Identification of Predictive Biomarkers Based on Cytokine Profiles for Molecular Relapse After Treatment-Free Remission in Chronic Myeloid Leukemia Patients.

Tyrosine kinase inhibitors (TKIs) have transformed chronic myeloid leukemia (CML) into a manageable disease, and discontinuation has become a feasible goal for many patients. However, molecular relapse after TKI cessation remains a challenge. This study investigated factors influencing molecular relapse-free survival (MRFS) in a real-world cohort of CML patients undergoing TKI discontinuation, focusing on clinical, molecular, and immune-related variables. Traditional prognostic tools, such as the Sokal score, showed limited capacity to predict relapse risk in this context. Instead, total treatment duration and depth of molecular response emerged as critical predictors, with longer therapy and deeper remission associated with improved outcomes. A key novel finding was the prognostic relevance of cytokine profiles, particularly interleukin-6 (IL-6) and monocyte chemoattractant protein-1 (MCP-1). These cytokines were strong predictors of relapse in a decision tree model that achieved high specificity and positive predictive value in an independent validation cohort. Notably, lower IL-6 and MCP-1 levels were associated with increased relapse risk, suggesting reduced immune surveillance may contribute to recurrence. Integrating cytokine signatures with molecular markers improved prognostic accuracy, supporting immune biomarkers as complements to established clinical parameters. These findings underscore the complexity of relapse mechanisms and the need for individualized risk assessment when considering TKI discontinuation.