Do Patients with Post-Essential Thrombocythemia and Post-Polycythemia Vera Differ from Patients with Primary Myelofibrosis?

Introduction: Clinical characteristics of post-essential thrombocythemia/polycythemia vera myelofibrosis "post ET/PV-MF" are not well defined as for primary myelofibrosis "MF".

Objective: We aimed to identify morphological, clinical and prognostic characteristics of patients with post ET/PV-MF seen at our center.

Methods: Retrospective chart review of 1120 patients with MF - 766 primary MF, 354 post ET/PV-MF, who were referred to our institution between years 1984-2013 was performed; 92% presented after the year of 2000. Fisher's exact and Mann-Whitney test were used for categorical and continuous variables; Cox proportional hazard model and Kaplan-Meier curves with log rank test for correlation between the variables and survival. Survival analysis were calculated after censoring patients for stem cell transplantation "SCT" (n=92).

Results: Overall median follow up was 36 months (0-411), 51% (n=573) patients had died. Progression to AML after median time of 32 months was not different between groups and occurred in 9.5% (n=106) patients over observation period of 5180 persons-years. Incidence rate was 3.4 cases per 100 persons-year. Causes of death were known in 55% of patients, and included progression of MF in 38% (n=122), infection in 27% (n=86), and other reasons with less than 10% occurrence (complications post SCT; secondary malignancy; other medical conditions). Demographics and disease characteristics are depicted in a table. 92% of patients were evaluable for karyotype; abnormalities were detected in 39% (n=404), of which 53% were unfavorable with monosomal (23%), complex (43%) and trisomy 8 (18%) being the most common. Molecularly high risk mutations "MHR" in genes ASXL1, EZH2, and IDH1/2 were positive in 35% of tested patients (n=161) regardless of presence of driver mutation. IPSS and DIPSS plus scores were similar between MF and post ET/PV-MF (with IPSS low in 8.7%, intermediate 1 in 19%, intermediate 2 in 28%, and high in 44%). By multivariate analysis, higher risk categories of IPSS and DIPSSplus predicted shorter overall survival "OS" for both cohorts (by DIPSSplus - high risk: HR 2.7, 95% CI 2.1-3.5; int-2: HR 1.6; 95% CI 1.3-2.0). Median OS stratified by IPSS was 160 (HR 1.8, 95%CI 1.0-3.0), 116 (HR 1.5, 95%CI 1.0 -1.5), 78 (HR 1.4, 95%CI 1.2-1.7) and 54 months, p=0.001. Univariate analysis identified age over 65, anemia, trombocytopenia, leukopenia, increased blasts, splenomegaly, constitutional symptoms, unfavorable karyotype, JAK2 mutation and triple negativity as predictors for inferior survival. Age over 65; hemoglobin below 10, platelets below 100 and peripheral blasts ≥1% showed significance for predicting OS by multivariate analysis. When stratified according to diagnosis, higher age, anemia, thrombocytopenia, high blasts, JAK2 positivity and triple negativity retained prognostic significance for MF whereas anemia, thrombocytopenia, and JAK2 mutation for post ET/PV-MF. Median OS was 73 months (range, 0.1-210) without difference between MF and post ET/PV-MF.

Conclusion: Post ET/PV-MF does not appear to have substantial different clinical characteristics than primary MF.