Myeloproliferative Neoplasm (MPN) Blastic Transformation Occurs at the Level of Both Short and Long Term Hematopoietic Stem Cells

Myeloproliferative neoplasms (MPN) are chronic hematopoietic malignancies characterized by the clonal proliferation of mature myeloid elements (Campbell PJ, Green AR. N Engl J Med. 2006; 355:2452). A subset of MPNs progress to MPN-blast phase (BP). The cells that are responsible for the progression to BP and the mechanisms and pathways that underlie this progression have not been well delineated. We therefore established an in vivo MPN-BP xenograft mouse model to address these questions.

Among the 22 patients studied, 11 were cytogenetically normal while the remainder had chromosomal abnormalities including del(5), del(20q), del(14), +1q, multiple copies of 1. Capture based next generation sequencing of MPN-BP mononuclear cells (MNCs) identified the following mutations in JAK2 (71%), ASXL1 (19%) , TET2 (19%), MPL (14%), SF3B1 (14%), RUNX1 (14%), U2AF1 (14%), PTPN11 (14%), IDH1/2 (10%), SRSF2 (10%) and TP53 (10%). 86% of the patients had at least 2 mutations in genes associated with oncogenesis. These findings suggest that MPN-BP is genetically heterogeneous.

In order to develop an in vivo assay for MPN-BP stem cells, CD3⁺ cell-depleted MNCs collected from the peripheral blood (PB) of 7 patients (Pt 1-4 and 6:1×10⁷ cells/mouse, Pt 5 and 7: 2.5×10⁶ cells/mouse; 3-5 mice/sample) were transplanted intravenously into sublethally irradiated NOD/SCID/IL-2Rγnull (NSG) mice. All of the transplanted cells from Pt 5 had del(20q) and 20% also had +8; 88% of cells from Pt 6 had +1q, while the remaining 5 patients were each cytogenetically normal. Four months after the transplantation, flow cytometric analysis showed that samples from 6 of 7 patients engrafted NSG mice [>0.5% hCD45⁺ cells in bone marrow (BM)] and generated myeloid blasts in the BM, spleens and PB to a varying degree. The degree of human chimerism did not appear to be correlated with the presence of particular genetic mutations or cytogenetic abnormalities. Morphological examination revealed that 20-30% (Pt 2) and 30-50% (Pt 4) of the cells in both the BM and the spleens of the transplanted mice were blasts. Mice receiving cells from Pt 2 and 4 were characterized by a reduced WBC and RBC count and there was a positive relationship between the level of hCD45⁺CD33⁺ cell chimerism and the degree of leukopenia. Mice receiving cells from Pt 2 and 4 experienced a 1.9- and 3.2-fold increase in splenic weight, respectively, as compared with control mice receiving PBS alone. Grafts from Pt 3 and 6 generated myeloid cells exclusively (CD33⁺, Gly A⁺, CD41a⁺), grafts from Pt 2 and 7 produced human cells belonging to the myeloid lineage but with few CD3⁺ T cells but no CD19⁺ B cells, while samples from Pt 4 and 5 produced a large number of cells belonging to both myeloid and lymphoid lineages (mostly CD3⁺ and few CD19⁺). Moreover, hCD33⁺, hCD19⁺, hCD3⁺ cells as well as hCD34⁺ cells selected from the BM of mice receiving the Pt 5 graft had the identical proportion of chromosomally abnormal [del(20q), +8] cells as primary cells. These findings suggest that long-term MPN-BP stem cells are capable of modeling human MPN-BP in a xenograft system.

We also demonstrated that the progression to BP from MPNs in some cases originates at the level of a short-term MPN-BP stem cell. Human cell engraftment characterized by a significant burden of myeloid blasts in the BM was observed as early as 1 month following the transplantation with Pt 6 cells (1X10⁷/mouse) and was further increased after 2 months, but the degree of human cell chimerism 4 months after the transplantation was diminished (human cell chimerism 1, 2, 4 months after the transplantation: CD45⁺: 40.5%, 55.6%, 2.8%; CD34⁺: 23.2%, 40.2%, 0%; CD45⁺CD33⁺:6.9%, 16.8%, 0.7%; CD45⁺CD19⁺: 0.1%, 0.4%, 0%; CD45⁺CD3⁺:11.7%, 11.3%, 0%). Although 4% of hCD45⁺CD33⁺ cells having +1q were detected in the BM of mice 1 month after the transplantation with Pt 6 cells, all the human cells were cytogenetically normal 4 months after the transplantation. These findings suggest that MPN-BP can originate in either long-term or short-term MPN-BP stem cells and that effective therapeutic strategies must target both classes of malignant stem cells.