Comparative 3- and 5-year outcomes among first-line imatinib, dasatinib, and nilotinib in chronic myeloid leukemia: A real world analysis

Abstract

Abstract Background: First-line tyrosine kinase inhibitor (TKI) selection in chronic myeloid leukemia (CML) impacts both molecular response and long-term safety, as well as healthcare utilization. Although imatinib, dasatinib, and nilotinib are all guideline-endorsed frontline agents, comparative real-world outcome data remain limited. We evaluated 3- and 5-year clinical outcomes using a large multicenter electronic health record dataset. Methods: We queried the TriNetX US Collaborative Network (2000–2022) for adult patients with CML who initiated monotherapy with imatinib, dasatinib, or nilotinib. Patients with prior TKI use or less than 3 years of follow-up were excluded. Propensity score matching adjusted for age, sex, race, comorbidities, and sociodemographic factors. Outcomes at 3 and 5 years included all-cause mortality, hospitalization, ICU admission, emergency department visits, hypertension, coronary artery disease (CAD), and diabetes mellitus. Risk ratios (RRs), 95% confidence intervals (CIs), and p-values were calculated. Results: Among 7,296 matched patients (2,346 each for imatinib and dasatinib; 1,302 each for imatinib and nilotinib), patients who received imatinib had the lowest 3-year mortality (9.2%) when compared to dasatinib (11.5%, RR 1.25, CI 1.05–1.48, p<0.05) and nilotinib (10.0%, RR 1.02, CI 0.81–1.28, p=0.89). At 5 years, imatinib recipients also had a lower 5-year mortality (11.2%) when compared to dasatinib (13.8%, RR 1.23, CI 1.06–1.44, p=0.008), while similar rates were observed with patients who were administered nilotinib (12.5% vs. 12.9%, RR 0.96, CI 0.79–1.18, p=0.715). Hospitalization rates at 3 years were modestly higher with imatinib (18.2%) when compared to nilotinib (16.7%, RR 0.91), while ICU admission rates were higher with dasatinib (7.5%) than imatinib (6.2%). At 5 years, imatinib recipients had the highest hospitalization rate (23.5%), while nilotinib had the lowest (21.2%), though this was not statistically significant. Dasatinib was associated with more ICU admissions (9.4%) and a significantly lower risk of diabetes (6.9% vs. 9.1%, RR 0.76, 95% CI 0.61–0.94, p=0.012). Rates of hypertension and coronary artery disease (CAD) were similar across all treatment groups. Conclusions: Imatinib showed the most favorable survival at 3 and 5 years. Dasatinib was associated with higher mortality but lower diabetes risk, while nilotinib offered similar survival and a trend toward reduced hospitalizations. These results support personalized TKI selection in frontline CML therapy.