MIPSS70: Mutation-Enhanced International Prognostic Score System for Transplantation-Age Patients With Primary Myelofibrosis.

Purpose To develop a prognostic system for transplantation-age patients with primary myelofibrosis (PMF) that integrates clinical, cytogenetic, and mutation data. Patients and Methods The study included 805 patients with PMF age ≤ 70 years recruited from multiple Italian centers and the Mayo Clinic (Rochester, MN), forming two independent learning and validation cohorts. A Cox multivariable model was used to select from among a list of 22 variables those that were predictive of overall survival (OS). Integrated clinical and genetic prognostic models with (MIPSS70-plus) or without (MIPSS70) cytogenetic information were developed. Results Multivariable analysis identified the following as significant risk factors for OS: hemoglobin 25 × 109/L, platelets < 100 × 109/L, circulating blasts ≥ 2%, bone marrow fibrosis grade ≥ 2, constitutional symptoms, absence of CALR type-1 mutation, presence of high-molecular risk mutation (ie, ASXL1, EZH2, SRSF2, IDH1/ 2), and presence of two or more high-molecular risk mutations. By assigning hazard ratio (HR)-weighted points to these variables, three risk categories were delineated for the MIPSS70 model; 5-year OS was 95% in low-risk, 70% in intermediate-risk, and 29% in high-risk categories, corresponding to median OS of 27.7 years (95% CI, 22 to 34 years), 7.1 years (95% CI, 6.2 to 8.1 years), and 2.3 years (95% CI, 1.9 to 2.7 years), respectively. In the MIPSS70-plus model, which included cytogenetic information, four risk categories were delineated, with 5-year OS of 91% in low-risk, 66% in intermediate-risk (HR, 3.2; 95% CI, 1.9 to 5.2), 42% in high-risk (HR, 6.4; 95% CI, 4.1 to 10.0), and 7% very high-risk categories (HR, 17.0; 95% CI, 9.8 to 29.2). Both models remained effective after inclusion of older patients in the analysis. Conclusion MIPSS70 and MIPSS70-plus provide complementary systems of risk stratification for transplantation-age patients with PMF and integrate prognostically relevant clinical, cytogenetic, and mutation data