How to Manage Essential Thrombocythemia

HOW TO MANAGEy How to manage essential thrombocythemia

G Finazzi

I use the hematological, morphological and molecular criteria recently established by the World Health Organization to diagnose essential thrombocytemia. In these patients, major causes of morbidity and mortality are represented by thrombosis and bleeding, whereas progression to myelofibrosis and transformation to acute leukemia are more rare. Myelosuppressive therapy can reduce the rate of vascular complications, but there is some concern about treatment-related toxicity. Therefore, I follow a risk-oriented therapeutic approach to avoid inappropriate exposure to cytotoxic drugs on one side or suboptimal treatment on the other. Established predictors of cardiovascular events are represented by older age and previous thrombosis, whereas recent data suggest a prognostic role for novel risk factors, including leukocytosis and JAK2V617F mutational status. There is no indication for therapeutic intervention in asymptomatic, low-risk patients, while I treat high-risk patients with hydroxyurea (HU) first. Other therapeutic options, such as interferon alpha or anagrelide, may find place in selected patients including those who are resistant or intolerant to HU. I follow a risk-oriented approach also for management of pregnancy. Low-risk women are given low-dose aspirin throughout pregnancy and prophylactic low-molecular-weight heparin (LMWH) post partum, whereas LMWH throughout pregnancy and/or interferon-alpha can be required in high-risk cases.

Leukemia (2012) 26, 875--882; doi:10.1038/leu.2011.306; published online 4 November 2011 Keywords: chronic myeloproliferative neoplasms; essential thrombocythemia; hydroxyurea; low-dose aspirin; alpha-interferon; anagrelide

Among the chronic myeloproliferative neoplasms (MPN), essential PMF9 and refractory anemia with ring sideroblasts and marked thrombocythemia (ET) is characterized by longer median survival as thrombocytosis,10 can be JAK2V617F positive and mimic ET in well as lower transformation rates into acute leukemia or post-ET their presentation. Therefore, bone marrow examination is often myelofibrosis.1 The clinical course is marked by thrombotic and necessary to distinguish ET from both reactive thrombocytosis hemorrhagic episodes that occur more frequently in older patients (in JAK2V617F-negative patients) and other myeloid neoplasms and those with previous vascular events.2 Recent advances in our (in JAK2V617F-positive cases).4 understanding of molecular pathogenesis3 and histological features4 Some investigators remain skeptical about the use of morpho- of this disease contributed to better define diagnostic criteria and logical criteria in distinguishing ET from prefibrotic PMF, main- risk factors for overall survival as well as for vascular and bleeding taining that these criteria are subjective and difficult to apply complications. A small number of randomized clinical trials showed reproducibly.11 However, a recent large international study that cytoreductive agents, particularly hydroxyurea (HU), are able to confirmed the prognostic relevance of distinguishing ET from control the myeloproliferation and avoid vascular complications. pre-fibrotic PMF.1 Out of 1104 enrolled patients, diagnosis was However, short and long-term side effects of these drugs should be confirmed as ET in 891 (81%) and was revised to early/prefibrotic borne in mind. Hence, the best strategy is to limit the cytotoxic PMF in 180 (16%) patients; 33 cases were not evaluable. In early/ therapy to patients stratified on the basis of their risk for developing prefibrotic PMF compared with ET, survival rates (59% and 80%, vascular events.5 In this article, I review the management of patients respectively), leukemic transformation rates (11.7% and 2.1%, with ET by discussing diagnosis, risk stratification, current therapeu- respectively), and progression to overt myelofibrosis rates (16.9% tic options and updated treatment recommendations. and 9.3%, respectively) after 15 years were significantly worse. The respective death, leukemia and overt myelofibrosis incidence rates per 100 patient-years for early/prefibrotic PMF compared with ET HOW I DIAGNOSE ET were 2.7 and 1.3% (relative risk (RR), 2.1; Po0.001), 0.6 and 0.1% The differential diagnosis of ET includes reactive thrombocytosis, (RR, 5.2; P ¼ 0.001), and 1 and 0.5% (RR, 2.0; P ¼ 0.04). These polycythemia vera (PV), primary myelofibrosis (PMF), chronic findings indicate that hematological transformations are uncom- myeloid leukemia and refractory anemia with ring sideroblasts mon in patients with ET strictly diagnosed according to WHO 2008 and marked thrombocytosis.6 Accurate diagnosis is important for criteria, and support current recommendations to focus manage- both prognostication and treatment, and I adopt the 2008 World ment on risk stratification, prevention and treatment of thrombo- Health Organization (WHO) classification system that include hemorrhagic complications.12 hematological, morphological and molecular criteria (Table 1).7 The presence of a clonal marker (for example, JAK2 or MPL mutation)3 reliably excludes the possibility of reactive thrombo- INCIDENCE AND TYPE OF THROMBOSIS AND HEMORRHAGE cytosis, but its absence does not define the diagnosis because up In uncontrolled studies, reported cumulative rates for thrombosis to 50% of patients with ET might be JAK2V617F negative.8 In and hemorrhage during follow-up ranged from 7 to 17% and 8 to addition, other chronic myeloid neoplasms, such as prefibrotic 14%, respectively.13 In one study that also evaluated a control

Department of Hematology, Ospedali Riuniti di Bergamo, Bergamo, Italy. Correspondence: Dr G Finazzi, Department of Hematology, Ospedali Riuniti di Bergamo, Largo Barozzi 1, 24128 Bergamo, Italy. E-mail: gﬁ[email protected] Received 31 August 2011; accepted 15 September 2011; published online 4 November 2011 Essential thrombocythemia G Finazzi 876 Table 1. Diagnostic criteria for ET according to the 2008 WHO classiﬁcation7

1. Sustained platelet count X450 Â 109/l 2. Bone marrow biopsy specimen showing proliferation mainly of the megakaryocytic lineage with increased numbers of enlarged, mature megakaryocytes; no signiﬁcant increase or left shift of neutrophil granulopoiesis or erythropoiesis 3. Not meeting WHO criteria for polycythemia vera, primary myeloﬁbrosis, BCR-ABL1--positive chronic myelogenous leukemia, myelodysplastic syndrome or other myeloid neoplasm 4. Demonstration of JAK2V617F or other clonal marker, or in the absence of JAK2V617F; no evidence for reactive thrombocytosis Diagnosis requires meeting all four criteria.

Table 2. Cohort studies of risk factors for thrombosis and bleeding in ET including at least 100 patients

Study Patients no. Risk factors for thrombosis (RR or P)

Cardiovascular Age 460 Previous thrombosis Platelet count Leukocytosis risk factorsa

Cortelazzo et al.2 100 10.3 (2.05--51.5) 13 (4.1--41.5) NS ----- N S Besses et al.62 148 3.3 (1.5--7.4) 3.0 (1.5--6.0) NS ----- 4 . 7 (1.8--11.8) Colombi et al.63 103 NS Po0.001 NS ------Jantunen et al.64 1 3 2 N S ----- N S ----- P ¼ 0.01 Bazzan et al.65 187 NS (age 45 5 ) ----- N S ----- N S Wolanskyi et al.21 322 1.51 (1.05--2.18) 2.3 (1.25--4.24) ----- 1.74 (1.15--2.66) NS (arterial only) (WBC X15 Â 109/l) Carobbio et al.22 439 2.3 (1.3--3.9) (age 460 and previous NS 2.3 (1.4--3.9) ----- thrombosis evaluated together) (WBC X8.7 Â 109/l) Passamonti et al.66 605 Po0.001 P ¼ 0.03 NS NS ----- Palandri et al.23 532 2.45 (1.35--4.43) NS 1.76 (1.95--2.97) ----- (WBC X11 Â 109/l

Study Patients no. Risk factor for bleeding Platelet count

van Genderen et al.67 200 (review of Po0.001 published cases) (platelets 41.000 Â 109/l) Fenaux et al.68 147 ‘higher risk’ (platelets 42.000 Â 109/l) Wolanskyi et al.21 322 NS Alvarez-Larran et al.43 300 Po0.01 (platelets 41.000 Â 109/l) Abbreviations: NS, not signiﬁcant; RR, relative risk. aAt least one of the following: smoking, hypertension, hypercholesterolemia and diabetes.

population, the incidence of thrombotic episodes was 6.6% per The age-related differences in the frequency of these events is patient-year in ET vs 1.2% in control subjects, and the rate of mostly due to the coexistence of vascular disease in older patients. major hemorrhagic complications was 0.33% per patient-year in However, younger patients are not free of vascular thrombosis, ET vs 0% in controls.2 sometimes in unusual sites, such as the portal or sagittal veins.16 The most frequent types of major thrombosis include stroke, Overall, the incidence of thrombotic and hemorrhagic complica- transient ischemic attack, myocardial infarction, peripheral arterial tions in asymptomatic patients with ET younger than 60 years of thrombosis and deep venous thrombosis often occurring in age who had a platelet count of o1500 Â 109/l is comparable to a unusual sites, such as hepatic (Budd--Chiari syndrome), portal and normal control population.17 mesenteric veins.14 In addition to large vessel occlusions, ET patients may suffer from microcirculatory symptoms, including vascular headaches, dizziness, visual disturbances, distal paresthe- Platelet count and function sia and acrocyanosis. The most characteristic of these disturbances Several large cohort studies have failed to define a relationship is erythromelalgia, consisting of congestion, redness and burning between the frequency of thrombotic complications and platelet pain to ischemia and gangrene of the distal portions of toes and number (Table 2). Thus, an elevated platelet count, per se, should fingers.15 The most frequent bleeding events are hemorrhages not be considered as an indication for a myelosuppressive therapy from the gastrointestinal tract followed by hematuria and other aimed at preventing thrombotic complications. Supporting this mucocutaneous hemorrhages. Hemarthrosis and large muscle view, in a retrospective study of 99 consecutive young patients hematomas are uncommon.14 (aged o60 years) who presented with extreme thrombocytosis (platelet count X1000 Â 109/l) and without a previous history of thrombohemorrhagic complications, the incidences of major RISK STRATIFICATION ACCORDING TO THROMBO-HEMOR- thrombosis and hemorrhage during the follow-up were similar RHAGIC RISK between those who were treated with prophylactic cytoreductive Age and previous thrombosis therapy and those who did not receive such therapy.18 Age over 60 and a previous thrombotic event were identified as The relationship between frequency of bleeding episodes and major risk factors for thrombosis in most studies (Table 2). high platelet counts is more consistent. Most studies have shown

Leukemia (2012) 875 --882 & 2012 Macmillan Publishers Limited Essential thrombocythemia G Finazzi 877 that the degree and duration of bleeding in this patient Table 3. Risk stratification in ET based on thrombohemorrhagic risk population correlate with the platelet count (Table 2). Bleeding events appear to occur when the platelet counts are excessively Risk category Age 460 years or history of thrombosis high and stop when the platelet count falls to normal. The clinical spectrum of bleeding in ET patients closely resembles that Low NO observed in von Willebrand disease.19 An increase in the number High YES of circulating platelets appears to favor the adsorption of larger Extreme thrombocytosis (platelet count 41000 Â 109/l) is a risk factor for von Willebrand multimers onto platelet membranes, resulting in bleeding, not for thrombosis. Increasing leukocyte count and JAK2V617F their removal from the circulation and their subsequent degrada- mutation have been identified as novel risk factors for thrombosis, but tion. The laboratory features of acquired von Willebrand syndrome confirmation is required. in ET is characteristic of a type II deficiency with prolonged bleeding time, normal factor VIII C:VWF Ag ratio, decreased ristocetin cofactor activity, and decrease or absence of large von wild-type patients with essential thrombocythemia.34 This study Willebrand factor multimers.14,19 showed that JAK2V617F patients have a two-fold risk of Platelets in patients with ET have been known for a consider- developing thrombosis (odds ratio (OR) 1.92, 95% confidence able time to be qualitatively abnormal.6 Although both increased interval (CI) 1.45--2.53) but a significant heterogeneity between and decreased platelet reactivities have been described, these studies has been pointed out. In addition to the prognostic role of findings have not been definitively associated with thrombo- JAK2 mutation for the first thrombotic episode, recent data would hemorrhagic complications with two noteworthy exceptions; indicate that the mutation has also a role to predict recurrent erythromelalgia, where the prompt relief of symptoms by thrombotic episodes in patients with ET.35 cyclooxygenase inhibitors provides direct evidence that prosta- Bone marrow histology is important for an accurate morpho- glandins have a role in the development of vascular occlusion,15 logic diagnosis of ET according to WHO criteria and for predicting and acquired von Willebrand syndrome, which is a major cause of survival and hematological transformations to myelofibrosis or bleeding in patients with ET.19 acute leukemia.1 However, its role as a risk factor for thrombosis is controversial. Campbell et al.36 have identified increased bone Leukocyte number and function marrow reticulin fibrosis at diagnosis as an independent predictor A prognostic role for leukocytosis in ET has been advocated.20 of subsequent thrombotic and hemorrhagic complications. At Three large cohort studies reported that an increased baseline variance, a recent analysis comparing 891 patients with WHO- leukocyte count was an independent risk factor for both diagnosed ET vs 180 patients clinically presenting like ET but with thrombosis and inferior survival.21 --23 The role of WBC count in an histological picture of prefibrotic PMF did not show any ET was mostly observed on the occurrence of myocardial difference in the rates of arterial (1.2--1.4% patient-year, respec- 24 25 tively) and venous (0.6% patient-year in both groups) thrombotic infarction, as also shown in patients with PV. In ‘low-risk’ ET 1 patients, (that is, below 60 years and without previous thrombosis) complications. Instead, prefibrotic PMF was a significant predictor of major bleeding, together with leukocytosis, previous hemor- leukocytosis conferred a thrombotic risk comparable to that of 37 treated ‘high-risk’ patients without leukocytosis.22 These findings rhage and aspirin use. The role of inflammation among the possible prognostic factors were not confirmed in a retrospective study of 407 low-risk 38,39 patients with ET.26 Leukocytosis at the time of diagnosis, defined in MPN has been recenly highlighted. Barbui et al. correlated 9 vascular complications with plasma levels of high-sensitivity by a cut-off level of either 15 or 9.4 Â 10 /l, did not appear to be predictive of either arterial or venous thrombosis during follow-up. C-reactive protein and pentraxin-3 in 244 ET and PV patients. However, in an analysis of 194 low-risk ET patients, the increase in Major thrombosis rate was higher in the highest C-reactive protein leukocyte count within 2 years of diagnosis (observed in 9% of tertile (P ¼ 0.01) and lower at the highest pentraxin-3 levels patients), rather than leukocytosis at diagnosis, was associated (P ¼ 0.045). These associations remained significant in multi- with an higher risk of vascular complications during follow-up.27 variable analysis and indicate that these inflammatory biomarkers independently and in opposite ways modulate the risk of In ET, an in vivo leukocyte activation has been consistently 38 documented in association with signs of activation of both cardiovascular events in patients with MPN. platelets and endothelial cells. Interestingly, the presence of the In conclusion, a clinically oriented scheme is proposed to stratify patients with ET in a ‘high-risk’ or ‘low-risk’ category according to JAK2 mutation is associated with higher platelet and leukocyte 12 activation in these patients.28 Thus, platelet and leukocyte their age and previous history of thrombosis (Table 3). Putative activation may have a role in the generation of the pre-thrombotic novel variables, such as leukocytosis and JAK2V617F mutational state that characterizes ET.29 However, whether leukocytosis status, might be incorporated in the risk classification, possibly should be simply considered a marker for vascular disease or allowing better definition of the low-risk group, once more whether elevated WBC levels actually contribute directly to information is available and when they have been eventually causing such disorders should be matter of prospective studies. validated in prospective studies.

Other risk factors The presence of the JAK2V617F mutation in about 60% of ET HOW I MANAGE ‘LOW-RISK’ PATIENTS patients raised the question whether mutated and non-mutated I avoid cytoreduction in low-risk ET patients (Figure 1). The natural patients differ in terms of thrombotic risk. The largest prospective history of such patients left untreated was prospectively evaluated study on 806 patients suggested that JAK2 mutation in ET was in a controlled study that compared 65 low-risk patients with 65 associated with anamnestic venous but not arterial events.30 An age- and sex-matched normal controls.17 Patients were followed increased risk of thrombosis in JAK2-mutated patients was up and cytoreductive therapy was introduced as soon as a major retrospectively observed by other investigators.31,32 However, clinical event was recorded. After a median follow-up of 4.1 years, the rate of vascular complications was not affected by the the incidence of thrombosis was not significantly higher in presence of the mutation in two relatively large retrospective patients than in controls (1.91% vs 1.5% per patient-year; age- and studies, including 150 and 130 ET patients respectively.8,33 sex-adjusted risk ratio 1.43, 95% CI 0.37--5.4). No major bleeding A systematic literature review was carried out to compare the was observed. These findings were confirmed in a cohort of 74 frequency of thrombosis between JAK2V617F-positive and young women followed untreated for more than 9 years who

& 2012 Macmillan Publishers Limited Leukemia (2012) 875 --882 Essential thrombocythemia G Finazzi 878 Diagnosis of ET*

No thrombosis / major bleeding History of / presentation with and age < 60 years thrombosis or major hemorrhagic complication or age > 60 years

No cytoreduction Cytoreductive treatment: Re-consider if complications Hydroxyurea(HU) as first choice; Low-dose aspirin Interferon in special situations (pregnancy); if microvascular symptoms Anagrelide as second-line therapy in patients intolerant or refractory to HU Low-dose aspirin if major thrombosis or microvascular symptoms

* according to WHO 20087 Figure 1. How I manage patients with ET

showed 1.2% per patient-year incidence of vascular events.40 Taking into account all these data, I currently do not use aspirin Thrombotic deaths seem very rare in low-risk ET subjects, and as primary anti-thrombotic prophylaxis in all ET patients, but I there are no data indicating that fatalities can be prevented by reserve the drug for patients with microvascular symptoms or starting cytoreductive drugs early. Admittedly, this policy is based severe, uncontrolled cardiovascular risk factors (Figure 1). on studies with relatively small samples, and further data from large clinical trials are needed. HOW I TREAT HIGH-RISK PATIENTS Low-dose aspirin The cytoreductive drugs most commonly used for the treatment of high-risk ET include HU, alpha-interferon and anagrelide. In ET, aspirin, 100 mg daily, has been found to control Practical recommendations for their management and dosing microvascular symptoms, such as erythromelalgia, and transient are reported in Table 4 and main results from randomized clinical neurological and ocular disturbances including dysarthria, hemi- trials comparing these drugs are summarized in Table 5. paresis, scintillating scotomas, amaurosis fugax, migraine and seizures.15. Higher doses, up to 500 mg daily, may be necessary in the acute phase of erythromelalgia. Translating evidence from the Hydroxyurea ECLAP randomized study in PV,41 the use of low-dose aspirin as HU has emerged as the treatment of choice in high-risk patients primary prophylaxis of vascular events can be considered. with ET because of its efﬁcacy in preventing thrombosis, as However, formal clinical trials addressing this issue in ET have demonstrated in a seminal randomized clinical trial.44 One not been produced so far. In a randomized study comparing HU vs hundred and fourteen patients were randomized to long-term anagrelide in high-risk patients with ET discussed in detail below, treatment with HU (n ¼ 56) or to no cytoreductive treatment low-dose aspirin was given to both groups.42 An increased rate of (n ¼ 58). During a median follow-up of 27 months, 2 thromboses major bleeding was registered in the anagrelide plus aspirin arm were recorded in the HU-treated group (1.6%/patient-year) and this may be due to a synergistic effect of the two drugs on compared with 14 in the control group (10.7% patient-year; platelet function inhibition. P ¼ 0.003). Notably, the anti-thrombotic effect of HU may A recent publication from Alvarez-Larran et al.43 questions the recognize additional mechanisms of action besides panmyelo- beneﬁt of low-dose aspirin in low-risk ET. These authors retro- suppression, including qualitative changes in leukocytes, spectively assessed the incidence rates of arterial and venous decreased expression of endothelial adhesion molecules and thrombosis in 300 low-risk ET patients either treated with enhanced nitric oxide generation.45 Some long-term follow-up antiplatelet drugs as monotherapy (n ¼ 198, follow-up 802 studies revealed that a proportion of ET patients treated with HU person-years) or subjected to observation only (n ¼ 102, 848 developed acute leukemia, particularly when given before or after person-years). They reported that overall rates of thrombotic alkylating agents or radiophosphorus.46 In other studies, however, events did not differ between these patient groups, whereas the use of this drug as the only cytotoxic treatment was rarely an increased risk of major bleeding was observed in patients with associated with secondary malignancies; in an analysis of 25 ET platelet count 41000 Â 109/l treated with antiplatelet therapy patients younger than 50 years and treated with HU alone for a (incidence rate ratio (IRR): 5.4; 95% CI 1.7--17.2; P ¼ 0.004). high risk of thrombosis, no case of leukemic or neoplastic Subgroup analysis also showed that two types of patients did transformation occurred after a median follow-up of 8 years worse with observation only: JAK2 V617F-positive patients had an (range 5--14 years).47 Interestingly, in a recent, large population- increased risk of venous thrombosis (IRR: 4.0; 95% CI 1.2--12.9; based nested case-control study in Sweden, the risk of AML/MDS P ¼ 0.02), and patients with cardiovascular risk factors had in MPNs was strongly associated with P32 (RR 3.39, 95% CI increased rates of arterial thrombosis (IRR: 2.5; 95% CI 1.02--6.1; 1.28--8.99, P ¼ 0.01) and alkylator treatments (RR 4.46, 95% CI P ¼ 0.047). 1.22--16.31, P ¼ 0.03). In contrast, in this survey HU (41000 mg)

Leukemia (2012) 875 --882 & 2012 Macmillan Publishers Limited Essential thrombocythemia G Finazzi 879 Table 4. Management and dosing of cytoreductive therapy in ET

Hydroxyurea The starting dose of HU is 15--20 mg/Kg/day until response is obtained (for response criteria, see Barosi et al.60). Thereafter, a maintenance dose should be administered to keep the response without reducing WBC count values below 2500 Â 109/l. Complete hemogram should be recorded every 2 weeks during the ﬁrst 2 months, then every month, and in steady state in responding patients every 3 months.

IFN-alpha IFN-alpha is contraindicated in patients with thyroid and/or mental disorders: for this reason an accurate evaluation of thyroid function and inquiry of previous or present mental disorders in candidate patients are recommended. IFN should be administered at the dose of 3 MU daily until a response is reached;60 then, therapy has to be adjusted at the lowest weekly doses, which maintain the response. Complete hemogram must be recorded every week during the ﬁrst month of therapy, every 2 weeks during the second month, then every month and in steady state in responding patients every 3--4 months. Pegylated-IFN-alpha is given at a starting dose of 0.5 mg/kg once weekly. In patients who failed to achieve a response after 12 weeks, the dose is increased up to 1.0 mg/kg/week. Then, therapy has to be adjusted at the lowest dose that maintain the response and monitored as above.

Anagrelide The recommended starting dosage of oral anagrelide for adults is 0.5 mg twice daily; this dosage should be maintained for at least 1 week and thereafter titrated individually to achieve a platelet count o600 Â 109/l and ideally 150--400 Â 109/l. Dosages should be increased by not more than 0.5 mg/day in a single week, and should not exceed 10 mg/day. The recommended maximum single dose is 2.5 mg. Patients with cardiovascular disease or hepatic dysfunction should be monitored closely. The drug is contraindicated in patients with severe cardiac insufﬁciency as well as hepatic (Child--Pugh classiﬁcation C) or renal (creatinine clearance o30 ml/min) impairment.

Table 5. Randomized clinical trials in high-risk patients with ET

Study Patients and follow-up Treatments and main results P

Cortelazzo et al.44 114 pts; 27 months (median) HUa No cytoreductiona Thrombosis 2 14 0.003

Harrison et al.42 809 pts; 39 months (median) HU+ASA Anagrelide+ASA Arterial thrombosis 17 37 0.004 Venous thrombosis 14 3 0.006 Major bleeding 8 22 0.008 Transformation to myeloﬁbrosis 5 16 0.01 Death for any cause 27 31 n.s.

Gisslinger et al.58 258 pts; 12 months HU Anagrelide Major thrombosis and bleeding 12 11 n.s. Abbreviations: ASA, Aspirin; HU, hydroxyurea; n.s., not signiﬁcant. aPlus ASA in 70% of patients in both groups.

did not significantly increase the risk for transformation to AML/ 3 months of at least 2 g/day of HU (2.5 g/day in patients with a MDS (RR 1.01, 0.28--3.6).48 Further support to the low, if any, body weight over 80 kg); platelet count 4400 000/ml and WBC leukemogenic potential of HU comes from a systematic review of count o2500/ml or Hb counto10 g/dl at any dose of HU; presence the efficacy and safety of this drug in sickle cell disease. This study of leg ulcers or other unacceptable mucocutaneous manifesta- analyzed HU toxicities not only in patients with sickle cell disease tions at any dose of HU; HU-related fever. but also in patients with other diseases, including MPNs, and concluded that, albeit limited, current evidence suggests that HU treatment in adults does not increase the risk for leukemia.49 Interferon alpha (IFN-alpha) Major side effects of HU include hematopoietic impairment, IFN-alpha was considered for the treatment of patients with MPDs leading to neutropenia and macrocytic anemia, and mucocuta- because this agent suppresses the proliferation of hematopoietic neuos toxicity, most frequently presenting as oral and leg ulcers, progenitors, has a direct inhibiting effect on bone marrow and skin lesions. In addition, it is estimated that about 10% of fibroblast progenitor cells and antagonizes the action of plate- patients receiving HU do not achieve the desired reduction in let-derived growth factor, transforming growth factor-beta and blood cell counts using recommended doses. Recently, an other cytokines, which may be involved in the development of international working group in the frame of the European myelofibrosis.51 Published reports concern small consecutive Leukemia Net was convened to develop a consensus formulation series of patients in whom hematological response and side of clinically significant criteria for defining resistance/intolerance effects were evaluated. The results of several cohort studies of ET to HU in ET.50 The Working Group proposed that the definition patients, reviewed in Lengfelder et al.,52 indicate that reduction of of resistance/intolerance should require the fulfillment of at least platelet count below 600 Â 109/l can be obtained in about 90% of one of the following criteria: platelet count 4600 000/ml after cases after about 3 months with an average dose of 3 million IU

& 2012 Macmillan Publishers Limited Leukemia (2012) 875 --882 Essential thrombocythemia G Finazzi 880 daily. IFN-alpha is not known to be teratogenic and does not cross international multicenter phase III study designed to evaluate the the placenta. Thus, it has been used successfully throughout non-inferiority of anagrelide vs HU in 258 high-risk ET patients pregnancy in some ET patients with no adverse fetal or maternal diagnosed according to the 2008 WHO diagnostic criteria. This outcome.53 The main problem with IFN-alpha therapy, apart from classification, at variance of PVSG criteria required in the PT-1 trial, its costs and parental route of administration, is the incidence of included a more homogenous category of patients excluding side effects. Fever and flu-like symptoms are experienced by most those with early myelofibrosis. During the whole study period, 11 patients and usually require treatment with paracetamol. Signs of major ET-related complications occurred in the anagrelide group chronic IFN-alpha toxicity, such as weakness, myalgia, weight and (5 arterial events, 2 venous thrombotic complications and 4 hair loss, severe depression, and gastrointestinal and cardiovas- bleedings) and 12 major events were seen in the HU arm cular symptoms, make it necessary to discontinue the drug in (5 arterial events, 5 venous thrombotic events and 2 bleedings). about one third of patients. Transformations to myelofibrosis were not reported. This study Pegylated forms of IFN-alpha allow weekly administration, provides preliminary evidence for non-inferiority of anagrelide potentially improving compliance and possibly providing more compared with HU in the treatment of ET diagnosed according to effective therapy. In patients with PV, pegylated IFN-alpha-2a the WHO classification. However, compared with PT-1, the number therapy was able to reduce the malignant clone as quantitated by of patients enrolled was small, duration of follow-up relatively the percentage of the mutated allele JAK2V617F.54 More limited short and considerably fewer end-point events were recorded. It is effects on JAK2 mutational status have been reported after therefore questionable whether this study has the statistical therapy with pegylated IFN-alfa-2b in a small group of patients power to detect the differences observed in the PT-1 study. with PV and ET.55 In a phase II study of pegylated IFN-alfa-2a in 79 patients with PV and ET, an overall hematologic response rate was JAK2 inhibitors observed in 80% of PV and 81% of ET (complete in 70% and 76% To date, clinical studies of new drugs with JAK2 inhibitory activity of patients, respectively).56 The molecular response rate was 38% have mainly included patients with PMF or PV. A small group of 39 in ET and 54% in PV, being complete (undetectable JAK2 V617F) in patients with ET, refractory or intolerant to HU, is currently treated 6% and 14%, respectively. The JAK2V617F mutant allele burden with Ruxolitinib (INCB018424). At the latest study update,59 after a continued to decrease with no clear evidence for a plateau in PV median follow-up of 15 months (range 4--21), 49% of enrolled patients, whereas this pattern was less clear in ET patients. The subjects had platelet counts normalized to the upper limit of tolerability of PEG-IFN-alpha-2a at 90 mg weekly was excellent. normal for a median duration of 3.5 months and 82% maintained Thus, this agent may have an important role in the treatment of platelet counts 600 Â 109/l for a median duration of 9.8 months. clonal MPN, particularly PV. o Palpable spleens were resolved in three out of four subjects and reductions in patient-reported symptom scores for pruritus, night Anagrelide sweats and bone pain were observed. Grade 3 adverse events potentially related to study medication included leukopenia (two Anagrelide, a member of the imidazoquinazolin compounds, has a patients), gastrointestinal disorder and peripheral neuropathy potent platelet reducing activity devoid of leukemogenic potential (one patient each). No grade 4 drug-related adverse event have and appeared to be the option to HU for reducing platelet counts in occurred. Clinical responses were unrelated to the presence/ younger ET patients at high risk for thrombosis. The largest analysis absence of JAK2V617F mutation at entry or to the allele burden reported so far comprised 3660 patients (2251 with ET).57 With changes following treatment. Although interesting, these results maximum follow-up of 7 years, anagrelide achieved platelet control are too preliminary to draw any conclusion on the current role of in over 75% of patients and did not increase the conversion to AL. JAK2 inhibitors in the management of ET. However, other complications of the drug include palpitations, congestive heart failure, headache and depression. HU and anagrelide (plus aspirin in both groups) have been Practical recommendations compared head to head in a randomized clinical trial (PT-1) My current practice for the treatment of patients with ET is including 809 ET patients.42 Patients in the anagrelide arm showed summarized in Figure 1 and largely overlaps with the recommen- an increased rate of arterial thrombosis (OR 2.16, 95% CI 1.04-- dations of a panel of experts recently appointed by the ELN.12 2.37, P ¼ 0.03), major bleeding (OR 2.61, 95% CI 1.27--5.33, Several areas of uncertainties still remain and call for appropriate P ¼ 0.008) and myelofibrotic transformation (OR 2.92, 95% CI clinical trials. In order to uniform the criteria to be used in clinical 1.24--6.86, P ¼ 0.01) but a decreased incidence of venous studies, a definition of response to treatment in PV and ET has thrombosis (OR 0.27, 95% CI 0.11--0.71, P ¼ 0.006) compared with been proposed.60 Clinico-hematologic, molecular and histologic HU. In addition, anagrelide was more poorly tolerated than HU response were selected and are expected to provide a means to and presented significantly greater rates of cardiovascular compare the results from different patient cohorts and to facilitate (Po0.0001), gastrointestinal (Po0.02), neurological (Po0001) communication within the scientific community. and constitutional (Po0.001) side effects. Transformation to AL was comparable between the two arms (4 anagrelide vs 6 HU), although the small number of transformations and short follow-up HOW I MANAGE PREGNANT PATIENTS WITH ET prevented firm conclusions about leukemogenicity. I take a detailed personal and family history, and I put the patient Therapy with anagrelide, but not with HU, was also associated also under the care of an obstetrician experienced in the with progressive anemia and an increase in bone marrow management of patients with high-risk pregnancies. The patient fibrosis.36 The increased fibrosis was reversible in a small number should stop any possibly teratogenic drugs at least 3 months of patients upon withdrawal of anagrelide, and follow-up trephine before conception. Depending on the risk of maternal vascular biopsies are therefore recommended for patients receiving this events and pregnancy morbidity, treatment options range from agent, perhaps every 2--3 years. It is important to note that the no therapy, aspirin alone, low-molecular-weight heparin (LMWH) diagnosis of ET in the PT-1 trial was made according to the PVSG to cytoreductive therapy (Table 6). classification and it remains questionable if these recommenda- In the absence of clear contraindications, I give aspirin (100 mg tions can be applied to ET patients diagnosed according to the daily) throughout pregnancy and LMWH 4000 U daily for 6 weeks WHO classification. post partum to all women with ET. However, not all studies agree In this connection, useful information is expected from the on the value of aspirin therapy in reducing miscarriage rates. Low- Anahydret trial.58 The Anahydret was a randomized single blind dose aspirin is considered safe in pregnancy and should preferably

Leukemia (2012) 875 --882 & 2012 Macmillan Publishers Limited Essential thrombocythemia G Finazzi 881 Table 6. How I manage pregnancy in ET 4 Thiele J, Kvasnicka HM, Orazi A. Bone marrow histopathology in myeloproliferative disorders --current diagnostic approach. Semin Hematol 2005; 1. Risk stratification 42: 184 --195. At least one of the following defines high-risk pregnancy: 5 Finazzi G, Barbui T. Evidence and expertise in the management of polycythemia Previous major thrombotic or bleeding complication vera and essential thrombocythemia. Leukemia 2008; 22: 1494 --1502. a Previous severe pregnancy complications 6 Finazzi G, Hu M, Barbui T, Hoffman R. Essential thrombocythemia. In: Hoffman R, 9 Platelet count 41500 Â 10 /l Benz Jr EJ, Shattil SJ, Furie B, Silberstein LE, McGlave P, Heslop H (eds). 2. Therapy Hematology Basic Principles and Practice, 5th edn Churchill Livingstone: (a) Low-risk pregnancy Philadelphia, 2009, pp 1149 --1166. Aspirin 100 mg/day 7 Tefferi A, Thiele J, Orazi A, Kvasnicka HM, Barbui T, Hanson CA et al. Proposals and LMWH 4000 U/day after delivery until 6 weeks postpartum rationale for revision of the World Health Organization diagnostic criteria for (b) High-risk pregnancy polycythemia vera, essential thrombocythemia, and primary myelofibrosis: As above, plus recommendations from an ad hoc international expert panel. Blood 2007; 110: If previous major thrombosis or severe pregnancy 1092 --1097. complications: LMWH throughout pregnancy (stop aspirin if 8 Wolanskyj AP, Lasho TL, Schwager SM, McClure RF, Wadleigh M, Lee SJ et al. JAK2 bleeding complications) V617F mutation in essential thrombocythaemia: clinical associations and long- If platelet count 41500 Â 109/l: consider IFN-a term prognostic relevance. Br J Haematol 2005; 131: 208 --213. If previous major bleeding: avoid aspirin and consider IFN-a to 9 Thiele J, Kvasnicka HM, Mu¨llauer L, Buxhofer-Ausch V, Gisslinger B, Gisslinger H. reduce thrombocytosis Essential thrombocythemia versus early primary myelofibrosis: a multicenter study to validate the WHO classification. Blood 2011; 117: 5710 --5718. aSevere pregnancy complications: X3 first-trimester or X1 second or 10 Schmitt-Graeff AH, Teo SS, Olschewski M, Schaub F, Haxelmans S, Kirn A et al. third-trimester losses, birth weight o5th centile of gestation, JAK2V617F mutation status identifies subtypes of refractory anemia with ringed pre-eclampsia, intrauterine death or stillbirth. sideroblasts associated with marked thrombocytosis. Haematologica 2008; 93: 34 --40. 11 Wilkins BS, Erber WN, Bareford D, Buck G, Wheatley K, East CL et al. Bone marrow be started before conception to facilitate placental and fetal pathology in essential thrombocythemia: interobserver reliability and utility for development. Bleeding complications are rare, but particular identifying disease subtypes. Blood 2008; 111:60--70. attention should be paid to patients with platelet count above 12 Barbui T, Barosi G, Birgegard G, Cervantes F, Finazzi G, Griesshammer M et al. 1000--1500 Â 109/l because the risk of bleeding may increase Philadelphia-negative classical myeloproliferative neoplasms: critical concepts significantly. and management recommendations from European LeukemiaNet. J Clin Oncol LMWH throughout all pregnancy is indicated for prophylaxis of 2011; 29: 761 --770. deep venous thrombosis and to reduce fetal morbidity in selected 13 Barbui T, Barosi G, Grossi A, Gugliotta L, Liberato LN, Marchetti M et al. Practice guidelines for the therapy of essential thrombocythemia. A statement from the high-risk ET women. The suggested dose of enoxaparin is 4000 U Italian Society of Hematology, the Italian Society of Experimental Hematology (40 mg) once daily, increasing to 4000 U twice daily for 16 weeks and the Italian Group for Bone Marrow Transplantation. Haematologica 2004; 89: and dropping to 4000 U daily for 6 weeks postpartum. To increase 215 --222. the antithrombotic efficacy in very high-risk situations LMWH can 14 Elliott MA, Tefferi A. Thrombosis and haemorrhage in polycythaemia vera and be used in combination with low-dose aspirin. essential thrombocythaemia. Br J Haematol 2004; 128: 275 --290. Cytoreductive therapy in pregnant women with ET is a 15 Michiels JJ, Abels J, Steketee J, van Vliet HH, Vuzevski VD. Erythromelalgia caused controversial area. During pregnancy, platelet count may undergo by platelet mediated arteriolar inflammation and thrombosis in a natural fall and this could reduce the need of cytoreductive thrombocythemia. Ann Intern Med 1985; 102: 466 --471. drugs. The target of platelet-lowering therapy is uncertain, 16 Mitus AJ, Barbui T, Shulman LN, Rosenthal DS, Viero P, Cortelazzo S et al. Hemostatic complications in young patients with essential thrombocythemia. because the available data did not indicate a relation between Am J Med 1990; 88: 371 --375. platelet count and adverse pregnancy outcome. According to 17 Ruggeri M, Finazzi G, Tosetto A, Riva S, Rodeghiero F, Barbui T. No treatment for 12 53,61 recent guidelines and expert judgement, candidates for low-risk thrombocythaemia: results from a prospective study. Br J Haematol 1998; platelet-lowering drugs are high-risk women, such as those with a 103: 772 --777. previous history of major thrombosis or major bleeding, 18 Tefferi A, Gangat N, Wolanskyj AP. Management of extreme thrombocytosis in particularly when platelet count is 41000--1500Â109/l, or when otherwise low-risk essential thrombocythemia: does number matter? Blood 2006; familial thrombophilia or cardiovascular risk factors are documen- 108: 2493 --2494. ted. If cytoreduction has to be given, IFN-alpha is probably the 19 van Genderen PJJ, Budde A, Michiels JJ, van Strik R, van Vliet HH. The reduction of safest option. Generally, patients should not be receiving HU when large von Willebrand factor multimers in plasma in essential thrombocythemia is related to the platelet count. Br J Haematol 1996; 93: 962 --965. they conceive or in the first trimester. Anagrelide may cause fetal 20 Barbui T, Carobbio A, Rambaldi A, Finazzi G. Perspectives on thrombosis in harm by crossing the placenta and result in severe thrombocy- essential thrombocythemia and polycythemia vera: is leukocytosis a causative topenia. It is therefore not recommended in pregnancy. factor? Blood 2009; 114: 759 --763. 21 Wolanskyj AP, Schwager SM, McClure RF, Larson DR, Tefferi A. Essential thrombocythemia beyond the first decade: life expectancy, long-term complication rates, and prognostic factors. Mayo Clin Proc 2006; 81: 159 --166. CONFLICT OF INTEREST 22 Carobbio A, Finazzi G, Guerini V, Spinelli O, Delaini F, Marchioli R et al. The author declares no conflict of interest. Leukocytosis is a risk factor for thrombosis in essential thrombocythemia: interaction with treatment, standard risk factors and JAK2 mutation status. Blood 2007; 109: 2310 --2313. REFERENCES 23 Palandri F, Polverelli N, Catani L, Ottaviani E, Baccarani M, Vianelli N. Impact of leukocytosis on thrombotic risk and survival in 532 patients with essential 1 Barbui T, Thiele J, Passamonti F, Rumi E, Boveri E, Ruggeri M et al. Survival and thrombocythemia: a retrospective study. Ann Hematol 2011; 90: 933 --938. disease progression in essential thrombocythemia are significantly influenced by 24 Carobbio A, Antonioli E, Guglielmelli P, Vannucchi AM, Delaini F, Guerini V et al. accurate morphologic diagnosis: an international study. J Clin Oncol 2011; 29: Leukocytosis and risk stratification assessment in essential thrombocythemia. 3179 --3184. J Clin Oncol 2008; 26: 2732 --2736. 2 Cortelazzo S, Viero P, Finazzi G, D’Emilio A, Rodeghiero F, Barbui T. Incidence and 25 Landolfi R, Di Gennaro L, Barbui T, De Stefano V, Finazzi G, Marfisi R et al. risk factors for thrombotic complications in a historical cohort of 100 patients with Leukocytosis as a major thrombotic risk factor in patients with polycythemia vera. essential thrombocythemia. J Clin Oncol 1990; 8: 556 --562. Blood 2007; 109: 2446 --2452. 3 Tefferi A, Vainchenker W. Myeloproliferative neoplasms: molecular 26 Gangat N, Wolanskyj AP, Schwager SM, Hanson CA, Tefferi A. Leukocytosis at pathophysiology, essential clinical understanding, and treatment strategies. J Clin diagnosis and the risk of subsequent thrombosis in patients with low-risk essential Oncol 2011; 29: 573 --582. thrombocythemia and polycythemia vera. Cancer 2009; 115: 5740 --5745.

& 2012 Macmillan Publishers Limited Leukemia (2012) 875 --882 Essential thrombocythemia G Finazzi 882 27 Passamonti F, Rumi E, Pascutto C, Cazzola M, Lazzarino M. Increase in leukocyte 47 Finazzi G, Ruggeri M, Rodeghiero F, Barbui T. Efficacy and safety of long-term use count over time predicts thrombosis in patients with low-risk essential of hydroxyurea in young patients with essential thrombocythemia and a high risk thrombocythemia. J Thromb Haemost 2009; 7: 1587 --1589. of thrombosis. Blood 2003; 101: 3749. 28 Arellano-Rodrigo E, Alvarez-Larran A, Reverter JC, Villamor N, Colomer D, 48 Bjorkholm J, Derolf AR, Ekstrand C, Kristinsson SY, Ekstrand C, Goldin LR et al. Cervantes F. Increased platelet and leukocyte activation as contributing Clinical risk for AML/MDS transformation in Philadelphia negative chronic mechanisms for thrombosis in essential thrombocythemia and correlation with myeloproliferative neoplasms. A population-based nested case-control study in the JAK2 mutational status. Haematologica 2006; 91: 169 --175. Sweden. J Clin Oncol 2011; 29: 2410 --2415. 29 Marchetti M, Castoldi E, Spronk HM, van Oerle R, Balducci D, Barbui T et al. 49 Lanzkron S, Strouse JJ, Wilson R, Beach MC, Haywood C, Park H et al. Systematic Thrombin generation and activated protein C resistance in patients with essential review: hydroxyurea for the treatment of adults with sickle cell disease. Ann Intern thrombocythemia and polycythemia vera. Blood 2008; 112: 4061 --4068. Med 2008; 148: 939 --955. 30 Campbell PJ, Scott LM, Buck G, Wheatley K, East CL, Marsden JT et al. Definition of 50 Barosi G, Besses C, Birgegard G, Briere J, Cervantes F, Finazzi G et al. A unified subtypes of essential thrombocythaemia and relation to polycythaemia vera definition of clinical resistance/intolerance to hydroxyurea in essential based on JAK2 V617F mutation status: a prospective study. Lancet 2005; 366: thrombocythemia: results of a consensus process by an international working 1945 --1953. group. Leukemia 2007; 21: 277 --280. 31 Cheung B, Radia D, Pantelidis P, Yadegarfar G, Harrison C. The presence of the 51 Silver RT. Long-term effects of the treatment of polycythemia vera with JAK2 V617F mutation is associated with higher haemoglobin and increased risk of recombinant interferon-alpha. Cancer 2006; 107: 451 --458. thrombosis in essential thrombocythaemia. Br J Haematol 2005; 132: 244 --250. 52 Lengfelder E, Griesshammer M, Hehlmann R. Interferon-alpha in the treatment of 32 Finazzi G, Rambaldi A, Guerini V, Carobbio A, Barbui T. Risk of thrombosis in essential thrombocythemia. Leuk Lymphoma 1996; 22 (Suppl.1): 135 --142. patients with essential thrombocythemia and polycythemia vera according to 53 Harrison C. Pregnancy and its management in the Philadelphia negative JAK2 V617F status. Haematologica 2007; 92: 135 --136. myeloproliferative diseases. Br J Haematol 2005; 129: 293 --306. 33 Antonioli E, Guglielmelli P, Pancrazzi A, Bogani C, Verrucci M, Ponziani V et al. 54 Kiladjian JJ, Cassinat B, Chevret S, Turlure P, Cambier N, Roussel M et al. Pegylated Clinical implications of the JAK2 V617F mutation in essential thrombocythemia. interferon alpha-2a induces complete hematologic and molecular responses with Leukemia 2005; 19: 1847 --1849. low toxicity in polycythemia vera. Blood 2008; 112: 3065 --3072. 34 Lussana F, Caberlon S, Pagani C, Kamphuisen PW, Bu¨ ller HR, Cattaneo M. 55 Samuelsson J, Mutschler M, Birgegard G, Gram-Hansen P, Bjo¨rkholm M, Pahl HL. Association of V617F Jak2 mutation with the risk of thrombosis among patients Limited effects on JAK2 mutational status after pegylated interferon alpha-2b with essential thrombocythemia or idiopathic myelofibrosis: a systematic review. therapy in polycythemia vera and essential thrombocythemia. Haematologica Thromb Res 2009; 124: 409 --417. 2006; 91: 1281 --1282. 35 De Stefano V, Za T, Rossi E, Vannucchi AM, Ruggeri M, Elli E et al. Increased risk of 56 Quintas-Cardama A, Kantarjian H, Manshouri T, Luthra R, Estrov Z, Pierce S et al. recurrent thrombosis in patients with essential thrombocythemia carrying the Pegylated interferon alpha-2a yields high rates of hematologic and molecular homozygous JAK2 V617F mutation. Ann Hematol 2010; 85: 97 --100. response in patients with advanced essential thrombocythemia and 36 Campbell PJ, Bareford D, Erber WN, Wilkins BS, Wright P, Buck G et al. Reticulin polycythemia vera. J Clin Oncol 2009; 27: 5418 --5424. accumulation in essential thrombocythemia: prognostic significance and relation 57 Fruchtman SM, Petitt RM, Gilbert HS, Fiddler G, Lyne A. Anagrelide: analysis of to therapy. J Clin Oncol 2009; 27: 2991 --2999. long term efficacy, safety and leukemogenic potential in myeloproliferative 37 Finazzi G, Carobbio A, Thiele J, Passamonti F, Rumi E, Ruggeri M et al. Incidence diseases. Leuk Res 2005; 5: 481 --491. and risk factors for bleeding in 1104 patients with essential thrombocythemia or 58 Gisslinger H, Gotic M, Holowiecki J, Penka M, Thiele J, Kvasnicka HM et al. Final prefibrotic myelofibrosis diagnosed according to the 2008-WHO-criteria. results of the ANAHYDRET study: non-inferiority of anagrelide compared to Leukemia; e-pub ahead of print 16 September 2011. hydroxyurea in newly diagnosed WHO essential thrombocythemia patients. Blood 38 Barbui T, Carobbio A, Finazzi G, Vannucchi AM, Barosi G, Antonioli E et al. 2008; 112, (abstract 661). Inflammation and thrombosis in essential thrombocythemia and polycythemia 59 Verstovsek S, Passamonti F, Rambaldi A, Barosi G, Rosen PJ, Levy R et al. Durable vera: different role of C-reactive protein and pentraxin 3. Haematologica 2011; 96: Responses with the JAK1/ JAK2 Inhibitor, INCB018424, in patients with 315 --318. polycythemia vera (PV) and essential thrombocythemia (ET) refractory or 39 Tefferi A, Vaidya R, Caramazza D, Finke C, Lasho T, Pardanani A. Circulating intolerant to hydroxyurea (HU). Blood 2010; 116, (abstract 313). interleukin (IL)-8, IL-2R, IL-12 and IL-15 levels are independently prognostic in 60 Barosi G, Birgegard G, Finazzi G, Griesshammer M, Harrison C, Hasselbalch HC primary myelofibrosis: a comprehensive cytokine profiling study. J Clin Oncol et al. Response criteria for essential thrombocythemia and polycythemia vera: 2011; 29: 1356 --1363. result of a European LeukemiaNet consensus conference. Blood 2009; 113: 40 Tefferi A, Fonseca R, Pereira D, Hoagland HC. A long-term retrospective study of 4829 --4833. young women with essential thrombocythemia. Mayo Clin Proc 2001; 76:22--28. 61 Tefferi A, Passamonti F. Essential thrombocyythemia and pregnancy: observations 41 Landolfi R, Marchioli R, Kutti J, Gisslinger H, Tognoni G, Patrono C et al. Efficacy and from recent studies and management recommenadtions. Am J Hematol 2009; 84: safety of low-dose aspirin in polycythemia vera. NEnglJMed2004; 350: 114 --124. 629 --630. 42 Harrison CN, Campbell PJ, Buck G, Wheatley K, East CL, Bareford D et al. 62 Besses C, Cervantes F, Pereira A, Florensa L, Sole´ F, Hernańdez-Boluda JC et al. Hydroxyurea compared with anagrelide in high-risk essential thrombocythemia. Major vascular complications in essential thrombocythemia: a study of the N Engl J Med 2005; 353:33--45. predictive factors in a series of 148 patients. Leukemia 1999; 13: 150 --154. 43 Alvarez-Larran A, Cervantes F, Pereira A, Arellano-Rodrigo E, Pe´rez-Andreu V, 63 Colombi M, Radaelli F, Zocchi L, Maiolo AT. Thrombotic and hemorrhagic Hernańdez-Boluda JC et al. Observation versus antiplatelet therapy as primary complications in essential thrombocythemia. A retrospective study of 103 prophylaxis for thrombosis in low-risk essential thrombocythemia. Blood 2010; patients. Cancer 1991; 67: 2926 --2930. 116: 1205 --1210. 64 Jantunen R, Juvonen E, Ikkala E, Oksanen K, Anttila P, Ruutu T. The predictive 44 Cortelazzo S, Finazzi G, Ruggeri M, Vestri O, Galli M, Rodeghiero F et al. value of vascular risk factors and gender for the development of thrombotic Hydroxyurea in the treatment of patients with essential thrombocythemia at complications in essential thrombocythemia. Ann Hematol 2001; 80: 74 --78. high risk of thrombosis: a prospective randomized trial. N Engl J Med 1995; 332: 65 Bazzan M, Tamponi G, Schinco P, Vaccarino A, Foli C, Gallone G et al. Thrombosis- 1132 --1136. free survival and life expectancy in 187 consecutive patients with essential 45 Maugeri N, Giordano G, Petrilli MP, Fraticelli V, de Gaetano G, Cerletti C et al. thrombocythemia. Ann Hematol 1999; 78: 539 --543. Inhibition of tissue factor expression by hydroxyurea in polimorphonuclear 66 Passamonti F, Rumi E, Arcaini L, Boveri E, Elena C, Pietra D et al. Prognostic factors leukocytes from patients with myeloproliferative disorders: a new effect for an old for thrombosis, myelofibrosis and leukemia in essential thrombocythemia: a study drug? J Thromb Haemost 2006; 4: 2593 --2598. of 605 patients. Haematologica 2008; 93: 1645 --1651. 46 Sterkers Y, Preudhomme C, La¨ı JL, Demory JL, Caulier MT, Wattel E et al. Acute 67 van Genderen PJJ, Michiels JJ. Erythromelalgic, thrombotic and hemorrhagic myeloid leukemia and myelodysplastic syndromes following essential manifestations of thrombocythaemia. Presse Med 1994; 23:73--77. thrombocythemia treated with hydroxyurea: high proportion of cases with 17p 68 Fenaux P, Simon M, Caulier MT, Lai JL, Goudemand J, Bauters F. Clinical course of deletion. Blood 1998; 91: 616 --622. essential thrombocythemia in 147 cases. Cancer 1990; 66: 549 --556.

Leukemia (2012) 875 --882 & 2012 Macmillan Publishers Limited