Thrombotic and hemorrhagic complications in chronic myeloid leukemia in the era of tyrosine kinase inhibitors: pathophysiology, predictors, and risk management strategies

Background. The long-term use of tyrosine kinase inhibitors (TKIs) has converted chronic myeloid leukemia (CML) into a largely controllable chronic disease. This therapeutic success has shifted clinical priorities toward complications that may compromise survivorship, particularly the concurrent risks of thrombosis and bleeding arising from both disease biology and treatment-related, off-target effects. Objective. To systematize and critically appraise contemporary evidence (2021–2025) on the incidence, pathophysiological mechanisms, and predictors of thrombotic and hemorrhagic complications in CML, and to integrate this knowledge into practical, cardio-oncology–oriented risk-management strategies. Methods. A scoping systematic review was performed in accordance with PRISMA-ScR principles. PubMed/MEDLINE, Embase, Scopus, and the Cochrane Library were searched for studies published from January 1, 2021, through June 30, 2025. Randomized trials, cohort studies, systematic reviews/meta-analyses, and major clinical practice guidelines were prioritized. Observational evidence was appraised using the Newcastle–Ottawa Quality Assessment Scale; diagnostic/prognostic validation studies were evaluated using QUADAS-2 principles. Findings were synthesized narratively across disease-intrinsic and therapy-induced mechanisms, and across clinically actionable predictor domains. Results. The review delineates a “hemostatic double paradox” in CML: (1) disease-intrinsic extremes that can manifest as hypercoagulability and leukostasis in hyperleukocytosis, yet paradoxical mucocutaneous bleeding in extreme thrombocytosis mediated by acquired von Willebrand syndrome with loss of high-molecular-weight multimers; and (2) a therapy-driven dichotomy determined by TKI kinome profiles. SRC family kinase inhibition is mechanistically linked to qualitative platelet dysfunction and a bleeding phenotype, whereas vascular endothelial toxicity–amplified by metabolic dysregulation and hypertension–underpins a pro-atherothrombotic phenotype observed with selected TKIs. The analysis consolidates predictors spanning CML phase and cytologic burden, baseline cardiometabolic risk, and drug-specific off-target signatures, and highlights the clinical utility of structured cardio-oncology risk stratification and biomarker surveillance. A particular emphasis is placed on allosteric ABL1 inhibition (STAMP mechanism) as a strategy with a potentially vascular-sparing profile, and on emerging sex-related differences in thrombotic phenotypes that may warrant tailored monitoring. Conclusions. Thrombotic and hemorrhagic complications in CML are best conceptualized as a spectrum driven by the interaction between disease-related hematologic extremes and TKI-specific off-target effects. Optimal long-term care requires integrated, risk-stratified decision-making that balances antileukemic efficacy against individualized hemostatic and vascular vulnerability, supported by standardized monitoring and periprocedural management pathways within a cardio-oncology framework.