A comparative assessment of molecular-based prognostic models in CMML

Area of Science:

• Hematology
• Oncology
• Molecular Biology

Background:

• Chronic myelomonocytic leukemia (CMML) requires accurate prognostic systems for risk stratification and treatment guidance.
• Existing systems integrate clinical, cytogenetic, and molecular data, but novel molecular-based models show promise.
• The molecular International Prognostic Scoring System (IPSS-M) was developed for myelodysplastic syndromes (MDS) and its utility in CMML warrants investigation.

Purpose of the Study:

• To evaluate the applicability and prognostic accuracy of the IPSS-M in patients with CMML.
• To comprehensively compare the IPSS-M against other major molecular-based prognostic models in CMML.
• To assess the predictive validity of these models in patients receiving hypomethylating agent therapy.

Main Methods:

• Retrospective analysis of clinical and molecular data from 340 CMML patients.
• Application and comparison of IPSS-M, Revised International Prognostic Scoring System (IPSS-R), CMML-specific prognostic scoring system (CPSS)-Mol, CPSS, Mayo molecular, and Groupe Francophone des Myélodysplasies models.
• Assessment of overall survival (OS) and acute myeloid leukemia (AML) evolution.

Main Results:

• IPSS-M stratified CMML patients into 6 risk categories with significantly different median OS and 4-year AML evolution rates (P < .001 and P = .008, respectively).
• Both CPSS-Mol and IPSS-M demonstrated superior OS discrimination compared to older models.
• CPSS-Mol showed the highest accuracy for predicting leukemic evolution, while IPSS-M exhibited comparable prognostic accuracy to CPSS-Mol and was reliably applicable in CMML.

Conclusions:

• The IPSS-M is a valid and reliable tool for risk stratification in CMML, comparable in accuracy to CPSS-Mol.
• Molecular-based prognostic models, including IPSS-M, improve risk assessment in CMML.
• These models maintain predictive validity in patients treated with hypomethylating agents, supporting their use in clinical decision-making, especially in community settings.