The granulocyte connection in MPD-associated thrombosis

Leukocyte count has been directly correlated with the incidence of thrombosis in the general population. In this issue of Blood, Landolfi and colleagues and Carobbio and colleagues describe a similar scenario in myeloproliferative disorders.

Both polycythemia vera (PV) and essential thrombocythemia (ET) are considered “hypercoagulable” states. Ten percent to 30% of patients with these disorders present with major thrombotic complications; a similar percentage of patients experience the particular complication during their disease course.¹  In both instances, arterial events, especially those involving the cerebral and coronary vasculature, are more frequent than venous events. For PV, the European Collaboration on Low-Dose Aspirin in Polycythemia Vera (ECLAP) database of over 1600 patients continues to provide clinically useful information, including that reported in the current issue of Blood; leu-kocytosis (< 15 × 10⁹/L as opposed to ≤ 10 × 10⁹/L) is identified as an independent predictor of myocardial infarction. In ET, the association between leukocytosis (≥ 15 × 10⁹/L) and thrombosis was first reported by Wolanskyj et al,²  and a similar observation with a leukocyte threshold count of 8.7 × 10⁹/L is now described in this issue of Blood by Carobbio and colleagues. As expected, the above studies in PV and ET also identified advanced age, history of thrombosis, and tobacco use as independent predictors of thrombotic events. It is interesting that a specific thrombosis history in the PV study predicted a similar thrombotic event in terms of both vessel type and location. Regardless, the suggested prognostic relevance of leukocytosis in myeloproliferative disorder (MPD)–associated thrombosis is consistent with the proven antithrombotic efficacy of hydroxyurea in high-risk ET and PV.

The pathogenetic contribution of granulocytes to thrombosis in MPDs might involve cross-talk between granulocytes and platelets and/or endothelial cells. Compared with healthy controls, patients with ET or PV display increased baseline/induced platelet P-selectin expression, platelet-granulocyte/platelet-monocyte complexes, granulocyte activation (increased CD11b expression, etc), and baseline/lipopolysaccharide-induced expression of tissue factor (TF) by both monocytes and neutrophils. Furthermore, these abnormalities might be more pronounced in patients with history of thrombosis and in those with 𝒥AK2V617F. Interestingly, a recent study suggested in vivo down-regulation of both neutrophil TF expression and number of neutrophil-platelet complexes but not platelet P-selectin expression, by hydroxyurea therapy, in patients with either ET or PV.³  This in vivo phenomenon was recapitulated in vitro, where hydroxyurea inhibited both P-selectin–mediated neutrophil TF expression and mixed aggregate formation in healthy subjects.³  Also, in vitro induction of platelet-neutrophil complex formation is decreased in ET patients receiving aspirin therapy.⁴  These observations provide additional novel mechanisms of action for both hydroxyurea and aspirin that might contribute to their proven efficacy in the treatment of high-risk patients with MPDs.

What is the practical relevance of the above-mentioned observations? First, there is now enough information to warrant prospective examination of the relationship between leukocyte count and thrombosis in MPDs. For example, one can envision a randomized study of hydroxyurea versus placebo in otherwise low-risk ET or PV associated with leukocytosis. In regards to current practice, the observations further undermine the role of anagrelide in the treatment of MPDs. Also, one could argue in favor of paying more attention to the leukocyte instead of platelet count in high-risk patients on cytoreductive therapy, although such practice requires prospective validation. Finally, in intermediate-risk patients in whom cytoreductive therapy is contemplated for one reason or another, the leukocyte count can now be considered as an additional factor in the treatment decision process.