New Predictive Model Developed for Leukemic Transformation in Primary Myelofibrosis

Published on: 

A recent study sought to develop a robust predictive model for patients whose disease will transform into acute myeloid leukemia, a complication that is associated with a poor prognosis.

Primary myelofibrosis (PMF) can be treated with allogenic stem cell transplant (which can cure the disease or prolong survival) or with drug therapy, splenectomy, and radiation. Determining the appropriate time for a transplant is an important objective in tailoring treatment to the individual patient, and currently, the mutation- and karyotype-enhanced prognostic scoring system (MIPSS70 + version 2.0), a risk model for overall survival, is one tool used to do so. A recent study sought to develop a complementary, robust predictive model for patients whose disease will transform into acute myeloid leukemia, a complication that is associated with a poor prognosis.

The study included 1306 patients with PMF who were treated at the Mayo Clinic between 1976 and 2017. Patients had a median age of 65 years (range, 19-92 years), and 63% were male. Median followup was 3.2 years (range, 0-31 years), and during this period, 149 (11%) cases of leukemic transformation (LT) were documented.

The investigators assessed the clinical and laboratory features of the patients who developed LT versus those who remained in chronic-phase disease at followup. They found that patients who experience LT were more likely to be male, to have higher incidence of excess circulating blasts, and to have ASXL1, SRSF2, and IDH1 mutations.


Multivariable logistic regression confirmed findings from a univariate analysis that there are independent prognostic contributions of the IDH1 mutation, the very high-risk karyotype, the ASXL1 mutation, age over 70 years, presence of the SRSF2 mutation, male sex, circulating blasts of 3% or greater, presence of anemia, and constitutional symptoms.

A parallel time-to-event Cox analysis confirmed inferior leukemia-free survival in patients who had mutations of IDH1, SRSF2, or ASXL1, in those who had circulating blasts of 3% or greater, age over 70 years, and anemia.

The investigators then developed a predictive model for LT, with point allocations that corresponded with hazard ratios. A total of 456 patients were informative for the 6 factors of LT, and a 3-tiered risk stratification was developed: high risk, intermediate risk, and low risk. In its predictive accuracy, write the investigators, the new model is superior to MIPSS70 + version 2.0, as well as superior to the genetically-inspired prognostic scoring system, or GIPSS.

Despite the promise of the model, wrote the authors, “Our observations require further validation, which might not be easy to accomplish, considering the difficulty in securing adequate number of informative cases.”


Vallapureddy RR, Mudireddy M, Penna D, et al. Leukemic transformation among 1306 patients with primary myelofibrosis: risk factors and development of a predictive model. Blood Cancer J. 2019;9(2):12. doi: 10.1038/s41408-019-0175-y.