Changes of JAK2 V617F Allele Burden Over Time in Patients with Polycythemia Vera or Essential Thrombocythemia: A Retrospective Study of 172 Patients.

Abstract 1892

Poster Board I-915

There are conflicting results about quantitative modifications of V617F allele burden in patients (pts) with myeloproliferative neoplasms (MPN) who are either therapy-naive or are treated with hydroxyurea (HU). In a retrospective single center study in 48 pts with polycythemia vera (PV) or essential thrombocythemia (ET) the granulocyte JAK2V617F allele burden remained stable over time (median follow-up was 34 months for PV and 23 for ET) irrespective of the pts being treated or not with cytotoxic therapy (Theocharides A et al, Haematologica 2008). Conversely, another study in 25 patients reported a significant reduction of V617F allele burden (>30% of baseline level) after HU therapy in 52% of the pts, becoming indetectable in 3 of them (Girodon F et al, Haematologica 2008). The aim of this study was to evaluate any modifications of JAK2V617F allele burden during long-term follow-up in patients with PV or ET and the effects of HU treatment. This two-center (Firenze and Bergamo) retrospective study concerned 172 patients with a diagnosis of PV or ET according to the WHO criteria. The only study inclusion criteria were the presence of JAK2V617F mutation and the availability of at least two sequential blood samples drawn at an interval time of at least 6 months. The JAK2-V617F allele load was measured by sensitive quantitative RT-PCR in granulocyte DNA according to the method of Lippert et al (Blood 2006). Differences between median values of JAK2 V617F allele burden were tested by the Wilcoxon matched-pairs signed-ranks test. Repeated measure test for JAK2 V617F mean change over time, irrespective of diagnosis, was also calculated to investigate a significant variance among ordered time measures. There were 103 pts with PV and 69 with ET; median age was 56 yr (range, 15–84), females were 49%. The median interval time between the baseline and follow-up sample in the whole pt population was 27 months (range 6–60), 26 and 28 months for ET and PV, respectively. The median patient follow-up was 3 years (range 0.5–25); no evolution to myelofibrosis or acute leukemia was recorded. According to previous reports, the mean V617F burden was significantly greater in PV than in ET pts (50 ± 26% and 32 ± 18%, respectively; p <0.0001). Sixty-nine pts (41 PV, 28 ET) remained untreated during follow-up (Group 1), whereas 103 received HU; of the latter, 60 pts were already on treatment at the time of first genotyping (Group 2) whereas 43 patients were chemotherapy-naive and started HU after the first blood sampling (Group 3). In Group 1 pts, the median interval between 1^st and 2^nd sample was 26 months (range 6–60)for PV pts and 24 mo (range 6–59) for ET pts. The JAK2 V617F allele burden was 47.7±22.3% and 48.3±18.9% in the 1^st and 2^nd sample in PV pts, respectively, and 25.2±13.1% and 28.6±13.5% in case of ET pts, without any significant difference. In pts included in Group 2, the median interval between 1^st and 2^nd sample was 28 months (range 6–54) for PV pts and 28 mo (range 6–60) for ET pts. The JAK2 V617F allele burden was 55.1±29.2% and 60.3±25.6% in the 1^st and 2^nd sample in PV pts, respectively, and 36.3±21.7% and 40.7±22.9% in case of ET pts (P=.039). Considering PV and ET pts together, there was a statistically significant increase of V617F allele burden over time from 47.9±28.0% in the 1^st sample to 52.8±26.3% in the 2^nd (P= .023, repeated measure test). Among pts of Group 3, the median interval time between 1^st and 2^nd sample was 32 months (range 10–48) for PV pts and 24 mo (range 7–58) for ET pts. The JAK2 V617F allele burden was 47.0±26.2% and 45.5±20.8% in the 1^st and 2^nd sample in PV pts, respectively, and 37.3±16.2% and 33.3±15.0% in case of ET pts (P= .024). In conclusion, within the observation period of this study we found no evidence for a time-dependent increase of V617F allele burden in untreated PV and ET pts, although we cannot exclude that these results might be related to the relatively short follow-up. Accordingly, in a previous study in ET pts we found that the mutant allele burden increased significantly after 10 years from diagnosis (Carobbio A et al., Exp Hematol, 2009). Furthermore, we observed a very modest reduction of V617F allele burden in HU newly treated ET patients, while in previously treated PV and ET patients the allele burden actually increased over time, suggesting that HU has very little if any likelihood to impact on the size of mutant hematopoietic cell clone.