Dynamic Prognostic Model to Identify Patients with Primary or Post-Polycythemia Vera/Essential Thrombocythemia Myelofibrosis (Mf) At the Highest Risk of Death and Transformation to Acute Myeloid Leukemia

While patients (pts) with primary or post-polycythemia vera (PV)/post-essential thrombocythemia (ET) myelofibrosis (MF) have a median overall survival of 5–6 years, the survival of particular subsets of pts with MF is highly variable. Pts transforming to acute myeloid leukemia (AML) represent a large proportion of those with the poorest prognosis. Prognostic tools capable of identifying pts at the highest risk of death and transformation may provide an opportunity for tailored early therapeutic intervention (e.g. allogeneic stem cell transplantation [allo-SCT]).

To identify risk factors that predict for short survival and for high risk to transformation to AML in a large cohort of pts with MF diagnosed at MD Anderson Cancer Center.

649 pts with MF referred from Feb-1966 to Jun-2010 were assessable for survival analysis, including 177 (27%) with post-PV or post-ET MF. The median follow-up was 19 months (range, 1–180). We first evaluated an extensive list of baseline risk factors for overall survival (OS) by univariate analysis (UVA): age, sex, prior exposure to alkylating agents, hydroxyurea, immunomodulatory drugs (i.e. thalidomide derivatives), and others, MF duration, performance status, splenomegaly, hepatomegaly, hemoglobin, white blood cell count (WBC), platelets, monocytes, peripheral blood (PB) or bone marrow (BM) blast percentage (continuous variable), chromosomal abnormalities, JAK2^V617F mutation (yes vs no, and as a continuous variable), total bilirubin, LDH, creatinine level. Multivariate analysis (MVA) identified age ≥60 (p=0.004;HR=1.63), baseline platelets <100×10⁹/L (p<0.001;HR=1.62), BM blast >10% (p=0.002;HR=2.18), worst karyotype (del17p, −5, −7, complex; p<0.001;HR=2.44), transfusion dependency (≥2 red cell transfusions; p<0.001;HR=2.64), PS ≥1 (p=0.003;HR=1.47), LDH >2000 U/L (p<0.001;HR=1.62), prior hydroxyurea (p<0.001;HR=1.69), and male gender (p=0.005;HR=1.41) as independent poor prognostic factors for OS. Using the corresponding hazard ratios, a dynamic risk model for survival in MF was derived based on weighted scores of 0–1, 2–3, 4–5, ≥6. The number of pts and median OS for the 4 risk groups were: Low: 84 (13%), median OS not reached; Intermediate-1: 191 (29%), 74 months; Intermediate-2: 208 (32%), 29 months; High :166 (26%), 11 months. We then examined the same baseline variables used in the OS analysis to identify independent risk factors for AML transformation. MVA identified WBC<3×10⁹/L (p=0.02), PB blasts ≥5% (p=0.01), BM blasts ≥5% (p=0.02), and unfavorable karyotype (p=0.04). The presence at baseline of BM blasts ≥10% and worst karyotype were identified as independent poor prognostic factors for both OS and AML-transformation. Pts with one or both of these two risk factors (n=80, 12% of the population) defined a subgroup of pts with a median OS of only 10 months. This subgroup was considered to have MF at significantly higher risk (compared to pts without any of these 2 adverse factors) for transformation to AML (1-year transformation-free survival 82% vs 98%, p<0.001). Ten of 80 (13%) of pts with any of these two factors progressed to AML within 12 months whereas only 10 of 568 (2%) pts without those two factors progressed to AML within 12 months. The yearly rates of transformation for the whole study group during the first, second, and third year of follow-up were 3%, 2%, and 3%, respectively.

Analysis of a large population of pts with either primary or post-PV or post-ET MF identified risk factors that at any time during the course of MF predict for short OS and for high risk for AML transformation AML. Pts with such factors may be candidates for more aggressive therapeutic approaches such as allo-SCT.

No relevant conflicts of interest to declare.