Real-world comparative outcomes of first-generation versus second- and third-generation tyrosine kinase inhibitors in chronic myeloid leukemia: A global propensity-matched analysis.

Abstract

e18595 Background: Second- and third-generation tyrosine kinase inhibitors (TKIs) are frequently used in chronic myeloid leukemia (CML) due to greater BCR–ABL inhibitory potency; however, real-world comparative effectiveness and safety relative to first-generation TKIs remain incompletely defined. We evaluated clinical outcomes associated with first-generation versus later-generation TKIs in a large multinational real-world cohort. Methods: We conducted a retrospective cohort study using data from 170 healthcare organizations in the TriNetX Global Collaborative Network. Adults (≥18 years) with BCR–ABL–positive CML treated with first-generation TKI (imatinib) or second-/third-generation TKIs (dasatinib, nilotinib, bosutinib, ponatinib, or asciminib) were identified. The index date was initiation of the qualifying TKI within three years of diagnosis. Outcomes were assessed from day 1 to day 1095 post-index. Propensity score matching (1:1) for demographics and comorbidities yielded 6,270 patients per cohort. Outcomes included hospitalizations, arterial thrombotic events, bleeding events, hepatotoxicity, blast crisis, and all-cause mortality. Risk and Kaplan–Meier analyses excluded patients with prior outcomes. Results: Baseline characteristics were well balanced after matching (all standardized differences <0.05). Over a median follow-up of 2.2 years, first-generation TKI therapy was associated with a lower risk of hospitalization (15.8% vs 17.7%; risk ratio [RR] 0.89, 95% CI 0.80–0.99; p=0.028) and improved hospitalization-free survival (hazard ratio [HR] 0.80, 95% CI 0.72–0.90; p<0.001). Bleeding events occurred less frequently with first-generation TKIs (4.7% vs 5.9%; RR 0.80; p=0.004), with superior bleeding-free survival (HR 0.74, 95% CI 0.63–0.87; p<0.001). Hepatotoxicity was rare but less frequent with first-generation therapy (0.2% vs 0.4%; HR 0.49, 95% CI 0.25–0.96; p=0.033). Rates of arterial thrombotic events were similar (4.5% vs 4.5%; p=0.98). Blast crisis occurred more often among patients receiving second-/third-generation TKIs (1.4% vs 2.9%; RR 0.50; p<0.001; HR 0.46, 95% CI 0.36–0.60). Overall mortality was similar by risk analysis (9.8% vs 10.6%; p=0.14), although time-to-death favored first-generation therapy (HR 0.86, 95% CI 0.77–0.96; p=0.005). Conclusions: In this large real-world analysis, first-generation TKI therapy was associated with lower hospitalization rates, fewer bleeding and hepatic adverse events, and comparable overall survival relative to second- and third-generation TKIs. The higher observed incidence of blast crisis among later-generation TKI recipients likely reflects residual confounding by indication, as these agents are preferentially used in patients with higher-risk or treatment-resistant disease.