Relationship Between Granulocyte JAK2 (V617F) Mutant Allele Burden and Risk of Progression to Myelofibrosis in Polycythemia Vera: a Prospective Study of 338 Patients.

Abstract 751

An identical gain-of-function mutation of JAK2 is found in about 95% of patients with polycythemia vera (PV). According to a two-step model [N Engl J Med. 2005 Apr 28;352(17):1779-90], the occurrence of JAK2 (V617F) gives rise to a clone that is heterozygous and expands to replace hematopoietic cells without the JAK2 mutation. A mitotic recombination in a hematopoietic cell that is heterozygous for JAK2 (V617F) later generates uniparental disomy and 9pLOH. The daughter cell that is homozygous for JAK2 (V617F) gives rise to a new clone that expands and replaces the previous heterozygous clone. Therefore, variable proportions of JAK2 (V617F) mutant alleles are found in myeloid cell populations from PV patients. A mutant allele dosage effect on phenotype has been described, and PV patients with high mutant allele burden have been found to have a more severe disease. Patients with post-PV myelofibrosis have the highest mutant allele burdens [median value of about 90% - Blood. 2008 Apr 1;111(7):3383-7]. Interestingly, JAK2 (V617F) activates circulating granulocytes, and by this means likely plays a role in the constitutive mobilization of CD34-positive cells into peripheral blood that characterizes the transformation of PV into post-PV myelofibrosis [Blood. 2006 May 1;107(9):3676-82]. Since all these observations may suggest that the mutant allele burden contributes to determining the myelofibrotic transformation of PV, we examined PV patients enrolled in a prospective observational cohort study. As of August 10, 2009, 338 patients diagnosed with PV according to the 2008 WHO criteria have been enrolled in this study. Of these patients, 320 (94.7%) carried JAK2 (V617F), 14 (4.1%) had JAK2 exon 12 mutations, and 4 (1.2%) did not carry JAK2 (V617F) nor exon 12 mutations despite a typical PV phenotype. Of the 320 patients carrying JAK2 (V617F), 146 were enrolled at diagnosis and 174 at follow-up. Patients were routinely treated with phlebotomy and low dose aspirin, while those at high risk for thrombosis (history of previous thrombosis and/or age greater than 60 years) were given also cytoreductive therapy. Diagnosis of post-PV myelofibrosis was based on the IWG-MRT criteria, while diagnosis of myelodysplastic syndrome or acute myeloid leukemia (AML) was done according the 2008 WHO criteria. In order to accurately assess the granulocyte mutant allele burden, we refined a previously described quantitative real-time polymerase chain reaction (qRT-PCR)-based allelic discrimination assay. This assay is now routinely calibrated using defined standards [Haematologica. 2009 Jan;94(1):38-45] and has a sensitivity equal to 0.2% mutant alleles. Within 320 JAK2 (V617F)-positive patients, the median mutant allele burden was 47% (range 1.1-100%); 167 (52%) patients had less than 50%, while 153 (48%) had more than 50% mutant alleles. PV patients at diagnosis had significantly lower mutant allele burdens than those enrolled in the study at follow-up (P = .002). During the study period, disease transformation occurred in 18 patients. Eight patients, all with more than 50% JAK2 (V617F) mutant alleles at study entry, progressed to post-PV myelofibrosis, while 10 patients developed AML. Since about half of the patients were enrolled at follow-up, survival analyses were carried out accounting for left censoring of the observation. Cox proportional hazard regression showed that the JAK2 mutant allele burden, analyzed as a continuous variable, was related to hematologic transformation-free survival (HR: 1.025, 95% CI 1.005-1.046; P = .015). By categorizing the mutant allele burden, patients with more than 50% mutant alleles had a significantly worse hematologic transformation-free survival (HR: 6.54, 95% CI 1.47-29.1; P = .013) compared with those with lower mutant allele burden. After adjusting for age in a multivariable analysis, the 50% cutoff retained statistical significance (P = .048). With respect to the risk of progression to post-PV myelofibrosis, the JAK2 mutant allele burden, considered as a continuous variable, was significantly related to myelofibrosis-free survival (HR: 1.04, 95% CI: 1.004-1.08; P = .029). In a multivariable analysis with allele burden and age as covariates, the mutant allele burden showed an independent effect on myelofibrosis-free survival (P= .038). By contrast, the risk of developing AML was not significantly related to the mutant allele burden. In conclusion, the findings of this study suggest that a high mutant allele burden represents a risk factor for progression to myelofibrosis in patients with PV.