Bone marrow and spleen neoangiogenesis is a relevant feature of patients with myelofibrosis (MF). We have previously reported that patients with MF have an increased percentage of circulating endothelial progenitor cells (EPC) assessed as CD34+CD133+VEGFR2+ cells compared with patients with other Ph-negative myeloproliferative disorders (polycythemia vera, PV, and essential thrombocytemia, ET) and healthy subjects. However, neither the functional activity of these putative EPC nor their belonging to the malignant clone have been yet fully characterized. In order to address these issues we have grown in vitro EPC-derived colonies from the peripheral blood (PB) of 36 patients with MF, 9 patients with PV or ET and 10 healthy subjects. Seventeen MF patients harbored a V617F JAK-2 mutation (8 heterozygous and 9 homozygous) whereas 2 patients showed a W515L MPL mutation (both heterozygous). Eight out of 9 PV/ET patients had a V617F JAK-2 mutation (5 heterozygous and 3 homozygous). Mononuclear cells were cultured in collagen coated 6 well plates in the presence of EBM-2MV medium according to Ingram et al (

). The endothelial origin of the colonies was ascertained by assessment of the expression of CD105, CD146, CD144, CD31, vWf, VEGFR-2, CD14 and CD45 antigens. V617F JAK-2 and W515L MPL mutations were assessed by PCR, followed by enzymatic digestion, of endothelial cells after tripsinization of the EPC-derived colonies. The median frequency (number of colonies per 107 mononuclear cells plated) of EPC-derived colonies was statistically higher in MF patients (0.25, range 0–8.1) compared to healthy subjects (0.05, 0–0.3; P=0.037), but not different form that of PV/ET patients (0, 0–4.4; P=NS). Immunophenotyping confirmed that the cells expressed the endothelial antigens CD105, CD146, CD144, CD31, vWf, and VEGFR-2 but not the hematopoietic specific antigens CD45 and CD14. The capacity of colony-derived endothelial cells of MF patients to form capillary-like structures in the Matrigel assay was not different from that of healthy subjects. No correlation was found between the number of colonies and the mutational status of either JAK-2 or MPL. In 11 MF patients harboring either a JAK-2 (n=9) or a MPL (n=2) mutation, colony growth was observed and PCR was performed on EPC-derived colonies. In 0/9 and 0/2 cases neither JAK-2 nor MPL mutations were found, respectively. In addition, no V617F JAK-2 mutation was found in the EPC-derived colonies of 8 PV/ET patients who carried the mutation in their granulocytes. Taken together, our data show that patients with MF have an increased frequency of EPC in their PB compared to healthy subjects and that these mobilized EPC are not clonally-related to the JAK-2 or MPL mutated clone. Whether or not circulating EPC derive from an earlier progenitor cell compared to the one in which the JAK-2/MPL mutations arise remains to be determined.

Author notes

Disclosure: No relevant conflicts of interest to declare