Limited Efficacy of Hydroxyurea in Lowering of the JAK2 V617F Allele Burden.

Introduction: The JAK2 V617F mutation has become a key molecular marker in the diagnostic setting in Philadelphia chromosome negative chronic myeloproliferative (CMPDs) disorders, virtually all patients with polycythemia vera (PV) and approximately half of those with essential thrombocythemia (ET) and primary myelofibrosis (PMF) being JAK2 V617F positive. Apart from being a clonal marker of CMPD, a quantification of the JAK2 V617F allele burden may allow a stratification of phenotype in regard to degree of myeloproliferation and - although still controversial - risk of thrombosis and transformation to a myelofibrotic disease state. The treatment of choice in elderly patients is hydroxyurea (HU). Preliminary data on the efficacy of HU in lowering the JAK2 V617F allele burden have been conflicting (1–3). In the present study we show that long-term treatment with HU for a median of two years does not induce significant reductions in the JAK2 V617F allele burden in patients with CMPDs.

Methods: 24 patients (newly diagnosed: PV=9, ET=3, PMF=2; non-newly diagnosed: PV=9, ET=1) were included. Quantitative PCR was performed on un-fractionized peripheral blood leukocytes every 3-6 months. Comparison of changes in median JAK2 V617F allele burdens was performed by linear regression analysis, calculating the slope of the regression line between medians at the various time points.

Results: A median of 4.7 samples (range 3–7) were collected from each patient. Median follow-up was 24.2 months (range 11–36). Seventeen patients were treated with HU and four patients with anagrelide (ANA). Three patients not receiving cytoreductive therapy were also included. The median JAK2 V617F allele burden at baseline in HU treated patients including both newly diagnosed patients and pre-baseline HU treated patients was 33% (95% c.i.: 11–36%) and 24 % (95% c.i.: 13–48%) at the end of follow-up (median 24.2 months, range 11–36). Although indicating a small decline in JAK2 V617F allele burden, the slope of the regression line between medians were 6.7 (p=0.32 (95% c.i.: −7.1–20.7)) and accordingly not significant. Looking solely on newly diagnosed patients no significant reduction was recorded in median JAK2 V617F allele burden at baseline 34% (9–47%) as compared to 21% (6–55%) at end of study, (slope -0.31 (p=0.96) (95%c.i.: −14.6–13.9))). Some patients on HU had an initial decline in JAK2 V617F allele burden during the first 3–6 months of therapy (pt. no. 1, 4, 8, 18, 19 and 24). However, the overall change in JAK2 V617F allele burden during the first 3–6 months of therapy in newly diagnosed patients was not significant with a slope of the regression line between medians (baseline and 3–6 months) of 1.7, (p= 0.75 (95%, c.i.: −13.3– 9.8)). In patients treated with anagrelide there was a trend towards increasing JAK2 V617F allele burden (p=0.16), whereas in patients followed without myelosuppresive therapy the JAK2 V617F allele burden significantly increased during follow up (p=0.04).

Conclusions: The present study of the dynamics of the JAK2 V617F allele burden during cytoreductive therapy with HU has clearly shown that HU has no or only modest effect on the JAK2 V617F allele burden. HU may have a capacity to prohibit clonal expansion to some extend. However, HU is cytostatic and only active against cycling precursors and has no selective clonal advantages. Our results are in accordance with the clinical experience of rapid increase in leukocyte and platelet counts when HU is discontinued contrasting the long lasting effects of immune therapy with alpha-interferon, which in a subgroup of patients has been shown to induce major and sustained molecular responses even after cessation of long-term treatment. Monitoring of molecular responses during conventional and novel targeted therapies is an important objective to be implemented in future clinical trials.