ELTS Score and Early "Warning" Define a Two-Stage Risk Framework for Failure-Free Survival in Chronic-Phase CML

Area of Science:

• Hematology
• Oncology
• Pharmacology

Background:

• Risk stratification for treatment failure in chronic-phase chronic myeloid leukemia (CP-CML) is challenging in the tyrosine kinase inhibitor (TKI) era.
• Existing models may not fully capture dynamic disease progression under modern therapies, particularly second-generation TKIs (2G-TKIs).

Purpose of the Study:

• To identify predictors of failure-free survival (FFS) in newly diagnosed CP-CML patients treated with TKIs.
• To evaluate the utility of the EUTOS long-term survival (ELTS) score and the European LeukemiaNet (ELN) 2025 'Warning' category in a dynamic risk framework.

Main Methods:

• Pooled analysis of 710 newly diagnosed CP-CML patients from the JALSG CML212 and New TARGET studies.
• Patients primarily received first-line 2G-TKIs (approx. 80%).
• Incorporated ELN 2025 'Warning' and 'Unfavorable' categories to define FFS events and analyzed transitions.

Main Results:

• The ELTS score consistently stratified FFS, outperforming the Sokal score in time-dependent discrimination for patients on 2G-TKIs.
• Baseline ELTS predicted initial treatment failure, but lost predictive power once a 'Warning' state was reached.
• Attaining a 'Warning' state within 6 months of TKI initiation independently predicted progression to 'Unfavorable' (HR 7.31).

Conclusions:

• A two-stage risk stratification framework is proposed, integrating baseline ELTS with the timing of 'Warning' events.
• This dynamic approach can guide risk-adapted management strategies for CP-CML patients.
• Early identification of 'Warning' states is critical for predicting subsequent treatment failure.