For Physicians

Thrombosis and the JAK2 Mutation

Polycythemia vera (PV) and essential thrombocytosis (ET) generally are considered to be rather benign diseases associated with a long life-expectancy; however, late in the disease, about 8-15 years after diagnosis, repeated complications usually impact survival. Survival is much shorter in primary myelofibrosis (PMF) and can range from a only few months to a decade or so depending on various risk factors and comorbidities. Cardiovascular events, both arterial and venous, are the leading cause of morbidity and mortality in these three classic myeloproliferative neoplasms (MPN). The reported cumulative rate of thrombosis in PV is 2.5-5% per patient-year, and in ET from 1.9-3% per patient-year. Data on frequency of thrombosis in PMF are limited, but one study found that at least 20% had experienced an event, and venous thrombosis was especially common. An important finding in all these studies has been that after an initial thrombotic event, the recurrence rate was substantially higher, about 5.6% per patient-year with a cumulative probability of a second event of 50% by year 10. Arterial events constituted 60-70% of all cardiovascular incidents¹.

JAK2 Mutation and Myeloproliferative Neoplasms

Normally JAK2 is a gene that encodes a tyrosine kinase (Janus kinase2) associated with certain cytokine receptors located in the cytoplasm of cells. When a specific growth factor (e.g. erythropoietin) becomes linked to its cytokine receptor, Janus kinase proteins, including the JAK2 tyrosine kinase, are activated and mediate intracellular signaling and cell growth. As recently as 6 years ago, a somatic mutation of the JAK2 gene was identified within hematopoietic cells. This mutation is acquired during life and has been shown to cause related diseases designated as myeloproliferative neoplasms¹. A point mutation in JAK2 exon 14 results in a valine to phenylalanine substitution at position 617 (V617F); this in turn causes mutant JAK2 tyrosine kinase to become autonomous and function independently of regulatory processes, constitutively activating downstream signaling with resultant stimulation of hematopoietic cell lines. Proof that this mutation causes MPN is illustrated by experiments showing that when these mutant cells are transplanted into murine recipients, a polycythemia vera-like disease (PV) results initially, and later a myelofibrosis-like disorder (PMF). Recently, transgenic mice have been developed that simulate essential thrombocytosis (ET). In humans JAK2V617F has been found in 65-97% of cases of PV, and 23-57% of cases of ET or PMF^1,3.

Splanchnic vein thromboses, including portal, mesenteric, and hepatic veins with Budd-Chiari syndrome, are unusually prevalent in MPN. A study of 247 random cases found that 45% of Budd-Chiari, and 34% of portal vein thromboses had the JAK2 mutation². For those patients presenting with overt PV, ET, or PMF at the time of splanchic thrombosis, JAK2 was detected in 94.7%³ in one series, and 96.5% in another². Nevertheless, more than 21% of abdominal cases and 5% of cerebral venous thromboses still had the mutation without overt signs of MPN². Similarly, a study of 42 patients with catastrophic intra-abdominal thromboses requiring visceral transplants found JAK2 mutations in 7 of 42 (17%) and only 3 of these 7 had MPN. An important observation in this particular series was that post-transplant survival was strikingly longer for those without the mutation, 116.4 months compared to 17.5 months when the mutation was present ⁴.

The natural history of carriers of the JAK2 mutation with splanchnic venous thrombosis without underlying MPN is not clear due to paucity of follow-up data. In one series where follow-up data were available (median 41 months), of 99 cases with splanchnic thrombosis 10 were JAK2 positive without disease, and only 3 of these 10 subsequently developed MPN ⁵. Whether carriers of the mutation without overt disease are destined to develop MPN still remains speculative.

A recent study of 444 patients with peripheral venous thrombosis and/or pulmonary embolism, and 60 patients with ischemic stroke, detected unexpected JAK2 mutations in 1.4% of venous thromboses, and 3.3% cases of stroke. The incidence in a healthy control population was 0.002% ⁶. A similar study of patients with various thromboses found JAK2 mutations in DVT (3%), and cerebral venous thrombosis (3%); no carriers were found in 111 controls⁷. These reports contradict earlier ones and are the first to find an increased prevalence of the mutation in patient groups with peripheral thromboses. They raise the controversial question of screening patient populations of arterial and venous thromboses with the goal of detecting JAK2V617F and presumably early cases of MPN.

Fatal cardiovascular events are the leading cause of morbidity and mortality for classic MPN. Many epidemiologic and clinical studies have indicated that MPN, including PV, ET, and PMF are thrombophilic conditions. Age >60 and prior thrombosis have been identified as the most predictive clinical risk factors for PV and ET, although generic cardiovascular risk factors may interact and increase thrombotic risk. Clinical risk factors associated with thrombosis in PMF are cardiovascular risk factors, thrombocytosis, and the cellular phase of the disease¹.

The pathogenesis of thrombosis in MPN is multifactorial and complex. Previous explanations have been too simple, emphasizing mainly elevated hematocrit in PV and thrombocytosis in ET as major factors. However, the explanation is not straightforward, as major studies in PV relating thromboses and hematocrit, and in ET relating thrombosis and platelet counts, have been inconsistent; some do, and others do not show a correlation¹. Factors such as blood viscosity (in PV), abnormal platelet function and numbers, activation of leukocytes, defective endothelial cell function, and an ill-defined hypercoagulable state may all coincide to produce a thrombotic diathesis. Recent studies have pointed to leukocytosis as being the most relevant risk factor for thrombosis¹, as this may be a marker for a cellular phase of disease associated with activated neutrophils, cytokines, and other undefined aspects of hyper- coagulability.

In addition, there is increasing evidence that JAK2V617F plays an important role in the pathogenesis of thrombosis. This mutation is present in about half of ET patients, which facilitated a study comparing the risk of arterial and venous thrombosis in those with and without the mutation. Mutated patients younger than 60, had a nearly four-fold increase in thrombosis risk ⁸, and if homozygous, the risk for recurrence also was increased. An extensive literature review confirmed that mutation positive patients had a significant increased risk of arterial and venous events in ET, but only a trend could be established for PMF⁹. Involvement of the microcirculation is particularly common in ET with symptoms of erythromelalgia, transient visual or hearing defects, headache, and peripheral paresthesia; all of these symptoms are more common in mutation positive individuals. It has not been possible to study PV comparing mutation positive to negative patients, since almost all individuals are mutation positive. However, a study of 173 patients found a definite correlation between the JAK2 mutation allele burden and high risk disease. Patients with > 75% mutant allele had higher hematocrits and white cell counts, larger spleens, more pruritis, and a seven-fold increased risk for a major cardiovascular event as compared to PV patients with <25% mutant allele¹⁰.

The data presented here indicate that JAK2V617F mutation positivity, as well as a high mutant allele burden and elevated leukocyte count, seem to be predictive of cardiovascular complications in patients with MPN. Since, this conclusion is based mainly on retrospective studies, it would be inappropriate to use this data alone to institute cytoreductive therapy; clinical risk factors also must influence this important decision. JAK2 inhibitors currently are available and in clinical trials. Data acquired from these trials regarding the impact of inhibitors on disease activity and their potential for reducing cardiovascular events should prove very interesting and important.

References:

1. Vannucchi A., Curr Hem Malig Rep, 2010, 5:22
2. Kiladjian J., Blood, 2008, 111:4922
3. De Stefano V. J Throm Haemost, 2007, 5:708
4. McMahon C., Am J Clin Path, 2007,127:736
5. Colaizzo D., J Thromb Haemost, 2007, 5:55
6. Zerjavic K., Acta Haematol, 2010, 124:49
7. Shetty S., Am J Clin Path, 2010,134:82
8. DeStefano V., Haematologica,2009,94:733
9. Caberlon L., Thromb Res, 2009 124:409
10. Vannucchi A., Leukemia, 2007, 21:1952