Patients diagnosed with polycythemia vera (PV) or essential thrombocythemia (ET), a subtype of myeloproliferative neoplasms (MPN), sometimes suffer disease transformation into myelofibrosis (MF) associated with poorer prognosis. Thus, predicting which patients have a risk of MF transformation is an important task. Following the identification of a driver mutation JAK2V617F in a majority of MPN patients, several studies was performed to investigate the potential of JAK2V617F allele burden as a diagnostic marker for MF-transformation. However, the results differ between cohorts presumably due to a lack of accurate JAK2V617F allele burden measurement. Since we have previously developed alternative-binding probe competitive polymerase chain reaction (ABC-PCR) that accurately determines the JAK2V617F allele burden, we assessed the predictive value of JAK2V617F allele burden in MF-transformation in Japanese MPN cohort.

In a retrospective study, we compared JAK2V617F allele burdens between formalin-fixed paraffin embedded-bone marrow (FFPE-BM) from initial diagnosis and peripheral blood (PB) from follow-up visits. We first examined whether the allele burdens in FFPE-BM and PB were comparable when they are collected at the same time. Determining the allele burdens in a set of FFPE-BM and PB taken from same patient within a 3-month period, we observed that allele burdens from these specimens are significantly correlated (n=26, R²=0.97), which is consistent with previous report with a larger cohort (Blood 122; 3784-6). Thus, in subsequent analyses, we set a base line of the mutant allele burden determined from FFPE-BM, which is then compared with allele burdens from PB during the disease duration.

We examined 14 PV and 20 ET patients (mean disease duration 69.2 months) defined by WHO 2008 MPN criteria. From first diagnosis to the time when MF-transformation was first recognized, JAK2V617F allele burden was significantly increased (mean increase 19.5±17.3%, p=0.044) in patients with MF-transformation (n=11). While patients with no MF-transformation (n=23) presented limited changes (mean increase 3.9±16.1%) over a similar duration period. When subclassifying patients into three groups based on the change or the base line value of JAK2V617F allele burden, MF-transformation was more frequently (p=0.034) observed in patients whose JAK2V617F allele burden was either increased by more than 10% during the follow-up (group A) or higher at first diagnosis than the mean values for each disease (PV; 71.7%, ET; 35.5%) (group B). MF-transformation was 2 out of 4 (50%) in the group A, and 9 out of 22 (41%) in the group B. In contrast, MF transformation in the rest of the patients (group C) was 0 out of 8 (0%). Hydoxyurea-treated (n=16, 6 PV and 10 ET) and –untreated (n=18, 8 PV and 10 ET) patients do not show significant difference in frequencies of MF-transformation, confirming that Hydoxyurea has no preventative effect against MF-transformation.

In conclusion, our study showed that higher JAK2V617F allele burden at first diagnosis or a dynamic increase in allele burden during the follow-up period is a predictive factor for MF-transformation. Thus, a routine measurement of the JAK2V617F allele burden by an accurate assay system is recommended to predict MF-transformation.

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution