Disease Burden in the Chronic Myeloproliferative Disorders Correlates with the Stem Level of JAK2 V617F Expression.

Abstract

The chronic myeloproliferative disorders (MPD), polycythemia vera (PV), essential thrombocytosis (ET) and idiopathic myelofibrosis (IMF) differ phenotypically but are linked by a common genetic lesion, JAK2V617F. Additional molecular lesions, host genetics or stem cell level of involvement are possible explanations for the differing myeloproliferative phenotypes associated with similar degrees of JAK2V617F. In this study, we examined the correlation between the stem cell level of JAK2V617F involvement and disease phenotype. The study population consisted of 35 PV, 14 ET and 7 IMF patients followed by us with a median disease duration of 8 years (range 1–27 years); all patients were clinically assessed between 2005 and 2006. Genomic DNA was extracted from neutrophils, T, B, and CD34+ cells purified from blood samples at the time of clinical assessment and in 4 patients, from erythroid colonies cloned from CD34+ cells cultured in semisolid media. The quantitative JAK2V617F allele percentage was obtained using a sensitive allele-specific, quantitative PCR assay. In PV patients with marked splenomegaly and leukocytosis, median CD34+ cell and neutrophil JAK2V617F allele percentages were the same (85 versus 87). By contrast, in PV patients without splenomegaly, median CD34+ cell allele percentages were lower than neutrophil JAK2V617F allele percentages (41 versus 62); these patients also had lower leukocyte counts (p<0.02) and lower JAK2V617F allele percentages in both CD34+ cells (p<0.002) and neutrophils (p < 0.02) compared to PV patients with splenomegaly, independent of disease duration. In 2 patients with substantial splenomegaly and elevated leukocyte counts, whose neutrophil and CD34+ cell JAK2V617F allele percentages were similar, 20 erythroid colonies from each were all JAK2V617F-positive and had the same JAK2V617F allele percentage as the neutrophils and CD34+ cells. In contrast, in two patients with normal leukocyte counts and no splenomegaly, whose CD34+ cell JAK2V617F allele percentages were lower than their neutrophil allele percentages, a significant number of erythroid colonies in each case were JAK2V617F-negative. JAK2V617F was detected in the T and B cells of 5/5 IMF (median T cell 25, range 18–50; neutrophil median 75, range 55–100), 11/23 PV (median T cell JAK2V617F allele percentage 30, range 14–82; neutrophil median 60, range 48–100), and 2/5 ET patients (median T cell 0, range 0–14; neutrophil median 54, range 35–60). Splenomegaly was more prevalent in PV patients with T cell JAK2V617F allele percentages greater than 20 (50%) compared to PV patients with T cell allele percentages less than 20 (14%); there was no association between T cell involvement and disease duration or neutrophil JAK2V617F allele percentage. These data indicate first, that T and B cells are involved in IMF and in PV but rarely in ET. Second, significant T cell JAK2V617F burden correlated with presence of extramedullary disease. Third, in PV patients, concordance between neutrophil and CD34+ cell JAK2V617F allele percentage was associated with extramedullary disease. The data suggest that JAK2V617F expression can differ with respect to stem cell level in MPD patients and this appears to account for differences in extramedullary disease. Measurement of T cell JAK2 V617F expression may be a useful prognostic indicator of for aggressive disease in MPD patients.