Outcomes of veterans diagnosed with chronic myeloid leukemia in the tyrosine kinase inhibitor era.

Abstract

e18567 Background: Since the development and approval of tyrosine kinase inhibitors (TKIs) targeting BCR-ABL, the treatment landscape for chronic myeloid leukemia (CML) has dramatically changed. The use of TKIs has improved long term outcomes for patients with CML, however, these medications have varied side effect profiles including increased risks of arterial thrombosis and bleeding. We assessed the outcomes of Veterans diagnosed with CML in the TKI era. Methods: We created a cohort of all Veterans diagnosed with CML (based on ICD-9 and ICD-10 codes) evaluated within the VA system between January 1, 2005, and December 31, 2023. We performed statistical analysis on the entire cohort—comparing outcomes to a large control group of 586,619 Veterans without CML—and a subset that had relevant medication data available. Survival analyses were conducted setting time 0 at age 55, subjects who had events before time 0 were excluded. CML was a time-dependent variable. The primary outcomes for this study were overall survival and the rates of arterial thrombosis and bleeding among Veterans diagnosed with CML compared to a control group. Secondary outcomes included a comparison of the different TKIs and the hazard ratio (HR) of arterial thrombosis and bleeding. Results: The total number of Veterans seen within the VA system with a diagnosis of CML included in the analysis was 10,869. Medication data for this cohort was available in 5,473 patients (50.4%). The risk of death after the age of 55 was 4.01 (95% CI 3.88-4.15) times higher among patients diagnosed CML compared to our control group. The risk of arterial thrombosis was 4.84 times (95% CI 4.59-5.09) and the risk of bleeding was 6.47 times (95% CI 6.11-6.86) greater in among the CML cohort when compared to the control group. Nilotinib conveyed the greatest risk of arterial thrombosis when compared to imatinib, HR of 1.34 (95% CI 1.01-1.78). Patients treated with dasatinib were also shown to have increased risk of thrombosis compared to imatinib, HR 1.29 (95% CI 1.00-1.66). The risk of bleeding was highest among patients treated with dasatinib, HR 1.46 (95% CI 1.21-1.90) compared to patients treated with imatinib. Bleeding risk was also elevated in patients who received nilotinib when compared to those treated with imatinib, HR 1.06 (95% CI 0.75-1.51). See Table 1. Conclusions: Our analysis demonstrated worse overall survival among patients diagnosed with CML compared to a control group in a time where TKIs became the standard of care. We also showed increased rates of arterial thrombosis and bleeding associated with 2nd generation TKIs compared to imatinib. Hazard ratios (with 95% CI in parentheses) comparing rates of arterial thrombosis and bleeding among patients treated with 2nd generation TKIs to those treated with imatinib (n=2828). Nilotinib (n=474) Dasatinib (n=955) Arterial Thrombosis 1.34 (1.01-1.78) 1.29 (1.00-1.66) Bleeding 1.06 (0.75-1.51) 1.46 (1.21-1.90)