Dosing variability of tyrosine kinase inhibitors in chronic myeloid leukemia

Abstract

Abstract Introduction Tyrosine kinase inhibitors (TKIs) have been the standard therapy for chronic myeloid leukemia (CML) since 2001. The current approved doses of these TKIs are largely derived from phase I studies yet the optimal starting and maintenance dose to ensure successful long-term outcomes remains unclear. Here, we report our experience of TKI dosing in patients with chronic phase (CP) CML treated at our institution. Methods We conducted a retrospective chart review to evaluate the patterns of choice and dosing of upfront and maintenance TKIs in patients with CP-CML at our institution from January 2005 to December 2024. Patients in treatment-free remission (TFR) were excluded from the analysis to focus on active TKI therapy. Institutional review board approval was obtained. Results We identified 216 patients from electronic medical records. We excluded 39 patients who attempted TFR with the final cohort consisting of 177 patients with complete data available. The median follow-up for the cohort was 7.5 years (range 0.6-36.9 years). The median age at diagnosis was 56 years (range 10-87 years) and majority of the patients were females (59%) and non-Hispanic Whites (69%). The median number of TKIs utilized were 2 (range 1-6) and 49%, 25%, 12% and 14% were on 1, 2, 3 and 4+ TKIs, respectively. Frontline TKI included imatinib, dasatinib, nilotinib and bosutinib in 46%, 32%, 19% and 3% of the patients, respectively. In the frontline setting 21% of the patients started on a lower TKI dose (imatinib <400mg/day=5%, dasatinib <100mg/day=47%, nilotinib <600mg/day=13% and bosutinib <400mg/day=33%). The latest TKI therapy included dasatinib, imatinib, bosutinib, nilotinib, asciminib and ponatinib in 34%, 29%, 18%, 11%, 6% and 2% of the patients, respectively. In patients on their latest TKI, more than half (n=100, 56%) were on a lower dose TKI at last follow up; dasatinib<100mg/day=67%, imatinib<400mg/day=41%, bosutinib<400mg/day=74%, nilotinib<600mg/day=42%, asciminib<80mg/day=41% and ponatinib<45mg/day=75%. Patients on a lower dose of the latest TKI, compared to standard dose, were older (56y vs. 46y) and had seen few prior TKIs (median 2 vs 3). The average daily dose of the latest TKI for dasatinib, imatinib, bosutinib, nilotinib, asciminib and ponatinib was 66mg/day, 246mg/day, 300mg/day, 397mg/day, 60mg/day and 30mg/day, respectively. For patients on a reduced dose of their latest TKI (n=100, 56%), 44% started at a reduced dose and 56% were reduced from a higher dose. The main reasons for dose reduction were intolerance (70%), decreased for good response (7% - mainly patient preference to remain on low dose rather than attempt TFR) or both (23%). At the time of data cut off, 35 patients were deceased. For all patients on their latest TKI, 55% were in MR4/MR4.5, 15% in MMR and 15% in CCyR. For those on a lower dose of the latest TKI, 58% were in MR4/MR4.5, 17% in MMR and 14% in CCyR, similar to those maintaining full dose. At last follow up, responses of MMR or better on a lower dose of the latest TKI were 75%, 81%, 78%, 75%, 40% and 67% for dasatinib, imatinib, bosutinib, nilotinib, asciminib and ponatinib, respectively. In patients on standard dose of their latest TKI, 64% were in MMR or better response at last follow up. Conclusion We report that approximately a fourth of the patients were started at a lower dose TKI for newly diagnosis CP-CML, with imatinib and dasatinib being the most commonly used agents. We observe that nearly a third of patients that started on imatinib and nilotinib had to switch TKIs and that bosutinib was more commonly used in later line settings. Most importantly, we note that more than 50% of the patients are on a lower dose TKI in the later line settings without compromising outcomes, highlighting that there is room for optimization of TKI dosing strategy in CP-CML for improved patient tolerance while maintaining outcomes. Larger registry studies are needed to better understand prescription practices of TKIs in CP-CML.