Development of a prognostic scoring system for chronic myeloid leukemia in blast phase

Lauseker, Michael ; Sacha, Tomasz ; Klamova, Hana ; Yakoleva, Julia; Lomaia, Elza ; Milojkovic, Dragana; Fabisch, Christian; Nicolini, Franck E.; Olsson-Stromberg, Ulla; Zackova, Daniela; Lewandowski, Krzysztof; Ernst, Philipp ; Castagnetti, Fausto ; Hirschbühl, Klaus ; Zupan, Irena; Voskanyan, Astghik; Gołos, Aleksandra; Frietsch, Jochen J. ; Gil, Justyna; Faber, Edgar ; Krause, Stefan W. ; Parry, Anne; Saussele, Susanne ; Heibl, Sonja; Franke, Georg-Nikolaus ; Casado, Luis Felipe; Paczkowska, Edyta ; Hebart, Holger ; Eder, Matthias; Anhut, Peter; Trautmann-Grill, Karolin ; Szczepanek, Elzbieta; Südhoff, Thomas; Patkowska, Elzbieta; Kiani, Alexander; Ernst, Thomas ; Apperley, Jane ; Mayer, Jiri; Hochhaus, Andreas ; Pfirrmann, Markus und Brioli, Annamaria (2026): Development of a prognostic scoring system for chronic myeloid leukemia in blast phase. In: Leukemia, Vol. 40: pp. 751-758

Abstract

Currently there is no established prognostic scoring system for patients with chronic myeloid leukemia (CML) in blast phase (BP). This study aimed to identify prognostic factors of CML-BP in a large cohort of prospectively and retrospectively collected patients and to develop a readily available prognostic scoring system at the onset of BP to enable future comparison between different trials and series. The analyses were based on 275 patients from thirteen countries, collected within the European LeukemiaNet Blast Phase Registry with a median observation time of 45 months and a median OS of 18.9 months. A Cox proportional hazards model for overall survival (OS) was employed, missing values were imputed. The study identified six independent prognostic factors: blast percentage, platelet count, age (all at onset of CML-BP), immunophenotype of BP, extramedullary disease, and previous history of CML. The low-risk group, encompassing 14% of patients, had a median OS of 97 months, the intermediate-risk group (59% of patients) of 22 months, and the high-risk group (27% of patients) of 9 months. Cross-validation demonstrated a good performance of the score, although external validation is strongly recommended. Despite the inherent limitations of registry data, the findings offer robust insights into prognostic factors for CML-BP.

• Item Type: Journal article
• Faculties: Medicine > Institute for Medical Information Processing, Biometry and Epidemiology
• Subjects: 600 Technology > 610 Medicine and health
• ISSN: 0887-6924
• Language: English