New blood vessel formation (angiogenesis) is a multistep process that involves extracellular matrix remodeling, endothelial cell migration and proliferation, capillary differentiation and anastomosis formation. Angiogenesis has a key role in tumor growth and spread. Vascular endothelial growth factor (VEGF) is a positive regulatory cytokine and has a key, rate-limiting dose in promoting tumor angiogenesis. Abnormal VEGF expression has been described in acute leukemias as well as lymphomas. The available data in chronic myeloproliferative neoplasms (MPN) suggests that microvascular density (MVD) and VEGF expression are involved in angiogenesis process.
The aim of this study was to evaluate MVD, VEGF expression and its correlation with clinical parameters.
The study was performed according to the regulation of the Ethics comitee. We included 41 patients with newly diagnosed MPN and 6 controls [28 PV, 23 PMF, 6 secondary eryhtrocytosis] from 2006 through 2009. G-band karyotype, FISH analysis (chromosomes 5, 7, 8 and 20) and JAK2V617F mutation were done according to standard techniques. Two experienced pathologists assessed morphology on hematoxylin-eosin stained slides, bone marrow fibrosis on Gomori silver slides and fibrosis was determined according to Thiele criteria (Thiele et al 2005). Bone marrow MVD was visualized in paraffin tissue sections using CD34 immunohistochemical staining and classified using a score system 0–4. Immunoexpression of VEGF was done in bone marrow core biopsies. A VEGF score was based on VEGF-positive megakaryocytes vs VEGF-immunohistochemistry staining. Statistical analysis was done using Chi-square test.
Most of PV patients had a low MVD (grade 0+1) (73%), whereas the majority of PMF patients had high MVD (grade 2+4) (76 %). MVD score was higher in PMF than PV (median grade 3 vs 1, respectively p=0.8). There were no differences in MVD score (both PV and PMF) according to JAK2V617F mutation status (respectively p=0.79 and p=0.8). Interestingly, PV and PMF patients with high MVD presented high platelet count (315 vs 694 × 109/L in PV, 664 vs 985 ×109/L in PMF); and a high incidence of thrombosis (18.1 vs 50% in PV p=0.9; and 0 vs 26.3% in PMF). VEGF score was higher in PV than PMF (median score 3 vs 2 p=0.22). Thus, PV and PMF patients with low MVD presented a higher rate of abnormal karyotype than high MVD patients (18% vs 0 in PV; 33 vs 10,5% in PMF p=0.0002). We combined data and analyzed together PV and PMF patients who presented low MVD and high MVD. High MVD patients (n=23) were older than low MVD group (68 vs 59 p=0.77), had low hemoglobin value (12.5 vs 18.5g/dL p=0.10), presented high platelet count (738 vs 351 × 109/L) and had a high incidence of thrombosis (30.4 vs 15.3% p=0.98) but there was no difference in VEGF score (median score 3 p=0.39). There was a correlation of high MVD and advanced fibrosis in (r=0.54 Pearson test). Low and high MVD patients had similar JAK2V617F rate mutation (61.5 vs 60.8%, p=0.4) and low MVD patients presented a slightly higher incidence of abnormal cytogenetic findings (31 vs 6.6%). There was no correlation with MVD and VEGF score (r=0.34), age (r=0.17), platelet count (r=0.27). There were no differences in VEGF score according to JAK2V617F mutation status. There was a correlation between VEGF score (grade 2+3) and the occurrence of massive splenomegaly in PV patients (r=0.52 Pearson test). PV patients who presented advanced fibrosis had a high platelet compared to mild fibrosis (567 vs 1031× × 109/L). There were no changes in platetet count regarding to fibrosis degree in PMF.
According to our results we found increase of platetet count and thrombosis in patients with high MVD. VEGF score did not relate to MVD findings neither thrombosis. The occurrence of high MVD correlated with high platelet count and thrombosis, suggest that MVD mechanisms may determine clonal changes and influence platelet differentiation and survival. Probably other biological interactions such as abnormal signaling network in bone marrow may play a role in determining MVD changes in MPN. This question remains to be further validated and investigated in other studies. Our current data show that stromal abnormalities were correlated with thrombosis and raise the question of the evolutive changes in the hematopoietic stem cell leading to abnormal changes in bone marrow during the course of the disease.
No relevant conflicts of interest to declare.
Asterisk with author names denotes non-ASH members.