Cardiovascular Profile of Tyrosine Kinase Inhibitors in Chronic Myelogenous Leukemia: A 16-Year Observational Study

Background: Chronic myelogenous leukemia (CML) is caused by a chromosomal translocation that gives rise to a chimeric BCR-ABL protein with constitutive tyrosine kinase (TK) activity. Tyrosine kinase inhibitors (TKIs) block this protein and have thus led to a paradigm shift in CML management in the early 2000s. Despite TKIs' efficacy in CML, their cardiotoxic side effects are a clinical challenge. This may result in TKI discontinuation and ultimately suboptimal response. Hence, we aim to explore the cardiotoxicity profile of TKIs, adjusting for pre-existing cardiovascular risk factors. Methods: We conducted a single-center retrospective cohort study to evaluate the cardiovascular complications of TKIs in chronic and accelerated phase CML adult patients (≥18 years) treated at Cleveland Clinic Foundation from 01/01/2007 to 12/31/2022. TKIs included were imatinib, dasatinib, nilotinib, bosutinib, ponatinib, and asciminib. Data collected included demographics, baseline comorbidities, cytogenetics, all treatment lines, and follow-up data. Overall survival (OS) and cardiovascular outcomes were collected in a time-to-event fashion. Major adverse cardiovascular events (MACE) included acute coronary syndrome (ACS), heart failure, cerebrovascular disease (CVA/TIA), and peripheral arterial disease (PAD) events. Additional events obtained included pulmonary effusion and pulmonary hypertension. Left ventricular ejection fraction (LVEF), global longitudinal strain (GLS), and pulmonary artery systolic pressure (PASP) were obtained from echocardiography reports at baseline and throughout the treatment course serially, when available. Multivariable Cox proportional hazards (M-CPH) regression models were employed for time-to-event outcomes, and multivariable linear mixed models were used for serial echocardiogram measurement estimation. Variables included in the models were age, gender, diabetes, hypertension, hyperlipidemia, and smoking status. Survival probabilities were estimated using the Kaplan-Meier method, with differences assessed by the log-rank test. Results: A total of 343 patients were included in the analysis, with a median age of 56 years (IQR: 43- 67); 197/343 (57%) were male and 293/343 (85%) were of White race. A total of 336 (98%) had chronic phase CML and 7 (2%) with accelerated CML. Baseline characteristics at diagnosis included 53/343 (15%) with diabetes, 112/343 (33%) with hypertension, 140/343 (41%) with hyperlipidemia, 157/343 (34%) were current or former smokers. The number of patients on asciminib was small (n=16) and they were not reported in the analysis. The median follow-up time was 63 months and the three-year OS (95% CI) probability by first-line treatment was 91% (87-96) for imatinib, 100% for bosutinib, 92% for dasatinib (86-98) and 96% (90-100) for nilotinib (log-rank P=0.3). The incidence rates of MACE per 100 patient-months (95%CI) were 1.18 (0.91, 1.51) for imatinib, 1.20 (0.84, 1.67) for nilotinib, 1.44 (0.77, 2.50) for ponatinib, 1.58 (1.16, 2.11) for dasatinib and 1.67 (0.97-2.68) for bosutinib. On M-CPH, new heart failure hazard ratio (HR) (95%CI) for bosutinib was 2.88 (1.52, 5.48), 1.21 (0.72, 2.04) for dasatinib, 0.86 (0.46, 1.62) for nilotinib, and 1.68 (0.59, 4.84) for ponatinib (reference: imatinib, P<0.01). On M-CPH, MACE HR (95%CI) were 2.54 (1.45-4.46) for bosutinib, 2.17 (1.52, 3.10) for dasatinib, 1.39 (0.94-2.06) for nilotinib, and 3.30 (1.72, 6.31) for ponatinib (reference: imatinib, P <0.01). On M-CPH; new pleural effusions HR (95%CI) was 1.87 (0.57, 6.10) with bosutinib, 2.68 (1.12, 6.41) with dasatinib, 0.82 (0.27-2.46) with nilotinib, and 2.86 (0.59-13.8) with ponatinib (reference: imatinib, P =0.14). Ponatinib was the only TKI associated with a significant decrease in LVEF compared to baseline with a coefficient (95% CI) of -4.56 (-8.4, -0.75) (P=0.02). The starting daily doses for patients on ponatinib were 15 mg (n=13), 30 mg (n=2), and 45 mg (n=20). Among TKIs analyzed, none was associated with significant changes in PASP or GLS. Conclusions: Tyrosine kinase inhibitor use is associated with an increased risk of major adverse cardiovascular events; the highest risk is associated with ponatinib. We also found that ponatinib was the only TKI associated with a significant decrease in LVEF over serial measurements. Cardiotoxicity of TKIs is a serious clinical concern that can lead life-threatening complications.