A PROGNOSTIC MODEL to PREDICT SURVIVAL In WHO-DEFINED ESSENTIAL THROMBOCYTHEMIA: A STUDY by the IWG-MRT (International Working Group for Myeloproliferative Neoplasms Research and Treatment)

Abstract 1746

Diagnosis of essential thrombocythemia (ET) has been updated in the last World Health Organization (WHO) classification, that provides a guideline to distinguish ET from early/prefibrotic primary myelofibrosis. These two conditions differ in term of disease evolution and survival (Barbui et al, J. Clin. Oncol. 2011). To design a prognostic model for survival in WHO-defined ET (International Prognostic Score for Essential Thrombocythemia, IPSET), we studied 891 patients with strict WHO diagnostic criteria. Demographics have been presented elsewhere (Barbui et al, J. Clin. Oncol. 2011). We tested the prognostic impact on survival of age, prior thrombosis, leukocyte count, platelet count, hemoglobin level, JAK mutational status, splenomegaly. Harrell's C concordance statistics were used to set the best cutoff levels of continuous variables in order to predict survival. Univariate Cox proportional hazard regression showed that age over 60 years (P<.0001), leukocyte count greater than 11 x10⁹/L (P<.0001), anemia (P=.04), platelet count greater than 1000 x10⁹/L (P<.004) and prior history of thrombosis (P< 0001) were significant risk factors for survival. Age over 60 years (hazard ratio-HR- 6.5, 95% CI: 3.9–10.7; P<.0001), leukocyte count greater than 11 × 10⁹/L (HR: 3.1, 95% CI: 2–4.7; P<.0001) and prior history of thrombosis (HR: 2.9, 95% CI: 1.9–4.5; P< 0001) retained statistical significance on survival by multivariable analysis and were included in the new prognostic model. We assigned each factor an integer weight close to the corresponding HR in the multivariable Cox regression: weight 2 for age over 60 years; weight 1 for leukocyte count greater than 11 x10⁹/L and for prior thrombosis. Finally, we defined the IPSET score by allocating patients into three risk categories with significantly different survival according to the sum of weights: low risk (sum of weights=0; n= 342; median survival not reached), intermediate risk (sum=1-2; N=374; median survival 24.5 years), and high risk (sum=3-4, n=158; median survival 13.8 years). Fisher exact test did not reveal a difference in the distribution of the causes of death among IPSET categories. We validated the model in two independent cohorts: one from Germany (132 WHO-defined ET) and another from France (234 ET patients not strictly diagnosed according to WHO). Log rank test and Cox regression showed significant differences in term of survival among IPSET categories in both series. Finally, we designed an age-specific IPSET model for patients younger than 60 years (IPSET-y). Using the same methodology, we found that age over 45 years (weight 2), leukocyte count greater than 11 × 10⁹/L (weight 1) and prior history of thrombosis (weight 1) were independent risk factors for survival. IPSET-y defined three risk categories (LR, sum=0; IR, sum=1-2; HR, sum=3–4) with a median survival not reached in LR and IR, and of 22.3 years in HR. In conclusion, IPSET, based on age over 60 years, leukocyte count greater than 11 x10⁹/L and history of thrombosis allows prognostic assessment of WHO-defined ET. The validation process makes IPSET applicable in all patients phenotypically appearing as ET. IPSET models may be useful for treatment decision-making.