Major adverse cardiovascular events (MACE) in patients with chronic myeloid leukemia treated with tyrosine kinase inhibitors: A target trial emulation study

Abstract

Abstract Background: Since tyrosine kinase inhibitors (TKIs) have transformed chronic myeloid leukemia (CML) into a chronic disease with near-normal life expectancy, cardiovascular toxicity has emerged as a leading cause of non-CML mortality. Yet, comparative risk of cardiovascular adverse events across TKIs remains poorly defined. Pivotal trials lack standardized cardiovascular endpoints, long-term follow-up, or sufficient power. We conducted a real-world, target trial emulation to assess major adverse cardiovascular event (MACE) risk across available TKIs. Methods: We emulated five target trials using TriNetX, a federated U.S. EHR network capturing patient data across 68 healthcare systems. Adults with BCR/ABL+ CML initiating a single TKI between 2013–2024 were included. Patients were excluded if they had prior myocardial infarction, stroke, or hepatic failure within 3 months of TKI initiation, ECOG ≥3, HIV infection, or pregnant. The primary trial compared imatinib to pooled second/third-generation TKIs (bosutinib, dasatinib, nilotinib, ponatinib); to account for treatment switching, secondary trials compared imatinib to each agent individually. Propensity score matching was used to adjust for confounding of 59 baseline covariates including demographics, cardiac risk factors, comorbidities, and prior medications. The primary outcome was 10-year composite, 3-point MACE (MI, stroke, cardiac death) and overall survival. Secondary outcomes included acute coronary syndrome (ACS), heart failure, and trends in lipid profile. Hazard ratios (HRs) with 95% confidence intervals (CIs) were estimated using Cox proportional hazards models in the matched cohorts. A Bonferroni correction (α = 0.0071) was applied across seven outcomes, maintaining an overall type I error rate of 0.0487. Sensitivity analyses included subgroup stratifications by age, sex, comorbidity burden, and treatment duration. Results: Among 10,462 eligible patients, 3,872 imatinib-treated patients were matched to 3,872 second/third-generation TKI recipients (mean [SD] age 59.6 [15.9] years; 43% female; 70% White; 12% Black; 3% Asian; 6% Hispanic/Latino), of which 59% received dasatinib, 28% nilotinib, 8% bosutinib, and 5% ponatinib. Mean follow-up was 51 months. Second-/third-generation TKIs were associated with increased risk of composite MACE (HR 1.73; 95% CI 1.49–2.01; p<0.001), all-cause mortality (HR 1.34; 95% CI 1.19–1.51), myocardial infarction (HR 2.21; 95% CI 1.75–2.78), stroke (HR 1.55; 95% CI 1.27–1.89), acute coronary syndrome (HR 2.15; 95% CI 1.73–2.67), and heart failure (HR 1.27; 95% CI 1.13–1.41), all p<0.001. No difference in cardiovascular death was observed (HR 0.98; 95% CI 0.69–1.42; p=0.932). Longitudinal lipid profiles were more atherogenic in second-/third-generation TKI users at 6, 12, 36-, 60-, 90-, and 120-month timepoints (LDL cholesterol: 98.1 vs. 83.4 mg/dL at 120 months, p<0.001; total cholesterol: 175.1 vs. 157.3 mg/dL at 120 months, p<0.001; HDL cholesterol: 50.5 vs. 45.4 mg/dL at 120 months, p<0.001). Among individual TKIs, ponatinib had the highest MACE risk (HR 3.38, 95% CI 1.50-7.59; p<0.001), followed by nilotinib (HR 2.06; 95% CI 1.59–2.66; p<0.001), dasatinib (HR 1.38; 95% CI 1.10–1.74; p=0.006); bosutinib showed nonsignificant risk after correction. Subgroup analyses confirmed elevated MACE risk across clinical strata, particularly with treatment >5 years (HR 2.47; 95% CI 1.75–3.49). Conclusions: In this large real-world target trial emulation, second- and third-generation TKIs were associated with significantly higher risk of MACE and mortality compared to imatinib. Among individual agents, nilotinib and dasatinib demonstrated the highest cardiovascular risks, while bosutinib showed trends toward increased risk without reaching statistical significance after correction. These findings highlight the importance of cardiovascular risk stratification when selecting TKIs and underscore the need for vigilant long-term cardiovascular monitoring, referral and evaluation by a cardio-oncologist.