Clonal Dynamics of JAK2V617F and Non-Driver Mutations in Polycythemia Vera and Essential Thrombocythemia Patients Receiving Hydroxyurea Therapy

Introduction:

Hydroxyurea (HU) is the most widely used cytoreductive treatment for patients with essential thrombocythemia (ET) and polycythemia vera (PV) at high risk of thrombosis. It remains unknown whether long-term HU therapy modulates or promotes the acquisition of mutations in non-driver (ND) genes, especially, when assessing hematological (HR) and molecular (MR) response. The objective of the study was to analyze the clonal dynamics of ND genes in HR and MR with HU in a cohort of JAK2V617F-mutated PV and ET patients.

Methods:

The study included 144 JAK2V617F positive patients (PV n = 73, TE n = 71) receiving HU as first-line cytoreductive treatment. The baseline sample (before HU treatment) and at the timepoint of best molecular response to JAK2V617F were analyzed. The allelic burden of J AK2V617F was assessed by allele-specific PCR and the mutational profile of ND genes was analyzed by next generation sequencing with a custom panel including 27 myeloid-associated genes. HR was defined according to the criteria of the European LeukemiaNet 2009 and MR of JAK2V617F was defined as complete, major, partial and no response (Table I).

Results:

Median molecular follow-up was 54.1 months for PV and 55.5 months for ET. Patients with PV were more likely to be males (p<0.001), and displayed higher leukocyte count (p<0.001) compared to those with ET. The respective numbers of deaths, leukemic transformations and fibrotic progressions were: 22 (30%), 4 (5%), 6 (8%) for PV cases, and 19 (27%), 1 (1%), 0 (0%) for ET patients.

At baseline, a total of 62 somatic mutations in ND genes were detected in 42/73 (57%) PV patients while 58 were detected in 36/71 (51%) ET patients.

Complete HR was observed in 102 patients: 44 (60%) PV and 58 (81%) ET. Partial MR in 67 cases: 35 (48%) PV and 32 (45%) ET and major or complete MR in 21 cases: 8 (11%) PV and 13 (18%) ET. The median duration of HU treatment was 45.8 months (range: 17.5-189.5) for PV and 45.6 months (range: 14.6-168.6) for ET.

The most frequently mutated genes detected at pre-therapy samples were TET2 (34%), ASXL1 (12%), SF3B1 (7%) and EZH2 (5%) in PV patients, and TET2 (34%), ASXL1 (13%), DNMT3A (13 %) and SRSF2 (5%) in ET patients. No significant differences were observed in the MR (p=0.358) or HR (p=0.917) according to the presence or absence of mutations in ND genes at baseline.

Clonal dynamics of DNMT3A, ASXL1, and TET2 (DAT) genes were not modulated by HU therapy to the same extent as JAK2V617F. Disappearance and emergence of additional mutations in DAT genes were observed independently of the molecular response achieved by the JAK2V617F clone. These findings suggest the existence of clones with mutations in ND genes independent from the pathogenic driver clone, and the lack of modulation by HU treatment.

Finally, an increase of allelic burden or the appearance of mutations in TP53, a gene related to progression, and in other DNA repair genes (PPM1D and CHEK2) was observed in 14 (19.1%) PV patients and 9 (12.6%) ET cases during HU treatment. However, no increased risk of myelofibrotic transformation or progression to acute myeloid leukemia was observed in these patients.

Conclusions: 

1. Pre-treatment ND mutations are not associated with HR and MR to HU in JAK2V617F-mutated patients. 2. The clonal dynamics of ND mutations (decrease, increase, appearance, disappearance) are not related to the evolutionary dynamics of JAK2V617F. 3. An increase or appearance of progression-related mutations in TP53 and/or other genes of the DNA repair pathway such as CHEK2 and PPM1D is observed during HU treatment.

Acknowledgments: Instituto de Salud Carlos III-FEDER, PI16/0153, PI19/0005, 2017SGR205, PT20/00023 and XBTC.

Figure 1

View largeDownload slide

Figure 1

View largeDownload slide

Close modal