Aquagenic pruritus is a typical feature of polycythemia vera (PV) reported by up to 65% of patients (pts) at diagnosis (Mesa R, Cancer 2007). It can either ante date diagnosis or appear during the course of disease; it is poorly responsive to phlebotomy or myelosuppressive therapy, while treatment with interferon-alpha or selective serotonin reuptake inhibitors can be sometimes successful. Its pathophysiologic mechanisms are poorly known, and eosinophils, basophils or dermal mast cells have all been hypothesized to contribute. An association with JAK2V617F burden has been reported, since pruritus was significantly more common in pts with >50% mutated alleles (Tefferi A, Cancer 2006; Vannucchi AM, Blood 2007), and the risk of having it was 4.2-fold greater in pts in the highest mutated allele quartile as compared to the lowest one (Vannucchi AM, Leukemia 2007). Therefore, we sought to focus on basophils to underscore their possible involvement in the pathogenesis of pruritus and the role of V617F mutation. According to previous reports, among the 75 pts included in this study the V617F allele burden was significantly greater in those referring pruritus (71±18%) than in those who denied it (48±19%; P=.002). We found that the V617F allele burden measured in immunomagnetically purified basophils was comparable to granulocytes (r=0.9), while the content of JAK2 mRNA was significantly higher in PV basophils than in both PV granulocytes (P<.0001) and normal basophils (P=.02) and granulocytes (P=.01). This was not due to a preferential representation of mutated RNA since the proportion of wild-type and V617F-mutated JAK2 mRNA transcripts was consistent with quantitative genotyping. CD63 is a tetraspanin contained in the granules of resting basophils, which is expressed on outer membrane following their activation. We measured by flow cytometry the number of circulating CD63+ basophils in PV pts. We found that median frequency of CD63+ cells in the CD123+/HLA-DR basophil gate was significantly greater in PV pts than in controls (13.0±10.5% versus 1.8±1.7%; P= .01), and their absolute number increased from 0.4±0.3×106/L in controls to 15.5±11.0 in PV pts (P=.003). Also the expression level of CD63, measured as the mean fluorescence intensity (MFI), was significantly greater in PV pts than in controls (1,405±856 vs 740±145.2; P=.04). PV pts suffering from pruritus had significantly greater percentage of activated basophils than those without: 18.8±10.6% vs 5.6±3.6% (P=.009). Finally, both the percentage and the absolute number of CD63+ basophils were greater in pts with >50% mutated allele compared to those with <50%: 15±9.5% vs 4.7±3.8 % (P= .02) and 21.2±10.8×106/L vs 7.1±6.7×106/L (P= .04), respectively, and they were linearly correlated with V617F burden (r=0.76, P<.01). Transmission electron microscopy analysis showed that, compared to healthy counterpart, PV basophils had reduced number of granules most of which were empty and displayed a disrupted membrane, while the cytoplasm presented abnormal electrondensity and vacuolization. To evaluate the role of JAK2V617F mutation in the activated status of circulating basophils, we analyzed their ex-vivo activation in response to the N-Formyl-Met-Leu-Phe (fMLP) peptide and IL-3 that is the most potent priming cytokine of basophils. Using progressive dilutions of fMLP in IL-3-primed cells, we found that the increase of CD63+ basophils was always significantly greater in PV pts than in controls, particularly in case of >50% V617F allele (P<.01); similarly, when basophils were challenged with varying amounts of IL3 (from 0.1 to 10 ng/mL) and a fixed dose of fMLP, the response of PV cells resulted significantly greater than control cells at all concentrations employed, overall pointing to an hypersensitivity of mutated cells. Finally, preincubation with the potent and selective JAK2 inhibitor AZD1480 (kindly provided by Astra Zeneca Ltd) largely prevented the ex-vivo activation of PV basophils which proved to be even more sensitive than normal cells. These data point to constitutively activated and hyper-reactive basophils as one mechanism contributing to pruritus in PV, and provide evidence for a direct role of JAK2V617F mutation. However, it is conceivable that additional circulating and/or dermal cells may be involved as well as in the pathogenesis of this agonizing symptom.

Disclosures: No relevant conflicts of interest to declare.

Author notes

Corresponding author

Sign in via your Institution