Comparative impact of tyrosine kinase inhibitor dose-reduction on ability to achieve treatment-free remission among TKI-intolerant patients with chronic myeloid leukemia

Abstract Background BCR-ABL1 tyrosine kinase inhibitors (TKIs) improve life expectancy to that of the general population for the majority of patients with chronic myeloid leukemia (CML). Second- and 3rd-generation TKIs improve rates of major molecular remission (MMR), with data suggesting treatment-free remission (TFR) is a feasible goal for patients sustaining a 2-year MMR. TFR is now an important outcome of interest when comparing CML treatment strategies. In patients who experience intolerable AEs, a switch in TKI has been traditionally recommended. However, TKI dose reduction (DR) is becoming a more common practice. More data are needed to determine whether this approach compromises the ability to achieve TFR. Methods We conducted a retrospective chart review of 301 patients with CML seen at the James Comprehensive Cancer Center from 2012-2022. All data were collected during standard clinical care. Patients who were intolerant to initial TKI and proceeded with DR or TKI switch were included. We hypothesized that the proportion of patients achieving TFR at the end of follow-up would be similar among these groups. We also compared time to and number of subsequent TKI switches as a surrogate for loss of effect or persistent toxicity. Differences in the distribution of demographic, clinical and treatment response were analyzed for significance using Mann-Whitney or Fisher's exact test. Time to subsequent TKI switch was calculated from the date of first TKI switch/DR to the start of the subsequent TKI switch with death and loss of follow-up without a switch treated as competing and censorable events, respectively. Competing risk regression was used to evaluate the association with the time to subsequent TKI switch between TKI DR and switch groups. Results Sixty-nine TKI-intolerant patients underwent initial DR and 232 patients underwent TKI switch. The median age for patients at diagnosis was 48. 49.7% of patients were female and 81.3% of patients were white. The percentage of patients initially started on imatinib was 60.8% in the TKI switch group, compared to 31.9% in the DR group, and the percentage of patients started on dasatinib was 27.2% in the TKI switch group, compared to 58% in the DR group (p<0.01). The remainder of patients were started on nilotinib or bosutinib. There were no significant differences in age, gender, race, white blood cell or platelet count at diagnosis. After initial TKI DR or switch, there was no statistically significant difference between the proportion of patients achieving an MMR when comparing the TKI switch and DR cohorts (63.8% vs. 70.7%, p=0.45) or deep molecular remission (DMR) (34.5% vs. 39.0%, p=0.71), nor was there a difference between the proportion of patients achieving an MMR (65.7% vs. 76.1%, p=0.13) or DMR (32.4% vs. 41.8%, p=0.19) at the end of follow up (median 4.8 years). Thirty-nine (56.5%) patients in the DR group and 131 (56.5%) patients in the switch group required a subsequent TKI switch. The mean number of subsequent TKI switches in the dose reduction group was 0.97, compared to 0.81 in the TKI switch group (p=0.32). The median time to subsequent TKI switch in the dose reduction group was 1.94 years compared to 2.25 years in the TKI switch group (p = 0.72). ​ At the time of last follow up 15.9% of patients in the DR group, and 9.5% of patients in the TKI switch group were in TFR (p=0.19). The median time in TFR for the DR group was 2.94 years compared to 2.51 years in the TKI switch group (p= 0.99). Discussion TFR is a desirable therapy goal for many patients with CML. In this 301-patient cohort, there was no statistically significant difference between the rate or duration of TFR when comparing those who initially underwent DR to those who underwent TKI switch. Patients treated with either approach were equally likely to achieve initial and long-term MMR and DMR, and subsequent TKI switch was not earlier in patients receiving an initial DR. This data adds to the currently available literature suggesting DR is a reasonable approach to TKI intolerance and does not put patients at risk of developing resistance that would impede the ability to achieve optimal CML treatment goals. Prospective randomized clinical trials are needed to more rigorously evaluate TKI DR as a treatment strategy.