Abstract 2864

Myeloproliferative neoplasms (MPN) are clonal disorders characterized by excessive production of mature blood cells and secondary stromal changes in the bone marrow leading to myelofibrosis. The role of a number of fibrogenic cytokines derived from megakaryocytes or monocytes in driving reactive secondary responses of bone marrow stroma cells (BMSC) in fibrosis has been repeatedly discussed. The aim of this study is the comparative analysis of BMSC from MPN patients and non-MPN donors in regard to hematopoietic stem and progenitor cell supporting capacity, extracellular matrix (ECM) remodeling in 3D collagen-based scaffolds as well as their in situ localization in the bone marrow niche.

BMSC from bone marrow routine aspirates were obtained from patients diagnosed with chronic myeloid leukemia (CML; n=5), essential thrombocythemia (ET; n=5) and polycythemia vera (PV; n=5) and compared to BMSC isolated from control BM (untreated initial diagnosis of non-Hodgkin′s lymphoma or from patients undergoing orthopedic joint replacement surgery). BMSC cultures from prefibrotic MPN patients (as determined by reticulin staining) were established and fulfilled MSC criteria according to common consensus comparable to BMSC cultures from control individuals. As BMSC have been shown to support hematopoiesis in vitro, we compared their constitutive production of hematopoietic-affiliated cytokines. BMSC from MPN and non-MPN patients expressed IL-7, M-CSF, FLT3-L, IL-6, TPO and LIF in the following order: control BMSC > ET BMSC > PV BMSC > CML BMSC. In myeloid colony formation unit (CFU) assays using healthy CD34+ hematopoietic stem and progenitor cells as readout, myeloid CFU activity was highest in the supernatant of control BMSC suggesting a decrease of hematopoiesis supporting capacity in BMSC from MPN patients.

When activated through contact with the collagenous matrix in 3D scaffolds, only BMSC from ET (in 3 of 5 cases studied) or CML (1/5) patients extensively remodeled and significantly contracted the collagenous matrix. A significant up-regulation of ECM proteins –fibronectin, collagen type I, collagen type IV, laminin and osteopontin - detected by qtRT-PCR and immunohistochemistry was seen in BMSC from MPN-patients. To evaluate if BMSC contribute to ECM remodeling in vivo we analyzed fibronectin deposition in corresponding bone punches of ET patients. Co-stainings with the recently identified BMSC markers CD271 and CD146 revealed that BMSC in ET patients strongly co-expressed fibronectin and were mobilized from their perivascular and endosteal niche. BMSC in ET patients diffusely localized in the bone marrow and seemed to be attracted by dysplastic megakaryocytes. In addition, local accumulation of unbound fibronectin was detected in prefibrotic ET patients. Remarkably, conventional reticulin staining was absent in these patients suggesting a predictive role for fibronectin staining in myelofibrosis. Therefore - and because recent evidence has shown that the amount of reticulin staining does not correlate with disease progression and prognosis - we analyzed matrix remodeling in bone marrow core biopsies on tissue microarrays including CML (n=16), ET (n=15) and PV (n=16) in comparison to primary myelofibrosis (MF, n=12) and non-Hodgkin′s lymphoma (n=19). Clinical data including several laboratory parameters associated with myeloproliferative disorders (spleen size, molecular diagnostics) have been assessed retrospectively and correlated with fibronectin and CD271 expression. There was a positive correlation for higher fibronectin and CD271 expression with lower haemoglobin levels (p = 0.001[both]) and higher blast counts (p = 0.006 [fibronectin]; p = 0.027 [CD271]) at diagnosis, but no correlation with JAK2 V617F or BCR/ABL expression or spleen size.

In conclusion, our data indicate that MSC from MPN patients actively participate in the process of myelofibrosis after being recruited and attracted by the malignant hematopoietic clone/dysplastic megakaryocytes. Further analysis will reveal if fibronectin and CD271-staining are suitable routine markers for disease prognosis and potential targets for therapeutic strategies aimed at the prevention of disease progression towards secondary MF.

Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution