REVEAL puts leukocytes into risk stratification

In this issue of Blood, Gerds et al identified persistent leukocytosis above the upper normal value as a risk factor for thrombosis in polycythemia vera. Furthermore, this association has been demonstrated in both arterial and venous thrombosis, in high- and low-risk patients, and even in patients in whom hematocrit was adequately controlled.¹ 

Thrombosis risk stratification is the basis for treatment selection in patients with polycythemia vera. Until now, the ECLAP (European Collaboration on Low-dose Aspirin in Polycythemia Vera) study, which included 1638 patients with a prospective follow-up of 2.7 years, provided the best evidence to guide treatment in this disease.² In that study, age and history of thrombosis were the main risk factors for thrombosis. Patients are usually stratified on the basis of this study into 2 risk categories: low risk (aged <60 years without a history of thrombosis) and high risk (aged >60 years or history of thrombosis). Since the publication of the ECLAP study in 2005, most clinical practice guidelines recommend treating low-risk patients with phlebotomies and high-risk patients with cytoreduction.^3,4 

The REVEAL (The Prospective Observational Study of Patients with Polycythemia Vera in US Clinical Practices) study included a total of 2510 patients with a median follow-up of 3.7 years, and it is the largest prospective study published to date in polycythemia vera.¹ In the 1768 high-risk patients, leukocytosis, thrombocytosis, and hematocrit >45% were associated with a higher probability of thrombosis, supporting that normalization of blood counts is associated with lower risk of thrombosis. This observation is in line with the results of a randomized clinical trial in patients with polycythemia vera resistant/intolerant to hydroxyurea, in which complete hematological response while on cytoreduction was associated with better event-free survival.⁵ Regarding the type of thrombosis, the REVEAL study confirms that there are different risk factors for arterial and venous thrombosis, an observation that has also been reported in essential thrombocythemia.⁶ Both leukocytosis and thrombocytosis, poor hematocrit control, and age were associated with a higher risk of arterial thrombosis, whereas female sex, history of previous thrombosis, and leukocytosis turned out to be the main risk factors for venous thrombosis. The fact that age was not associated with the risk of venous thrombosis could be explained by the excess burden of this type of thrombosis in patients with polycythemia vera of any age, which could cancel out the risk described for age in the general population. In this sense, venous thrombosis has been associated in polycythemia vera with a neutrophil-to-lymphocyte ratio >5 and a JAK2V617F variant allele frequency >50%, suggesting an important role for the biological characteristics of the mutated clone in this type of thrombosis.^7,8 

Some of the limitations of the REVEAL study include the fact that the median time elapsed from diagnosis to inclusion was 4 years. Thus, the proportion of newly diagnosed patients was small, which excludes from the analysis the period with higher thrombotic risk. On the other hand, cardiovascular risk factors have not been considered in the different analyses, precluding a precise evaluation of risk factors for arterial thrombosis. Finally, there was no information available on the mutational load of JAK2 or the neutrophil-to-lymphocyte ratio that could help confirm previous observations in this regard, as well as exploring its interaction with leukocytosis.

Previous studies had postulated the possible role of leukocytosis as a risk factor for thrombosis and there is an important body of evidence on the role of leukocyte and platelet activation in the pathogenesis of thrombosis in patients with myeloproliferative neoplasms.⁹ Despite this evidence, current clinical practice guidelines do not include leukocytosis in risk stratification.^3,4 Although a prospective clinical trial proving that normalization of leukocytosis reduces this risk could be of interest, low thrombotic rate and the need for a high number of patients with long follow-up make this option unfeasible. In the meanwhile, data from REVEAL support the inclusion of leukocytosis in the definition of high-risk polycythemia vera, and 2 aspects of clinical practice could be changed. First, classic low-risk patients with persistent leukocytosis could be elevated to the high-risk category and, therefore, be candidates for cytoreduction. Second, normalization of leukocytes should be included as an objective of treatment in patients receiving cytoreduction.

Conflict-of-interest disclosure: The author declares no competing financial interests.