or elderly patients with comorbidities, poor venous access, and/or difficulties with follow-up).

Hydroxyurea.-Hydroxyurea is an antimetabolite inhibiting ribonucleoside reductase that has become the cytotoxic agent of choice in P vera owing to its efficacy and safety. PVSG 08 demonstrated its efficacy in controlling red blood cell and platelet counts within 12 weeks in the vast majority of patients, with reduction in thrombotic complications to one third of those seen with the historical phlebotomy only group.5 The major side effect is myelosuppression, especially leukopenia, but it is uncommon for this to be severe in adequately monitored patients (typically monitor every month or two once a stable effect is achieved) and cytopenias that may occur reverse rapidly because of the drug’s short-lived effects. Another uncommon side effect is leg ulcers. Hydroxyurea’s leukemogenic potential or lack thereof remains controversial. In PVSG 08, 51 patients without prior exposure to myelosuppressive agents took hydroxyurea a median of 6.2 years and had a leukemia incidence of 5.9% after 11 years, not statistically different from phlebotomy alone. It is clear that if there is any increased risk for acute leukemia with hydroxyurea therapy in MPD patients, the increase is small,56 and no excess leukemic risk has been observed with hydroxyurea in other situations such as sickle cell anemia.

Interferon; Anagrelide.-These medications are discussed in the “Essential Thrombocythemia” section. Interferon can control blood counts in most patients with P vera, but its use is limited by side effects and cost. Anagrelide is very effective for thrombocytosis in P vera and can affect red cell proliferation as well, but it too is limited by its side effect profile.

Aspirin; Antiplatelet Agents.-With most deaths in P vera being due to arterial thrombotic disease and with thrombocytosis a common feature, the rationale for antiplatelet agents is clear. In 1977 the PVSG 05 study randomized 150 phlebotomized patients to receive 3 times daily aspirin 300 mg and dipyridamole 75 mg or to receive P32.5 There was no reduction of thrombotic complications in patients receiving aspirin (7 severe events vs 2), but gastrointestinal bleeding episodes requiring hospitalization and transfusion were increased (6 events vs none). Hemorrhage was particularly common in patients with extreme thrombocytosis. Because such high doses of aspirin are known to diminish endothelial production of thromboprotective prostacyclin in addition to platelet-derived prothrombotic thromboxane A2, lower doses of aspirin (100 mg daily) were used in the European Collaboration on Low-Dose Aspirin in Polycythemia trial, which randomized 518 mainly low-risk patients.7 Aspirin therapy resulted in a statistical decrease in the combined end point of cardiovascular death, myocardial infarction, stroke, and venous thromboembolism, with only a mild trend for increased bleeding. Low-dose aspirin is also extremely effective for microvascular symptoms such as erythromelalgia. Aspirin in low dose should now be standard therapy for most patients with P vera (and probably ET), excepting those with bleeding propensities. The addition of other antiplatelet agents (eg, clopidogrel) seems reasonable in patients who have suffered arterial thromboembolic events.

Imatinib; Future Agents.-JAK2 tyrosine kinase mutations have recently been uncovered in the great majority of patients with P vera and in significant numbers of patients with other MPDs. Imatinib, developed for its efficacy against the Bcr-Abl tyrosine kinase, has some effect against the JAK2 kinase and may have some clinical efficacy in P vera.s It is anticipated that the day is close for targeted therapies against identified pathogenic mutations, leading to highly effective and nontoxic therapythe availability of such agents could render moot many of the current controversies of who, when, and how to treat.


Case 2

A 21-year-old college student was incidentally discovered to have a platelet count of 890000/mm^sub 3^ in 1989. The diagnosis of ET was established after an evaluation that included normal leukocyte alkaline phosphatase score, normal serum ferritin, “hyperaggregable” platelets by aggregation panel, computed tomography scan showing moderate splenomegaly, and bone marrow showing increased megakaryocytes in clusters and normal cytogenetics. The patient was treated with aspirin 81 mg daily. In 1998 and 2000 there were uncomplicated deliveries of healthy term infants. The highest platelet count was 1600000/mm^sup 3^ in mid-1992. In the last 2 years, her spleen size has varied from 5.5 to 8 cm below the costal margin, and hematocrit and leukocyte counts have been normal, platelets have varied from 638000 to 883000/mm^sup 3^, and blood smear shows moderate teardrop poikilocytes and 2% myelocytes. The patient has never had symptoms attributable to ET and has never received treatment beyond aspirin.

Case 3

In 1998, a 24-year-old woman had a platelet count of 967000/mm^sup 3^ at the time of cholecystectomy. Her bone marrow was hypercellular with increased megakaryocytes and absent iron stores. Platelet aggregation was markedly impaired to high doses of epinephrine and moderately impaired to adenosine diphosphate. Platelets remained greater than 800000/mm^sup 3^ after oral iron repletion. Chronic headaches worsened. Treatment was initiated with anagrelide in mid-1999. Through early 2001 the platelet count was controlled between 197000 and 321000/mm^sup 3^, but there were intermittent visual scotoma, paresthesias, and anagrelide-induced leg edema. At a time that platelets were 429000/mm^sup 3^ in 2001, the patient had a several hour episode of numbness of the left side of her body and face-low-dose aspirin was added at that time. In 2003, the patient stopped anagrelide while trying to conceive, and blood counts rose promptly (hematocrit 49%, platelets 930000/mm^sup 3^). Interferon alfa 3 million units subcutaneous daily was then given for the duration of pregnancy, with side effects mainly substantial skin dryness and itching. Platelets were controlled between 359000 and 534000/mm^sup 3^, and a healthy infant was delivered at term. When placed back on anagrelide, headaches increased and platelets were 600 to 700000/mm^sup 3^, necessitating addition of hydroxyurea. Symptoms improved with a platelet count of 550000/mm^sup 3^.

Case 4

A 22-year-old woman in excellent health, taking birth control pills since her recent marriage, presented with fulminant acute hepatic failure. She was encephalopathic and jaundiced, with ascites and hepatosplenomegaly. Imaging studies documented hepatic venous thromboses (BuddChiari syndrome). Initial blood counts revealed hematocrit 32%, leukocytes 15200 cells/mm^sup 3^ and platelets 243000/mm^sup 3^. The hematology department was asked to explore underlying hypercoagulability. No hereditary or acquired thrombophilic disorder was uncovered, except the blood smear showed moderate teardrops, elliptocytes, many giant and bizarre hypogranular platelets, and 5% myelocytes and metamyelocytes. An MPD was suspected, and bone marrow exam revealed hypercellularity, increased and dysplastic megakaryocytes, moderate fibrosis, and cytogenetics showing 2Oq deletion in one third of metaphases. With conservative measures, liver function stabilized, encephalopathy cleared, and platelets rose to greater than 700000/mm^sup 3^ as acute hypersplenism diminished. Anticoagulation was initiated (heparin, then warfarin, after the fall from international normalized ratio 4.5 at presentation) and hydroxyurea was given. Platelet counts and anticoagulant level were well controlled for 8 months, then orthotopic liver transplantation was performed. The patient is well 6 years after transplant with normal blood counts and normal liver function, taking anagrelide, warfarin, and antirejection medications.

“Watch and Wait”-The first issues that must be addressed in most patients presenting with an elevated platelet count are whether the etiology is reactive or a MPD, and if the latter, whether treatment is warranted. The evaluation performed in case 2 is illustrative to ruleout common causes of reactive thrombocytosis (sometimes extreme), such as iron deficiency, bleeding, occult cancer, and inflammatory disorders.1′ A bone marrow examination (as with P vera) is not a necessary part of the evaluation, but when MPD is a serious consideration one should perform a chromosome study to exclude the presence of atypical chronic myeloid leukemia.1″ Some reported series of ET include cases in which hematopoiesis is polyclonal, and the classification of such cases engenders controversy.

The decision to initiate treatment for ET is easy if the patient is symptomatic (case 3) or has already suffered a serious complication, but knowing when to intervene for asymptomatic patients can be difficult. Patients can be stratified for risk of thromboembolic complications just as with P vera, the major risk factors again being age older than 60 to 65 years and prior thrombosis.h As with P vera, the height of the platelet count is not a clear-cut indicator of thrombotic risk, but very high platelet counts (especially in excess of 1500000/mm^sup 3^) do correlate well with bleeding risks and with the development of platelet dysfunction including acquired type IIA von Willebrand disease.h Contrary to dogmas of 20 years ago, widespread use of automated platelet counters has revealed that ET is far from a rare disease, and cases 2, 3, and 4 attest that ET is particularly common in young women as well as in older individuals. Some data indicate that the prognosis is better in affected young women”; case 2 may be illustrative in that the patient has been living 16 years with her disease with no therapy and no symptoms; on the other hand, the specter of complications has emerged with progressive splenomegaly, case 2 illustrates the propensity to transform to MF and case 3 to P vera. Pregnancies were uncomplicated without therapy in case 2 and with interferon in case 3, but case series indicate there is risk to mother and fetus.12 Decision-making challenges are illustrated by case 4, a young woman whose first manifestation of ET was life-threatening hepatic vein thrombosis. She is 1 of 8 young patients that we have personally cared for whose presenting manifestation of ET or P vera was hepatic venous or portal vein thromboses (7 women; 6 are doing well, 3 after liver transplants and 1 after surgical gastric devascularization; L.R., unpublished data, 2005). The decision to treat case 4 after her complication was straightforward, but it is sobering to realize that had these patients presented before their complication, they may not have manifest accepted indications for treatment.

Once a decision to cytoreduce is made, evidence supports a platelet count target less than 600000/mm^sup 3^.10 -1 Some data and many anecdotes indicate that lower targets to the normal range (400000/mm^sup 3^ or less) may further reduce thrombotic complications.14 Current trends, facilitated by the availability of nonmutagenic agents, are to treat earlier and more aggressively.

Hydroxyurea and Anagrelide.-Hydroxyurea and anagrelide have been the main agents used for cytoreduction in ET, but hydroxyurea has emerged as the drug of first choice for most patients based on a randomized controlled trial against no cytoreductive therapy and a comparative randomized trial of the 2 agents. Anagrelide is a quinazolin originally developed as a platelet function inhibitor but found to reduce platelet counts because of predictable effects on megakaryocyte maturation. It is rapidly effective in lowering platelet counts in the great majority of patients with ET or other MPDs.1″ Common side effects include headaches, palpitations, orthostatic hypotension, diarrhea, nausea, and edema (rarely cardiomyopathy), which limit dosage and lead about 20% to discontinue therapy.

In an Italian trial of 114 high-risk patients randomized against no cytoreductive therapy and targeting platelet count less than 600000/mm^sup 3^ hydroxyurea reduced the risk of thrombosis from 24% to 3.6% for 27 months of median follow-up.13 In the United Kingdom, 809 high-risk patients given low-dose aspirin were randomized to receive hydroxyurea or anagrelide.”‘ Although the platelet count was well controlled with either agents, after more than 3 years of follow-up, the anagrelide arm had statistically more arterial thromboses (9.1% vs 4.2%), serious bleeding (5.4% vs 2%), transformation to MF (4% vs 1.2%), and drug discontinuation because of side effects (36.5% vs 20%), whereas venous thromboses were more common in the hydroxyurea arm (3.5% vs 1%).

Interferon alfa.-Interferons can control platelet counts and often proliferation of other cell lines in most patients with MPDs.17 Modest doses (3 million units 3 times weekly) may be adequate. Its utility is limited by side effects that are not dangerous but can greatly impact quality of life, such as depression, anorexia, asthenia, and the need for parenteral injections. These factors make it a “thirdline” agent in ET and other MPDs. As in case 3, it has some utility in pregnancy because of fears that hydroxyurea and anagrelide may engender teratogenic risks.12

Plateletpheresis.-Because the platelet-lowering effect is very transient (few days), plateletpheresis should be reserved for unusual situations and emergencies (eg, cerebrovascular or cardiac ischemia with extreme thrombocytosis).18 A rapid-acting myelosuppressive agent should generally be given with it, such as hydroxyurea or subcutaneous cytosine arabinoside.


Case 5

A 33-year-old woman was found to be anemic in 1997. In late 1998, the spleen was palpable 3 cm below the costal margin, hemoglobin was 8.7 g/dL, leukocytes 4200 cells/mm1 and platelets 321 000/mm^sup 3^. The blood smear showed marked teardrop poikilocytes and leukoerythroblastic changes. Bone marrow aspirate was dry, and biopsy showed MF; cytogenetic analysis was unsuccessful. The patient’s 6 siblings were human leukocyte antigen typed and 1 was identical. B6 and folic acid were given. By late 2002, the spleen had increased to 10 cm below the costal margin and hemoglobin declined to 8.0 g/dL. Serum erythropoietin was 312 units/mL; erythropoietin injections were administered but without response. By mid-2004, the spleen was 12.5 cm below the costal margin, hemoglobin 6.9 g/dL, leukocytes 3000 cells/mm^sup 3^ and platelets 129000/mm^sup 3^. The patient denied any symptoms at all, working long hours at her job, and refused to consider marrow transplant. Hydroxyurea was given in an attempt to shrink the spleen and improve blood counts but did not help. In late 2004, the patient admitted to being “light-headed at times,” hemoglobin was 5.9 g/dL and thalidomide was started with rapidly tapered prednisone. This therapy has had some success with the spleen reduced by 4.5 cm, hemoglobin 8.1 g/dL, and mild peripheral neuropathy.

Tailored Therapy.-All individuals with MPDs should have therapy individualized for their disease status and personal preferences, but this concept is most apparent with MF because this disease is more variable in its presentation and course than the related MPDs. Therapeutic strategies vary greatly depending on the complications encountered in a given patient, and it is very reasonable to adopt a “watch and wait” approach for the 20% to 30% of patients who are asymptomatic at presentation, illustrated by the first 4 years of case 5. This case conforms with reports that MF in young people may often have a more indolent course,1″ even though this patient had poor prognostic factors of anemia and (shortly after diagnosis) leukopenia.2″ A variety of symptoms and complications may emerge as the initial indications for treatment, including bleeding or thrombotic problems, manifestations of portal hypertension, systemic symptoms such as fever and weight loss, cardiopulmonary perturbations, or symptoms related to extramedullary hematopoietic depositions. Most commonly, the symptoms that first require therapy are related to anemia or splenomegaly.

Erythropoietic Stimulators.-Before assuming anemia is because of disease-associated marrow and splenic factors, correctible causes such as iron deficiency or autoimmune hemolytic anemia (which has increased incidence in these patients) should be considered. Vitamins (B6, folic acid) are benign interventions that have sometimes improved anemia. Androgen therapies, reported useful in the past, have been largely supplanted by erythropoietic hormones, although danazol continues to show utility in recent reports.21 Erythropoietic hormones (erythropoietin, darbepoietin) have reversed transfusion dependence and improved quality of life in close to half of anemic MF patients; inappropriately low endogenous erythropoietin levels predict response.22 Unfortunately, responses generally do not last more than 1 to 2 years.

Cytotoxic Agents.-These may have a role in controlling thrombocytosis, leukocytosis, systemic hypermetabolic symptoms, and occasionally splenomegaly. Although older reports used alkylating agents with modest success, most would avoid these agents now because of the risk of hastening leukemic transformation. Hydroxyurea is the preferred cytotoxic agent at present.21 Depending on the treatment goal, cytotoxic agents generally have only moderate success rates (illustrated in case 5), and with more advanced disease, the risk of exacerbating cytopenias is significant.

Splenectomy.-Nothing incites heated argument among expert hematologists more reliably than debating the wisdom of splenectomy in a given MF patient, proving again that opinions are strongest where data are weakest. (An attempt by the PVSG to randomize patients with symptomatic splenomegaly between splenectomy and melphalan accrued 8 patients in 10 years.) Some have advocated early splenectomy for patients presenting with massive splenomegaly, others would avoid splenectomy except when forced by the failure of alternative measures to achieve therapeutic goals, and other opinions fall in between. The best established indications for splenectomy in patients with very large spleens are refractory anatomic symptoms (discomfort, repetitive infarcts, early satiety), severe thrombocytopenia, transfusion-dependent anemia, and portal hypertension/esophagogastric varices.24 Complication rates have been high in the past, including postoperative bleeding, thrombosis, and infection, and later progressive liver dysfunction with hepatomegaly. Perioperative mortality near 10% is reported. Postsplenectomy median survival of 2 years has also been reported, so most would couch a decision for splenectomy in terms of improving quality of life rather than longevity. Splenic irradiation has been used in poor surgical candidates, but it variably and temporarily controls local pain, rarely improves cytopenias, is surprisingly myelotoxic, and makes future operative intervention difficult at best. (Low-dose radiation can be very effective for symptomatic metaplastic hematopoietic implants.)

Thalidomide.-Believed to have antiangiogenic and immunomodulatory properties, thalidomide has shown efficacy in improving anemia, thrombocytopenia, and splenomegaly (as in case 5). Reports, including 1 prospective study of 36 patients, find some response in about 50% often at low doses (high doses are poorly tolerated), aided by temporary use of low-dose corticosteroids.25

Bone Marrow Transplantation.-Allogeneic bone marrow transplant is the only therapeutic option affording the possibility of disease cure, but the early mortality and possibility of ongoing morbidity from graft-versus-host disease are daunting. It is difficult to recommend such intervention to even young people with P vera or ET given their prospects for long-term survival with low morbidity (and one is a poor transplant candidate once serious complications intervene). It is not surprising that most bone marrow transplantation experience is with MF, and the first lesson ringing loud and clear is that all marrow changes in this disorder are potentially reversible. A recent retrospective review found 48% 5-year survival after allogeneic transplant (62% in patients younger than 45 years) and 27% transplant-related mortality in the first year.26 Hopes are that advances in transplantation, including the use of nonmyeloablative preparative regimens, will improve on these results.27


Patients presenting with P vera should be phlebotomized to a normal hematocrit. Asymptomatic patients with P vera or ET can be stratified into high or low risk for thrombosis, based mainly on age and prior thrombosis. Platelet count greater than 1 500000/mm^sup 3^ may be another indication to initiate treatment. High-risk patients generally benefit from cytoreductive therapy, and hydroxyurea has emerged as the agent of choice, because alkylating agents (and P32) have high leukemogenic potentials. Anagrelide and interferons are second-line agents. The addition of low-dose aspirin is beneficial for most, helping to prevent arterial thrombotic complications. Therapy in any of these disorders should be tailored to the unique characteristics of the individual patient. With MF, therapeutic options run the gamut from observation, erythropoietic stimulators, cytotoxic agents, splenectomy, and bone marrow transplantation. Thalidomide and imatinib have shown some utility. The expectation is that our therapeutic armamentarium will soon include agents targeted specifically to pathogenic molecular derangements such as the recently identified JAK2 kinase mutation.


1. Barbui T, Barosi C, Grossi A, et al. Practice guidelines for the therapy of essential thrombocythemia: a statement from the Italian Society of Hematology, the Italian Society of Experimental Hematology and the Italian Group for Bone Marrow Transplantation. Haematologica. 2004:89:215-232.

2. Gruppo ltaliano Studio Policitemia. Polycythemia vera: the natural history of 1213 patients followed for 20 years. Ann Intern Med. 1995:123:656-664.

3. Gardner FH. Early approaches in the treatment of polycythemia vera. In: Wasserman LR, Berk PD, Berlin Nl, eds. Polycythemia Vera and the Myeloproliferative Disorders. Philadelphia, Pa: WB Saunders: 1995:147-153.

4. Thomas Dl, Marshall J, Ross Russell RW, et al. Effect of hematocrit on cerebral blood flow in man. iancef. 1977:2(80451:941-943.

5. Berk PD, Wasserman LR, Fruchtman SM, Goldberg JD. Treatment of polycythemia vera: a summary of clinical trials conducted by the polycythemia vera study group. In: Wasserman LR, Berk PD, Berlin Nl, eds. Polycythemia Vera and the Myeloproliferative Disorders. Philadelphia, Pa: WB Saunders: 1995:66-194.

6. Finazzi G, Barbui T. Risk-adapted therapy in essential thrombocythemia and polycythemia vera. Blood Rev. 2005;19:243-252.

7. Landolfi R, Marchioli R, Kutti J, et al. Efficacy and safety of low-dose aspirin in polycythemia vera. N Engl I Med. 2004;350:114-124.

8. Pardanani A, Tefferi A. lmatinib targets other than bcr/abl and their clinical relevance in myeloid disorders. Blood. 2004;104:1931-1939.

9. Buss DH, Cashell AW, O’Connor ML, Richards F II, case LD. Occurrence, etiology, and clinical significance of extreme thrombocytosis: a study of 280 cases. Am J Med. 1994:96:247-253.

10. Rice L, Popat U. Every case of essential thrombocythemia should be tested for the Philadelphia chromosome. Am J Henwtol. 2005:78:71-73.

11. Tefferi A, Fonseca R, Pereira DL, Hoagland HC. A long-term retrospective study of young women with essential thrombocythemia. Mayo CUn Proc. 2001; 76:22-28.

12. Elliott MA, Tefferi A. Thrombocythaemia and pregnancy. Besf Praci Res din Haematol. 2003:16:227-242.

13. Cortelazzo S, Finazzi C, Ruggeri M, et al. Hydroxyurea for patients with essential thrombocythemia and a high risk of thrombosis. N Engl/Med. 1995; 332:1132-1136.

14. Regev A, Stark P, Blickstein D, Lahav M. Thrombotic complications in essential thrombocythemia with relatively low platelet counts. AmJ Hematol. 1997; 56:168-172.

15. Tefferi A, Silverstein MN, Petitt RM, et al. Anagrelide as a new plateletlowering agent in essential thrombocythemia: mechanism of action, efficacy, toxicity, current indications. Semin Thromb Hemost. 1997;23:379-383.

16. Harrison CN, Campbell PJ, Buck G, et al. Hydroxyurea compared with anagrelide in high-risk essential thrombocythemia. N Engl J Med. 2005;353:3345.

17. Elliott MA, Tefferi A. Interferon-alpha therapy in polycythemia vera and essential thrombocythemia. Semin Thromb Hemost. 1997;23:463-472.

18. Baron BW, Mick R, Barron JM. Combined plateletpheresis and cytotoxic chemotherapy for symptomatic thrombocytosis in myeloproliferative disorders. Oncer. 1993:72:1209-1218.

19. Cervantes R, Barosi G, Demory JL, et al. Myelofibrosis with myeloid metaplasia in young individuals: disease characteristics, prognostic factors and identification of risk groups. Br I Haematol. 1998; 102:684-690.

20. Dupriez B, Morel P, Demory JL, et al. Prognostic factors in agnogenic myeloid metaplasia: a report on 195 cases with the new scoring system. Blood. 1996:88:1013-1018.

21. Cervantes F, Hernandez-Boluda JC, Alvarez A, et al. Danazol treatment of idiopathic myelofibrosis with severe anemia. Haemato/ogica. 2000:85:595-599.

22. Cervantes F, Alvarez-Larran A, Hernandez-Boluda JC, Sureda A, Torrebadell M, Montserrat E. Erythropoietin treatment of the anaemia of myelofibrosis with myeloid metaplasia: results in 20 patients and review of the literature. BrI Haematol. 2004:127:399-403.

23. Lofvenberg E, Wahlin A. Management of polycythaemia vera: essential thrombocythaemia and myelofibrosis with hydroxyurea. Eur J Haematol. 1988; 41:375-381.

24. Tefferi A, Mesa RA, Nagorney DM, et al. Splenectomy in myelofibrosis with myeloid metaplasia: a single-institution experience with 223 patients. Blood. 2000:95:2226-2233.

25. Mesa RA, Elliott MA, Schroeder G, Tefferi A. Durable responses to thalidomide-based therapy for myelofibrosis with myeloid metaplasia. Mayo CHn Proc. 2004:79:883-889.

26. Deeg HI, Appelbaum FR. Stem-cell transplantation for myelofibrosis. N Engl 1 Med. 2001;344:775-776.

27. Rondelli D, Barosi G, Bacigalupo A, et al. Allogeneic hematopoietic stemcell transplantation with reduced-intensity conditioning in intermediate- or highrisk patients with myelofibrosis with myeloid metaplasia. Blood. 2005; 105:41154119.

Lawrence Rice, MD; Kelty R. Baker, MO

Accepted for publication April 4, 2006.

From the Department of Medicine/Hematology-Oncology section, Baylor College of Medicine, Houston, Tex.

The authors have no relevant financial interest in the products or companies described in this article.

Reprints: Lawrence Rice, MD, Department of Medicine/Hematology-Oncology section, Baylor College of Medicine, 1709 Dryden Rd, M620, Suite 675, Houston, TX 77030 (e-mail:

Copyright College of American Pathologists Aug 2006

Provided by ProQuest Information and Learning Company. All rights Reserved