Essential Thrombocythemia (ET) is a common myeloproliferative neoplasm (MPN) that usually has an indolent disease course. Over the long term, ET can progress into myelofibrosis, acute myeloid leukemia (AML), or myelodysplastic syndrome (MDS). Incidence rate of transformation to MDS or AML is 1-5% with median time to transformation reported between 84 and 138 months. Although transformation of ET to MDS or AML has been reported previously, there is little information on prognosis in these patients and if specific ET related therapy may lead to increased risk of transformation to MDS or AML. Our study aims to investigate a cohort of ET patients who transformed to MDS or AML in order to better understand their clinical course and identify prognostic factors that may impact overall survival.


After obtaining institutional review board approval, cases of ET between 1990 and 2014 with pathological confirmed transformation to MDS or AML were identified through our pathology database. Patient demographic data, pathological data, and therapy information were recorded. Pathology slides were reviewed by two hematopathologists in order to confirm the diagnosis. Kaplan Meier estimates were used for overall survival assessment and Cox regression modeling for univariate analysis.


A total of 40 patients were identified who met the study criteria. Nineteen (48%) were male, and the median age at time off ET diagnosis was 59 years (19-72 years). Sixteen patients (40%) transformed to MDS, 13 (33%) to AML, and 11 (27%) to MDS followed by AML. Median time from ET diagnosis to transformation was 76 months (26-481). Overall, with time to transformation to MDS being 69 months (69-223) and to AML being 88 months (39-481), respectively. After a median follow up of 15 years (10.1-19.9 years), the median OS from ET diagnosis was 10 years (7.4-12.6 years). The median OS from time of transformation was 14 months (5-22.9 months). Patients who had MDS showed longer OS from time of transformation than those who had AML, 35 months (14.2-55.7) versus 10 months (4.4-15.6). Jak2 V617F-negative patients (12 pts) had a better OS from time of diagnosis of ET than Jak2 V617F-positive positive patients (14 pts) (20 yrs vs 8 yrs P = 0.05) but the OS did not differ from time of transformation (P = 0.5). Nine of 40 pts (33%) had normal cytogenetics and 31/40 pts (77%) had abnormal cytogenetics at the time of transformation to MDS or AML. More than half of them had complex cytogenetic 18/31 (58%). At the time of transformation, complex versus non-complex cytogenetic abnormalities did not have an impact on patients’ survival from disease evolution (P = 0.67). The use of hydoxyurea or anagrelide did not have an impact on transformation (P = 0.96, P = 0.58), nor did either impact OS from transformation (P = 0.54, and P = 0.67, respectively). Patients with a grade 2-3 fibrosis identified in the bone marrow at time of transformation had worse OS from transformation than those with grade 0-1 fibrosis (P = 0.05). A higher white blood cell count, but not platelet count, hemoglobin (Hgb) or blast count, at transformation to AML was associated with worse overall survival (P=0.04). Advanced age was also associated with worse OS at transformation of all patients (P = 0.01).


Little is known about prognostic factors in ET transformed to MDS or AML. In our forty patients there was no difference in OS from transformation based on having received hydroxyurea or anagrelide therapy for ET. Patients with moderate to severe myelofibrosis (grade 2 - 3 fibrosis) or a higher WBC count at time of transformation had worse overall survival from transformation likely indicating a more aggressive MDS or AML disease biology. Interestingly, OS from diagnosis of ET was 8 years versus 20 years in patients with JAK2 V617F-positive versus JAK2 V617F-negative ET indicating different disease biology in Jak2 negative ET patients. Collaborative multi institutional efforts are needed to further assess prognostic factors in this rare transformation and establish a genetic link especially in JAK 2 negative patients.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.

Sign in via your Institution