Blast Transformation in a Patient with Primary Myelofibrosis Initiated from JAK2 V617F Progenitor.

Myeloproliferative disorders (MPDs) are caused by clonal proliferation arising from a single multi-lineage stem cell. The JAK2 V617F mutation has been reported in greater than 90% of patients with polycythemia vera (PV), and ∼50% of patients with essential thrombocythemia (ET) and primary myelofibrosis (PMF). However, several studies have indicated that the JAK2 V617F mutation is not an MPD initiating mutation but rather represents clonal evolution of these MPDs. Jelinek and colleagues first reported that most PV transformed acute leukemias are JAK2 V617F negative (Jelinek, Blood, 2005 106:3370). More recently, the role of the JAK2 V617F mutation in leukemic transformation in 27 patients with MPDs revealed that most JAK2 V617F -positive MPD patients transformed to a JAK2 V617F -negative AML (Theocharides, Blood, 2007 110:375); however, in the 4 patients with an apparent JAK2 V617F -positive leukemia clonality of leukemic blasts and mature granulocytes was not determined. Two models proposed by Theocharides et al may explain these findings. First, MPD and AML represent 2 independent clones that arose de novo from different progenitors. Second, MPD and AML are 2 subclones derived from a common progenitor. Here, we describe a woman with PMF with transformation to AML. We determined her JAK2 V617F mutation status by sensitive and quantitative real-time PCR (Nussenzveig, Exp Hematol, 2007 3:32). At the time of her transformation to AML, her normal appearing peripheral blood granulocytes were purified and the frequency of mutant JAK2 allele T was 6%. However, all FACS-sorter isolated CD34+ cells (enriched to 95% purity) were heterozygous for the JAK2 V617F mutation. To determine if MPD and AML clones arose de novo or from the same progenitor, we performed clonality studies using a newly developed sensitive and quantitative real-time PCR based on the X-chromosome inactivation principle using transcriptional clonality assays in granulocytes and CD34+ purified cells from peripheral blood at both stages of disease (see Swierczek et al, abstract, this meeting). When this woman’s PMF was first discovered, hematopoiesis was clonal, based on heterozygosity for three X-chromosome genes, FHL1, G6PD and IDS (Liu, Blood, 2003 101:3294) and their single allelic expression in granulocytes and platelets. At the time of leukemic transformation, both her granulocytes and leukemic CD34+ cells expressed all three identical isoforms from the same parental X chromosome. Our findings indicate that leukemic transformation does not invariably arise from a JAK2 V617F negative progenitor. This has important implication for therapy of MPDs with JAK2 V617F inhibitors, as these would not prevent leukemic transformation. It remains to be determined if the JAK2 background of leukemic progenitors is variable, and if there are differences between PV and PMF.